56807 FAST TRACK BRIEF September 21, 2009 The IEG report "World Bank Engagement at the State Level: The Cases of Brazil, India, Nigeria and Russia," was discussed by CODE on September 21, 2009 World Bank Engagement at the State Level: The Cases of Brazil, India, Nigeria and Russia Beginning in the late 1990s, the World Bank significantly expanded its engagement at the state level in Brazil, India, Nigeria, and Russia. This pilot cross-country study reviews the selected cases of World Bank's lending and analytic work at the state level in those four large federated countries. In each case, state governments were the Bank's principal development partners. The study looks at the evolution of the four country strategies and the Bank's mode of engagement at the state level, in order to draw lessons from that experience both for the Bank and for its federal and state partners. State-level engagement posed several strategic and operational questions, among them which states to engage, the scope of engagement, and the modalities of engagement. The Bank set out its approach to selecting states in country strategy documents. Two tendencies--often in tension-- featured in most approaches. One was to support better-performing, reformist states ("lead or focus states" approach). The other was to support the poorest states as a more direct route to reducing poverty. Concerning the scope of engagement, the initial and principal area of engagement was typi- cally fiscal reform - fiscal sustainability, medium-term fiscal frameworks, strengthening the public financial management capacity of state governments, and fiscal federalism. In some states, Bank in- volvement extended beyond fiscal reform to multisector engagement focused on the growth and poverty reduction agenda. The modalities of engagement and the instruments deployed by the Bank evolved over the review period and included state-level development policy loans, multisector results- based investment lending, and reimbursable technical assistance. There was considerable successful innovation in this area, yet little knowledge sharing between countries. The following findings are worth highlighting. First, the study confirms the desirability of continued selective Bank lending in a few states. However, the poverty impact of those interventions could be enhanced by balancing states' propensity to reform and the concentration of poverty within them, giving greater weight to the needs of poorest states. Second, continued focus on public finance management as the core area appears sound, irrespective of whether engagement is confined to this area or serves as an entry point for broader engagement. And third, there is considerable scope for greater impact from analytic work, knowledge transfer, and expanded knowledge sharing -- not so much concepts and theories as practical experience of what works and what does not. state level in these countries more rewarding given the client's Evolution of Bank Strategy greater interest in the Bank's financial and knowledge re- Why State-Level Engagement? sources. O ver the past decade, the World Bank significantly Which States? expanded its engagement at the state level in four The shift to the state level presented the Bank with a number large federations: Brazil, India, Nigeria, and Russia, of operational issues. Among them was which states the Bank mainly through lending, but also through policy dialogue, should engage. The four countries have large numbers of technical advice, and analytic work. Both demand and supply states--26 states and a Federal District in Brazil, 28 states and factors contributed to this expansion. 7 Union Territories in India, 36 states and a Federal Capital On the demand side, federal governments focused on fiscal stabi- Territory in Nigeria, and 83 Regions ("subjects of the federa- lization following the financial crises of the late 1990s. They tion"--republics, oblasts, krays, and okrugs) in Russia. Working saw the potential for the Bank to provide state governments in all of them would obviously be beyond the Bank's budgetary with incentives for reform through financing, while encourag- and human resource capacity. The Bank defined the strategic ing discipline through agreed policy measures and providing approach to the selection of states in its Country Assistance technical support for implementation. Federal governments Strategies (CASs). Some attempts were made to develop had limited scope for differentiating between states based on quantified criteria for selective engagement, but the Bank factors such as commitment to reform and saw in the Bank's generally preferred to keep the criteria broad to allow for flex- capacity to do this a means of assisting states willing to take a ibility. It is clear from all the countries reviewed that there was lead, with the possibility of a demonstration effect for other tension between the Bank's interest in identifying progressive, states. State governments were also eager to borrow from the reforming states that could serve as demonstrations to others Bank because its loans, denominated in foreign currency, and its interest in supporting poverty reduction by assisting the generally came at lower rates than provided by the federal poorest states. In addition to these two key criteria-- government or the domestic market. Bank loans, while often effectiveness of assistance and need (poverty)--another no less financially modest at the federal level, could be a major source important criterion was the political economy, unique in each of financing at the state level. State governments welcomed the country and including (but not limited to) relations between the Bank's focus on their economy as well as the associated dialo- federal center and the states; capacity and political affiliation of gue, advice, and analytic work. In addition, with the increasing the state-level leadership; level of trust and relationship of the concern about meeting or achieving outcomes with regard to Bank teams with the clients; and local politics and electoral the Millennium Development Goals (MDGs), both federal and cycles. state governments saw the Bank as having a comparative ad- In Brazil, the shift toward states was proposed in the 1995 vantage in supporting better service delivery in the relevant CAS, directing lending to creditworthy reforming states. The social and economic sectors at the state level. 2000 CAS continued this approach using criteria established by On the supply side, with the combination of fiscal stabilization the federal Fiscal Responsibility Law (FRL) and stressed the and improvement in the fiscal situation of the four countries intention of providing assistance to the states of the North- during the commodity boom of 2000-07, there was a limited East region with highest poverty levels. Lending to states be- appetite to borrow from the Bank for federal programs (this came more multisectoral, and significant innovations were in- trend was more pronounced in Brazil and Russia; Nigeria, an troduced, such as state-level sectorwide lending (multisector IDA borrower, was an exception; in India, federal level SWAp) and state-level development policy loans (DPLs). Both borrowing increased slightly in 2004-07). A level of Bank instruments were applied in states that had turned the corner engagement commensurate with the size and importance of fiscally, despite significant disparity in their income levels. The these countries therefore almost mandated the shift to the 2008 Country Partnership Strategy (CPS) focused on a technic- state level, where demand remained buoyant. The increasing al assistance program of modest size with the federal govern- focus of the Bank on poverty reduction after 1995 was also ment and a major demand-driven lending program with states, an important factor. There is a distance between federal-level conditioned on perceived commitment, ownership of reforms, programs and results on the ground in such large countries. and fiscal responsibility. The Bank engaged with some of the Most of the public expenditure categories most closely asso- more prosperous and reformist states. Although initially the ciated with poverty reduction in the short and long term Bank attempted to expand investment lending, this proved usually are state responsibilities in these countries. Therefore, cumbersome given the two-tier approvals required by the state increasing the Bank's impact on poverty reduction meant and federal government. As a result of joint consultations with increasing the focus on and activities at the state level. In ad- state and federal counterparts, by 2008 the composition of dition, many Bank country and sector staff found work at the state-level lending shifted toward cross-sectoral operations in 2 support of economic policies and public sector reforms (DPLs wide sectoral programs. The decision about which states to and multisector SWAps). engage, for example, in support for agriculture and rural devel- opment in India or education in Brazil, was sometimes a matter In India, the Bank shifted its focus to the state level following of strategic choice, but more often a matter of historical the 1997 CAS. At that time, states were facing financial prob- engagement or the availability of analytic work, based in turn lems, and both the federal and state authorities were keen to on opportunistic involvement of the Bank in particular states. tap the Bank's resources and take advantage of technical assis- In the new context, the Bank was steering toward a new model tance. The Bank opted for major involvement in progressive of engagement. This tended to have two elements: support for reforming states ("focus states" approach). The 2004 CAS fiscal reform and broader multisector engagement at the state level. signaled a change of strategy, noting that the focus on reform- ing states was resulting in neglect of the lagging states. There- Fiscal Reform fore, the CAS proposed the provision of technical assistance to the lagging states and to try shifting lending to them as well. The Bank expanded its involvement in fiscal reform in This proved to be difficult. As the 2009 CAS completion selected states. With the focus on stabilization and the need report noted, while lending at the state level remained a large to reduce growing state deficits or enhance state public share of the overall program, the share of lagging states in the expenditure management capacity, the Bank engaged in two program had actually declined. areas. The first of these was fiscal federalism. Fiscal relations between the federal and state governments are politically In Nigeria, the Bank reactivated and intensified its lending highly sensitive, since this is often at the core of balancing activities after return to civilian rule in 1999. During that regional interests. From a strictly economic perspective the period, Bank strategy passed through two phases: a period of resources available to the state governments need to be interim strategies (FY00-05) following re-engagement; and balanced with their expenditure responsibilities. For many the FY05-FY09 CPS, when the Bank adopted a medium-term taxes it is more efficient to centralize collections. Hence, focus. Engagement at the state level in Nigeria was largely generally there is an imbalance between the revenues driven by the social and poverty reduction agenda, with focus collected by the states and their developmental mandate. In on improving infrastructure and providing support for agri- Brazil this imbalance is relatively modest, but in Nigeria it is cultural and rural development. During the CPS FY05-09 very large. This means that the federal government is required period, the Bank's strategy formally moved to focus on well- to transfer resources to the state, generally based on various performing states ("lead states"), seeking to leverage state formulae that take account of population, per capita income, efforts and resources by granting them access to a "perfor- and the state government's own tax effort. In the countries mance package." Five states were selected based on the gov- reviewed, federal transfers have not been very effective in ernment-led SEEDS benchmarking process. reducing disparities in expenditure capacity between states. In Russia, the 1997 CAS emphasized regional investment A second key issue in fiscal federalism concerns discretionary projects (despite recognition that they were expensive to transfers from the center to the states, usually intended to prepare and supervise). The next CAS (1999) outlined a phased provide an incentive to states to undertake high priority shift in lending away from investment projects in infrastructure programs. The Government of India has used such schemes a and energy in favor of increased emphasis on systemic aspects great deal. For the Bank, the political sensitivities make fiscal of institutional development. The 2002 CAS continued the federalism a difficult area for intervention unless there is an strategy shift, emphasizing support for reforms at the regional explicit request from the federal government. The Bank has level, particularly in strengthening public sector management. undertaken substantial analytic work in this area, with Russia, The 2005 CAS Progress Report stated that work at the regional where new fiscal relations were being defined in the past level was to be carried out in a multisectoral manner and would decade, as a very good example. concentrate on a small number of regions in agreement with A second and far larger part of Bank engagement in fiscal the federal government. The 2007 CPS added a finishing reform is its direct support to public finance management at the touch: the plan for gradual shift to the new modalities of coop- state level, including enhancements in tax capacity, moderniz- eration and instruments, such as the subnational facility that ing the tax structure, developing a sustainable fiscal policy and allows the Bank Group to provide funds without a sovereign medium term expenditure framework, and improving budget guarantee to states and municipalities; and provision of tech- and expenditure management. The Bank's engagement model nical assistance on reimbursable basis (fee for service). generally started with a trigger mechanism that required states to show commitment to fiscal reform. This requirement was The Scope of Engagement highly formalized in Brazil where the federal government The second issue for the Bank was the scope of its engagement. requires strict adherence to the Law on Fiscal Responsibility In Brazil, India, and Nigeria, the Bank had had numerous state- (FRL). In India and Nigeria the requirements were less forma- specific projects in the past as part of its support for nation- 3 lized but generally related to timely budgeting and reporting. In the Bank had struggled to find an instrument to attach its policy Russia, criteria were established for participation in the fiscal dialogue and strategic approach to at the state level. The multi- reform projects supported by the Bank. Once the triggers were sector restructuring loan in Andhra Pradesh, an investment loan met, the Bank was able to further support fiscal reform undertaken in 1998, was a way to accomplish this, but it was an through an engagement model that combined analytic work enormously costly operation to prepare and supervise. Adjust- with multisector lending and focused technical assistance in the ment lending rapidly became the instrument of choice to areas of fiscal and governance reform. Elements of this model support fiscal reform and statewide strategies in Brazil and India, are present in each country. In Lagos, Nigeria, for example, in- but in Brazil there was a sense that it was less effective in reach- tensive analytic work at the state level was combined with in- ing out to the line ministries in key sectors. The Bank's Brazil vestment lending and technical assistance, but the Bank is only country team developed the innovative approach of a multisec- now considering the possibility of using multisector lending in tor SWAp, a results-based instrument with target indicators support of its approach. In Andhra Pradesh and Orissa in India, defined for each sector and disbursements associated with the Bank carried out analytic work and subsequently used achievement of the targets. This had very positive outcomes in multisector loans to support fiscal reform, but technical advice, that it brought to the fore the linkages required to achieve except insofar as it was embedded in the analytic work and results, such as the need for improved water supply in order to lending activities, has not been a focus of the approach. The reduce infant mortality. Another important innovation was the Bank's involvement with tax policy has been relatively light, pioneering of reimbursable technical assistance at the state level with the focus of its efforts being on budgeting and public in Russia. Bank budgets rarely allow the level of analytic work expenditure management systems and increasingly on gover- demanded by intensive engagement in three to five states, and an nance issues associated with transparency and accountability approach whereby states can pay for additional work has consi- for efficient service delivery. derable promise for other middle-income countries. Multisector Engagement Findings In addition to fiscal engagement, the Bank has also undertaken Overall, the analysis leaves little doubt that the Bank engage- a broader multisector engagement at the state level, which has ment at the state level did add value. There was a great deal of focused on the growth and poverty reduction agenda. This has enthusiasm at both the federal and state levels in these coun- involved a mix of analytic work, adjustment lending, and tries regarding the Bank's contribution and a large number of investment lending. In only a few cases has the Bank produced specific achievements, ranging from fiscal reform in Orissa, an explicit strategy for its activities in a particular state. Usually India, to effective technical assistance and capacity building in these were covered in broad-brush fashion in the CAS. At its Lagos, Nigeria, to a wide range of achievements in Ceará, most developed, as in Andhra Pradesh in India and in Ceará in Brazil, and improved fiscal management in St. Petersburg, Brazil, the Bank programs combined investment lending in Russia. Although state level engagement often requires most of the core economic and social sectors with multisector additional effort and can be resource-intensive, it is usually lending. The objective was to derive synergies from the combi- worth the cost. In many states and regions the Bank's nation of activities. In Andhra Pradesh, for example, the Bank program was pivotal in reinforcing the efforts of pro-reform was explicit in its view that the difficult measures required for government officials and helped to develop an expert fiscal reform needed to be matched by increased investment in community and a pool of local consulting institutions that agriculture, rural development, health, and education to pro- were drawn into the process of public policy formulation and vide a politically acceptable package of reforms. The selection implementation. of states for this broader engagement focused more on fiscal reform than on the broader poverty issue. The Bank spent a The main findings of this review, which may be helpful in great deal of time in these countries supporting relatively high- guiding the organization of future work at the state level, income, high-capacity states (such as Minas Gerais in Brazil, include: St. Petersburg in Russia, and Karnataka in India). While this support added value, it came at the expense of Bank efforts in On selection of states: poorer states that lack capacity. · The strategy to concentrate lending services on a few states to enhance the impact of the Bank's program is Modalities of Engagement right in principle, but selection criteria and the mode of How to engage was also a concern for the Bank during the period, implementation could give greater weight to the needs with considerable evolution in the approach. The first bridge to of the poorest states. be crossed was the use of adjustment (development policy) lend- ing at the state level (the first such adjustment loan was made in · Bank engagement with high-performing states clearly March 2000 to the state of Uttar Pradesh in India). Until that point added value, both strengthening in-state capacity and 4 encouraging state-to-state knowledge transfer, albeit mainly between the high performers. However, there is little evidence that it had the desired demonstration effect on poor, lagging states on the scale that the Bank hoped for, or that the Bank had an exit strategy About Fast Track Briefs to permit increased focus on poorer states over time. Fast Track Briefs help inform the World Bank Group (WBG) At the same time, experience shows that it is possible managers and staff about new evaluation findings and recom- to achieve results in the poorest states through persis- mendations. The views expressed here are those of IEG and tent work with committed state counterparts and should not be attributed to the WBG or its affiliated organiza- strong partnership with the federal government and tions. Management's Response to IEG is included in the pub- other donors. lished IEG report. The findings here do not support any general inferences beyond the scope of the evaluation, including any infe- · It is important to stay engaged not only in states that rences about the WBG's past, current or prospective overall are able to borrow from the Bank, but also in states performance. that have no fiscal space to borrow but demonstrate a genuine commitment to development that can be supported through analytic work and technical assis- The Fast Track Brief, which summarizes major IEG evalua- tance. tions, will be distributed to selected World Bank Group staff. If you would like to be added to the subscription list, please email On the scope of engagement: us at ieg@worldbank.org, with "FTB subscription" in the subject line and your mail-stop number. If you would like to · Continued focus on public finance management as the stop receiving FTBs, please email us at ieg@worldbank.org, core area for state-level work appears sound, whether with "FTB unsubscribe" in the subject line. engagement is confined to this area or it serves as an entry point for broader engagement. · The lending programs and Bank budgets in some Contact IEG: states are often larger than for many countries the Director-General, Evaluation: Vinod Thomas Bank engages with. For those states where the Bank Director: Cheryl Gray (IEG-WB) plans or has a major engagement, a brief state strategy Manager: Ali M. Khadr (IEGCR) document could be a useful tool for defining the scope Task Manager: Konstantin Atanesyan (IEGCR) of engagement and developing a medium-term outlook. On modalities of engagement: Copies of the report are available at: · There is considerable scope for greater impact from http://www.worldbank.org/ieg/sle knowledge transfer and expanded knowledge services. IEG Help Desk: (202) 458-4497 In particular, there is strong demand for better know- E-mail: ieg@worldbank.org ledge sharing, both within the Bank and across the countries concerned. This is not so much a matter of sharing of concepts and theories as of practical experience regarding what is working and what is not. · Widening the scope and increasing the amount of ana- lytical work at the state level could be helpful in identify- ing high-impact, high-priority areas. There seems to be a potential for closer partnership between state gov- ernments and the Bank in this area. 5