o1n01~ NOU'f'nl'f'A)UiiONioliONI v2 54435 \ie)IW I J:ll I >IN\18 Ol~OM !)31 THE WORLD BANK GROUP WORKING FOR A WORLD FREE OF POVERTY The World Bank Group consists of five institutions-the International Bank for Reconstruction and Development (IBRD ), the International Finance Corporation (IFC), the International Development Association (IDA), the Multilateral Investment Guarantee Agency (MIGA), and the International Centre for the Settlement of Investment Disputes (ICSID). Its mission is to fight poverty for lasting results and to help people help themselves and their environment by providing resources, sharing knowledge, building capacity, and forging partnerships in the public and private sectors. THE INDEPENDENT EVALUATION GROUP ENHANCING DEVELOPMENT EFFECTIVENESS THROUGH EXCELLENCE AND INDEPENDENCE IN EVALUATION The Independent Evaluation Group (IEG) is an independent, three-part unit within the World Bank Group. lEG-World Bank is charged with evaluating the activities of the IBRD (The World Bank) and IDA, IEG-IFC focuses on assessment of IFC's work toward private sector development, and IEG-MIGA evaluates the contributions of MIGA guar- antee projects and services. lEG reports directly to the Bank's Board of Directors through the Director-General, Evaluation. The goals of evaluation are to learn from experience, to provide an objective basis for assessing the results of the Bank Group's work, and to provide accountability in the achievement of its objectives. It also improves Bank Group work by iden- tifying and disseminating the lessons learned from experience and by framing recommendations drawn from evaluation findings. Inside- Acknowledgments Foreword Executive Summary Chairperson's Comments: Informal Subcommitee on Development Effectiveness (CODE) Abbreviations Table of Contents to the Complete Volume lEG WORLD BANK IFC MIGA World Bank Engagement at the State Level The Cases of Brazil, India, Nigeria, and Russia -Evaluation Summary- hltp:/lwww/ifc.org.lcg 20091] The World Bank http:/\vww.miga.org.ieg Washington, D.C. ©2009 The International Bank for Reconstruction and Development I The World Bank 1818 H Street NW Washington, DC 20433 1Clephone: 202-473-1000 Internet: www. worldbank.org E-mail: feedback@, worldhank.org All rights reserved 1 2 3 4 5 12 11 10 09 This volume, except for the "Chairperson's Comments," is a product of the staff of the Independent Evaluation Group 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 docs 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. The World Bank Group docs 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 judgement on the part of The World Bank Group concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development I The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center Inc., 222 Rosewood Drive, Danvers, MA 01923, USA; telephone: 971-l-750-8400; fax: 978-750-4470; Internet: \Vww.copyright.com. All other queries on rights and licenses, including subsidiary rights, should he addressed to the Office of the Publisher, The World Bank, 1818 H Street Nw, Washington, DC 20433, USA; fax: 202-522-2422; e-mail: puhrights~l. worlclhank.org. Cover: Illustration by Nip Rogers, IDimages.com/CORBIS, reproduced by permission. ISBN :978-0-8213-H224-0 e-ISBN: 978-0-8213-H225 DOl: 10.1596/97H-O-H213-H224-0 Library of Congress Cataloging-in-Publication data have been applied for. World Bank InfoShop Independent Evaluation Group E-mail: pic(h;worldhank.org Communication, Learning, and Stratq,'y (IEGCS) 1Clephone: 202-458-5454 E-mail: icg((~worldhank.org Facsimile: 202-522-1500 'lelephone: 202-451-l-4497 Facsimile: 202-522-3125 0 Printed on Recycled Paper Acknowledgments This report was prepared by the Independent (Brazil), Tapas Sen (India), Olufemi Taiwo Evaluation Group Country Evaluation and Regional (Nigeria), and Galina Kurlyandskaya (Russia). Relations evaluation team: Konstantin Atanesyan (fask Team Leader, Senior Evaluation Officer), Basil Ravi Kanbur (Cornell University), Steven Webb Kavalsky (consultant), and Sarwat]ahan (consul- (Adviser, LCSPS), and Thomas O'Brien (Manager, tant). It is based on the results of field visits to all SARDE) were the peer reviewers. Sarwat Jahan four countries and interviews with key local (federal and Tim De Vaan provided data and research and state) counterparts and Bank teams. assistance. Cecilia Tan and Corky de Asis provided administrative support and William The study draws extensively on four country Hurlbut provided editorial support. Caroline background papers prepared by Joao Oliveira McEuen edited the report for publication. Director, Independent Evaluation Group-World Bank: Cheryl Gray Senior Manager, lEG, Country Evaluation and Regional Relations: Ali Khadr Task Manager: Konstantin Atanesyan Note: Vinod Thomas, Director-General, Evaluation, was recused from management oversight of the evaluation owing to his prior association with the Brazilian program as country director. iii The Women in Movement Project sponsors an AIDS prevention clinic. BraziL Photo by Alejandro Lipszyc, courtesy of the World Bank Photo Library. Foreword This report summarizes the past 10 years lending, the aim being to derive synergies from the (1998-2008) ofWorld Bank engagement at the mix. The instruments deployed by the Bank state level in selected large federal countries: evolved over the review period and included state- Brazil, India, Nigeria, and Russia. This pilot cross- level development policy loans, multisector results- country study combines elements of a country based investment lending, and reimbursable assistance evaluation and a thematic review, technical assistance. There was considerable looking at the evolution of four country strate- successful innovation in the development of the gies as well as the Bank's state-level operations. instruments used, yet little knowledge sharing The evaluation limited its review to selected among countries. cases of lending and analytic work where the state governments were the principal partners of The report identifies lessons and good practice the Bank and the primary party responsible for examples that warrant further examination and development outcomes. wider dissemination. First, the study confirms the desirability of continued selective lending in Evaluating state-level engagement posed several a few focus states. The Bank's engagement with strategic and operational questions, among them progressive, reformist states has added value and the selection of states, the scope, and the modali- has been highly appreciated, but to enhance the ties of engagement. Two ideas-often at odds poverty impact of state-level interventions, with one another-featured in most approaches greater weight should be given to the needs of to selection of states for direct engagement. One the poorest states by balancing states' propensity was to support better-performing, reformist to reform and the concentration of poverty states (the lead or Jocu._<; states approach), while within them. Experience shows that it has been the other was to support the poorest states as a possible to achieve results in some of the poorer, more direct route to reducing poverty low-capacity states through persistent work with committed state counterparts and partnerships The initial area of engagement was typically fiscal with other donors. Second, continued focus on reform, where the Bank generally helped to public finance management appears sound, enhance the capacity of state governments for irrespective of whether engagement is confined public financial management. In some states Bank to this area or serves as an entry point for involvement extended to multisector engagement broader engagement. Third, there is consider- that usually involved a mix of analytic work, able scope for greater impact from knowledge development policy lending, and investment transfer and expanded knowledge services. Cheryl W Gray Director, lEG-World Bank v Gathering for a meeting in India. Photo by Curt Carnemark, courtesy of the World Bank Photo Library. Executive Summary eginning in the late 1990s, the World Bank significantly expanded its B engagement at the state level in Brazil, India, Nigeria, and Russia. This pilot cross-country study reviews selected cases of World Bank lend- ing and analytic work at the state level in those four large, federated countries. In each case, state governments were the Bank's principal development part- ners. The study looks at the evolution of the four country strategies, and the Bank's mode of engagement at the state level, to glean lessons from that ex- perience for both the Bank and its federal and state partners. State-level engagement posed several strategic The following findings are worth highlighting: and operational questions, among them which states to engage, the scope of engagement, and • First, the study confirms the desirability of con- the modalities of engagement. The Bank set out tinued selective Bank lending in a few states. But its approach to selecting states in country the poverty impact of those interventions could strategy documents. Two tendencies-often at be enhanced by balancing the propensity of odds-featured in most approaches. One was to states to reform and the concentration of support better-performing, reformist states (the poverty within them, giving greater weight to lead or focus states approach). The other was to the needs of the poorest states. support the poorest states as a more direct route • Second, continued focus on public finance to reducing poverty. management as the core area appears sound, irrespective of whether engagement is con- Concerning the scope of engagement, the initial fined to this area or it serves as an entry point and principal area of engagement was typically for broader engagement. fiscal reform-fiscal sustainability, medium- • Third, there is considerable scope to gain term fiscal frameworks, strengthening the public greater impact from analytic work, knowledge financial management capacity of state govern- transfer, and expanded knowledge sharing- ments, and fiscal federalism. In some states, not so much in concepts and theories as in the Bank involvement extended beyond fiscal practical experience of what works and what reform to multisector engagement focused on does not. the growth and poverty reduction agenda. The modalities of engagement and the instruments Evolution of Bank Strategy deployed by the Bank evolved over the review period and included state-level development Why State-Level Engagement? policy loans, multisector results-based invest- Over the past decade, the World Bank signifi- ment lending, and reimbursable technical cantly expanded its engagement at the state level assistance. There was considerable successful in four large federations: Brazil, India, Nigeria, innovation in this area, yet little knowledge and Russia, mainly through lending, but also sharing among countries. through policy dialogue, technical advice, and WORLD BANK ENGAGEMENT AT THE STATE LEVEL India 72 107 67 12.7 22.9 57 Brazil 47 101 47 5.2 15.9 33 China 71 112 63 11.5 15.7 73 Argentina 18 57 32 3.6 11.6 31 Mexico 5 49 10 0.85 11.5 7 Pakistan 26 40 65 2.0 7.4 27 N' a 19 32 59 1.8 3.1 58 Source: World Bank data. lmageBank. analytic work. Both demand and supply factors delivery in the relevant social and economic contributed to this expansion. sectors at the state level. On the demand side, federal governments On the supply side, wit h the combination of focused on fiscal stabilization following the fiscal stabilization and improvement in the financial crises of the late 1990s. They saw the fiscal situatio n of the four countries during the potential for the Bank to provide state govern- commodity boom of 2000-07, there was ments with incentives for reform through financ- limited appetite to borrow from the Bank for ing, while encouraging discipline through agreed federal programs (this trend was more policy measures and provision of technical pronounced in Brazil and Russia; Nigeria, an support for implementation . Federal govern- International Development Association (IDA] ments had limited scope to differentiate among borrower, was an exception; in India, federal- states based on factors such as commitment to level borrowing increased slightly in 2004-07). reform. They saw in the Bank's capacity to do A level of Bank engagement commensurate this a means of assisting states willing to take the with the size and importance of these countries lead, with the possibility of a demonstration almost mandated the sh ift to the state leve l, effect for other states. where demand remained buoyant. The increas- ing focus of the Bank on poverty reduction State governments were also eager to borrow after 1995 was also an important factor. There from the Bank because its loans, denominated in is a distance between federal-level programs foreign currency, generally came at lower rates and res ults on the ground in such large than those provided by the federal government cou ntries. The majority of the public expendi- or the domestic market. Bank loans, while often ture categories most closely associated with financially modest at the federal level, could be a poverty reduction in the short and lo ng terms major source of financing at the state level. State are usually state responsibilities in these governments welcomed the Bank's focus on countries. Therefore, increasing the Bank's their economy as well as the associated dialogue, impact on poverty reduction meant increasing advice, and analytic work. In addition, with the the focus on and activities at the state level. In increasing concern about meeting or achieving addition, many Bank country and sector staff outcomes with regard to the Mi llennium fou nd work at the state level in these countries Development Goals, both federal and state more rewarding , given the clients ' greater governments saw the Bank as having a compara- interest in the Bank's financial and knowledge tive advantage in supporting better service resources. 2 EXECUTIVE SUMMARY Which States? corner fiscally, despite significant disparity in The shift to the state level presented the Bank their income levels. The 2008 Country Partner- with a number of operational issues. Among them ship Strategy (CPS) (World Bank Group 2008) was which states the Bank should engage. The focused on a technical assistance program of four countries have large numbers of states-26 modest size with the federal government and a states and a Federal District in Brazil, 28 states major demand-driven lending program with and 7 Union Territories in India, 36 states and a states, conditioned on perceived commitment, Federal Capital Territory in Nigeria, and 83 ownership of reforms, and fiscal responsibility. Regions ("subjects of the federation"-republics, The Bank engaged with some of the more oblasts, krays, and okrugs) in Russia. Working in prosperous and reformist states. Although all of them would obviously be beyond the Bank's initially the Bank attempted to expand invest- budgetary and human resource capacities. ment lending, this proved cumbersome given the two-tier approvals required by the state and The Bank defined the strategic approach to the federal governments. As a result of joint consul- selection of states in its Country Assistance tations with state and federal counterparts, by Strategies (CASs). Some attempts were made to 2008 the composition of state-level lending develop quantified criteria for selective engage- shifted toward cross-sectoral operations in ment, but the Bank generally preferred to keep support of economic policies and public sector the criteria broad to allow for flexibility. It is clear reforms (DPLs and multisector SWAps). from all the countries reviewed that there was tension between the Bank's interest in identify- In India, the Bank shifted its focus to the state ing progressive, reforming states that could serve level in the mid-1990s. At the time, states were as demonstrations to others and its interest in facing financial problems, and both the federal supporting poverty reduction by assisting the and state authorities were keen to tap into the poorest states. In addition to these two key Bank's resources and take advantage of technical criteria-effectiveness of assistance and need assistance. The Bank opted for major involve- (poverty)-another equally important criterion ment in progressive reforming states (the focus was the political economy, unique in each states approach). The 2004 CAS (World Bank country and including (but not limited to) Group 2004) signaled a change of strategy, noting relations between the federal center and the that the focus on reforming states was leading to states; capacity and political affiliation of the neglect of the lagging states. Therefore, the CAS state-level leadership; level of trust and the proposed the provision of technical assistance to relationship of the Bank teams with the clients; the lagging states and an effort to shift lending to and local politics and electoral cycles. them as well. This proved to be difficult. As the 2009 CAS completion report noted, while lending In Brazil, the shift toward states was proposed in at the state level remained a large share of the a mid-1990s CAS, directing lending to creditwor- overall program, the share of lagging states in the thy reforming states. The next CAS, prepared at program actually declined. the turn of the millennium, continued this approach, using criteria established by the In Nigeria, the Bank reactivated and intensified federal Fiscal Responsibility Law (FRL) and stress- its lending activities after return to civilian rule in ing the intention of providing assistance to the 1999. During that period, Bank strategy passed states of the Northeast region with highest through two phases: a period of interim strate- poverty levels. Lending to states became more gies (fiscal 2000-05) following re-engagement; multisectoral, and significant innovations were and the fiscal 2005-09 CPS (AfDB, DFID, USAID, introduced, such as state-level sectorwide and World Bank Group 2009), when the Bank lending (multisector SWAp) and state-level adopted a medium-term focus. Engagement at development policy loans (DPLs). Both instru- the state level in Nigeria was largely driven by the ments were applied in states that had turned the social and poverty reduction agendas, with focus 3 WORLD BANK ENGAGEMENT AT THE STATE LEVEL on improving infrastructure and providing finishing touch: the plan for gradual shift to the support for agricultural and rural development. new modalities of cooperation and instruments, During the CPS period of fiscal 2005-09, the such as the subnational facility that allows the Bank's strategy formally moved to focus on well- Bank Group to provide funds without a performing states (lead states), seeking to sovereign guarantee to states and municipalities leverage state efforts and resources by granting and provision of technical assistance on them access to a performance package. Five reimbursable basis (fee for service). states were selected based on the government- led State Economic Empowerment and Develop- The Scope of Engagement ment Strategy (SEEDS) benchmarking process. The second issue for the Bank was the scope of its engagement. In Brazil, India, and Nigeria, the In Russia, the mid-1990s CAS emphasized Bank had carried out numerous state-specific regional investment projects (despite recogni- projects as part of its support for nationwide tion that they were expensive to prepare and sectoral programs. The decision about which supervise). The next CAS (World Bank Group states to engage-for example, in support for 1999) outlined a phased shift in lending, away agriculture and rural development in India or from investment projects in infrastructure and education in Brazil-was sometimes a matter of energy in favor of increased emphasis on strategic choice, but more often a matter of histor- systemic aspects of institutional development. A ical engagement or the availability of analytic subsequent strategy in the early 2000s continued work, based in turn on opportunistic involvement the strategy shift, emphasizing support for of the Bank in particular states. In the new reforms at the regional level, particularly to context, the Bank was steering toward a new strengthen public sector management. The 2005 model of engagement. This tended to have two CAS Progress Report (World Bank Group 2005) elements: support for fiscal reform and broader stated that work at the regional level was to be multisector engagement at the state level. carried out in a multisectoral manner and would concentrate on a small number of regions in Fiscal Reform agreement with the federal government. The The Bank expanded its involvement in fiscal reform 2007 CPS (World Bank Group 2007) added a in selected states. With the focus on stabilization Figure 1: Distribution of Projects by State A. Brazil B. India Bahia 13% Other states 49% b.::--------! Andhra ,.-----1 Pradesh Minas 1~i% Gerais 10% 8% Sao Paulo 8% Rajasthan 14% 8% 8% 4 EXECUTIVE SUMMARY Box 1: Analytic Work on Fiscal Federalism in Russia With the dissolution of the Soviet Union and the need to put in Budgeting in Russia: Preempting a Potential Crisis (Freinkman, place new instruments and rules for the relationship between the Treisman, and litov 1999). central government and the regions and republics, the Bank un- With a significant change in the approach to fiscal federalism dertook a substantial program of analytic work designed to stim- that took place in 1999, the Bank published a Research Working ulate and contribute to internal debate and policy formulation. Paper on Decentralization in Regional Fiscal Systems in Russia: The studies published by the Bank included: Russia and the Chal- Trends and Links to Economic Performance (Freinkman and Yos- lenge of Fiscal Federalism (World Bank 1994), which outlined the sifov 1999). This was generally regarded as a major study on fis- nature ofthe center-regional issues and tensions and how a well- cal federalism in Russia and promoted an intensive dialogue designed intergovernmental system could help reduce them; Fis- between the Bank and the federal and regional governments on cal Management in the Russian Federation (Le Houerou 1995); these issues and laid the basis for the Regional Technical Assis- "Federal Transfers in Russia: Their Impact on Regional Revenues tance Loan at end-1999 and the Fiscal Federalism and Regional Fis- and Incomes" (Le Houerou and Rutkowski 1998); and Sub-national cal Reform Loan in early 2002. Source: World Bank documents. and the need to reduce growing state deficits or resources available to the state governments enhance state public expenditure management need to be balanced with their expenditure capacity; the Bank engaged in two areas. responsibilities. For many taxes it is more efficient to centralize collections. This discon- The first of these was fiscal federalism. Fiscal nect generally creates an imbalance between the relations between the federal and state govern- revenues collected by the states and their ments are politically highly sensitive, since this is development mandate. In Brazil this imbalance often at the core of balancing regional interests. is relatively modest, but in Nigeria it is very large. From a strictly economic perspective, the This means that the federal government is Box 2: The Fiscal Adjustment Program in Brazil As part of their 1997 debt restructuring agreement, states signed stabilization policies. Since the PAFs were signed, states have a Fiscal Adjustment Program (PAF) and committed to meet ne- shown significant improvement in most of the relevant indicators. gotiated targets on financial debt/net real revenue ratio, a min- There has been a reduction of debt ratios, increasing revenue imum primary surplus, maximum wage bill expenditure, minimum collections, and primary surpluses as a ratio of state net revenue. own-revenue collection, structural reforms and/or assets di- For some states, the proceeds from privatization were a significant vestiture, and level of investment expenditures. The PAF is athree- contributor to amortization of debt. The program has shown mixed year rolling fiscal program, annually monitored by the federal results with respectto structural reforms ofthe states, such as the government, and revised as needed, up to the 30 years during social security system. which the state is under the obligations of the debt restructur- Although specifically designed to resolve the states' debt cri- ing agreement. Of 26 states and the Federal District, only 2 states sis ofthe 1990s, PAFs have been critically important, paving the way (Tocantins and Amapa), which had no significant outstanding for the adoption of the Fiscal Responsibility Law (FRL) in 2000, debt, did not sign the agreement. which brought a more strategic approach for a sustainable fiscal The PAFs have been a powerful instrument for fiscal and struc- policy in the Brazilian federation. tural reforms at the state level and for supporting macroeconomic Source: lEG consultant reports and mission interviews. 5 WORLD BANK ENGAGEMENT AT THE STATE LEVEL required to transfer resources to the state, analytic work at the state level was combined generally based on various formulas that take with investment lending and technical account of population, per capita income, and assistance, but the Bank is only now considering the state government's own tax effort. In the the possibility of using multisector lending in countries reviewed, federal transfers have not support of its approach. In Andhra Pradesh and been very effective in reducing disparities in Orissa in India, the Bank carried out analytic expenditure capacity between states. work and subsequently used multisector loans to support fiscal reform, but technical advice, A second key issue in fiscal federalism concerns except that embedded in the analytic wo-k and discretionary transfers from the center to the lending activities, has not been a focus of the states, usually intended to provide an incentive approach. The Bank's involvement with tax to states to undertake high-priority programs. policy has been relatively light,; the a focus of its The government of India has used such schemes efforts has been on budgeting and public a great deal. For the Bank, the political sensitivi- expenditure management systems, and increas- ties make fiscal federalism a difficult area for ingly on the governance issues associated with intervention unless there is an explicit request transparency and accountability for efficient from the federal government. The Bank has service delivery. undertaken substantial analytic work in this area. Russia, where new fiscal relations were being Multisector Engagement defined in the past decade, is a very good In addition to fiscal engagement, the Bank has example. also undertaken a broader state-level multisec- tor engagement, which has focused on the A second and far larger part of Bank engagement growth and poverty reduction agenda. This has in fiscal reform is its direct support to public involved a mix of analytic work, adjustment finance management at the state level, including lending, and investment lending. The Bank enhancements in tax capacity, modernizing the produced an explicit strategy for its activities in a tax structure, developing a sustainable fiscal policy particular state in only a few cases. Such strate- and medium-term expenditure framework, and gies were usually covered in broad-brush improving budget and expenditure management. fashion in the CAS. The Bank's engagement model generally started with a trigger mechanism that required states to At its most developed, as in Andhra Pradesh in show commitment to fiscal reform. India and in Ceara in Brazil, the Bank programs combined investment lending in most of the This requirement was highly formalized in Brazil core economic and social sectors with multisec- where the federal government requires strict tor lending. The objective was to derive synergies adherence to the FRL. In India and Nigeria the from the combination of activities. In Andhra requirements were less formalized, but generally Pradesh, for example, the Bank was explicit in its related to timely budgeting and reporting. In view that the difficult measures required for fiscal Russia, etiteria were established for participation reform needed to be matched by increased in the fiscal reform projects supported by the investment in agriculture, rural development, Bank. Once the triggers were met, the Bank was health, and education to provide a politically able to further support fiscal reform through an acceptable package of reforms. engagement model that combined analytic work with m ultisector lending and focused technical The selection of states for this broader engage- assistance in the areas of fiscal and governance ment focused more on fiscal reform than on the reform. broader poverty issue. The Bank spent a great deal of time in these countries supporting Elements of this model are present in each relatively high-income, high-capacity states (such country. In Lagos, Nigeria, for example, intensive as Minas Gerais in Brazil, St. Petersburg in Russia, 6 EX E CU T IV E SUMMA R Y Box 3: The Bank Program in Andhra Pradesh Andhra Pradesh is considered a middle-income state in India. helped Andhra Pradesh to become the frontrunner in a number In the late 1990s, Andhra Pradesh became one ofthe first Indian of reform areas, with significant demonstration effect on other states to promote reform and fiscal adjustment and to reach out states: introducing a single-window clearance system for new in- for Bank assistance. At the time, the Bank's strategy in India was vestments; a-procurement on a wide scale, and so on. to targetthe reforming states. Andhra Pradesh was selected as Andhra Pradesh has also implemented a number of invest- the initial focus state. The culmination of the Bank- Andhra ment projects (in the rural poverty, forestry, water, and power sec- Pradesh partnership was the Andhra Pradesh Economic Re- tors) over fiscal 199~2008. The Bank has also supported structuring Project (APERP, 1998), a massive multisectoralloan significant nonlending activities in Andhra Pradesh. It started with underpinned by an agreed multiyear fiscal framework with atotal the "Andhra Pradesh: Agenda for Economic Refo rms" (World loan/credit value of $540 million. This was essentially a DPL that Bank 1997) report, which underpinned the subsequent lending had to be constructed as an investment loan, since Bank policy program.ln addition there have been several Andhra Pradesh Pol- did not approve of subnational DPls at that time. icy Notes covering issues such as fiscal and debt management After Bank policy on subnational DPls changed, Andhra and analysis of Andhra Pradesh's growth potential and public en- Pradesh received three DPls (fiscal 2002, 2004, and 2007). The terprise reform. Bank's first state-level policy-based loan (S-PBL) to Andhra The portfolio of Andhra Pradesh has changed with the Bank's Pradesh addressed not only the need for fiscal adjustment but also shift in strategy away from focus states to lagging states. While public expenditure management reform and restructuring of pub- Bank's commitments to Andhra Pradesh had reached $1.5 billion lic enterprises, including privatization. The later DPLs added three by 2000, after the strategy shift in fiscal 2004, the Andhra Pradesh specific·sectoral foci: power, health, and education. The Bank's portfolio gradually fell. In June 2004 it comprised 10 percent of S-PBL program was controversial in India because it censored the total lending volume ($1 .2 billion) and fell to 5 percent ($0.74 some populist measures such as free power and irrigation water billion) in June 2008. At the time of the 2008 C AS, only one proj- for farmers. Despite the initial setback in the power and irrigation ect planned for the state was considered firm ( Andhra Pra desh sectors, the S-PBLs not only ushered in fiscaf correction but also Water Sector Improvement). Source:World Bank documents and mission interviews. and Karnataka in Ind ia). Wh ile th is support Adjustment lending rapidly became the instru- added value, it came at the expense of Bank ment of choice to support fiscal reform and efforts in poorer states that lack capacity. statewide strategies in Brazil and India, but in Brazil there was a sense that it was less effective Modalities of Engagement in reaching out to the line ministries in key How to engage was also a concern for the Bank sectors. The Bank's Brazil country team during the period, and there was considerable developed the innovative approach of a multisec- evolution in the approach. The first bridge to be tor SWAp, a results-based instrument with target crossed was the use of adjustment (development indicators defined for each sector and disburse- policy) lend ing at the state level (the first such ments associated with achievement of the adjustment loan was made in March 2000 to the targets. Th is had very positive outcomes: it state of Uttar Pradesh in India) . Until that point brought to the fore the linkages required to the Bank had struggled to find an instru ment to achieve resu lts, such as the need for improved attach to its policy dialogue and strategic water supply in order to reduce infant mortality. approach at the state level. The mu ltisector Another important innovation was the pioneer- restructuring loan in Andhra Pradesh, an invest- ing of reimbursable technical assistance at the ment loan u ndertaken in 1998, was a way to state level in Russia. Bank budgets rarely allow accomplish this, but it was an enormously costly the level of analytic work demanded by intensive operation to prepare and supervise. engagement in th ree-to-five states, and an 7 WOR L D BANK ENGAG E MEN T AT T HE STAT E LEVEL Box 4: Multisector SWAp SWAps traditionally have been associated with a single sector. programs. It allowed the government to manage its fiscal re- From this perspective, the SWAps in Brazil are innovative be- sources and allocations while forging a partnership between cause they were adapted to integrate several sectors and they the central and the line secretaries. were delivered atthe state level.a Ceara was the first state to im- This loan had varying degrees of success in achieving the de- plement a multisector SWAp.b This loan aimed to strengthen sired outcomes in the targeted sectors. It was very successful social inclusion while preserving fisca l sustainability. It had a in leveraging the Bank's support to ensure that sector expendi- unique design, because it was an Adaptable Program loan mak- ture levels, which had been about 28 percent of budgeted amounts ing use of a results-based SWAp loan modality. before the start of the project, were up to 70 percent. The loan The key innovative features of this loan were: also helped to reinforce other traditional investment projects. Two • Multisectoral Model and Activities. The loan had two com- of the three investment loans in Ceara were upgraded to satis- ponents. The SWAp component supported nine eligible ex- factory as a result of cross-conditionality. The downside of the penditure programs across five key line sectors: health, modality of this loan is that it is complex and it can be success- education, water and sanitation, water source management, fully implemented only if there is sufficient institutional capacity.c and environment. The technical assistance component pri - Multisector SWAps in Other Brazilian States: The Ceara marily supported public sector management. Targeting six SWAp model was sufficiently successful that a second SWAp sectors assured that institutional reform synergies would was approved ($240 million, 2008). The model is also being repli- 'take place across sectors. cated in Minas Gerais ($976 million, 2008), the Federal District • Results-Based Disbursement The project emphasized the use ($130 million, 2009), and Pernambuco ($155 million, 2009). Each of resu lts-based disbursement. Disbursement depended on: of these operations foll ows essentially the same SWAp struc- (a) the borrower had reached specific disbursement-linked ture, combining support to public sector management reforms indicators mapped to each sector; (b) borrowers' primary sur- with support for selected sector programs. In addition, each plus was above a certain threshold; (c) performance of Ceara one links disbursements to the borrower's abil ity to meet spe- on th ree other World Bank loans (water, education, and rural cific performance targets over the course of the loan. The na- development), ensuring that these loans meet stated physi- ture of the indicators varies in accordance with the institutional cal as well as disburseme nt targets; and (d) the borrower had development of the state/sector. For ex~mple , the Minas Gerais actually spent at least 70 percent of the amount budgeted for SWAp relies to a greater degree than the others on outcome the spec ific sectors. measures: app ropriate " stretch goals" for one of Brazil's most • Disbursement to Treasury and Not to Sectors. The Bank re- developed states. In contrast, Ceara SWAp-II will use mainly out- leased its funds directly to the state treasury as a single put measures rather than outcome targets as disbursement in- tra nche for reimbursement of expenditure under the various dicators for the sector prog rams. Source: World Bank documents and lEG mission materials. a. Before Brazil, only Mexico had implemented a multisector SWAp (fiscal 2004), but it was at the federa l level. b. Ceara Multi-Sector Soc ial Inclusion Development Project ($150 million, fiscal2006). c. Des pite being a poor state, Cea ra was considered to be relatively well managed and reform minded making it a viable candidate for implementing the SWAp . approach that permits states to pay for additional a large number of specific achievements. work has considerable promise for other middle- Although state-level engagement often requires income countries. additional effort and can be resource-intensive, it is usually worth the cost. Findings Overall, the analysis leaves little doubt that the The main findings of this review, which may be Bank's engagement at the state level did add helpful in gu iding the organ ization of future value. There was a great deal of enthusiasm at work at the state level, include: both the federa l and state levels in these countries regarding the Bank's contribution and On selection of states: 8 EXECUTIVE SUMMARY Box 5: Client Views Federal- and state-level clients broadly considered the Bank's approve the operations and provide the sovereign guarantee re- state-level engagement useful and timely. But views differed from quired by the Bank. The states take full responsibility for the fi- country to country on specific aspects, such as modalities of en- nancial execution and implementation. gagement and mix of products. In all four cases, federal governments have been supportive In Brazil, for example, both federal and state authorities of the Bank's strategy of state-level engagement. Most federal seemed to prefer DPLs and multisector SWAps. Among the ad- governments value the discipline associated with Bank lending vantages they noted were flexibility; disbursement speed; and, to the states. In Brazil, for example, the federal government ap- most important, the role of these loans in helping to establish and pears to put considerable weight on Bank lending as a mecha- institutionalize policy coordination in the state government. nism for reinforcing compliance with the FRL. In Nigeria, federal authorities expressed strong reservations In all four countries, a detailed and structured process at the fed- about adjustment lending in general, yet a few high-level state eral level for approving proposals for state borrowing is in place.• · officials expressed interest in multisector SWAps, and possibly In Brazil and Russia, the federal governments have seen borrow- even state-level DPLs. ing by states and regions as a way of maintaining substantial Bank In all cases, the federal governments have the final say about engagement and·presence in the country, despite the sizeable re- whether the Bank will lend to a particular state, because they source inflows and rising per capita income level of the past decade. Source: lEG mission interviews. a. While it would be na·ive to argue that political considerations do not enter into the state selection process, in practice decisions on the eligibility of states for selec· tion are dominated by technocratic considerations. In the choice of which of the eligible states to support, both governments and the Bank have generally tried to avoid the perception of bias by including both pro-government and opposition-led states. • The strategy to be selective and concentrate On the scope of engagement: lending services on a few states to enhance the impact of the Bank's program is right in prin- • Continued focus on public finance manage- ciple, but selection Cliteria and the mode of im- ment as the core area for state-level work ap- plementation cou ld give greater weight to the pears sound, whether engagement is confined needs of the poorest states. to this area or it serves as an entry point for • Bank engagement with high-performing states broader engagement. clearly added value, both strengthening in-state • The lending programs and Bank budgets in capacity and encouraging state-to-state knowl- some states are often larger than for many edge transfer, albeit mainly between high per- Bank clients. For states where the Bank plans formers. However, there is little evidence that or has a major engagement, a brief state strat- it had the desired demonstration effect on egy document could be a useful tool for defin- poor, lagging states on the scale the Bank hoped ing the scope of engagement and developing for, or that the Bank had an exit strategy to a medium-term outlook. permit increased focus on poorer states over time. At the same time, experience shows that On modalities of engagement: it is possible to achieve results in the poorest states through persistent work with committed • There is considerable .scope for greater im- state counterparts and strong partnership with pact from knowledge transfer and expanded the federal government and other donors. knowledge services. In particular, there is • It is important to stay engaged not only in strong demand for better knowledge sharing, states that are able to borrow from the Bank, both within the Bank and across the countries but also in states that have no fiscal space to concerned. This is not so much a matter of borrow but demonstrate a genuine commit- sharing of concepts and theories as it is of ment to development that can be supported communicating practical experience regard- through analytic work and technical assistance. ing what is working and what is not. 9 WORLD BANK ENGAGEMENT AT THE STATE LEVEL • Widening the scope and increasing the amount Le Houerou, Philippe, and Michal Rutkowski 1996. "Fed- of analytical work at the state level could be em! Transfers in Russia: Their Impact on Regional Rev- helpful in identifying high-impact, high-priority enues and Incomes." Comparative Economic Studies areas. There seems to be a potential for closer 38:21-44. partnership between state governments and World Bank. 1997. Andhra Pradesh: Agenda for Eco- the Bank in this area. nomic Rtiforms. Report No. 15901. Washington, DC: World Bank. References - - - . 1994. Russia and the Challenge ofFiscal Fed- AffiB (African Development Bank), DFID (U.K. De- eralism. Washington, DC: World Bank. partment for International Development), USAID World Bank Group. 2008. "Country Partnership Strategy (U.S. Agency for International Development) and for Brazil, 2008-11." Report No. 42677. World Bank, Wbrld Bank Group. 2009. "Country Partnership Strat- Washington, DC. Photocopy. e;.,')' for Nigeria, 2010-13." Report No. 46818-NG. - - - . 2007. "Russian Country Partnership Strategy. " World Bank, Washington, DC. Report No. 37901-RU. World Bank, Washington, DC. Freinkman, Lev, and Plamen lbssifov: 1999. Decentral- Photocopy. ization in Regional Fiscal Systems in Russia: Trends - - - . 2005. "Russian Federation Country Assistance and Links to Economic Peiformance. World Bank Strategy Progress Report." Report No. 31579. Wi:xld Policy Research Working Paper \X!PS 2100. Washing- Bank, Washington, DC. Photocopy. ton, DC: World Bank. - - - . 2004. "Country Strategy for India." Report No. Freinkman, Lev, Daniel 1reisman, and Stepan 'litov. 29374. World Bank, Washington, DC. Photocopy. 1999. Subnational Budgeting in Russia: Preempting - - - . 1999. "Country A~sistance Strategy for the Russ- a Potential Crisis. World Bank 1cchnical Paper No. ian Federation." Report No. 19897. World Bank, 452. Washington, DC: World Bank. Washington, DC. Photocopy. Le Houerou, Philippe. 1995. Fiscal Management in the Russian Federation. Report No. 14862. Washing- ton, DC: World Bank. 10 Chairperson's Comments: Informal Subcommittee of the Committee on Development Effectiveness (CODE) n September 21, 2009, the Informal Subcommittee of the Commit- 0 tee on Development Effectiveness (CODE) considered an Inde- pendent Evaluation Group (lEG) report entitled World Bank Engagement at the State Level: The Cases ofBrazil, India, Nigeria, and Rus- sia. Staff representatives from the four countries considered in the report were present at the meeting. Summary Recommendations and Next Steps Members commended lEG for the well-written The Subcommittee recommended the following and informative report and welcomed manage- to management: ment's broad agreement with its findings. The Subcommittee held a rich discussion, where • Consider this lEG report as a basis for further members noted the importance of this report as thinking toward a more comprehensive frame- a basis for further considering the direction and work to guide the Bank's engagement at the potential of state-level engagement by the Bank. state level, and to continue adjusting its in- Some members remarked on the considerable struments to meet the needs of the states. innovation and creativity in this area, which was • Maintain a flexible approach to selection of considered a critical aspect of a country's states, to accommodate different country con- development. Going forward, the need to bear texts while keeping in mind the poverty focus. in mind the poverty focus in engaging at the state • Consider ways to strengthen systematic knowl- level and to strengthen knowledge sharing edge sharing and learning from ongoing work within and across countries was noted. at the state level. Members' interventions focused on lEG's findings related to the need for the Bank to have Main Issues Discussed a strategic approach to state-level support, Framework for Bank Engagement. Some members selection of states, modalities of engagement, noted the potential benefit of lEG's suggestion and knowledge transfer, which are elaborated to prepare brief state-level strategy documents below. The comparative analyses of the experi- for states with significant long-term Bank ences in the four countries covered by the report engagement. Management clarified that the and staff elaboration of country experiences strategic anaZysis and dialogue with client were appreciated. countries on the Bank's engagement at the state 11 WORLD BANK ENGAGEMENT AT THE STATE LEVEL level takes place as part of the CAS process. It and approaches to addressing foreign exchange also expressed concerns about adding another risks at state levels. lEG and management layer of strategy paper, but noted that integrat- stressed that Bank engagement at the state level is ing separate strategies for the most important predicated on federal government support. state clients within a CAS could be a useful way Referring to the example ofBrazil, management to approach this issue. also noted that the federal government has encouraged Bank involvement at the state level Selection of States. Members supported selectiv- as a way to incentivize states to carry out ity in engaging at the state level and discussed difficult reforms (such as reducing the fiscal the approach to selecting states, i.e., whether the deficit). It also indicated that the Treasury Bank should work with more progressive, Department has offered assistance to Indian reformist states or with poorer states in light of states interested in addressing exchange rate lEG's findings, taking into consideration the risks. Bank's mandate to reduce poverty and the potential added value of focusing on lagging Scope and Modality of Support. Some members states. There was general consensus to preserve highlighted the potential for Bank support for flexibility to accommodate different country fiscal federalism and governance and capacity contexts and political economies and enable the building at the state level. A member sought more Bank to work with active state actors at the state clarity on the outcomes of fiscal reforms and and sub-state levels, while keeping in mind the another encouraged conducting Public Expendi- poverty focus. ture Reviews at the state level. Others emphasized the importance of analytical work at the state level, The importance of transparency in selecting including in lagging states. A few members sug- states was emphasized. lEG noted the tension gested that the IFC model of state-level support between the two approaches and reiterated that be reviewed. Management described some of the to enhance the pover~y impact of state-level ongoing ana~ytical work at the state level such as intervention, greater weight to the needs of the the Doing Bu..'>iness reports in Russia and Nigeria, poorest states was merited since there was no Public Expenditure Reviews in Nigeria, and clear evidence of the desired demonstration Social Expenditure Reviews in Russia. effect. lEG added that it is possible to achieve results through persistent work, even in the Demonstration Effect and Knowledge Sharing. A few poorest and low-capacity states. Management members sought more clarity regarding the commented on the country realities that have limited demonstration effect on lagging states, driven the selection of state~~ as in the case of noting that the experience in Russia seems to India, where the Bank is making an effort to have been more positive. Speakers stressed the work more with lagging states. Reference was importance of strengthening knowledge sharing also made to the CPS in Nigeria that describes across states and countries. For example, the the principles of engagement at the state level; a potential for other countries to learn from key principle is the level ofhuman development Russia's experience with fee-based services was indicators. noted. A few members urged that the Bank should give the highest priority to demand for Federal Government and States. Some questions services where the recipient was willing to pay a were raised about ownership and involvement of fee, because this indicated the seriousness and the federal government with regard to Bank the will to carry out reform and change. The assistance to states and any tensions encountered possibility of a Global Expert Team to improve between the federal and state levels. In addition, knowledge sharing among staff was suggested. interest was expressed in the Bank's comparative Management elaborated on the demonstration advantage in working directly with states, the effect in Nigeria and Russia. The Nigeria costs and benefits of working directly with states, country team representative described their 12 CHAIRPERSON'S COMMENTS efforts in reaching out to other regional depart- that there have been institutional efforts to ments to learn from their experience and about share experiences on provision of fee-based sharing of experience across states through the services. It was noted that Brazil is now consid- Governor's Forum in Nigeria. It was also noted ering fee-based services at the state level. Giovanni Majnoni, Chairman 13 View down the boulevard with high-rise construction in the background. Russian Federation. Photo by Yuri Kozyrev. courtesy of the World Bank Photo Library. Abbreviations CAS Country Assistance Strategy CODE Committee on Development Effectiveness CPS Country Partnership Strategy DPL Development policy lending/loan FRL Fiscal Responsibility Law IDA International Development Association lEG Independent Evaluation Group (World Bank Group) IFC International Finance Corporation SEEDS State Economic Empowerment and Development Strategy SWAp Sector wide approach All dollar($) amounts are in U.S. dollars. 15 Contents of the Complete Volume Abbreviations Acknowledgments Foreword Executive Summary Chairperson's Comments: Informal Subcommittee of the Committee on Development Effectiveness (CODE) 1. Introduction Background Scope and Objective of the Evaluation Structure 2. Which States? Evolution of the Bank Strategy Brazil India Nigeria Russia Conclusions 3. The Scope of Bank Engagement Fiscal Reform Fiscal Federalism Fiscal Capacity at the State Level Multisector Engagement at the State Level State-Level Strategies 4. Modalities of State-level Engagement Evolution of Instruments Multisector SWAps and State-Level DPLs Investment Lending at the State Level Capacity Building and AAA Implementation Arrangements and Staffing Partnership 5. Summary of Findings Appendixes A: Portfolio Performance B: Key Fiscal Indicators and Legal Framework at the State Level C: The Global Financial Crisis and State-Level Lending D: Countries at a Glance E: Key State Social and Economic Indicators F: Distribution of Projects by States 17 WORLD BANK ENGAGEMENT AT THE STATE LEVEL G: Fiscal Responsibility Law-Brazil H: Partnerships with Other Development Agencies I: People Met Endnotes Bibliography Boxes 3.1: Increasing Use of Earmarked Transfers in India and Russia 3.2: The Fiscal Adjustment Program in Brazil 3.3: Analytic Work on Fiscal Federalism in Russia 3.4: Improved Governance in the Fiscal Area in Orissa 3.5: Bank Program in Andhra Pradesh 4.1: Multisector SWAp 4.2: Minas Gerais-From Management Shock to Results-Oriented State 4.3: Bank Engagement in Bihar: An Example of Effective Partnership 4.4: Client Views Figure 1.1: Distribution of Projects by State Tables 1.1: Bank Lending to State/Provincial Governments (1998-2008) 2.1: Comparative Data on Population and Gross Domestic Product per Capita in States/Regions 2.2: Evolution of Bank Engagement in Brazil, Fiscal1998-2009 CASs 2.3: Evolution of Bank Engagement in India, Fiscal1998-2008 2.4: Evolution of Bank Engagement in Nigeria, Fiscal1998-2008 2.5: Evolution of Bank Engagement in Russia, Fiscal1998-2008 4.1: Average Preparation and Supervision Costs of State- and Federal-Level Projects, Fiscal 1998-2008 18 lEG PUBLICATIONS 2006 Annual Report on Operations Evaluation Annual Review of Development Effectiveness 2006: Getting Results Addressing the Challenges of Globalization: An Independent Evaluation of the World Bank's Approach to Global Programs Assessing World Bank Support for Trade, 1987-2004: An lEG Evaluation Books, Buildings, and Learning Outcomes: An Impact Evaluation of World Bank Support to Basic Education in Ghana Brazil: Forging a Strategic Partnership for Results-An OED Evaluation of World Bank Assistance Bridging Troubled Waters: Assessing the World Bank Water Resources Strategy Capacity Building in Africa: An OED Evaluation of World Bank Support China: An Evaluation of World Bank Assistance The CGIAR at 31: An Independent Meta-Evaluation of the Consultative Group on International Agricultural Research Committing to Results: Improving the Effectiveness of HIV/ AIDS Assistance-An OED Evaluation of the World Bank's Assistance for HIV/AIDS Control Country Assistance Evaluation Retrospective: OED Self-Evaluation Debt Relief for the Poorest: An Evaluation Update of the HIPC Initiative A Decade of Action in Transport: An Evaluation ofWorld Bank Assistance to the Transport Sector, 1995-2005 The Development Potential of Regional Programs: An Evaluation of World Bank Support of Multicountry Operations Development Results in Middle-Income Countries: An Evaluation of the World Bank's Support Economies in Transition: An OED Evaluation of World Bank Assistance Engaging with Fragile States: An lEG Review of World Bank Support to Low-Income Countries Under Stress The Effectiveness of World Bank Support for Community-Based and -Driven Development: An OED Evaluation Evaluating a Decade of World Bank Gender Policy: 1990--99 Evaluation of World Bank Assistance to Pacific Member Countries, 1992-2002 Extractive Industries and Sustainable Development: An Evaluation of World Bank Group Experience Financial Sector Assessment Program: lEG Review of the Joint World Bank and IMF Initiative From Schooling Access to Learning Outcomes: An Unfinished Agenda-An Evaluation of World Bank Support to Primary Education Hazards of Nature, Risks to Development: An lEG Evaluation of World Bank Assistance for Natural Disasters How to Build M&E Systems to Support Better Government lEG Review of World Bank Assistance for Financial Sector Reform Improving Investment Climates: An Evaluation of World Bank Group Assistance Improving the Lives of the Poor Through Investment in Cities Improving the World Bank's Development Assistance: What Does Evaluation Show? Maintaining Momentum to 2015? An Impact Evaluation of Interventions to Improve Maternal and Child Health and Nutrition Outcomes in Bangladesh New Renewable Energy: A Review of the World Bank's Assistance Pakistan: An Evaluation of the World Bank's Assistance Pension Reform and the Development of Pension Systems: An Evaluation of World Bank Assistance Poland Country Assistance Review: Partnership in a Transition Economy The Poverty Reduction Strategy Initiative: An Independent Evaluation of the World Bank's Support Through 2003 The Poverty Reduction Strategy Initiative: Findings from 10 Country Case Studies of World Bank and IMF Support Power for Development: A Review of the World Bank Group's Experience with Private Participation in the Electricity Sector Putting Social Development to Work for the Poor: An OED Review of World Bank Activities Small States: Making the Most of Development Assistance-A Synthesis of World Bank Findings Social Funds: Assessing Effectiveness Sourcebook for Evaluating Global and Regional Partnership Programs Water Management in Agriculture: Ten Years of World Bank Assistance, 1994-2004 World Bank Assistance to the Financial Sector: A Synthesis of lEG Evaluations The World Bank in Turkey: 1993-2004-An lEG Country Assistance Evaluation World Bank Lending for Lines of Credit: An lEG Evaluation All lEG evaluations are available, in whole or in part, in languages other than English. For our multilingual selection, please visit http:/ /www.worldbank.org/ieg ISBN 978-0-8213-8224-0 lEG WORLD BANK INDEPENDENT EVALUATION G~OUP IFC MIGA 9 SKU 18224