54435 The Cases of Brazil, India, Nigeria, and the Russian Federation IEG 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, 1987­2004: An IEG 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 of World Bank Assistance to the Transport Sector, 1995­2005 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 IEG 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: 1990­99 Evaluation of World Bank Assistance to Pacific Member Countries, 1992­2002 Extractive Industries and Sustainable Development: An Evaluation of World Bank Group Experience Financial Sector Assessment Program: IEG 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 IEG Evaluation of World Bank Assistance for Natural Disasters How to Build M&E Systems to Support Better Government IEG 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, 1994­2004 World Bank Assistance to the Financial Sector: A Synthesis of IEG Evaluations The World Bank in Turkey: 1993­2004--An IEG Country Assistance Evaluation World Bank Lending for Lines of Credit: An IEG Evaluation All IEG evaluations are available, in whole or in part, in languages other than English. For our multilingual selection, please visit http://www.worldbank.org/ieg World Bank Engagement at the State Level The Cases of Brazil, India, Nigeria, and the Russian Federation http://www.worldbank.org/ieg 2010 http://www.ifc.org/ieg The World Bank http://www.miga.org/ieg Washington, D.C. ©2010 The International Bank for Reconstruction and Development / the World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org E-mail: feedback@worldbank.org All rights reserved 1 2 3 4 13 12 11 10 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 does not support any general inferences beyond the scope of this evaluation, including any references about the World Bank Group's past, cur- rent, or prospective overall performance. The World Bank Group does not guarantee the accuracy of the data included in this work. The boundaries, colors, de- nominations, and other information shown on any map in this work do not imply any judgment 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 / 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: 978-750-8400; fax: 978-750-4470; Internet: www.copyright.com. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, , The World Bank, 1818 H Street NW Washington, DC 20433, USA; fax: 202-522-2422; e-mail: pubrights@worldbank.org. Cover: Illustration by Nip Rogers, ©Images.com/Corbis, reproduced by permission. ISBN-13: 978-0-8213-8224-0 eISBN: 978-0-8213-8225-7 DOI: 10.1596/978-0-8213-8224-0 Library of Congress Cataloging-in-Publication data have been applied for. World Bank InfoShop Independent Evaluation Group E-mail: pic@worldbank.org Communication, Learning, and Strategy (IEGCS) Telephone: 202-458-5454 e-mail: ieg@worldbank.org Facsimile: 202-522-1500 Telephone: 202-458-4497 Facsimile: 202-522-3125 Printed on Recycled Paper Contents v Abbreviations vii Acknowledgments ix Foreword xi Executive Summary xvii Chairperson's Comments: Informal Subcommittee of the Committee on Development Effectiveness (CODE) 1 1 Introduction 4 Scope and Objective of the Evaluation 5 Structure 7 2 Which States? Evolution of the Bank Strategy 11 Brazil 13 India 16 Nigeria 18 Russia 20 Conclusions 23 3 The Scope of Bank Engagement 25 Fiscal Reform 34 Multisector Engagement at the State Level 37 4 Modalities of State-Level Engagement 39 Evolution of Instruments 46 Partnership 47 5 Summary of Findings 49 Future Research Agenda 50 Findings Appendixes 53 A Portfolio Performance 55 B Key Fiscal Indicators and Legal Framework at the State Level 57 C The Global Financial Crisis and State-Level Lending 59 D Countries at a Glance 69 E Key State Social and Economic Indicators iii W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L 75 F Distribution of Projects by States 95 G Fiscal Responsibility Law--Brazil 97 H Partnerships with Other Development Agencies 101 I People Met 107 Endnotes 113 Bibliography Boxes 26 3.1 Increasing Use of Earmarked Transfers in India and the Russian Federation 27 3.2 The Fiscal Adjustment Program in Brazil 28 3.3 Analytic Work on Fiscal Federalism in the Russian Federation 33 3.4 Improved Governance in the Fiscal Area in Orissa 35 3.5 The Bank Program in Andhra Pradesh 40 4.1 Multisector SWAp 41 4.2 Minas Gerais--From Management Shock to Results-Oriented State 45 4.3 Bank Engagement in Bihar: An Example of Effective Partnership 50 5.1 Client Views Figure 11 2.1 Distribution of Projects by State Tables 4 1.1 Bank Lending to State/Provincial Governments (1998­2008) 10 2.1 Comparative Data on Population and Gross Domestic Product per Capita in States/Regions 12 2.2 Evolution of Bank Engagement in Brazil, Fiscal 1998­2008 15 2.3 Evolution of Bank Engagement in India, Fiscal 1998­2009 17 2.4 Evolution of Bank Engagement in Nigeria, Fiscal 1998­2008 19 2.5 Evolution of Bank Engagement in the Russian Federation, Fiscal 1998­2008 44 4.1 Average Preparation and Supervision Costs of State- and Federal-Level Projects, Fiscal 1998­2008 iv Abbreviations AAA Analytic and advisory activities ADB Asian Development Bank AfDB African Development Bank AusAID Australian Agency for International Development CAE Country Assistance Evaluation CAF Andean Development Corporation CAS Country Assistance Strategy CASPR Country Assistance Strategy Progress Report CIDA Canadian International Development Agency CMSIDP Ceará Multisector Social Inclusion Development Project CODE Committee on Development Effectiveness CPS Country Partnership Strategy DFID Department for International Development (U.K.) DPL Development policy lending/loan EBRD European Bank for Reconstruction and Development EC European Commission ESW Economic and sector work FFRFRP/L Fiscal Federalism and Regional Fiscal Reform Project/Loan FRL Fiscal Responsibility Law GDP Gross domestic product GSDP Gross state domestic product IBRD International Bank for Reconstruction and Development IDA International Development Association IDB Inter-American Development Bank IEG Independent Evaluation Group (World Bank Group) IFC International Finance Corporation ILO International Labor Organization IMF International Monetary Fund JBIC Japan Bank for International Cooperation MDGs Millennium Development Goals MGDPL Minas Gerais Development Policy Loan NGO Nongovernmental organization PAF Fiscal adjustment program PBL Policy-based lending PER Public Expenditure Review RFRF Regional Fiscal Reform Fund RFTAP Regional Fiscal Technical Assistance Project SAL State-level adjustment loan SEEDS State Economic Empowerment and Development Strategy SIDA Swedish International Development Agency SWAp Sectorwide approach UN United Nations USAID U.S. Agency for International Development WHO World Health Organization All dollar ($) amounts are in U.S. dollars. v View down central boulevard with high-rise construction in the background, Russian Federation. Photo by Yuri Kozyrev, courtesy of the World Bank Photo Library. Acknowledgments This report was prepared by the Independent Tapas Sen (India), Olufemi Taiwo (Nigeria), and Evaluation Group Country Evaluation and Regional Galina Kurlyandskaya (the Russian Federation). Relations evaluation team: Konstantin Atanesyan (Task Team Leader, Senior Evaluation Officer), Ravi Kanbur (Cornell University), Steven Webb Basil Kavalsky (consultant), and Sarwat Jahan (con- (Adviser, LCSPS), and Thomas O'Brien (Manager, sultant). It is based on the results of field visits to SARDE) were the peer reviewers. Sarwat Jahan and all four countries and interviews with key local Tim De Vaan provided data and research assis- (federal and state) counterparts and Bank teams. tance. Cecilia Tan and Corky de Asis provided administrative support and William Hurlbut pro- The study draws extensively on four country back- vided editorial support. Caroline McEuen edited ground papers prepared by Joao Oliveira (Brazil), the report for publication. Director, Independent Evaluation Group-World Bank: Cheryl Gray Senior Manager, IEG, Country Evaluation and Regional Relations: Ali Khadr Task Manager: Konstantin Atanesyan Note: Vinod Thomas, Director-General, Evaluation, The World Bank Group, was recused from management oversight of the evaluation owing to his prior association with the Brazilian program as country director. vii 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 (1998­2008) of World Bank engagement at the the mix. The instruments deployed by the Bank state level in four selected large federal coun- evolved over the review period and included tries: Brazil, India, Nigeria, and the Russian Fed- state-level development policy loans, multisector eration. This pilot cross-country study combines results-based investment lending, and reim- elements of a country assistance evaluation and bursable technical assistance. There was consid- a thematic review, looking at the evolution of erable successful innovation in the development four country strategies as well as the Bank's state- of the instruments used, yet little knowledge level operations. The evaluation limited its re- sharing among countries. view to selected cases of lending and analytic work where the state governments were the prin- The report identifies lessons and good practice ex- cipal partners of the Bank and the primary party amples that warrant further examination and responsible for 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 modal- has been highly appreciated, but to enhance the ities of engagement. Two ideas--often at odds with poverty impact of state-level interventions, greater one another--featured in most approaches to weight should be given to the needs of the poor- selection of states for direct engagement. One was est states by balancing states' propensity to reform to support better-performing, reformist states and the concentration of poverty within them. Ex- (the lead or focus states approach), while the perience shows that it has been possible to achieve other was to support the poorest states as a more results in some of the poorer, low-capacity states direct route to reducing poverty. through persistent work with committed state counterparts and partnerships with other donors. The initial area of engagement was typically fis- Second, continued focus on public finance man- cal reform, where the Bank generally helped to agement appears sound, irrespective of whether enhance the capacity of state governments for engagement is confined to this area or serves as public financial management. In some states Bank an entry point for broader engagement. Third, involvement extended to multisector engage- there is considerable scope for greater impact ment that usually involved a mix of analytic work, from knowledge transfer and expanded knowl- development policy lending, and investment edge services. Cheryl Gray Director, IEG-World Bank ix Gathering for a meeting, 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 the Russian Federation. This pilot cross-country study reviews selected cases of World Bank lending and analytic work at the state level in those four large, federated countries. In each case, state governments were the Bank's principal devel- opment partners. The study looks at the evolution of the four country strate- gies and the Bank's mode of engagement at the state level to glean lessons from that experience 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. its approach to selecting states in country strat- But the poverty impact of those interventions egy documents. Two tendencies--often at odds-- could be enhanced by balancing the propen- featured in most approaches. One was to support sity of states to reform and the concentration better-performing, reformist states (the lead or of poverty within them, giving greater weight focus states approach). The other was to support to the needs of the poorest states. the poorest states as a more direct route to re- · Second, continued focus on public finance ducing 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 serves as an entry point for and principal area of engagement was typically fis- broader engagement. cal reform--fiscal sustainability, medium-term · Third, there is considerable scope to gain fiscal frameworks, strengthening the public fi- greater impact from analytic work, knowledge nancial management capacity of state govern- transfer, and expanded knowledge sharing-- ments, and fiscal federalism. In some states, Bank not so much in concepts and theories as in the involvement extended beyond fiscal reform to practical experience of what works and what multisector engagement focused on the growth does not. and poverty-reduction agenda. The modalities of engagement and the instruments deployed by Evolution of Bank Strategy the Bank evolved over the review period and in- cluded state-level development policy loans, multi- Why State-Level Engagement? sector results-based investment lending, and Over the past decade, the World Bank signifi- reimbursable technical assistance. There was con- cantly expanded its engagement at the state level siderable successful innovation in this area, yet lit- in four large federations: Brazil, India, Nigeria, tle knowledge sharing among countries. and Russia, mainly through lending, but also xi W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L through policy dialogue, technical advice, and The increasing focus of the Bank on poverty re- analytic work. Both demand and supply factors duction after 1995 was also an important factor. contributed to this expansion. There is a distance between federal-level pro- grams and results on the ground in such large On the demand side, federal governments fo- countries. The majority of the public expendi- cused on fiscal stabilization following the financial ture categories most closely associated with crises of the late 1990s. They saw the potential for poverty reduction in the short and long terms the Bank to provide state governments with in- are usually state responsibilities in these coun- centives for reform through financing, while tries. Therefore, increasing the Bank's impact on encouraging discipline through agreed policy poverty reduction meant increasing the focus on measures and provision of technical support for activities at the state level. In addition, many Bank implementation. Federal governments had limited country and sector staff found work at the state scope to differentiate among states based on fac- level in these countries more rewarding, given the tors such as commitment to reform. They saw in clients' greater interest in the Bank's financial the Bank's capacity to do this a means of assist- and knowledge resources. ing states willing to take the lead, with the possi- bility of a demonstration effect for other states. Which States? The shift to the state level presented the Bank with State governments were also eager to borrow a number of operational issues. Among them was from the Bank because its loans, denominated in which states the Bank should engage. The four foreign currency, generally came at lower rates countries have large numbers of states--26 states than those provided by the federal government and a Federal District in Brazil, 28 states and 7 or the domestic market. Bank loans, while often union territories in India, 36 states and a Federal financially modest at the federal level, could be a Capital Territory in Nigeria, and 83 Regions ("sub- major source of financing at the state level. State jects of the federation"--republics, oblasts, krays, governments welcomed the Bank's focus on their and okrugs) in Russia. Working in all of them economy as well as the associated dialogue, ad- would obviously be beyond the Bank's budgetary vice, and analytic work. In addition, with the in- and human resource capacities. creasing concern about meeting or achieving outcomes with regard to the Millennium Devel- The Bank defined the strategic approach to the se- opment Goals, both federal and state govern- lection of states in its Country Assistance Strategies ments saw the Bank as having a comparative (CASs). Some attempts were made to develop advantage in supporting better service delivery in quantified criteria for selective engagement, but the relevant social and economic sectors at the the Bank generally preferred to keep the criteria state level. broad to allow for flexibility. It is clear from all the countries reviewed that there was tension between On the supply side, with the combination of fis- the Bank's interest in identifying progressive, re- cal stabilization and improvement in the fiscal forming states that could serve as demonstrations situation of the four countries during the com- to others and its interest in supporting poverty re- modity boom of 2000­07, there was limited ap- duction by assisting the poorest states. In addition petite to borrow from the Bank for federal to these two key criteria--effectiveness of assis- programs (this trend was more pronounced in tance and need (poverty)--another equally im- Brazil and Russia; Nigeria, an International De- portant criterion was the political economy, unique velopment Association [IDA] borrower, was an ex- in each country and including (but not limited ception; in India, federal-level borrowing increased to) relations between the federal center and the slightly in 2004­07). A level of Bank engagement states; capacity and political affiliation of the state- commensurate with the size and importance of level leadership; level of trust and the relation- these countries almost mandated the shift to ship of the Bank teams with the clients; and local the state level, where demand remained buoyant. politics and electoral cycles. xii E X E C U T I V E S U M M A RY In Brazil, the shift toward states was proposed in overall program, the share of lagging states in a mid-1990s CAS, directing lending to creditwor- the program actually declined. thy reforming states. The next CAS, prepared at the turn of the millennium, continued this ap- In Nigeria, the Bank reactivated and intensified proach, using criteria established by the federal its lending activities after return to civilian rule in Fiscal Responsibility Law (FRL) and stressing the 1999. During that period, Bank strategy passed intention of providing assistance to the states of through two phases: a period of interim strategies the Northeast region with highest poverty levels. (fiscal 2000­05) following re-engagement; and Lending to states became more multisectoral, the fiscal 2005­09 CPS (AfDB, DFID, USAID, and and significant innovations were introduced, such World Bank Group 2009), when the Bank adopted as state-level sectorwide lending (multisector a medium-term focus. Engagement at the state SWAp) and state-level development policy loans level in Nigeria was largely driven by the social and (DPLs). Both instruments were applied in states poverty reduction agendas, with focus on im- that had turned the corner fiscally, despite sig- proving infrastructure and providing support for nificant disparity in their income levels. The 2008 agricultural and rural development. During the Country Partnership Strategy (CPS) (World Bank CPS period of fiscal 2005­09, the Bank's strategy Group 2008a) focused on a technical assistance formally moved to focus on well-performing states program of modest size with the federal govern- (lead states), seeking to leverage state efforts and ment and a major demand-driven lending program resources by granting them access to a perfor- with states, conditioned on perceived commit- mance package. Five states were selected based ment, ownership of reforms, and fiscal responsi- on the government-led State Economic Empow- bility. The Bank engaged with some of the more erment and Development Strategy (SEEDS) prosperous and reformist states. Although ini- benchmarking process. tially the Bank attempted to expand investment lending, this proved cumbersome, given the two- In Russia, the mid-1990s CAS emphasized re- tier approvals required by the state and federal gional investment projects (despite recognition governments. As a result of joint consultations with that they were expensive to prepare and super- state and federal counterparts, by 2008 the com- vise). The next CAS (World Bank Group 1999) position of state-level lending shifted toward outlined a phased shift in lending, away from in- cross-sectoral operations in support of economic vestment projects in infrastructure and energy policies and public sector reforms (DPLs and mul- in favor of increased emphasis on systemic as- tisector SWAps). pects of institutional development. A subsequent strategy in the early 2000s continued the strat- In India, the Bank shifted its focus to the state egy shift, emphasizing support for reforms at level in the mid-1990s. At the time, states were fac- the regional level, particularly to strengthen pub- ing financial problems, and both the federal and lic sector management. The 2005 CAS Progress state authorities were keen to tap into the Bank's Report (World Bank Group 2005b) stated that resources and take advantage of technical assis- work at the regional level was to be carried out tance. The Bank opted for major involvement in in a multisectoral manner and would concen- progressive reforming states (the focus states ap- trate on a small number of regions in agreement proach). The 2004 CAS (World Bank Group 2004a) with the federal government. The 2007 CPS signaled a change of strategy, noting that the (World Bank Group 2006) added a finishing focus on reforming states was leading to neglect touch: the plan for a gradual shift to the new of the lagging states. Therefore, the CAS pro- modalities of cooperation and instruments, such posed the provision of technical assistance to the as the subnational facility that allows the Bank lagging states and an effort to shift lending to Group to provide funds without a sovereign them as well. This proved to be difficult. As the guarantee to states and municipalities and pro- 2009 CAS completion report noted, while lend- vision of technical assistance on a reimbursable ing at the state level remained a large share of the basis (fee for service). xiii W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L The Scope of Engagement states to undertake high-priority programs. The The second issue for the Bank was the scope of government of India has used such schemes a its engagement. In Brazil, India, and Nigeria, the great deal. For the Bank, the political sensitivities Bank had carried out numerous state-specific make fiscal federalism a difficult area for inter- projects as part of its support for nationwide sec- vention unless there is an explicit request from the toral programs. The decision about which states federal government. The Bank has undertaken to engage--for example, in support for agriculture substantial analytic work in this area. Russia, where and rural development in India or education in new fiscal relations were being defined in the Brazil--was sometimes a matter of strategic past decade, is a very good example. choice, but more often a matter of historical en- gagement or the availability of analytic work, A second and far larger part of Bank engagement based in turn on opportunistic involvement of the in fiscal reform is its direct support to public fi- Bank in particular states. In the new context, the nance management at the state level, including Bank was steering toward a new model of en- enhancements in tax capacity, modernizing the tax gagement. This tended to have two elements: structure, developing a sustainable fiscal policy and support for fiscal reform and broader multisec- medium-term expenditure framework, and im- tor engagement at the state level. proving budget and expenditure management. The Bank's engagement model generally started Fiscal Reform with a trigger mechanism that required states to The Bank expanded its involvement in fiscal re- show commitment to fiscal reform. form in selected states. With the focus on stabi- lization and the need to reduce growing state This requirement was highly formalized in Brazil, deficits or enhance state public expenditure man- where the federal government requires strict ad- agement capacity, the Bank engaged in two areas. herence to the FRL. In India and Nigeria the re- quirements were less formalized, but generally The first of these was fiscal federalism. Fiscal re- related to timely budgeting and reporting. In Rus- lations between the federal and state governments sia, criteria were established for participation in are politically highly sensitive, since this is often the fiscal reform projects supported by the Bank. at the core of balancing regional interests. From Once the triggers were met, the Bank was able to a strictly economic perspective, the resources further support fiscal reform through an en- available to the state governments need to be bal- gagement model that combined analytic work anced with their expenditure responsibilities. For with multisector lending and focused technical many taxes it is more efficient to centralize col- assistance in the areas of fiscal and governance lections. This disconnect generally creates an im- reform. balance between the revenues collected by the states and their development mandate. In Brazil Elements of this model are present in each coun- this imbalance is relatively modest, but in Nigeria try. In Lagos, Nigeria, for example, intensive ana- it is very large. This means that the federal gov- lytic work at the state level was combined with ernment is required to transfer resources to the investment lending and technical assistance, but state, generally based on various formulas that the Bank is only now considering the possibility take account of population, per capita income, and of using multisector lending in support of its ap- the state government's own tax effort. In the coun- proach. In Andhra Pradesh and Orissa in India, the tries reviewed, federal transfers have not been Bank carried out analytic work and subsequently very effective in reducing disparities in expendi- used multisector loans to support fiscal reform, ture capacity among states. but technical advice, except that embedded in the analytic work and lending activities, has not been A second key issue in fiscal federalism concerns a focus of the approach. The Bank's involvement discretionary transfers from the center to the with tax policy has been relatively light; the focus states, usually intended to provide an incentive to of its efforts has been on budgeting and public xiv E X E C U T I V E S U M M A RY expenditure management systems, and increas- proach at the state level. The multisector re- ingly on governance issues associated with trans- structuring loan in Andhra Pradesh, an invest- parency and accountability for efficient service ment loan undertaken in 1998, was a way to delivery. accomplish this, but it was an enormously costly operation to prepare and supervise. Multisector Engagement In addition to fiscal engagement, the Bank has also Adjustment lending rapidly became the instru- undertaken a broader state-level multisector ment of choice to support fiscal reform and engagement, which has focused on the growth statewide strategies in Brazil and India, but in and poverty reduction agenda. This has involved Brazil there was a sense that it was less effective a mix of analytic work, adjustment lending, and in reaching out to the line ministries in key sec- investment lending. The Bank produced an ex- tors. The Bank's Brazil country team developed plicit strategy for its activities in a particular state the innovative approach of a multisector SWAp, in only a few cases. Such strategies were usually a results-based instrument with target indicators covered in broad-brush fashion in the CAS. defined for each sector and disbursements asso- ciated with achievement of the targets. This had At their most developed, as in Andhra Pradesh in very positive outcomes: it brought to the fore India and in Ceará in Brazil, the Bank programs the linkages required to achieve results, such as combined investment lending in most of the core the need for improved water supply in order to economic and social sectors with multisector reduce infant mortality. Another important inno- lending. The objective was to derive synergies vation was the pioneering of reimbursable tech- from the combination of activities. In Andhra nical assistance at the state level in Russia. Bank Pradesh, for example, the Bank was explicit in its budgets rarely allow the level of analytic work view that the difficult measures required for fis- demanded by intensive engagement in three-to- cal reform needed to be matched by increased in- five states, and an approach that permits states to vestment in agriculture, rural development, health, pay for additional work has considerable prom- and education to provide a politically acceptable ise for other middle-income countries. package of reforms. Findings The selection of states for this broader engage- Overall, the analysis leaves little doubt that the ment focused more on fiscal reform than on the Bank's engagement at the state level did add broader poverty issue. The Bank spent a great deal value. There was a great deal of enthusiasm at both of time in these countries supporting relatively the federal and state levels in these countries re- high-income, high-capacity states (such as Minas garding the Bank's contribution and a large num- Gerais in Brazil, St. Petersburg in Russia, and Kar- ber of specific achievements. Although state-level nataka in India). While this support added value, engagement often requires additional effort and it came at the expense of Bank efforts in poorer can be resource-intensive, it is usually worth the states that lack capacity. cost. Modalities of Engagement The main findings of this review, which may be How to engage was also a concern for the Bank helpful in guiding the organization of future work during the period, and there was considerable at the state level, include: evolution in the approach. The first bridge to be crossed was the use of adjustment (development On selection of states: policy) lending at the state level (the first such ad- justment loan was made in March 2000 to the · The strategy to be selective and concentrate state of Uttar Pradesh in India). Until that point lending services on a few states to enhance the Bank had struggled to find an instrument the impact of the Bank's program is right in to attach to its policy dialogue and strategic ap- principle, but selection criteria and the mode xv W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L of implementation could give greater weight to pears sound, whether engagement is confined the 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- · The lending programs and Bank budgets in state capacity and encouraging state-to-state some states are often larger than for many knowledge transfer, albeit mainly between high Bank clients. For states where the Bank plans performers. However, there is little evidence or has a major engagement, a brief state strat- that it had the desired demonstration effect egy document could be a useful tool for defin- on poor, lagging states on the scale the Bank ing the scope of engagement and developing hoped for, or that the Bank had an exit strat- a medium-term outlook. egy to permit increased focus on poorer states over time. At the same time, experience On modalities of engagement: shows that it is possible to achieve results in the poorest states through persistent work with · There is considerable scope for greater impact committed state counterparts and strong part- from knowledge transfer and expanded knowl- nership with the federal government and other edge services. In particular, there is strong de- donors. mand for better knowledge sharing, both within · It is important to stay engaged not only in the Bank and across the countries concerned. states that are able to borrow from the Bank, This is not so much a matter of sharing of con- but also in states that have no fiscal space to bor- cepts and theories as it is of communicating row but demonstrate a genuine commitment practical experience regarding what is working to development that can be supported through and what is not. analytic work and technical assistance. · Widening the scope and increasing the amount of analytical work at the state level could be help- On the scope of engagement: ful in identifying high-impact, high-priority areas. There seems to be a potential for closer · Continued focus on public finance manage- partnership between state governments and ment as the core area for state-level work ap- the Bank in this area. xvi Chairperson's Comments: Informal Subcommittee of the Committee on Development Effectiveness (CODE) n September 21, 2009, the Informal Subcommittee of the Commit- O tee on Development Effectiveness (CODE) considered an Inde- pendent Evaluation Group (IEG) report entitled World Bank Engagement at the State Level: The Cases of Brazil, India, Nigeria, and the Russian Federation. Staff representatives from the four countries considered in the report were present at the meeting. Summary The comparative analyses of the experiences in Members commended IEG for the well-written the four countries covered by the report and and informative report and welcomed manage- staff elaboration of country experiences were ment's broad agreement with its findings. The appreciated. Subcommittee held a rich discussion, where members noted the importance of this report Recommendations and Next Steps as a basis for further considering the direction The Subcommittee recommended the following and potential of state-level engagement by the to management: Bank. Some members remarked on the con- siderable innovation and creativity in this area, · Consider this IEG report as a basis for further which was considered a critical aspect of a coun- thinking toward a more comprehensive frame- try's development. Going forward, the need to work to guide the Bank's engagement at the bear in mind the poverty focus in engaging at state level, and to continue adjusting its in- the state level and to strengthen knowledge struments to meet the needs of the states. sharing within and across countries was noted. · Maintain a flexible approach to selection of Members' interventions focused on IEG's find- states, to accommodate different country con- ings related to the need for the Bank to have a texts while keeping in mind the poverty focus. strategic approach to state-level support, selec- · Consider ways to strengthen systematic knowl- tion of states, modalities of engagement, and edge sharing and learning from ongoing work knowledge transfer, which are elaborated below. at the state level. xvii W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L Main Issues Discussed Federal Government and States. Some ques- Framework for Bank Engagement. Some tions were raised about ownership and involve- members noted the potential benefit of IEG's ment of the federal government with regard to suggestion to prepare brief state-level strategy Bank assistance to states and any tensions en- documents for states with significant long-term countered between the federal and state levels. Bank engagement. Management clarified that In addition, interest was expressed in the Bank's the strategic analysis and dialogue with client comparative advantage in working directly with countries on the Bank's engagement at the state states, the costs and benefits of working directly level takes place as part of the CAS process. It also with states, and approaches to addressing for- expressed concerns about adding another layer eign exchange risks at state levels. IEG and man- of strategy paper, but noted that integrating sep- agement stressed that Bank engagement at the arate strategies for the most important state state level is predicated on federal government clients within a CAS could be a useful way to ap- support. Referring to the example of Brazil, man- proach this issue. agement also noted that the federal government has encouraged Bank involvement at the state Selection of States. Members supported selec- level as a way to incentivize states to carry out tivity 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 De- Bank should work with more progressive, re- partment has offered assistance to Indian states formist states or with poorer states in light of interested in addressing exchange rate risks. IEG's findings, taking into consideration the Bank's mandate to reduce poverty and the po- Scope and Modality of Support. Some mem- tential added value of focusing on lagging states. bers highlighted the potential for Bank support There was general consensus to preserve flexibility for fiscal federalism and governance and capacity to accommodate different country contexts and building at the state level. A member sought more political economies and to enable the Bank to clarity on the outcomes of fiscal reforms and an- work with active state actors at the state and sub- other encouraged conducting Public Expendi- state levels, while keeping in mind the poverty ture Reviews at the state level. Others emphasized focus. the importance of analytical work at the state level, including in lagging states. A few members The importance of transparency in selecting states suggested that the IFC model of state-level sup- was emphasized. IEG noted the tension between port be reviewed. Management described some the two approaches and reiterated that to en- of the ongoing analytical work at the state level hance the poverty impact of state-level inter- such as the Doing Business reports in Russia and vention, greater weight to the needs of the poorest Nigeria, Public Expenditure Reviews in Nigeria, states was merited since there was no clear evi- and Social Expenditure Reviews in Russia. dence of the desired demonstration effect. IEG added that it is possible to achieve results through Demonstration Effect and Knowledge Shar- persistent work, even in the poorest and low- ing. A few members sought more clarity regard- capacity states. Management commented on the ing the limited demonstration effect on lagging country realities that have driven the selection states, noting that the experience in Russia seems of states, as in the case of India, where the Bank to have been more positive. Speakers stressed is making an effort to work more with lagging the importance of strengthening knowledge shar- states. Reference was also made to the CPS in ing across states and countries. For example, the Nigeria that describes the principles of engage- potential for other countries to learn from Russia's ment at the state level; a key principle is the level experience with fee-based services was noted. A of human development indicators. few members urged that the Bank should give xviii INFORMAL SUBCOMMITTEE OF THE COMMITTEE ON DEVELOPMENT EFFECTIVENESS (CODE) the highest priority to demand for services where regional departments to learn from their expe- the recipient was willing to pay a fee, because this rience and about sharing of experience across indicated the seriousness and the will to carry states through the Governor's Forum in Nigeria. out reform and change. The possibility of a Global It was also noted that there have been institutional Expert Team to improve knowledge sharing among efforts to share experiences on provision of fee- staff was suggested. Management elaborated on based services. It was noted that Brazil is now con- the demonstration effect in Nigeria and Russia. sidering fee-based services at the state level. The Nigeria country team representative de- scribed their efforts in reaching out to other Giovanni Majnoni, Chairman xix Chapter 1 Evaluation Highlights · The Bank began to think of the state as a strategic unit in the mid-1990s. · State effectiveness in planning, budgeting, and implementing pro- grams contributes to development effectiveness. · The four country programs exam- ined--in Brazil, India, Nigeria, and Russia--had a similar aim of engag- ing at the state level. · In each case, state governments were the Bank's principal develop- ment partners. Town gathering, Nigeria. Photo by Curt Carnemark, courtesy of the World Bank Photo Library. Introduction T he World Bank has traditionally focused its lending, analytic work, and policy dialogue on federal governments.1 This is a requirement of the Bank's Articles of Agreement, which specify that the Bank lend either to a federal government or with a sovereign guarantee. It has also been a prac- tical matter. National projects with a single implementing min- ingly to engagement with states as the The Bank's engagement istry or agency are less costly to prepare and su- logical evolution of its programs in sup- with states in federations pervise than multiple subnational projects. This port of poverty alleviation in large fed- began to change in the did not prevent the Bank from preparing and im- eral countries. Despite some initial mid-1990s. plementing projects at the subnational level, but, hesitation, federal governments soon in general, subnational units were seen as ad- warmed to the approach. They saw a number of ministrative agents, with responsibility for their de- potential benefits from the Bank's involvement, velopment residing with the federal government. including the Bank's capacity to differentiate be- In large countries where the bulk of economic and tween states, to reward reforming states by trans- social expenditures that affect the living stan- ferring additional resources, or equalize funding dards of the poor are under the control of state to help states with weaker social service coverage governments2 or other decentralized units, Bank or quality to improve them--something that can efficiency in fulfilling its poverty reduction man- be politically difficult for some federal govern- date was impeded. ments. In addition, the Bank can both create in- centives for increased state-level expenditure For much of its history, the Bank has provided in- through lending in specific areas and provide vestment loans for projects located within the knowledge services to add to effectiveness in ad- territory of particular states in federations. State dressing the MDGs. governments have often been signatories to proj- ect agreements in the irrigation, roads, water sup- In some of the larger federal countries State governments have ply, agriculture, and education sectors and have where the Bank is active, state gov- gradually emerged as shared responsibility for the implementation of ernments have emerged as important important players in these projects. However, until the mid-1990s, the development actors, with major re- development. Bank did not think about the state as a strategic sponsibility for infrastructure and social unit. Nor did the Bank consider its package of state service provision. Table 1.1 shows the size of Bank lending a vehicle for reform and poverty reduc- lending at the state (or provincial) level in some tion or as elements in a coherent state-level de- of its biggest borrowing countries. The effective- velopment program. ness of state governments in planning, budgeting, and implementing programs has therefore In the late 1990s, with a renewed focus on poverty emerged as an important determinant of con- and the adoption of the Millennium Develop- straint in overall development effectiveness. Con- ment Goals (MDGs), the Bank turned increas- sequently, the Bank's programs in countries such 3 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L Table 1.1: Bank Lending to State/Provincial Governments (1998­2008) Number of loans Amount (US$ billions) Percentage of Percentage of Countries/operations State Total state lending State Total state lending 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 Nigeria 19 32 59 1.8 3.1 58 Sources: World Bank data, ImageBank. as Brazil, India, Nigeria, and Russia have increas- tries (see table 1.1), given time and resource lim- ingly been geared toward providing support for itations, this evaluation focused on selected cases the state governments. in four countries that are major clients of the Bank: Brazil, India, Nigeria, and Russia.3 These four The Bank has engaged with state governments in countries were selected taking into account ge- two broad areas. First, it sought to improve fiscal ographic representation, regional importance, outcomes and governance at the state size of state-level programs, and use of innovative The Bank has engaged level. Second, it tried to increase the instruments and approaches by the country teams. them on fiscal outcomes focus on poverty reduction by steering The study covers the period from 1998 to 2008 and and governance and resources and advisory services to combines elements of a Country Assistance Eval- sought to increase its states where poverty is greatest and uation (CAE), looking at four country programs focus on poverty those most critical for achievement of with a similar aim of engaging at the state level in reduction. the MDGs. a large federal system, and a thematic review based on a limited number of case studies. The The increase in the Bank's state-level engage- study looked at a number of projects in two states ment has not been a coordinated effort. State-level in each country. Therefore, it neither claims uni- activities have been, and continue to be, driven versal knowledge of all specifics and outcomes of by individual country programs. During the pe- Bank lending and nonlending activities in all large riod of state-level engagement the Bank did not federal countries, nor does it suggest universal examine its approach from a cross-country per- recipes for other large federal countries not in- spective. Each country team has had to find its cluded in this review. Instead, it looked at the own way and develop programs from first prin- evolution of four country strategies and the Bank's ciples, often "reinventing the wheel." By now, mode of engagement in order to draw lessons however, there is a significant body of experi- from that experience. ence in dealing with state governments and some emerging good practice that warrants further The evaluation limited its review to state-level examination and wider dissemination. This was lending and analytic work where the state gov- the motivation for this Independent Evaluation ernments were the Bank's principal partners, and Group (IEG) cross-country review. therefore a primary party responsible for devel- opment outcomes. These include loans support- Scope and Objective of the Evaluation ing improved fiscal management and governance Although the Bank has had substantial programs in particular states and loans that were strategic of state and province-level lending in many coun- at the state level--that is, loans designed to sup- 4 INTRODUCTION port a state-level economic development pro- Chapter 3 reviews the scope of the This evaluation looks at gram or a state-specific sectoral or thematic issue Bank's engagement at the state level, in- the evolution of state (including state-level development policy lend- cluding support for fiscal reform and ad- engagement in four ing and multisector sectorwide approaches). Sev- dressing poverty reduction and the country strategies from eral federal-level loans that supported improved growth agenda through multisector en- 1998 to 2008. systems of fiscal federalism were also included. gagement and policy dialogue. Chapter 4 covers the modes of Bank engagement at the state level, including selection of lending instru- Structure ments, analytic and advisory activities (AAA) and Chapter 2 summarizes the evolution of thinking capacity building, and the process im- in adapting the Bank's policies to encompass lend- plications of working at the state level. In these cases, the states ing and other work at the state level. It also cov- Chapter 5 summarizes the findings from were the Bank's principal ers how the Bank dealt with selecting states with the four country case studies. development partners. which to engage in different country contexts. 5 Chapter 2 · After the financial crises of the late 1990s, states gained the fiscal space to allow them to seek financing for their investment programs. · Engagement in each country was conditioned by demand, capacity, and ownership of reforms. · In most cases the decision to en- gage at the state level was reason- able and based on demand and state capacity. · The most difficult issue has proven to be selectivity--which states to choose--with tension between en- gaging those with capacity and will- ingness and those that are the poorest. Women harvesting cotton, India. Photo by Ray Witlin, courtesy of the World Bank Photo Library. Which States? Evolution of the Bank Strategy T he 1998­2008 period saw a major expansion of Bank engagement at the state level in four large federations: Brazil, India, Nigeria, and Rus- sia, mainly through lending but also through policy dialogue, techni- cal advice, and analytic work. Both demand and supply factors seem to have fed this expansion. On the demand side, federal governments fo- On the supply side, with the com- In the late 1990s, cused on fiscal stabilization following the financial bination of fiscal stabilization and federal governments saw crises of the late 1990s. They saw the potential for improvement in the fiscal situation potential for the Bank to the Bank to provide state governments with in- of the four countries during the support state-level centives for reform through financing, while en- commodity boom of 2000­07, there reforms, and states forcing discipline through conditionality and was a limited appetite to borrow themselves were eager to providing technical support for implementation. from the Bank for federal programs borrow from the Bank. Federal governments have limited scope for dif- (this trend was more pronounced in ferentiating between states on the basis of non- Brazil and Russia; Nigeria, an International De- tangible factors such as commitment to reform velopment Association [IDA] borrower, was an ex- and saw the Bank's capacity to do this as a means ception; in India, federal-level borrowing increased of assisting states willing to take the lead, with the slightly in 2004­07). A level of Bank engagement possibility of a demonstration effect for other commensurate with the size and importance of states. these countries almost mandated the shift to the state level, where demand remained buoyant. As a rule, state governments were also eager to The Bank's increasing focus on poverty reduction borrow from the Bank. Bank loans, denominated after 1995 was also an important factor. There is in foreign currency, typically came at lower rates a great distance between federal-level programs than those provided by the federal government and results on the ground in such large countries. or the domestic market.1 Bank loans, while often Most of the public expenditure cat- The Bank saw in the financially insignificant at the federal level, could egories most closely associated with states an opportunity be a major source of financing at the state level. poverty reduction in the short and to increase its impact on Finally, state governments welcomed the Bank's long term are usually state respon- poverty. focus on their economy and the associated dia- sibilities in these countries. Hence, logue, advice, and analytic work. In addition to in- increasing the Bank's impact on poverty reduc- creasing concern about meeting MDGs, or at least tion meant increasing the focus on and the ac- achieving respectable outcomes, both federal and tivities at the state level. A final point is that Bank state governments saw the Bank as having a com- country and sector staff found work at the state parative advantage in supporting better service de- level in these countries more rewarding given livery in the relevant social and economic sectors the generally greater interest of the client in the at the state level. Bank's financial and knowledge resources. 9 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L Table 2.1: Comparative Data on Population and Gross Domestic Product (GDP) per Capita in States/Regions Average GDP Average GDP Average per capita of per capita of Population Number of population GDP the three the three (millions, states/ per state per capita richest states poorest states Country 2007) regions (millions) (PPP, 2007) (2005/06) (2005/06) Brazila 191.6 26 states 6.1 $9,034 $10,464 $1,829 1 Federal District Indiab 1,123.3 28 states 30.2 $2,598 $1,376 $229 7 union territories Nigeriac 147.9 36 states 3.8 $1,867 $4,021 $131 1 Federal Capital Territory Russian Federationd 141.6 83 regions 1.6 $14,743 $27,107 $1,078 Sources: Population and GDP per capita (purchasing power parity [PPP], 2007) data are from World Bank data. GDP per capita of states/regions are from: Brazil, Instituto Brazileiro de Estatistica (IBGE), Regional Accounts 2005; India, Central Statistical Organization of the government of India and the State Statistical Bureaus; Nigeria, Economic Associates (economic consulting firm; the Nigeria National Bureau of Statistics does not yet publish state GDP data); and Russia, Federal State Statistics Service. Data for 2005/06 are used because that is the most recent year available for all four countries. Note: GDP = gross domestic product; PPP = purchasing power parity. a. Three richest states in Brazil are Săo Paulo, Rio de Janeiro, and Federal District. Săo Paulo's GDP would rank 22nd largest in the world. The poorest (by per capita GDP) are Alagoas, Maranhăo, and Piauí. b. Three richest states (union territories) in India are Goa, Chandigarh, and Ponicherry. The poorest are Bihar, Uttar Pradesh, and Orissa. c. Three richest states in Nigeria are Bayelsa, Cross Rivers, and Delta. The poorest are Adamawa, Taraba, and Jigawa. d. The three richest regions in Russia are Khanty-Mansi Okrug, Yamal-Nenets Okrug, and Sakhalin Oblast--all remote and sparsely populated oil- and gas- producing regions. The poorest are Ingush Republic, the Republic of Chechnya, and the Republic of Kalmykia. The problem for the Bank The shift to the state level presented sisting the poorest states. The motivation and tim- was the matter of which the Bank with a number of operational ing of the Bank's engagement at the state level states to engage. issues. First among these was which differed by country. states the Bank should engage. Most of these countries have large numbers of states-- In terms of distribution of state-level lending 26 states and a Federal District in Brazil, 28 states within particular countries, the Bank generally and 7 union territories in India, 36 states and a tended to concentrate its engagement on a select Federal Capital Territory in Nigeria, and 83 sub- number of states, instead of spreading its efforts jects of the federation (republics, oblasts, krays, thinly across the entire country. In Brazil (figure and okrugs) in Russia (table 2.1). The Bank's 2.1),2 approximately 51 percent of the state-level budgetary and human resource capacity gener- lending (in number of projects) was allocated to ally allowed engagement in perhaps three-to-five five states. In India (figure 2.1), approximately states at any given time. Selection was thus strate- 62 percent of the state-level lending (in number gic and was usually defined in the Bank's coun- of projects) was allocated to five states. In Nige- try strategy documents. Some of these documents ria, four states--Bauchi, Cross River, Kaduna, and attempted to develop quantified criteria for se- Lagos--implemented projects that were specifi- lective engagement, but generally the Bank pre- cally tailored for individual states. The other states ferred to keep the criteria broad to allow for in Nigeria were evenly distributed in implement- flexibility. It is clear from all the countries re- ing state programs designed to roll out to all viewed that there was tension between the Bank's states. In Russia, the scale of state- and regional- interest in finding progressive, reforming states level engagement was much smaller (only a few that could serve as demonstrations to others and loans), but it nonetheless provided important its interest in supporting poverty reduction by as- lessons about competitive selection of states 10 W H I C H S TAT E S ? E V O L U T I O N O F T H E B A N K S T R AT E G Y Figure 2.1: Distribution of Projects by State A. Brazil B. India Bahia 13% Tamil Nadu Other 8% states 49% Rajasthan 8% Ceará Other 12% states 38% Karnataka 14% Minas Andhra Gerais Săo Paulo Pradesh 10% 8% Pernambuco Uttar Pradesh 15% 8% 17% Source: World Bank data. based on fiscal performance and the use of new forms in public finance and social The initial Bank strategy engagement modalities. expenditure/anti-poverty policies.7 The in Brazil was driven by 1997 CAS advocated continuing the the desire to engage with Brazil same state-centered assistance strat- likely reformers, but it The Constitution of 1988 created a Brazilian fed- egy (table 2.2). was limited by high eration, which today consists of 26 states, one Fed- state-level debt. eral District, and 5,564 municipalities.3 After the The CASs of the mid-1990s pointed decentralization drive fostered by the constitution, out the risks to stabilization posed by Brazilian states were given responsibilities for the the growing fiscal imbalances at the state level and provision of key social services. At the time, it signaled the Bank's willingness to support mean- was believed that devolution of responsibility to ingful reforms through adjustment lending. The the states would lead to more efficiency. Instead federal government was not yet ready to start a it led to unsustainable spending4 and the accu- major reform at the state level, but requested mulation of large state-level debts.5 This began to Bank assistance in its efforts to reform the fi- change with the adoption of the Real Plan in 1994, nances of Brazilian states. In response to these re- which set the stage for economic stabilization6 and quests, the Bank initiated policy dialogue with increased transparency in the finances of the several states, culminating in the preparation of Brazilian states. four state reform loans (de facto development pol- icy loans [DPLs]).8 These loans accounted for 10 With stabilization, federal borrowing declined, percent of new commitments to Brazil and the Bank faced a new situation in Brazil. To and supported the privatization of In the late 1990s, the stay active with a sizeable program, and to be rel- banks and several infrastructure en- Bank began to engage evant, it formally re-oriented its lending strategy terprises owned by the states. with reforming states. in the mid-1990s Country Assistance Strategy (CAS), proposing to shift the focus of lending, pol- The millennium strategy continued direct lending icy dialogue, and advisory work toward the states. to states within fiscal eligibility criteria established Because state indebtedness had reached crisis by the federal government's Fiscal Responsibility proportions, the Bank suggested that lending be Law (FRL), adopted in May 2000.9 This was a time directed to creditworthy states undertaking re- of transition for many states, because they had to 11 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L Table 2.2: Evolution of Bank Engagement in Brazil, Fiscal 1998­2008 2008 Country Mid-1990s strategies Millennium strategy 2003 CAS Partnership Strategy Bank strategy Focus on creditworthy Explicit focus on the Reinforcement of the Continue special focus on reforming states to support Northeast (NE) region commitment to the creditworthy states, with stabilization (states with the highest Northeast region. state governors defining poverty levels) the priorities Alleviate poverty in Special focus on states Emergence of a state- Northeast states that were "turning the dominated lending portfolio corner" fiscally Major instruments/ Policy-based investment Bank's lending First subnational SWAp is Expand the SWAp milestones loans are given to four commitments to states developed (Ceará) instrument to several states decline because of the states need to comply with the FRL Eligible NE states receive First subnational DPL is Greater coordination of loans for poverty given (Minas Gerais) Bank lending cycle and alleviation or improving Brazil's political cycles at social services the state level Portfolio States: States: States: States:a (in dollars) 33 percent 19 percent 18 percent 70 percent Federal: Federal: Federal: Federal: 65 percent 81 percent 82 percent 30 percent Source: World Bank country strategy documents. Note: DPL = development policy loan; FRL = Fiscal Responsibility Law; SWAp = sectorwide approach. a. The portfolio for fiscal 2008­11 shows the planned allocation between state and federal governments. prove their creditworthiness by producing a pri- been dramatic,11 and the federal government mary surplus in the preceding year and a debt/ needed to increase these reserves quickly to pro- revenue ratio of 1 or less, or be in line with the tect itself from external shocks. The Bank re- agreed time path for debt reduction toward a sponded by approving several large programmatic debt/revenue ratio of 1. During this period of DPLs to the federal government.12 Because of the state-level fiscal adjustment, direct lending to dual effect of the federal need for Bank funds and states by the Bank was expected to decline tem- the limitation placed on states' borrowing through porarily. The Bank intended to focus its lending the FRL, the Bank's actual lending to states from and nonlending services on the states of the fiscal 2001 to 2003 was small, about 19 percent of Northeast region, including states with the high- the total lending volume, down from 33 percent est poverty levels.10 Meanwhile, the Bank also during the previous CAS period (fiscal 1998­2000). planned to step up its nonlending services to a broader group of states. The 2003 CAS reinforced the commitment to focus on the Northeast and started building a During the early 2000s, Brazil's economy was on new focus on states with higher levels of poverty a downward spiral prompted by the currency cri- through a concerted effort to strengthen institu- sis of 1998. The drain on foreign reserves had tional capacity. Since virtually all states showed de- 12 W H I C H S TAT E S ? E V O L U T I O N O F T H E B A N K S T R AT E G Y mand for Bank support, an additional filter for would define their development pri- After 2003, the Bank state selectivity was needed. orities and main challenges, while the focused on lending to Bank would present the options (in- states with higher levels of To accomplish this, the 2003 CAS set out a frame- struments) it could offer, based on its poverty and shifted away work for lending to states, with selection criteria analytical work and experience in Brazil from adjustment lending covering sectoral integration (in particular, em- and elsewhere. Another significant in favor of multisectoral phasizing "horizontal" support to public sector change in the 2008 CPS was the at- instruments. management13), and with special focus (agreed tempt to better coordinate the Bank's with the federal treasury) on states that were lending cycle with the political cycle at the state "turning the corner" fiscally.14 In some cases, level.20 poverty criteria gave way to other considerations, such as the demonstration effect.15 India Lending to states under the 2003 CAS (fiscal The 1950 constitution established the Indian Re- 2004­07) became more multisectoral. Significant public with a two-tier federation of states and innovations in lending instruments16 were in- unions; today it consists of 28 states and 7 union troduced, including state-level DPLs and a state- territories.21 The center is traditionally strong and level sectorwide lending approach (multisector exercises control over the most significant spheres SWAp).17 Both instruments were applied in states in the economy. Since the early 1990s, several that had "turned the corner" fiscally, despite sig- factors have contributed to greater decentraliza- nificant disparity in their income levels,18 and tion and strengthening of the federal system. First, supported state teams that had defined innova- the overall trend to liberalize the trade and in- tive, well-tailored reform programs. vestment regime in India led to increased com- petition between states to attract new businesses Despite introduction of new instruments, over- and investments. Second, the decline of the once all lending to states in fiscal 2004­07 remained nationally dominant Congress Party and modest--comprising only about 18 percent of the emergence of regional political par- More recently, the focus new commitments.19 Bank lending was domi- ties distributed power. Third, a 1992 on state-level lending has nated by large, federal-level DPLs targeted at mit- amendment to the Constitution added increased as the need for igating vulnerability to external shocks. Toward the a third tier of local self-government-- federal-level lending has end of the CAS period of fiscal 2004­07, with the the panchayats. declined. sharp rise in primary commodity prices, the fed- eral government's position improved dramati- Until the 1990s Bank engagement at the state cally, which sharply reduced demand for Bank level was mainly a matter of the location of sec- lending. toral projects in infrastructure, agriculture, and the social sectors. This was determined by the sectoral Reflecting on these changes, the 2008 Country dialogue with individual states and the interests Partnership Strategy (CPS) (World Bank Group of the central government.22 Only after informal 2008a) proposed a much smaller federal pro- direct negotiations (subsequently formalized in tri- gram, largely consisting of technical assistance partite agreements) were allowed could loans, some SWAps, and programs that promoted the Bank consider the option of direct In the 1990s, synergies across federal and state lending (for engagement with the state govern- decentralization and example, education and water resource manage- ments. A mid-1990s CAS emphasized, other factors created ment). The majority of financing (70 percent) for the first time, the need to focus ef- an environment was to be directed to state programs, which were forts on state-level sector and fiscal ad- conducive to state-level in compliance with the FRL. This emerging justment issues. engagement in India. state­dominated program was branded as one based on "principled opportunism," where state A second mid-1990s CAS introduced a strategy of governments interested in working with the Bank focusing on reforming states (later called focus 13 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L states). The timing of this shift was significant. In The major change in strategy around the millen- general, the states were in serious financial diffi- nium relative to the mid-1990s was the official en- culties, with combined fiscal deficits gradually dorsement of adjustment lending as an instrument climbing from 2.28 percent of gross domestic to pursue the focus state strategy.26 The Bank's product (GDP) in 1993­94 to 2.85 percent in support to cross-cutting reform programs in the 1997­98, and eventually to 4.17 percent in focus states also generated interest from other 1998­99. The central government faced increas- states. In two cases (Orissa and Tamil Nadu) this ing pressure from states for special bailout pack- interest led to agreement in principle with the ages. Given the fiscal problems of the central central government to move toward adjustment government, it was not only open but keen to find lending.27 In other cases (Maharashtra, Punjab, ways of sorting out the problems of state finances. and Bihar) it was agreed instead that the Bank Hence, when the Bank decided to direct its as- should provide analytical support for cross-cutting sistance to reforming states, both the reforms. The Bank initially focused central government and several states on progressive states. were eager to do business.23, 24 Even so, there was growing concern that lagging states were being overlooked in the Bank pro- The concentration of Bank effort in a few pro- gram. The implementation review of the millen- gressive states had four main justifications. First, nium CAS (World Bank Group 2004a, prepared in aid was most effective when used in a good pol- 2004) found that the focus state approach was too icy environment (Burnside and Dollar 1997). Sec- narrow and left the bulk of the poor unaffected ond, it would have a demonstration effect, signaling by Bank activities.28 At about the same time, India to other states the benefits of reform and demon- adopted ambitious targets to achieve the MDGs, strating the returns to specific policy changes or and the performance of the lagging states with institutional development in the selected states, large concentrations of poverty was essential in which could then be rolled out in others. Third, this regard. Hence, the core strategic issue for the in a vibrant democracy such as India, it seemed to 2004 CAS became how to scale up impact in a na- make more sense to support willing reformers. tional context (table 2.3). Fourth, it enabled the Bank to be selective: it was likely to be more efficient and effective for the Bank The 2004 CAS proposed a change from the focus to focus on a limited number of states rather than states approach to a broader focus on the 12 to spread itself thinly across many. largest and poorest states. It proposed four ways of engaging in states. First, it proposed opening Among the reforming states, Andhra Pradesh was dialogue with the largest and the poorest states one of the first to promote reform and fiscal ad- on the cross-cutting reforms that are the focus of justment and the first to reach out to the Bank for adjustment lending (fiscal management, gover- assistance in this endeavor. As a result, Andhra nance, service delivery, the power sector, and in- Pradesh was the natural choice for a focus state. vestment climate). Second, more resources were The culmination of the Bank­Andhra Pradesh promised for the four poorest states with the partnership led to the Bank's first subnational weakest public institutions--Bihar, Jharkhand, policy-based loan in India:25 the Andhra Pradesh Orissa, and Uttar Pradesh. Third, state-level DPLs Economic Restructuring Project. This massive were targeted to reach 15 percent of total lend- multisectoral investment project, underpinned ing. Finally, there would no longer be an upfront by an agreed multiyear fiscal frame- decision to concentrate substantial state-level in- As concern began to grow work with a total loan/credit value of vestment lending on focus states that were also that lagging states were $540 million, was approved in May 1998. receiving policy-based loans in support of cross- being overlooked, the Andhra Pradesh also benefited from a cutting reforms. Instead, investment lending Bank shifted to a broader number of other investment projects. would be channeled to states that were able to focus on the largest and In 2000, the Bank reported $1.5 billion comply with the federal guidelines for specific poorest states. in commitments to the state. sector engagement. 14 W H I C H S TAT E S ? E V O L U T I O N O F T H E B A N K S T R AT E G Y Table 2.3: Evolution of Bank Engagement in India, Fiscal 1998­2009 1998­2001 2002­04 2005­08 2009­12 Millennium CAS 2004 CAS and Mid-1990s and 2003 CAS 2007 CAS Category strategy Progress Report Progress Report 2009 CAS Bank strategy Focus States Focus States Lagging States Lagging States Engage with states that Reinforce reforms in Open up new Intensify engagement with have chosen to embark on focus states with engagements with the low-income states that are a comprehensive program selective interventions largest and poorest states. lagging but reform-minded. of economic in other states. reforms. Priority states Andhra Pradesh Andhra Pradesh Bihar Bihar Karnataka Jharkhand Uttar Pradesh Uttar Pradesh Uttar Pradesh Orissa Tamil Nadu Orissa (added later) Major instruments/ The first subnational Subnational DPLs are Lending to Centrally Programmatic DPLs and milestones policy-based lending is introduced. Sponsored Schemes sector investment loans for delivered using a increases significantly. lagging low-income states multisector investment (Bihar, Orissa, and Uttar lending approach. Pradesh). Sector investment loans for advanced low-income states. Portfolio States: 54% Focus states: 34% Low-income states: 9% Low-income states: 29% (in dollar terms) Central/federal: 46% Other states: 44% Other states: 36% Other states: 27% Central/federal: 22% Central/federal: 54% Central/federal: 44% Source: World Bank country strategy documents. Note: The portfolio for fiscal 2009­12 shows expected allocations In the end, the buildup of state lending envisaged ing volume) was linked to the signifi- The envisaged buildup during the 2004 CAS period did not take place. Out cant expansion of Centrally Sponsored in state lending did not of the seven state-level adjustment lending op- Schemes.31 occur and in the 2008 erations (SALs) planned, five were approved, but CAS the Bank intensified the total was below the target of 15 percent of total At the request of the central govern- its focus on the poorest lending.29 Single-state loans to the four poorest ment, the Bank intensified its program states. states fell from about 14 percent of total lending for low-income/lagging states under volume to about 9 percent during the same pe- the 2008 CAS (World Bank Group 2008c). This time riod.30 In contrast, the stock of outstanding com- the Bank stressed selectivity and the need to de- mitments directed to the three previous focus velop intensive engagement (programmatic DPLs, states--Tamil Nadu, Karnataka, and Andhra sector investments, technical assistance, and In- Pradesh--increased from 22 to 24 percent. Much ternational Finance Corporation services) in only of the operational engagement during fiscal three low-income states, which were lagging but 2004­08 (an average of 20 percent of total lend- also reform minded. Programs in other low-income 15 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L states would be limited to sector investments and in project supervision and contributed to weak dialogue with the Bank. In more advanced/higher- portfolio performance.33 There was no incentive income states, the focus would be on helping to mechanism in place for the state governments to strengthen institutions so they could deal with compete for resources, and no correlation be- emerging middle-income challenges, relying on In- tween state reform efforts and the resources avail- ternational Bank for Reconstruction and Devel- able to them. opment (IBRD) lending, cutting-edge analytical work, and International Finance Corporation (IFC) The Bank soon realized that it was spreading its activities for private sector clients. resources too thinly to the states and that a more selective approach, based on competition among Nigeria states for the Bank's resources, could be benefi- Nigeria is a three-tier federation consisting of 36 cial for the overall impact of Bank interventions. states, the Federal Capital Territory, and 774 local As a result, the Bank made significant efforts to governments. The three tiers have overlapping re- engage reformist states, Lagos in particular.34 The sponsibilities; subnational governments account identification of Lagos as a priority under the last for a substantial part of fiscal activity but have be- Interim Strategy Progress Report (fiscal 2004; come over-dependent on federal transfers. World Bank Group 2004a) was an attempt to co- ordinate the various efforts of the Bank's sector Bank activities in Nigeria came to a standstill dur- units to derive synergies in Nigeria's largest and ing 1993­99, amid deteriorating governance under most important state. the military regime. During this period, the Bank remained engaged through supervision of proj- The Bank's strategy formally moved to focus on ects already being implemented and a limited well-performing states (or lead states) during the amount of analytic work. The Bank quickly reac- period of the fiscal 2005­2009 CPS. Following the tivated its program after a democratically elected lead states approach, financial and technical as- government assumed power in May 1999. Since sistance would be designed to leverage state ef- that re-engagement, Bank strategy has forts and resources in the selected states to boost The Bank re-engaged with passed through two phases: the pe- economic activity and improve social service Nigeria at the end of the riod of interim strategies, 1999­2003, delivery. Lead states would be able to access a "per- 1990s with state-level with short-term focus;32 and the pe- formance package," a more programmatic, cross- lending disbursed across riod following the adoption of a full sectoral approach to both analytical work and the country. CPS in 2004 (World Bank Group 2005a). financing, drawing on both IDA and U.K. De- partment for International Development (DFID)35 Under the interim strategies, Bank lending at the resources. It was expected that SWAp-type oper- state level was dispersed across the country: proj- ations would be developed for the lead states ects were essentially sectoral programs. Since it as well. was impractical to carry these out in all 36 states, the Bank attempted to choose a subset of states, In selecting lead states, the Bank relied on the usually determined on the basis of the Bank's benchmarking process of the government-led knowledge of and dialogue with the state and State Economic Empowerment and Develop- the interest of key ministers or officials. How- ment Strategy (SEEDS). Participation in the ever, the Bank was constantly under pressure process was voluntary for the states, which were from the central government to expand the num- ranked in four areas: policy; fiscal management ber of states covered under any project, with the and budget process; service delivery; and com- federal government insisting that one munication, transparency, and corruption. Five To better focus its state in each of the six broad geo- states--Cross River, Enugu, Kaduna, Kano, and resources, the Bank opted graphic regions of the country be in- Lagos--were selected based on poverty level, re- to have the states compete cluded for parity reasons. This often gional importance, and previous engagement for them. created serious problems for the Bank (table 2.4). 16 W H I C H S TAT E S ? E V O L U T I O N O F T H E B A N K S T R AT E G Y Table 2.4: Evolution of Bank Engagement in Nigeria, Fiscal 1998­2008 1993­99 2000­04 2005­09 2009 CPS and Category Interim Country Strategy 2008 CPS Progress Report Bank strategy The Bank has no new lending Reengagement in Nigeria Lead State activities in Nigeria as a The Bank reengages in Nigeria in May The Bank formally shifts its focus to consequence of poor governance 2000 with a wide-ranging program of top-performing, reform-minded states; during the military dictatorship. investment lending after a that is, lead states. Lead states are democratically elected government selected in early 2006 based on the assumes power in May 1999. government's SEEDS benchmarking process. Gradual focus to reforming states Semi-lead states Initial attempts to guarantee national Several nonlead states demonstrate coverage diluted the Bank resources strong demand and commitment to available for each state. Lagos sector reforms. As a result, projects emerged as a priority state (after 2002) designed for lead states are in an attempt to coordinate efforts in implemented in nonlead states, known one large and important state for as semi-lead states. greater impact. Priority states None Lagos Lead states Cross River, Enugu, Kaduna, Kano, and Lagos Semi-lead states Bauchi, Kwara, Osun Portfolio None Federal projects: 31 percent Federal projects: 38 percent (expected) (percentage State projects--centrally designed State projects--centrally designed of total volume, for rollout to all states: 58 percent for rollout to all states: end of period) 39 percent (expected) State project--tailored to the needs of specific states: 11 percent State project--tailored to the needs of specific states: 23 percent (expected) Sources: World Bank documents. Nigerian counterparts had opinions that differed tradicted the common knowledge This approach evolved from those of the Bank on the value and modal- about front-runners in both perfor - into a focus on well- ities for the lead states approach. First and fore- mance and quality of services. Third, in performing lead states. most, the concept was challenged for excluding addition to the 36 states with a high the poorest states. In practice, the focus on re- level of independence and authority, the federal forming and well-performing states limited ac- government attached a great deal of weight to the cess for poor states to the assistance and support balance among the 6 geopolitical zones in devel- they needed from the Bank, since almost by def- oping its policies. Because only four of the six geo- inition they have less technical and financial ca- political zones were represented in the selection pacity.36 Second, there was an arbitrary element of the five lead states, the concept was perceived to the ranking of states, which in some cases con- as flawed. 17 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L Nigerian counterparts As soon as the selection was com- The period after 2000 was marked by the strength- argued against the pleted, some nonlead states started to ening of the federal government's position: the approach because it challenge the selection. In many cases, federation was subdivided into seven federal excluded the poorest they were making the case for en- okrugs,40 each headed by a plenipotentiary states. gagement in specific sectors, where presidential envoy; the governors were to be ap- these states' governments had man- pointed by regional legislatures following presi- aged to demonstrate commitment to reforms dential nomination. Income disparities among and even achieve visible results. As a result, the regions became especially pronounced. In 2005, lead states approach rapidly became diluted, and per capita gross regional product in one of the projects that were designed for lead states were richest regions (the oil-producing Tyumen oblast implemented in nonlead states (referred to as in Siberia) was about 69 times higher than that of semi-lead states). Examples include an educa- the poorest region (Republic of Ingushetia in the tion project in Kwara, a rural roads project in Northern Caucasus). In 2006, per capita revenue Osun, and a state governance and capacity build- before transfers in the richest region was 151 ing project in Bauchi. Concern was not limited to times that in the poorest region.41 neglect of the lagging states; it extended to the possibility of the annual selection process re- Uneven distribution of industrial capacity and the sulting in frequent changes in classification and colorful tapestry of multiethnic and multilingual instability in donor interventions from the view- groups add another layer of complexity to re- point of a state. For example, Cross River, ini- gional inequality. Industrial production is con- tially chosen as one of the lead states, will cease centrated in the European part of Russia, mostly to be so with the new CPS coming into force. From in Moscow and St. Petersburg and their vicini- Cross River's perspective, this created a disrup- ties, as well as along the Volga River and in the Ural tion in the amount of Bank lending it could ex- Mountains, while a large portion of the country's pect. At the same time, many observers admit that natural resources, including oil and gas, is lo- the lead state principle has had several cated in Siberia. Hence, the lead positive effects, including inducing state approach was some nonlead states to invest in im- The Bank's first two CASs for Russia in the mid- quickly diluted. proving public institutions. 1990s emphasized the role of the newly emerg- ing private sector in promoting Russia's transition The CPS Progress Report (fiscal 2008; World Bank to a market economy. One of the specific objec- Group 2008b) noted that the Bank would con- tives of the second of the two CASs (World Bank tinue to implement the lead state approach and Group 1999) was to promote private sector de- would use engagement in other states to inform velopment through regional infrastructure proj- its decision on whether to expand into other lead ects. Regions' participation in these projects was states when the next CPS is prepared.37 conditional on demonstrated commitment to the reforms needed to improve long-term credit- Russia worthiness. Lending to the regions, however, The Russian Federation consists of 83 constituent proved to be difficult, because only one of the four units that are referred to as subjects of the fed- planned large infrastructure projects at the re- eration or regions.38 The intergovernmental sys- gional level was approved.42 tem has been undergoing a series of changes since the early 1990s. Subnational units, which Based on this experience, the 1999 CAS found that traditionally had a marginal set of the strategy of lending to the regions was pre- The first CASs for Russia responsibilities, suddenly found them- mature. It concluded that "regional approaches focused on the private selves having to undertake more man- may well become appropriate at a later time sector through regional dates with fewer resources.39 In the when efforts to improve their basic financial projects, an approach 1990s, excessive expenditure obliga- management and soundness have been clarified that turned out to be tions led regional governments to ac- and fiscal and administrative relations between premature. cumulate overdue liabilities. Regions and the federal government have ma- 18 W H I C H S TAT E S ? E V O L U T I O N O F T H E B A N K S T R AT E G Y Table 2.5: Evolution of Bank Engagement in the Russian Federation, Fiscal 1998­2008 1998­2009 2000­02 2003­06 2007­09 Early 2000s CAS 2007 CPS 1999 CAS and and 2005 CAS and 2009 CPS Category Mid-1990s CAS 2001 CASPR Progress Report Progress Report Bank strategy Focus on reforming regions. Regional approaches Focus on well-performing Deepen direct involvement in Large regional investment are not appropriate for and creditworthy regions. the regions by identifying a loans in infrastructure. Russia Federation. subset of 6­10 regions. Subnational reforms Type of Bank support would would be targeted be based on the income level through systemic reforms of the selected regions. through the federal level. Progress/major Lending to the regions does Two large federal Two region-specific DPLs New instruments for developments not progress as planned, projects (fiscal 2000 and are approved, one in St. subnational engagement are resulting in approval of 2002) targeting regional Petersburg (fiscal 2003), developed and implemented: only one regional project. public sector reforms are the other in Kazan (fiscal 1. Subsovereign loans approved. 2005). without sovereign For the first time, regions guarantees. have to compete for 2. Reimbursable technical selection. assistance (fee-for- service). Source: Bank country strategy documents. tured. (World Bank Group 1999, paragraph 50). subnational level.45 The 2007 CPS (World Bank Accordingly, the CAS recommended a shift in em- Group 2006) added a finishing touch: the plan phasis toward activities aimed at rationalizing for gradual shift to the new modalities of coop- federal-regional fiscal relations and improving eration and instruments based on the Bank-wide basic management at the local level. In addition, strategy for engagement with middle-income the CAS stressed that the Bank "should address countries. While the core program of activities was the key agenda of subnational reform more to focus on traditional instruments,46 some re- through systemic reforms via the federal level sources were to be invested in developing new than through large, regional based projects which areas of cooperation, such as direct are very expensive to prepare" (World Bank subnational lending without sovereign CASs in the early 2000s Group 1999, paragraph 103). guarantee (jointly with IFC) and reim- put greater emphasis on bursable technical assistance (fee for subnational development During this period, two federal projects43 that fo- service) (table 2.5). issues. cused on fiscal reforms and public sector man- agement at the regional level were approved. The The 2007 CPS identified groups of regions based novelty of these projects was that they piloted a on five criteria: willingness and commitment to competitive approach for selecting the partici- work with the Bank, history of successful coop- pants. Regions were to be chosen based on demon- eration, reform-orientation and competence, strated improvement of quality and efficiency of strategic importance and the possibility of scaling regional fiscal management mechanisms.44 up in other similar regions, and creditworthiness and potential interest in Bank operations (for Both the 2005 CAS Progress Report (World Bank wealthier or middle-income regions). Regions Group 2005b) and an earlier CAS inched toward were chosen from three groups--high income, more emphasis on development issues at the middle income, and poor. For the three high- 19 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L From 2007 to 2009, income participants, cooperation was conditioned on the perceived commitment and the Bank deepened focused on fee for service, with possi- ownership of reforms and fiscal responsibility. its involvement in the ble direct subnational lending. The 10 regions, but with middle-income regions would be eli- Although the Bank initially attempted to expand selectivity based gible for subnational lending, with fee investment lending, this proved cumbersome on willingness for service on selected issues. Finally, given the two-tier approvals required by the state and commitment, the low-income group, including 11 and the federal governments. As a result, by 2008 creditworthiness, and regions from the Southern Federal the composition of state lending in Brazil had reform orientation. Okrug (not eligible for subnational shifted toward cross-sectoral operations in support borrowing) would be able to benefit of economic policies and public sector reforms, from Bank-supported federal projects imple- such as the first DPL for the state of Minas Gerais mented in these regions, as well as grants and AAA. (2006), and the innovative Ceará and Minas Gerais' Reimbursable technical assistance has multisector SWAps (2006 and 2008, respectively). As the states' fiscal recently attracted a great deal of in- stances improved in terest among Russia's regions.47 Sub- In India, following the mid-1990s CAS, the Bank the 1990s, they were sovereign loans without sovereign substantially shifted its focus to state-level en- allowed to seek financing guarantees are also slowly gaining gagement. The timing of this shift was significant, from the Bank. ground.48 because states were facing financial problems and both the federal and state authorities were keen Conclusions to tap into Bank's resources and take advantage Following the crises of late 1990s, the economic of technical assistance. The Bank decided to ini- situation improved substantially in all four coun- tiate major involvement in progressive reform- tries reviewed, and federal governments became ing states--the focus states approach--initially less dependent on foreign financing.49 At the selecting Andhra Pradesh, Karnataka, and Uttar same time, considerable improvement in states' Pradesh for this purpose. fiscal stances (increasing primary surpluses and declining debt/revenue ratios) generated fiscal In 2004, the Bank CAS signaled a change of strat- space. In this new environment, states were al- egy, noting that the focus on reforming states lowed to seek financing for their investment pro- was causing neglect of the lagging states. The grams, as long as fiscal constraints, established at CAS signaled the intention to provide technical as- the federal level, were satisfied. For the Bank, sistance to the lagging states and to try shifting this was an important development, because fed- lending to them as well. This proved difficult to eral governments were scaling back on borrow- achieve. The 2009 CAS completion report noted ing from the Bank (Brazil, Russia). Adapting to this that, in practice, while lending at the state level new policy environment, Bank strategies in large remained a large share of the overall program, the federal countries proposed a major expansion in share of lagging states had actually declined. state-level engagement. In Nigeria, the Bank reactivated and intensified its In Brazil, the fiscal 2008 CPS (World Bank Group lending activities after return to civilian rule in 2008a) followed the approach of "principled op- 1999, and the project portfolio rose from $80 portunism," which focused on a technical assis- million in 2000 to $750 million in 2007. During that tance program of modest size with the federal period, Bank strategy passed through two phases: government and a major demand-driven lending the first was a period of interim strategies (fiscal program with states. Bank assistance focused on 2000­05) following reengagement; the second some of the more prosperous (such as Rio de was covered by the fiscal 2005­09 CPS, when the Janeiro, Rio Grande do Sul, Minas Gerais, and Bank adopted a medium-term focus. Engage- Săo Paulo) or reformist states (Bahia, Ceará). ment at the state level in Nigeria was driven largely Bank assistance was generally demand-driven, by social and poverty reduction agendas, with 20 W H I C H S TAT E S ? E V O L U T I O N O F T H E B A N K S T R AT E G Y focus on improving infrastructure and providing was sensible. In most cases it was an easy The main issue was how support for agricultural and rural development. decision to make, because federal gov- to do this effectively and During the period of the fiscal 2005­09 CPS (World ernments significantly reduced their how to select states for Bank Group 2008b), the Bank's strategy formally borrowing from the Bank and provi- engagement. moved to focus on well-performing states, called sion of many important public services lead states, seeking to leverage state efforts and is the responsibility of state governments in most resources by granting them access to a perfor- of these countries. In this context, reforming the mance package. Five states were selected based capacity of state governments to manage their fis- on the government-led SEEDS benchmarking cal resources was a critical entry point for state-level process. The lead-state approach is currently engagement; in some cases it was a prerequisite. under review. Once the Bank decided to engage at the state level and defined its objectives to support fiscal re- While a mid-1990s CAS for Russia proclaimed form and poverty reduction in those states, the greater emphasis on regional investment proj- issue became how to do this effectively. It was im- ects (despite recognition that they were expen- possible to engage in every state, so The Bank's intent to sive to prepare and supervise), a CAS near the end the Bank had to be selective. But the invest more in reducing of the decade outlined a phased shift in lending basis for selectivity has proved one of poverty and building away from investment projects in infrastructure the most difficult issues to resolve. Ten- capacity by engaging at and energy in favor of increased emphasis on sion often arose between supporting the state level was systemic aspects of institutional development. A progressive states to carry on with re- sensible. Russia CAS in the early 2000s continued the strat- forms and helping the poorest states. egy shift, emphasizing support for reforms at the Various approaches have been adopted to take regional level, particularly to strengthen public sec- these factors into account. The tension is most ap- tor management. The 2005 CAS Progress Report parent in India, where the pendulum has swung (World Bank Group 2005b) stated that work at the from one criterion to the other during the period. regional level was to be multisectoral in nature and would concentrate on a small number of regions. One approach to reconciling the two selection cri- The 2007 CPS (World Bank Group 2006) added teria is to treat them as two different strands or a finishing touch: the plan for a gradual shift to objectives in the Bank's work. First, the Bank has the new modalities of cooperation and new in- supported fiscal reform both with regard to cen- struments, such as the subnational facility that al- ter/state fiscal relations and reform of state-level lows the Bank Group to provide funds without a finances. Second, the Bank engaged strategically sovereign guarantee to states and municipalities at the state level, trying to adapt the instruments and provision of technical assistance on a reim- at its disposal (loans and analytics) to support bursable basis (fee for service). the broader development/poverty reduction agenda. The different approaches taken by the Overall, the Bank's intent to invest more in re- Bank in the four countries will be looked at from ducing poverty and building capacity through en- these two perspectives in the next chapters of this gagement at the state level in large federal countries report. 21 Chapter 3 · The Bank provided support for fiscal federalism and for improved state capacity to plan and manage rev- enues and expenditures. · Political economy issues make it dif- ficult for the Bank to engage in fiscal federalism, but it can do more than it has in the four countries studied. · The Bank's efforts in support of fiscal reforms at the state level have re- sulted in positive reforms in all four countries, but the outcomes are unclear. · While the Bank had a clear and ac- cepted federal-level strategy for state engagement, it rarely had a state-level strategy for engagement. Where it did, the strategy did not cover all the states with which the Bank engaged. Workers from the Medniy copper processing plant walking home after their shift. Photo by Yuri Kozyrev, courtesy of the World Bank Photo Library. The Scope of Bank Engagement nother issue for the Bank was the scope of engagement at the state A level. In Brazil, India, and Nigeria, the Bank had carried out numerous state-specific projects as part of its support for nationwide sectoral programs. The decision about which states to engage with usually either formula-based or discretionary (for example, in support for agriculture and rural transfers to fund federally mandated programs. development in India or education in Brazil) was The third is allocation of responsibilities for pub- sometimes a strategic choice, but more often a lic expenditures between the federal and state matter of historical engagement or the availabil- governments, which covers most of the major ity of analytic work, based in turn on opportunistic economic and social sectors. The fourth is the ca- involvement of the Bank in particular states. In the pacity of state governments to run deficits and bor- new context, the Bank was increasingly steering row to meet those deficits.3 toward a new model of engagement with two distinct elements: support for fiscal reform and Revenues. The central government normally col- broader multisector engagement at the state level. lects all taxes on external trade. Practice differs with regard to personal income tax and value added and Fiscal Reform excise taxes. In general, state and local govern- In the fiscal area, the Bank provided two categories ments are permitted to levy sales taxes and taxes of support: that for fiscal federalism, addressing on property, both immoveable and moveable, as the fiscal relationship between the federal and well as to collect for local licenses and fees. state governments; and specific support for state governments to improve their capacity to plan and However, there is a vertical imbalance in the four manage their revenues and expenditures. countries studied--that is, the share of tax rev- enues collected by the central government sub- Fiscal Federalism1 stantially exceeds the federal share of expenditures, The fiscal relations between federal and state gov- and vice-versa for the state level. There are three ernments can be broken down into four broad rationales for this. First, it is generally areas. The first is the allocation of revenue au- more efficient for the federal govern- In the countries studied, thority between the federal and state govern- ment to collect taxes; that is, the cost the share of taxes ments. In most countries, the bulk of taxes are of collections per dollar raised will be collected by the federal collected centrally, and even where sharing rev- lower. Second, there needs to be equal- government is much enues with the states is automatic, the states have ization across states; that is, part of larger than its share of no legislative control over the policies and ad- the resources raised from taxation expenditures, requiring ministration that govern taxes. The second is should be transferred from richer states significant revenue transfers from the center to states,2 which are to poorer ones. Third, the central gov- transfers to the states. 25 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L Box 3.1: Increasing Use of Earmarked Transfers in India and the Russian Federation Since 2000, the Indian government increasingly has used ear- Renewal Mission was made conditional on state governments re- marked matching grants to states for special purposes--called ducing stamp duty rates. Centrally Sponsored Schemes. The largest of these is the Edu- In Russia, the number of designated National Projects has cation for All Initiative. The incentive provides major funding for been increasing since 2005. These are sets of targeted programs primary education, but instead of being channeled through state cofinanced by regional governments. They are aimed at imple- government budgets, it goes directly to special parastatals menting the main constitutional guarantees that are under the joint (district-level societies) that the state governments were re- jurisdiction of the federal government and the regions--that is, quired to set up to be eligible to receive these funds. Other major education, health care, affordable housing, and agricultural de- Centrally Sponsored Schemes now include the countrywide velopment. These are funded through earmarked capital transfers. school meal program, the National Village Roads Scheme, the Na- In both countries there is considerable debate about whether tional Rural Health Mission, and the National Rural Employment the high proportion of funding through these schemes and the Guarantee Scheme. The Indian government has also used these loss of autonomy in expenditure allocation at the state or repub- schemes to secure reforms in other areas, such as state taxes. lic level, as well as the conditionalities associated with them, run For example, funding for the Jawaharlal Nehru National Urban contrary to the spirit of federalism. Sources: IEG consultant reports and mission interviews. ernment needs a role in directing resources at the pace of achieving the education MDG led to the state level to programs that are considered national involvement of the national government at the pri- priorities. mary level (box 3.1). Federal responsibility in- creases as one ascends the education ladder. In Transfers from the center to the states. Trans- Nigeria, this happens early, with joint responsibility fers usually have two components. The first is for secondary schooling. In the other countries, block or unconditional grants that are provided secondary schooling tends to be a state respon- based on formulas that generally give the great- sibility, but there are exceptions for specialized est weight to population, but also include equal- schools and disadvantaged regions. There is gen- ization mechanisms, which allocate more funds erally a major federal role in tertiary education. In on a per capita basis to states with lower per many areas, states are subject to unfunded man- capita incomes.4 Sometimes the transfers also in- dates from central legislation that require them to clude an incentive element that ties them to the undertake programs, but no funding is provided. state's own revenue efforts or its willingness to en- gage in fiscal reform. A second set of transfers are Another element of expenditure that is not fully ad hoc and are for specially defined purposes. under the control of the state government is that These may have some criteria or formulas asso- of public sector salaries. While in theory state bu- ciated with them, such as matching grant pro- reaucrats' salaries are independent of those paid grams, or they may be discretionary grants from by the federal government, in practice it is im- line ministries and not necessarily nationwide in possible for state governments to resist the pres- scope (box 3.1). sure to match salaries with those of federal officials. In the past few years, federal govern- Expenditures. The allocation of expenditure re- ment salary increases in India and Nigeria have se- sponsibilities between the center and states usu- riously worsened the financial situation of the ally relates to the locus of programs. For example, state governments. primary education is typically a responsibility of state or local governments, although in India the Deficits, borrowing, and debt management. federal government's frustration with the slow Practice varies on the treatment of deficits and the 26 THE SCOPE OF BANK ENGAGEMENT Box 3.2: 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 nego- shown significant improvement in most of the relevant indicators. tiated targets on the financial debt/net real revenue ratio, a mini- There has been a reduction of debt ratios, increasing revenue mum 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 divesti- For some states, the proceeds from privatization were a significant ture, and level of investment expenditures. The PAF is a three-year contributor to amortization of debt. The program has shown mixed rolling fiscal program, annually monitored by the federal govern- results with respect to structural reforms of the states, such as the ment, and revised as needed, up to the 30 years during which the social security system. state is under the obligations of the debt restructuring agreement. Although specifically designed to resolve the states' debt cri- Of 26 states and the Federal District, only 2 states (Tocantins and sis of the 1990s, PAFs have been critically important, paving the way Amapá), which had no significant outstanding debt, did not sign for the adoption of the Fiscal Responsibility Law (FRL) in 2000, 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 Sources: IEG consultant reports and mission interviews. borrowing capacity of state governments over GDP By the mid-1990s inflation was Where state governments . time both within and across countries. Borrow- above 2000 percent a year and the Real have been able to borrow ing presents a particular problem. On the whole, Plan was introduced, which succeeded from the banking system, federal governments are open to the idea that in establishing financial control and they have sometimes better-performing state governments or their op- price stabilization. However, price sta- incurred high levels erating enterprises should be able to go to the bilization limited the growth of the of debt requiring the market to borrow without a sovereign guarantee states' revenues, and the failure to ad- federal government to and for the markets to impose discipline. Where just levels of state public expenditure, bail them out. state governments are able to borrow from the along with high interest rates, led to an banking system without federal guarantees, they unsustainable fiscal situation for many of them. have sometimes incurred high levels of debt that This led to a third restructuring of state debts in they were unable to service. In these cases the fed- 1997. This time the restructuring was associated eral government has invariably been forced to with formal agreements (the Fiscal Adjustment step in and meet the states' deficits so that, in ef- Programs, or PAF), legal measures, and fiscal and fect, there has been a sovereign guarantee for structural reforms, which contributed to the sig- such borrowing. For this reason, where states or nificant improvement in state finances that took their operating enterprises are allowed to borrow, place after 2000 with the implementation of the there are usually federal oversight mechanisms in FRL (box 3.2). place requiring approval of the borrowing. In India, the government maintains tight control In Brazil, there was a considerable expansion of of all external borrowing and market borrowing subnational indebtedness from the 1980s to the through the issue of state government bonds. mid-1990s, mainly through loans from state offi- These require approval of the Department of Ex- cial banks and bond issues. In 1989 the federal gov- penditure. The constitution forbids direct exter- ernment assumed part of the states' external nal borrowing by states and requires states to . debt, equivalent to 2 percent of GDP This bailout seek central approval for domestic borrowing as proved insufficient, and in 1993 there was a sec- long as they have outstanding debt to the central ond debt bailout equivalent to 7.2 percent of government. 27 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L Box 3.3: Analytic Work on Fiscal Federalism in the Russian Federation 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 Titov 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 Policy Research ulate and contribute to internal debate and policy formulation. Working Paper on Decentralization in Regional Fiscal Systems in The studies published by the Bank included Russia and the Russia: Trends and Links to Economic Performance (Freinkman and Challenge of Fiscal Federalism (World Bank 1994), which outlined Yossifov 1999). This was generally regarded as a major study of the nature of the center-regional issues and tensions and how a fiscal federalism in Russia. It promoted an intensive dialogue be- well-designed intergovernmental system could help reduce them; tween the Bank and the federal and regional governments on Fiscal 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 1996); and Subnational cal Reform Loan in early 2002. Source: World Bank documents. In Nigeria, the debt management framework also mechanism. Nigeria represents a particularly clear distinguishes between external and domestic case of the use of transfers to counteract such ten- debt. States cannot engage in external borrowing dencies. Second, there are often more direct po- without federal approval and guarantees. The litical agendas dealing with the relationship of federal government borrows on behalf of states the party in power in the center with the party in and on-lends to them. States are expected to ex- power at the state level. Third, there is the inter- ecute a subsidiary loan agreement with the fed- est of the federal bureaucracy in maintaining sub- eral government. In domestic borrowing, some stantial say in how resources are allocated and domestic debts require federal guarantees, while used. Finally, there may be governance issues re- others do not. However, there are guidelines for lated to the interest at various levels in capturing each type of domestic debt. States have to com- rents from revenue and expenditure authority. ply with the guidelines that involve federal guar- antees. The level of compliance with guidelines These factors make fiscal federalism a difficult that do not involve federal guarantees is suspect. area for Bank engagement. Where it has sensed openness on the part of the authorities to lend- In Russia, state governments are allowed to bor- ing or analytic work, the Bank has attempted to row without a federal guarantee, and indeed the engage. In Russia, the Bank had an intensive pro- federal government has refused to issue guaran- gram of work on fiscal federalism in the late-1990s tees for state government borrowing and has (see box 3.3). In Brazil, the Bank has undertaken stated clearly that it will not bail out states that en- some analytic work on these topics, including a counter problems in servicing their debt.5 study of state debt (World Bank 2002a). A study carried out in India on state-level fiscal reform Political economy issues There is a substantial political econ- has some discussion of fiscal federalism issues make it difficult for the omy dimension that makes fiscal (World Bank 2005b). In India, the Bank has also Bank to engage in fiscal federalism especially sensitive for gov- facilitated technical discussions of the Indian Fi- federalism. ernments. First, there is the problem of nance Commissions with internationally reputed the cohesion of the federation as a researchers and has contributed to the national de- whole. Most federations are culturally diverse and bate through occasional seminars and confer- there are centrifugal tendencies that the central ences. In Nigeria, the Bank has kept away from this government tries to counter through the transfer topic, given its extreme political sensitivity. 28 THE SCOPE OF BANK ENGAGEMENT While the Bank clearly needs to consult closely cal Federalism Reform Project) were However, it can do with the federal governments of the countries passed on from the federal govern- more--ensuring that it concerned to ensure the timing and nature of ment to the regions as grants, with the has the appropriate any interventions in this area, there would seem intention of maximizing their incen- capacity to analyze the to be three ways in which the Bank can go further tive effect. The Bank's lending for fis- way the current system than it has to date. First, it can ensure that it has cal reform in Russia was based on works and by fostering the in-house analytic capability and information selecting participating regions based cross-country knowledge in this area and can indicate to the federal gov- on fiscal performance. sharing. ernment its willingness to act as an honest bro- ker should the government find this useful. The Bank's support for fiscal reform has Second, the Bank can undertake more analysis of four major components: increasing the way in which the current system is working; state revenues, use of an adjustment ap- Bank lending was an that is, what the impact is on state-level income proach, improved public expenditure inducement for states to and expenditure, without taking a position on management, and support for im- engage in reforms that whether or not the system should be changed. proved governance. The next sections would improve their Third, the Bank can foster cross-country knowl- discuss each of these. fiscal capacity. edge sharing on issues of fiscal federalism by bringing together policy makers and officials to dis- Increasing State Revenues cuss the problems they have encountered and The Bank has a rather uneven record in sup- their approach to addressing them. porting the efforts of state governments to in- crease their revenues. In Russia the Bank has had Fiscal Capacity at the State Level major involvement in this area. In 1995, the Bank In the mid-1990s the federal governments of financed a first Tax Administration Project, fo- Brazil and India, facing serious state deficits that cused on the modernization of local tax offices in contributed to overall macroeconomic instability, Nizhny Novgorod and Volgograd. This was fol- showed increasing interest in Bank engagement lowed by a second Tax Administration Project in at the state level to support efforts for fiscal 2002. In Nigeria, in the State Governance Proj- reform.6 The broad criterion was that the state ect, support for improving tax policy and admin- concerned had to have the fiscal space to ac- istration was included in two of the three states commodate borrowing from the Bank, and the covered by the project (Bauchi and Kaduna) at the carrot of Bank lending could therefore induce specific request of the state governments. In the states to adopt the reforms needed to estab- India, the SAL/DPL programs have had a revenue- lish and maintain fiscal space and facilitate the reform component. In both Andhra Pradesh and process through easing the adjustment.7 Orissa, the Bank expedited the introduction of the value added tax by making it part of the agreed The determination of whether a state has fiscal conditions. In Andhra Pradesh there was also space varies from country to country. In India agreement on the establishment of a Revenue this is assessed case-by-case by the Department Reforms Commission to talk out a tax and non- of Economic Affairs, which reviews states' re- tax revenue reform program. In Orissa, the Bank quests for Bank loans. In Brazil the determination required tax reforms in motor vehicle taxes, stamp is rule-based, with clear criteria established by duty, property tax, and profession tax. Despite this, the FRL. In Nigeria, since Bank loans are passed the view expressed by the Indian authorities is that on by the federal government as grants, this is not this is an area where greater support from the an issue. The overall Bank lending program is Bank would be welcomed. In Brazil this has not the subject of consultation with the Ministry of been a component of the Bank's program, partly Finance, and once this is agreed, the Bank goes because of the support being provided ahead with the individual projects identified as part in this area by the Inter-American De- The Bank's record on of the CPS. In Russia, Bank funds intended to velopment Bank (IDB) and the Inter- support for increasing promote fiscal reform at the regional level (the Fis- national Monetary Fund (IMF). state revenues is uneven. 29 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L One of the general impressions from all four coun- an adjustment loan from the Bank. However, there tries was that the Bank could well do more in was no follow-up, because the Bank's partnership helping state governments improve and expand with Uttar Pradesh weakened amid frequent po- their revenue-generation activities. Almost all state litical changes in state government and weaken- governments need to reduce their dependence on ing commitment to a program of fiscal reform. federal transfers and generate additional resources for investments in physical and social infrastruc- In Andhra Pradesh the Bank determined that a ture. Perhaps even more important, the relation substantial part of the fiscal problem was the pro- between revenue raising and the quantity and vision of free or heavily subsidized electricity in quality of expenditure at the state level needs to the rural areas. Attempts to condition lending on be made clearer to the population of the state. De- a lowering of subsidies proved politically infeasi- mand for good governance is closely related to the ble. In 2007 a detailed evaluation was done of concern of the public that the money that they pay the impact on fiscal adjustment of "Ten Years of in taxes is effectively used. World Bank Sub-National Policy-Based Lending to India" (Howes, Mishra, and Ravishankar 2007). Use of an Adjustment Approach The study (pp. 41­68) concludes that: The Bank has generally linked its adjustment loans at the state level to sound macroeconomic It is impossible to give a rigorous answer management. In Brazil and India this linkage has (to the question of the impact of state sometimes been associated with an effort to turn policy-based lending on state reforms). Per- around a poor-performing state, so that the Bank formance against fiscal targets was largely loan was made at a time when the fis- on-track. While non-PBL states also have Adjustment lending to cal condition of the state was still prob- achieved some fiscal adjustment, overall states generally was lematic. The counterfactual of Bank PBL states adjusted faster and further than linked to sound involvement in fiscal adjustment is very non-PBL ones. We attribute this largely to the macroeconomic difficult to determine, since relatively effective screening put in place which management. rapid growth in all the countries in this avoided the adverse selection problem often study during the period from 2000 to 2007 has associated with PBL, but also to the com- meant that state finances have also improved sub- bination of lending, dialogue and moni- stantially. In Brazil all the states where the Bank toring which helped place reform higher on has been engaged have come into compliance the political and bureaucratic agenda than with the FRL. Meeting the relevant fiscal indica- it would have been otherwise. tors was a trigger for the operations. In the Ceará Multi-Sector Social Inclusion Development Proj- The authors claim that part of the achievement was ect there were indicators for the primary surplus that this supported the Bank's focus state strategy, and revenue/GDP ratio. The Minas Gerais Devel- which "succeeded in its aim of creating a demon- opment Policy Loan followed a similar stration effect across India's states to build support The effects of the model, with up-front requirements on for reforms." But the evidence for this demon- Bank's efforts have meeting the conditions of the FRL and stration effect is not provided, and the Bank itself been obscured by the the separate agreement with the Brazil- has abandoned the focus state approach as inef- high level of growth ian government (the so-called PAF) fective and as limiting its capacity to operate in in all four countries that the state had signed as a condition poorer states. On balance, the judgment must be during the period. of debt write-downs. that while it is likely that there is a positive impact on fiscal adjustment through policy lending at the In India the evidence of the impact of Bank lend- state level, it is a modest impact at best. ing on state-level fiscal adjustment is somewhat mixed. In the early period of its state strategy, the In Nigeria, the Bank adopted the lead states ap- Bank decided that it would try to engage with the proach (along the lines of the India focus states government of Uttar Pradesh, the first state to get approach), which proposed to allocate Bank lend- 30 THE SCOPE OF BANK ENGAGEMENT ing to states that met certain criteria for fiscal re- proving the quality of public expendi- Although improved public sponsibility, such as producing budgets and ac- ture management--that is, enhanced expenditure management counts within a certain time frame. The hope was and timely budgeting, the use of is a key part of the Bank's that this would have a demonstration effect on Medium-Term Expenditure Frame- engagement at the state other states, which would then also become eli- works and outcome targeting, the ac- level, it has done very few gible for lending from the Bank. There is no evi- counting and auditing procedures, Public Expenditure dence to suggest that this has led either the lead oversight by state legislatures, the im- Reviews for states. states or the other states to pursue fiscal reform plementation of treasury systems, state with more vigor (or less vigor in most cases) than procurement, monitoring and evaluation, and so they would have done otherwise. The Bank is on. In Nigeria and Russia in particular, this has now considering the use of development policy been the core of the Bank's efforts to support fis- lending or multisector lending at the state level, cal reform at the state level. which could be triggered by the passage of FRLs. Brazil. The Bank's adjustment and multisector A similar approach was taken in Russia, where el- loans in Brazil can be characterized as mainly sup- igibility for the Regional Fiscal Technical Assis- porting more effective state government pro- tance Loan and the Fiscal Federalism and Regional grams for growth and poverty reduction. The Fiscal Reform Project (FFRFRP) was determined specific content of the loans was not fiscal, but, on a competitive basis, with criteria related to fis- as indicated above, the operations were con- cal performance and the quality of the reform ceived with the fundamental objective of sup- programs put forward. In the Russian case, the evi- porting the adjustment of the states' investment dence seems much stronger for a positive impact and expenditure programs to render them con- on fiscal performance of the participating states. sistent with the requirements of the FRL. Meet- ing the relevant fiscal indicators was the trigger for Improved Public Expenditure Management each operation. A key part of the Bank's engagement at the state level is to improve the quality of public expendi- Although the Ceará Multi-Sector Social Inclusion ture management. This is an important area of the Development Project was considered very suc- Bank's comparative advantage. In this context it cessful in meeting the agreed social and eco- is surprising that the Bank has carried out very nomic indicators, it did not address, for example, few state-level Public Expenditure Reviews (PERs), the existing state financial management and pro- with the possible exception of India, where PERs curement systems, which are regarded as weak. were often part of state-specific economic re- The Implementation Completion Report noted ports. While carrying out PERs for states is costly, that a more proactive approach to supervision of it needs to be remembered that many of these financial and procurement management would states are substantially larger than most of the have been beneficial (World Bank 2008a). The countries where the Bank carries out analytic Minas Gerais DPL laid the foundations for im- work on public expenditures. In some countries proved public sector management, helped in the there is still ambiguity about the relative respon- state's fiscal turnaround, and enabled the state sibilities for expenditure between federal and government to successfully implement a results- state governments. This is particularly so in Nige- based management system. These operations in ria, where, for example, both levels are responsible Ceará and Minas Gerais have had a powerful for secondary education. Again, this is an area demonstration effect on other states, which have where the Bank has not undertaken analytic work shown a great deal of interest in learning more or taken a clear line, even in its lending programs, about the implementation of results-based ap- on the appropriate allocation of responsibility. proaches to development management. Aside from the issue of allocations, however, there India. In the two focus states examined as part of is also the much more significant question of im- this study, Andhra Pradesh and Orissa, the Bank 31 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L agreed with the state government concerned to off to a very slow start, but has now begun to support a substantial set of reforms in public ex- show some progress. A series of targets was es- penditure management. A first set of reforms re- tablished for the project objectives, and there has lated to budget management, including the setting been progress toward them, though none have of hard budget ceilings, the aggregation of sub- been met yet. These projects represent a very heads for small programs to allow flexibility, and small and tentative first step in tackling one of Nige- the de-linking of departmental salary budgets ria's most serious development issues. from programs to allow personnel reallocation and the closing down of dysfunctional programs. Russia. In 1999, the Russian government took sig- nificant steps to reform intergovernmental fiscal A second set of reforms related to the introduc- relations. An interministerial working group for in- tion of e-governance and e-procurement, in- tergovernmental fiscal reform was established, a cluding computerization of document registration, government fiscal concept paper was adopted, and payrolls, pensions, and the government employee the government resumed dialogue with the re- database and the introduction of an Internet- gions on key issues of intergovernmental reform. based tendering system for procurement above The Bank supported this process with two key op- a designated value. erations: the Regional Fiscal Technical Assistance Project (RFTAP) and the Fiscal Federalism and A third set of reforms related to the identification Regional Fiscal Reform Loan (FFRFRL). of public enterprises for reorganization, pri- vatization, or closure; the implementation of the The $30 million RFTAP was approved in Decem- program; and the introduction of a voluntary re- ber 1999. The purpose of the project was to sup- tirement scheme, with provisions for retraining port participating regions in a wide range of and reallocation of the employees. Other mea- institutional development programs.8 Access to sures included the establishment of the Center for RFTAP funds was determined competitively on the Good Governance in Hyderabad, with support basis of performance. The principles were widely from DFID, and the introduction of Medium- advertised. On the basis of this competition, six term Expenditure Framework Plans for selected regions were selected for participation in the government departments. project.9 Despite slow implementation, the proj- ect has produced a formidable list of achieve- Nigeria. The Bank's major intervention at the ments, including enhancements in federal and state level has been the 2005 State Governance and regional fiscal legislation, federal monitoring ca- Capacity Building Project. This was originally rolled pacity, accounting and budgeting in the selected out for three states--Bauchi, Cross River, and regions, and the carrying out of sectoral PERs. Kaduna--but a component in the 2006 Lagos Metropolitan Development and Governance Proj- The objectives of the FFRFRL of 2002 were to ect uses the same model. The model has a stan- support the regions in the implementation of fis- dard package that covers all participating states. cal reform programs that promoted financial It includes public financial legislation, budget transparency and budgetary accountability and preparation, an accounting and reporting system, strengthened fiscal management policies and external audit, budget and treasury information practices at the regional level. This was an ad- systems, and human resource management. In ad- justment loan intended to support the imple- dition there is a state-specific component for mentation of the Regional Fiscal Reform Fund which each state was asked to identify its priority (RFRF) set up by the federal government. The needs. In Bauchi and Kaduna the emphasis was RFRF, in turn, was a window of the federal budget on revenue through the modernization of the to reward regions that demonstrated effective taxpayer identification system, while in Cross River fiscal performance and put in place fiscal reform the state requested the rehabilitation of the Man- programs. Regions competed for this Fund by agement Development Institute. The project got meeting a set of specified standards. 32 THE SCOPE OF BANK ENGAGEMENT Over a three-year period (2002­05), the FFRFRL opment Policy Loan (MGDPL) had In some cases, fiscal made a significant contribution to the evolution components for simplifying the tax reform is closely related of fiscal management at the regional level. Al- regime for small and medium-size en- to improved governance; though the overall amount of the loan may seem terprises, creation of a one-stop shop, in general, however, the small in relation to the size of the Russian econ- streamlining the environmental li- Bank's approach to state- omy, it was leveraged by the resources of the RFRF. censing procedure, and promotion of level governance issues And, for each participating region, it represented public-private partnerships. has been indirect. a sizeable share of their budget and a substantial incentive to improved performance. A set of quan- Perhaps the most interesting operation of this titative proxy indicators of quality was established type is the support the Bank provided to St. at the outset of the project, and the value of these Petersburg in 2003 for an Economic Develop- had doubled by its completion. Among the achieve- ment Project that included reform of the business ments at the regional level were the introduction licensing regime, divestment of city-owned en- of new procedures to control budget commit- terprises, reform and increased transparency of ments, reduction in the number of off-budget ac- procurement procedures, and adoption of zon- counts, monitoring and capping of tax exemptions, ing legislation. This model of providing support improvement of the information base and moni- to a state government that shows genuine com- toring of debt, improved asset management tech- mitment to tackling corruption and improving niques, and strengthened internal audit of the governance through institutional development use of budget funds. and state legislation could be an important ele- ment of Bank operational engagement at the Support for Improved Governance state level in the future. In most cases the Bank's approach to gover- nance issues at the state level appears to have On balance, the efforts of the Bank to The Bank's efforts in been indirect, through the various aspects of im- support fiscal reform have yielded pos- support of fiscal reforms proved public expenditure management de- itive returns in all four countries ex- at the state level have scribed earlier. The Bank has used Investment amined by the study (box 3.4). The resulted in positive Climate Assessments at the state level in some traction that Bank funding provided reforms in all four countries to identify bureaucratic and, by impli- for fiscal reform at the state level was countries. cation, governance obstacles to private sector clear in all cases. While at the federal development. In Brazil, the Minas Gerais Devel- level the amounts of funding the Bank Box 3.4: Improved Governance in the Fiscal Area in Orissa In Orissa, a contributory factor to the state's fiscal problems until But more than any specific action or condition, the general 2000 was the lack of transparency about the seriousness of the perception is that the Bank's involvement and focus on these is- problem. The government preferred to maintain silence about sues contributed to the changes that took place. The current sit- the fiscal problems, and instead to publicize such achievements uation is now almost the reverse of the previous situation, and state as the size of the approved (but not implemented) annual plans. government and officials now take pride in the openness of the The Bank's involvement helped to illuminate the seriousness of Budget-at-a-Glance document and the disclosures on the Internet. the problem. An unpublished Bank report on fiscal reform and eco- A lesson from this experience is the importance of sustained nomic growth in Orissa cited the importance of issuing a White and continuous engagement with the state government on these Paper as the first requirement of launching the reform program. issues. This contributed to the decision of the state government to prepare an issues paper on the state's finances. Sources: IEG consultant report and mission interviews. 33 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L can provide are relatively small for these large stability (Andhra Pradesh in India or Minas Gerais countries, at the state level they represent real fi- in Brazil). nancial additionality. In some cases the Bank has labeled these re- In all the countries the Bank's role as an honest forming states "focus" or "lead" states, with the broker in providing incentives and support for implication that it would design a package of in- state governments to participate in the federal fis- terventions intended to support the develop- cal reform agenda was significant, and all the state ment effort. The two cases that were most fully representatives interviewed appreciated the ex- worked through are the states of Ceará in Brazil ternalities of a direct dialogue with the Bank and and Andhra Pradesh in India. In both states the the sense that the Bank was committed to, and Bank program combined investment lending in interested in, their success. most of the core economic and social sectors with multisector lending. The objective was to de- When one moves to outcomes of Bank engage- rive synergies from the combination of activities. ment in fiscal reform, measurement and attribu- In Andhra Pradesh, for example, the Bank was ex- tion become difficult. As indicated for all the plicit in its view that the difficult measures re- countries concerned, 2000 to 2007 were years of quired for fiscal reform needed to be matched by growth with improving finances at all levels of increased investment in agriculture, rural devel- government. The Bank selected states where the opment, health, and education to provide a po- government had shown commitment and already litically acceptable package of reforms. started to move in the proposed direction. In every case where the state followed through to the It is noteworthy that the states in which the Bank point of borrowing from the Bank, there is evi- had a broad strategic engagement were generally dence of improved fiscal management. In some not the poorest. Instead, the Bank has spent a cases (Russia and Brazil) the improvements went great deal of time in these countries supporting beyond Bank projections. In India, improvement relatively high-income, high-capacity states (Minas went beyond expectations in Orissa and was per- Gerais, St. Petersburg, Karnataka). While this sup- haps in line elsewhere, and in Nigeria it fell short, port has added value, it has come at the expense but with forward momentum. of Bank efforts in poorer states that lack capacity. The selection of states for broader engagement Multisector Engagement at the often focused more on the fiscal reform crite- State Level rion than on the broader poverty issue. The state In addition to the specific focus on fiscal reform of Alagoas in Brazil is a good example. The re- (discussed in the previous section), another aspect forming government of the state is saddled with of the Bank's engagement with states in the past a large debt overhang from past administrations decade has been to take a holistic view of devel- and lacks the fiscal space, required by the federal opment at the state level--in essence FRL for external borrowing. The Bank, with the In several states the Bank treating the state as a country--and support of the federal government, is ready to pro- has developed a broad developing a strategic approach to vide assistance to Alagoas, but making an excep- strategic approach to supporting poverty reduction at the tion in this case might put at risk the discipline poverty reduction, state level. The entry point for this that the Brazilian authorities have worked so hard designating such states broad engagement has generally been to achieve. "lead" or "focus" states. a state government commitment to fiscal reform, evidenced either through On the whole, however, the fiscal entry point has its record over the years (as in Ceará been a good proxy for the commitment of the in Brazil or Karnataka in India) or state government to development and poverty re- But these states were through specific agreements with the duction. For example, when the Bank initially at- generally not the poorest Bank as part of a multisector lending tempted to work with Orissa, one of India's poorer in their countries. operation, designed to restore fiscal states, the poor quality of fiscal management did 34 THE SCOPE OF BANK ENGAGEMENT Box 3.5: The Bank Program in Andhra Pradesh Andhra Pradesh is considered a middle-income state in India. In helped Andhra Pradesh to become the frontrunner in a number the late 1990s, Andhra Pradesh became one of the 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, e-procurement on a wide scale, and so on. to target the reforming states. Andhra Pradesh was selected as Andhra Pradesh also implemented a number of investment the initial focus state. The culmination of the Bank- Andhra Pradesh projects (in the rural poverty, forestry, water, and power sectors) partnership was the Andhra Pradesh Economic Restructuring over fiscal 1998­2008. The Bank has also supported significant non- Project (APERP, 1998), a massive multisectoral loan underpinned lending activities in Andhra Pradesh. It started with the "Andhra by an agreed multiyear fiscal framework with a total loan/credit Pradesh: Agenda for Economic Reforms" (World Bank 1997) report, value of $540 million. This was essentially a DPL that had to be con- which underpinned the subsequent lending program. In addition structed as an investment loan, since Bank policy did not approve there have been several Andhra Pradesh Policy Notes covering of subnational DPLs at that time. issues such as fiscal and debt management and analysis of its After Bank policy on subnational DPLs changed, Andhra growth potential and public enterprise reform. Pradesh received three DPLs (fiscal 2002, 2004, and 2007). The The portfolio of Andhra Pradesh has changed with the Bank's Bank's first state-level policy-based loan (S-PBL) to Andhra shift in strategy away from focus states to lagging states. While Pradesh addressed not only the need for fiscal adjustment, but also Bank's commitments to Andhra Pradesh had reached $1.5 billion public expenditure management reform and restructuring of pub- by 2000, after the strategy shift in fiscal 2004, the Andhra Pradesh lic enterprises, including privatization. The later DPLs added three portfolio gradually fell. In June 2004 it comprised 10 percent of the specific sectoral foci: power, health, and education. The Bank's total lending volume ($1.2 billion) and fell to 5 percent ($0.74 billion) S-PBL program was controversial in India because it censored in June 2008. At the time of the 2008 CAS, only one project planned some populist measures such as free power and irrigation water for the state was considered firm (Andhra Pradesh Water Sector for farmers. Despite the initial setback in the power and irrigation Improvement). sectors, the S-PBLs not only ushered in fiscal correction but also Sources: World Bank documents and mission interviews. not allow for adjustment lending and broad en- State-Level Strategies The Bank's strategy had gagement. The Bank worked patiently with the au- Many of the states with which the Bank the approval of most thorities until there was evidence of progress in is working closely are larger than most stakeholders. fiscal reform, and then moved rapidly to provide of the Bank's borrowing member coun- a wide-ranging support program. tries. The Bank's lending programs in Ceará, Minas Gerais, Andhra Pradesh, Orissa, and Lagos The Bank's broad strategy of state-level inter- are far larger than overall lending in many coun- ventions generally enjoyed the approval of most tries. Yet these programs have proceeded with- stakeholders. It was closely coordinated with and out the preparation of strategy documents. endorsed by the federal government--or rather, followed the federal policy. The specific contri- The overall state strategy is discussed in CAS doc- bution of the Bank was in identifying specific uments, but there is no room for outlining indi- measures to implement this strategy, prioritizing vidual state-level strategies. To do this would clearly various components, and adjusting modalities to add considerably to the work program, but there suit individual states. The state governments, are a number of factors that argue for periodic often feeling somewhat constrained in imple- preparation of brief state strategies for the three- menting reforms by political factors, realized that to-five states in each country where the Bank has Bank engagement provided a way of reducing or is likely to have a major engagement. This is the political risks of supporting reform.10 likely to help on several fronts. First, there is the 35 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L Although some of the matter of ownership of the program. It an economic blueprint (Lagos Ten-Point Agenda) states the Bank has is easier to secure a clear public com- with poverty alleviation and job creation as the worked with are larger mitment to the approach, as well as central themes. Later (since 2004), all state gov- than most of its bringing on board nongovernmental ernments in Nigeria developed comprehensive de- borrowing countries, its organizations (NGOs) and civil society, velopment strategies--SEEDS. programs in those states if the Bank can use a strategy docu- have not benefited from ment as the basis for a broad dialogue. The Bank's CPS later built on the Lagos strategy fully articulated Second, in most states and regions the and SEEDS and focused on improved governance, strategies. Bank is not the only donor, and strat- private sector­led growth, and human develop- egy documents could be useful for ment at the state level. A promising approach donor coordination. Third, this exercise would to establishing partnership relations with regions help to identify gaps in the knowledge framework in Russia are the memoranda of cooperation, at the state level and for the Bank and the state gov- signed with a few regions (Khanti-Mansiisk, ernment to plan follow-up analytic work and bet- Tatarstan, and Voronezh). In order to become a ter integrate it into the lending strategy. signatory to such a memorandum, a region should have a record of cooperation with the Bank. In In several cases, either the Bank or the concerned Brazil, the Bank prepared state economic mem- state government did try to put forward addi- orandums for a few states in the late 1990s, but tional state-specific strategy documents. But these this effort was not pursued in all states. In India, efforts did not cover all states in which in addition to a number of high-quality state eco- Where there were the Bank had major engagements, nomic reports, (for example, World Bank 1997, state-specific strategy and they were not pursued in a con- which underpinned the subsequent lending pro- documents, they did not sistent manner, despite an enthusiastic gram), several informal state strategies have been cover all the states with reception at the state level.11 In Nige- prepared and discussed with the authorities, but which the Bank was ria, in the initial period of Bank re- they were never formalized and/or spread to engaged. engagement (2000­03), Lagos prepared other states. 36 Chapter 4 · Finding the right lending instrument for states was an early challenge. · In most cases, multisector lending proved most effective for getting funding to the right areas at the state level. · The fee-for-service approach also offers promise, but it is relatively new. · The Bank's implementation arrange- ments were adequate, though there were concerns about the length and speed of the Bank's processes. · The Bank's analytic work and tech- nical assistance at the state level have been timely and competent, but it has not been considered a strate- gic part of the Bank's state-level engagement. · The Bank has generally partnered effectively with other donors at the state level. Road work in Brazil. Photo by Thomas Sennett, courtesy of the World Bank Photo Library. Modalities of State-Level Engagement T he issue of how to engage has also been a challenge for the Bank, and there has been considerable evolution in the approach. The first chal- lenge was the use of adjustment lending at the state level. The first such adjustment loan was made to Uttar Pradesh, India, in 2000. Until that point the Bank had struggled to find an instrument to attach to its policy dialogue and strategic approach at the state level. Evolution of Instruments also brought to the forefront the Adjustment lending The multisector restructuring loan in Andhra intersectoral linkages required to quickly became the Pradesh, an investment loan undertaken in 1998, achieve results, such as the need for preferred instrument at was a way of squaring the circle, but it was an enor- improved water supply to reduce in- the state level. mously costly operation to prepare and supervise. fant mortality. With the adoption of the policy to allow for ad- justment lending at the state level, the Bank was Another important innovation has been the pio- much better placed to conduct a policy dialogue neering of reimbursable technical assistance (fee with the state governments and to focus on some for service) at the regional level in Russia. Bank of the key policy constraints, such as governance, budgets rarely allow the level of analytic work the investment climate, and the quality of public demanded by intensive engagement in three-to- expenditure management. five states, and an approach that permits states to pay for additional work has considerable prom- Adjustment lending rapidly became the instru- ise for other middle-income countries. ment of choice to support fiscal reform and statewide strategies in Brazil and India. In Brazil, As a result of these new developments, state gov- however, there was a sense that adjustment lend- ernments recognized that they had an inde- ing was not as effective in reaching out to the pendent role in promoting policy reforms and line ministries in key sectors. The Brazil country financial execution, in addition to their normal re- team developed the innovative Multisector SWAp sponsibility for project execution. In Brazil, State (see box 4.1), a results-based instrument with Reform Loans (de facto DPLs) for the target indicators defined for each sector and dis- states of Mato Grosso, Rio Grande do Multisectoral lending bursements associated with achievement of the Sul, and Rio de Janeiro (1997­98) in- helped to attract the targets. cluded considerable policy compo- attention of the right nents on fiscal adjustment (cuts in staff ministries, and in Multisector lending has proved an indispensable and salary of civil servants, pension Nigeria it helped component for drawing the core ministries-- and tax reforms) and structural reform. overcome federal finance and planning--and the line ministries This new trend better responded to authorities' concerns into a dialogue on development priorities. It states' own investment and reform about loss of control. 39 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L Box 4.1: Multisector SWAp SWAps traditionally have been associated with a single sector. From It allowed the government to manage its fiscal resources and al- this perspective, the SWAps in Brazil are innovative because they locations while forging a partnership between the central and were adapted to integrate several sectors and they were delivered the line secretaries. at the state level.a Ceará was the first state to implement a multisector SWAp.b This loan had varying degrees of success in achieving the desired This loan aimed to strengthen social inclusion while preserving fis- outcomes in the targeted sectors. It was very successful in lever- cal sustainability. It had a unique design, because it was an Adapt- aging the Bank's support to ensure that sector expenditure levels, able Program Loan making use of a results-based SWAp loan which had been about 28 percent of budgeted amounts before the modality. start of the project, were up to 70 percent. The loan also helped The key innovative features of this loan were: to reinforce other traditional investment projects. Two of the three investment loans in Ceará were upgraded to satisfactory as a re- · Multisectoral Model and Activities. The loan had two compo- sult of cross-conditionality. The downside of the modality of this nents. The SWAp component supported nine eligible expendi- loan is that it is complex and it can be successfully implemented ture programs across five key line sectors: health, education, only if there is sufficient institutional capacity.c water and sanitation, water source management, and environ- ment. The technical assistance component primarily supported Multisector SWAps in Other Brazilian States: The Ceará SWAp public sector management. Targeting six sectors assured that model was sufficiently successful that a second SWAp was ap- institutional reform synergies would take place across sectors. proved ($240 million, 2008). The model is also being replicated in · Results-Based Disbursement. The project emphasized the use Minas Gerais ($976 million, 2008), the Federal District ($130 million, of results-based disbursement. Disbursement depended on the 2009), and Pernambuco ($155 million, 2009). Each of these opera- following: (a) the borrower had reached specific disbursement- tions follows essentially the same SWAp structure, combining linked indicators mapped to each sector; (b) the borrowers' pri- support to public sector management reforms with support for mary surplus was above a certain threshold; (c) performance of selected sector programs. In addition, each one links disbursements Ceará on three other World Bank loans (water, education, and to the borrower's ability to meet specific performance targets over rural development), ensuring that these loans met stated phys- the course of the loan. The nature of the indicators varies in ac- ical as well as disbursement targets; and (d) the borrower had cordance with the institutional development of the state/sector. actually spent at least 70 percent of the amount budgeted for the For example, the Minas Gerais SWAp relies to a greater degree specific sectors. than the others on outcome measures: appropriate "stretch · Disbursement to Treasury and Not to Sectors. The Bank re- goals" for one of Brazil's most developed states. In contrast, Ceará leased its funds directly to the state treasury as a single tranche SWAp-II will use mainly output measures rather than outcome tar- for reimbursement of expenditures under the various programs. gets as disbursement indicators for the sector programs. Sources: World Bank documents and IEG mission materials. a. Before Brazil, only Mexico had implemented a multisector SWAp (fiscal 2004), but it was at the federal level. b. Ceará Multi-Sector Social Inclusion Development Project ($150 million, fiscal 2006). c. Despite being a poor state, Ceará was considered to be relatively well managed and reform minded, making it a viable candidate for implementing the SWAp. plans, thus strengthening their sense ownership ernment's commitment to reform. There has been and commitment and improving capacity. no such proposal since then. Federal authorities have been skeptical about the use of DPLs in Nige- The only country among the four that received ria, including at the state level, and have expressed only investment loans (on IDA terms) was Nige- concerns about possible loss of control over the ria. Proposed adjustment lending (World Bank public finance system. At the same time, some Group 2000) never materialized because of con- state-level counterparts have expressed interest in cerns about macroeconomic stability and the gov- state-level DPLs or multisector SWAps. 40 M O D A L I T I E S O F S TAT E - L E V E L E N G A G E M E N T Box 4.2: Minas Gerais--From Management Shock to Results-Oriented State After years of economic mismanagement, the budget of Minas These data were entered in the central project monitoring sys- Gerais reached a deficit of R$2.3 billion (about $1 billion) in 2003. tem for regular monitoring of their performance, which enabled Public investment dropped from an average of R$5 billion/year the government to limit wasteful expenditures. The combination (2003 prices) in 1995­98 to R$1.5 billion by 2003. The net consoli- of the traditional policies with the impact of GERAES turned the dated debt of the state amounted to 238.87 percent of the net cur- fiscal situation around. Starting in 2004, the state had four con- rent revenue in 2003. Because of its indebtedness, the state could secutive years of fiscal surplus, after one entire decade of fiscal not count on financial support from the federal government to deficit. contract credit operations (this was prohibited by law). Faced with extremely adverse fiscal conditions, the newly Second Phase (2007­to date) elected state government initiated a set of bold adjustment mea- Taking advantage of the improvement in public accounts, the sures, the so-called choque de gestăo (management shock), in Jan- Second Phase was launched in 2007, under the label of Results- uary 2003. This management shock can be divided into two stages: Oriented State. This meant that fiscal balance would be a prereq- fiscal balance and fiscal quality. uisite for government action, and government performance would be measured by improving outcomes; that is, it had to be results- First Phase (2003­06) oriented. To move toward achieving this goal, the governor of The first set of measures concentrated on traditional policies: re- Minas Gerais asked for the Bank's support. The Bank responded gaining fiscal sustainability by increasing revenue and rationaliz- with a two-tranche DPL of $170 million, which was fully disbursed ing expenditures; reorganizing the state government's macro in May 2007. This DPL had three pillars--fiscal stabilization, structure (administrative reform); integrating planning and man- public sector reform, and private sector development. All three pil- agement instruments; and establishing a new human resources lars had a well-defined action matrix and impact indicators. It policy--for example, creating a ceiling for remuneration. helped the government to formulate its actions so that it would be The innovative feature was the creation of the Strategic Re- results-oriented. source and Action Management. GERAES, as it was known, was To continue with his program, in July 2008 the governor of the operational framework that guided the allocation of resources Minas Gerais asked the Bank for a loan of $1 billion (the entire bor- to structural projects. In 2005, the government defined and agreed rowing capacity of the state in its agreement with the federal on performance goals for all 31 of the state's structural projects. treasury). Sources: Government of Minas Gerais; IEG mission materials and interviews. Multisector SWAps and State-Level DPLs of credit operations, the newly elected state gov- In Brazil the pioneer states in piloting the new in- ernment approached the federal government and struments were Ceará and Minas Gerais. The the Bank for support of its program to turn around Ceará multisector Social Inclusion Development the inherited negative economic and financial Project (SWAp) was developed in response to the trends and to put the state back on a sustainable severe fiscal constraints faced by the state,1 which fiscal path. The project was designed to put in threatened key public investment programs and place the state's overall budget constraints and to delivery of core public services. The implicit cri- bring Minas Gerais back in line with the national teria for selecting Ceará for this initiative were stabilization program. More important, it adopted poverty, environmental challenges, fiscal sus- the state government's choque de gestăo (man- tainability concerns, and the strong commitment agement shock) approach (box 4.2) to reforms. of state leadership to modernization and reform. Apart from laying the foundations for improved A slightly different situation led to development public sector management, the project helped the of the Minas Gerais2 Partnership for Develop- state to complete a fiscal turnaround and en- ment DPL. In 2003, after a period of suspension abled the government to regain its investment 41 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L The Ceará SWAp and capacity. Overall, besides their obvi- by federal law to carry out policy and institutional Minas Gerais DPL not ous contributions to macroeconomic reforms. Originally the loan was contemplated only contributed to stability, a sustainable fiscal situation, as an investment operation, but given the time macroeconomic and improved performance of other in- constraints,6 it was transformed into a DPL. There stability, but also were vestment loans, the Ceará SWAp and was considerable skepticism in the Bank regard- instrumental in Minas Gerais DPL were instrumental ing the feasibility of reforming Kazan's financial introducing and in introducing and institutionalizing management system. Some viewed the loan as a institutionalizing the culture of results-based manage- political order from the central government. Nev- the culture of results- ment at the state level and strength- ertheless, the importance of engaging an impor- based management ening coordination within the state tant region within an important client country at the state level. government. overweighed the risks. In hindsight, the Bank's de- cision to proceed with the loan was rewarded In Russia the only two regional DPLs were im- with successful implementation and satisfactory plemented in the cities of St. Petersburg3 and outcomes,7 and it serves as an example of a close Kazan.4 Both projects were designed to support and productive partnership with the client at the the implementation of the federal government's local level. Medium-Term Program of Social and Economic Development for 2002­04 by ensuring conform- Despite the overall success of both regional DPLs ity of federal and regional legislation and by in- in Russia, they are unlikely to be replicated. First, troducing clear functional divisions of authority the financial approach of the Kazan project, with and financial resources between the federal and loan proceeds provided to the city on a grant regional levels. Both loans were designed as sub- basis, is unlikely to be repeated in other munici- national DPLs, with disbursement tied to progress palities. Second, the City of St. Petersburg ac- in the implementation of reform under specified complished the needed reforms "for free," with terms and conditions. no incentive other than its own good will. It is worth noting, however, that regional counter- The objectives of the St. Petersburg DPL included parts in both cases admitted that the key value of improving the business climate, stimulating ex- the loans was not the monetary incentive, but pansion of private sector participation and pro- the authorities' understanding of the usefulness moting development of the land and real estate of fulfilling the loan conditions--together with the markets, and strengthening the city's fiscal man- acquired capacity and skills. With hindsight, the agement and ensuring the long-term stability of projects could have been even more successful its fiscal revenue base. The "city component" was had they been supplemented with sizeable tech- closed under somewhat unusual circumstances: nical assistance for implementation of conditions. despite meeting all loan conditions, the City of In the absence of a technical assistance compo- St. Petersburg declined both tranches and re- nent, such assistance was limited to what could quested reallocation of the first tranche to the "fed- be provided by the Bank staff in the course of eral component" and cancellation of the second preparing and supervising the loan. one. The formal reason given was that a contin- uous budget surplus (2004­06) made the Bank re- Investment Lending at the State Level sources redundant. Investment operations in all four countries com- prised the bulk of the Bank's lending. The Bank's Unlike St. Petersburg, Kazan had not been an method of operation did not differ greatly from independent subnational entity, but an adminis- its practices in a regular investment operation at trative unit within the Republic of the federal level, or in a country of a size compa- The Russia DPLs, though Tatarstan,5 and obtained the status of rable to a state in a federation. The main value was successful, are unlikely a municipal formation only in 2004 probably the greater chance of adopting lessons to be replicated. (after loan approval), which enabled it and using the model of a successful state-level in- 42 M O D A L I T I E S O F S TAT E - L E V E L E N G A G E M E N T vestment project in a similar state within the same analysis. In the mid-1990s the Bank un- Much of the Bank's federation, compared with country-to-country dertook a series of state economic investment lending to exchange of experience. memoranda in Brazil that were much states was for appreciated by the state authorities. infrastructure and Among examples of such success is the Basic Ed- In most cases these were collaborative operated in much the ucation Quality Improvement Project in Minas efforts, and the outputs were viewed as same way as operations Gerais (Brazil). This project helped the state gov- joint products. For reasons that are un- in small-country ernment to create an enabling policy environ- clear, the Bank has not followed up on borrowers. ment to increase school productivity through this model, either in Brazil or in other progressive reform measures aimed at process im- countries. provements and decentralization. The culture of monitoring and evaluation introduced to the sec- Analytic work does not appear to have The Bank's analytical tor by the project made it easier for the state to been considered a strategic part of the work and technical develop an appropriate set of education indica- Bank's state-level interventions--there assistance--focused tors for the later SWAp operation. The Minas was no organized attempt to identify mainly on fiscal issues-- Gerais education program is currently viewed by key knowledge gaps and to develop were timely, competent, other states as a potential model that can be ap- partnerships with state institutions to and widely appreciated. plied elsewhere. meet these gaps. In some cases, stud- ies have been carried out, but have not Most of the Bank's investment loans at the state been made publicly available. Given Analytic work was not level supported infrastructure development. In the paucity of analysis and information considered a strategic Ceará (Brazil), the Bank provided support for an at the state level in many countries, part of the Bank's state- irrigation program that is credited with a major en- this is an obvious gap to be filled by the level interventions. hancement in the state's economic prospects. In Bank. Nigeria, with its much smaller states, infrastruc- ture has been a difficult area for the Bank: a water In general, wherever the Bank's analytic work supply project covering six state capitals proved preceded the projects in states, it clearly con- extremely difficult to implement and had limited tributed to bridging the knowledge gap and to im- impact. In Nigeria's larger states, the Bank has proving program design. In Russia, the Bank's been able to intervene more effectively in infra- work on intergovernmental financial relations structure, and the Lagos Metropolitan Transport (World Bank 1994; Le Houerou 1995; Le Houerou Project is one of the success stories of the Bank's and Rutkowski 1996; Freinkman, Treisman, and program there. Titov 1999; Freinkman and Yossifov 1999) provided a foundation for a future comprehensive pro- Capacity Building and AAA gram of reform and restructuring (see box 3.3). Overall, the Bank's analytical work and technical In Nigeria, the Bank had an intense AAA program, assistance at the state level were widely appreci- albeit mainly at the federal level (only 8 ESW ated and regarded as timely and competent inputs products out of 75 focused on the state level), in building capacity. At the same time, there is an which facilitated modernization of the tax sys- appetite for more, deeper, and more state-specific tem, adoption of the Medium-Term Expenditure economic and sector work (ESW) and technical Framework, and the design of the FRL at the state assistance, as well as room for improving client par- level. In India,8 the Bank's programs in Orissa ticipation and strengthening the link between and Andhra Pradesh to some extent owed their the analytical work and lending. success to the high quality of analytical and di- agnostic work preceding such engagements. The Bank's analytic work at the state level has fo- cused mainly on fiscal issues. There has been rel- To improve the link with local demand, the Bank atively little other state-specific sector or thematic might want to consider formally eliciting stake- 43 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L Table 4.1: Average Preparation and Supervision Costs of State- and Federal-Level Projects, Fiscal 1998­2008 Preparation cost Supervision cost Total cost Country (US$ '000) (US$ '000) (US$ '000) Brazil Federal 237.7 360.1 597.8 State 283.9 407.6 691.5 India Federal 421.8 563.2 985.0 State 517.6 622.5 1,140.1 Nigeria Federal 469.4 840.9 1,292.9 State (designed to roll out to all states) 650.3 853.4 1,503.7 State (tailored for individual states) 944.5 804.6 1,749.1 Russia Federal 773.6 703.4 1,477.0 Region 952.8 552.4 1,505.2 Source: World Bank data, April 4, 2009. holder views on suitable subjects for AAA. Knowl- sory services to help develop public-private part- edge of international best practices, a clear com- nership arrangements in four investment proj- parative advantage of the Bank, needs to be ects. There are similar requests from a number of brought into analytical work to a greater extent. other regions (Khanti-Mansiisk Okrug, Chuvash Republic, Volgograd Oblast, and others).9 It is Technical assistance associated with Bank loans noteworthy that reimbursable technical assis- is often seen by the state-level clients as a primary tance (fee-for-service) arrangements implemented advantage of borrowing from the Bank, as op- at the regional level in Russia, which were quite posed to other lenders (especially in middle- popular with both central and state authorities, income countries such as Brazil and Russia). have yet to be tested in other countries, despite However, despite a declared focus on capacity seemingly fertile ground for this modality of co- building, the Bank rarely tested state-specific operation, most notably in Brazil. Borrowers often see technical assistance loans, although the Bank's technical some loans did include a component Implementation Arrangements assistance as a major for technical assistance and/or train- and Staffing benefit of borrowing. ing. Russia is the only country among In most cases, the Bank followed standard pro- the four included in this review where cedures for implementing its projects at the state reimbursable technical assistance (or fee-for- level--which are quite similar to those used in fed- service) is being implemented. It includes prepa- eral projects, with rare exceptions. At the same ration of regional strategies, sector analyses, and time, there were several good practice examples technical support for specific investment activi- that deserve mention. In Brazil, the partner agency ties (for example, public-private partnerships in for Ceará SWAp was the parastatal Economic Re- infrastructure). search and Strategy Institute, which set up a proj- ect implementation unit staffed by regular state Growing demand for this new instrument can be employees. This decision proved important for seen in the regions, in part because of the fiscal smooth implementation of the project. The Bank federalism loans (RFTAP and FFRFRL, and government counterparts identified and mon- The Bank's discussed earlier) that were instru- itored disbursement-linked indicators, with di- implementation mental in demonstrating the high qual- rect involvement of central and line secretariats. arrangements were ity of Bank's technical assistance to the This arrangement improved the interagency co- similar to those it used at regional administrations. The City of St. operation among line secretariats, because achieve- the federal level. Petersburg used the Bank's paid advi- ment of many indicators required joint efforts. 44 M O D A L I T I E S O F S TAT E - L E V E L E N G A G E M E N T Box 4.3: Bank Engagement in Bihar: An Example of Effective Partnership Bihar is India's third-most populous state and one of its poorest. tance ($5 million through a DFID-financed trust fund, comple- About 41.4 percent of the state's population lives in poverty com- mentary to the DPL). The approval of the DPL was especially pared with 27.5 percent for India as a whole. Average state per significant because it accounted for over 20 percent of the state's capita income is about a quarter of the all-India level. Bihar's so- own revenues. cial indicators are among the lowest in the Indian states, and if pres- Several national projects supported by the Bank are currently ent trends continue, it will attain only two of the eight MDG targets being implemented in Bihar, including one-third of the Bank's by 2015 (reduction in child malnutrition and access to safe drink- Lucknow-Muzaffarpur Highway Project ($620 million). Altogether, ing water). Bihar's share in the Bank's net total commitments is currently 6 per- Bihar was a nonperforming state with limited Bank engagement. cent. Looking ahead, the Bank is developing its strategy to inten- Between fiscal 1998 and 2004, Bihar received support through sify and scale up its engagement with Bihar. Projects currently in two projects--one for district primary education and the other for preparation may include rural roads, local governance, and de- strengthening immunization through a national program. velopment policy lending to support reforms. Future lending will de- The 2004 India CAS proposed a strategy for stronger engage- pend on both the pace of reforms and the state's absorptive ment with lagging states, which paved the way for greater resource capacity. allocation to Bihar. This change in Bank strategy almost coincided Bank engagement in Bihar has taken place with significant with the November 2005 election of a new coalition and political donor collaboration. In November 2006, the Bank, DFID, the ADB, leadership strongly committed to a developmental agenda. and the government of Japan agreed to pilot a strategic partner- The Bank initiated engagement in Bihar by delivering a report ship between donors in Bihar. Since then, the ADB, DFID, and the entitled "Bihar: Towards a Development Strategy" (World Bank Bank have developed a joint strategy for Bihar that was presented 2005a). This document presented a basic development strategy that to the Bihar chief minister in February 2008. rested on two pillars: enhancing Bihar's growth performance by This partnership has enabled the donors to better serve the establishing a healthy investment climate and supporting basic client through reduced transactions costs and to offer the best pos- human resource development through improved quality in basic sible package of support to Bihar based on a rational division of social services. labor and innovative partnership efforts. For example, the DFID- With a strategy in hand, Bank lending to Bihar began to de- World Bank trust fund has played an important role in supporting velop more strongly, which led to the approval of the first Bihar capacity building. An often-cited example of the work supported DPL ($225 million, fiscal 2008), an investment project (Bihar Rural by DFID trust funds in Bihar is the development of a Management Livelihood Project fiscal 2007, $63 million), and technical assis- Information System to track flooding in some districts in Bihar. Sources: World Bank documents and mission interviews. One concern surfaced consistently in conversations four countries are now located in the State authorities often with state authorities in all countries (although it field. The drive to hire native speakers complained of the length is not necessarily specific to the state level only): and train expatriate staff in the language and pace of preparation the lengthy process and slow pace of project spoken in the country (especially in and implementation. preparation and implementation, which was often Brazil and Russia) was highly effective blamed on the Bank's procurement and dis- and appreciated by the authorities. bursement procedures. Creation of parallel struc- tures for project management was also mentioned Staff resources allocated by the Bank for state- as an impediment to more effective use of the level projects seem adequate. Unlike some de- Bank's loans to strengthen institutional capacity at velopment partners, the Bank did not pursue the state level (Nigeria). The issue of the pacing establishment of field offices in any of the states-- of project preparation is well recognized by Bank regardless of engagement intensity or geographic management, and most operational staff in all remoteness (for example, Ceará in Brazil, Andhra 45 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L Pradesh in India, or the Far East region in Russia). India presents an interesting case of potentially ef- Nevertheless, that did not seem to be an imped- fective and efficient joint effort of several donors iment, nor was it mentioned as such by the local at the state level11 in a poor lagging state (box 4.3). (state) counterparts. Also in India, the Bank has jointly worked with the Asian Development Bank (ADB) in coordinating In all four countries, country management used selected projects in infrastructure.12 an informal system of designating state/regional coordinators--staff members who would serve as The Bank's strategy in all the countries normally focal points for information in a region, as well as involved a tripartite arrangement between the those in charge of maintaining instant contact federal government, the Bank (in some cases in with counterparts. This arrangement also proved partnership with other donors), and state gov- to be particularly effective in Brazil and useful, by ernments. The participation level of other parties all accounts, though not of major importance, in was uneven: in Brazil and Nigeria, for example, the the other countries. The authority of these co- nongovernmental sector played only a marginal ordinators was largely moral authority; they lacked role in the design of engagement strategy, while specific portfolio or budget oversight responsibility in Russia several NGOs and private think tanks or accountability. took part in developing regional strategies. Partnership10 The same can be said about the degree of in- The Bank has generally partnered effectively with volvement of state authorities, which varied widely other donors at the state level. In Brazil, the major from one country to another, as well as from state partner is the IDB. A division of labor to state within the same country. In India, the The Bank has partnered has emerged between the two institu- Andhra Pradesh state government started the re- effectively with other tions whereby the IDB works mainly form process, while in Orissa it was the Bank that donors at the state level, with municipal governments and the played the role of initiator until a more commit- usually in a three-way Bank works with state governments. ted chief minister took office. In Nigeria, input from agreement including the In Nigeria the Bank and DFID have a the states at the preparation stage varied from federal government. formal partnership and prepared a joint states that actually initiated the project (Kaduna) CAS in 2004. In India, DFID often com- and actively participated in the design process plements the Bank's loans with grants (Kano and Kwara) to quite limited interest, despite The degree to which state for technical assistance. The recently inputs from various line agencies (Lagos LMDGP). authorities were involved adopted modality of multidonor strate- The last was a quite complicated case because of varied widely. gic engagement in the state of Bihar in the unique and complex set of challenges.13 46 Chapter 5 Village market, Nigeria. Photo by Curt Carnemark, courtesy of the World Bank Photo Library. Summary of Findings T here is little doubt that the Bank has added value at the state level. In the countries studied, there was a great deal of enthusiasm at both the federal and state levels regarding the Bank's contribution. A large number of specific achievements are indicative: ranging from successful fis- cal reform in Orissa to effective technical assistance and capacity building in Lagos, to a wide range of achievements in Ceará, and improved fiscal man- agement in St. Petersburg. IEG therefore concludes that although state-level engagement often requires additional effort and can be resource-intensive, it is usually worth the cost. In all four countries, the Bank tried to engage first · Introduction and institutionalization of a results- with progressive reformist states, aiming to based management approach at the state level. achieve a demonstration effect. The programs in · Progress in reforming public finance manage- most selected reformer states and regions proved ment and introduction of progressive practices successful and brought about many positive of budget planning and execution (elements of changes. But there is little evidence that the in- performance budgeting, multiyear budgeting, tended demonstration effect was achieved (with debt and risk management, and the like). a few small-scale exceptions) on the scale desired · Improvement in local capacity to manage re- by the Bank. Nonetheless, the Bank's engage- gional fiscal resources. Newly introduced ment in some of the poorer, lagging states (such tools--such as an integrated government ac- as Orissa, and, more recently, Bihar) shows that count, evaluation of tax expenditure efficiency, it is possible to achieve results through persistent Public Expenditure Reviews, a subnational fis- work with state-level counterparts and strong cal sustainability tool, and independent audit partnerships with other donors. of government expenditures--frequently be- came everyday practices of state governments. In many states and regions the Bank's program was pivotal in bringing together pro-reform gov- Future Research Agenda ernment officials. It also helped to develop an ex- This study has demonstrated the potential for pert community and a pool of local consulting research into the Bank's development experi- institutions that were drawn into the formula- ence at the subnational level. Additional cross- tion and implementation of public policy (Insti- country thematic evaluations like this one could tute of Economy in Transition, Institute for Urban provide lessons that are applicable on a wider Economics, Center for Fiscal Policy, and Leontief scale. Among the themes and directions that Center in Russia;1 IPECE in Brazil; and the Cen- could be pursued are: ter for Good Governance in India). · Evaluations of new modes of engagement that Among other specific accomplishments are the are often piloted at the state level, such as following: multisector SWAps, reimbursable technical 49 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L Box 5.1: Client Views Federal- and state-level clients broadly considered the Bank's quired by the Bank. The states take full responsibility for the financial state-level engagement useful and timely. But views differed from execution and implementation. country to country on specific aspects, such as modalities of en- In all four cases, federal governments have been supportive gagement and mix of products. of the Bank's strategy of state-level engagement. Most federal In Brazil, for example, both federal and state authorities seemed governments value the discipline associated with Bank lending to to prefer DPLs and multisector SWAps. Among the advantages they the states. In Brazil, for example, the federal government appears noted were flexibility; disbursement speed; and, most important, to put considerable weight on Bank lending as a mechanism for the role of these loans in helping to establish and institutionalize reinforcing compliance with the FRL. policy coordination in the state government. In all four countries, a detailed and structured process at the In Nigeria, federal authorities expressed strong reservations federal level for approving proposals for state borrowing is in about adjustment lending in general, yet a few high-level state of- place.a In Brazil and Russia, the federal governments have seen ficials expressed interest in multisector SWAps, and possibly even borrowing by states and regions as a way of maintaining sub- state-level DPLs. stantial Bank engagement and presence in the country, despite the In all cases, the federal governments have the final say about sizeable resource inflows and rising per capita income level of the whether the Bank will lend to a particular state, because they ap- past decade. prove the operations and provide the sovereign guarantee re- Source: IEG mission interviews. a. While it would be naďve to argue that political considerations do not enter into the state selection process, in practice decisions on the eligibility of states for se- lection 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. assistance, and lending jointly with the IFC approach could be rebalanced toward poor, without sovereign guarantees. low-capacity states that are willing to address · Looking at another set of countries (such as development problems. Engagement with China, Argentina, Mexico, and Pakistan) to pro- better-off states can still be pursued through the vide a comparison with the findings of this study. use of development policy lending and reim- · An in-depth study of specific sector interven- bursable technical assistance. tions (such as health or education) at the sub- · Bank engagement with high-performing states national level, with possible comparison with added value both in strengthening the in-state similar federal-level programs. capacity and in encouraging state-to-state knowledge transfer, albeit mainly between the Findings high performers. But there is little evidence that Based on the cases examined by this assessment, it had the desired demonstration effect on some general findings emerge, which may be poor, lagging states on the scale the Bank helpful in guiding the organization of future work hoped for, or that the Bank had an exit strat- at the state level: egy to permit increased focus on poorer states over time. Nevertheless, experience shows that On selection of states: it is possible to achieve results in the poorest states through persistent work with committed · The strategy to concentrate lending services state counterparts and strong partnership with on few states to enhance the impact of the other donors. Bank's program is the right one in principle, but · It is important to stay engaged, not only in selection criteria and the mode of implemen- states that are able to borrow from the Bank but tation could give more weight to the needs also in poorer states that demonstrate a gen- of the poorest states. The lead/focus state uine commitment to development that can be 50 S U M M A RY O F F I N D I N G S supported through analytic work and techni- On the modalities of engagement: cal assistance--even if there is no fiscal space to lend to them. · There is considerable scope for greater im- pact from knowledge transfer and expanded On the scope of engagement: knowledge services. There is particularly strong demand for better knowledge sharing, both · Continued focus on public finance manage- within the Bank and across the countries con- ment as the core area for state-level work ap- cerned. This is not so much a matter of shar- pears sound, whether engagement is confined ing of concepts and theories as it is of sharing to this area or it serves as an entry point for practical experience about what is working broader engagement. and what is not (for example, multisector in- · For states where the Bank plans or has a major vestment lending in Brazil, reimbursable tech- engagement (such as Orissa, Ceará, and Lagos), nical assistance in Russia, and so on). Similarly, it might consider preparing a brief state strat- government officials in all countries have egy document. The lending programs and Bank little knowledge of approaches being taken budgets in these states are often larger than the elsewhere. programs for many Bank borrowing countries, · The Bank's analytical work at the state level is and the use of a strategy document to focus the of high quality and is appreciated by the au- dialogue and get buy-in from counterparts, thorities. Its relative scarcity, however, could donors, and civil society could contribute to the hamper the effective identification of high- Bank's effectiveness. impact, high-priority areas. There seems to be · Lower-cost lending instruments, such as DPLs, potential for closer partnerships between state multisector SWAps, and reimbursable technical governments and the Bank in this area. There assistance (fee-for-service) seem to be a more are three dimensions to this: (i) widening the natural choice for high-capacity, better-off scope of ESW at the state level; (ii) stronger states, especially in middle-income countries. partnerships with the state governments in This will help to redirect more Bank resources the Bank's ESW; and (iii) wider dissemination to work with poorer and low-capacity states. and better marketing of the Bank's ESW to · The Bank's new instruments (the state-level state officials. More client participation in an- DPL and multisector SWAp) can be especially alytical work will improve the sense of owner- effective at the state level when they address ship, which will increase the chances of linking cross-sector issues (such as public sector re- it with lending and ensure consideration of form). However, the results frameworks un- local conditions. derpinning these instruments need to be · Partner state governments become increas- prepared in a manner commensurate with the ingly exposed to the risks of foreign exchange capacity available at the state level, with tech- rate fluctuation--something they normally are nical assistance carefully designed to mitigate not equipped to handle, which puts their debt related risks and capacity constraints. sustainability at risk. It will be quite useful for · Providing advisory services could be a con- the Bank to explore ways of helping the state venient entry point for the Bank to expand di- governments to develop mechanisms to hedge rect cooperation with state governments. With risks related to foreign exchange fluctuation, as a relatively small price tag, these types of in- well as maturity, interest, and liquidity risks, pos- terventions are highly valued at the state level sibly through provision of technical assistance and are useful for forging closer partnerships. or training. In the eyes of state governments, the Bank · It is critical to have the federal government often has a comparative advantage over private firmly on board through assurance that state- sector providers, because it has access to in- level lending will not only stay within the fed- ternational best practice and has gained cred- eral rules, but will also help reinforce the ibility through previous engagement. implementation of federal laws and programs. 51 Walking over a bridge, India. Photo by Ray Witlin, courtesy of the World Bank Photo Library. APPENDIX A: PORTFOLIO PERFORMANCE To understand the performance of the portfolio, 97 percent of projects rated satisfactory. When dis- this review looked at projects that were approved aggregated, the state-level projects were seen to during or after fiscal 1998 and closed during perform close to the Region's average. In the the evaluation period. The data do not provide evaluation period, only two state-level projects did clear evidence of state-level projects outper- not receive a satisfactory rating,1 while only one forming federal projects (or vice versa), but some of the federal-level projects did not receive a sat- country-specific results do emerge: isfactory rating. This shows that projects in Brazil generally perform well once they are approved. Brazil: Forty-five projects closed in Brazil over the evaluation period; 33 of these were federal proj- India: Thirty-nine projects closed in India over ects and the remaining 12 were state-level proj- the evaluation period; 13 of these were federal ects. At the aggregate, Brazil outperformed the projects and the remaining 26 were state projects. Latin America and Caribbean Region, with over At the aggregate, India outperformed the South Table A.1: Percentage of Projects Rated Satisfactory or Higher for Approval Years Fiscal 1998­2008 Total number Outcome Sustainability evaluated satisfactory (%) likely (%) Latin America and Caribbean Region 304 84.1 84.0 Brazil overall 45 99.3 97.6 Federal 33 93.4 87.8 States 12 99.2 83.3 South Asia Region 115 81.9 79.1 India Overall 39 90.0 87.2 Federal 13 91.2 92.3 States 26 89.2 84.6 Africa Region 308 75.1 69.1 Nigeria overall 4 58.1 75 Federal 1 100.0 100.0 States 3 77.6 66.6 Europe and Central Asia Region 320 87.0 85.5 Russian Federation overall 8 48.9 75.0 Federal 7 22.5 71.4 Region 1 100.0 100.0 Bank-wide 1,309 83.4 79.4 Source: World Bank data as of April, 2009. 53 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L Asia Region, with over 87 percent of projects forms. There is currently another education proj- rated satisfactory. Only five projects did not receive ect that targets only three states and the latest a satisfactory rating; of these, four were in the en- Interim Status Report shows that its performance ergy sector, one at the federal level (coal reha- is satisfactory. This could indicate that state pro- bilitation project), and the other three were at the grams that include a smaller number of states state level (Haryana Power Adaptable Program deliver better results. Loan, Uttar Pradesh Power Sector Reform, and Rajasthan Power Sector Reform).2 This indicates Russia: Of the eight projects that closed in Rus- that although India has generally performed well, sia over the evaluation period, seven were federal it has had persistent problems with the power sec- projects. The only subnational project was the tor, both at the state and the federal level. Kazan municipal project,3 which received a satis- factory rating. Russia's portfolio showed a lower Nigeria: Four projects closed in Nigeria over the percentage of satisfactory ratings (75 percent) evaluation period; one was a federal project and compared with the Europe and Central Asia the remaining three were state programs de- Region (85 percent) mainly because two large signed to roll out to all states. Only one, the uni- federal projects (Structural Adjustment Loan [SAL] versal education project, a state program, received II and II) closed during this time with unsatis- an unsatisfactory rating. This project targeted factory ratings. Both SAL I (fiscal 1998) and SAL 16 of Nigeria's 36 states. One of the reasons this II (fiscal 1999) failed to reach the structural re- project failed to reach its objectives was that there forms planned because Russia was going through was a limited number of states committed to re- a financial crisis during this time. 54 APPENDIX B: KEY FISCAL INDICATORS AND THE LEGAL FRAMEWORK AT THE STATE LEVEL 55 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L Table B.1: Key Fiscal Indicators and the Legal Framework at the State Level Indicator/framework Brazil Indiaa Nigeria Russian Federation Tax allocation 9.3% of GDP (2008) 6.2% of GDP (2007­08) 1.4% of GDP (2007) 9.9% of GDP 25.5% of total tax 36% of total tax revenue revenue (2008) (2007­08) State/regional share of 7.0% of total tax 26.6% of total tax consolidated tax revenue revenue (2007) revenue (as of Jan. 2007) Expenditure allocation 10.7% of GDP (2007) 15.6% of GDP (2007­08) 9.3% of GDP (2007) 15% of GDP State/regional share of 56.6% of total 39.2% of total 29% of total expenditure total expenditure expenditure (2007­08) expenditure (2007) (2008). Federal transfer to states 2.7% of GDP (2006) 5.2% of GDP (2007­08) 7.7% of GDP (2007) 2.3% of GDP (2007) Legal framework Fiscal Responsibility Fiscal Responsibility National Debt Budget code Law Acts of individual Management (i) Budget deficit of Prohibits state states Framework regional government borrowing if: The State acts had three Domestic debt for state cannot exceed 15% focal points: governments: of current revenues (i) the net consolidated (10% for heavily debt exceeds twice (i) zero revenue deficit (i) Federally guaranteed subsidized regions). net current revenue (or surplus) by 2008­09, domestic loans: From (RLC--Receita Liquida time to time the Debt (ii) The net consolidated (ii) maximum 3% of Corrente); Management Office will debt of regional govern- gross state domestic establish limits on ment cannot exceed (ii) new credit operations product as fiscal deficit borrowings with an current revenues (it can exceed 16% of RLC; by 2008­09 and official guarantee. reach 50% for heavily (iii) debt service exceeds (iii) a mid-year report on subsidized regions). (ii) Capital markets: 11.5% of RLC; progress to the state Total amount of loans (iii) Budget deficit of legislature. (iv) spending exceeds outstanding at any municipal government 60 percent of the net The present set of acts particular time shall cannot exceed 10% current revenue; and will be valid until the not exceed 50% of the of current revenues (5% end of 2009­10 only. actual revenue of the for heavily subsidized (v) the debt reduction body concerned for the municipals). schedules set by the preceding 12 months. debt renegotiation (iv) The net consolidated contracts are violated. (iii) Commercial Banks: debt of municipal gov- The monthly debt service ernment cannot exceed ratio of a subnational current revenues (it can should not exceed 40% reach 50% for heavily of its monthly federation subsidized municipals). accounts allocation of the preceding 12 months. Sources: Brazil: Instituto Brasileiro de Planejamento Tributario (IBPT), February 2009; consultant report. India: Ministry of Finance 2008; consultant report. Nigeria: Central Bank of Nigeria various years; consultant report. Russia: www.roskazna.ru--the official site of the Russian Federation Treasury with the RF Ministry of Finance and consultant report. Note: Only the highlights of the legal framework in each country are shown. a. All data for India are based on budget estimates. 56 APPENDIX C: THE GLOBAL FINANCIAL CRISIS AND STATE-LEVEL LENDING The countries included in this study differ in their with October 2007), while federal revenues grew integration into the world markets, with conse- by 1.5 percent. In the period of October 2008 quent variation in their degrees of exposure to the through January 2009, compared with the same calamity caused by the global financial crisis. In four months of the previous year, federal gov- general, all four countries have seen an outflow ernment revenues on the whole decreased by of foreign capital, smaller export volume, falling 10 percent, while regional revenues decreased by commodity prices, and falling revenues. It is the 14 percent. fall in revenue collection that has one of the most significant implications for the federal-state fiscal How will this decline in the growth of revenue af- relationship through transfer of revenues, and fect Bank lending to the states? One would expect consequently the demand for foreign borrowing. state governments to ultimately increase their At the same time, there is a possibility (noted by demand for external loans. Although there have several state-level counterparts to the IEG mission) not been any explicit examples of this happening that growing demand at the federal level will de- yet, there are indications that both the federal and crease the amount of resources potentially avail- state governments are taking measures to facili- able for direct lending to state governments. tate state-level lending. For example, in Brazil several initiatives to amend the Fiscal Responsi- The global meltdown has resulted in a gradual bility Law (FRL) are running in the National Con- erosion of state government revenues. In India, gress. The recent successful debt rescheduling for because of both the recessionary tendencies and Rio Grande do Sul, which ultimately made this tax cuts, growth in tax revenues is estimated to state eligible to borrow from the World Bank, is slow in fiscal 2008­09 and 2009­10. Since a por- one such example. tion of the central tax revenue is shared with states, the total revenues of the states are likely Similarly, the government of India has allowed to suffer significantly. states to amend their Fiscal Responsibility and Budget Management Act (FRBM) target of a 3 In Nigeria, state government revenues are largely percent fiscal deficit to 3.5 percent for 2009­10. dependent on federal transfers, which in turn While this amendment has allowed the states in depend largely on movement in crude oil prices. India an additional borrowing of 0.5 percent of The fall in crude oil prices has resulted in a sub- gross state domestic product (GSDP), they are stantial reduction in federal transfers to states-- careful about borrowing because the exchange by about half of the allocation made in July 2008. rate risk now falls on the state governments be- cause of the back-to-back on-lending system that In the Russia, regional governments were the was introduced in 2004.1 The depreciation of the first to demonstrate reduction of tax proceeds on rupee brought about by the global financial cri- a monthly basis. In October 2008, their revenues sis means that borrowing from the Bank is getting were the first to shrink (by 24 percent compared more and more expensive for the states. 57 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L For now, state governments in Nigeria are partially In Russia, some of parties interviewed by the IEG insulated from the global crisis. The pre-crisis oil mission believed that the new economic situation boom allowed the accumulation of substantial may soon bring about a turn in the federal gov- external reserves--about US$60 billion--with 30 ernment's borrowing policy. The Federal Min- months of import cover. The federal government istry of Finance has been considering the idea of also built up excess crude reserves on behalf of resuming borrowing from the international fi- all tiers of government. So far, the federal gov- nancial institutions. However, it is quite likely ernment has used the excess reserves to augment that the bulk of new borrowing will occur at the the statutory allocations to state governments. federal level. Another hurdle to expanding sub- However, this cushion will not be sustainable if the national lending is that the most creditworthy crisis persists. Federal transfers to the states have regions, such as the Republic of Tatarstan, have already fallen by half compared with the previous the least fiscal space for new borrowing and will year, and the states will require external funding not be in a position to undertake new debt lia- if the crisis continues. bilities in the near future. 58 APPENDIX D: COUNTRIES AT A GLANCE 59 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L Brazil at a Glance 60 A P P E N D I X D : C O U N T R I E S AT A G L A N C E 61 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L India n e India at a Glance 62 A P P E N D I X D : C O U N T R I E S AT A G L A N C E 63 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L Nigeria i Glance c Nigeria at a Glance 64 A P P E N D I X D : C O U N T R I E S AT A G L A N C E 65 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L Russian Federation at a Glance 66 A P P E N D I X D : C O U N T R I E S AT A G L A N C E 67 APPENDIX E: KEY STATE SOCIAL AND ECONOMIC INDICATORS Table E.1: Brazil: Key State Social and Economic Indicators Human Infant Anticipated GDP Population Development Literacy mortality life per capita Area (in (2005, Index rate rate (2007, expectancy (2005, PPP, State million km2) millions) (2005) (2003) out of `000) (2007) in US$) Acre 0.15 0.66 0.75 0.84 30.70 71.40 2,928 Alagoas 0.03 3.02 0.68 0.70 50.00 66.80 2,051 Amapá 0.14 0.59 0.78 0.91 23.90 70.40 3,214 Amazonas 1.57 3.23 0.78 0.94 25.90 71.60 4,516 Bahia 0.56 1.82 0.74 0.79 33.40 72.00 2,880 Ceará 0.15 8.10 0.72 0.78 29.70 70.30 2,211 Distrito Federal 0.01 2.33 0.87 0.96 16.80 75.30 16,502 Espírito Santo 0.05 3.41 0.80 0.90 18.90 73.70 6,059 Goiás 0.34 5.62 0.80 0.90 19.40 73.40 3,935 Maranhăo 0.33 6.10 0.68 0.77 39.20 67.60 1,816 Mato Grosso 0.90 2.80 0.80 0.90 20.40 73.10 5,849 Mato Grosso do Sul 0.36 2.26 0.80 0.91 18.50 73.80 4,182 Minas Gerais 0.59 19.24 0.80 0.89 20.40 74.60 4,381 Pará 1.25 6.97 0.76 0.90 24.40 72.00 2,458 Paraíba 0.06 3.60 0.72 0.75 38.00 69.00 2,052 Paraná 0.20 1.26 0.82 0.93 18.60 74.10 5,400 Pernambuco 0.10 8.41 0.72 0.79 38.40 68.30 2,595 Piauí 0.25 3.01 0.70 0.72 28.20 68.90 1,619 Rio de Janeiro 0.04 15.38 0.83 0.96 19.60 73.10 7,024 Rio Grande do Norte 0.05 3.00 0.74 0.77 34.80 70.40 2,603 Rio Grande do Sul 0.28 10.85 0.83 0.95 13.50 75.00 5,824 Rondônia 0.24 1.53 0.78 0.92 23.70 71.20 3,679 Roraima 0.22 0.39 0.75 0.91 19.10 69.90 3,554 Santa Catarina 0.10 5.87 0.84 0.95 16.10 75.30 6,362 Săo Paulo 0.25 40.44 0.83 0.95 15.50 74.20 7,867 Sergipe 0.02 1.97 0.74 0.90 33.80 70.90 2,985 Tocantins 0.28 1.31 0.76 0.83 27.30 3,044 Sources: The Brazilian Institute for Geography and Statistics (IBGE), Ministry of Planning; World Bank data. Note: PPP = purchasing power parity. 69 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L Table E.2: India: Key State Social and Economic Indicators Per capita Per capita net state net state Literacy domestic product domestic product Density Literacy rate (2004­05, (2004­05, Area Population (per sq. rate (2004­05 current price, current price, State (`00 sq. km.) (`000) km.) (2001 census) census) in rupees) in US$) Andhra Pradesh 275 76,210 277 61 53 23,153 526.2 Arunachal Pradesh 84 1,098 13 54 60 19,724 448.3 Assam 78 26,656 340 63 71 13,633 309.8 Bihar 94 82,999 881 47 46 5,772 131.2 Chhattisgarh 135 20,834 154 65 57 13,013 295.8 Goa 4 1,348 364 82 75 58,184 1,322.4 Gujarat 196 50,671 258 69 66 28,355 644.4 Haryana 44 21,145 478 68 62 32,712 743.5 Himachal Pradesh 56 6,078 109 77 72 27,486 624.7 Jammu and Kashmir 222 10,144 100 56 61 16,190 368.0 Jharkhand 80 26,946 338 54 52 23,945 544.2 Karnataka 192 52,851 276 67 61 27,048 614.7 Kerala 39 31,841 819 91 83 14,069 319.8 Madhya Pradesh 308 60,348 196 64 54 15,073 342.6 Maharashtra 308 96,879 315 77 70 32,170 731.1 Manipur 22 2,294 103 71 76 14,901 338.7 Meghalaya 22 2,319 103 63 78 19,572 444.8 Mizoram 21 888 42 89 90 Nagaland 17 1,990 120 67 78 Orissa 156 36,804 236 63 57 13,601 309.1 Punjab 50 24,359 484 70 68 30,701 697.8 Rajasthan 342 56,507 165 60 50 16,212 368.5 Sikkim 7 541 76 69 75 24,115 548.1 Tamil Nadu 130 62,406 480 74 70 25,965 590.1 Tripura 10 3,199 305 73 74 Uttar Pradesh 241 166,198 690 72 64 11,477 260.8 Uttaranchal Pradesh 53 8,489 159 56 51 West Bengal 89 80,176 903 69 66 22,497 511.3 Union Territory A.& N. Islands 8 356 43 81 77 Chandigarh 0.1 901 7,900 82 84 67,370 1,531.1 D. & Nagar Haveli 0.5 220 449 58 58 Daman & Diu 0.1 158 1,413 78 76 Delhi 1.5 13,851 9,340 82 80 53,976 1,226.7 Lakshadweep 0.03 61 1,895 87 78 Pondicherry 0.5 974 2,030 81 74 56,034 1,273.5 All India 3,287 1,028,737 325 65 60 22,946 521.5 Source: Economic Survey, 2007­08, Office of the Registrar General; State Statistical Bureaus and Central Statistical Organization, government of India. 70 A P P E N D I X E : K E Y S TAT E S O C I A L A N D E C O N O M I C I N D I C AT O R S Table E.3: Nigeria: Key State Economic and Social Indicators GDP Access to Net (2005) GDP primary primary Adult Youth Population billions per capita State Zone school enrolment literacy literacy (millions) of naira (US$) 1 Abita South East 78.9 82.6 79.2 94.2 2.83 234.6 627.0 2 Adamawa North East 76.1 64.0 56.1 71.3 3.17 40.2 96.0 3 Akwa Ibom South South 56.8 78.5 81.6 92.7 3.93 1,129.4 2,178.2 4 Anambra South East 65.1 85.1 77.8 94.0 4.18 171.9 3,11.4 5 Bauchi North East 68.7 40.1 39.5 49.0 4.68 204.3 3,31.0 6 Bayelsa South South 88.0 72.2 64.3 73.3 1.70 1,419.9 6,315.2 7 Benue North Central 63.2 77.4 67.0 83.7 4.22 438.0 786.5 8 Borno North East 72.6 34.3 27.4 36.0 4.15 176.7 322.5 9 Cross River South South 72.4 77.0 75.5 90.7 2.89 228.7 599.6 10 Delta South South 71.9 78.1 72.9 86.4 4.10 1,453.9 2,687.4 11 Ebonyi South East 32.7 75.0 57.7 85.9 2.17 126.8 442.0 12 Edo South South 79.7 76.8 77.0 91.6 3.22 226.4 532.8 13 Ekiti South West 89.0 89.1 75.0 98.6 2.38 142.0 451.2 14 Enugu South East 62.7 79.9 75.6 96.2 3.26 129.9 302.0 15 Gombe North East 82.6 33.4 54.3 53.1 2.35 114.1 367.2 16 Imo South East 52.3 84.4 75.4 92.7 2.93 280.8 724.7 17 Jigawa North West 72.9 29.6 38.7 40.9 4.35 103.6 180.5 18 Kaduna North West 84.2 66.1 66.4 76.8 6.07 364.2 454.7 19 Kano North West 76.5 47.8 57.5 62.6 388.9 20 Katsina North West 74.0 45.1 36.5 45.0 5.79 167.4 218.9 21 Kebbi North West 72.2 32.9 51.1 61.1 3.24 100.1 234.0 22 Kogi North Central 88.1 80.5 64.5 89.7 3.28 297.1 686.5 23 Kwara North Central 83.8 79.8 55.6 78.0 2.37 163.6 522.8 24 Lagos South West 93.9 81.8 89.9 97.5 9.01 1,701.0 1,429.6 25 Nassarawa North Central 79.6 66.5 53.7 68.0 1.86 90.4 367.6 26 Niger North Central 88.6 57.5 36.5 53.8 3.95 204.6 392.4 27 Ogun South West 85.3 83.6 69.6 93.2 3.73 131.6 267.4 28 Ondo South West 86.2 84.5 76.6 97.4 3.44 604.1 1,330.0 29 Osun South West 86.7 84.1 74.8 97.5 3.42 178.0 393.9 30 Oyo South West 84.5 77.1 73.3 94.7 5.59 467.2 632.9 31 Plateau North Central 74.9 79.3 61.6 77.0 3.18 145.5 346.7 32 Rivers South South 70.1 75.9 82.6 93.9 5.19 2,125.8 3,105.8 33 Sokoto North West 80.5 32.1 70.3 76.7 3.70 95.7 196.0 34 Taraba North East 70.2 59.0 55.7 65.2 2.30 35.8 117.7 35 Yobe North East 62.2 35.5 25.3 37.6 2.32 60.6 197.7 36 Zamfara North West 64.7 26.1 53.4 59.5 3.26 109.6 2,54.7 FCT North Central 94.9 83.4 79.0 82.8 1.41 395.4 2,131.9 North East 71.9 43.7 42.2 52.5 North Central 79.7 72.5 58.3 77.3 North West 76.4 42.2 54.5 63.3 South East 60.6 81.6 75.7 94.1 South South 71.7 76.8 78.0 91.3 South West 88.0 82.3 79.2 96.5 National 75.9 61.5 65.7 80.2 Source: National Bureau of Statistics, 2006 Core Welfare Indicator Questionnaire (CWIQ) Survey and Economic Associates (an economic consulting firm) for state GDP data. 71 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L Table E.4: Russian Federation: Key State Economic and Social Indicators Infant Population with deaths Life money incomes GDP, US under 1 expectancy Number below subsistence Area, $/per year old at birth of physicians minimum as 1,000 capita, Population per 1,000 (number per 10,000 percentage of the Constituent regions sq. km 2006 (1,000) 2006 live births of years) population total population Central Federal Okrug 1 Belgorod Oblast 27.1 4,545 1,512 8.6 69.27 40.7 15.8 2 Bryansk Oblast 34.9 2,350 1,346 7.8 65.3 36.7 19.2 3 Vladimir Oblast 29.0 2,899 1,487 7.8 64.44 34.4 29.6 4 Voronezh Oblast 52.4 2,691 2,334 8.2 67.11 54.3 19.6 5 Ivanovo Oblast 23.9 1,821 1,115 9.1 64.49 52 35.2 6 Kaluga Oblast 29.9 3,183 1,022 10.8 66.03 39.4 17.2 7 Kostroma Oblast 60.1 2,854 717 14.5 64.62 36.8 20.0 8 Kursk Oblast 29.8 3,241 1,199 9.9 66.06 50.4 15.7 9 Lipetsk Oblast 24.1 6,056 1.19 8.1 66.73 41.9 11.3 10 Moscow Oblast 46.0 5,370 6.63 8.6 66.4 37.4 12.6 11 Oryol Oblast 24.7 2,857 842 10.1 66.39 39 21.6 12 Ryazan Oblast 39.6 3,329 1,195 11.3 65.23 54.6 19.4 13 Smolensk Oblast 49.8 3,010 1,019 10.3 63.01 60.8 19.7 14 Tambov Oblast 34.3 2,652 1,145 9.7 66.84 35 15.1 15 Tver Oblast 84.1 3,410 1,425 10.5 62.85 53 14 16 Tula Oblast 25.7 3,423 1,622 8.9 64.23 35 15.1 17 Yaroslavl Oblast 36.4 4,488 1,339 8.2 66.11 58.8 13.8 18 City of Moscow 1.0 18,73 10,407 7.9 71.81 78.6 13.5 North-West Federal Okrug 19 Republic of Karelia 172.4 4,719 703 7.6 63.79 49.3 15.7 20 Republic of Komi 415.9 8,214 996 7 64.21 46.1 15.4 21 Arkhangelsk Oblast 410.7 6,097 1,263 10.2 64.84 53.2 17.6 22 Nenets AO 176.7 -- 42 15.2 62.24 39.3 8.83 23 Vologda Oblast 145.7 6,410 1,245 8.6 65.36 35.5 17.5 24 Kaliningrad Oblast 15.1 4,042 945 7.1 64.13 36.9 14.6 25 Leningrad Oblast 85.3 6,143 1,653 7.9 63.06 31.2 14.7 26 Murmansk Oblast 144.9 6,893 873 10.3 65.17 48.3 18.7 27 Novgorod Oblast 55.3 4,203 674 11.4 62.66 40.4 17.8 28 Pskov Oblast 55.3 2,610 737 13.2 61.22 34.4 18.5 29 City of St. Petersburg 0.6 6,737 4.6 4.7 68.9 83.5 9.7 South Federal Okrug 30 Republic of Adygeya 7.6 1,799 445 8 68.27 38.4 34.3 31 Republic of Dagestan 50.3 1,695 2,622 14.8 73.35 40 11.9 32 Ingush Republic 4.3 0.664 482 31.4 76.02 23.4 57.4 33 Kabarda-Balkar Republic 12.5 1,778 897 16.1 70.14 44.2 19.8 72 A P P E N D I X E : K E Y S TAT E S O C I A L A N D E C O N O M I C I N D I C AT O R S Table E.4: Russian Federation: Key State Economic and Social Indicators (continued) Infant Population with deaths Life money incomes GDP, US under 1 expectancy Number below subsistence Area, $/per year old at birth of physicians minimum as 1,000 capita, Population per 1,000 (number per 10,000 percentage of the Constituent regions sq. km 2006 (1,000) 2006 live births of years) population total population 34 Republic of Kalmykia 76.1 1,623 290 14.2 67.52 50.1 48.7 35 Karachai-Circassian Republic 14.1 2,002 435 10.7 70.19 36.4 19.7 36 Republic of North Ossetia 8.0 2,316 704 9.8 70.74 67.4 16.3 37 Republic of Chechnya 15.0 0.946 1,141 16.7 73.08 20.8 ... 38 Krasnodar Krai 76.0 3,470 5.1 8.2 68.74 43.1 22.9 39 Stavropol Krai 66.5 2,509 2,718 10.1 68.25 45.2 21.6 40 Astrakhan Oblast 44.1 3,235 998 10.7 66.14 67.1 17 41 Volgograd Oblast 113.9 3,601 2,655 11.2 67.84 49.4 11.2 42 Rostov Oblast 100.8 2,975 4,334 13.2 67.61 38.7 18.1 Privolgskiy Federal Okrug 43 Republic of Bashkortostan 143.6 4,734 4,079 10.9 67.47 42.7 14.7 44 Republic of Mari El 23.2 2,277 717 11.1 64.82 34.5 30.7 45 Republic of Mordovia 26.2 2,528 866 6.8 67.75 51.7 29.0 46 Republic of Tatarstan 68.0 6,115 3.769 8.2 69.04 45.3 10 47 Republic of Udmurtia 42.1 4,017 1,553 10.8 66.01 58.3 19.7 48 Chuvash Republic 18.3 2,706 1,299 9.1 66.98 48.1 22.0 49 Kirov Oblast 120.8 2,552 1,461 9.9 65.8 45.7 23.9 50 Nizhny Novgorod Oblast 74.8 4,260 3,445 11.5 64.6 47 16.4 51 Orenburg Oblast 124.0 5,339 2.15 9.9 66.17 49.1 19.4 52 Penza Oblast 43.2 2,467 1,423 10.6 67.25 38.7 23.8 53 Perm krai 160.6 5,452 2.77 11.6 63.99 53.7 14.7 54 Samara Oblast 53.6 5,847 3,201 7.3 66.57 49.9 17.2 55 Saratov Oblast 100.2 2,930 2,626 9.1 67.37 52 22.3 56 Ulianovsk Oblast 37.3 2,896 1,351 10.1 66.33 36.4 24.9 Ural Federal Okrug 57 Kurgan Oblast 71.0 2,590 992 14.2 65.52 27.7 24.3 58 Sverdlovsk Oblast 194.8 5,648 4,428 8.7 66.47 42.5 12.2 59 Tyumen Oblast 161.8 9,441 1,316 8.4 67.95 49.7 11.5 60 Khanty-Mansi AO 523.1 33,408 1,469 7.5 68.84 50.7 7.93 61 Yamal-Nenets AO 750.3 29,183 523 13 68.86 49 7.03 62 Chelyabinsk Oblast 87.9 4,768 3,551 9.1 66.17 41.1 12.4 Sibir Federal Okrug 63 Republic of Altai 92.6 2,066 204 15.3 62.49 39.7 37.8 64 Republic of Buryatia 351.3 3,576 969 12.5 62.43 39.3 29,7 (Table continues on next page) 73 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L Table E.4: Russian Federation: Key State Economic and Social Indicators (continued) Infant Population with deaths Life money incomes GDP, US under 1 expectancy Number below subsistence Area, $/per year old at birth of physicians minimum as 1,000 capita, Population per 1,000 (number per 10,000 percentage of the Constituent regions sq. km 2006 (1,000) 2006 live births of years) population total population 65 Republic of Tyva 170.5 1,822 308 15.1 58.43 42.9 38.7 66 Republic of Khakassia 61.9 3,606 541 15.6 64.51 38.8 25.1 67 Altai Krai 169.1 2,517 2,565 11.2 66.64 46.8 20.1 68 Krasnoyarsk Krai 2,339.7 7,673 2,925 12.9 65.58 49.6 19.3 69 Irkutsk Oblast 767.9 4,872 2,545 11.8 63.06 46.2 19.6 70 Kemerovo Oblast 95.5 4,524 2,855 10.3 63.04 47 11.7 71 Novosibirsk Oblast 178.2 4,119 2,662 9.6 66.38 58.8 21.9 72 Omsk Oblast 139.7 4,631 2,047 8.8 66.17 55.9 16 73 Tomsk Oblast 316.9 6,853 1,037 13.8 66.5 67.6 14.5 74 Zabajkalsk Krai 431.5 2,958 1,136 10.2 61.43 54.9 24.5 Far-East Federal Okrug 75 Republic of Sakha (Yakutia) 3,103.2 8,229 951 10.6 65.55 54 20.3 76 Primorski Krai 165.9 3,941 2,036 10.7 64.4 52.9 23.8 77 Khabarovsk Krai 788.6 5,289 1.42 12.3 63.67 59 18.2 75 Amur Oblast 363.7 3,949 887 17.4 62.23 60.2 31.1 79 Kamchatka Krai 472.3 352 12.2 65.19 52.5 27.1 80 Magadan Oblast 461.4 6,670 175 14.2 63.4 56.2 19.0 81 Sakhalin Oblast 87.1 11,794 532 12.9 62.79 45.8 16.2 82 Jewish AO 36.0 3,647 189 14.3 61.27 36.4 25.4 83 Chukotka AO 737.7 11,008 51 23.2 58.93 81.6 13.1 Sources: Federal State Statistics Service of Russian Federation; World Bank data. 74 APPENDIX F: DISTRIBUTION OF PROJECTS BY STATES 75 76 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L Table F.1: Brazil: Distribution of Projects by States Sector Agriculture Economic policy and rural governance, Urban State development and finance Education Health Environment Transportation development Water Summary Acre Total projects = 0 Alagoas Total projects = 0 Amapa Amapa Sustainable Total projects = 1 Communities Environment = 1 (fiscal 2005) Amazonas Alto Solimoes Basic Total projects = 1 Services and Health = 1 Sustainable Devel- opment Project I (fiscal 2008) Bahia Rural Poverty Bahia, Basic Bahia, State Health Bahia Integrated Bahia Poor Bahia Water Total projects = 9 Alleviation Project, Education Project System Reform State Highway Urban Areas Resources Agriculture = 3; Bahia (approved Phase I (fiscal 2001) (fiscal 2003) Management Integrated Management Education = 2; fiscal 1995; exited (fiscal 2007) Development (fiscal 1998) Health = 1; fiscal 2001) Bahia Education (fiscal 2006) Transport = 1; Project (APL), Urban Bahia, Rural Poverty Second Phase development = 1; Reduction Project (fiscal 2001) Water = 1 (fiscal 2001) Bahia, State Integrated Project: Rural Poverty (fiscal 2006) Ceará Ceará, Rural Poverty Ceará Multisector CEARÁ Basic Ceará Urban Ceará Integrated Total projects = 8 Reduction Project Social Inclusion Education Development & Water Resources Agriculture = 3 +1 (fiscal 2001) Development APL Quality Project Water Resource Management (additional (fiscal 2006) (fiscal 2001) (approved fiscal Project financing); Ceará Water Pilot 1995, exited (fiscal 2000) Economic policy & Project (approved fiscal 2004) governance = 1; fiscal 1997; exited Education = 1 ; Urban fiscal 2002) development = 1; Water = 1 Rural Poverty Alleviation Project, Ceará (approved fiscal 1995; exited fiscal 2001) Ceará, Rural Poverty Reduction-- additional financing (fiscal 2006) Distrito Total projects = 0 Federal A P P E N D I X F : D I S T R I B U T I O N O F P R O J E C T S B Y S T AT E S Espírito Espírito Santo Total projects = 2 Santo Water and Coastal Water = 2 Pollution (approved fiscal 1994, exited fiscal 2003) Espirito Santo Water and Coastal Pollution Management (fiscal 2005) Goiás Goias State Total projects = 1 Highway Transport = 1 Management (fiscal 2002) (Table continues on next page) 77 78 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L Table F.1: Brazil: Distribution of Projects by States (continued) Sector Agriculture Economic policy and rural governance, Urban State development and finance Education Health Environment Transportation development Water Summary Maranhăo Maranhăo, Rural Second State Total projects = 3 Poverty Alleviation Highway Agriculture = 2; Project Management Transport = 1 Maranhăo-- Project--States of Integrated Program: Maranhao, Piaui, Rural Poverty and Tocantins Reduction (approved fiscal (fiscal 2004) 1994, exited fiscal 2000) Mato Grosso Total projects = 0 Mato Mato Grosso State Total projects = 1 Grosso do Privatization Project Finance = 1 Sul (approved fiscal 1997, exited fiscal 2002) Minas Minas Gerais Rural Minas Gerais State Minas Gerais Basic Minas Gerais Minas Gerais Total projects = 7 Gerais Poverty Reduction Privatization (fiscal Education (approved Municipal Water Quality and Agriculture = 1; (fiscal 2006) 1998, cancelled fiscal 1994, exited Management and Pollution Control Economic policy & after Board fiscal 2002) Environmental Project (approved governance = 3; approval) Infrastructure fiscal 1993, exited Education = 1; (approved fiscal fiscal 2000) Urban Minas Gerais 1994, exited development = 1; Partnership for fiscal 2002) Water = 1 Development (fiscal 2006, DPL) Minas Gerais Development Partnership II-- SWAp (fiscal 2008) Pará Pará Integrated Total projects = 1 Rural Environment = 1 Development-- APL (fiscal 2007) Paraíba Paraíba--Rural Total projects = 1 Poverty Alleviation Agriculture = 1 Project Paraná PARANÁ--Rural Parana Basic Total projects = 2 Poverty Alleviation Education (approved Agriculture = 1; and Natural fiscal 1994, exited Education = 1 Resources fiscal 2002) Management Project (approved fiscal 1996, exited fiscal 2006) Pernambuco Rural Poverty Pernambuco Recife Urban Total projects = 5 Alleviation Project, Integrated Upgrading Project Agriculture = 2 +1 Pernambuco Development: (fiscal 2003) (additional (approved fiscal Education Quality financing); 1997, exited Improvement Project Education = 1; A P P E N D I X F : D I S T R I B U T I O N O F P R O J E C T S B Y S T AT E S fiscal 2002) (fiscal 2008) Urban development = 1 Pernambuco--Rural Poverty Reduction Project (fiscal 2001) Pernambuco Rural Poverty Reduction-- additional financing (fiscal 2007) Piauí Rural Poverty Second State Total projects = 4 Alleviation Project, Highway Agriculture = 2 +1 Piauí (approved Management (additional fiscal 1997, exited Project, States of financing); fiscal 2002) Maranhao, Piaui, and Transport = 1 Tocantins (approved fiscal 1994, exited fiscal 2000) (Table continues on next page) 79 80 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L Table F.1: Brazil: Distribution of Projects by States (continued) Sector Agriculture Economic policy and rural governance, Urban State development and finance Education Health Environment Transportation development Water Summary Paiui Piauí, Rural Poverty (continued) Reduction Project (fiscal 2001) Piauí Rural Poverty Reduction-- additional financing (fiscal 2006) Rio de Rio de Janeiro Rio de Janeiro Total projects = 3 Janeiro State Reform­ Mass Transport Finance = 1; Privatization (fiscal 1998) Transport = 2 (fiscal 1998) Rio de Janeiro Mass Transport-- additional financing (fiscal 2008) Rio Grande Rural Poverty Rio Grande do Total projects = 3 do Norte Alleviation Project, Norte-- Integrated Agriculture = 1 +1 Rio Grande do Norte Water Resources (additional financing); (approved fiscal Management Water = 1 1997, exited (fiscal 2008) fiscal 2002) Rio Grande do Norte Rural Poverty Reduction-- Additional Financing (fiscal 2002) Rio Grande Natural Resources Rio Grande do Sul Rio Grande do Sul Total projects = 3 do Sul Management and State Reform Project State Highway Agriculture = 1; Rural Poverty (approved fiscal Management Finance = 1; Alleviation Project, 1997, exited Project (approved Transport = 1 Rio Grande do Sul fiscal 1999) 1997, exited (approved fiscal fiscal 2006) 1997, exited fiscal 2006) Rondônia Total projects = 0 Roraima Total projects = 0 Santa Santa Catarina-- 2nd Land Santa Catarina Total projects = 3 Catarina Natural Resources Management State Highway Agriculture = 1; Management and Project--Santa Management Environment = 1; Rural Poverty Catarina (approved Project (approved Transport. = 1 Reduction fiscal 1990, exited fiscal 1993, exited (fiscal 2002) fiscal 1999) fiscal 2000) Săo Paulo Innovations in Basic Sao Paulo Total projects = 5 Education Project, Integrated Urban Education = 1; Sao Paulo (approved Transport Project Transport = 3+1 fiscal 1991, exited (fiscal 1998) (additional financing) A P P E N D I X F : D I S T R I B U T I O N O F P R O J E C T S B Y S T AT E S fiscal 1999) Sao Paulo Metro Line 4 (fiscal 1998) Sao Paulo Metro Line 4--additional financing (fiscal 2008) Sao Paulo Trains and Signaling (fiscal 2008) Sergipe Rural Poverty Total projects = 2 Alleviation Project, Agriculture = 2 Sergipe (approved fiscal 1995, exited fiscal 2001) (Table continues on next page) 81 82 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L Table F.1: Brazil: Distribution of Projects by States (continued) Sector Agriculture Economic policy and rural governance, Urban State development and finance Education Health Environment Transportation development Water Summary Sergipe Sergipe, Rural (continued) Poverty Reduction (fiscal 2002) Tocantins Second State Total projects = 2 Highway Transport = 2 Management Project, States of Maranhao, Piaui, and Tocantins (approved fiscal 1994, exited fiscal 2000) Tocantins Sustainable Regional Development (fiscal 2004) Table F.2: India: Distribution of Projects by States Sector Agriculture Economic policy and rural governance, Energy Urban State development and finance Education Health and mining Transportation development Water Summary Andaman & Total projects =0 Nicobar Andhra Andhra Pradesh Andhra Pradesh Andhra Pradesh Andhra Pradesh Total projects =11 Pradesh District Poverty Economic Reform Economic Power Sector Agriculture & Initiatives Project Loan/Credit Restructuring Restructuring rural = 5 +1 (fiscal 2000) (fiscal 2002) (fiscal 1998) Project (fiscal 1999, (additional financing); APL I) Economic policy = 3; Andhra Pradesh Second Andhra Health = 1; Rural Poverty Pradesh Economic Energy & mining = 1 Reduction Project Reform Loan (fiscal 2003) (fiscal 2004, SAL II) Andhra Pradesh Andhra Pradesh Economic Reform A P P E N D I X F : D I S T R I B U T I O N O F P R O J E C T S B Y S T AT E S Rural Poverty Project (fiscal 2007, Reduction DPL III) additional financing (fiscal 2008) Andhra Pradesh Community Forest Management (fiscal 2003) Chhattisgarh District Rural Poverty Project (fiscal 2003, DRPP) Andhra Pradesh Community Tank Management Project (fiscal 2007) (Table continues on next page) 83 84 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L Table F.2: India: Distribution of Projects by States (continued) Sector Agriculture Economic policy and rural governance, Energy Urban State development and finance Education Health and mining Transportation development Water Summary Arunchal Total projects = 0 Pradesh Assam Total projects = 1 Agriculture & rural = 1 Bihar Bihar Rural Bihar DPL District Primary Total projects = 3 Livelihoods Project (fiscal 2008) Education Project III Agriculture & (fiscal 2007) (fiscal 1998, Bihar rural = 1; and Jharkhand) Economic policy = 1; Education = 1 Chandigarh Total projects = 0 (Union Territory) Chhattisgarh Total projects = 0 Dadra and Total projects = 0 Nagar Haveli (Union Territory) Daman and Total projects = 0 Diu (Union Territory) Delhi (Union Total projects = 0 Territory) Goa Total projects = 0 Gujarat Gujarat Highways Gujarat Emergency Total projects = 2 (fiscal 2001) Earthquake Transport = 1; Reconstruct Urban (fiscal 2002) development = 1 Haryana Haryana Power Haryana Power Total projects = 2 Sector Restructuring Sector Restructuring Agriculture = 1; APL-I APL-I Energy & mining = 1 Himachal Mid-Himalayan (HP) Himachal Pradesh Himachal Pradesh Total projects = 3 Pradesh Watersheds DPL I (fiscal 2008) State Roads Project Agriculture & (fiscal 2006) (fiscal 07) rural = 1; Economic policy = 1; Transport = 1 Jammu and Total projects = 0 Kashmir Jharkhand Total projects = 0 Karnataka Karnataka Technical assistance Karnataka Health Karnataka Highways Karnataka UWS Total projects = 10 Watershed for Economic Systems (fiscal 2001) Improvement Agriculture & Development Reform Project (fiscal 2007) Project rural = 4; (fiscal 2001) (fiscal 2000) (fiscal 2004) Economic policy = 3; Health = 1; Karnataka First Karnataka Transport = 1; Community Economic Water = 1 Based Tank Restructuring Management (fiscal 2001, SAL I) A P P E N D I X F : D I S T R I B U T I O N O F P R O J E C T S B Y S T AT E S (fiscal 2002) Second Karnataka Karnataka Economic Panchayats Restructuring Strengthening Project (fiscal 2002, Project SAL II) (fiscal 2005) Karnataka Tanks (fiscal 2008, supplement) Kerala Kerala Forestry Kerala State Kerala Rural Total projects = 4 (fiscal 1998­2004) Transport Water Supply and Agriculture & rural = 1; (fiscal 2002) Environmental Transport = 1; Sanitation Water = 2 (fiscal 2001) (Table continues on next page) 85 86 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L Table F.2: India: Distribution of Projects by States (continued) Sector Agriculture Economic policy and rural governance, Energy Urban State development and finance Education Health and mining Transportation development Water Summary Second Karnataka Rural Water Supply and Sanitation Project (fiscal 2002) Lakshad- Total projects = 0 weep (Union Territory) Madhya Madhya Pradesh Total projects = 2; Pradesh District Poverty Agriculture & Initiatives Project rural = 2 (fiscal 2001, DPIP I) Madhya Pradesh Water Sector Restructuring (fiscal 2005) Maharashtra Maharashtra Water Maharashtra Health Maharashtra Rural Total projects = 3 Sector Improvement Systems Water Supply and Agriculture & (fiscal 2005) Development Sanitation rural = 1; (fiscal 1999) "Jalswarajya" Health = 1; Project Water = 1 (fiscal 2004) Meghalaya Total projects = 0 Mizoram Mizoram State oads Total projects = 2 Project (fiscal 2002) Transport = 1+1 (additional financing) Mizoram Roads-- Additional Financing (fiscal 2007) Nagaland Total projects = 0 Orissa Orissa Socio- Orissa Health Total projects = 3 Economic Systems Economic policy = 2; Development Loan (fiscal 1998) Health = 1 (fiscal 2005, SAL I) Orissa Socio- Economic Development Loan II (fiscal 2007) Puducherry Total projects = 0 (Union Territory) Punjab Punjab State Roads Punjab Rural Total projects = 2 Project (fiscal 2007) Water Supply & Transport = 1; Sanitation Water = 1 (fiscal 2007) Rajasthan Rajasthan District Rajasthan District Rajasthan Health Rajasthan Power I Total projects = 6 Poverty Initiatives Primary Education Systems (fiscal 2001) Agriculture & Project (fiscal 2000) Project (fiscal 1999, Development Project rural = 2; A P P E N D I X F : D I S T R I B U T I O N O F P R O J E C T S B Y S T AT E S DPEP) (fiscal 2004) Education = 2; Rajasthan Water Health = 1; Sector Restructuring Rajasthan Second Energy & mining = 1 Project (fiscal 2002, District Primary RWSRP) Education (fiscal 2001, DPEP II) Sikkim Total projects = 0 Tamil Nadu Tamil Nadu Tamil Nadu Health Tamil Nadu Road Tamil Nadu Urban Total projects = 6 Empowerment and Systems Project Sector Project Development II Agriculture & Poverty Reduction (fiscal 2005) (fiscal 2003) (fiscal 1999) rural = 2; (fiscal 2006) Health = 1; Tamil Nadu Urban Transport = 1; Tamil Nadu Irrigated Development III Urban Agriculture Mod. (fiscal 2006) development = 2 and Water-Bodies Rest and Management (fiscal 2007) (Table continues on next page) 87 88 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L Table F.2: India: Distribution of Projects by States (continued) Sector Agriculture Economic policy and rural governance, Energy Urban State development and finance Education Health and mining Transportation development Water Summary Tripura Total projects = 0 Uttaranchal Uttaranchal Total projects = 1 Decentralized Agriculture & rural = 1 Watershed Development (fiscal 2004) Uttar Pradesh Uttar Pradesh Technical assistance Uttar Pradesh Basic UP Health Systems Uttar Pradesh Uttar Pradesh State Uttaranchal Total projects = 12 Forestry for Economic Education Project II Development Power Sector Roads Project Rural Water Agriculture & (fiscal 1998­2004) Reform Project (fiscal 1998) Project Restructuring (fiscal 2003) Supply and rural = 4; (fiscal 2000) (fiscal 2000) Project Sanitation Economic policy = 2; (fiscal 2000) (fiscal 2007) Education = 2; Uttar Pradesh Sodic Uttar Pradesh Fiscal Uttar Pradesh Third Health = 1; Lands Reclamation II Reform and Public District Primary Energy & (fiscal 1999­2008) Sector Restructuring Education mining = 1, Program (fiscal 2000, Transport = 1; Uttar Pradesh Water (fiscal 2008) DPEP III) Water = 1 Sector Restructuring Project (fiscal 2002) Uttar Pradesh Diversified Agricultural Support Project (fiscal 1998­2004) West Bengal Total projects = 0 Table F.3: Nigeria: Distribution of Projects by States Community- Number of driven Transport, Urban State projects Health Education HIV/AIDS development Water Agriculture power development Governance Abia 5 Health System HIV/AIDS Community-Based Fadama 3 Development 2 Poverty Reduction, Community Social Development Adamawa 5 Health System HIV/AIDS Local Empowerment Fadama 2, Development 2 and Environment Fadama 3 Management Akwa Ibom 5 Health System HIV/AIDS Fadama 3 Community- Based Development 2, Urban Development Malaria A P P E N D I X F : D I S T R I B U T I O N O F P R O J E C T S B Y S T AT E S Anambra 3 Health System HIV/AIDS Development 2, Malaria Bauchi 8 Health System HIV/AIDS Local Empowerment Fadama 2, Community- Based State Governance Development 2, and Environment Fadama 3 Urban Development and Capacity Malaria Management (WB) Building Bayelsa 4 Health System HIV/AIDS Local Empowerment Fadama 3 Development 2 and Environment Management Benue 4 Health System HIV/AIDS Local Empowerment Fadama 3 Development 2 and Environment Management Borno 4 Health System HIV/AIDS Fadama 2, Development. 2 Fadama 3 Cross River 6 Health System HIV/AIDS Community-Based Urban Fadama 3 State Governance Development. 2 Poverty Reduction Water 2 and Capacity Building Delta 3 Health System HIV/AIDS Fadama 3 Development 2 (Table continues on next page) 89 90 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L Table F.3: Nigeria: Distribution of Projects by States (continued) Community- Number of driven Transport, Urban State projects Health Education HIV/AIDS development Water Agriculture power development Governance Ebonyi 6 Health System HIV/AIDS Community- Based Fadama 3 Community- Based Development 2 Poverty Reduction, Urban Community Social Development Development Edo Health System HIV/AIDS Fadama 3 Community- Development 2 Based Urban Development Ekiti 5 Health System HIV/AIDS Community-Based Fadama 3 Development 2 Poverty Reduction, Community Social Development Enugu 5 Health System HIV/AIDS Local Empowerment Urban Fadama 3 Development 2 and Environment Water 1 Management Gombe 5 Health System HIV/AIDS Fadama 2, Development 2, Fadama 3 Malaria Imo 5 Health System HIV/AIDS Local Empowerment Fadama 2, Development 2 and Environment Fadama 3 Management Jigawa 4 Health System HIV/AIDS Community Development 2, Based Urban Malaria Development Kaduna 8 Health System State Education HIV/AIDS Urban Fadama 2, Rural Access State Governance Development 2 Sector Project Water 1 Fadama 3 and Mobility and Capacity Project 1 Building Kano 4 Malaria State Education HIV/AIDS Fadama 3 Sector Project Katsina 3 Health System HIV/AIDS Local Empowerment Development 2 and Environment Management Kebbi 6 Health System HIV/AIDS Community-Based Fadama 2, Development 2 Poverty Reduction, Fadama 3 Community Social Development Kogi 4 Health System HIV/AIDS Community-Based Development 2 Poverty Reduction, Community Social Development Kwara 4 Health System State Education HIV/AIDS Community -Based Development 2 Sector Project Poverty Reduction, Community Social Development Lagos 8 Health System HIV/AIDS Urban Fadama 2, Lagos Urban Lagos Lagos Metropolitan Development 2 Water 2 Fadama 3 Transportation Metropolitan Development and Project Development and Governance Governance A P P E N D I X F : D I S T R I B U T I O N O F P R O J E C T S B Y S T AT E S Nasarawa 4 Health System HIV/AIDS Fadama 3 Community- Development 2 Based Urban Development Niger 5 Health System HIV/AIDS Local Empowerment Fadama 2, Development 2 and Environment Fadama 3 Management Ogun 6 Health System HIV/AIDS Urban Fadama 2, Community- Development 2 Water 1 Fadama 3 Based Urban Development Ondo 4 Health System HIV/AIDS Fadama 3 Community- Development 2 Based Urban Development Osun 3 Health System HIV/AIDS Fadama 3 Development 2 Oyo 5 Health System HIV/AIDS Local Empowerment Fadama 2, Development 2 and Environment Fadama 3 Management (Table continues on next page) 91 92 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L Table F.3: Nigeria: Distribution of Projects by States (continued) Community- Number of driven Transport, Urban State projects Health Education HIV/AIDS development Water Agriculture power development Governance Plateau 2 Health System HIV/AIDS Development 2 Rivers 4 Health System HIV/AIDS Fadama 3 Development 2, Malaria Sokoto 3 Health System HIV/AIDS Fadama 3 Development 2 Taraba 4 Health System HIV/AIDS Fadama 2, Development 2 Fadama 3 Yobe 5 Health System HIV/AIDS Community-Based Fadama 3 Development 2 Poverty Reduction, Community Social Development Zamfara 3 Health System HIV/AIDS Fadama 3 Development 2 FCT 4 Health System HIV/AIDS Fadama 2, Development 2 Fadama 3 A P P E N D I X F : D I S T R I B U T I O N O F P R O J E C T S B Y S T AT E S Table F.4: The World Bank Fee-for-Service Activities at the Subnational Level in the Russian Federation Region/Area Description Development of Public-Private Partnership System City of St. Petersburg Advising the government on establishing a regional public-private partnership unit (expansion of Pulkovo Airport, and building of the Western High Speed Diameter, Orlovsky Tunnel, and Nadzemny Express Projects), as well as managing the public-private partnership projects City of Kazan Advising on development of a mid-term public-private partnership­based management strategy for the Kazan Airport Volgograd Oblast Advisory services to the government a public-private partnership unit Education Khanty-Mansi Autonomous A comprehensive assessment of the regional education system and provision of analytical services Okrug-Yugra to help upgrade the Okrug's preschool education system City of Kazan A diagnostic study of the higher education system to further design of a balanced education development strategy for the city City of Moscow Developing a strategy to improve the international status and to strengthen the research capacity of the Higher School of Economics Tver Oblast Cooperation on designing a package of measures for vocational education system development Health Khanty-Mansi Autonomous Assistance in restructuring the health care delivery system and strengthening the institutional Okrug-Yugra capacity for medical education and clinical research City of Kazan Designing measures to strengthen the healthcare system Social policy/assistance Republic of Tatarstan Monitoring and evaluation of living standards and poverty indicators Tver Oblast Cooperation in improving targeted social assistance Kalmyk Republic Advisory services to improve social assistance efficiency Infrastructure development City of Nizhnevartovsk Supporting the development of a Housing and Communal Services Strategy and Action Plan (2008­12) Perm Krai Designing a regional infrastructure development program Strengthening of regional administration capacity Khanty-Mansi Autonomous Assistance on improving the government's institutional structure and performance, including a horizontal function Okrug-Yugra review in executive authorities of the Okrug Tomsk Oblast A study of administrative barriers to investment and advice on improvement of the investment climate Republic of Tatarstan A study of administrative barriers to investment and advice on investment climate improvement Source: World Bank Russia Country Office, as of April 6, 2009. 93 APPENDIX G: FISCAL RESPONSIBILITY LAW--BRAZIL Enacted in May 2000, the Fiscal Responsibility · Transparency: Budget outturns and compli- Law (FRL) is at the core of efforts to strengthen ance with the FRL--including a statement of fiscal institutions in Brazil. The main issues ad- corrective measures if the relevant provisions dressed by the LRF are in six areas: are breached--are reported on a regular basis. Municipalities and states are also required to re- · Debt: The FRL contains strict provisions on in- port the fiscal outturns of the previous year to debtedness and issuance of public debt by the Ministry of Finance. The legislative branch the central bank, prohibiting creditor debt- of each level of government, aided by their re- restructuring operations among the various spective Court of Accounts, monitors obser- levels of government. In accordance with Ar- vance with the fiscal targets and ceilings. ticle 30 of the law, a Senate resolution estab- · Legal framework: The FRL introduced more lished limits of indebtedness for each level stringent requirements on fiscal targets in the of government. When in breach of the debt preparation of the Budget Guidelines Law (Lei ceilings, new financing and discretionary trans- de Dotaçőes Orcamentárias--LDO), strength- fers to subnational governments are banned. ening its role in budget preparation and fiscal · Expenditures: The creation of permanent management in general. The LRF also calls spending mandates without a corresponding for a detailed assessment of the government's increase in permanent revenue or a reduction contingent liabilities and strengthens the link in other permanent spending commitments between the Annual Budget Law (Lei de Orça- is prohibited, as are new spending commit- mento Annual--LOA) and the LDO. A com- ments that cannot be executed before the end plementary Fiscal Crime Law is applied to all of the incumbent's term in office. levels of the public administration, with the · Personnel expenditures: FRL sets separate possibility of detention for public officials who ceilings for personnel spending, including pen- fail to comply with the FRL. sions and payment of subcontractors. Spend- · Golden rule: To prevent financing of current ing is limited at 50 percent of net current spending by borrowing, the amount of new revenues at the federal level, and 60 percent for loans contracted is limited to the amount of the states and municipalities. In case of noncom- capital expense. In practice, it means that any pliance, the jurisdiction is not allowed to engage loans contracted will only be used for expenses in new credit line operations, and subnational related to investments. governments will not be allowed to receive transfers or credit guarantees from the federal government. 95 APPENDIX H: PARTNERSHIPS WITH OTHER DEVELOPMENT AGENCIES Brazil · Areas where demand is so large that both banks engage independently (in areas such as public Overview of Other Donor Agency Activities sector management and infrastructure). Along with the World Bank, the major external fi- nancing agencies operating in Brazil are the Inter- The Bank also works closely with several bilateral American Development Bank (IDB) and the agencies--for example, the European Union and Andean Development Corporation (CAF). The Germany in conservation in the Amazon, and the IDB and the Development Corporation both work United Kingdom on climate change and public sec- to contribute to the process of economic and so- tor management. cial development in Latin America and actively carry out projects in Brazil. The Development India Corporation has concentrated its efforts on: (i) supporting municipal governments; (ii) re- Overview of Other Donor Agency Activities gional integration; (iii) socioeconomic develop- There were a large number of donor agencies in ment; (iv) infrastructure; and (v) the private India, and the Bank had close collaboration with sector. At the subnational level, IDB works mainly them in most sectors, including energy (ADB, with municipal governments. Its operational strat- Japan Bank for International Cooperation [ JBIC], egy has rested on three major components, in- U.K. Department for International Development cluding: (i) promoting sustained, stable, and [DFID], U.S. Agency for International Develop- environmentally sustainable growth; (ii) reducing ment [USAID], Kreditanstalt für Wiederaufbau poverty, promoting social inclusion, and en- [KfW], Canadian International Development hancing social and regional equity; and (iii) sup- Agency [CIDA]), health and nutrition (USAID, porting institutional strengthening and promoting World Health Organization [WHO], European democracy and citizen participation. Commission [EC], DFID, concerned UN agencies, Australian Agency for International Development Donor Coordination/Collaboration [AusAid], International Labor Organization [ILO], The Bank program in Brazil is based on an infor- and other institutions, including the Gates Foun- mal structure with IDB that determines which dation), and small and medium-size enterprise fi- agency will do what. There are: , nancing (DFID, KfW German Agency for Technical Cooperation [GTZ], and International Finance · Areas where the two banks have worked to- Corporation [IFC]). However, in September 2003, gether effectively (for example, in financing the government of India announced new guide- the Bolsa Familia program and, in the future, lines for development cooperation with bilateral helping with restructuring state debts) partners. In keeping with the new guidelines, · Areas where each bank "leaves it to the other." India will receive direct bilateral assistance only For example, the World Bank (IBRD) special- from Japan, the United Kingdom, Germany, the izes in rural development and the IDB in United States, the European Commission, and the tourism. Russian Federation. All other bilateral assistance 97 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L would be routed either through multilateral or- lead states. DFID has supported lead state gov- ganizations or, for projects of economic and social ernments to develop a plan to build the capacity importance, directly to universities, NGOs, or au- of local governments. USAID is implementing tonomous bodies registered under the Foreign its economic growth and governance program Contribution (Regulation) Act. With this shift, the in Cross River, Kano, and Kaduna, and is also in- Bank has lost some valuable partners in India, and creasingly focusing resources on Kano and Kaduna now has fewer donor partners to work with. Be- in particular. Both of these agencies have also cause of the changed donor landscape, particular worked in the nonlead states: DFID has worked focus has been placed on coordination with the on girls' education, immunization, malaria re- largest external financing sources (ADB, EC, DFID, duction, and HIV/AIDS. USAID has worked within Japan, and Germany). its strategic framework on governance and democ- racy, growth, and human development. Donor Coordination/Collaboration A good example of exploiting synergies with part- Donor Coordination/Collaboration ners is the World Bank's work with DFID. At the The Bank-DFID partnership dominates the donor suggestion o f DFID, the Bank and DFID-India en- scene in Nigeria. Over the period of fiscal tered into a strategic partnership agreement in 1998­2008, the Bank and DFID cofinanced two April 2004. DFID has provided trust fund re- state projects (Universal Basic Education, fiscal sources to the World Bank to work on common 2003, and the State Education Project, fiscal 2007), priorities. DFID and the Bank have worked to- both in the education sector. The World Bank has gether on direct budgetary support and power a joint CPS (fiscal 2005) with DFID. Joint diag- sector reform in Andhra Pradesh and Orissa, and nostics and analysis included Country Economic also cofinanced (with other agencies) the federal Memorandum, the Education Public Expenditure government's centrally sponsored education pro- Review, and the Investment Climate Program, grams. World Bank and DFID staff worked par- which assisted in designing DFID and World Bank ticularly closely in Bihar and Orissa, building on interventions to support growth and private sec- a shared vision of the development process and tor development at the state level. In addition to the primacy of client ownership. a main office in Abuja, DFID also has three regional offices (Lagos, Kano, and Enugu) that coordinate There are considerable overlaps in the areas of programs in several of Nigeria's 36 states. Though activity between the Bank and ADB in India. Both the Bank does not have regional offices, it was able institutions have a substantial and growing in- to work in cooperation with DFID's significant volvement in infrastructure and in energy, and presence on the ground in Kano (the relationship both are also engaged in adjustment lending to with Kaduna is also managed from this office) states. To avoid overlap, ADB, the Bank, and the In- and more limited presence in Enugu (also cover- dian government completed a Coordinated Assis- ing Cross River) and Lagos. tance Strategy for roads in June 2001, focusing on state and national highways. The ADB and World , Other agencies, such as USAID, the UNDP and the Bank have continued to exploit complementarities African Development Bank (AfDB), have varying in several other sectors, including finance, power, degrees of partnership, but most have limited and railways. ADB has also been proactive in lend- engagement. Nigeria's main development partners ing to lagging states, with commitments to fund the (USAID, EC, African Development Bank, and the six most poorly connected states. UN system) have been working closely and plan to sign a joint 2009 CPS. Nigeria Russia Overview of Other Donor Agency Activities Along with the World Bank, DFID and USAID Overview of Other Donor Agency Activities have taken the lead in state-level work in Nigeria. International donors have been actively involved DFID, USAID, and the World Bank signed the in Russia, but the scope of international donor as- Principles of Partnership agreements with the sistance is becoming more limited and selective 98 A P P E N D I X H : PA RT N E R S H I P S W I T H O T H E R D E V E L O P M E N T A G E N C I E S as Russia's economic performance improves. more than 20 projects in Russia. Most of these Russian itself is emerging as an international projects are in financial markets, general manu- donor and active member of multilateral organ- facturing, and private equity funds. The World izations. Even so, there are a few external Bank and the EBRD have collaborated on se- financing agencies operating in Russia; the Eu- lected analytical work, such as work on the busi- ropean Bank for Reconstruction and Develop- ness climate. Examples of World Bank partnership ment (EBRD) is a prominent donor in this group. with other development agencies include the Since 2002, the EBRD has maintained annual in- program on public administration and budget re- vestment levels in Russia of over EUR 1 billion, 80 form supported by DFID, CIDA, the Swedish In- percent of which has gone to the private sector. ternational Development Agency [SIDA], and the It has an active program focused on nonsovereign European Union. The Bank has collaborated lending, including lending to Russia's regions. closely with the Organization for Economic Co- The EBRD has been particularly active in the fi- operation and Development (OECD) on issues of nancial sector; energy; infrastructure; manufac- tertiary education. The Bank and WHO jointly turing; and, through the Russia Small Business coordinated a partnership for work on improving Fund, the small and medium-size enterprise sec- governance in the Southern Federal Okrug, which tor. Like IFC, the EBRD has been introducing included the active participation of USAID, the Eu- new financial instruments in Russia, including ropean Union, and DFID. A multisector Finnish ruble-denominated bond issues and loans in trust fund has supported a diverse set of regional 2005, and working increasingly with regional development activities in the regions of the North- administrations. west. DFID has also supported the Bank's poverty work in Russia. IFC's Private Enterprise Partner- Donor Coordination/Collaboration ship in Russia has been funded by the govern- The EBRD remains a vital working partner for ments of Austria, Canada, Denmark, Finland, the World Bank Group, particularly for IFC. To Switzerland, and Saxony (of Germany), as well as date, IFC and the EBRD have jointly supported by the Global Environment Facility. 99 APPENDIX I: PEOPLE MET People Interviewed/Consulted Brazil Idilvan Alencar Executive Secretary, State Secretariat of Education, Ceará Rodrigo Apgaua Secretariat of Education (SEE), Minas Gerais Luiz Antonio Athayde Under Secretary, Head of Public-Private Partnership Unit, State Secretariat for Economic Development, Minas Gerais Andre Barrence Director of Superintendancy for Central Coordination, Secretariat of Planning and Management (SEPLAG) Marcus Augusto Coordinator, State Secretariat of Finance, Ceará Vasconcelos Coelho Carlos Eduardo Lampert Costa Deputy Secretary, Federal Ministry of Planning and Management, Secretariat of International Affairs Edoardo Coutinho Deputy Secretary, National Treasury Juliana Damasceno Secretariat of Planning and Management (SEPLAG) Accacio Ferreira Coordinator of EXPORTMINAS, State Economic Development Secretariat (SEDE), Minas Gerais Marcela Ferreira Projects Coordinator, State Economic Development Secretariat (SEDE), Minas Gerais Jurandir Gurgel Gondim Filho Coordinator, State Secretariat of Finance, Ceará Joao Filocre Secretariat of Education (SEE), Minas Gerais Francisco Gaetani Deputy Executive Secretary, Federal Ministry of Planning, Budget and Management Executive Secretariat Marcos Costa Holanda Director General, Economic Research and Strategy Institute (IPECE), State Secretariat of Planning and Management, Ceará Henri Kistler Adviser, Federal Ministry of Finance, SAIN Monica Salles Lanna Legal Adviser, State Economic Development Secretariat, Minas Gerais Leonardo Mauricio Deputy Secretary of Finance, State Secretariat of Finance, Colombini Lima Minas Gerais Thais Amaral Lucena State Secretariat of Planning and Management, Ceará Joao Marcos Maia Deputy Secretary, State Secretariat of Finance, Ceará Tiago Moraes Institute for Governance Studies (IGS) 101 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L Bernardo Moretzsohn Secretariat of Planning and Management (SEPLAG), Minas Gerais Jorge Duarte de Oliveira Director, EXPORTMINAS, State Economic Development Secretariat (SEDE), Minas Gerais Silvana Parente Secretary of Planning, State Secretariat of Planning and Management, Ceará Cesar Augusto Pinheiro Secretary, State Secretariat of Hydro Resources, Ceará Gilberto Resende Secretariat of Education (SEE), Minas Gerais Fernanda Cimini Salles Assistant in International Projects and Operations, State Economic Development Secretariat (SEDE), Minas Gerais Accacio F. Santos Jr. Superintendent of Foreign Trade Affairs, State Secretariat for Economic Development, Minas Gerais Aline Dieguez B. de Program Director, Federal Ministry of Planning, Budget and Meneses Silva Management Executive Secretariat Frederico Amaral E Silva Public Policy and Management Specialist, Government of State of Minas Gerais Rafael Siqueira Secretariat of Planning and Management (SEPLAG) Bernardo Tavares Deputy Secretary of Planning, Secretariat of Planning and Management (SEPLAG) Thiago Toscano Under Secretary of Planning, Secretariat of Planning and Management (SEPLAG) Alvaro Vereda Federal Ministry of Finance, SAIN Renata Vilhena Secretary of Planning, State Secretariat of Planning and Management, Minas Gerais India K. C. Badu Team Member, Orissa Economic Recovery Credit M. Brahmaiah Resource Group Director, Financial Management, Center for Good Governance, Andhra Pradesh Ch. Channareddy BE Director (Transmission), Andhra Pradesh Transco Soumya Chattopadhyay Under Secretary to the Government of India, Department of Economic Affairs, Ministry of Finance K. Damayanthi Secretary, Andhra Pradesh Social Welfare Residential Educational Institutions Society S. Galeb Professor, Center for Economic and Social Studies (CESS), Andhra Pradesh Sri Azhar Hussain Deputy Executive Engineer, Roads Department, Andhra Pradesh K. Kamayanthi Secretary, Andhra Pradesh Social Welfare Residential Educational Institutions Society A. Srinivas Kumar Deputy Executive Director (Finance & Projects), Centre for Good Governance, Andhra Pradesh T. Vijay Kumar Chief Executive Officer, Society for Elimination of Rural Poverty, Andhra Pradesh 102 APPENDIX I: PEOPLE MET Vasudha Mishra Secretary, Finance (PMU) Department Government of Andhra Pradesh G.K. Mitra Faculty, Center for Economic and Social Studies (CESS), Andhra Pradesh B.C. Mohapatra Additional Secretary, Finance Dept., Government of Orissa Partha Mukhapadhay Centre for Policy Research, New Delhi C.S. Murthy Professor, Center for Economic and Social Studies (CESS), Andhra Pradesh Er. Jayamangal Nayak Chief Engineer, World Bank Projects, Orissa State Roads Project (fiscal 09), Project Management Unit Works Department, Govt. of Orissa K. Nijayanand Joint Managing Director, Transmission Corporation of Andhra Pradesh Ltd. Manoj Panda Director, Center for Economic and Social Studies (CESS), Andhra Pradesh Naresh Penumaka Additional Secretary, Finance Department, Government of Andhra Pradesh R. Radhakrisna Professor, Center for Economic and Social Studies (CESS), Andhra Pradesh M.S. Srenivasa Rao Joint Director, Medical, Health, Andhra Pradesh S. Bhale Rao Principal Secretary--Public Enterprise Department, Andhra Pradesh . P Padmanabhe Rao Register, Center for Economic and Social Studies (CESS), Andhra Pradesh I.Y.R. Krishna Rao Principal Secretary to Government, Finance Department, Andhra Pradesh Sri M. Venkateroara Ras Roads Department, Andhra Pradesh Satyapriya Rath O.S.D., Finance Dept, Government of Orissa C. Ravi Faculty and Joint Director, Center for Economic and Social Studies (CESS), Andhra Pradesh G.B. Redd Director Watershed Mission and Ex-Officio Special Secretary to Govt., Government of Orissa . P Prudvikas Reddy Faculty, Center for Economic and Social Studies (CESS), Andhra Pradesh G. B. Reddy Director, Watershed Mission & Ex-Officio Special Secretary to Government, Orissa I. Rama Chandra Reddy Deputy General Manager, State Bank of India, Andhra Pradesh M. Gopinath Reddy Professor, Center for Economic and Social Studies (CESS), Andhra Pradesh R. Venkat Reddy M Venkatarangaiya Foundation, Andhra Pradesh Sri M.K. Rehman Engineer in Chief, Irrigation Department, Andhra Pradesh Alok Sheel Joint Secretary, Department of Economic Affairs, Ministry of Finance, Andhra Pradesh 103 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L Dipak Kumar Singh Director, Department of Economic Affairs, Ministry of Finance . R.V Singh Special Secretary to Government, P&C Department, Orissa Prasann Thatte Program Manager, Centre for Good Governance, Andhra Pradesh K. Vijayanand Joint Managing Director, Transmission Corporation of Andhra Pradesh Ltd. Nigeria Ben Akabueze Commissioner for Economic Planning and Budget, Lagos State John Bezard Commissioner for Economic Planning and Budget, Kaduna State Olaoye Abdulkareem Deputy Director, Federal Ministry of Finance Abiodun Alao Director, Federal Ministry of Finance Bayamin Bagaiya DAF, Ministry of Planning and Budget, Kaduna State Peter K. Bakam DPRS, Ministry of Planning and Budget, Kaduna State Shehu Moh'd Bambale DIA, Ministry of Finance, Kaduna State Dawuda Danbaki Director, Treasury Operation, Ministry of Finance, Kaduna State S. Eloho Assistant Director, National Planning Commission Bulus D. Emishe Permanent Secretary, Ministry of Finance, Kaduna State Funso Esan Deputy Director, National Planning Commission Yusuf Moh'd Hayatuddeen PFMU, Ministry of Finance, Kaduna State Nuruddeen Ibrahim Director, Budget, Ministry of Planning and Budget, Kaduna State Balarabe Shehu Kudan D/MOFI, Ministry of Finance, Kaduna State Tunde Lawal Deputy Director, National Planning Commission Sylvester O. Monye Secretary to the Council Esther J. Myahwegi , Project Coordinator, SESP Ministry of Education, Kaduna State Ayodele Omotosho Director, National Planning Commission Stephen Oronsaye Permanent Secretary, Federal Ministry of Finance Zahari Aminu Salihu Accountant-General, Ministry of Finance, Kaduna State Danladi D. Sanda Permanent Secretary, Ministry of Planning and Budget, Kaduna State Daniel K. Sankey DPME, Ministry of Planning and Budget, Kaduna State Abbas Yahaya Sanusi Director, Final Accounts, Ministry of Finance, Kaduna State D.B. Sule Assistant Director, Federal Ministry of Finance Joel Usman , M&E Officer, SESP Ministry of Education, Kaduna State Shamsuddeen Usman Minister, National Planning Commission Sani Zorro SA (Communications), National Planning Commission Russia Eduard Batanov Chairman, St. Petersburg Finance Committee Murat Gadelshin Adviser to the Prime Minister, Tatarstan Republic (former Vice-Mayor of Kazan) 104 APPENDIX I: PEOPLE MET Vladimir Gritskikh Director General, Directorate of Extra Budgetary Programs for Municipal Development, Municipality of Kazan, Tatarstan Republic Leonid Limonov Director General, Leontief Center Nina Oding Head, Research Department, Leontief Center Vladislav Onishchenko Deputy Director, Foundation for Enterprise Restructuring and Financial Institutions Development Alexander Puzanov Director General, The Institute for Urban Economics Larisa Yeroshkina Acting Director, Intergovernmental Relations Department, Federal Ministry of Finance World Bank Abuja u Onno R hl Country Director, Nigeria Bayo Awosemusi Lead Procurement Specialist, AFTPC Steven R. Dimitriyev Sr. Private Sector Development Specialist, AFTFP Simeon K. Ehui Sector Leader, AFTAR Foluso Okunmadewa Lead Social Protection Specialist, AFTH3 Volker Treichel Lead Economist, AFTP3 Brasilia John Briscoe Country Director, Brazil (ret.) Alexandre Abrantes Manager, Portfolio & Operations Antonio Rocha Magalhaes Former Bank staff Madalena do Santos Consultant, World Bank (former World Bank staff) Jennifer Sara Sector Leader, Infrastructure and Urban Deborah Wetzel Lead Economist and Sector Leader Dhaka Vinaya Swaroop Adviser, MDW New Delhi Rachid Benmessaoud Operations Adviser, SACIN Roger Grawe Consultant, SACIN Mandakini Kaul Country Officer, SACIN Rajna Khanna Senior Economist, SASGP Gerard M. La Forgia Lead Health Economist, SASHD Vikram Menon Senior Public Sector Specialist, SASGP Giovanna Prennushi Economic Adviser, SASEP . V J. Ravishankar Lead Economist, SASGP 105 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L Moscow Klaus Roland Country Director, Russia Andrei Darusenkov Former Bank staff Lev Freinkman Lead Economist Andrei Markov Sr. Human Development Specialist, ECSHD Tatyana Shadrunova Urban Specialist, ECSSD Samir Suleymanov Senior Operations Officer, ECSSD Stepan Titov Senior Economist, ESCPE Washington, DC Mauricio Carrizosa Adviser, IEGCR Shahrokh Fardoust Senior Adviser, IEGDG Roy Gilbert Lead Evaluation Officer, IEGSE Mohinder Gulati Country Sector Coordinator, ECSSD Poonam Gupta Country Program Coordinator, AFCNG Nalini Kumar Senior Evaluation Officer, IEGSE Larisa Leshchenko Senior Country Officer, AFCZA Jorge Munoz Lead Rural Development Specialist, LCSAR Tawhid Nawaz Operations Adviser, HDNOP Marsha Olive Country Program Coordinator, ECCU1 Chris Parel Consultant, LCSPS Jose Guillerme Reis Lead Private Sector Development Specialist, LCSPF Mark Sundberg Manager, IEGCG Vinod Thomas Director General, IEGDG Fahrettin Yagci Consultant, AFTPM Donors Julian Barr Director, ITAD (DFID contractor) Shamit Chakravarti Programs Officer, ADB, New Delhi Emma Donnelly Deputy Head, DFID Nigeria Tetsu Ito Social Economist (Financial Sector), ADB, New Delhi Jaime Mano, Jr. Sector Specialist, IDB, Brasilia Susanna Moorehead Former Director, DFID-India Shigehiko Muramoto Head, Project Administration Unit, ADB, New Delhi Sujatha Viswanathan Social Economist, ADB, New Delhi Joseph Umoabasi Regional Coordinator, South-West Nigeria, DFID 106 ENDNOTES Chapter 1 in taming inflation, which had numerous collateral ef- 1. In some countries, federal governments are called fects. It made spending at all levels of government "central" or "national." This study uses the term "fed- more transparent, which prevented governors from eral government" for both. engaging in unsustainable spending patterns (Dillinger 2. The largest subnational units in three of the four and Webb 2001). countries included in this review (Brazil, India, and 7. The Bank proposed a three-pronged strategy of Nigeria) are called "states." In Russia, the "subjects of state lending: investment loans in social sectors and in- federation," as they are known, can include republics, frastructure in creditworthy states; reform loans in okrugs, oblasts, and others. The more common term states where fiscal and sector policy reforms were being used to describe all of them is "region." This study implemented; and, for states with low creditworthiness, uses both terms, "state" and "region," to describe the the Bank would lend through federal loans (requiring subnational units in Russia. federally funded counterpart financing). 3. The evaluation commissioned background ma- 8. These loans were Minas Gerais State Privatization terial for each country, prepared by Joao Oliveira (Brazil), (fiscal 1998, later dropped because of the state gov- Tapas Sen (India), Olufemi Taiwo (Nigeria), and Galina ernment's moratorium on state debt); Rio de Janeiro Kurlyandskaya (Russia). The report draws extensively State Reform Privatization (fiscal 1998); Rio de Grande on this material. do Sul State Reform Project (fiscal 1997); and Mato Grosso State Privatization Project (fiscal 1997). Chapter 2 9. See Chapter 3 for details. 1. During most of the period reviewed, foreign ex- 10. State selectivity was based on a more systematic change risks appeared negligible. approach using such criteria as the total number of poor, 2. Detailed information on state-level lending for average income levels, fiscal performance, social per- each country can be found in Appendix F. formance and innovation, structural reform perfor- 3. Over 1,000 municipalities were created after 1988. mance, policy continuity, and implementation capacity. 4. The main reasons were poor management of Based on these benchmarks, Bank lending for all north- public resources by state (and local) governments, eastern states was expected to be approximately 55 per- lack of fiscal discipline, and no clear division of re- cent of total Bank lending, assistance for the north sponsibilities among different government levels, often and center-west was expected to be about 10 percent leading to overlap in spending across different levels of total Bank assistance, and assistance for selected of government. states of the southeast and south with about 15 percent 5. State debt in bonds, as percentage of GDP more , and 10 percent of Bank assistance, respectively. Re- than doubled between 1990 and 1996. maining Bank assistance was to be allocated case-by-case 6. According to many sources, the Real Plan had no to support innovations or special opportunities. less significant impact on the Brazilian political econ- 11. Foreign reserves had fallen from $74 billion in omy than the Constitution of 1988. While the consti- April 1998 to $30 billion in January 1999. tution simply codified the power of subnational forces, 12. These loans included Programmatic Financial the reforms of 1993­95 dramatically reconfigured fis- Sector Adjustment Loan I and II (fiscal 2001 and fiscal cal power. The Real Plan succeeded first and foremost 2002); Programmatic Fiscal Reform I and II (fiscal 2001 107 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L and fiscal 2003); and Programmatic Human Develop- government and signing of the loan needs to be short- ment Sector Reform Loan I (fiscal 2003). ened drastically. 13. Projects would include a component to address 21. Several states were created after 1950. cross-cutting public sector management issues, such as 22. Selection of state projects was done on project capacity building for better policy analysis and expen- and sector grounds, without much concern for the diture review, and focused support for specific policy overall policy of the state. Sometimes there was polit- reforms: Maranhao Rural Poverty Reduction (fiscal ical pressure to spread Bank operations over as many 2004); Pernambuco Education Quality Improvement states as possible, without a strategic rationale at the Project (fiscal 2005); Bahia Rural Poverty (fiscal 2006); level of individual states. and others. 23. Bank engagement was not universally welcomed 14. The state and municipal projects would even- because the political costs of such reforms were feared tually have to be approved by Brazil's inter-ministerial to be large (reduction of subsidies on irrigation water, commission (COFIEX). power, and the like, reducing the wage bill and public 15. One often-cited example is the Minas Gerais enterprise reforms), and there were genuine concerns Rural Poverty Reduction Project (fiscal 2006). Although about their impact on the poor (user charges for health there are significant numbers of poor living in its north- services). There was resistance to reforms from the left- ern region, Minas Gerais is one of Brazil's wealthiest and ist states almost as a matter of principle as well, because best-performing states. The Bank believed that suc- the Bank-supported reforms involved privatization. cessful engagement in Minas Gerais would produce a 24. The central government sought the Bank's help demonstration effect throughout Brazil. in disciplining the states, specifically in inducing them 16. See Chapter 3 for details on Brazil's multisector to accept and abide by a hard budget constraint, in SWAp. exchange for quick disbursement and relatively cheap 17. Because of a change in the Bank's policy in financing. 1998, DPLs could be used as a lending instrument in 25. Formally, this was an investment project, but it subnational governments. had the kind of macro-conditionality normally associ- 18. A subnational DPL supporting an ambitious pub- ated with a policy-based loan. lic sector modernization program was implemented in 26. Adjustment lending never took off to the extent Minas Gerais--a well-performing high-income state. A envisaged in the 2001 CAS, but there was a steady flow subnational SWAp supporting a comprehensive sectoral of about one adjustment loan a year, or around $150 program was implemented in Ceará--a relatively poor million a year. but well-managed state. 27. The DPL for Tamil Nadu was later dropped. 19. Almost half of the projects were at the state 28. CAS objectives had been achieved in the re- level. forming focus states. 20. The Brazilian experience showed three things. 29. Two to Orissa (fiscal 2005, 2007), one to Andhra First, there is little chance for new lending in the first Pradesh (fiscal 2007), one to Bihar (fiscal 2007), and year of the four-year term, because it usually takes one to Himachal Pradesh (fiscal 2008), for a total of months to formulate a program. Second, in the next $1.1 billion. two-to-three years, there is high demand for fast re- 30. This was driven largely by a decline in the Uttar sponse from the Bank, to show results before the Pradesh portfolio, reflecting the hiatus in dialogue term expires. Third, demand declines in the last year, until 2007. Among the four states, the Bank's portfo- because the law prohibits the state government from lio increased only in Bihar. Outstanding loans to Orissa borrowing in the last nine months of the term. For the and Jharkhand remained roughly the same. Bank, this means that commitments are going to be 31. Centrally sponsored schemes are programs de- uneven, with more lending in the first and second signed and financed by the central government but im- years of the Country Partnership Strategy (corre- plemented by state and local governments. Bank support sponding to the second and third years of terms of state has taken various forms: support at the national level administrations), and less in the third and fourth years. (for example, the Sawa Shiksha Abhiyan Education for In addition, the long (on average, 30 months) lag be- All scheme); support to individual states participating tween approval of the project concept by the federal in a national scheme (for example, on rural roads); and 108 ENDNOTES support to states in parallel to the national scheme (for 42. Only the St. Petersburg Rehabilitation Project (fis- example, on rural water supply and sanitation). cal 1997) was approved during this period. The other 32. Country strategy documentation for 2000 and three dropped infrastructure projects were Siberia and 2001 (World Bank Group 2000, 2001, 2004b). Far East Highways Rehabilitation (planned for fiscal 33. For example, three of the Bank's projects--the 1998), Moscow Urban Transport (planned for fiscal HIV/AIDS Program Development Project, the Second 1998), and Krasnodar Power (planned for fiscal 1998). Health System Development Project, and the Partner- 43. The Regional Fiscal Federalism Project (fiscal ship for Polio Eradication Project--covered all 36 states 2000) and Fiscal Federal and Regional Fiscal Reform of the federation. Several other projects, such as the Uni- Project fiscal (2002) are discussed in more detail in versal Basic Education Project, covered more than 10 Chapter 3. states. Where it was obvious that the Bank could not 44. Regions that met the Bank's criteria were fairly cover many states, consideration was given to the achieve- diverse in terms of geographic location and income. For ment of a balance among the six geopolitical zones. example, three out of six regions selected for one of 34. The Bank helped to produce the Lagos State Eco- them were relatively rich regions, while the others nomic Development Strategy as well as fiduciary work were well below the Russian average per capita budget on procurement and financial management in Lagos. revenue level. 35. The fiscal 2005 Country Partnership Strategy 45. The term "subnational" instead of "regional" was a joint strategy document with DFID. was specifically introduced to underline the possibil- 36. The 2004 Country Partnership Strategy was a joint ity of Bank involvement not only with the subjects of CAS with DFID, which signaled close coordination of federation (regions), but also with other subnational lev- the DFID and Bank programs. DFID was particularly els of government, such as municipalities. concerned that the lead states approach might divert 46. Analytical and advisory activities, IBRD loans in donor interventions away from the poorest parts of the support of federal programs, World Bank Institute ca- country. (IEG forthcoming.) pacity building activities, carbon financing, and grants. 37. During IEG mission interviews, some Nigerian 47. Fourteen fee-for-service agreements have been counterparts suggested keeping the principle of the lead signed, one has been completed. state approach to enforce competition, but packaging 48. The World Bank Group (IFC and IBRD) has it in a modified way to avoid the controversy associated committed $85 million in three subnational projects: with the lead state title. Among the ideas being con- two in the Chuvash Republic for improvements in rural sidered by the Bank is the possibility of dropping the road networks and water utility infrastructure and a local lead states approach for investment lending, but using currency loan to the Municipality of Petropavlovsk- DPLs as a way to support states that pass Laws of Fis- Kamchatsky to finance rehabilitation of about 23 per- cal Responsibility and show evidence of commitment cent of its road network. to higher standards of public sector management and 49. For example, in Brazil the stock of foreign re- improved service delivery. serves grew from $37 billion in 2002 to $205 billion in 38. Those are 21 republics (native territories), 46 2008. oblasts, 9 krais, 4 autonomous okrugs, 1 autonom- ous oblast, and the 2 federal cities of Moscow and Chapter 3 St. Petersburg. 1. The topic of fiscal decentralization is part of the 39. The problem of unfunded mandates was not fully broad area of fiscal federalism, but is not treated here. resolved until 2005, the year of expenditure assign- In practice the Bank has not systematically advocated ment reform. fiscal decentralization, but it has supported fiscal de- 40. Federal okrugs are best described as decon- centralization when federal governments have deter- centrated federal government units that are supposed mined that this is the direction in which they want to to provide some oversight and control over subjects of proceed. Among the four countries, Brazil has moved the federation. Their actual legal status remains elusive. most clearly to decentralize fiscal decision making to 41. This is not precisely representative, as most of the state level during the period under review; the sit- the highest-income regions in Russia are the sparsely uation in Nigeria and Russia has been somewhat am- populated northern territories rich in natural resources. biguous; and in India, arguably the growth of Centrally 109 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L Sponsored Schemes moves in the direction of greater eting and fiscal management; (ii) financial planning, centralization. treasury, and cash management; (iii) budgetary ac- 2. The relationship of the state governments to local counting, reporting, and audit; (iv) expenditure and governments is an important issue in federations. In public sector restructuring; (v) debt management sys- most cases there are mandatory requirements for shar- tems; (vi) regional budget procurement system; (vii) ing revenues of the allocated transfers from the center regional fiscal management guidelines, best practice to the states. In Brazil the municipalities have a direct standards, and regional and local public finance man- relationship with the central government, and transfers ual; (viii) computer equipment and design of software to them bypass the state government. This is an im- programs; and (ix) training of local staff needed for a portant subtopic that merits a separate examination. successful installation and subsequent application of in- 3. Appendix B compares the fiscal framework in tegrated financial management systems in the selected each of these areas among the four countries studied. regions. In addition, the project was to assist partici- 4. Nigeria is a special case here. About a third of all pating regions with the design of expenditure reform transfers go to the four oil-producing states, reflecting plans by financing a number of Public Expenditure Re- the oil derivation payments. Fifty-four percent of trans- views, which would primarily focus on the sectors that fers are distributed equally across states, which obvi- had been the largest recipients of subsidies from con- ously works to the disadvantage of the more populous solidated regional budgets, such as housing and utili- states, and a small share--2.5 percent--is distributed ties, agriculture, public transportation, education, and according to the revenue-generation efforts of the health. This long list of components has resulted in very states, which tends to be regressive. The consequence slow implementation: separate terms of reference are is that per capita income plays little role in the even- required for each component and the procurement tual distribution. arrangements must be made for numerous small con- 5. This has caused some problems for the Bank, sultant contracts. which normally requires a sovereign guarantee. In re- 9. The participating regions were: Vologda Oblast, cent years the Bank has developed a program with Samara Oblast, Belgorod Oblast, Chelyabinsk Oblast, IFC for a special Subnational Facility to lend directly to Khabarovsky Krai, and the Republic of Chuvashia. A sev- states or municipalities. Two Russian entities, the Chu- enth region, St. Petersburg City, was subsequently vash Republic (Rural Roads Project, $50 million) and the added to the FFRFRP's first cohort of competitors, but Municipality of Petropavlovsk-K ($28 million) have thus was not eligible for technical assistance under the far received two loans from this Facility. RFTAP . 6. State deficits in India rose from a total of 2.28 per- 10. However, in some specific areas (such as food cent of GDP in 1993/94 to 4.17 percent in 1998/99, subsidy, pricing agricultural supply of power in Andhra while in Brazil the state primary deficit rose from 0.63 Pradesh) political economy factors proved to be too percent of GDP in 1995 to 2.12 percent in 1998. strong to overcome. 7. In Brazil this lending was provided on IBRD 11. Most high-level state officials interviewed by terms, which meant substantially lower interest rates IEG in the four countries welcomed the idea of brief, than would be applied to domestic borrowing. In India, separate, state-specific assistance strategies. loans were provided on IDA terms, but at that point the Indian government treated all external and internal Chapter 4 loans equally and passed them on to the states at the 1. The state of Ceará has confronted serious poverty domestic interest rate of 9.5 percent, with the federal and development challenges, including tough climatic government taking the foreign exchange risk. In 2004 conditions, frequent droughts, low levels of educa- a new policy was adopted of passing on external loans tion, high infant mortality, and scarce social and phys- on a "back-to-back" basis--that is, at the actual inter- ical infrastructure and basic services. At the same time, est rate, but with the state government taking the for- in the past 30 years, Ceará has been one of the better eign exchange risk of the loan. managed and more progressive states in Brazil, with a 8. Proposed project activities included (i) diagnos- long and successful history of cooperation with the tic reviews and development of reform plans in budg- Bank. 110 ENDNOTES 2. Minas Gerais is considered a microcosm of Brazil. 6. Federal and regional authorities were eager to use The per capita income is close to Brazil's national av- the loan financing before the celebration of the 1000th erage and personal and regional income disparities re- anniversary of the city of Kazan. flect those of the country. It is also one of Brazil's largest 7. IEG rated the outcomes of the project satisfactory. and wealthiest states, though with significant poverty 8. The 2002­06 assessment of Bank AAA in India pockets in the north. It has substantial managerial ca- rated it satisfactory overall, but moderately unsatisfac- pacity within the state government and is often com- tory with respect to coherence and internal integration. pared favorably with the federal government in this It also noted the wide variation in quality and the regard. After the fiscal and financial disarray of the late supply-driven nature of the program, with little client 1990s, the new state government (since 2003) showed participation and client-level dissemination, which a strong commitment to reaching fiscal sustainability, made follow-up activities rather uncertain. It also rec- public sector reforms, poverty reduction, and creating ommended greater usage of available local research ca- an enabling environment for private sector develop- pacity as well as capacity-building activities in this area. ment. The Bank now has a sizeable program in the 9. The Bank has a competitive edge over potential state, including a Multisector SWAp, and the state offi- private sector competitors in Russia: the 2005 Federal cials have been very professional and skillful in devel- Law on Public and Municipal Procurement does not oping the results indicators and managing the program apply to the World Bank or to a number of other in- in general. However, it is difficult to point to obvious ternational institutions, making Bank's technical as- value added derived from the Multisector SWAp. Per- sistance more attractive to the client. haps the part of the program that is most appreciated 10. See Appendix H for a more detailed account of by the local authorities is the Bank-financed technical partnership with other donor agencies. assistance, which provides flexible resources to indi- 11. The joint strategic approach in Bihar is a relatively vidual ministries and agencies to use to deepen and fine- recent development and is not covered by this review. tune their programs. 12. In June 2001, ADB and the Bank completed a Co- 3. St. Petersburg has the status of a "Subject of the ordinated Assistance Strategy for roads, focusing on Federation"--equal to the status of a "state" in other state and national highways. The Bank also coordi- countries included in this review. The St. Petersburg nated in other sectors, including inland and coastal project was a hybrid: a combination of a regional DPL waterways. (city component) and a standard investment loan (fed- 13. Lagos is a megacity faced with the challenges of eral component). The rationale was not clear, because sprawling urban growth and infrastructural decay. It is the two components had very little in common. We are also the commercial and industrial capital of Nigeria. not discussing here the investment, or so-called "fed- Slum areas account for about 70 percent of the state's eral" component of the project in St. Petersburg, aimed population, with population growing at 4.8 percent at rehabilitation of city's cultural assets. per annum, and density of 260 people per hectare. 4. Kazan is the capital of the Republic of Tatarstan. Slum areas are generally under-serviced and have lim- At the time of the project was prepared, Kazan did not ited access to basic social services, including roads and have the status of a municipality, which was granted later water. Flooding is a major phenomenon, which has a in the life of the project. strong association with poor solid waste management. 5. Tatarstan is a region of special significance for the The objective of the LMDGP was to increase sustain- federal center. It is one of the largest and most power- able access to basic services through investments in crit- ful national republics, and a stronghold of Islam in Rus- ical urban infrastructure. This project is not progressing sia. In the 1990s Tatarstan enjoyed a high degree of satisfactorily due largely to management and procure- autonomy, including fiscal autonomy (with almost all ment challenges. taxes retained in the republic). To ensure the smooth return of Tatarstan into the system of intergovernmental Chapter 5 fiscal relations, the federal center introduced an ambi- 1. Although the Institute of Economy in Transition, tious program of economic development of Tatarstan. Institute for Urban Economics, and Center for Fiscal Pol- The Bank effectively became a part of that program. icy are physically located in Moscow, they are closely 111 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L involved in regional development issues in Russia. Appendix C Leontief Center is based in St. Petersburg. 1. In 2004, in response to long-standing requests from states, the Indian government decided to change Appendix A the terms on which Bank and IDA financing is on-lent 1. At the state level, only two projects closed with to them. Instead of providing these funds to states on an unsatisfactory rating. One was the Minas Gerais uniform terms (70 percent grant, 30 percent loan at a State Privatization Project, which was cancelled 18 benchmarked interest rate for most states), funds are months after Board approval (and after three years of now passed through on the same terms on which project preparation) because the loan was never signed they are received from the Bank and IDA--in rupees, with the counterparts. At this time, the governor of but including the foreign exchange risk. This arrange- Minas Gerais declared a moratorium of the state debt ment, known as the back-to-back lending system, has to the federal government. As a consequence, the Bank raised some new issues: it transfers the management suspended the right of the state to make withdrawals of foreign exchange fluctuations to the states, and it from the existing loan accounts and stopped preparing increases the complexity of loan administration and of new operations. The other unsatisfactory project tar- managing states' access to IDA. Initially, the new sys- geted the education sector (in Bahia, a lagging state). tem did not diminish the states' interest in borrowing 2. There were only two other projects in the energy from the Bank. The present Global Financial Crisis, sector that closed during this period with a satisfactory however, raised concern on how the state would be rating, one at the federal level (power grid system de- able to finance the additional cost of borrowing due velopment) and one at the state level (Andhra Pradesh to the exchange rate depreciations India has been Power APL). experiencing. 3. The St. Petersburg Loan (SPDP) discussed in this review, is still formally active, because its "federal com- ponent" is still being implemented. 112 BIBLIOGRAPHY AfDB (African Development Bank), DFID (U.K. De- ing a Potential Crisis. World Bank Technical Paper partment for International Development), USAID No. 452. Washington, DC: World Bank. (U.S. Agency for International Development) and .J. Howes, Stephen, Deepak Mishra, and V Ravishankar. World Bank Group. 2009. "Country Partnership 2007. "A Decade of World Bank Sub-National Policy- Strategy for Nigeria, 2010­13." Report No. 46818-NG. Based Lending to India: A Retrospective." In Ragh- World Bank, Washington, DC. bendra Jha, ed., The Indian Economy Sixty Years Burnside, Craig, and David Dollar. 1997. Aid, Policies, after Independence. London and New York: Pal- and Growth. Policy Research Working Paper WPS grave Macmillan. 1777. Washington, DC: World Bank. IFC (International Finance Corporation). 2006. Doing DFID (Department for International Development, Business in Brazil. Washington, DC: World Bank. U.K.). 2004. "Drivers of Change." Public Information IEG (Independent Evaluation Group, The World Bank Note, DFID, London. http://www.gsdrc.org/docs/ Group). Forthcoming. Nigeria Country Assistance open/DOC59.pdf Evaluation. IEG Study Series. Washington, DC: Dillinger, Bill. 1995. "Brazil--State Debt: Crisis and World Bank. Reform." Report No. 14842-BR. Washington, DC: India, Ministry of Finance. 2008. "Indian Public Finance World Bank. Statistics 2007­2008." New Delhi. Dillinger, Bill, and Steven Webb. 2001. Fiscal Manage- Le Houerou, Philippe. 1995. Fiscal Management in ment in Federal Democracies: Argentina and the Russian Federation. Report No. 14862-RU. Wash- Brazil. Policy Research Working Paper No. 2121. ington, DC: World Bank. World Bank, Washington, D.C. Le Houerou, Philippe, and Michal Rutkowski 1996. Elaigwu, Isawa. 2008. Fiscal Federalism in Nigeria: "Federal Transfers in Russia: Their Impact on Re- Facing the Challenges of the Future. London: Ado- gional Revenues and Incomes." Comparative Eco- nis and Abbey Publishing. nomic Studies 38:21­44. Fardoust, S., and A. K. Lahiri. 2003. "Indian States: the Martinez-Vazquez, Jorge, and Jameson Boex. 2001. Rus- Road to Reform." In S. Howes, A.K. Lahiri, and N. sia's Transition to a New Federalism. WBI Learn- Stern, eds., State Level Reforms in India: Towards ing Resources Series Report No. 22026. Washington, More Effective Government. New Delhi: Macmillan. DC: World Bank. Freinkman, Lev. 2007. Intergovernmental Relations Mishra, Deepak. 2003. Unlocking the Growth Oppor- in Nigeria: Improving Service Delivery in Core tunities of Andhra Pradesh. New Delhi: World Bank. Sectors. MPRA Paper No. 10032, University Library Nigeria, Central Bank of Nigeria. Various years. "An- of Munich. http://mpra.ub.uni-muenchen.de/10032/ nual Report and Statement of Accounts." Abuja: Freinkman, Lev, and Plamen Yossifov. 1999. Decen- Central Bank of Nigeria. WWW .cenbank.org tralization in Regional Fiscal Systems in Russia: UNDP (United Nations Development Program). 2006. Trends and Links to Economic Performance. World Niger Delta Human Development Report. Lagos, Bank Policy Research Working Paper WPS 2100. Nigeria: UNDP . Washington, DC: World Bank. Webb, Steven. 2004. "Fiscal Responsibility Laws for Freinkman, Lev, Daniel Treisman, and Stepan Titov. Subnational Discipline: The Latin American Expe- 1999. Subnational Budgeting in Russia: Preempt- rience." WPS 3309. Washington, DC: World Bank. 113 W O R L D B A N K E N G A G E M E N T AT T H E S TAT E L E V E L World Bank. 2008a. "Brazil ­ Ceará Multi-Sector Inclu- ------. 2004e. State Fiscal Reforms in India, Progress sion Development Project." Report No. ICR817. and Prospects. Report No. 28849. Washington, DC: World Bank, Washington, DC. World Bank. ------. 2008b. Brazil within Brazil: Testing the Poverty ------. 2003a. India: Sustaining Reform, Reducing Map Methodology in Minas Gerais. Policy Research Poverty. Report No. 25797. Washington, DC: World Working Paper WPS 4513. Washington, DC: World Bank. Bank. ------ 2003b. Next Steps for Education in Four Se- ------. 2008c. "Doing Business in Nigeria 2008: Com- lected States in Brazil. Report No. 24343. Wash- paring Regulations in 10 States and Abuja, FCT." ington, DC: World Bank. Report No. 44049. World Bank, Washington, DC. ------. 2003c. Nigeria--Policy Options for Growth ------. 2008d. "Orissa in Transition: Challenges for and Stability: Public Expenditure Management 2006­2010." Report No. 44612. World Bank, Wash- and Financial Accountability Review. Report No. ington, DC. 26215. Washington, DC: World Bank. ------. 2007a. "Minas Gerais World Bank Partnership: ------ . 2003d. Nigeria States Finances Study. Report Building on a Strong Foundation and Leading to Next No. 25710. Washington, DC: World Bank. Steps." World Bank Policy Note. Report No. 40036. ------. 2003e. Strategies for Poverty Reduction in Washington, DC: World Bank. Ceará--The Challenge of Inclusive Moderniza- ------ .2007b. Nigeria--A Fiscal Agenda for Change. tion. Report No. 24500. Washington, DC: World Public Expenditure and Financial Accountability Re- Bank. view. Report No. 36496. Washington, DC: World ------. 2002a. Brazil--Issues in Fiscal Federalism. Bank. Report No. 22523. Washington, DC: World Bank. ------. 2007c. Nigeria--Lagos State Finances Review ------. 2002b. Stabilizing Intergovernmental Trans- and Agenda for Action. Report No. 38569. Wash- fers in Latin America: A Complement to Na- ington, DC: World Bank. tional/Subnational Fiscal Rules? World Bank Policy ------. 2007d. Towards Sustainable Mineral-Intensive Research Working Paper WPS 2869. Washington, Growth in Orissa. Report No. 39878. Washington, DC: World Bank. DC: World Bank. ------ . 2002c. State and Local Governance in Nige- ------. 2006. Fiscal Responsibility Legislation and ria. Report No. 24477. Washington, DC: World Bank. Fiscal Adjustment: The Case of Brazilian Local ------ . 2001a. Broadening the Base for Growth: A Re- Governments. World Bank Policy Research Working port on the State of Bahia. Report No. 21377. Wash- Paper WPS 3812. Washington, DC: World Bank. ington, DC: World Bank. ------. 2005a. "Bihar: Towards a Development Strat- ------. 2001b. Public Expenditures for Poverty Alle- egy." World Bank Working Paper No. 32819. World viation in Northeast Brazil--Promoting Growth Bank, Washington, DC. and Improving Services. Report No. 22425. Wash- ------. 2005b. State Fiscal Reforms in India: Progress ington, DC: World Bank. and Prospects. New Delhi: Macmillan. ------. 2001c. Rural Poverty Alleviation in Brazil: ------. 2004a. "Growth and Poverty Reduction in Rio Towards an Integrated Strategy. Report No. 21790. Grande Do Norte: A State Economic Memorandum." Washington, DC: World Bank. Report No. 24891. Washington, DC: World Bank. ------. 2000. Poverty Reduction, Growth, and Fiscal ------. 2004b. India Fiscal Decentralization to Rural Stability in the State of Ceará--A State Economic Governments. Report No. 26654. Washington, DC: Memorandum. Report No. 19217. Washington, DC: World Bank. World Bank. ------. 2004c. Nigeria--State of Lagos Financial Ac- ------. 1999a. Fiscal Management in Federal Democ- countability Assessment. Report No. 29147. Wash- racies: Argentina and Brazil. Policy Research Work- ington, DC: World Bank. ing Paper WPS 2121. Washington, DC: World Bank. ------. 2004d. Piaui State Economic Memorandum: ------. 1999b. Social Exclusion and Land Adminis- Managing a Natural Inheritance. Report No. 24484. tration in Orissa, India. Policy Research Working Washington, DC: World Bank. Paper WPS 2124. Washington, DC: World Bank. 114 BIBLIOGRAPHY ------. 1997. Andhra Pradesh: Agenda for Economic ------. 2005b. "Russian Federation Country Assistance Reforms. Report No. 15901. Washington, DC: World Strategy Progress Report." Report No. 31579. World Bank. Bank, Washington, DC. Photocopy. ------. 1994. Russia and the Challenge of Fiscal Fed- ------. 2004a. "Country Strategy for India." World eralism. Washington, DC: World Bank. Bank, Washington, DC. Photocopy. World Bank Group. 2008a. "Country Partnership Strat- ------. 2004b. "Second Joint Interim Strategy Progress egy for Brazil, 2008­11." Report No. 42677. World Report for the Federal Republic of Nigeria." Re- Bank, Washington, DC. Photocopy. port No. 29060. World Bank, Washington, DC. ------. 2008b. "Country Partnership Strategy Report, Photocopy. Nigeria, FY 2005­09." Report No. 42771. World Bank, ------. 2001. "Interim Strategy Update for the Federal Washington, DC. Photocopy. Republic of Nigeria." World Bank, Washington, DC. ------. 2008c. Country Strategy for India, FY 2009­12. Photocopy. World Bank, Washington, DC. Photocopy. ------. 2000. "Nigeria--Country Strategy Note." Re- ------. 2006. "Russian Country Partnership Strategy. " port No. 20309. World Bank, Washington, DC. Report No. 3790. World Bank, Washington, DC. Photocopy. Photocopy. ------. 1999. "Country Assistance Strategy for the ------. 2005a. Country Partnership Strategy--Nige- Russian Federation." World Bank, Washington, DC. ria, 2005­09. Report No. 32412. World Bank, Wash- Photocopy. ington, DC. Photocopy. 115 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. IEG-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. IEG 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. ISBN 978-0-8213-8224-0 90000 9 780821 382240 SKU 18224