Document of The World Bank Report No. 14394-BR STAFF APPRAISAL REPORT BRAZIL NORTHEAST RURAL POVERTY ALLEVIATION PROGRAM RURAL POVERTY ALLEVIATION PROJECT - SERGIPE June 6, 1995 Natural Resources Management and Rural Poverty Division Country Department I Latin America and the Caribbean Regional Office CURRENCY EQUIVALENTS (as of April 12, 1995)' Currency Unit = Real (R$) R$0,91 = US$1.00 R$1,00 = US$1.10 R$1 million = US$1.1 million FISCAL YEAR (State of Sergipe) January 1 to December 31 WEIGHIS AND MEASURES The metric system has been used throughout the report. Appraisal mission. BRAZIL NORTHEAST RURAL POVERTY ALLEVIATION PROJECT - SERGEPE GLOSSARY OF ACRONYMS APCR Apoio as Pequenas Comunidades Rurais (Program of Support to Small Rural Communities under the original Northeast Rural Development Program) CAS Country Assistance Strategy EMATER Empresa de Assistencia Tecnica e Extensao Rural (Technical Assistance and Rural Extension Agency) FAO/CP World Bank/FAO Cooperative Program FUMAC Conselho Municipal de Apoio Comunitirio (Municipal Community Schemes) FUMAC-P Fundo Municipal de Apoio Comunitario - Piloto (Pilot Municipal Community Funds) GOB Governo Federal do Brasil (Federal Government of Brazil) IBGE Funda,co Instituto Brasileiro de Geografia e Estatistica (Brazilian Institute of Geography and Statistics) HICA Instituto Interamericano de Coopera,co para a Agricultura (Inter-american Institute for Agricultural Cooperation) IPEA Instituto de Pesquisa Econ6mica Aplicada (Institute of Applied Economic Research of the Secretariat of Planning of the Presidency of the Republic) MIS Sistema Gerencial de Informa,6es (Management Information System) NCB Licitaqao Nacional (National Competitive Bidding) NGO Organizagao Nao-Governamental (Non-Governmental Organization) NRDP Programa de Apoio ao Pequeno Produtor Rural do Nordeste (Northeast Rural Development Program) O&M Operaqao e Manutencao (Operation and Maintenance) PAC Programa de Apoio Comunitario (State Community Schemes) PCU Unidade Coordenadora do Projeto (Project Coordinating Unit) POA Plano Operativo Anual (Annual Operating Plan) POLONORDESTE Programa de Desenvolvimento Integrado do Nordeste (Northeast Integrated Development Program) PPR Physical Performance Review PRONESE Projeto Nordeste do Estado de Sergipe (Northeast Program of the State of Sergipe) SEAIN Secretaria de Assuntos Internacionais (Secretariat of International Affairs of the Ministry of Planning of the Presidency of the Republic) SEPLAN Secretaria do Planejamento e da Ciencia e Tecnologia (State Secretariat of Planning, Science and Technology) SOEs Prestaqao de Contas (Statements of Expenditure) SSMP Sistema Simplificado do Monitoramento do Projeto (Simplified Project Monitoring System) UT Unidade Tecnica Technical Unit BRAZIL NORTHEAST RURAL POVERTY ALLEVIATION PROJECT - SERGIPE TABLE OF CONTENTS Page No. I. RURAL POVERTY IN BRAZIL. THE NORTHEAST AND SERGIPE ... .... 1 The Rural Poor and the Economy .............................. 1 Rural Poverty in the Northeast ........................... ..1 Experience with Past Bank Lending. 3 The State of Sergipe. 5 Main Lessons Learned from the Northeast Rural Development Program. 7 II. THE PROJECT..9 .................... ........ 9 Origin and Rationale for Bank Involvement .. .. 9 Project Objectives .... 9 Project Components ...................................... 10 A. Community Subprojects .. 10 Municipal Community Schemes (FUMAC) ......... 11 Pilot Municipal Community Funds (FUMAC-P) ... ...... 11 State Community Schemes (PAC) .. 11 B. Institutional Development .. 12 C. Project Administration, Monitoring and Evaluation .. . 13 Decentralization and Thrgeting ................................ 14 Decentralization .. 14 Project Targeting ................ 14 Project Coordination Arrangements .. 14 Froject Coordinating Unit .. 14 Project Municipal Councils .. 16 This report incorporates the findings of a Bank mission which visited Brazil in March-April, 1995. The mission consisted of Messrs./Mmes. L. Coirolo (Mission Leader, Portfolio Manager), T. Barbosa (Task Manager, Economist), R. Caminha (Agriculturalist), J. Barbosa (Agriculturalist), M. Rieth (Economist), J. van Zyl (Agricultural Economist), L. Sonn (Economist, FAO/CP), J.C. Sallier (Civil Engineer, Consultant) and V. Valle (Staff Assistant). The report also benefited from contributions from A. Roumani (Rural Development Specialist) and C. Bryant (Institutional Specialist). The Division Chief is M. Wilson (LAINR) and the Department Director, G. T. Nankani (LA1DR). Peer reviewers are J. Tendler, (Professor of Political Economy, MIT) and T. Jones (LA1HR). TABLE OF CONTENIS (Continued) Project Implementation Arrangements ........................... 16 Community Subproject Cycle ......................... 16 Project Operational Manual . ........................ 17 Operation and Maintenance (O&M) ........................ 18 Operational Checks and Balances ........................ 18 M onitoring ........ ......... . . .................. 18 Project Costs . .......... 19 Financing Plan ........ ... . . . . . . . . .19 Procurement ......................21 Disbursements and Special Account .......................... 23 Accounts and Audits ......................... 23 Environmenital Aspects ......................... 24 Poverty Category ......................................... 24 Women's Participation.. . .......... 25 Project Benefits.. . ......... 25 Project Risks.. . ......... 26 II1. SUMMARY OF AGREEMENTS REACHED AND RECOMMENDATION .... 27 ANNEXES: I. Socio-Economic Indicators . 29 2. The Reformulated NRDP: Performance and Evaluation .31 3 Targeting Mechanisms and Project Municipalities .52 4. Institutional Development .55 5. The FUMAC-P Pilot Program .60 6. Subproject Operation and Maintenance . 63 7. Project Monitoring, Reporting and Evaluation .65 8. Project Costs, Financing Plan and Disbursement Schedule .73 9. Project Operational Manual (Outline) .77 10. Supervision Plan .79 11. List of Documents Available in Project Files .81 MAP: IBRD No. 26945 BRAZII, RURAL POVERTY ALLEVIATION PROJECT - SERGIPE LOAN AND PROJECT SUMMARY Borrower: State of Sergipe Guara'tor: Federative Republic of Brazil Implementing Agency: Administrative Unit for the Sergipe Northeast Project (PRONESE), of the Sergipe State Secretariat of Planning, Science and Technology (SEPLAN) Beneficiaries: Poor rural community associations, including a target population of 330,000 people (about 66,000 families) Poverty: Program of Targeted Interventions Amount: US$36 million equivalent Terms: Repayable over 15 years, including 5 years of grace, at the Bank's standard variable interest rate Commitment Fee: 0.75% on undisbursed loan balances, beginning 60 days after signing, less any waiver Onlending Terms: Not applicable Financing Plan: See Annex 8, Table 8.2 Economic Rate of Return: Not applicable Staff Appraisal Report: Report No. 14394-BR Map: IBRD No. 26945 Project Identification No.: BR-PA-38885 I. RURAL POVERTY IN BRAZIL, THE NORTHEAST AND SERGIPE The Rural Poor and the Economy 1.1 Poverty in Brazil has a strong rural and regional dimension. Around 50 percent of the Brazilian poor live in rural areas, and the incidence of poverty among the rural population is more than double that in large cities and urban areas. Economic instability over the past two decades has been a major obstacle to redressing rural poverty. The implicit tax resulting from inflation has penalized the poor disproportionately. Real economic growth fell from 9 % annually in the 1970s to 2.7% in the 1980s, and for more than a decade there was little improvement in standards of living. Economic and sectoral policies did not provide support for sustainable investment, as a consequence savings fell, inflation accelerated, and consumer prices increased as much as 30% a month. Six separate attempts at adjustment between 1986 and 1994 failed. Austerity programs associated with the different adjustment attempts cut back investment for rural development, in particular for primary education, targeted health interventions, rural rcads, water supply and small farner agricultural services. Migration from rural areas in the 1980s and early 1990s--motivated by unemployment, underemployment and lack of social and other services--created an enormous new burden on urban centers, and underscored the urgency of addressing rural poverty to stem its conversion into an urban poverty phenomenon. 1.2 Introduction of the Plano Real in July 1994 has achieved some degree of macroeconomic stabilization, with the annual inflation rate already decreasing to a fraction of its former level. However, in the short to medium term, stabilization also implies a very strong fiscal adjustment with tightening of government expenditures, high real interest rates and, hence, a deepening of the investment pause in the farm and rural sector. The consequences in terms of rural employment and social services may be quite serious, and clearly call for offsetting safety net actions. While successful macroeconomic adjustment should promote longer-term economic growth with benefits extending to rural areas, international experience shows that targeted programs and policies for the socio-economic development of the poor remain an essential complement to the adjustment process.' Rural Poverty in the Northeast 1.3 The Northeast has long constituted the single largest area of rural poverty in Latin America. This Region, covering nine states and part of a tenth,2 accounts for 19% of Brazil's total land area and 30% of its 147 million population. More than half of all Brazilians living in poverty, and almost two-thirds of the country's rural poor, live in the Northeast. According to a recent study by the Ministry of Planning and Budget, some 12 million rural inhabitants of 1. Brazil: A Poverty Assessment, Human Resources Operations Division, March 31, 1995. 2. The Northeast Region includes the States of Maranhao, Piaui, Ceara, Rio Grande do Norte, Paraiba, Pernambuco, Alagoas, Sergipe and Bahia and the northern part of the State of Minas Gerais. -2- the Northeast live in extreme poverty, with annual per capita incomes under US$214 (less than one-tenth of the national average). Among the underlying causes of rural poverty in the Northeast are the relatively poor resource base of large parts of the region and variable agro- climatic conditions which make them vulnerable to drought. Skewed access to land and the virtual absence of a functioning rural financial system for the poor are additional constraints. Northeast agriculture is characterized by limited input use and slow rates of technology adoption. Productivity is low, with output per farm worker less than half that of other regions. Overall Northeast agricultural GDP growth was negligible during 1991-1994. 1.4 About 40% of the Northeast's rural population lives in the interior semi-arid zone characterized by poor soils and severe, cyclical and often protracted drought. The "drought polygon" in the semi-arid region includes most of Ceara, Rio Grande do Norte, Paraiba, Pemambuco and Bahia, and smaller proportions of all the other Northeastern states except Maranhao. The remainder of the rural population lives in areas which are generally more humid and have better soils, the coastal belt and the drought-prone transitional zone between the coast and the semi-arid area. The rural economy revolves around food crops including fruits and vegetables, extensive livestock grazing by larger farmers, and, as a consequence of the uncertain climate and market fluctuations, basic food production and small-scale animal husbandry by tenants, sharecroppers, landless laborers and small farm owners. The rural poor in the Northeast rely increasingly on a complex of activities--traditional subsistence agriculture, cash crops (mainly cotton and cashew), casual agricultural and non-agricultural work, and remittances from family members living in cities. 1.5 Health and social indicators attest to the poor quality of life in the Northeast, compared to the rest of Brazil. The region has: (i) an infant mortality rate of 75 per 1000 live births, compared to a national average of 58; (ii) a rate of child malnutrition at 27% (of children under five), compared to 13% nationwide; (iii) a literacy rate for heads of household of 56%, compared to 78% nationwide; (iv) 57% of households lacking water supply compared to 28% nationwide; and (v) 73 % of households without access to proper sanitation facilities, compared to 46% nationwide.3 1.6 The new Brazilian administration is preoccupied with the level of poverty nationwide, but particularly in the Northeast and its rural areas, and has declared the Northeast a priority for public development programs. The cornerstone of the Federal Government's effort to combat poverty and hunger in Brazil is the recently announced Comunidade Solidaria program, a broad framework for the coordination of rural and urban poverty programs and projects, focusing on municipalities where poverty is most severe. Implementation would be decentralized, stressing community self-help and local empowerment, and concentrating on four main themes: rural development, income and employment generation, food and nutrition, and urban services. The proposed Sergipe Rural Poverty Alleviation Project is fully consistent with this strategy, parts of which draw heavily on the Bank-supported NRDP experience in Northeast Brazil (para 1.12-1.16). 3. Brazil: A Poverty Assessment, Human Resources Operations Division, March 31, 1995. - 3 - 1.7 The proposed project would also be a timely complement to important changes in the political economy of Brazil, stemming from the new federalism embodied in the Constitution of 1988. Fiscal resources, priority setting and program implementation responsibilities are increasingly being decentralized from the Federal Government to the states, municipalities and local populations. This encouraging trend should help avoid many of the financial and managerial problems which generations of Northeast poverty programs encountered under centralized coordination and implementation arrangements. Much of the Bank's dialogue and lending for poverty, including in rural development, health and education, have now shifted to the states, with the role of the Federal Government being redefined to stress more strategic functions, such as broad policy and norm setting and financial transfer schemes to improve equity. Experience with Past Bank Lending 1.8 The Federal and State Governments have been involved since the early 1970s in financing schemes to alleviate rural poverty in the Northeast. The World Bank has been associated for many years with these efforts, through its financing of a series of programs and projects, including area-specific and state rural development projects, various sectoral operations and a regional rural development program. 1.9 Historically, rural development programs in the Northeast have coalesced around two themes: drought relief and the integrated development of selected areas. The first involved emergency relief programs or projects to increase the productivity of scarce water resources, including large public irrigation schemes. While the budgetary impact of these projects has been tremendous, their poverty effects have been limited and often temporary. The second approach, that of integrated subregional development programs, included initiatives like the Sertanejo project and the Bank-financed POLONORDESTE projects of the 1970s and early 1980s. 1.10 POLONORDESTE. The POLONORDESTE operations concentrated on specific microregions within each of the nine Northeast states and northern Minas Gerais. They were comprehensive, multisectoral projects designed to improve agricultural efficiency, raise rural incomes and increase employment. They typically included components for agricultural modernization (extension, research, credit, marketing, land administration), infrastructure (rural water supply, roads and irrigation) and social investments (education, health and community development). Many important sectoral initiatives were first piloted under the POLONORDESTE projects: farm management extension techniques; integrated farm unit research for small-scale agriculture, rather than crop-based research; a wide range of land redistribution initiatives and modem land titling technologies; educational curriculum and school calendar reforms for rural areas; and community-based primary health care schemes. 1.11 Nevertheless, evaluation of POLONORDESTE by the Bank in the early 1980s showed that, in the aggregate, they had only a limited impact on poverty. The micro-regions supported by the Bank projects were relatively small areas in the context of the whole Northeast; poverty targeting mechanisms were weak; and delivery mechanisms relied exclusively on federal and state bureaucracies. The Federal Government frequently failed to provide adequate - 4 - counterpart funds and delayed the release of budgeted amounts, with damaging effects on the projects given Brazil's chronically high inflation. Consequently, less than one-third of project funds reached intended beneficiaries, as seriously reduced budgets were absorbed by administration and overheads. 1.12 Sectoral Loans and the Northeast Rural Development Program. These considerations prompted the Bank and Federal Government to change strategies in the mid- 1980s. On the one hand, the lessons learned from some of the pilot efforts under POLONORDESTE were promising, but it was felt that these needed to be expanded geographically and supported by broader sector policy reforms, within the framework of single- sector operations unencumbered by the coordination and implementation difficulties of multiple components. A series of sectoral operations such as Northeast land titling, education, health, and irrigation resulted in the next few years. 1.13 With respect to integrated rural development, this was still considered to be a viable and necessary approach but, again, it was felt that the effort should be expanded statewide in order to have a broader impact, and that the number of components should be drastically reduced. As a result, in 1985, the Northeast Rural Development Program (NRDP) was established to alleviate rural poverty and improve the living standards of small farm families in the Northeast. Under this program, the Bank approved during 1985-87 individual loans for the ten states in an amount totalling US$826.7 million. But, through 1993. with the exception of one component supporting small community projects (known as the APCR), implementation of the NRDP lagged behind schedule and performance in meeting basic objectives was weak. Some of the main causes were strikingly similar to the causes of poor performance under the earlier POLONORDESTE projects: chronic fiscal deficits and other macroeconomic distortions led to almost constant counterpart funding shortages and widespread deterioration of many of the public institutions responsible for carrying out the NRDP. The Federal Government devolved some responsibilities to the states, but, by and large, institutional arrangements were still highly centralized. Decision-making, administration and financing arrangements were generally non- participatory, and the sense of project "ownership" at both state and local levels was limited. 1.14 On the other hand, the innovative APCR community-based components (para. 1. 13), which relied on local planning and implementation, stressed community organization and contracted out needed technical expertise, were noticeably more successful. They acknowledged the expressed needs of the communities, their poverty alleviation impact was positive, the cost per family relatively low, and disbursement and commitment rates far exceeded those of other components. 1.15 Reformulated NRDP. On the Bank's initiative, in 1993 the NRDP was reformulated and transformed in its entirety into a community-based development program, drawing both on the successful experience of the APCR small community projects component and on lessons learned with similar schemes elsewhere in Latin America, particularly the Mexican Solidaridad program. Performance has since improved markedly. Rural communities are participating in the identification, financing and implementation of subprojects that meet their most pressing needs, increase their sense of project ownership, and have a much sharper focus on poverty alleviation. The reformulated NRDP is covering all members of poor rural - 5 - communities and not only those with productive assets, extending beyond production and income, and based on a "matching grant" mechanism linked to beneficiary contribution towards subproject cost. The State Governments are fully responsible for counterpart funding, and the corresponding sense of ownership of the program has also improved. 1.16 Although only 18 months have passed since the reformulation, based on implementation so far (as well as the earlier experience of the APCR components), certain facts are clear. The program has generated unprecedented enthusiasm among beneficiaries and favorable multiplier effects, and has mobilized additional funds from municipalities and communities. As much as 95% of disbursements are reaching targeted beneficiaries, and diversion of funds for non-intended uses has been sharply reduced. The NRDP's participatory approach has benefitted community organization and empowerment, increased transparency, and demonstrated that rural communities can influence the allocation and use of resources at the municipal level to alleviate poverty. Investments funded by the program have been of satisfactory quality and less costly than those executed by public agencies. While the reformulated program does not expect to change the basic agricultural patterns in the Northeast, it has successfully reached a large number of landless families and land reform beneficiaries. Several field investigations, surveys and studies have assessed the impact of NRDP. The consensus, from a broad sample of completed and operational subprojects, is that they have had a positive impact on the quality of life, and in the case of productive subprojects, on employment and incomes of beneficiaries. They have relieved the effects of a severe drought in 1993. Implementation of productive subprojects and rural water supply works under the program enabled those families involved to take fuller advantage of improved climatic conditions in 1994 than was possible for families not participating. A summary evaluation of the reformulated NRDP is provided in Annex 2. The State of Sergipe 1.17 With an area of only 22,000 km2, Sergipe is Brazil's smallest state. The level to gently rolling land is mostly devoted to agricultural production (86% of total area). Three broad zones define the State: the coast, which enjoys moderate temperatures and rainfall, and plentiful, good quality underground water; the inland semi-humid area with variable rainffall; and the semi-arid area with highly variable rainfall and prolonged dry periods. Sergipe's coastline has infertile soil, its original forest having been destroyed, and the coastal population therefore lives from fishing and coconut production, and working as wage laborers on large sugarcane farms. The semi-humid region farther inland is marked by fruit plantations due to the favorable climate, relatively fertile soil and increasing use of irrigation. It also has large areas of pasture for raising cattle. About 47% of the total land area in Sergipe constitutes part of the "drought polygon" and in this semi-arid region, farthest inland, subsistence farming predominates. The semi-arid region has lately been encroaching upon the semi-humid region, making irrigation an increasingly critical investment. 1.18 Sergipe's industrial sector, concentrated in mineral extraction and processing, accounts for nearly half of the State's GDP, but agriculture (10% of GDP) is still vital to the 489,000 rural dwellers among Sergipe's 1.5 million inhabitants. Rural poverty in Sergipe, as - 6 - elsewhere in the Northeast, is closely linked to agro-climatic conditions, the poor natural resource base and skewed farm size distribution: 84% of farms are under 30 ha but occupy only 25% of the available land.4 Small farmers, lacking adequate technology, grow mainly low productivity crops which generate minimal working capital. 1.19 Income patterns reflect these precarious conditions: 68% of rural agricultural workers in Sergipe earn one minimum wage (about US$80) or less per month, compared to 30% in industry and 45% in the services sector (1990). During periodic droughts, the rural poor not only suffer serious crop losses but also often lose their employment on the larger farms as well, provoking intermittent migration to urban areas. Social conditions are also more difficult. Rural illiteracy is 60 %, compared to 25 % for urban areas. Nearly two-thirds of rural households lack access to proper sanitation facilities, compared with 27% in the State as a whole. Finally, 62 % of all hospital beds are located in the metropolitan area of Aracaju. 1.20 State Government Strategy to Alleviate Poverty. The State Government of Sergipe is very concerned with the growing disparity between its urban and rural areas. The strategy of the current administration to combat rural poverty is to improve infrastructure and basic services (e.g. health and education) and to create sustainable, employment and income- generating alternatives in rural areas, in order to improve the standard of living of rural dwellers and reduce their periodic migration to urban centers in the interior and the metropolitan area of Aracaju. The Government intends to implement a comprehensive program of community projects in rural areas which respond to the identified priorities of the communities themselves, thereby generating a greater poverty alleviation impact with State Government resources. In partnership with the World Bank, the proposed project would serve as the vehicle to implement this demand-driven program of community investments, and would emphasize productive subprojects as the basis for increasing employment and income-generating activities in rural areas. 1.21 Previous Experience with Rural Development Projects. Under the POLONORDESTE program, Sergipe implemented three area development programs of which one, Tabuleiro Sul, was financed by the World Bank from 1977 to 1984 (Loan 1714-BR). Despite implementation problems common to all states under POLONORDESTE (paras. 1.10- 1.11), the Tabuleiro Sul project met with relative success, making notably important progress in the provision of rural infrastructure. The follow-up program under the NRDP was a statewide rural development project (Loan 2523-BR) which, from 1985 until its refonnulation in 1993, was less successful under Brazil's generalized macroeconomic crisis and other implementation problems noted for the NRDP program in general (paras. 1.13-1.14). 1.22 Nonetheless, Sergipe's overall disbursement performance in both the original and reformulated NRDP has been among the best of the Northeast states. Of the original loan amount of US$61.3 million, nearly US$40 million had been disbursed by the time of reformulation in 1993. Since then, and as of February 1995, an additional US$7.4 million have been disbursed, and US$6.7 million fully committed, leaving a balance of only about US$6.3 4. Source: SEPLAN, based on INCRA data (1985). - 7 - million for disbursement before the scheduled loan closing on December 31, 1995. Both subproject approval and completion rates have been high: of 430 subproject proposals received, some 330 have been approved, of which almost half (156) have been completed (see Annex 2). This positive perfonnance is attributed in large part to the commitment of the Government to NRDP and the longstanding experience of the State Project Coordinating Unit with rural development. Main Lessons Learned from the Northeast Rural Development Program 1.23 The World Bank has 20 years of experience in supporting rural development and poverty alleviation in Northeast Brazil starting with the POLONORDESTE Program (1975-84), a variety of sectoral projects and more recently with the NRDP (1985 to the present). The NRDP, following its reformulation in 1993, has focused entirely on client-driven subproject investments implemented by the communities themselves, with decentralized project management at the state and municipal levels and more transparent decision-making and accountability for project performance. Implementation has provided important lessons for the proposed project (see Annex 2 for further details): (a) Decentralization of fiscal and investment decision-making from Federal to state and local government ensures more efficient project administration. Decentralization of resource allocation and investment decisions to rural municipalities and communities should be accompanied by a clearly-defined and well-disseminated system of incentives and penalties to discourage the misuse of funds; (b) Participation in the financing of subprojects generates a sense of ownership and a willingness to share responsibility for the future operation and maintenance of project investments. Beneficiary participation in the selection, execution, supervision and financing of project investments ensures that investments respond to true perceived needs, generates cost savings and increases accountability at the local level; (c) Sustainability of project investments has greater potential when the municipalities and communities contribute to subproject financing in a cost-sharing arrangement and when there is increased beneficiary participation. For example, the participatory process introduced in the pilot Municipal Council component (FUMAC) of the reformulated NRDP ensured better selection and prioritization of subprojects by beneficiaries; (d) Povertv targeting mechanisms should be simple, explicit and monitorable; be based on objective criteria; foster transparency and minimize political interference in project resource allocation; and ensure that project resources do reach the poorest communities; - 8 - (e) Technical assistance should be accessible to rural communities to enable them to identify, prepare and implement their own subprojects, thereby augmenting their capacity to compete for investment funds. Technical assistance should also be targeted to weaker municipalities to improve their planning, management and financial capacity; (f) Standardization of subproject documents. technical designs and unit costs simplifies the subproject preparation and evaluation process, facilitates the procurement of goods and works, prevents overdesign and improves the quality of subprojects, thereby encouraging greater participation by poorer communities and reducing bottlenecks in the subproject cycle; and (g) A user-friendly monitoring and evaluation system facilitates the subproject evaluation process, provides feedback and necessary information to improve targeting and efficiency, and is an essential management and planning tool. 1.24 The design of the proposed Sergipe Rural Poverty Alleviation Project has taken into account these lessons, which are consistent in many respects with lessons learned elsewhere in Latin America and other regions with similar community-based development programs. The project will expand the FUMAC approach and initiate a pilot municipal funds program (para. 2.7), to promote decentralization of decision-making to municipal and community levels, to relieve the administrative logjam resulting from state intervention in the minutiae of subprojects, and to further encourage municipal cost-sharing of subprojects. The project will employ a poverty targeting methodology based on a series of poverty-related criteria at the municipal, community and beneficiary levels, and backed by a strong system of operational checks and balances to thwart mistargeting and misappropriation of resources. As a condition of loan effectiveness, an improved monitoring and evaluation system will be adopted to reinforce the targeting mechanism and to strengthen the management of the subproject cycle. In addition, for simplicity and efficiency in subproject preparation, evaluation and supervision, the Project Operational Manual will establish standard subproject designs, technical and financial parameters, and cost indicators for the most frequently requested subprojects. Finally, funds for technical assistance to rural communities and municipalities will be facilitated to enable participation of the poorest areas, foster community capacity and empowerment, and encourage the active participation of NGOs and local grassroots organizations. -9- II. THE PROJECT Origin and Rationale for Bank Involvement 2.1 The proposed project is based on the positive results of the reformulated NRDP. Joint evaluation by the Brazilian Government and the World Bank has confirmed that the reformulated NRDP, itself the culmination of two decades of experience, has succeeded in introducing an effective model for rural poverty alleviation in the Northeast. This model has now been field tested for about one and a half years, and the Northeast State Governments are developing projects which would expand its coverage. The proposed project would be one of the first follow-on operations. As the principal donor engaged in a sustained, long-term partnership with the Brazilian Government to address rural poverty issues in the Northeast, the Bank is particularly well placed to support this next phase of development initiatives for the Region. 2.2 The new Brazilian administration that came to office in January 1995 is extremely concerned with the level of poverty in the country, and has declared the Northeast Region a priority area for public development programs. The proposed project, based on a more participatory and cost-effective approach than previous programs, would fit well within the Federal Government's Comunidade Solid4ria strategy for poverty alleviation (see also para. 1.6). 2.3 The project is also consistent with the Bank's Country Assistance Strategy (CAS) for Brazil, discussed by the Board on June 29, 1993. A central theme of the CAS is to reduce poverty and inequity both through resumption of broad-based economic growth based on structural reforms and appropriate sector-specific policies and through proper targeting and delivery of services to the poor. The project would contribute to economic growth through income- and employment-generating investments in rural areas of Sergipe. It would also support decentralization of decision-making and expenditure responsibility from the Federal to subnational government level, in this case, to the the State Government of Sergipe and its rural municipalities. The project would expand social expenditures directed to the poorer segments of the rural population based on community mobilization efforts and active local beneficiary participation. Project Objectives 2.4 The project would assist the State Government of Sergipe to alleviate rural poverty and its consequences by: (a) providing basic social and economic infrastructure and employment- and income-generating opportunities for the rural poor; (b) supporting rural community groups in planning and implementing their own subprojects; (c) providing a safety net for the rural poor during a period of strong fiscal adjustment when the Goverment's ability to finance essential investments and services is constrained; and (d) leveraging revenue - 10 - mobilization at the community and municipal levels. The project would ensure that funds are targeted and reliably transferred to the poorest communities, and would promote further decentralization of decision-making to State, municipal and local levels. Project Components 2.5 The proposed project would include three main components, as follows: (a) Community Subprojects (US$55. 1 million or 93 % of total base cost), which would support small-scale investments selected and subsequently operated and maintained by the beneficiaries themselves; (b) Institutional Development (US$2.4 million or 4% of total base cost!, which would provide all implementing entities and communities with technical assistance and training to increase their capacity and improve implementation of the project; and (c) Project Administration, Monitoring and Evaluation (US$1.8 million or 3% of total base cost), which would finance project coordination and activities to provide feedback on project performance and impact. A. Community Subprojects 2.6 The project would provide matching grants to rural community associations to finance small-scale subprojects (up to US$50,000) identified by these groups as priority investments that would improve community well-being. Subprojects would be of many types, broadly classified as infrastructure (such as rural water supply, electrification, local road improvements, and small bridges), social (e.g., day care centers for the young and elderly, school or health post rehabilitation) and productive (such as small-scale processing and minor irrigation schemes). There would also be a negative list of subprojects types which would not be eligible for financing. Although demand-driven, the project would set indicative targets for the main categories of investments (e.g. social, productive and infrastructure) and would track them targets via the project's monitoring system (Annex 7). Subproject proposals would observe standard documentation and technical, economic, environmental and sustainability criteria established in a Project Operational Manual, which sets forth procedures and guidelines for implementation. The Operational Manual will build on the manual already in use during the past year and a half for the reformulated NRDP project, with some fine-tuning and additions to reflect the lessons learned during the past two years implementation. Agreements have been reached on the needed improvements to the NRDP Operational Manual. During negotiations the Bank reviewed a complete revised draft of the Operational Manual, and assurances were provided by the Borrower that all project components and subprojects would be executed in accordance with the Project Operational Manual (para. 3.1(a)). Adoption of the Operational Manual would be a condition of loan effectiveness (para. 3.2(a)). 2.7 Three different delivery mechanisms would be used to screen, approve, and implement community subprojects, as reflected in the three subcomponents described below: - 11 - (a) Municipal Community Schemes (FUMAC). These subprojects would be identified and prepared by rural communities and presented to project Municipal Councils for approval. The Municipal Councils, composed of community members and representatives of local organizations, would promote local consensus-building on priority needs, screen and approve subprojects for financing by the State Project Coordinating Unit (PCU). It is expected that at least some US$35.6 million equivalent or 60% of total base cost would be invested under this subcomponent. (b) Pilot Municipal Community Funds (FUMAC-P). This subcomponent would be a pilot variant of FUMAC, in which select Municipal Councils would prioritize and approve subprojects submitted by rural communities and then would finance them within the context of an annual budget, which the State PCU would delegate to the Municipal Councils to manage themselves. This annual budget envelope would be allocated to the Municipal Councils according to a distribution formula, based on clear and measurable criteria, for implementation of an Annual Operating Plan (POA) prepared by each Municipal Council and approved by the PCU (Annex 5). The Municipal Council would transfer funds to beneficiary associations to finance approved community subprojects. It is expected that at least US$5.9 million equivalent or 10% of total base cost would be invested under this subcomponent. (c) State Community Schemes (PAC). This subcomponent would provide the opportunity for rural communities to submit their subproject investment proposals directly to the State PCU. This would avoid placing at a disadvantage those communities within municipalities which lack either the capacity or the political will to adopt the FUMAC or FUMAC-P approach. The continued existence of the PAC would also serve as an incentive to encourage local mayors to join FUMAC knowing that otherwise their communities could access project funds directly at State level. The PCU would screen, approve and release funds for subprojects, interacting directly with the beneficiary communities. It is expected that up to some US$13.6 million equivalent or 23% of total base cost would be invested under this subcomponent. 2.8 Given the overwhelmingly positive response to FUMAC under the reformulated NRDP, it is expected that the need for the PAC subcomponent would diminish over the course of project implementation, with the FUMAC and FUMAC-P subcomponents gradually expanding. Flexibility would be built into the project to reallocate funds among subcomponents to accommodate this expansion. 2.9 In order to receive financing under the project, beneficiaries would form legally- registered associations, in compliance with Brazilian law, and would enter into agreements (conv&ios) with the PCU and/or project Municipal Councils. Understandings have been reached on draft model agreements between the PCU, Municipal Councils and community associations for the implementation of community subprojects under the FUMAC, FUMAC-P and PAC subcomponents. The final model agreements would form part of the Project Opreational - 12 - Manual. Beneficiary associations would contribute to subproject costs, either in cash, kind or labor, and would be responsible for the operation and maintenance of the investments. The level of contributions of the beneficiaries, municipalities and the State Government for the different types of subprojects would be specified in the Project Operational Manual. B. Institutional Development 2.10 Given the decentralized implementation framework of the project, capacity building through technical assistance and training of the various participating entities--beneficiary associations, Municipal Councils, and PCU--will be central to its success. The Institutional Development component (US$2.4 million or 4% of total base cost) will finance appropriate techncial assistance and training in order that each group acquire the capabilities needed to effectively carry out project implementation. For example, the beneficiary associations will need accessible support for community organization and subproject preparation and execution, as well as training in operation and maintenance techniques, to derive optimal benefit from subproject investments. The project Municipal Councils will require assistance to orient their decision- making on resource allocation among community investments and, in the FUMAC-P pilot program, to create capacity for investment planning and financial administration. In addition to providing or making available from other sources a wide range of technical and managerial advice to the project Municipal Councils and beneficiary associations, the PCU will need training to enhance its own project coordination and supervision skills. To this effect, this component would finance consultant services, training materials and courses, seminars, workshops, and related operational costs. 2.11 The technical assistance and training would be packaged into four categories: (a) a basic introductory capacity-building program for Municipal Councils. Organized by the PCU, this program would include intensive training on the project purpose, subproject cycle, role of the Municipal Councils, guidelines of the Operational Manual, and related topics, for each Municipal Council under the project; (b) specialized skills training for the PCU and Municipal Councils. Based on needs, each group would contract "on-the-job training" from local experts in areas such as environmental assessment, monitoring techniques, and financial administration; (c) technical assistance and training for participating communities, organized and contracted by the PCU and the Municipal Councils, on such topics as community mobilization and organization, subproject preparation, operation and maintenance of investments, and expert advice (e.g. for water use); and (d) workshops and seminars for the Municipal Councils and beneficiary associations, to exchange experiences under the project. 2.12 Potential service providers include local consultants, public institutions, universities, NGOs and international organizations. The Municipal Councils, provided with model Terms of Reference prepared by the PCU, would contract technical assistance based on programs agreed with the PCU, which would then release funds for that purpose. - 13 - C. Project Administration. Monitoring and Evaluation 2.13 This component (US$1.8 million or 3% of total base cost) would support overall project coordination and supervision and would help to strengthen the effectiveness and quality of project operations. It would finance the incremental operating costs of the PCU, mainly for supervision of subprojects, and a statewide information campaign to disseminate information about the project and its guidelines to increase awareness, transparency and participation. For this campaign, the project would finance: (a) local technical assistance in the design and development of the campaign; and (b) the implementation costs, including posters, leaflets, radio spots, and videos. 2.14 This component would also finance the establishment of a monitoring system, including the maintenance and use of a management information system (MIS). To function as an efficient management tool, the project MIS (see Annex 7), originally designed under the NRDP, would be improved and made more user-friendly. It would contain key technical, financial and socio-economic information on subprojects, as well as cost and other physical indicators. MIS-generated project reports, including monthly disbursement reports and semi- annual progress reports, would facilitate subproject tracking and generate appropriate feedback on project performance. In addition to financing operating costs for monitoring activities, the project would finance experts in information technology and project management: (i) to improve the MIS; and (ii) to collect and analyze cost indicators and physical parameters from the most frequently approved subprojects, so as to calculate a range of values for these indicators and to adjust subproject eligibility criteria accordingly. These adjustments would then form part of the Project Operational Manaual. 2.15 Finally, the project would finance consultant services to develop and implement studies to evaluate the impact of the subprojects and provide feedback to improve project operations: (a) annual physical performance reviews, to assess the quality and sustainability of common types of subprojects financed by the project, including reviews of community-based procurement; (b) a Mid-Term evaluation of project performance, carried out in the third year of implementation; and (c) a comprehensive impact evaluation, which would include a baseline evaluation to establish indicators of socio-economic impact, a mid-term and a final evaluation of the project using these indicators. Annex 7 contains a detailed description of the studies to be financed under the project. 2.16 During negotiations, assurances were provided by the Borrower that: a project information and monitoring system, with reports and format agreed with the Bank, would be maintained throughout the project period. The Bank would receive monthly information on disbursements of project funds and semi-annual reports on subproject activities and performance of project investments on July 31 and January 31 of each year (para. 3.1(c)); a baseline evaluation study would be completed within one year of loan effectiveness, and a Mid-Term Review of the project would be carried out during the third year of the project (para. 3. 1(d)). The establishment, by the State, of effective project monitoring and evaluation arrangements would be a condition of loan effectiveness (para. 3.2(c)). - 14 - Decentralization and Targeting 2.17 Decentralization. Decentralization of decision-making and the transfer of project implementation responsibilities from the Federal Government to the State and from the State to the local (municipal and community) levels, started under the reformulated NRDP with the introduction of the community-based PAC and FUMAC programs and will be further advanced in the proposed project. First, FUMAC coverage will expand significantly. Second, the pilot FUMAC-P program will promote a transfer of financial management responsibilities to select Municipal Councils. 2.18 Project Targeting. Beneficiary targeting would occur at four levels: (a) geographically by poverty level and other characteristics of the municipality; (b) geographically within municipalities to target rural settlements and communities; (c) community-based selection of poor beneficiaries and particularly vulnerable groups by the project Municipal Councils; and (d) project-based selection through the incentive structure and cost-sharing matrix established for major types of subprojects. Targeting at the first and second levels defines the project area. Project municipalities are selected based on poverty-related criteria such as rate of indigence, availability of municipal resources and other socio-economic indicators (Annex 1), which are linked with the poor natural resource base, harsh agro-climatic conditions, and concentration of small-farm families. Within municipalities, the second level of targeting reduces the potential beneficiary population to only those communities located in settlements with population under 7,500 people, resulting in a target population of 660,800.5 The third and principal targeting level is community-based. The project assumes that local representatives can best judge which community groups are poorest and what investment priorities are in each area. The fact that 80% of the members of the Municipal Councils would be representatives from rural communities and the local civil society, as well as the clear and transparent rules and procedures for the Councils' functioning, are also important for targeting poor beneficiaries at the lowest level. Finally, the cost-sharing matrix favors subprojects with greater poverty alleviation potential and discourages those thought to have somewhat lesser development impact. The effectiveness of the targeting methodology in directing project benefits to the poor would be assessed through the project monitoring system and through annual performance reviews and the Mid-Term Review (Annex 7). Details on targeting mechanisms are included in Annex 3. Agreements have been reached with the State of Sergipe on these targeting mechanisms and on the municipalities to be included under the project; these understandings would be reflected in the final Project Operational Manual. Project Coordination Arrangements 2.19 Project Coordinating Unit. The Unidade de AdministraCao do Projeto Nordeste de Sergipe (PRONESE), an autonomous entity under the State Secretariat of Planning, Science and Technology (SEPLAN), would act as the Project Coordinating Unit (PCU) and have overall responsibility for project coordination and administration. PRONESE was established in 1984 5. Including inhabitants of small urban centers. - 15 - to oversee the implementation of the NRDP in Sergipe; most of its staff had also had previous experience with the Bank-financed POLONORDESTE program. Under the direction of a General Coordinator, assisted by a Technical Coordinator, PRONESE's structure is being reorganized to accommodate the needs of the new, community-based strategy of the reformulated NRDP and the proposed project. Its three divisions, each headed by a Manager, have a professional staff totalling 47. The Community Projects Management Division (formerly the Supervision and Monitoring Division) will be responsible for analyzing community demand, appraising and supervising subprojects, and assisting (both on technical and legal matters) and training communities and project Municipal Councils. The Planning, Monitoring and Evaluation Division will also work with community subprojects, but will focus on planning, and monitoring and evaluation activities, in particular the preparation of POAs and the maintenance of the project's MIS. The third unit is the General Administration Division. While PRONESE has no field offices, the state's small size--most rural communities are only 50-100 km away from the state capital--readily permits field evaluation and supervision from headquarters in Aracaju. Although PRONESE's monitoring and supervision functions would be increased under the proposed project, its executive role would be reduced given the expansion of the FUMAC component of the reformulated NRDP and the delegation of many decision-making and implementation responsibilities transferred to project Municipal Councils, particularly that of evaluation, approval and supervision of subprojects under FUMAC and FUMAC-P. 2.20 Specific project functions of PRONESE as PCU would be to: (a) prepare project POAs; (b) work closely with the State Secretariats of Planning and Finance to ensure timely and adequate availability of loan and state counterpart funds for the project; (c) promote the project and encourage participation of potential beneficiary communities, NGOs and private sector entities; (d) review and appraise subproject proposals submitted by PAC communities; (e) review POAs submitted by FUMAC-P Municipal Councils and allocate the corresponding approved budgets; (f) release funds to the community associations whose proposals have been approved, and to the participating FUMAC-P Municipal Councils; (g) supervise the implementation of approved subprojects and monitor compliance with project guidelines; (h) provide technical assistance to the FUMAC and FUMAC-P Municipal Councils; (i) prepare statements of expenditure (SOEs) and applications for reimbursement; and (j) ensure that procurement, disbursement and auditing procedures applicable to the Bank loan are followed. A condition of loan effectiveness would be the signing of a satisfactory agreement between the State of Sergipe and PRONESE for implementation of the project (para. 3.2(b)). 2.21 Assurances were provided by the Borrower at negotiations that the POAs and the corresponding budgets for each project year would be made available to the Bank for review and comments by October 31 of each year; that the Bank would receive a copy of the final plan and budget as soon as they are approved; and that adequate counterpart funding would be made available for the project in a timely manner by the State (para. 3. 1 (b)). Approval of targets for any given year will be contingent upon performance in the previous year, including a thorough analysis of the actual demand from communities, of subproject costs actually observed, and of the number of uncompleted subprojects. The use of standardized technical designs and cost indicators would substantially facilitate the evaluation process by the PCU. - 16 - 2.22 Project Municipal Councils. The FUMAC and FUMAC-P Municipal Councils would have expanded responsibilities in decision-making and project implementation, compared with current practice under the NRDP. The Councils would comprise representatives from: (a) rural communities, civil groups and organizations such as NGOs, religious organizations and rural labor unions; and (b) publicly-elected representatives and municipal government officials. The Mayor (prefeito) would chair the Municipal Council. While the composition of these Councils may vary from one municipality to another, representatives from rural communities and the civil society would make up at least 80% of the Councils' membership. PRONESE would participate in the Councils in an advisory capacity. Decisions would be made at open public meetings, previously announced, and would be taken by majority of votes. Detailed criteria, rules and procedures for the Municipal Councils structure and operations are included in the Project Operational Manual. 2.23 The Municipal Councils would be entrusted with the following tasks/responsibilities: (a) publicizing and promoting the project in their areas; (b) providing technical assistance to the communities for subproject preparation and execution; (c) prioritizing, through local consensus-building, community subproject proposals and reviewing them for approval or rejection; and (d) monitoring and supervising the implementation of approved subprojects. Under FUMAC, the Municipal Councils will present subprojects which they have approved to the PCU for financing; under FUMAC-P, they will themselves be responsible for managing the funds allocated to them for their community subprojects (Annex 5). Project Implementation Arrangements 2.24 Community Subproject Cycle. Beneficiaries submit proposals for subproject investments based on priority needs of their communities. These proposals are discussed and screened within the Municipal Council under both FUMAC and FUMAC-P, and by the Project Coordinating Unit under PAC. Beneficiary associations would participate in the financing of subprojects according to cost-sharing arrangements defined for the different types of investments in the Project Operational Manual. They also take on the responsibility for subproject preparation, implementation, operation and maintenance. Standard documentation for subproject proposals, developed by the PCU, would be distributed to the beneficiary associations to facilitate the subproject preparation and screening process. To ensure the quality of the investments financed and to provide guidance to the beneficiaries when preparing subprojects, the Project Operational Manual would contain standardized technical designs, including environmental assessments and cost paramters (e.g., cost per physical unit or per beneficiary) for the most frequent types of investments. Beneficiary associations may solicit technical assistance for more sophisticated or less common subproject types. 2.25 Various standard designs already exist, and others would be developed under the project. In some cases, specialists would be contracted to prepare the designs, based on terms of reference agreed with the Bank. In others, the PCU would gather and adapt existing designs (e.g., for pipes or box culverts, retaining walls, fords, and decks of bridges) and make them available to the Municipal Councils and associations. Standard designs will include the following considerations, where appropriate: (a) engineering aspects; (b) technical, financial and - 17 - economic feasibility; (c) operation and maintenance; (d) recommendations on environmental impact; and (e) social demand. This process will be facilitated by the fact that other Northeast States will be developing similar projects and, hence, there will be considerable scope for dissemination and exchange of designs among the professionals of PRONESE and the other State PCUs. 2.26 The appraisal of community subprojects would be carried out by the Municipal Councils (under FUMAC and FUMAC-P) or the PCU (under PAC). The appraisal procedures would take into account: (a) compliance with the project's guidelines and eligibility criteria, including those related to the technical, environmental and cost specifications contained in the Operational Manual; (b) the beneficiary association's commitment to provide the required contribution and to take full responsibility for the execution, operation and maintenance of the investments to be funded; and (c) in the case of an association which has already benefited from other subprojects, assessment of past performance. Using standardized cost indicators, reasonable costing for subproject implementation would be ensured. Departures from standard designs would have to be fully justified, as would proposed investment costs which fall outside the range of standardized costs. 2.27 In order to disburse funds for subproject implementation, agreements (convenios) would be signed: (a) under FUMAC, between the PCU, the Municipal Council and the beneficiary association for subproject funding and implementation; (b) under FUMAC-P, between the Municipal Council and the association, for subproject funding and implementation, and between the PCU and the Municipal Council for release of project funds; and (c) under PAC, between the PCU and the beneficiary association for subproject funding and implementation. Such agreements would spell out the terms and conditions for the funding, execution, ownership, operation and maintenance of the approved subprojects. Model convenios would form part of the Project Operational Manual. 2.28 To supervise subproject implementation at the community level, a special committee, composed of at least three members of the beneficiary association not belonging to its administration and freely elected by all members, would be responsible for overseeing the execution of the subproject, including contracting. The PCU and the Municipal Councils would visit the subproject sites regularly during subproject execution to monitor physical and financial progress. The PCU would have overall responsibility for subproject supervision through the project MIS, which would track subproject progress from initial submission to completion (Annex 7), and through accounting and auditing arrangements. A Project Implementation Plan was developed and agreed at appraisal (Annex 7, para. 12 and Table 7.1). It includes project performance indicators, responsible institutions and annual targets, by component. The key elements of this Plan are included in the Loan Agreement. The Bank's Recife Office would supervise subproject progress on a sample basis in the field and through review of monthly disbursement summaries (see Annex 10 for Bank project Supervision Plan). 2.29 Project Operational Manual. The NRDP Operational Manual would be improved and expanded to reflect the need for fine-tuning and additions based on experience to date. Agreements were reached with the State during appraisal on the following points to update and/or include in the Operational Manual: (a) indicative quantitative and qualitative targets to - 18 - be tracked with the Project MIS; (b) the project area and the criteria used to select the municipalities included in the project area; (c) the initial distribution of municipalities among the three components (FUMAC, FUMAC-P and PAC) and the criteria for future redistribution; (d) guidelines for the pilot component (FUMAC-P); (e) arrangements for institutional strengthening and project administration, monitoring and evaluation; (f) FUMAC and FUMAC-P Municipal Council composition, responsibilities and modus operand>, notably in relation to FUMAC-P; (g) clear identification and description of responsibilities of beneficiaries and municipalities in the implementation of the subprojects; (h) model agreements to be signed between the PCU, the Municipal Councils and beneficiary associations; (i) new standard agreements between the PCU and the FLUMAC-P Municipal Councils and between the FUMAC-P Municipal Councils and the beneficiary associations; (j) standard forms for proposals to be submitted by potential beneficiaries; (k) the cost-sharing matrix for the various types of subprojects; (1) the fund release mechanism to beneficiaries for FUMAC-P; (m) a scheme of incentives and penalties; (n) monitoring arrangements, including a new classification of the various categories of subprojects; (o) standard subproject designs and their role in the appraisal of proposals submitted by beneficiaries; and (p) a revised standard form of acceptance of works completed (ata de recebimento). 2.30 Operation and Maintenance (O&MV Community associations would bear the responsibility for O&M for all investments, as specified in the convenio for release of project funds. These associations may request technical assistance to develop O&M programs and techniques. Operation and Maintenance arrangements for community subprojects are described in Annex 6. Experience to date under the NRDP has been very positive with respect to O&M of investments when local communities have a strong sense of ownership. 2.31 Operational Checks and Balances. A system of operational checks and balances, included in the Project Operational Manual, would promote transparency and accountability in the use of project funds. Increased decentralization of resources and decision-making to beneficiary associations and Municipal Councils, with only ex-post control by the State through audit of accounts and supervision and monitoring by the PCU, is central to the project strategy. However, this advancement requires disincentives against departures from project guidelines, based mainly on education/joint problem solving and, as a last resort, penalties for misappropriation, mistargeting, lack of participation or improper O&M. 2.32 Monitoring. Analysis of implementation will depend on a database of subproject information from the project Management Information System (MIS). Input to the database would be made from a standard data sheet at key stages of the subproject cycle (e.g. identification, appraisal, procurement and disbursement. See also Annex 7). Periodic processing of the database information will permit the monitoring of the characteristics and evolution of project implementation. Project performance indicators to be monitored over the course of project implementation are included in Annex 7. Monitoring and reporting arrangements are discussed in paras. 2.14-2.16 above. - 19 - Project Costs 2.33 The total project cost, calculated in April 1995 prices and including contingencies, is estimated at US$60.0 million. Foreign exchange costs are estimated to be US$15.4 million, equivalent to about 26% of total project cost. The baseline cost is estimated at US$59.3 million, and contingencies for the period between loan negotiations and project completion are estimated at US$0.7 million. Contingencies are low because they apply only to the two components of institutional development and project administration, monitoring and evaluation. For these activities, physical contingencies of 10% and 5% have been included, as well as annual price increases for local and foreign costs (both expressed in US dollars). No contingency provisions have been applied to subprojects component. Since the number of community investments is not defined ex ante, it was considered preferable that any change in physical or price attributes of subproject investment would be accommodated by a change in the number of subprojects approved and implemented. 2.34 Investments in subprojects represent some 93 % of total base cost (60% for FUMAC, 10% for FUMAC-P and 23% for PAC). The distribution among different categories of subprojects, as well as the phasing of expenditures over the five-year implementation period, are indicative only, and there would be flexibility to reallocate funds from the PAC to FUMAC or FUMAC-P subcomponents as Municipal Councils gain experience. Financing Plan 2.35 The Borrower would be the State Government of Sergipe, with the guarantee of the Federative Republic of Brazil. The Bank loan will be repayable over 15 years, including a five-year grace period, and will bear interest at the Bank's standard variable rate. The loan of US$36 million equivalent would finance 60% of the total project cost, and would disburse against 100% of expenditures for technical assistance, training and evaluation studies, 59% of expenditures for community subprojects and 20% of incremental project administrative costs (except salaries) and monitoring. The State Government would be responsible for the remaining 40% of total project cost, with beneficiaries contributing to these counterpart funds. The State Government of Sergipe is expected to meet its counterpart funding obligations without difficulty. The State has consistently met all of its counterpart funding obligations under the reformulated NRDP. The level of annual funding required under the new project would be similar, and would be equivalent to only about 1 percent of the total State budget. Beneficiaries would provide a minimum of 10% of the total costs of community-managed investments, in cash, labor or kind, amounting to US$5.5 million (paras. 2.36 and 2.43). Detailed project costs and the financial plan are presented in the table below: - 20 - ESTIMATED PROJECT COSTS AND FINANCING PLAN Local Foreifn Total % of Total % Foreijn Baseline Costs -------- (US$ Million) --------- Estimated Costs A. Community Subprojects 1. FUMAC 27.1 8.5 35.6 60.0 24.0 2. FUMAC-P 4.5 1.4 5.9 10.0 24.0 3. PAC 10.3 3.3 13.6 23.0 24.0 Subtotal 41.9 13.2 55.100 93.0 72.0 B. Institutional Development 1.0 1.4 2.4 4.0 58.0 C. Project Administration, 1.3 0.5 1.8 3.0 28.0 Monitoring and Evaluation Total Baseline Costs 44.2 15.1 59.3 100.0 100.0 - Physicai contingencies 0.2 0.1 0.3 - Price contingencies 0.2 0.2 0.4 Total Project Costs 44.6 15.4 60.0 Financins Plan State Government of Sergipe 18.5 -- 18.5a' Beneficiaries 5.5 -- 5.5 IBRD 20.6 15.4 36.0 Total 44.6 15.4 60.0 a/ Including US$2.4 million in local taxes. 2.36 Cost Sharing and Sustainability. The cost estimates and average size of community-managed subproject investments have been determined from the experience gained under the reformulated NRDP and the operation of its PAC and FUMAC components. On this basis, the project is expected to finance some 2,100 community-managed subprojects over 5 years. Since the NRDP is still on-going, no retroactive financing is being considered. The amount of the beneficiary communities' contribution to subproject costs would depend on the type of subproject and the financial resources available to the community, the final determination being made by the PCU. It is expected to be a minimum of 10% for social investments, 15% for infrastructure investments and 20% for productive investments. As in the existing NRDP, ownership of project facilities by the beneficiaries should enhance the sustainability of community investments. A key element would be the establishment of sound arrangements - 21 - under the project for the provision of operating, maintenance and capital replacement expenditures, the requirements for which would vary significantly between types of subprojects and the availability of which would vary by municipality and community association. The Project Operational Manual and the convenios would contain specific guidelines concerning the funding of these costs. Procurement 2.37 Procurement for community subprojects costing the equivalent of US$50,000 or less and procured by local communities, would be carried out mainly through direct contracting. This procurement procedure is appropriate because most subprojects: (a) would be small and/or implemented in scattered or remote areas and therefore it will be difficult to obtain competitive proposals; (b) can be managed directly by rural communities which will contribute to the work through the donation of unskilled labor and local materials; (c) will be selected on the basis of willingness of the beneficiary communities to contribute to and physically supervise their execution; and (d) would provide a vehicle for communities to play an active role in the local development process. There may be cases when the community lacks the capacity to make the necessary purchases of materials. For such subprojects, the municipalities or the PCU would make purchases on behalf of the communities, packaging the procurement for several subprojects whenever possible. Procurement by the municipalities or the PCU would follow: (i) National Shopping procedures for goods, and procedures acceptable to the Bank for procurement of small works under lump sum, fixed price contracts awarded on the basis of price quotations (under US$100,000 and up to an aggregate amount of US$2.8 million for goods and US$1.0 million for small works); or (ii) National Competitive Bidding -- NCB (over US$100,000). For NCB the standard bidding documents agreed between the Bank and the Federal Government of Brazil will be used. ICB is not anticipated for any goods or works under the project. Although the costs of individual subprojects may not exceed US$50,000, the average size of the subprojects is expected to be less than US$25,000. Approximately 400 subprojects would be implemented yearly using these procurement methods. 2.38 Prior review of procurement documentation by the Bank would be made for all NCB contracts. Although the level of Bank prior review of procurement would be low, it would be compensated in several ways. First, a study of procurement by community contracting would be carried out during the first year of the project, under terms of reference agreed during negotiations. Second, cost comparisons of similar subprojects would be conducted using the project information and monitoring system (para. 2.14) in order to detect possible procurement problems. Third, the project physical performance evaluation (Annex 7) would analyze procurement issues; and finally, during Bank supervision, random reviews would be conducted of subprojects, including field visits and review of subproject documentation. 2.39 Consultant services to provide technical assistance and training to the beneficiary communities, the Municipal Councils and the PCU would be procured in accordance with Bank guidelines for the use of consultants. Contracts, short lists, and selection procedures for technical assistance and training would receive prior review by the Bank when the value of the contract is US$100,000 or more for consulting firms or US$50,000 or more for individual - 22 - consultants. All consultant terns of reference would be subject to Bank prior review. During negotiations, assurances were provided by the State of Sergipe that: (a) the above procurement arrangements would be followed (para. 3.1(g)); and (b) the study of procurement by community contracting referred to in para. 2.38 would be carried out during the first year of the project and any agreed recommendations adopted expeditiously (para. 2.1 (h)). Procurement arrangements are summarized in the following table: Procurement Arrangements (US$ million) Procurement Method Category Total NCB Other Cost Community Subprojects * Minor Works -- 24.8"' 24.8 (under US$50,000) (14.7) (14.7) * Goods and Materials -- 24.81' 24.8 (under US$50,000) (14.7) (14.7) * Minor Works 1.8 -- 1.8 (over US$50,000) (1.1) -- (1.1) * Minor Works -- 1.o2' 1.0 (over US$100,000) (0.6) (0.6) * Goods and Materials 2.82' 2.8 (over US$50,000) (1.7) (1.7) Consulting Services -- 2.9 2.9 (2.9) (2.8) Project Administration -- 1.9 1.9 (0.3) (0.4) TOTAL 1.8 58.2 60.0 (1.1) (34.9) (36.0) Figures in parentheses are the respective amounts expected to be financed by the Bank. 1) Direct contracting. 2) Shopping. Disbursements and Special Account 2.40 The proposed Bank loan would be disbursed over a period of 5 years, and the project is expected to be completed by June 30, 2000, and the project's Closing Date is - 23 - December 31, 2000. The disbursement period shown in the projected Disbursement Schedule (Annex 8, Table 8.4) is shorter than the Bank's standard disbursement profile for rural development projects in Brazil, but is in line with the recent experience under the reformulated NRDP in Brazil and with other Bank-financed social fund projects in Latin America. The allocation of loan proceeds by disbursement category is shown in Annex 8, Table 8.3. 2.41 During negotiations, assurances were provided by the State of Sergipe that it would open and maintain a Special Account in US Dollars in a commercial bank for Bank disbursements (para. 3.1(e)). The authorized allocation would be US$3.0 million, adequate to cover the expected average four-month flow of disbursements under the project. Disbursements for all expenditures would be made on the basis of statements of expenditure (SOEs), except for goods and works exceeding US$100,000 equivalent; contracts with consulting firms above US$100,000 equivalent; and with individuals above US$50,000 equivalent. 2.42 The information required for the compilation of SOEs would be maintained by the Project Coordinating Unit in the MIS data base. All SOEs would be transmitted by the PCU to the Bank in Washington. Simultaneously, the PCU would send to the Bank's Recife Office a copy of SOEs and reimbursement applications submitted to the Bank in Washington, as well as updated information on the status of all subprojects, enabling a rapid review of subproject eligibility. This procedure would allow the Bank and PCU management to maintain a cumulative account of the progress of project implementation. 2.43 The communities' contribution to subproject investments would be defined in the agreement (convenio) between the PCU and the community association (PAC), between the PCU, the Municipal Council and the association (FUMAC) or between the Municipal Council and the community association (FUMAC-P). It would be computed as part of counterpart finance of the project. Simple, standard records, whose format would be included in the Operational Manual, would be completed by a designated community representative. They would be used to record cash contributions, materials and labor inputs, and would be subject to project audit procedures. Disbursements to the beneficiary associations for subproject execution would be contingent on physical evidence of work progress (i.e., the flow of funds to the community would be result-driven; records of counterpart contributions would be maintained as back-up information). Accounts and Audits 2.44 The PCU would maintain separate accounts for project expenditures, which would be audited annually by independent auditors acceptable to the Bank. The accounts of the Municipal Councils (for the FUMAC-P subcomponent) and of the community associations would also be subject to review by the independent auditors on a sample basis. Finally, independent auditors would also audit the Special Account (para. 2.41). Copies of the audit reports would be provided to the Bank within six months of the end of each fiscal year. The audit reports would convey the auditor's opinion and comment as necessary on the methodology employed in the compilation of the statements of expenditures, their accuracy, the relevance of supporting documents, eligibility for financing in terms of the project's legal agreements and standards of - 24 - record keeping and internal controls related to the foregoing. The State of Sergipe provided assurances during negotiations that: (a) the Special Account and the project accounts would be established, maintained and audited annually in accordance with appropriate auditing principles, by independent auditors acceptable to the Bank; (b) with respect to withdrawals on the basis of SOEs, that such audits would contain a separate opinion as to whether the SOEs, together with the procedures involved in their preparation, support the related withdrawals; and (c) certified copies of the audited accounts and of the auditors' reports would be submitted to the Bank within six months of the close of each project fiscal year (para. 3.1 (f)). Environmental Aspects 2.45 The proposed project has been assigned a "B" classification. Because of their relatively small size, most PAC, FUMAC, and FUMAC-P subprojects would not have a significant effect on the environment. The project would, however, ensure proper environmental screening of subprojects and enforcement measures, to prevent certain types of productive or infrastructure subprojects from producing negative effects on the environment. The Operational Manual would specify key environmental criteria and screening and enforcement procedures to be followed. 2.46 Based on specific criteria for different subproject types, each subproject proposal presented by the community associations would include a simple environmental assessment in the form of a "checklist", following a standard format. For standard subproject types, environmental screening at the subproject approval stage would be the responsibility of the PCU and Municipal Councils. For exceptional subproject types, for which technical standards have not been developed, the PCU would screen the subproject proposals in detail for environmental impact, contracting technical assistance when necessary, and this assessment and any subsequent recommendations would be incorporated into the subproject proposal. For the FUMAC-P subprojects, the Municipal Councils would conduct the environmental screening together with either the PCU or with contracted technical assistance. Enforcement of environmental criteria would be ensured via subproject supervision by the PCU and Municipal Councils (paras. 2.20 and 2.23). Poverty Category 2.47 The proposed project is part of the Bank's Program of Thrgeted Interventions. The project centers its investment on the rural poor of Sergipe, a state of the Northeast Region which, in turn, has the greatest incidence of poverty in Brazil. The project also employs a targeting methodology at the beneficiary, community and municipal levels, which takes into account poverty-related criteria and other key parameters (Annex 3) to ensure that project resources reach those communities which contain a higher concentration of rural poor than in the rest of the state. - 25 - Women's Participation 2.48 The full participation of women is critical to the poverty objectives of the project. Bank-supported operations and studies in the Northeast of Brazil show the strong correlation between the problems and needs of rural women and those of producers and family units in general. Interestingly enough, they also show that community leaders in the Northeast favor women's participation due to their willingness to provide assistance and services, receptiveness to change, and greater likelihood of repaying debt.6 Responding to lessons concerning women's participation, this project will seek to ensure that women's activities (both traditional and innovative) are included; to support subprojects which foster maternal/child health and day care facilities (the latter have proven invaluable under the ongoing PAC/FUMAC operations in freeing rural women for income generating activities); to target groups and activities in which female participation has proven constructive; to provide technical assistance for women's subproject proposals and implementation; and to monitor women's participation in the project and its benefits. Project Benefits 2.49 The proposed project would result in an improved quality of life for some 330,000 people (about 50% of the target population), by providing better access to water supply and other basic social and economic infrastructure and services, and by contributing to an increase in their production, incomes and employment. Project-financed rural water supply and minor irrigation works will be particularly important in reducing vulnerability to drought. Productive subprojects will provide a vehicle for financing improvements in input use, technology and value-adding activities, thus raising productivity and incomes. Female community members will benefit especially from subprojects such as water supply and some productive activities and social investments (e.g., day care centers) which enable rural women to seek paying jobs and contribute to increased family cash incomes. Annex 2 provides an assessment of the kind of benefits which have been achieved to date for the most common types of subprojects under the reformulated NRDP. These are illustrative of the kinds of benefits to the target population which are expected, albeit on a larger scale, from the proposed project. 2.50 Another important project benefit will be the strengthening of the capacity of poor municipalities and their communities to define and address priority needs, and to plan and execute their own development projects. It is also expected that the new project will have a similar impact in the mobilization of additional community and municipal resources, as has the reformulated NRDP thus far (close to 10% of resources invested by the program), and this will improve the longer-term sustainability of investments. Finally, the proposed project will provide a safety net for some of the poorest segments of Sergipe's rural population who will likely be affected by decreased public investments and services during a period of strong fiscal adjustment. 6. A Study of Popular Participation in the Brazil Northeast Rural Development Program (NRDP-PAPP), Conrad Kottak, February 7, 1994. - 26 - Project Risks 2.51 Bank experience in Latin America and elsewhere suggests that the main sources of risk for this type of program arise from: (a) irregular funding, which can erode community interest and participation; (b) technical weaknesses and inadequate supervision of large numbers of widely dispersed subprojects; (c) mistargeting of program resources; and (d) over- centralization of decision-making, which often has the corollary effect of depressing local initiative and community involvement. The funding risk will be mitigated by the fact that the Bank loan will be taken directly by the State, thereby eliminating those disbursement delays previously encountered at the Federal level. Secondly, counterpart funding responsibilities would be shared to a larger extent by the communities, decreasing the relative proportion of the State's own participation. Third, and perhaps most important, the strong sense of ownership by the State, municipalities and communities should help to protect project counterpart funding needs during the process of allocating scarce budget resources. The risks of overdesign and weak supervision at the subproject level should be substantially reduced with the establishment of technical standards and cost criteria for the most common types of subprojects, and a reinforced institutional strengthening component. Problems of mistargeting and lack of community participation have been reduced under the reformulated NRDP. However, to further minimize such problems, the project design has introduced a system of strong operational checks and balances against such behavior. For instance, a community or municipality would receive fewer funds the following year, or be cut off entirely from the program, in cases of misappropriation, mistargeting or insufficient participation. - 27 - mI. SUMMARY OF AGREEMENTS REACHED AND RECOMMENDATION 3.1 During Negotiations, the Bank reviewed the complete revised draft of the Project Operational Manual (para. 2.6). The State of Sergipe provided assurances that: (a) all project components and subprojects would be executed in accordance with the Project Operational Manual (para. 2.6); (b) Annual Operating Plans (POAs) and budgets for each project year would be made available to the Bank for review and comments by October 31 of each year, and that the Bank will receive a copy of the final plan and budget as soon as they are approved, and that adequate counterpart funding would be made available for the project in a timely manner by the State (para. 2.21); (c) a project information and monitoring system, with reports and formats agreed with the Bank, would be maintained throughout the project. The Bank would be presented monthly information on disbursements of project funds and semi-annual reports (on July 31 and January 31 of each year) on subproject activities and performance of project investments (para. 2.16); (d) a baseline evaluation study would be completed within one year of loan effectiveness, and a Mid-Term Review of the project would be carried out during the third year of project implementation (para. 2.16); (e) a Special Account would be opened and maintained in US Dollars in a commercial bank for Bank disbursements (para. 2.41); (f) (i) the Special Account and the project accounts would be maintained and audited annually according to standards and procedures satisfactory to the Bank; (ii) terms of reference for the auditors would include, inter alia, detailed procedures for the examination and verification of the statements of expenditure for the project accounts; and (iii) copies of the audit reports would be submitted to the Bank within six months of the close of each project fiscal year (para. 2.44); (g) procurement procedures would be as specified in paras. 2.37 - 2.39; and (h) a study on procurement by community contracting would be carried out during the first year of the project and any agreed recommendations adopted expeditiously (para. 2.39). - 28 - 3.2 Conditions of Loan Effectiveness would include: (a) adoption of the Project Operational Manual (para. 2.6); (b) signing of a satisfactory agreement between the State of Sergipe and PRONESE for implementation of the project (para. 2.20); and (c) establishment, by the State, of effective project monitoring and evaluation arrangements (para. 2.16). 3.3 Subject to the above agreements and conditions, the proposed project would be suitable for a Bank loan of US$36.0 million equivalent at the Bank's standard variable interest rate with a maturity of 15 years, including five years of grace. June 6, 1995 BRAZIL NORTHEAST RURAL POVERTY ALLEVIATION PROJECT - SERGIPE Socio-Economic Indicators BRAZIL NORTHEAST!' SERGIPE Total Rural Total Rural Total Rural Pbpulation (mn) 146.8 35.8 42.5 16.7 1.5 0.5 Illiteracy2' 20% 41% 38% 57% 36% 60% Heads of 33% 60% 56% -- 50% 68% household earning I minimum wage or less Heads of 24% 49% 46% -- 43% 71% household with less than I year of schooling % of households 74% 33% 46% -- 60% 13% with running water % of households 15% 47% 38% -- 27% 64% with access to proper sanitation facilities ca I/ Except Minas Gerais. 2/ Population 15 years and over. Sources: Censo Demoerdfico 1991 - Brasil, IBGE Censo Denogrdfico 1991 - Serripe, lBGE BRAZIL NORTHEAST RURAL POVERTY ALLEVIATION PROJECT - SERGIPE Socio-Economic Indicators Total Rural Pop. Rate of ICS Federal Pop. Rural Pop. Area Density Indigence Index3' Dependency Ratio (N°)y MYXf (kM2) °N/krn) Mv! %p State 1,491,876 32.8 21,843 22.4 31.4 0.36 Project Area 961,676 50.2 20,989 23.0 44.1 0.52 64.2 Proiect Area by Rezion: Agreste de Itabaiana 120,734 49.4 1,131 52.8 43.3 0.43 71.2 Agreste de Lagarto 209,732 58.8 3,122 39.5 43.9 0.54 55.1 , Cotinguiba 121,623 33.8 2,064 19.9 43.3 0.44 48&9 ° Litoral Sul Sergipano 110,267 40.8 2,159 20.8 41.1 0.49 54.8 Nossa Senhora das Dores 138,776 53.0 4,296 17.1 44.8 0.57 77.5 Propria 84,455 46.8 1,459 27.1 43.3 0.48 70.4 Sertao do Rio Real 77,453 51.4 2,071 19.2 46.5 0.59 75.0 Sertao Sergipano do Sao Francisco 98,636 61.3 4,686 12.9 47.7 0.57 79.0 1/ Censo DemoRerfico 1991, IBGE. 2/ 0 Maya da Fome, IPEA, 1993. 3/ The ICS (Indice de Condi,cao de Sobrevivencia) is a composite index calculated, for all municipalities, by UNICEF and IBGE to measure the adequacy of conditions pertinent to the survival of children under 6, based on characteristics of their families--household heads having incomes below the minimum wage, heads of household having less than one year of schooling or households without running water. These conditions are precarious for an ICS of 0.50 to 1.00, intermediate for an ICS of 0.30 to 0.49 and good for an ICS of 0 to 0.29. State and Project Area figures calculated as weighted averages. Source: Municdios Brasileiros - Criancas e suas Condico5es de Sobrevivencia, UNICEF/IBGE, 1994. 4/ Federal Dependency Ratio (1993): Total transfers from the Federal Fundo de Parficipacao as percent of municipal budget. Source: Tribunal de Contas do Esiudo do Sergipe, Coordinadoria de Contas Municipais (CCM), Secretaria de Estado da Fazenda. - 31 - Annex 2 Page I of 13 BRAZIL NORTHEAST RURAL POVERTY ALLEVIATION PROJECT - SERGIPE The Reformulated NRDP: Performance and Evaluation A. Introduction 1. The Federal and State Governments have been involved since the early 1970s in financing programs to alleviate rural poverty in the Northeast. The Federal Government established the Northeast Rural Development Program (NRDP) in 1985 to reduce rural poverty and improve the living standards of small farm families in the Northeast. Through 1993, with the exception of one component supporting small community projects, implementation of the NRDP lagged behind schedule, and performance in meeting basic project objectives was weak. The innovative community-based components (Apoio tis Pequenas Comunidades Rurais, APCR), which relied on community planning and implementation, stressed community organization and contracted out needed technical expertise, were more successful. They acknowledged the expressed needs of the communities, their poverty alleviation impact was positive, the cost per family relatively low, and disbursement and commitment rates far exceeded those of other components. The NRDP was reformulated and transformed in its entirety into a community- based development program--effective September 28, 1993--drawing both on the successful experience of the small community projects components and on lessons learned with similar schemes elsewhere in Latin America. 2. Under the reformulated NRDP, the project provides matching grants to rural community associations to finance small-scale subprojects identified by these groups as priority investments to improve community well-being. Choosing among eligible subprojects, the beneficiaries solicit investments which respond to their most critical needs. There are two different delivery mechanisms for screening, approving and implementing community subprojects: (a) State Community Schemes (PAC) in which rural communities submit their subproject investment proposals directly to the State. The State screens, approves and releases funds for subprojects, interacting directly with the beneficiaries; and (b) Pilot Municipal Community Schemes (FUMAC), in which subprojects identified and prepared by rural communities are presented to project Municipal Councils for approval. The Councils, composed of community members and representatives of the civil society, promote local-level consensus-building on priority needs, and screen and submit subprojects for subsequent financing by the State. - 32 - Annex 2 Page 2 of 13 3. This Annex presents the results of a performance and impact analysis of the reformulated NRDP, as described above. The analysis is based on information and findings of several studies and is presented as follows: (i) the information base and methodology on which the analysis and conclusions are based; (ii) an analysis of the performance to date; (iii) project evaluation, particularly an impact analysis; (iv) a summary of the major findings and results of other studies; (v) lessons for follow-up project design and implementation; and (vi) conclusions. 4. Lessons learned since reformulation show that after several years of experience with rural development efforts in the Northeast, a large measure of success has finally been achieved in effectively reaching the rural poor with targeted interventions. While the analyses have shown certain aspects that need improvement in a minority of subprojects, these can be easily adjusted by modifications to the project design. The design of the proposed Rural Poverty Alleviation Project has explicitly taken into account these lessons, which are consistent in many respects with lessons learned elsewhere in Latin America and other regions with similar community-based development programs. B. Information Base 5. The analysis presented in this Annex draws upon the following sources: (a) Data supplied by the Northeast States to the Simplified Project Monitoring System (SSMP) of the Bank's Recife Office. The SSMP is the major data base for the reformulated NRDP, from which the performance of the project is monitored. The SSMP stores key information on each subproject and is updated on a monthly basis; (b) Sample surveys by the Unidades Tecnicas (UTs)' of participating states. These surveys, the major source of information on the NRDP's impact since reformulation, evaluate the impact of a group of 52 PAC/FUMAC projects (8 different types), obtained by using the random sampling methodology proposed by the Bank.2 The eight major types of projects analyzed are: water supply; rural electrification; manioc flour mills (casas de farinha); tractors; house improvement I. The Project Coordinating Units (PCUs) under the proposed project. 2. This methodology builds on statistical random sampling procedures from a universe of all subprojects under implementation and/or completed, as obtained from the SSMP Due to the wide variation in project characteristics, the maximum acceptable difference between sample estimated value and the true population value was set at 20% -- the total number subprojects selected for data collection following this procedure was 177. The Bank's Recife Office identified specific subprojects randomly by employing a random number generator. The distribution of the sample for a given type of subproject among the states was based on the distribution frequency of such projects. A questionnaire was designed for each type of project, focussing on: (a) subproject identification; (b) nature of the subproject (investments funded); (c) results/impact of the subproject; and (d) special infbrmation. (For additional and detailed information on the procedure, see "A Methodological Note on the NRDP Survey on Impact," Recife Office, World Bank). - 33 - Annex 2 Page 3 of 13 for community health purposes; rice mills; clothes-making; and child day care centers. These eight project types account for more than 50% of all projects completed or under implementation in the ten States. (c) A desk review of the following studies: (i) special evaluation studies of FUMAC, also conducted by UTs for each of the ten States, based on their implementation experience and specific case studies; (ii) the University of Michigan's "Baseline and Popular Participation Study", initiated in 1993 by Kottak and Costa, and followed by field surveys in July-October 1994, covering 38 communities in 23 municipalities of 9 states (Minas Gerais was not included); (iii) a series of studies under the "ARIDAS Project on Regional/Municipal Development in Semi-arid Areas of the Northeast"; (iv) an evaluation of NRDP, commissioned in 1994 by the Federal Secretariat of International Affairs (SEAIN); and (v) reports from several World Bank and FAO missions reviewing the experience of the reformulated NRDP. C. Performance to Date 6. Statistical information from the SSMP on the performance of all 10 participating states in the reformulated NRDP is presented in Thbles 2.1 to 2.6. Depending on the state, the reformulated NRDP became operational between October 1993 and January 1994. The general performance of the program as of February 1995 is reviewed below. Disbursements 7. Thble 2.1 presents disbursement progress (including funds already committed) since reformulation for each participating state, as of February 15, 1995. In just over a year, a total of US$108.5 million has been disbursed or committed under the program, consisting of US$71.6 million actually disbursed and another US$36.9 million already committed by the States. The program's current disbursement rate, overall, is significantly higher than its historical equivalent before reformulation. During the eight years of implementation of the original NRDP (1985- 1993), US$342 million, or less than US$43 million per year, was disbursed. 8. Individual disbursement profiles have also been analyzed. In the second half of 1994, disbursements increased markedly for six of the states (Bahia, Minas Gerais, Sergipe, Maranhao, Piauf and Paraiba), but remained lower for the others (Pernambuco, Alagoas, Ceari and Rio Grande do Norte). Bottlenecks in the flow of loan funds from the Federal Treasury to the states were responsible for most of the lower disbursement rates. Subprojects and Beneficiaries 9. As of February 25, 1995, more than 21,000 PAC and FUMAC subproject proposals have been submitted by communities in the ten participating states. Of the total, more than - 34 - Annex 2 Page 4 of 13 15,000 subprojects have already been approved, of which 5,931 are completed and/or under implementation, and another 379 awaiting the final transfer of funds to the beneficiary associations. Negotiations with beneficiary associations on the terms of agreement and other issues were in process on the other 8,925 approved subprojects (Thble 2.2). 10. Some 120 different types of subprojects have been implemented and/or completed. Of these, 55% are infrastructure subprojects, 42% productive subprojects and 3% social subprojects. Subprojects related to water supply constitute the largest single category of submissions (19.3%), followed by rural electrification (17.6%), tractors (9.7%), manioc flour mills (8.0%) and a variety of others (Table 2.3). 11. Of the approved subprojects under implementation and/or completed, some 13 % are under FU1MAC. Although the costs of PAC and FUMAC subprojects are similar (ITble 2.4), FUMAC subprojects involve, on average, 40% more beneficiaries, resulting in a significantly lower implementation cost per beneficiary. 12, The average cost per subproject (around US$21,000) varies greatly among project types, ranging from US$10,000 for tubewells to US$36,500 for tractors, but never exceeding US$40,000 (TIble 2.5). Costs for the same type of subprojects often differ substantially between states. 13. Subprojects were being implemented (or completed) in 898 municipalities (Table 2.6), or 71 % of the total eligible municipalities (1,258). About 11% of these are FUMAC municipalities. On average, 5.9 subprojects have been approved for different communities within each municipality under FUMAC, compared with 4.7 under PAC. D. Evaluation Socio-economic Impact Evaluation 14. A full impact evaluation of community investments financed under the reformulated NRDP is difficult, as the program has been operational for a little more than one year and definitive conclusions can be drawn only from those subprojects which have been completed and are fully operational (Table 2.2). However, by taking a sample and using data obtained from field surveys of the subprojects and their beneficiaries by the different UTs, a socio-economic impact evaluation was conducted in November-December 1994 for eight main project types, which account for more than half of all subprojects under implementation and/or completed and of total subproject costs (para. 5(b) contains a description of the procedure followed). The analysis of sample cases surveyed by the UTs indicates that PAC and FUMAC have generally made a positive impact on the quality of life and, in the case of productive subprojects, on employment and incomes of beneficiary communities or associations. In addition, the analysis shows that the social internal rates of return for productive subprojects are extremely high (>50%), with the cost-effectiveness results also impressive--in terms of both employment -35 - Annex 2 Page 5 of 13 creation and social benefit/cost ratio. The prospects for financial sustainability of these subprojects are also more than satisfactory. 15. Data were obtained for 52 sample subprojects in the various states, of which 30 were under PAC and 22 under FUMAC. First, quantitative benefits per project type were assessed based on weighted averages obtained from the subprojects surveyed in that category, and results (appropriate to each subproject type) derived per beneficiary. Second, these results were extrapolated to all the subprojects under implementation or completed in each subproject category, with analysis of sustainability of the communal investments, in terms of operation and maintenance as well as capital replacements, for some of the productive subproject types. The analysis, made in financial terms, covered the ten states as a whole. Given the sampling nature of the exercise and the variations that exist between individual states in terms of subproject costs and performance, results which have been quantified in the overall evaluation (Tables 2.7 and 2.8) should be taken as illustrative only: (a) Rural water supply (19% of subprojects submitted). Since drought is a frequent occurrence and one of the major threats to life in the Northeast, communities express great demand for this type of subproject. Due to variation in physical conditions among the different areas, works implemented differ according to the source of water (surface or groundwater) and the type of infrastructure required (new or rehabilitated well, reservoir, fountain, etc.). The average cost is about US$16,000 per subproject' in the sample or US$142 per beneficiary. Rural water supply projects provide a precious resource which previously was obtainable only through long hours of walk or had to be supplied by the municipality (by carro pipa or water bowser), usually at great cost. The subproject results are savings in time, effort and cost, as well as improved health through better sanitation. However, these gains are difficult to quantify. It is estimated that approximately 140,000 families will benefit from the 976 subprojects under implementation and/or completed in this category. (b) Rural electrification (18% of subprojects submitted). This project type on average costs US$20,000 or about US$7,800 per km of network, and on average benefits 48 families. Providing electricity to roads, households, shops and small-scale processing units, the subproject contributes not only to the obvious improvement in quality of life of the local inhabitants (including access to radio, television and the use of domestic appliances), but also generates additional employment and incomes from the increased operation of local businesses and industries. In total, it is estimated that more than 36,000 families will benefit from the 758 rural electrification subprojects under implementation and/or completed, with more than 24,000 houses connected to electricity. 3. The variation between average costs per project type given in this section and those of Table 2.5 is explained by the smaller number and characteristics of the sample subprojects when compared with the total under the program. - 36 - Annex 2 Page 6 of 13 (c) Manioc mills (8% of subprojects submitted). Casas de faninha are a familiar structure in rural areas in the Northeast. They are popular with a population for whom manioc continues to be a major food crop, particularly in the form of flour. These subprojects cost around US$20,000 and, on average, involve 108 producers of whom 68 are members of the association which owns the mill and the rest are farmers who come to the mill for processing. The principal benefit from this type of project is a reduction in processing costs (plus, in some cases, savings in transport to other mills located outside the project area), leading to increased production (both of manioc flour and subproducts) either for sale or family consumption, and therefore higher incomes (in some cases, the quality of the product is also improved through particular care of the association members). Because faster processing allows the producers more time for planting and harvesting, the installation of manioc mills is associated with an increase in the crop areas, at least where access to land is not a problem. At the same time, the subprojects generate additional jobs for both men and women working in the mills, whose mechanization generally results in better working conditions. It is estimated that the 380 subprojects which were under implementation and/or completed will benefit around 30,000 families, produce annual incremental income of about US$377 per family, and generate more than 11,000 jobs (including additional farm employment). The comparison between receipts and operating and maintenance costs associated with manioc mills indicates that this type of investment is financially sustainable, leaving the association with enough funds to amortize the mill and its equipment in 5 years as compared to its average useful economic life of 12 years. (d) Tractors for communal use (10% of subprojects submitted). This subproject type, which costs roughly US$33,300 or US$440 per beneficiary, benefits on average some 76 farmers. Not surprisingly, tractor use has facilitated increases in the area under cultivation for a number of crops (manioc, maize, beans, cashew, horticulture crops and, to a lesser extent, sugarcane and dry season rice) and in their yields and productivity, resulting in a significant gain in the incomes of the corresponding association members. Also, as elsewhere in the world, tractors are labor-using when used primarily in pre-harvest production activities. In total, the 198 tractor subprojects under implementation and/or completed will benefit 15,000 producers and create 9,900 jobs (including the equivalent of additional person-days from crop cultivation). The net income of the association owning the tractor, after deduction of all operation and maintenance expenditures, is sufficient to cover the initial investment after a period of 5 years, which is much less than its average useful economic life of 10 years. (e) Rice mills (less than 2 % of subprojects submitted). The benefits from rice mills are similar to those of manioc mills, i.e., reduced processing costs, savings in time and effort, increased areas, production (including for subproducts for animal feed), incomes and employment. In addition, rice mills allow association members to produce and sell milled rice rather than unhusked rice at a much lower cost than -37 - Annex 2 Page 7 of 13 prior to the project. At an average cost of US$11,000 or US$234 per beneficiary (on average 47 association members, plus an additional 7 producers who process their rice at the project mill), this type of subproject brings an annual incremental income of US$330 per beneficiary, resulting from savings in processing costs and increased production of rice and its subproducts. The 62 subprojects under implementation and/or completed in this category will benefit more than 2,900 producers and create some 400 jobs. Milling revenues net of operation and maintenance costs allow the rice mill association to fully amortize its plant after 5 years, compared to an average useful economic life of 12 years normally assumed for this type of investment. (f) Clothes making (2 % of subprojects submitted). The most popular type of non- agricultural productive projects, clothes-making subprojects cost US$9,100 or US$109 per beneficiary, with each association comprising an average of 84 members. The clothes produced are sold on the local market (substituting for imports, which in some cases were the only kind found before installation of the subproject) or at weekly fairs in the area. They generate an additional annual income of about US$190 per beneficiary. The 88 such subprojects under implementation and/or completed will benefit a total of 7,400 people, create some 1,600 jobs, mostly for women, and generate a gross value of production of about US$5 million per annum. (g) House improvement (less than 4% of subprojects submitted). The dwellings of many rural families in the Northeast are in poor condition and often associated with lack of hygiene which can lead to very serious health problems (e.g., the often fatal disease, chagas, which is spread by an insect favoring materials like the thatched roofs of traditional rural houses in the interior). The average cost of making the necessary reforms is about US$33,000 per project--US$460 per beneficiary family (71 per subproject) and US$90/m2, which is low. In specific areas, when the subproject involves strictly rehabilitation and employs local materials and beneficiaries' own labor, the cost can even be as low as US$33/m2. It is estimated that the 116 house improvement subprojects under implementation and/or completed will benefit some 8,200 families in total, and indications are that consequent improvements in the living conditions of these families have an important positive impact on their health. (h) Day care centers (less than 1 % of subprojects submitted). Another social subproject in relatively high demand is the establishment of nurseries or kindergartens. The 29 subprojects already under construction and/or completed, at an average cost of US$26,000 or US$290 per family (91 beneficiary families per subproject), will benefit some 2,600 families. By freeing the mothers from child care responsibilities during the day, these subprojects increase the capacity of mothers to work and earn additional income, estimated at about US$230 per beneficiary per year and totalling some US$605,000. Other benefits include an increase in their time for leisure, educational activities and food preparation. - 38 - Annex 2 Page 8 of 13 16. Cost-effectiveness was determined by estimating, where appropriate, the total investment cost per additional job created by the subprojects, as well as social benefit-cost ratios. Benefit-cost ratios are high (>3.0) for all subprojects analyzed, and the initial investment per additional job created was low for all productive subprojects (more than 10 times lower than in the industry and service sectors). Social internal rates of return were also estimated for the four types of productive subprojects--they all exceed 50%. Both analyses were made assuming constant benefits over an eight-year subproject life cycle (Table 2.7). In addition to these positive impacts, benefits are largely concentrated in the subprojects' beneficiary communities. 17. Sustainability. The financial sustainability analysis of productive subprojects shows that while beneficiary associations receive a one-time investment grant from the program, this investment is financially sustainable because cost recovery through user fees by the average beneficiary association is adequate to cover both O&M and replacement of the original investment long before the end of its useful economic life (Thble 2.8). Technical Evaluation 18. In most PAC and FUMAC subprojects, works are generally of good quality, particularly when implemented by private contractors (75 % of subprojects). With the remaining 25% of subprojects (20% executed by municipalities, with unskilled labor often provided by beneficiaries, and 5% implemented by the communities themselves), the quality varies but remains generally fair to good. Only for a very small proportion (less than 5 % of all subprojects) is construction quality rated as poor. 19. Nevertheless, some shortcomings were detected in a number of subprojects, namely the overdesign of works due to lack of technical criteria and/or competent technical assistance, and leading to discrepancies in investment size and costs per beneficiary for the same type of projects. In addition, technical assistance provided by UTs or other entities to communities in planning and executing projects, has sometimes been found to be inadequate. Despite the availability of funds under NRDP to hire professional assistance with planning or implementation, such assistance is hard to secure in rural areas--in part because under the reformulated NRDP technical assistance funds were tied to the subproject and could not be released to remunerate contracted professionals if the association's proposal was rejected. 20. Because of their relatively small size, PAC/FUMAC subprojects generally do not produce significant effects on the environment. However, certain types of projects, by their very nature, produce environmentally undesirable by-products (e.g. from the processing of manioc) or lead to increased use of products that may have a negative environmental impact (fertilizers and pesticides associated with the use of tractors, wood for fuel in cassava and rice mills, etc.). Additional consideration would be given to these aspects under the proposed new project. -39 - Annex 2 Page 9 of 13 Institutional Assessment 21. Institutional analysis of the reformulated NRDP and of the role played by the various actors in the program yields a generally favorable picture: (a) Once initiated, community associations generally function well, apart from some isolated cases of apathy or of takeover by individuals trying to monopolize benefits. Appointment of office-holders, payment of dues or user fees and accounting all seem to be taken seriously by the members. However, associations do have difficulty with paperwork and the cost of legal registration. They also find it hard to comply with the level of detail required to submit proposals under PAC or FUMAC, although the more than 21,000 subprojects submitted would seem to belie this claim; (b) Development and performance of the FUMAC Municipal Councils are affected by the attitude of the local political power base, especially that of the mayor (prefeito). However, in most cases mayors are supportive and the disparate interests represented in the FUMAC councils have found a modus vivendi; (c) NGOs have played a mixed role. In many cases, churches, rural labor unions and some technical NGOs are playing a useful part in subproject implementation in certain states--stimulating and helping with the initial formation of community associations or providing assistance with submission or execution of community proposals. NGOs in these categories see the NRDP as a source of additional funds to support their work. On the other hand, some of those contacted via initial publicity campaigns were unsuited or unwilling to become involved in the reformulated NRDP. Others would only participate if project funds were channelled through them; (d) Many rural municipalities have limited annual budgets and very little revenue of their own. Since their funds barely cover operating costs, most project municipalities view the reformulated NRDP as a welcome source of additional resources for investment. For the most part, mayors, municipal legislature members (vereadores) and municipal administrations have supported and participated in NRDP, although at times mayors or legislators have attempted to subvert the program to their own ends. Only in a few cases has the weakening of the traditional clientelismo implied by the FUMAC approach aroused open opposition or boycott. In such cases, PAC still operated satisfactorily and allowed beneficiary associations to bypass the municipal authorities and submit proposals directly to the state UTs; and (e) The technical units (UTs) are located within the Planning Secretariat of each state. Most have established several field offices, each with one or more technicians. In some cases UTTs also handle other development programs in the state but usually these are programs of minor importance. That the UTs have generally been - 40 - Annex 2 Page 10 of 13 successful in promoting NRDP is obvious from the overwhelming community response. However, despite their size (generally 40-60 technicians), UTs find it hard to process the large number of subproject proposals and to provide technical assistance for the preparation and implementation of subprojects4. To provide sufficient technical support to communities the Ul's have in some cases forged successful alliances with NGOs (e.g. the Associacdo de Apoio as Comunidades do Campo (AACC) in Rio Grande do Norte). Support has also been provided by different state extension services (EMATER), but their response has generally been poor. In some states, while there is no official alliance between the UT and EMATER, individual extensionists may still support NRDP ad personam. UTs themselves receive project technical assistance, mostly contracted from international agencies, such as the Instituto Interamericano de Coopera9ao para a Agricultura (IICA) and FAO. Advantages of FUMAC over PAC 22. The data on subprojects implemented under the program show that FUMAC has outperforned PAC in a number of ways. First, although only 12 % of total subproject proposals submitted by communities are under FUMAC (the pilot nature of this component necessarily limited its area of intervention), 13% of all subprojects under implementation and/or completed, and only 4% of those rejected, belong to this component and account for 16% of all beneficiaries under the program. Second, although the average cost per subproject is similar for PAC and FUMAC (Table 2.4), the number of beneficiaries per subproject is on average 40% higher in FUMAC than in PAC, resulting in a cost per beneficiary which is considerably less for FUMAC subprojects. 23. Although socio-economic benefits produced by the two components are similar in many respects, there is enough evidence to assert that greater community participation and transparency have been achieved through FUMAC and that FUMAC-generated projects, most likely through better selection and prioritization by the beneficiaries, are able to meet the program's criteria better than PAC subprojects. In addition, FUMAC has contributed to increased community organization and capacity to identify, plan and implement their own projects. E. Summary of Results and Findings from Other Studies/Reports 24. Other studies of the reformulated NRDP provide qualitative analyses which, focusing on FUMAC, primarily address institutional issues such as decentralization, municipalization, community organization and participation, transparency in decision-making, and the role of the 4. Subproject proposal processing has not been a major limiting factor, considering that more than 15,000 subprojects have been approved in little more than a year. -41 - Annex 2 Page 11 of 13 State in the provision of training and technical assistance to municipalities. The following positive findings are cited in the various studies as the reformulated NRDP's main achievements: improvement in the living conditions and nutritional situation of the rural poor; positive multiplier effects of successful subprojects; support to rural communities and associations, and recognition of their potential; incentives and other positive contributions to community organization and participation; increases in value added of rural activities, production, incomes, and employment generation; and containment of rural/urban migration. On the other hand, some aspects which need improvement are: bureaucratic procedures and excessive documentation requirements; delays in subproject approval and fund disbursements; lack of technical assistance on the part of UTs; funding Limitations and subproject cost ceilings; unavailability of local technical staff to assist communities in preparing and implementing subprojects; lack of municipal participation and funds to contribute to subproject costs; weak participation of communities in prioritization of subprojects; and insufficient knowledge of the program by communities. 25. Reports of World Bank and FAO missions all stress that the program's overall impact on the rural poor is generally positive and that available evidence would indicate that FUMAC subprojects meet the program's stated objectives better than PAC subprojects. While there is consensus that the reformulated NRDP as a whole and its constituent elements PAC and FUlMAC are successful, there are also some areas that need improvement, mostly under PAC, which are similar to those already listed, specifically: the need for improved targeting and greater transparency; design problems, particularly with regard to insufficient technical criteria, excessive bureaucracy, and lack of adequate community participation in subproject selection and execution; unsustainability of project investments and of beneficiary associations; lack of counterpart funds; and, lack of subproject supervision and follow-up. However, the reports also concur that these problems have not affected the majority of subprojects and can be corrected or circumvented with improved design criteria aimed specifically at these issues. F. Lessons for Project Design and Implementation 26. The World Bank has 20 years of experience in supporting rural development and poverty alleviation in Northeast Brazil, starting with the POLONORDESTE Program (1975-84) and more recently under the NRDP (Northeast Rural Development Program, 1985-93). The NRDP, following a reformulation in 1993, focused on small, client-driven subproject investments implemented by the communities themselves, with decentralized project management at the state and local levels and more transparent decision-making and accountability for project performance. Implementation since 1993 has provided important lessons for the proposed project: (a) Decentralization of fiscal and investment decision-making from federal to state and local governments tends to result in more efficient project administration. Decentralization of resource allocation and investment decisions to rural municipalities and communities should be accompanied by a clearly-defined and well-disseminated system of checks and balances to discourage the misuse of funds; -42 - Annex 2 Page 12 of 13 (b) Participation in the financing of subprojects generates a sense of ownership and a willingness to share responsibility for the future operation and maintenance of project investments. Beneficiary participation in the selection, execution, supervision and financing of project investments ensures that investments respond to a true need, generate cost savings and increase accountability at the local level; (c) Sustainability of project investments has greater potential when the municipalities and communities contribute to the subproject financing in a cost-sharing arrangement and when there is increased beneficiary participation. For example, the participatory process introduced in the FUMAC component of the reformulated NRDP ensured better selection and prioritization of subprojects by beneficiaries; (d) Poverty targeting mechanisms should be simple, explicit and monitorable; be based on objective criteria; foster greater transparency and minimize political interference in project resource allocation; and ensure that project resources reach the poorest communities; (e) Tbchnical assistance should be facilitated to rural communities to enable them to identify, prepare and implement their own subprojects, thereby augmenting their capacity to compete for investment funds. Technical assistance should also be targeted to project Municipal Councils to improve their planning, management and financial capacity; (f) Standardization of subproject documents, technical designs and unit costs simplifies the subproject preparation and evaluation process, facilitates the procurement of goods and works, prevents overdesign and improves the quality of subprojects, encouraging greater participation by poorer communities and reducing bottlenecks in the subproject cycle; and (g) A user-friendly monitoring and evaluation system facilitates the subproject evaluation process, provides feedback and necessary information to improve targeting and efficiency, and serves as an effective management and planning tool. G. Conclusions 27. The lessons learned from the NRDP, especially since reformulation, show that after several years of experience with rural development efforts in the Northeast, a large measure of success has been achieved in effectively reaching the rural poor with targeted interventions. While the analyses have also shown some deficiencies in a minority of specific subprojects, these can be easily rectified by modifications to the project design. The design of the proposed Rural Poverty Alleviation Project has explicitly taken into account all of these lessons, which are consistent in many respects with lessons learned elsewhere in Latin America and other regions with similar community-based development programs: -43- Annex 2 Page 13 of 13 (a) The project will expand FUMAC and initiate a pilot municipal funds program, to promote further decentralization of decision-making to municipal and community levels, to relieve the administrative logjam resulting from state intervention in the minutiae of subprojects, and to further encourage municipal cost-sharing of subprojects; (b) The project will employ a poverty targeting methodology based on a series of poverty-related criteria at the municipal, community and beneficiary levels, backed by a strong system of operational checks and balances to thwart mistargeting and misappropriation of resources; (c) As a condition of loan effectiveness, the project will enhance its monitoring and evaluation system to reinforce the targeting mechanism and to facilitate better control and project management throughout the subproject cycle by the State Project Coordinating Units; (d) For simplicity and efficiency in subproject preparation, evaluation and supervision, as weUl as for minimizing design deficiencies, the Project Operational Manual will establish standard engineering designs, technical and financial parameters, and cost indicators for the most frequently requested subprojects. Also, environmental assessment criteria will be part of the Operational Manual; (e) Funds for technical assistance and training will be facilitated to rural communities and Municipal Councils to enable participation of the poorest areas, foster community capacity and empowerment, and encourage the active participation of able and competent NGOs and local development agencies; (f) In addition to the measures above, participation and transparency will be further enhanced by setting clear rules for the composition and modus operandi of project Municipal Councils, with more than 80% of members coming from community representatives and the local civil society; and (g) A system of operational checks and balances will be included to promote transparency. Disincentives and penalties against departures from project guidelines--misappropriation, mistargeting, faulty project design, lack of participation or lack of proper O&M--will accompany the increased decentralization of resources and decision-making responsibilities to beneficiaries and project Municipal Councils, with ex post control by the State through auditing of accounts, and supervision and monitoring. BRAZIL NORTHEAST RURAL POVERTY ALLEVIATION PROJECT - SERGIIPE Disbursement Performance of the NRDP by State, US$ million (as of February 15, 1995) State Original Loan Disbursement at Cancellations 2) Actual Already Outstanding Amount Reformulation ') Disbursement Committed after Balance 4) after Reformulation 3) Reformulation Sergipe 61.3 39.9 1.0 7.4 6.7 6.3 Rio Grande do Norte 61.4 36.1 10.0 2.7 3.9 8.7 Pernambuco 92,0 51.8 20.0 4.1 1.6 14.5 Bahia 171.0 59.4 60.0 28.2 9.5 13.9 Ceara 78.0 34.0 25.0 7.4 1.9 9.7 Piauf 122.0 45.1 30.0 5.8 4.3 36.8 Paraiba 60.0 24.0 0.0 5.1 2.7 28.2 Minas Gerais 55.0 14.1 0.0 2.4 2.7 35.8 Maranhao 84.0 25.0 0.0 7.0 2.3 49.7 Alagoas 42.0 12.6 0.0 1.5 1.3 26.6 Total 826.7 342.0 146.0 71.6 36.9 230.2 ; Notes: ') Reformulation became effective on September 28, 1993. 2) Cancellation became effective on December 21, 1994. _ 3) Figures do not include a large number of community subprojects for which the first installment has been made but which are awaiting transfer of more loan funds, nor subprojects already approved but which have yet to be implemented. 4) As of February 15, 1995. BRAZIL NORTHEAST RURAL POVERTY ALLEVIATION PROJECT - SERGIPE PAC/FUJMAC Subprojects by State, at Various Stages of Processing (as of February 25, 1995) State Number of subprojects Subproject distribution according to stage of processing '; Total PAC FUMAC 0 1 2 3 4 5 6 Sergipe 430 370 60 0 9 88 3 1 173 156 Rio Grande do Norte 1,152 1,035 117 0 607 0 62 156 327 0 Pernambuco 1,894 1,682 212 1,094 184 172 12 0 250 182 Bahia 11,575 10,099 1,476 0 0 8,231 237 22 2,259 826 Piaui 1,385 1,329 56 1,001 0 0 0 140 49 195 Ceara 1,243 1,106 137 231 144 315 52 0 367 134 Paraiba 721 607 114 477 1 2 0 13 150 78 Minas Gerais 580 442 138 302 0 1 8 46 194 29 Maranhao 1,703 1,497 206 742 302 116 61 1 363 118 Alagoas 403 389 14 295 26 0 1 0 76 5 Total 21,086 18,556 2,530 4,142 1,273 8,925 436 379 4,208 1,723 Note: I) Processing stages: 0 = proposals received by UT waiting for appraisal I= proposals being appraised by UT 2 proposals approved, negotiating with beneficiary associations/communities on terms of agreements to be signed 0 3 rejected proposals 4= proposals approved, agreements prepared, awaiting for loan funds 5 proposals approved, funds released to associations, projects being implemented 6 = completed projects Total demand = 0+1+2+3+4+5+6 - 46 - Annex 2 Table 2.3 BRAZIL NORTHEAST RURAL POVERTY ALLEVIATION PROJECT - SERGIPE Community Demand -- Distribution of Subprojects Submitted by Type, NRDP (as of February 25, 1995) Total NRDP NRDP Type of Project (All states) Sergipe 1%) (%) Infrastructure 55.2 73.5 Productive 41.9 12.5 Social 2.9 14.0 Total 100.0 100.0 Water supply 19.3 11.4 Rural electrification 17.6 41.6 Farm tractor 9.7 2.3 Manioc flour mill 8.0 6.5 House improvement 3.5 2.1 Irrigation 2.6 0.0 Sanitation 2.3 1.2 Clothes making 2.2 0.0 Bridges 2.1 4.0 Brick making 2.1 0.0 Rice processing 1.5 0.0 Localized road 2.7 11.1 rehabilitation Maize processing 1.2 0.0 Child day care centers 0.5 0.7 Cashew processing 0.5 0.0 Other 24.2 19.1 Total 100.0 100.0 -47 - Annex 2 Table 2.4 BRAZIL NORTHEAST RURAL POVERTY ALLEVIATION PROJECT - SERGIPE Average Cost of Subprojects under Implementation and/or Completed (US$) State Number of Subprojects Average Cost2' (Stages 5 & 6)'1 (US$) PAC FIJMAC Total PAC FUMAC Sergipe 278 51 329 27,328 24,157 Rio Grande do Norte 266 28 294 24,706 23,537 ernambuco 147 21 168 34,517 36,064 Bahia 1,484 119 1,603 16,103 16,390 Piaui 237 7 244 14,647 12,312 Ceara 423 78 501 27,465 28,488 Para-ba 157 71 228 11,711 12,588 Minas Gerais 175 48 223 30,529 24,729 Maranhio 386 95 481 19,731 20,487 Alagoas 49 6 55 21,302 21,717 Total 3,602 524 4,126 21,714 21,838 Notes: D Projects under implementation and/or completed 2) Includes beneficiary contributions BRAZIL NORTHEAST RURAL POVERTY ALLEVIATION PROJECT - SERGIPE Average Cost of PAC/FUMAC Subprojects, by Type and by State Average Cost by Type 1)2) STATE (US$) Manioc Farm Water Rural Small Thbe Rice House Small-scale Clothes Type of Project: flour mill tractor supply electrification dam well process. improvement irrigation Bridge making Sergipe 11,688 33,038 19,800 25,387 --- --- --- 25,542 --- 28,332 --- Rio Grande do Norte 20,489 35,752 15,991 19,605 25,145 --- --- --- 29,371 --- 19,841 0 Pemambuco 26,488 38,939 21,369 32,758 30,595 --- --- 37,633 35,352 --- --- Bahia 10,705 --- 17,286 22,995 12,012 10,697 12,377 20,509 19,837 18,888 20,537 Piauf 3,927 27,312 10,317 21,940 16,745 6,782 9,285 22,198 14,168 28,301 4,207 Ceara 14,796 38,576 20,760 32,412 28,321 12,946 29,043 --- 30,504 --- 28,278 Parai'ba 11,190 39,648 10,307 16,584 --- 2,932 2,147 --- 14,033 --- 7,843 Minas Gerais 29,570 35,862 --- 30,947 7,326 4,619 --- --- 16,892 37,110 --- Maranhao 10,277 35,925 17,824 30,145 24,142 21,042 11,004 39,083 22,113 15,265 15,633 Alagoas 16,176 27,751 --- 33,608 --- --- --- 16,439 --- --- --- Total 12,123 36,520 18,274 25,279 13,943 10,049 11,925 30,023 21,963 22,350 19,835 PAC 11,847 36,340 19,078 25,716 13,927 10,123 11,910 31,994 22,606 22,168 19,855 FUMAC 14,956 37,194 15,078 22,878 14,208 7,622 11,978 17,488 18,227 23,613 19,671 | Notes: ' Costs include beneficiary contributions 2) Nature and technical specification for the same type of project vary within and among states BRAZIL NORTHEAST RURAL POVERTY ALLEVIATION PROJECT - SERGIPE Number of Municipalities Reached by NRDP (as of February 25, 1995) Number of Municipalities ') State Implemented/completed projects per In the State In the With implemented/completed projects municipality Project Area Total PAC FUMAC Total PAC FUMAC Sergipe 75 74 68 61 7 4.8 4.66 7.3 Rio Grande do Norte 159 158 56 50 6 5.3 5.3 4.7 Pemambuco 174 167 99 87 12 4.4 4.2 5.5 Bahia 415 264 252 234 18 12.2 11.6 21.1 Piauf 143 124 77 75 2 3.2 3.2 3.5 Ceara 178 120 100 89 11 5.0 4.8 7.1 Paraifba 171 109 71 61 10 3.2 2.6 7.1 Minas Gerais 50 50 54 40 14 4.1 4.4 3.4 Maranhio 136 135 98 88 10 4.9 4.4 9.5 Alagoas 97 57 23 18 5 3.5 3.9 2.0 Total 1,598 1,258 898 803 95 4.8 4.7 5.9 Note: ') The sum of the PAC and FUMAC municipalities may exceed the total number of municipalities in the project area because some original PAC municipalities later became FUMAC BRAZIL NORTHEAST RURAL POVERTY ALLEVIATION PROJECT - SERGIPE Socio-economic Benefits of PAC/FUMAC Subprojects by Main Subproject Type Project Type Total No. of Total No. of Cost per Total No. Total Net Net Total Social Cost Subprojects Beneficiaries Benefiary of Jobs Incranentel Incremental Incrental Internal Effecti eness being (US$) Created Income per Income per Crop Area Rate of Implemented Yeaar Beneficiary (ha) Retbrn Total so and/or (US$ '000) per Yar (%) Investment per B_t- Completed (UiSS) Job Created Cost (USS) Ratio ") Rund water supply 976 138,592 142 -- -- - _ _ Rura electrification 758 36,331 400 - ---- -- - - Manioc nmlls 380 39,520 297 11,460 14,890 377 7,900 > 50 1,273 >3.0 Tractors fr conununal use 198 15,048 438 9,900 11,587 770 36,080 > 50 816 >3.0 Rice mills 62 2,932 234 398 968 330 1,220 > 50 2,895 >3.0 Clothes nukiug 88 7,360 109 1,583 1,400 190 - > 50 925 >3.0 House inipwvemeat 116 8,236 461 ---- ---- ---- -- -- Child Care centers 29 2,639 290 ---- 605 229 > 50 >3.0 Note: ')Real discount mte is 10%. - 51 - Annex 2 Table 2.8 BRAZIL NORTHEAST RURAL POVERTY ALLEVIATION PROJECT - SERGIPE Financial Sustainability Analysis of Productive Subprojects Iten/Project Manioc Mill Rice Mill Farm Tractor Number of associations 380 62 198 Average net income per 3,737 2,131 6,631 association (US$)" Average cost of subproject (US$) 20,000 11,000 33,300 Averaze nwunber of yars: Of useful economic life (years) 12 12 10 To build replacement fund (years)) 5 5 5 Notes: " Total income from association fees and cost recovery net of all O&M and other recurrent costs. 2) Number of years after which the association has accumulated enough funds to replace the original investment, which is considerably less than the useful economic life of the investment. The real interest rate is assumed to be 10%. - 52 - Annex 3 Page I of 3 BRAZIL NORTHEAST RURAL POVERTY ALLEVIATION PROJECT - SERGIPE Targeting Mechanisms and Project Municipalities 1 Targeting of the rural poor is of critical importance in the project particularly as justification for the use of grants for subproject financing. Under the project, it is proposed that targeting occurs at four levels: (a) geographically by poverty level and other characteristics of the municipalities; (b) geographically within municipalities to target rural settlements and communities; (c) community-based selection of poor beneficiaries and particularly vulnerable groups by the project Municipal Councils; and (d) project-based selection through the incentive structure and cost-sharing matrix. 2. Municipalities. Targeting at the first and second levels defines the project area. All municipalities, with exception of the metropolitan area of Aracaju', are included in the project area. Thrgeting of the 70 project municipalities takes into consideration the concentration of income and economic activity in the metropolitan area and the higher incidence of poverty in the municipalities of the interior, manifested by such indicators as the rate of indigence2, availability of municipal resources, and other socio-economic indicators (Annex 1). According to IPEA (1993), 91 % of the project municipalities suffer a rate of indigence greater than 50%. In addition, since rural poverty in Sergipe is closely correlated with the poor natural resource base, the harsh agro-climatic conditions and concentration of land ownership, the project area includes the entire semi-arid (sertao) region of the state and includes the area currently assisted by the reformulated NRDP, characterized as smaHl farm areas--84.4% of rural fanns in Sergipe are considered minifindios (under 30 ha) yet occupy only 25.4% of available land, while the remaining 15.6%, principally large farms, occupy 76.4% of the total area. A list of the project municipalities is included in this Annex. 1. The metropolitan area of Aracaju comprises the municipalities of Barro dos Coqueiros, Nossa Senhora do Socorro, and Sao Cristovio. 2. Rate of indigence equals the number of families below the official poverty line established by IPEA divided by the total number of families in the municipality. See 0 Mapa da Fome, IPEA, March 1993. - 53 - Annex 3 Page 2 of 3 3. Within the project municipalities, the selection of PAC and FUMAC municipalities will be based on criteria such as past performance under the reformulated NRDP, estimation of the organizational capacity of communities within the municipality, and "progressiveness" of the mayor and municipal administration, among other criteria. The criteria for selection of FUMAC-P municipalities is included in Annex 5. 4. Rural Settlements and Communities. The second level of targeting, that within municipalities, focusses on rural settlements and communities. Beneficiary settlements or communities are limited to those with a population under 7,500 people. 5. Community-based Selection. The third and principal level of targeting is community-based selection. The project assumes that local communities can best judge which community groups are poor and what investment priorities are in each municipality. The role of consensus-building at the local level by project Municipal Councils aims to ensure that only poor communities benefit from the project. During the open meetings to set priorities, criteria such as the socio-economic situation of the participating communities, including existing infrastructure and availability of services, would be discussed as part of the ranking of subprojects within the municipality. The composition of the Municipal Councils, in which 80% are representatives from rural communities and the local civil society, as well as the clear and transparent rules and procedures for the Councils' functioning, are important elements which facilitate targeting poor beneficiaries. Finally, indicative targets set for social projects and projects which benefit women in particular, should be considered by the Municipal Councils as criteria for subproject selection, insofar as the principles of need and perceived benefits guide ranking decisions. 6. Project-based Selection. A fourth level of targeting relates to the incentive structure which favors certain types of projects, and to standard project designs which would permit Municipal Councils and the PCU (in the case of PAC) to readily appraise and approve subprojects. A cost-sharing matrix would be employed which would distinguish between four different types of subprojects: social, infrastructure, productive and leisure subprojects. Adjusting the incentive structure to require different amounts from communities will favor some types of subprojects. The following cost-sharing matrix is proposed: Project Type Beneficiaries/ Municipality State Bank Community Government Social infrastructure > 10% > 30% < 60% Economic > 15% > 25% < 60% infrastructure Productive projects > 20% > 0% < 20% < 60% Leisure projects, > 25% < 25% < 50% e.g., community halls - 54 - Annex 3 Paze 3 of 3 7. The effectiveness of the targeting methodology would be assessed through the project monitoring system and through ex-post evaluation studies. For example, annual physical performance reviews (Annex 7) would monitor the targeting of project benefits. Also, the Mid- Term Review (Annex 7) would measure the ability of the project to target and reach the poor. Adjustments and fine-tuning of the targeting methodology would be based on these ex-post reviews, in agreement with the Bank. List of Municipalities in Project Area Azreste de Itabaiana 13. Siriri 3. Canhoba 1. Areia Branca 4. lIha das Flores 2. Campo do Brito Litoral Sul Sergipano 5. Neopolis 3. Itabaiana 1. Estancia 6. Nossa Senhora de Lourdes 4. Macambira 2. Indiaroba 7. Pacatuba 5. Malhador 3. Itaporanga d'Ajuda 8. Propria 6. Moita Bonita 4. Santa Luzia do Ithany 9. Telha 7. Sao Domingos 5. Umbauba Sertao do Rio Real Aereste de La2arto Nossa Senhora das Dores 1. Cristinapolis 1. Araua 1. Aquidaba 2. Pogo Verde 2. Boquim 2. Carira 3. Tobias Barreto 3. Itabaianinha 3. Cedro de Sao Joao 4. Tomar do Geru 4. Lagarto 4. Cumbe 5. Pedrinhas 5. Feira Nova Sertao Seraipano do Sao 6. Riachao do Dantas 6. Frei Paulo Francisco 7. Salgado 7. Gracho Cardoso 1. Caninde de Sao Francisco 8. Simao Dias 8. Itabi 2. Gararu 9. Japoati 3. Monte Alegre de Sergipe Cotineuiba 10. Malhada dos Bois 4. Nossa Senhora da Gloria 1. Capela 11. Muribeca 5. Poqo Redondo 2. Carm6polis 12. Nossa Senhora Aparecida 6. Porto da Folha 3. Divina Pastora 13. Nossa Senhora das Dores 4. General Maynard 14. Pedra Mole 5. Japaratulba 15. Pinhio 6. Laranjeiras 16. Ribeir6polis 7. Maruim 17. Sio Francisco 8. Pirambu 18. Sio Miguel do Aleixo 9. Riachuelo 10. Rosario do Catete Propria 11. Santa Rosa de Lima 1. Amparo de Sao Francisco 12. Santo Amaro das Brotas 2. Brejo Grande - 55 - Annex 4 Page I of 4 BRAZIL NORTHEAST RURAL POVERTY ALLEVIATION PROJECT - SERGIPE Institutional Development Project Coordinating Unit 1. The Unidade de Administra§o do Projeto Nordeste de Sergipe (PRONESE), an autonomous entity under the State Secretariat of Planning, Science and Technology (SEPLAN), would continue as the Project Coordinating Unit (PCU), with overall responsibility for project coordination and administration. Although the unit itself was formally established in 1984 to oversee the implementation of the NRDP in Sergipe, most of its small cadre of professionals have had many years of previous experience with implementing rural development programs in the state, including the Bank-financed POLONORDESTE program. 2. Under the direction of a General Coordinator, assisted by a Technical Coordinator, PRONESE currently comprises three divisions: General Administration, which includes Personnel, Finance and Administrative Support units, and is in charge of administrative and financial, including disbursement, matters; Planning and Evaluation, responsible for technical coordination and supervision of the PAC and FUMAC programs under the on-going NRDP; and Supervision and Monitoring, in charge of the old "Subprojects component"' of the NRDP. The three divisions, each headed by a Manager, have a professional staff totalling 47, of which 29 are high-level tecnicos and 18 mid-level tecnicos, plus related administrative and support staff. Upon completion of the NRDP, the Supervision and Monitoring division is expected to be reorganized into a "Community Projects Management" division, with responsibility for analyzing community demand, appraising and supervising subprojects, and assisting (both on technical and legal matters) and training communities and project Municipal Councils under the proposed project. The Planning and Evaluation division (which in the future will be called the "Planning, Monitoring and Evaluation" division) will continue to work with community subprojects but will focus on planning, and monitoring and evaluation activities, in particular the preparation of project annual operating plans (POAs) and the maintenance of the project management information system (MIS). PRONESE's organizational structure is presented in Figure 1. While PRONESE has no field offices (its 4 regional units were closed in 1991), the state's small size--most rural communities are only 50-100 km from the state capital--readily permits field evaluation and supervision from headquarters in Aracaju by some 13 mid-level 1. These were principally medium-scale irrigation projects implemented by State agencies. - 56- Annex 4 Page 2 of 4 professionals (mostly agricultural technicians) supported and supervised by PRONESE's high-level professionals. 3. While the monitoring and supervision functions of PRONESE as PCU would be increased under the proposed project, its executive role would be considerably reduced given the expansion of the FUMAC subcomponent and the delegation of many decision-making and project implementation responsibilities to project Municipal Councils, particularly that of evaluation, approval and supervision of subprojects under FUMAC and FUMAC-P. The PCU will prepare the project annual operating plans (POAs), having reviewed and consolidated the POAs submitted by the FUMAC-P Municipal Councils, and will ensure timely and adequate availability of loan and state counterpart funds for project implementation. It will appraise and approve PAC community subproject proposals, facilitated by the use of standardized technical designs and cost indicators for subprojects; administer and release funds for subprojects approved by FUMAC Municipal Councils; and allocate and release annual budget envelopes to FUMAC-P Municipal Councils. The PCU will also provide technical advice and training to project Municipal Councils. Municipal Councils 4. FUMAC and FUMAC-P Municipal Councils will carry the primary responsibility for subproject approval and for technical and organizational assistance to beneficiary associations in their respective communities. The project Councils comprise representatives from the rural communities, civil groups and local organizations such as NGOs, churches, and labor unions, and publicly-elected representatives and municipal government officials. While their composition may vary from municipality to municipality, representatives from rural communities and the civil society must make up at least 80% of the Councils' members. The PCU will be represented in the Municipal Councils but only in an advisory capacity. The Municipal Councils receive subproject proposals from the communities and through open public meetings, previously announced, build consensus at the local level on priority investments to be financed under the project. Council decisions for the approval of subprojects are subsequently made by majority of votes during the open meetings. The project Councils then monitor and supervise the implementation of the approved subprojects. Detailed criteria and guidelines for the structure and operation of the project Municipal Councils are included in the Project Operational Manual, and assurances were provided by the State Government at negotiations that these criteria and guidelines would be fully observed during project implementation. 5. The composition and the decision-making procedures of the Municipal Councils are the same both under the FUMAC program and the FUMAC-P pilot program. Nevertheless, in addition to the regular functions of project Councils, the FUMAC-P Municipal Councils will also be responsible for administering subproject funds and releasing them to the beneficiary associations. On the basis of POAs which they prepare for review and approval by the PCU, the FUMAC-P Councils receive an annual budget envelope from which to finance subprojects. They therefore must program and monitor the use of the funds - 57 - Annex 4 Page 3 of 4 from this annual budget, and their consensus-building and priority-setting tasks therefore take on the added dimension of allocation and rationalization of a fixed amount of resources. Institutional Capacity Building 6. Within a decentralized framework, the implementation capacity of the various participating entities--beneficiary associations, Municipal Councils and the PCU--is central to the success of the project. The Institutional Development component will finance appropriate technical assistance and training in order that each group acquire the capabilities needed to effectively perform their functions under the project. For example, beneficiary associations will need accessible expertise on community organization and subproject preparation and execution, as well as training in operation and maintenance techniques to derive optimal benefit from subproject investments. The project Municipal Councils will require assistance to orient their decision-making on resource allocation among community investments and, in the FUMAC-P pilot program, to build capacity for investment planning and financial administration. In addition to providing or making available from other sources a wide range of skilled technical and managerial advice to the project Municipal Councils and beneficiary associations, the PCU will need training to enhance its own project management and supervision skills, as well as specific skills such as environmental assessment. 7. The instruments for achieving this institutional capacity building would include a combination of workshops, technical exchanges, on-site training, and more traditional technical assistance, drawing upon local expertise within the state: private technical and management specialists, other public entities, universities, NGOs, and international organizations. As not all entities need the same specific skills, the PCU will develop a Capacity Building Program in a menu format which lists potential areas of technical skills and other training and potential service providers for each. The menu would package technical assistance and training into four categories: (a) a basic introductory capacity- building program for project Municipal Councils, including intensive training on the project philosophy, subproject cycle, role of the Municipal Councils, guidelines of the Operational Manual, among other themes; (b) specialized skills training for the PCU and the project Councils on a demand basis, including "on-the-job training" in areas such as environmental assessment, monitoring techniques, financial administration, etc.; (c) assistance to communities and associations in identification of needs, subproject preparation, implementation, and operation and maintenance; and (d) workshops and seminars to exchange experiences and disseminate project results. 8. From this menu, the PCU would develop and cost an annual technical assistance and training program for its own staff, for project Municipal Councils and, under PAC, for beneficiary associations. Also on an annual basis, the project Municipal Councils would plan specific skills training and beneficiary training, using the menu as a guide but supplementing it with particular needs when necessary, and submit this plan to the PCU for costing and approval. For each item in the program, the PCU would aid the Municipal Councils in the development of appropriate terms of reference and program outline, and in the identification and contracting of service providers. In those instances where a large number of beneficiary - 58 - Annex 4 Page 4 of 4 associations or Municipal Councils need similar skills--for example, accounting and budgeting for FUMAC-P Councils, or operation of irrigation subprojects for several beneficiary associations--the PCU would organize intensive short training workshops at the state or regional level on that particular subject or hire technical experts on a retainer contract to assist in a series of training events across the state. 9. The funds for institutional development will be administered by the PCU. The PCU will contract all technical assistance and training for the PAC program and in cases of packaging or upon request by the Municipal Councils within the FLTMAC program. FUMAC and FUMAC-P Municipal Councils will contract technical assistance and training for themselves or beneficiary associations and communities, with the assistance of the PCU and on the basis of approved terms of reference. The PCU would monitor the use of institutional development funds by the project Municipal Councils through review of statements of expenditure (SOEs) and periodic supervision. - 59 - Annex 4 Figure I Figure 1: Project Organizational Chart State Governor Secretary of Planning (SEPLAN) PRONESE 3 General Coordinator Technical Coordinator 3 I General Administration Planning, Monitoring and Community Projects Division Evaluation Division Management Division (GERAG) (ASPLAN) (GEPROC) - 60 - Annex 5 Page I of 3 BRAZIL NORTHEAST RURAL POVERTY ALLEVIATION PROJECT - SERGIPE The FUMAC-P Pilot Program Background 1. Decentralization of decision-making and the transfer of project implementation responsibilities from the State to the local (municipal and community) levels, started under the reformulated NRDP with the introduction of the community-based PAC and FUMAC programs, will be further advanced in the proposed project. First, FUMAC coverage will expand significantly under the project. Second, a further step towards decentralization will be reflected in a pilot program, FUMAC-P, which will promote the transfer of financial management responsibilities to several FUMAC Municipal Councils. 2. Generally, municipalities which have participated in FUMAC to date tend to be those in which the environment for participatory decision-making is more favorable, which distinguishes them from more traditionally-run municipalities with a strong "clientelistic" approach. This environment has been the result of a conjunction of cultural/political factors such as the emergence of a movement for the development of civil society (usually induced by religious organizations, NGOs and more recently by some political parties) and of a new generation of more open-minded political leaders. 3. Several important lessons have emerged from the FUMAC experience in the reformulated NRDP. First, it has proved to be an efficient, transparent approach for identifying and prioritizing, through local-level consensus building, multiple demands for investments by needy communities. Second, it has provided the opportunity to experiment not only with increased transfer of decision-making, including approval of subprojects, from the State-level Project Coordinating Unit (PCU) to Municipal Councils representing communities, but also with the management of community subproject funds by these very Councils. This experience is reflected in the proposed project in the design of FUMAC and, in particular, FUMAC-P where the feasibility of financial management at the grassroots level will be tested. FUMAC-P: Main Features 4. The FUMAC-P pilot program would account for at least US$5.9 million--or 10% of the total base cost of the project--over the five-year implementation period. If the approach proves successful with the communities and meets its objectives of local-level - 61 - Annex 5 Page 2 of 3 investment selection, execution and financial management, FUMAC-P would be expanded during project implementation with funds initially earmarked for the FUMAC and PAC subcomponents (see para. 10). 5. Initially, the PCU would select FUMAC-P municipalities/communities from among the best FUMAC performers under the on-going NRDP, i.e. those with the best record in administering and executing community subprojects. Additionally, other municipalities which have a favorable environment for participatory work (extensive community organization, municipal officials open to participative decision-making, etc.), and are interested and can commit themselves to following project guidelines, could be included. There should also be some indication that local capacity exists to manage financial resources, for instance in communities or existing associations already managing small-scale businesses (on an individual or group basis) and whose members could potentially form part of FUMAC-P Municipal Councils. As the proposed project gets underway and the FUMAC-P approach proves successful, other well-performing FUMAC municipalities could gradually be upgraded into FUMAC-P municipalities. 6. Once the municipalities are selected, allocation of funds would be done based on a number of criteria to which different weights will be assigned: (a) size of the municipality, in terms of both population and area; (b) ratio of rural to total population; (c) rate of indigence; (d) number and degree of dispersion of poor rural communities; (e) availability of public resources per capita; (f) number of subprojects approved under the reformulated NRDP (small discriminatory factor); (g) performance in the previous year; and (h) willingness of the mayor (prefeito) to participate. 7. FUMAC-P Municipal Councils would have the same composition and modes of operation as FUMAC Municipal Councils (Annex 4). They would be responsible for: (a) assisting rural communities in identifying priority needs and translating them into investment proposals; (b) screening and reviewing these proposals, and ensuring that they meet eligibility criteria as contained in the Project Operational Manual; - 62 - Annex 5 Page 3 of 3 (c) monitoring and supervising, together with the PCU, the implementation of approved subprojects; (d) providing, directly or through contracting, technical assistance and training to beneficiary associations in the design and operation and maintenance of project- financed investments; (e) managing subproject funds and transferring funds to beneficiary associations with approved subprojects; and (f) preparing indicative annual operating plans (POAs) for review and approval by the PCU. The POA would contain a list of community investment priorities, indicative targets to be met, a tentative timetable, the indicative budget and the technical assistance required to implement the proposed program. The PCU would review and approve the POA at the end of the year preceding implementation. 8. An agreement (conYenio) between the PCU and the Municipal Council would establish the terms and conditions as well as rights and obligations of each party with respect to the release and use of the approved funds. These (both loan and state counterpart funds) would be deposited into an account in a commercial bank, and would be withdrawn solely upon formal order by the Municipal Council. The Council would screen community subproject proposals on the basis of the approved POA and the funds made available by the PCU. An agreement (convenio) between the Municipal Council and the beneficiary association would be required. 9. Supervision of the operations of the Municipal Councils would be the responsibility of the PCU. 10. A Mid-Term evaluation of the FUMAC-P program would be carried out by independent evaluators. The basic purpose of this evaluation will be to assess performance under the pilot program and, if necessary, to adjust its design and mechanisms in order to improve its procedures and impact on beneficiaries. If the pilot program is found successful and its expansion is justified, part of the PAC and FUMAC funds under the project could be transferred to the FUMAC-P program. - 63 - Annex 6 Page 1 of 2 BRAZIL NORTHEAST RURAL POVERTY ALLEVIATION PROJECT - SERGIPE Subproject Operation and Maintenance 1. Although the reformulated NRDP has financed many of the same types of community investments (e.g., manioc mills or small-scale rural water supply projects) that have been supported under past programs or projects in the Northeast, one of its innovative features has been to ensure that these investments are operated and maintained by the beneficiaries themselves, thereby enhancing their sense of ownership and the prospects for sustainability of the investments. Under the proposed project, this feature will be maintained and strengthened. 2. Community associations will bear the primary responsibility for O&M for most project-funded investments, and this provision will be included in the agreements (convenios) with the associations. The relative responsibilities of the associations for O&M would vary according to the type of investment. As a general rule, beneficiaries would be fully responsible for operating and maintaining their own productive subprojects and some social subprojects, such as day care centers or home improvements for health purposes (e.g. for chagas control). In the case of schools or health posts rehabilitated under the project, and most infrastructure subprojects, the responsibility for O&M would normally be shared between the communities and the relevant State or municipal institution. In exceptional cases of relatively complex subprojects, beneficiary associations might contract private firms or NGOs to assist them in O&M tasks. 3. Based on experience under the reformulated NRDP, several types of arrangements would be implemented in order to ensure the financial sustainability of project investments, as follows: (a) Productive Subprojects (e.g., manioc and rice mills, clothes making, small- scale irrigation schemes, farm machinery, etc.): 4. Once a subproject has been approved for financing, the beneficiary association elects a committee to be in charge of subproject implementation and management. This committee has the task, among others, of determining the different dues and user fees to be charged to association members (as well as to non-association members using project facilities), based on estimated requirements for O&M and capital replacement or eventual expansion. User fees have to be approved by all association members, and may be paid in cash, labor or kind, depending on the particular situation. The committee is responsible for collecting the fees, placing them in a "reserve fund", managing the reserve fund, and making - 64 - Annex 6 Page 2 of 2 payments for recurrent expenditures such as electricity consumption, fuel, inputs, repairs, and employees' salaries. (b) Social Subprojects (e.g., day care centers, housing improvement, school or health post rehabilitation): 5. The degree of community involvement in the O&M of social subprojects depends on the type of investment. In the case of subprojects related to public services, such as health posts or schools, the municipality or relevant public/private entity could undertake O&M following an agreement with the community. For other types of subprojects such as day care centers or house improvement, the community will be responsible for O&M, each of the community members contributing either financially or in the form of labor. (c) Infrastructure Subprojects (e.g., small bridges, water supply systems, fords, road repairs): 6. Different arrangements are to be followed for O&M of infrastructure subprojects. In the case of services (e.g., water supply) connected to individual households, beneficiaries could transfer the corresponding assets to the relevant state or municipal utility company, which then undertakes O&M and recovers costs through user fees. Simplified water supply systems (without connection to households) will be operated and maintained by the community itself, which makes arrangements for collection of dues or fees from individual community members to pay for O&M expenses. These arrangements are commonly found in the rural water supply subprojects financed under the reformulated NRDP. For infrastructure subprojects such as roads, bridges, and street paving, etc., which serve a larger public, responsibility for O&M would normally be shared with the municipality under agreement with the beneficiary association. - 65 - Annex 7 Page 1 of 5 BRAZIL NORTHEAST RURAL POVERTY ALLEVIATION PROJECT - SERGIPE Project Monitoring, Reporting and Evaluation Monitoring 1. The establishment of an efficient project monitoring system will be instrumental in following the progression of the investment program and ascertaining its consistency with project objectives. Project implementation relies heavily on a decentralized system of decision-making and management, and is driven by demand from rural community associations which is filtered through a selection process based on established criteria. This process involves a large number of small, diverse subprojects over an extensive geographic area. Hence, since the ex ante determination of the pattern of project investment and other key parameters is not possible, effective supervision depends on the existence of a capable monitoring system. 2. The objectives of the project monitoring system are to: (a) provide information regarding the progress of the project and the status of the subproject cycle in its different stages (e.g., subproject approval, procurement, implementation and disbursement); and, (b) verify that project activities are consistent with the guidelines, targeting methodology and selection criteria included in the Project Operational Manual. 3. The project monitoring system would depend on a Management Information System (MIS), created under the reformulated NRDP and to be refined under the project. The project MIS is a data base which contains pertinent physical and financial information for each subproject, organized in the following format: by program (PAC, FUMAC, and FUMAC-P); by type (Infrastructure, Productive and Social subprojects); by category (e.g., water supply, transportation, energy, etc.); by sub-category, in particular for water supply subprojects (e.g., dams, wells, tanks, etc.); and by location (municipality, community and beneficiary association). Information for the MIS is inputted from data sheets for each subproject, separated into three modules which correspond to key steps of the subproject cycle: Module No. I would include information for the subproject regarding its location, identification, and technical, social and financial characteristics, as well as a set of indicators such as cost/capita, cost/m2, and other relevant parameters. Module No. 1 data sheets would be partially completed at the time of presentation of the subproject proposal and finalized after its appraisal. Module No. 2 data sheets would include information regarding - 66 - Annex 7 Page 2 of 5 procurement procedures and contracting, and would be completed after signature of the contract!convenio with the beneficiary association. Module No. 3 data sheets would present disbursement data and would be updated at the time of each payment installment to beneficiary associations and completed at the time of final installment at the conclusion of subproject execution. The Project Coordinating Unit (PCU) would be responsible for operating and maintaining the MIS and for providing the input corresponding to PAC subprojects. The project Municipal Councils would be responsible for completing the data sheets for FUMAC and FUMAC-P subprojects and, upon completion of each module, would present them to the PCU which would subsequently update the MIS. 4. Given the estimated number of subprojects to be financed during the five years of project implementation, the MIS is an essential tool for directing the course of project investment. Information from the MIS is also fundamental to measure compliance with and the effectiveness of the project's targeting methodology. By monitoring the direciion of project benefits and types of subprojects being financed, the PCU and the Municipal Councils can reorient investment incentives or make appropriate changes to the cost-sharing matrix, in agreement with the Bank. Progress in meeting indicative investment targets set for certain types of subprojects, such as subprojects with special environmental features, social subprojects and subprojects which benefit women, will be tracked regularly with the MIS and the information subsequently used to program technical assistance or promotion to corresponding groups. Cases of mistargeting or misappropriation of funds--resources channeled outside the project area or for ineligible investments--will be verified using the MIS in order for appropriate remedies to be applied. 5. Within the framework of the project monitoring system, the PCU and the project Municipal Councils (under FUMAC and FUMAC-P) are responsible for supervising the application of the guidelines from the Operational Manual in the development and implementation of all subprojects. Using the information from the MIS, combined with field visits and inputs from project-contracted studies and audits, the PCU and the Municipal Councils would monitor project characteristics and trends, identify implementation problems and accomplishments and undertake or promote appropriate actions to improve project implementation. The MIS will also serve as a basis from which to derive Annual Operating Plans (POAs), regulate implementation, and make necessary adjustments to the POAs and corresponding budgets during the course of each fiscal year. The delineation of monitoring responsibilities between the PCU and the Municipal Councils would be clearly reflected in the Project Operational Manual. Reporting 6. The PCU is responsible for reporting on project progress to the Bank. Four types of reports would be generated by the PCU and presented to the Bank: (a) a copy of the content of the MIS data base on diskette would be provided on a monthly basis to the Bank's Field Office in Recife. The information from the - 67 - Annex 7 Page 3 of 5 MIS will facilitate Bank supervision efforts and monitoring of the investment program; (b) an MIS-generated report which would include the updated status of project implementation (USPI) and an updated list of subprojects with their physical, procurement and disbursement status would be provided to the Bank's Field Office in Recife at the same time that the statements of expenditure (SOEs) are sent to Bank Headquarters. The USPI and subproject list will pennit the Recife Office to review the eligibility of subprojects presented for reimbursement; (c) a Semi-annual Project Progress Report, using standard formats agreed with the Bank, would be presented by July 31 of each year of project implementation; and (d) an Annual Report would be presented by January 31 of each year of project implementation. The Annual Report would use the standard formats from the semi-annual report but would also include a comprehensive review of: (i) the status of the project performance indicators and the specific activities of the institutional development program; (ii) subproject appraisal indicators, the application of the targeting methodology and subproject investment eligibility criteria, and other implementation guidelines included in the Operation Manual; and (iii) other relevant analyses and recommendations from project-contracted studies and audits which may be applied to improve project implementation. Evaluation Studies 7. The project would develop and implement a range of evaluation studies to measure the impact of the subproject investments and to provide appropriate feedback to improve project methodology and implementation. The PCU would be responsible for drafting Terms of Reference (TORs), which would form part of the Project Operational Manual, and contracting local and other consulting services, both steps in agreement with the Bank, for the following studies: (a) annual physical performance reviews; (b) a Mid-Term Review; and (c) a comprehensive impact evaluation. These studies are described below. During the course of project implementation, the PCU may also identify other needed evaluation studies which can be financed under the project in agreement with the Bank. 8. Annual Physical Performance Reviews. The objective of the Physical Performance Review (PPR) is to evaluate the success of investments financed by the project, in terms of their achievement of specific subproject objectives, and the quality, efficiency, and sustainability of the investments. For each subproject included in the sample, the study would focus on the following points: (a) the adequacy of the physical design in relation to the subproject's objective; (b) the quality and cost of subproject execution; (c) the efficiency of procurement procedures used; and (d) the targeting of project benefits. The PPR would provide indication of the effectiveness of community-based procurement; compliance with and/or needed adjustments to cost indicators and other physical/financial appraisal indicators - 68 - Annex 7 Page 4 of 5 and to the Standard Designs included in the Operational Manual; and the quality of project investments and the profile of distribution of project benefits. The PPR will also be an important input to the project's Mid-Term Review and to project Annual Reports. 9. The PPR will be conducted on the basis of TORs agreed with the Bank and included in the Operational Manual. The first PPR will be conducted in 1997 for a sample of subprojects representative of the 1996 investment program (PAC, FUMAC, and FUMAC-P). To conduct the PPR, the PCU would: (i) establish, with Bank agreement, the sample of subprojects to be reviewed; (ii) supply counterparts to the independent consultants contracted by the PCU to implement the study; (iii) participate in the diagnosis and generation of appropriate recommendations; and (iv) incorporate the recommendations into the implementation of the project. Subsequent PPRs would be conducted in 1998, 1999 and 2000, for the exercises corresponding to 1997, 1998 and 1999, respectively, and would review physical performance issues identified during the first PPR and monitor improvements. TORs for the remaining PPRs would be agreed with the Bank by no later than December 31 of each previous corresponding year. 10. Mid-Term Review. The purpose of the Mid-Term Review is to assist the State Government, the PCU and the Bank in the identification of positive experiences and implementation problems and bottlenecks, in order to recommend needed modifications to project methodology and procedures for incorporation during the second half of the project implementation period. Among the various aspects of project implementation to be reviewed, the Mid-Term Review would evaluate the efficiency and impact of the different subcomponents (PAC, FUMAC and FUMAC-P), the effectiveness of institutional development activities, the level of community participation, and the ability of the project to target and reach the poor. The Mid-Term Review would be carried out in 1998 by independent consultants, under T)Rs and contracting procedures agreed with the Bank. Inputs to the Mid-Term Review would include: the semi-annual and annual reports, MIS- generated analyses and reports, results of the PPRs, recommendations from project external and internal audit reports, and additional reviews or studies contracted by the PCU. 11. Project Impact Evaluation. In addition to the previously mentioned two studies, which focus primarily on issues of subproject implementation, the project would finance a series of three (baseline, mid-term and final) evaluations which measure the socio-economic impact of the subproject investments. This comprehensive evaluation, based on a random sample survey of representative subproject types, would select important socio-economic indicators, such as employment, income, health, and education, and measure the impact of subproject investments on these indicators and consequently the impact of the project on the quality of life of the beneficiary population. The evaluation would also relate impact back to specific aspects of the project in order to identify those conditions or activities which resulted in positive impacts. The indicators would be identified and measured during the baseline evaluation in the first year of project implementation, and monitored thereafter in the Mid- Term impact evaluation (during the third year of implementation) and in the final evaluation at the conclusion of implementation. The impact evaluation would be carried out by independent consultants and contracted under TORs and procedures acceptable to the Bank. - 69 - Annex 7 Page S of 5 If under the new project the same patterns continue, as under the reformulated NRDP, productive subprojects could result in the creation of about 46,000 new jobs and could generate net incremental income of about US$890 per year. About 34,000 families could have increased access to safe water. These projections are, however, only illustrative and are not targets, since the project will be entirely demand-driven, and therefore community preferences could vary significantly. Implementation Plan and Project Performance Indicators 12. The implementation plan and project performance indicators for each project component over the five-year implementation period are shown in Table 7.1 at the end of this annex. Given the demand-driven nature and flexibility of the project, these indicators are mostly indicative targets. They can be expected to be achieved at least in the initial year of the project but may well develop somewhat differently in subsequent years, particularly in terms of the proportion that each main category of subprojects (infrastructure, social or productive) would represent out of the total investments financed by the project. NORTHEAST RURAL POVERTY ALLEVIATION PROJECT - SERGIPE Project Implementation Plan and Performance Indicators- Project Components and Activities Responsible Institution Unit PYO PYI PY2 PY3 PY4 PY5 Total A. Community SubyroiecLs Subprojects ImplementedZ; Beneficiary Associations No. 468 468 426 404 364 2130 -Infrastructure " No. 130 130 118 112 100 590 -Social No. 48 48 44 42 38 220 -Productive " No. 290 290 264 250 226 1320 Beneficiaries -Total1' Families 14 14 13 14 11 66 ('000) -Total People 70 70 65 70 55 330 ('000) -women as % of productive % 30 30 30 30 30 30 subproject beneficiaries Community Associations' Beneficiary ° Associations No. (cum.) 234 468 681 883 1065 1065 Municipalities Assisted PCU No. (cum.) 55 70 70 70 70 70 Municipal Councils -FUMAC No. (cum.) 30 45 45 45 45 45 -FUMAC-P No. (cum.) 8 8 8 8 8 8 B. Institutional Development Preparation of annual program of technical PCU, Municipal assistance and training Councils (i) PCU No. I I I I 1 5 (ii) Municipal Councils No. 38 54 54 54 54 254 No. of workshops/seminars for - Beneficiaries 1' 55 15 -- -- -- 70 - Project Municipal Councils 6 8 4 -- -- -- 12 (FUMAC and FUMAC-P) - FUMAC-P Municipal Councils only6' 2 -- -- -- -- 2 4< Project Components and Activities Responsible Institution Unit PYO PY1 PY2 PY3 PY4 PY5 Total No. of courses "' offered to: - Beneficiary associations no. 23 47 68 88 107 333 - Project Municipal Councils no. 38 53 53 53 53 250 (FUMAC and FUMAC-P) 2' - FUMAC-P Municipal Councils only 9/ no. 8 8 8 8 8 40 - PCU staff no. 2 2 2 2 2 10 C. Proiect Administrtion. Monitoring and Evaluation Special Account PCU x Publicity Campaign PCU -Presentation to Bank x -Campaign Launched x x Operational Manual PCU No. x -Preparation x x x x x -Review and Adjustments (with Bank apprval) Stuadad Designs -Preparation PCU No. 6 15 21 -Review and Updating x x x x Annual Operating Plans (POAs) -Preparation of POA for FUMAC-P FUMAC-P Councils 8 8 8 8 8 40 -Consolidation and Preparation PCU I 1 1 I 1 5 of Project POA Monitoring Reports (sent to Bank) -Mondtly Disbursement PCU No. 12 12 12 12 12 60 Summaries and updating of MIS database -Annual/Semi-annual Reports PCU No. 2 2 2 2 2 10 -External Audits PCU No. 1 1 1 1 1 5 Project Components and Activities Responsible Institution Unit PYO PYI PY2 PY3 PY4 PY5 Total Evaluation Studies -Physical Performance Reviews PCU No. 1 1 1 1 4 -Mid-Term Review PCU No. 1 1 -Impact Evaluation PCU No. (i) Baseline 1 (ii) Mid-Term 1 1 (iii) Final 1 1 . Because of the demand-driven nature and flexibility of the project, these targets are indicative and will be re-assessed and revised during the Mid-Term Review. 2. Based on average costs of approved subprojects in each category in Sergipe. 3. Estimated on the basis of 30 beneficiary families per subproject. 4. Assumes an average of 2 subprojects per association (based on NRDP-Sergipe experience). 5. At least I initial workshop for all communities within each project municipality. 6. 1 workshop for every 5 new Municipal Councils. 7. Specialized skills training. 8. 1 basic training course initially and at least 1 specialized skills training course per year subsequently for groups of 10 associations. 9. 1 basic training course initially and at least I specialized skills training course per year subsequently for each Municipal Council. to -J CD3 BRAZIL NORTHEAST RURAL POVERTY ALLEVIATION PROJECT - SERGIPE Annual Phasing of Project Costs by Component (US$ million) Project Component Totals Including Contingencies 1996 1997 1998 1999 2000 Total A. Community Subprojects 1. FUMAC 7.8 7.8 7.1 6.8 6A1 35.6 2. FUMAC-P 1.3 1.3 1.2 1.1 1.0 5.9 3. PAC 3.0 3.0 2.7 2.6 2.3 1316 Sub-Total 12.1 12.1 11.0 10.5 9.4 55.1 B. Institutional Development 0.7 0.7 0.5 0.3 0.2 2.4 (Technical Assistance and Training) C. Project Administration, Monitoring and Evaluation 0.3 0.3 0.4 0.4 0.4 1.8 TOTAL BASELINE COST 13.1 13.1 11.9 11.2 10.0 59.3 Physical Contingencies 0.1 0.1 0.1 0.0 0.0 0.3 Price Contingencies 0.0 0.0 0.1 0.1 0.2 0.4 TOTAL PROJECT COST 13.2 13.2 12.1 11.3 10.2 60.0 - 74- Annex 8 Table 8.2 BRAZIL NORTHEAST RURAL POVERTY ALLEVIATION PROJECT - SERGIPE Financing Plan (US$ million) Project Component Amounts to be Financed by State Goverrnent/ Total Cost IBRD Municipalities Beneficiaries&' A. Community Subprojects 1. FUMAC Sub-projects 35.6 21.0 11.1 3.5 Technical Assistance and Training 0.9 0.9 0.0 0.0 Sub-total 36.5 21.9 11.1 3.5 2. FUMAC-P Sub-projects 5.9 3.5 1.8 0.6 Technical Assistance and Trining 0.3 0.3 0.0 0.0 Sub-total 6.2 3.8 1.8 0.6 3. PAC Sub-projects 13.6 8.0 4.2 1.4 Technical Assistance and Training 0.4 0.4 0.0 0.0 Sub-total 14.0 8.4 4.2 1.4 B. Institutional Development 0.8 0.8 0.0 0.0 (Technical Assistance and Training) C. Project Administration, Monitoring 1.8 0.6 1.2 0.0 and Evaluation TOTAL BASELINE COST 59.3 35.5 18.3 5.5 Physical Contingencies 0.3 0.2 0.1 0.0 Price Contingencies 0.4 0.3 0.1 0.0 TOTAL PROJECT COSTb 60.0 36.0 18.5 5.5 a/ Minimum contribution. b/ Including about US$2.4 million of taxes and duties not financed by IBRD. - 75 - Annex 8 Table 8.3 BRAZIL NORTHEAST RURAL POVERTY ALLEVIATION PROJECT - SERGIPE Allocation of Loan Proceeds Category Item and % of Expenditures Total Amount (US$ million) I. Community Subprojects A. FUMAC 18.9 B. FUMAC-P 3.1 C. PAC 7.2 59% of expenditures 11. Institutional Development including Technical Assistance, Training and Evaluation Studies 2.5 100% of expenditures in. Project Administration and Monitoring System 0.7 20% of expenditures (except salaries) IV. Unallocated 3.6 TOTAL 36.0 - 76 - Annex 8 Table 8.4 BRAZIL NORTHEAST RURAL POVERTY ALLEVIATION PROJECT - SERGIPE Estimated Schedule of Bank Disbursements (US$ million) Bank Quarter Disbursement Cumulative Disbursement Balance FLscal Ending During of Year Quarter Amount % of Total Loan 1996 March 31, 1996 -- -- -- 36.0 June 30, 1996 5.5.' 5.5 15.3 30.5 1997 Sept. 30, 1996 1.5 7.0 19.4 29.0 Dec. 31, 1996 1.6 8.6 23.9 27.4 March 31, 1997 1.7 10.3 28.6 25.7 June 30, 1997 1.7 12.0 33.3 24.0 1998 Sept. 30, 1997 1.7 13.7 38.1 22.3 Dec. 31, 1997 1.7 15.4 42.8 20.6 March 31, 1998 1.8 17.2 47.8 18.8 June 30, 1998 1.8 19.0 52.8 17.0 1999 Sept. 30, 1998 1.9 20.9 58.1 15.1 Dec. 31, 1998 1.9 22.8 63.3 13.2 March 31, 1999 1.9 24.7 68.6 11.3 June 30, 1999 1.8 26.5 73.6 9.5 2000 Sept. 30, 1999 1.8 28.3 78.6 7.7 Dec. 31, 1999 1.8 30.1 83.6 5.9 March 31, 2000 1.7 31.8 88.3 4.2 June 30, 2000 1.7 33.5 93.1 2.5 2001 Sept. 30, 2000 1.5 35.0 97.2 1.0 Dec. 31, 2000 1.0 36.0 100.0 0.0 a! Including the authorized allocation of US$3.0 million into the Special Account. - 77 - Annex 9 Page 1 of 2 BRAZIL NORTHEAST RURAL POVERTY ALLEVIATION PROJECT - SERGIPE Project Operational Manual (Outline) The Operational Manual would have the following table of contents: I. Introductory remarks II. The Organization of the Manual III. The Project A. Objectives B. Strategy C. Targets D. Project Area E. Components F. Eligibility criteria 1. Compliance with poverty targeting methodology 2. Consistency with standard technical designs 3. Consistency with financial/cost parameters 4. Socio-economic evaluation 5. Environmental assessment G. Operational guidelines 1. Project coordination 2. FUMAC 3. FUMAC-P 4. PAC 5. The subproject cycle: identification, preparation, submission, evaluation, implementation, conclusion, technical assistance, operation and maintenance 6. Procurement 7. Cost-sharing matrix 8. Fund release and statements of expenditure 9. Operational Checks and Balances 10. Monitoring and evaluation - 78 - Annex 9 Page 2 of 2 Annexes: A. Standard subproject designs B. Standard agreements C. Standard forms (for proposal submittal, ata de recebimento, etc.) D. Guidelines for FUMAC and FUMAC-P Municipal Councils E. Management Infornation System (MIS) and reporting F. Flow of funds and statements of expenditure G. Project administration H. Technical assistance, training and studies - 79 - Annex 10 Page I of 2 BRAZIL NORTHEAST RURAL POVERTY ALLEVIATION PROJECT - SERGIPE Supervision Plan 1. The principal responsibility for supervision of the Rural Poverty Alleviation Project will rest with the State Government through the Project Coordinating Unit (PCU), and with the Bank through its Recife Office in conjunction with staff from Headquarters. The Municipal Councils will also supervise the execution of community subprojects within their respective municipalities. International agencies, such as IICA or FAO, are expected to play an important role in performing supervision activities which require specific expertise. Selected NGOs may also be contracted by the State or the FUMAC-P Municipal Councils to assist with supervision activities and evaluation of the status of community organization and mobilization. 2. Supervision will be performed at two levels: at the project level, where the PCU will monitor global trends across all project municipalities; and, at the subproject level, where the PCU staff and the project Municipal Councils will monitor the performance of individual subprojects. The PCU will set up a team to supervise the subproject cycle on a continuous basis. At the project level, the State's responsibilities for project supervision are: (i) to focus on global project implementation and administration; and, (ii) to review activities of the FUMAC and FUMAC-P Municipal Councils. Supervision of project Municipal Councils by the PCU will focus on: the Councils' composition, to ensure satisfactory beneficiary representation; the frequency and openness of Councils' meetings; and their effectiveness in providing assistance to the beneficiary communities. In addition to these points, the PCU's supervision of the FIJMAC-P Municipal Councils, in particular, will focus on the criteria applied to screen and prioritize subproject proposals and the administrative and financial arrangements made by the Councils to effectively control and release project funds to the community associations. 31 At the subproject level, the PCU, together with the corresponding Municipal Councils, will carry out the following subproject supervision tasks: (i) verification of the status and quality of subproject implementation (PAC by the PCU and FUMAC and FUMAC-P by corresponding Municipal Councils); (ii) assessment of community participation in the identification, preparation and execution of subprojects; (iii) evaluation of the extent to which communities are organized to undertake the operation and maintenance of the investments; and (iv) determination of whether communities satisfy the targeting criteria. Information collected during supervision, including indicators of subproject implementation performance and identification of principal bottlenecks or problems, will be incorporated into the standard data sheets completed by the PCU for each subproject (Annex 7). 4. Based on the information collected at the project and subproject levels, the PCU will prepare and submit to the Bank semi-annual and annual reports on the overall financial and -80- Annex 10 Page 2 of 2 physical status of project implementation and performance. The PCU will also provide the Bank's Recife Office with copies of all SOEs and reimbursement applications when these are submitted to Bank Headquarters (Annex 7). In addition, the PCU will present the Annual Operating Plan (POA) and corresponding budget for each subsequent fiscal year of implementation to the Bank for its review and comment. 5. The Bank will carry out at least two regular supervision missions per year. Regular supervision will entail: (i) a discussion of project implementation problems identified by the PCU; (ii) field visits to a random sample of community subprojects; and (iii) a review of the performance of all project components, and particularly an intensive assessment of the FUMAC- P pilot program, to address problems and make appropriate recommendations. In addition, based on frequent communication with the PCU, the Recife Office will conduct short supervision missions to address problems which require prompt action. Also, in view of the diversity of community subprojects, consultants of different expertise would be contracted to provide occasional input and advice to Bank missions. Two full supervision reports will be issued annually, as well as Back-to-Office Reports for each short supervision mission. It is estimated that about 45 staff-weeks will be required for project supervision, or about 9 staff-weeks annually during the five years of project implementation. 6. The schedule of project supervision is as follows: (a) by July 31 and January 31 of each year, presentation of semi-annual and annual reports from the PCU to the Bank; (b) by February and September of each year, full Bank supervision missions completed; (c) by October 30 of each year, presentation of the POA for the subsequent year to the Bank for review and comment by November 15. - 81 - Annex 11 PaLe 1 of I BRAZIL NORTHEAST RURAL POVERTY ALLEVIATION PROJECT - SERGIPE List of Documents Available in Project Files 1. Anufirio Estatistico de Sergipe: 1992, Secretaria do Estado do Planejamento (SEPLAN)/Superintendencia de Estudos e Pesquisas (SUPES). 2. Boletim Conjuntural--Nordeste do Brasil: 1994, Superintendencia do Desenvolvimento do Nordeste (SUDENE)/ Fundacao Joaquim Nabuco (FUNDAJ), 1994. 3. Brazil: NRDP and a Possible Follow-Up Program, draft, FAO/CP, February 1995. 4. Censo Demografico 1991: Resultados do Universo Relativos as Caracteristicas da bpula,o e dos Domicilios--Brasil, Numero 1, IBGE, 1991. 5. Censo Demografico 1991: Resultados do Universo Relativos is Caracteristicas da Pbpulaco e dos Domicilios--Sergipe, Numero 16, EBGE, 1991. 6. Informac,es sobre o Estado de Sergipe, PRONESE/SEPLAN, 1995. 7. 0 Mapa da Fome, IPEA, three volumes, 1993. 8. Municipios Brasileiros--Criancas e suas Condioes de Sobrvivencia, UNICEF/IBGE, 1994. 9. Programa de Governo Albano Franco, 1995. 10. Projeto de Desenvolvimento Municipal Sustentfivel, ARIDAS, October 1994. 11. Safety Nets and Social Funds to Alleviate Poverty: Performance Problems and Policy Options, Issues Note, UNCTAD, October 29, 1993. 12. A Study of Popular Participation in the Brazil Northeast Rural Development Program (NRDP-PAPP), Conrad Kottak, Febniary 7, 1994 (Draft). J8 o0 37530 ,- ( VENEZUE3A7 /r ~~~~~ ~~~Alt,..... o1,ci B R A Z I L r'y , h r AMAA RURAL POVERTY ALLEVIATION PROJECT ' --. S ERGIPE _J AMAZONAS ,/ PARA c - , . s < . A r ' ~~~PIAUI 7.,s, ' t; i yI'nrA,B R A Z I SERGePE ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~PE RU _!AiAlo10 ) ZA,I,_ 9 30' °,0 10 20 3? 40 50 930 RU GR0550 S S k 0 KILOMETERS . BoLivIA I An MINAS GfRAIS .,sMg,, Ih eAOul A/omo , u_ r a { 5 , D C J j o u kHIIEJ P Ge.p; ,,a JANEIRO ,/ S Frc.nc,.-Ot I\ ,, 7 0u { J' X~~~~~~~1. OUNDArlE5 3700' 3 0 A L A G 0 A S 10'00 10 00' A4, . .gp F~~~ Aqu.d.~Coli bo B A H I A N.-< < | >s7) ( ~~~~~~~~~~~~~AREA EXCLUDED FROM THE PROJECT Pi JJ.p/ ,., .t I,.\ AVERAGE ANNUAL RAINFALL IN MILLIMETERS S, a | \ S/ / Ut~~~~~Sonlo I$ijo o Mn1, CLIMATIC ZONES: M { \ }__ ./Umboabo / s)t . ( gd HUMID SEMI-ARID fiI, baandwolls, cala,s. 2 h d- y / Dmu SELECTED CITIES AND TOWNS lI'30' L,norn,nol,an. and anr N II33' n,iA.,i.tinpmdaaa ;bcwn , - l so$+ndooo<,- MAIN ROADS imply aton aai d aol Nf EOf pLbida 'I 't 1 STATE CAPITAL Pha WaddbL GRE Oa rp. any jadganoni on hd bogof STATE ROUNDARIES_ r orocccepIonce of SiCE m h,acndnnoe, ai O a b-. 38 00' 37E3' 37100 36 30 a I~~~~~~. IMAGING Report No: 14394 BR Type: SAR