Mexico - Decentralization and Regional Development Project For the Disadvantaged States

0ocument of The World Bank FOR OFF0 IAIL USE ONLY' Report No. 8786-ME STAFF APPRAISAL REPORT MEXICO DECENTRALIZATION AND REGIONAL DEVELOPMENT PROJECT FOR THE DISADVANTAGED STATES MARCH 4, 1991 Agriculture Operations Division Country Department II Latin America and the caribbean Regional Office This document has a restricted distribution and mas be used bv recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENTS Currency Unit = Mexican Peso (Mex$) US$1 = Mex$2,950 Mex$1 million = US$340 (December 1990) FISCAL YEAR January 1 - December 31 WEIGHTS AND MEASURES 1 meter (m) - 3.28 feet (ft) 1 kdlometer (km) = 0.62 mile (mi) 1 hectare (ha) = 10,000 m2 = 2.47 acres 1 square kilometer (kin2) = 0.38 square miles (mi2) = 100 ha 1 metric ton (m ton) = 2,205 pounds  FOR OFFICIAL USE ONLY AGSAtL AgricultU) 9fSector Adjustment La ANOAP% Asocist4o Xactoaal de Oqianismoi Operadore de Agua Potable (Nationa Association of Potable U tter Agecis) BANORRAS CAO Catnmos y Autopistas de Oaxaca (Roads and liigbway of Oaxaca) CAPAEG Comisi6a de Agu Potable y Akautarillado del Estado de Guerrero (State % ater and Sewerage Comtismo of Guerrerol CAPfCE ComRi Administrador Wd Programs Federal de Construccidn de Eauela (balagemeat Comait.ee of the Federal School Cotutnticon Program) CEC Comisifa &Etatal de Caminos de Chiapu (Cliapas State Road Com on) CEM Country Economk Memorandum cmSP Comisid Estat de Savids Ptbikos de Hiidigo (State Water Comminsion of Hidatgo) CFE Coaisi;o Fedl de Electricidad (Federal Commission for E&ectncty) CLC Cuu pr Liqudar Ceilfcadas (C Fed Accounts Payable) CNA Comisioo Nacioal del Aguas (Natioal Water Commision) COCSOSA Comisift Coordinadora de lo Sstema de Obras Servicis de Agus (Coordinating Commi for Worh and Water Senices) CONACAL Comesit Nacional de Carreteras Alimeatadeoras (National Commission for Feeder Roads) COPLADE Comld60 de Planeadd dd Dearroo Estat (State Commision for Deveopment Planing) COPLADEM Comisi6o de PlaneaciUn W Desarrollo Municipal (Muni;Cpal Commission or Developmet Planning) COPLADER Comisido de Planeacido de DourroUo de Is Rqeono (Regiwal Commission for Development Plannig) COPLAMAR Coordinacido Genral del Plan Naclon de Zone Deprimidas y Grupos Marginales CONAFRUT Comisid Nacioal de Fruticultura (National Commisn for Fruit Crops) CONAPO CoeJo Nadconal de Poblacidn (National Populato Council) CONtASPO Comisido Nacion de Alimentacifn (Natioal Commuision for Nutrition) CUD Convenio Unico de Desrrdoo (Development agremtent bween federal govenent and eacb state) CUDEMUN Coaveaio Unico de Dearrollo Municipal (Development Agrmenet between State ad Muaicipality) DAPA Direcida de Age Potable y Alcatarillado (Sate Water Agency of Chiapas) DGCT Direcido Genral de Comuniicaioe y Traisporte de Hidaelo (Genea Directorate of Communications and Transport of fidalg) DDR Rural Development Districts DDRI Disrito de DesorroDo Rural Intetrdo (Integrated Rural Developament rid) ERR Economic Rate of Retum FIRCO Fideiaomiso de Rieg Compartido (Trwt Fund for &iIsharing) FMS Fondo Munwicia de Solidaridad (Municipal Solidarity Fund) GIRA Gera Interest Rate AgmeDt IMTA lontituto Meaco de Teaologia del Aga (Mexican lnsttute of Water Technology) INAi Ins;ttto Naional de Antropologla e Hishoia (National Insiute of Athropoloy and History) INAP Instituto Nacional de AdministraciOn Psblica (National Public Adminstraton Isthut) INEGI National Instute of Statistics, Geography ad Information This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  INI Itteaial get INWAP I I _dy e NaialIstitute for Crop. Anmle Forat INNMCAFE Ib eot Mano dl Ca3f (eca Coffee lnstue) MAQUINOP M4ide p Obrus P de Guoa. SteRAgW4y of Gterr) NARN Naionl Fc, SNC (Nao tl e CepMY) NGO Noqoverumet orgaaizatLom PACTO Pact. Eo de Sod PAEO Potableg AastsYl d EttAdo& d OauCi (Stae War Mum of Oaaa PASSPA A S p Pebtodi Mzirfa PECE Pato de Estabizacida y Crecimiemto Eeoedwic. (Stbilizat nd CGre*b Fact) PIDER Proraiantg d Desarb5 Rue PdER p PLANAT Pan Nodnal de Apwo pr A#actr de TeDporal (Nation Asdstance for Rded Arem) PME Pr ade Modersaodn Edouco (Egdaonal Modenzatino Progrm) PNS rm N de Soldda -N n S d Progw) POA Pregama Opeo Aa (Annua Operatie Prm) PROCATI Progrma de Capacitaci6o, AsiTaecia Tcniae nvtigacida (Program for Tramung, Etenos and Reesrub) PRODERITH Prgm for the Development of Humid Tropic PRONASOL PNas Naboal de S (National Solndanty Program) PSE Prgas d Solidaida Esata (Ste Soliday Program) RP.C Regon Plnni Center SARH Secar dA A y Recusos lribuls (Sertrt of Agiutrie and Water Reisre.) SCG Special Coordinatfg Grop SCT Secretarfa de C _mnacon y Tr_anspmoie (Secretara of CommuQ atmos ad T_port) SDUC Secretaria de Desarrollo Urbano y Cokadone de Chi?as (Secretariat Of Urban Developoei4t andCnmuni ofCbipas SECOGEF SeCretart do ls Contrlora G a do Is Federacion (Ger Sr f Comptroers of the Fedeation) SEDUCOP Secretaria de Desarrollo Urbane, C k y Obms Publcas de HPao (Secrtaat of Urba Devetopent, Q t , and Pblc Works of hidalgo) SEDUE Secretar' d Desarroto Urbano y EcdqS (Secrait of Ura Developunaet g d_ SEP Secetarfa de EdwcaiOn Pubblc (Secretariat f Plc Education) SHCP Scretarts de H y Cr Pdblio (Secretariat of Finance and PbiCC Credik) SOE Statnait of Expaditue SPP Seretarfa de Progracia y Presput (Seetariat of rgramming and Bidgeing SWS Suo de Agua Potabe y Seneato (St. Waer Supply and Saitation Subcomm)tte) WSSP Water Supply and Santain Secor Prjec S Aho kjun a PRONASOL  MEXICO DECENTRAUZATION AND REGIONAL DEVELOPMENT PROJECT FOR THE DADVANTAGEDSTATES Tblk of ConznX Pag No. L LOAS AND PROJECT SUMMARY ............................................. I M. REGIONAL AND RURAL DEVLOPMENt POLIC,S ................................. 4 A. Introduction ... ..................................................... 4 1. Poverty and Regional Dynamics .......................................... S 2. Th Role of a Regiona Investment Progrm in an Overall Stwagy of Poverty Alleviation ....... 5 3. Why Focus on the FourPooetStes? ......Poores............................... 6 4. Doentrzation ................................................... 7 5. Equityin AccesstoPublic Resours ....................................... 8 B. Incoporating Lesons from Experience.. 9 C. Recnt Policy Reformns ........................1.......................... 11 1. Improving the Regional Planning System ..................... 11 2. Increasing Flexibility in Resource Channelng ............... .. .................. 13 3. Administrative Decentralizaton ........................................... 15 D. Governtent Stwategy ........................ ........................... 16 E. Rationale for Bank Involvement ...................... ....................... 17 Il. THE PROJECT .......... ............................................... 19 A. Origin of the Project ................................................... 19 B. Project Objectives ..................................................... 19 C. Project Description and Components .......................................... 20 1. The Development Agreements (CUDs) and the Regional Planning Mechanism ..... ........ 21 2. Providing Resources for Poverty Alleviation through the Investment Prgratms of the Four Poorest States ......... ............ ........................ 22 (a) Maintaining the Poverty Focus of Investment Subprograms ..................... 23 (b) Project Appraisal and Selecion ...................................... 24 (c) Financing and Budgeting for Investment Cost ............................. 26 (d) Beneficiary Contributions to Investment Cost ............................. 27 3. Financing Recurrent and Opeation and Maitenance Costs ...................... .. 27 4. Coordination with Sectoral Policies ....................................... 28 (a) Agriculture .................................................. 28 (b) Social Services ............................................... 30 (c) State and Rural Roads ........................................... 31 (d) Ruml k ha t ..........l................................... 33 (e) Protecting the Environment ........................................ 34 (f) Protection and Rehabiliation of Cultural Patrimony ......................... 35 5. A Four-Year Proposed Investment Program ............................ 35 (a) Preparation of investment Progmams ............................. 36 (b) Proposed Investment Programs ......... ................... 36 (c) Implementation Capacity ............................ 38 6. Project-Supported Institutional Developm.ent .. 39 (a) Strenbhening State Institutions ......... ................... 40 (b) StrengtSening MunicipalWits ............................ 42 (c) Strengthening Institutions to Implnent Envirmental PoLcies ................... 42 Tbis report is based on the findings of an apptaisal mission that visitd Mexico May 7-12, 1990. Mission members were Mews. Abel Mateus (Task Manager), Gian Carlo Guarda, Carmen Hamann, Alin Tobelem, Hubaus von Pogrel and Maritta Koch-Wesr. It also incorporates the finding of ealier prepsation work by Messrs. Shelton Davis, Angel Gonzalez- Maaxechevarria, Christopher Hennin, Caoline Mascarell, and Jean-Claude Sallier, as wel as the following consultants: Chars MeLure, Mario Ramos, Ingrid Bmxell, Frncico Hoyo, Eduardo Gdlo, Robin Boadway, and Alicia Barcea.  -u - Pare No. D. Projct Organization andl Management . ..... . 42 1 Project Morlitoriig and Evaluation (M&E) .................... I .......... 43 2. Procuement 44 3. Disbumment . ....... .............. 46 4. Audit ..48 5. Bank Supervision ..49 E. Project Costs . . .. 49 F. Benefita and Risks . .. .. 0 IV. AGREEMENIS REACHED AND RECOMMENDATIONS ... . . 53 ANNEXES 1. POLICY LETTER ............................................ ........... 56 Attachment l-Decentralization Matrix ... .... . . . 63 Attachment 2-Perfornance Indicators . 67 2. THE REVENUE-SHARING SYSTEM ... . 74 3. STATE AND MUNICIPAL FINANCES ... . 79 Attachment 1-Projections of State Finances of the Four States .83 Attachment 2-State Finances and Financing of Investment Programs (CtS$) .84 4. CRITERIA FOR INVESTMENT ALLOCATION AND PROJECT SELECTION ... 86 Attachment 1-Project Selection: Institutional Functions and Products .. . 91 Attachment 2-Road Component .. ........... 93 Attachment 3-Rules for Subproject Environmental Assessment . ........ 95 5. COST RECOVERY RULES FOR SUBPROJECTS FINANCED BY THE PROPOSED PROJECT .. .. 96 6. OPERATION AND MAINTENANCE GUIDELINES ... ... . 98 7. CORE INVESTMENT PROGRAM-PROJECT COSTS ................................ . 101 8. RECURRENT COSTS IMPLICATIONS OF INVESTMENT PROGRAMS ... 104 9. ENVIRONMENTAL COMPONENT ...... . 106 Attachment 1-Cost of Environmental Component . . I ..... . 112 10. IMPLEMENTATION ARRANGEMENTS, CAPACITY AND THE INSTITUTIONAL DEVELOPMENT COMPONENT .113 Attachment 1- Table 1: Technical Assistance Plan &0' 'osts .122 Table 2: Training Plan and Costs . ... ....... . . 123 Table 3: Municipal Institutional Capacity per Function .... 124 Attachment 2-Project Imputs: Institutional Development Component ... ............. . 125 II. PROJECT ORGANIZATION AND MANAGEMENT ................ I ................. 127 Attachment 1-Project Organization Chart ........... 130 Attachment 2-Flow of Funds Chart ... 131 Attachment 3-Flow of Funds ... I ... 132 12. PROPOSED MONITORING AND EVALUATION SYSTEM ... ....... 133 Attachment 1-Terms of Reference ... 141 Attachment 2-Institutional Development Obicetive ..... 143 13. SUPERVISION PLAN ........ 145 Attachment 1-Civil Works Supervision Schedule ..... 147 14. STUDIES TO BE UNDERTAKEN UNDER THE PROJECT ..1.0...... lS0 1S. ESTIMATED SCHEDULE OF BANK DISBURSEMENTS .. ........ 153 16. DOCUMENTS AVAILABLE IN THE PROJECT FILES ..1..5.4.......... 1S4 CHARTS: Chart No. WB22312R: State of Hidalgo. Index of Poverty . .5.5........... lSS Chart No. WB22410R: State of Chiapas. Index of Poverty . ........ 156 Chart No. WB22411R: State of Guerrero- Index of Poverty . ..... .... ...... 157 Chart No. WB22412R: State of Oaxaca. Index of Poverty . ..1.8.... I MA: IBRD Map No. 22311R: Mexico - Decenralization and Regional Development Project.  MEXICO I)ECENTRALIZATION AND RE(GIONAL DEVELOPMENt LPRO.ECT FOR THE DTSAIVANTAGED STATF-S 1. LOAN AND PROJECT SUMMARY BoffQwer: Nacional Financiera, S.N.C. (NAFIN) Guaraator: United Mexican States Egecutillg Secretariat of Planning and Budgeting (SPP), other Federal Entities, the Project Ager1gy: States of Chiapas, Guerrero, Hidalgo and Oaxaca, and Participating Municipalities. Amount: US$35G million equivalent Tenns: Seventeen y!ars, including five-year grace period, at the standard variable interest rate. Beneficiarigs: Target groups in rural poverty areas and groups of producers (about five million persons) in the states of Chiapas, Guerrero, Hidalgo, and Oaxaca. Public works under the project would generate about 230,000 man-years of employment in the four states. PuMogse. To contribute to the reduction of poverty in Mexico's four most disadvantaged states, through a program of decentralization and regional development; complementary loan components include environmental preservation and the protection of archeological sites. Backgund: Of the 21 million people living below the poverty line in Mexico, two-thirds live in rural areas. About one-third of Mexico's poor (seven million persons) inhabit Chiapas, Guerrero, Hidalgo and Oaxaca. The poor constitute about 65 percent of the population of these states. The four states have large disadvantaged Indian populations, and a concentration of poverty among women and children. While poverty in these areas has longstanding structural roots, it has worsened during Mexico's economic crisis beginning in 1982. Between 1982 and 1988, public investment declined by about 60 percent. Cuts in rural, regional and social sectors were even deeper, reaching about 70 percent and leading to the collapse of poverty reduction programs in the four target states. The reduction of investment contributed to a decline of about 20 percent in rural real incomes between 1983 and 1988. Past Government policies have hampered the economic and social development of the four states: (a) exchange rate, trade and pricing policies have had an anti-agricultural bias; (b) public expenditures on social services have favored the better-off states; (c) despite the remoteness and dispersion of many settlements, infrastructure investments have been less than the national per capita averages; (d) the excessively centralized administrative, financial, and fiscal systems have not allocated resources efficiently to the very heterogeneous and peripheral regions of the states; and (e) agricultural services (research, extension, marketing) have been underfunded and poorly developed. Current Government strategy and the proposed project are directed at redressing some of these past policies. Project The principal objective of the project is to increase the access of poor and indigenous Djcription: populations in the four poorest states of Mexico to basic infrastructure, social services, larger markets, and agricultural technologies. The specific project objectives are: (a) to  2 - assist the states in developing, executing and financing investment programs with a strong poverty adleviation focus, (b) to assist the Federal and State planning and budgeting ministries in making their project selection, environmental assessment, and monitoring and evaluation functions more efficient and the financing more flexible; and (c) to strengthen the decentralized state and municipal institutions to identify, prepare, build, operate, and maintain the investments in a more participatory manner. Environmental preservation and the protection of archeological sites are complementary project objectives. T'e project has three components: (a) investment; (b) environmental and archeological site protection; and (c) institutional development. The inykmmne component would finance a US$1.3 billion invement program in the four target states. 'he Bank loan would allocate US$274 million to finance 50 percent of the cost of projects in the core investment program. About 46 percent of the program would be for infrastructure projects (rural roads and rural electrification), 30 percent for productive projects (agriculture, forestry and agroindustries), and 24 percent for social projects (education, water supply and sewerage systems). Projects will be selected on the basis of their targeting of the poor, technical feasibility, benefits relative to costs, and beneficiary contribution and commitnent to operation and maint3nance. The environmental preservation and archeological protection component of the project would: (a) finance 100 percent of the eligible costs (US$43 million) of a program (the Lacandona Protection Plan) which would protect the last remaining humid forest area in Mexico; (b) strengthen the institutional capacity of federal offices, state agencies, and municipalities in order to improve project assessment techniques and environmental policy implementation; and (c) finance the restoration of selected archeological sites. The institutional development component would allocate US$18 million of loan funds to finance 100 percent of eligible costs for strengthening: (a) project executing agencies (mainly in the agricultural and roads sectors); (b) state and municipal planning and budgeting systems; and (c) the state and municipal entities to assume additional administrative functions that they have and will be receiving under the Government's program of decentralization. US$15 million of loan funds are as yet unallocated and will be available for filture allocation among the three components described above. The project provides funds for civil works and materials; furniture and equipment; training, studies, and technical assistance. Tne total cost of the project is esdmated at US$1,363 million equivalent, with a foreign exchange component of US$425 million (31 percent). Retroactive financing of up to US$35 million for expenditures incurred after May 1, 1990 will be made available. -A breakdown of costs and the financing plan are shown in Schedule A. The loan would be made to the Government financial agent, Nacional Financiera (NAFIN), and executed by the Secretariat of Programming and Budgeting (SPP), other FeJeral Entities, the Project States, and Participating Municipalities. Loan fimds would be passed on as grants and would only be available for investment costs and technical assistance, but not for recurrent costs or credit. The project would be implemented over a five-year period. B The proposed project would help decrease by an estimated 40 percent the gap in and Risks: access to roads, schools and other basic social goods of the populations in the four poorest states, as well as improve economic opportunities to the mostly indigenous rural populations. Public works generated by the project are estimated at about 230Q000 man-  3 - years. Possible risks arise from the fact that some states and municipalities have relatively weak administrative suctres, and that their geography and social composition are quite varied. Environmental risks are important in some areas with fragile systems. Investment decisions may be subject to political influences, and money could be misappropriated. To mitigate these risks, the project emphasizes strengthening existing safeguards and building new ones into subproject selection, monitoring, evaluation, and audit. Decentralization would be carefully paced. The regional development mechanism used in the project is the development agreement between the Federal Government and each state called the Convenio Unico de Desarrollo (CUD), which has been improved over many years and is designed tn solve interagency conflicts in the design and execution of the investment programs. Institution building components and technical assistance would strengthen administrative and implementation capacities of the states and municipalities as they receive additional functions and revenmes. The environmental component would strengthen environmental proetion capability at the state level. Estimated The estimated costs of the proposed investunent program in the four states are Cost: shown below: L.ocI Foreian Total -------- in USS millioes- Estilated Costs Productive Subprojects 245.3 81.8 3Z?.1 Infrastructure Subdprojects 316.5 173.2 489.7 Sociat Subprojects 171.9 86.5 258.4 Institutional Devlopmnt 14.1 4.7 18.8 EnvirorvOental and Cultural Patrieony 35.8 6.5 42.3 Base Cost RLZ 1136J Physical Contingencies 73.4 34.1 107.5 Price Contingencies 80.9 38.0 118.9 Total Project Cost 937.9 424.8 1368 finnira Plan Government 600.4 171.3 771.7 Beneficiaries 241.0 0.0 241.0 IBRD 70.0 280.0 350.0 lotat 911-4 451 3 21 I1RD Fiscal Year fosixte -9Lv - 521 293 a 95 96 8ank Disburseents ---...............in US$ millions . .-.-- Ar,tl 70.0 100.0 100.0 45.0 27.5 7.5 CujlIative 70.0 170.0 270.0 315.0 342.5 350.0 ... ........ ................... 1. Retroactive financing of USS35.0 million to be provided for alt project expenditures incurred after May 1, 1990. Finaning Elan: The investment program again which Bank disbursements can be claimed is estimated at US$1.4 billion. Bank share of project costs would be 30.8 percen. The remaing financing would be provided by the Government of Mexico in the annual budgets (52.8 percent) and the beneficiaries' contribution (16.4 percent). Rate of Te selection criteria used for screening subprojects assures an economic rate of Retutn: return for productive and large infrastructure projects of at least 12 percent. For social and small infrastructure projects, social and minimum-cost justifications will be provided.  HI. REGIONAL AND RURAL DEVElOPMENTF PQO1U * ~~~~A. Intrgductign 2.01 Between 1950 and 1974, Mexico enjoyed a remarkable period of high growth, low inflation and moderate external debt accumulation. Real growth averaged 6.4 percent, while the poverty rate (percentage of households below the poverty line) declined by 15-20 percentage points. The discovery of large oil reserves led to a public expenditure boom financed largely with external debt. The Government's share in total value added increased by almost one-third, the real exchange .ate appreciated, and Mexico's external debt increased from USS16 billion to USS86 billion between 1975 and 1982. 2.02 The situation changed dramatically in 1982 when rising world interest rates and falling oil prices put an end to Mexico's expansionary policies. The gap between the real interest rates on external debt and real GDP growth went from -6.3 percent in 1980-1981 to a full + 10.5 percent in 1983. A stabilization program reduced the fiscal deficit by one-half, international reserves were rebuilt, and the underlying rate of inflation reduced. However, the earthquake in late 1985 and continuing deterioration in the terms of trade led the authorities to Introduce a major adjusmnent and stabilLation program centered upon trade liberalization and strong fiscal retrenchment. 2.03 A restrictive demand management policy reduced inflation significantly, but public investment was strongly curtailed (from 10 percent of GDP in 1982 to 4.4 percent in 1988) and real minimum wages fell by one-half. There was no cumulative real growth between 1982 and 1988, leading to a severe decline in per capita income, and rough estimates indicate that there was no further improvement in the poverty rate. 2.04 In March 1990, as the first case in debt reduction of the Brady initiative, Mexico and its commercial creditors implemented an agreement to restructure USS52.7 billion of Mexico's external debt that reduced the stock of debt by an estimated 30 percent. In Febmary 1990, the restructuring of the agricultural financial system led to a cut in annual losses of as much as US$1.5 billion, while significantly reducing subsidies to poor farmers. Deregulation and privatization also improved the efficiency of ,.a economy. Through a series of social contracts with labor and entrepreneurs, complemented by anchoring the nominal exchange rate, inflation was kept under control; but the use of price controls led to deterioration in the terms of trade to farmers. Then, in May 1990, the Govermnent increased energy prices, removed some price controls, and announced that negotiations on a bilateral free trade agreement with the United States were being accelerated. 2.05 Stabilization has been successful, with real interest rates dropping from 30 percent in 1987 to about 12 percent most recently. But economic growth slowed somewhat to a two percent annual rate in 1990, reflecting the tightening of aggregate demand mwgement policies. With an increasing primary fiscal surplus, the Government announced that more emphasis would be given to social policies. In fact, throughout the adjustnent period, there was a substantial deterioration in social conditions. Real per capita expenditures in the social sectors were cut by about 7 percent per year, and public investment in rural and regional development dropped by 80 percent to 90 percent in the last decade. Real incomes in rural areas declined about 20 percent between 1983-1988 from levels that were already low.  1. Pove~rtvJn ae_ 1a ynmc 2.06 About 21 million people live below the poverty line in Mexico (currently US$925 equivalent annual household income in urban areas and US$830 oiv4 in rural areas). About two-thirds of the poor live in rural areas; about one-third live in the four project states-Chiapas, Guerrero, Hidalgo, and Oaxaca. The Bank's poverty and regional/rural development studies have found that poverty in these states is structural, but the situation of the poor has been exacerbatot by the economic crisis, particularly by the fall in real wages, the decrease in infrastructure and social investnents, the increase in prices of agricultural inputs, and control of prices of some agricultural outputs. Estimates show a decrease of real income between 9 percent and 16 percent due to the fiscal retrenchment and external terms of trade deterioration. 2.07 Sixty-five percent of the population of the four states (about 10.3 million persons) fill below the poverty line (below one minimum wage). Many lack access to high-quality agicultural resources, and even where agricultural resources are good, productivity is low. Agricitural endowments vary widely among and within the states. Lack of education also correlates strongly with poverty; in rural areas completion rates for primary schooling are only about 30 percent. Settlements are widely scattered and are often poorly served by roads. Communications are generally poor. 2.08 Indigenous populations predominate in poor regions and their cultures are very diverse. Oaxaca has over one million Indians speaking more than a dozen native languages. Indians live in the most adverse microclimatic regions and agroecological zones. Linguistic, cultural, and ecological diversity also characterize Chiapas, Guerrero, and Hidalgo; together the four project states have nearly 40 percent of the indigenous population of Mexico. The higher cost of providing social and infrastructure services to the indigenous communities has put these states at a disadvantage. The Bank project would improve aceess to services by targeting a major share of the investment to poorly served areas. It would also strengthen institutional mechanisms to give indigenous persons better access to public resources. 2. The Role of a Regional Investment Program in an Overall Strateg of Poverty Alleviation 2.09 As elements of a poverty alleviation strategy, trade reform and agricultural policy reform (supported by earlier trade policy loans and by the First Agricultural Sector Adjustment Loan - AGSAL 1) are rapidly reducing urban bias. Financial Sector Reform (supported by the Financial Sector Adjustment Loan - FSAL) and tax reform have virtually eliminated the subsidy of capital-intensive production. Together, these reforms will increase urban and rural employment. Second, equal access among states and microregions to basic social and productive infrastructure is being pursued through this project. Third, the poorest individuals will be targeted directly by enhanced primary health care (supported by the primary health project), and improved nutrition and food subsidy programs (to be supported by AGSAL II). 2.10 Regional investment programs and targeted health, nutrition, and food subsidy programs are not exclusive but complementary. Within the World Bank program, these complimentarities will be exploited by including the four poorest states not only in this project but also in the recently approved Basic Health Care Project, the proposed Primary Education Project, and the pilot Nutrition Program to be supported by AGSAL II.  -6 - 3. Why Focus out the Four Poprest States? 2.11 The Goverunen's investments in poverty alleviation under the National Solidarity Progran. (PNS)' are not confined to these four states, but their choice reflects the desire of the Mexican Government to provide special financial and technial support to these particularly poor and administratively underdeveloped states. 2.12 Why concentrate effort on the poverty states? Why not concentrate investments at growth poles and other areas of high economic potential and wait for economic de Jlopment and out- migration to take care of the poverty regions? Mexico pursued such a strategy between 1950 and 1988, but the high national economic growth rates of that period did litde to alleviate the poverty of many rural people. 2.13 Despite rapid outmigration, high population growth rates have persisted in the marginal areas, and population will continue to increase in absolute numbers. During the last two decades, with extreme urban bias in agricultural policies and social programs, the rural population of the four states rose at an annual rate of about 2.5 percent. These growing populations must be given access to basic social services where they live. 2.14 According to early 1980s da ., about half the population in these four states lacks a primary education; about a third of primary school leavers have no access to secondary schooling. Preschool mortality rates are three to four times higher than in better-off states. At least 40 percent of the population lacks communications and 60 percent to 70 percent lacks acceptable housing. In the richer states, only 5 percent to 10 percent of the population lives under such conditions. By emphasizing social and infrastructure investments, the project would contribute significantly to the improvement of these indicators. 2.15 Regional development policy will continue developing growth poles for industrialization and for 'deconcent.-ation" of the largest congested cities. Growth poles are likely to be located in major urban centers and in the northern regions. Migration from the poorest states to those areas will help reduce poverty in the disadvantaged states. But unless those who leave have acQuired human capital at home, they will swell the number of illiterate and menial workers in the capital and in the north. The educational system must be able to provide that human capital. Health, education, and nutrition are the foundations for human capital whether the workers remain in their home regions or migrate for job opportunities. 2.16 While social services and the accompanying infrastructure should be provided on an equity basis, productive investments will have to be concentrated in areas with a high potential for economic success and a comparative advantage, e.g., agriculture and forestry, mining, tourism, and cottage industry. This, of course, applies to potential growth poles within the four states. In some disadvantaged ecological zones, where the ec3nomic potential is severely limited, it makes little sense to force investments beyond the few high payoff opportnities that almost always exis4.. - Projects that do not produce a sufficiendy high return cannot alleviate poverty and are quicldy abandoned by their beneficiaries after the initial phase of investment. Selection criteria for production-oriented investment projects will therefore screen out projects below a minimum level of efficiency. 1. Soo disuion below, Saction D.  -7 - 4. Decentralization 2.17 Decentralization is an instrument for improving efficiency in project identification and operation. An integrated strategy cannot be effectively coordinated in such wide and diversified regions unless the populations are given control of their own resources to solve local problems. Some functions, like technical norms and accountability, have to be exercised at the federal level. Others, like expenditure control for statewide health and education, should be the responsibility of state governments. Still others, like maintenance of local primary schools or rural roads, are better performed by municipalities or local communities, provided that these are above the minimum size necessary to conduct these funotions. 2.18 Mexico is a federation of 32 states with about 3,000 municipalities. The sweeping constitutional reform of 1983 gave municipalities responsibility for administering water supply and sewer systems, public lighting, municipal markets, security, streets, cemeteries, public gardens, and slaughterhouses.' While some institutionally stronger municipalities have assumed all these responsibilities, others are not yet capable of doing so (see Annex 10). The project, through its institutional development component, would enable significantly more municipalities to assume those functions. For the remainder, state agencies would have to continue providing these services, or communities would have to continue to forego them. 2.19 In the last seven years, important steps have been taken toward administrative decentralization. A decentralized planning system has been adopted, one that would be the backbone of the present project (see paras. 2.38 to 2.45). In addition to state planning agencies, a development committee (COPLADE), headed by the state governor, has been created for each state to coordinate planning by state and federal line agencies. These COPLADEs effectively control the CUDs between the central government and each state. Although the execution of the majority of large projects is still the responsibility of the federal line agencies, through the Programa Normal,3 the Government hopes to accelerate the process of transferring responsibility for projects with local or state impact from the Programa Normal to the CUD. 2.20 Decentralization of health services is already under way in about 12 states. It has been completed successfully in Guerrero and has significantly improved the primary health care system.' Decentralization of education began last year and is an explicit objective of the Programa de Modernizacion Educativa of the present administration. Construction and maintenance of state (feeder) roads is now undertaken by state agencies, as is most of the construction and rehabilitation of rural roads. States and municipalities are assuming ever-increasing responsibilities in mainenance of infrastructure due to the financial retrenchmnent of the central government. The proiect would help rationalize and accelerate administrative decentralization, concentrating on the priority sectors: agriculture extension services and rural roads. 2.21 Federal officials are sometimes reluctant to transfer functions to their state counterparts because their administrative capacity is weak, but the latter cannot mature except through 2. Aiticle 1 15 of the Contittion, Regulatoiy Deree, article m. 3. Ima Nod oI th exclusively federally controlled and funded invetmt pgram. 4. Cimple dcemirlization of the hea qstem is axeced to taplce p in th ote the das in the next three yeem, sppoied by the Bsic Hallth Loan.  experience. To break this stalemate. the project includes dentralization timetables and a major institutional development component to assist the four states in acquiring the necessary capacity. 2.22 Expenditure control. In federated industrialized countries, state and local governments usually control 45 percent to 65 percent of public expenditures, but in Mexico they control only 20 percent. In th, four poorest states, the federal and/or autonomous nationwide agencies control about 80 percent of current expenditures, while the states control about 1S percent and municipalities five percent; 75 percent of the investment trogram remained under central government control in 1988. In 1989 investments through the CUDs increased to about half of the federally funded investment. Moreover, a significant share of central government investment is being coordinated with state governments (in the four states, approximately two-thirds). The C;overnment expects to continue shifting control of public investment to the state and municipal go imunents. 2.23 Revenue raising. For efficient administration, states and municipalities should mobilize sufficient local resources to cover their expenditures. nTat is not the case for Mexico where, since the 1980 fiscal reform, states and municipalities have had little tax authority. The poor states remain dependent on federal grants for about 70 percent to 90 percent of their total budgets.! Most resources come through the revenue-sharing system as block grants. While the Salinas government is giving top priority to improving tax administration, it does not plan to transfer significant revenue authority to the states in the near future. States and municipalities have started to increase collections from the few local taxes they do control, e.g., property taxes and water charges, and will continue to do so (see Annex 3). However, the potential to generate revenues from these sources is very limited. The reform of the revenue-sharing formula introduced in 1990 would increase substantially the block grant available to finance the state share of investmnents and recurrent expenditures for social services in these states. In addition. when functions are decentralized, the transfer of additional resources has to be effecte through selective block giants specified in the s e , as it was implemented in the health sector. S. Equity in Access to Public Resources 2.24 Public resources are not allocated equitably across states and municipalities. Until January 1990, the formula for reenue sharing (block grants) among states had no equity elements and merely reflected revenues from the former state sales tax (replaced in 1980 by the federal Value Added Tax). As a result, the Federal District and the State of Mexico, with 25 percent of the population of the country, received 36 percent of the Fondos de Participacion, while the four poorest states, with 12 percent of the population, received only seven percent. 2.25 In addition, most current ex2enditures in the states (about half of regionalized state expenditures are channeled through the Programa Normal), are skewed toward the better-off jurisdictions. 2.26 Finally, er caita pubic investment is highly inequitable. Of the total regionalized public investment, the four project states received only 4.8 percent. In the 1985-88 period, Mexico invested, on average, only US$77 per capita/per year in the four states, compared with US$118 per capita nationally. However, the Salinas government's new emphasis on equity and poverty alleviation is changing the distribution pattern. Revision of the revenue-sharing system, introduced in November 5. In the poores jurisdictions that dependence is as high as 95 percent.  -9- 19E9, will increase the share for the poorest states by 20 percent to 30 percent in the next five years. Investment funds for the poorest swes are now growing faster than the average increase in the country. - B. InaQrporating LessoN from Experielne 2.27 Mexico has long and wide-ranging experience with rural and regional development programs. Bank assistance in rural and regional development began in 1975 with the first Integrated Rural Development (PIDER) project. Two additional PIDER projects and the more agriculturally- oriented National Assistance Plan for Rainfed Areas (PLANAT) and Program for the Development of Humid Tropics (PRODERITH) projects provided additional experience. These projects fostered improvements in administrative structures, mechanisms, and procedures at the federal, state, and municipal levels that today are the backbone of federal-state investment planning and financing. For example, the original PIDER operating manual has been continuously revised and updated. A cadre of development professionals and administrators have lived through successful and unsuccessful programs and have worked at the field level, as administrators of the projects and programs and as evaluators. At the close of the PIDER III project in 1988, Mexico and the Bank conducted in-depth reviews of this entire experience. Mexico surveyed PIDER MI subprojects (the survey of the National Institute of Statistics, Geography and Information - INEGI) and commissioned a strategy report (Estrategias para Areas MarLinadas v su Instrumentacion . The Bank assessed the entire regional and rural development experience (Mexico: Strategy Proposal for Rural/Regional Development in the Disadvantaged States, Report 7746-ME). Project completion reports of PIDER, PRODERITH, and PLANAT provided additional insights. All these studies were largely completed when the Salinas administration took office two years ago. 2.28 Many of the recommendations that emerged from these studies were incorporated in the design of the PNS and the regional development program, components of which this project would support. The following sections describe some of the key lessons learned in the review process and applied during preparation of this project. 2.29 The rate of satisfaction with small social and infrastructure projects is high. The INEGI survey of PIDER III found that about 65 percent of subprojects were judged successful by the beneficiaries, with social and infrastructure projects rated above 70 percent. Experience with productive projecs, on the other hand, is mixed and shows clearly that the public sector does not have comparative advantage in such areas. For example, agroclimatic conditions were inappropriate for some small tree crop projects, and marketing arrangements were undeveloped or technical assistance was lacking for some small irrigation projects. Therefore, while the PIDER projects unsuccessfully mandated that at least 61 percent of the projects financed should be productive projects, the proposed project sets up no such rigid objective. About 25 percent of the overall state investment program and about 26 percent of the programs designed by the communities, municipalities, and states are productive projects. Many of the public investments will, of course, support private productive investments and activities. 2.30 The PIDER III Project lost its poverty focus. Execution of PIDER m coincided with the economic crisis that began in 1982. Investment budgets for regional and rural development were progressively slashed. The remaining investment resources were spread thinly among many communities and invested in highly visible, low-cost projects, such as development of community stores. By focusing supervision narrowly on the subprojects, neither the Secretariat cf Programming and Budget (SPP) nor the Bank had the means to evaluate the overall poverty impact of investments  on poor communities, municipalities, and stat. To increase the poverty focus of regional development, four changes have been or would be made under the present project: (a) formulas for revenue-sharing have been given greater equity weight (para. 2.36); (b) SPP and the Bank would now monitor the poverty focus of each participating state's entire investment program rather than just that of the subprojects (a major shift for both Mexico and for the Bank (Par Il, Section C.2.(a)); (c) a revised project appraisal system introduces a poverty weight into the scoring system of each project (Part m1, Section C.2.(b)); and (d) grant eligibility rules and cost recovery requirements would be more stringent for richer communities (Part Il, Section C.2.(d)). 2.31 Significant poverty impact is not feasible without greatly enhacd resource levels. The reviews show clearly that projects such as PIDER m, which spread limited resources over a vast number of beneficiaries (US$43 per beneficiary per year) have little impact on communities in which they are carried out. In contrast, more intensive projects, such as the PRODERITH and PLANAT projects that spent between US$1,000 and US$2,000 per beneficiary per year, achieved greater and more lasting impact. 2.32 Between 1982 and 1988, public investment in Mexico declined by 60 percent. Cuts in rural, regional, and social sectors were even deeper, reaching about 70 percent, leading to the collapse of the investment programs for poverty alleviation in the four states where the core investment program decreased from about US$1 billion in 1985' to US$240 million in 1989. 2.33 In view of the acute basic needs in the four project states, the investment program for poverty alleviation urgently needs expanding. The core investment program to be supported by the project increased by 20 percent in real terms between 1989 and 1990 and is projected to increase by another 20 percent in the next two years. Although the expansion would be rapid, it starts from an extremely low base and by 1992 would still amount to only 66 percent of the 1985 level. 2.34 Creating parallel administrative structures leads to unsustainable pro=am=s. By their very nature, regional and rural development programs involve integrating components from different sectoral agencies and different levels of government. They tend, therefore, to spawn parallel administrative structures or executing agencies to carry out the programs. While this may enhance coordination in the short run, it becomes disruptive when changes in administration or economic circumstances occur. The most glaring example in Mexico was the collapse of the General Committee of the National Plan for Depressed Areas and Marginal Groups (COPLAMAR), administered out of the President's office during the Lopez Portillo administration, but quickly abandoned when the new administration took office in 1982. PIDER Im, executed by the regional development subsecretariat of the Ministry of Planning and Budget, was the first project that became fully integrated into the regular regional planing system of the Mexican Government. Because of the experience with COPLAMAR, the PNS of the Salinas administration is fully integrated into the same regional planning system (described in the following section). During preparation of the proposed project, further improvements were made to the regional planning system, and neither the project nor PNS created any parallel structures. 2.35 Community participation reauires the transfer of real responsibility and authority to lower-level jurisdictions and communitiea. All past programs proposed the goal of community participation and carried out numerous microregion studies. However, highly centralized decision- making, administrative procedures, and controls ovet spending, in fact, acted as barriers to community participation in subproject selection, execution, and operation. Execution was often 6. The first year for which data is available on a regionalized foQwt.  entrusted to federal line agencies operating from Mexico City, further increasing coordination proMems. Achievng genuine community participation requires a shift of authority over subproject selecion, execution, spending, and operation and maintenance to lowerlevel jurisdictions and communities themselves. Reforms already carried out to accomplish this goal are discussed in the following sections. C. Recent Policy Reforms 2.36 Changes in the revenue-sharing formula introduced in November 1989 during project preparation represent a significant improvement in equity (Annex 2 describes the old and new systems). Starting in 1990, about half the revenue is to be allocated on per capita basis and the other half on the basis of historical shares that change according to the rate of excises, property taxes, and water charges collected in each state. The per capita component introduces an acceptable equity criterion. Projections show that the four poorest sxates will increase their share portion of shared federal revenues between 20 percent and 30 percent, thus increasing substantially their capacity for cost-sharing in investment programs and for covering recurrent costs in the social sectors.7 2.37 The tables in Attachment 3 of Annex 3 illustrate the administration's major reorientation of inveatment effort towards poverty alleviation. Nationwide, investments in social, rural, and regional programs will increase at a faster pace than overall investment; by 1990 their growth rate was already almost double that of total investment. Within the core program, the four poorest states will increase their share of investment in the national total from 10 percent in 1989 to 14 percent in 1992. Within the overall investment program of the four states, the share for poverty alleviation will increase from 38 percent in 1989 to 48 percent by 1992. These resources are starting to become available, making a significant impact feasible. 1. Improving the Refional Planning System 2.38 Mexico has a very elaborate regional planning system based on the COPLADEs, the state coordinating agencies that were created as a consequence of the PIDER projects (1975-1984). Funds for investment are allocated to the states through the CUDs while fiscal resources for r pnditure are mostly allocated through revenue-sharing arrangements. 2.39 The formulation of state p!ans is a programming exercise for pbic investmen. I starts by the middle of the preceding year, when the COPLADE initiates the process through its various sectoral committees (education, health, agriculture, infrastructure, etc.) with representatives of all federal and state sectoral institutions.! The sectoral committee proposals are based on subregional diagnostics and reflect priorities expressed in the state development plans. They also take into account national sectoral programs, such as the Plan of Educational Modernization. The COPLADE then formulates a first proposal. By December, SPP sends to all governors an indicadon of national 7. Revene shading repres 18 percent of total federal taxes and between 60 pect nd 80 percent of toal mo financing. S. As dcwribed in Section m, below, there ws wide popular participation, involving all municipalities an condmuni. in the prpation of this project, the fust experience of its kind in Maxico. Each municipality wa asked to idMn* a priority list of "proj, ideas' and a ranking syaem was used for ordering thoft projet b_ on quick baeecost analysis.  - 12 - priorities, defines the aggregate allocation to the regional budget, and specifies what rules have to be followed in its final formulation. DistribWion by state is based on last year allocation, demonstrated implementation capacity and an overall needs assessment. Sectoral programs must be reviewed and approved by the line secretariats. While this process is going on, the various sectoral agencies prepare appraisals of the subprojects. Finally, the SPP and the states negotiate overall investment allocaions, specific programs, and sources of financing. By February-March most of the development agreements have been concluded. These CUD contracts between the federal and state goverrnents specify the legal framework under which the matching grants and the investment program will be administered. All investments to be carried out are included in the Programa Operativo Atual (POA), which is an annex to the CUD. 2.40 In order for the POA to be approved, each investment must be supported by a finalized appraisal report by the responsible sectoral agency proving that the investment meet the appraisal criteria. The appraisal must follow established SPP guidelines (Guias para la Elaboracion de Expedientes Tecnicos) and must contain a basic project description, evaluation of beneficiaries, a budget and financing plan, a technical study, a feasibility and engineering study (with confirmation by the federal line agency that it satisfies all technical norms and that such agency would supervise the works and give technical assistance for its execution), an economic evaluation, an agreement between the community and the implementing agency (through which the community guarantees cofinancing in the fbrm of either land, labor, materials, or monetary contributions), and an environmental impact assessment. Sectoral guidelines establish a basic appraisal format for each type of project. The appraisals are reviewed by the federal line agency and then submitted by the COPLADEs for approval by the SPP state delegate. The environmental impact assessment is done by the state delegation of the Secretariat of Urban Development and Environment (SEDUE). The appraisal mission reviewed a sample of appraisal reports and found them satisfactory. 2.41 The budgetary control of the CUDs is exercised through category XXVI (Ramo XXVI) of the Mexican Public Administration Budget. The disbursement of funds to each state is made through the correspondent banks of the Banco de Mexico, and is subject to quarterly statements on physical and financial status of the investment programs by the states. Rules set down in the Manual Unico de Operacion require that implementing agencies maintain proper accounting of all programs and subprojects. The release of funds is controlled by means of accounts-payable certificates (CLCs) that must be reviewed and approved by the COPLADE Coordinator and the SPP state delegate. The procurement procedure for each project is specified at the time of programming; it may be by direct administration, contracting through local or international competitive bidding, or under agreement with the community.' Under the Cuernavaca agreement on procurement concluded in April 1990, those rules generally conform to Bank guidelines. Sections on disbursement and procurement lay down rules for operating the project (paras. 3.98-3.106). 2.42 The present monitoring and evaluation system has its origins in the PIDERs and is broadly satisfactory in terms of physical and financial implementation. Financial control is further exercised through public accounting and auditing systems of the General Secretariat of Comptrollers of the Federation (SECOGEF). An overall agreement between SECOGEF and the Bank establishes the rules for auditing of Bank projects. Subproject monitoring is based on timetables in the appraisal report that establish targets for physical and financial implementation. Monitoring tools also include all supporting documents as well as the project authorizations issued by federal agencies that enforce technical and environmental rules. States send quarterly reports to SPP on the status of project execution. State and federal line agencies also monitor the subprojects and raise with SPP any 9. Thase procedures an reguleed by the Ley de Obras PubL;as.  - 13 - problems that may occur in project opsertion. The projt would introduce an improved system for analyzing economic and social impact and for processing information for policy decisions (see Annex 12). 2.43 In the course of preparing the project, significant improvements in the planning process were introduced. First, detailed diagnostics were undertaken at the state and regional levels."0 Second, strategic state and regional five-year plans were formulated through an interactive process. Third, communities' needs and municipal investment priorities were elicited through a participatory process involving the Secretaria de Planeacion Estatal and every municipality. Fourth, the program was screened on the basis of a new project scoring system designed for the states by SPP and the Bank (Annex 4). Fifth, sectoral agencies in each of the states were involved in selecting projects, studying the feasibility of communities' proposals, and proposing additional programs or projects with interregional and statewide impact. Sixth, the planning mechanism was further decentralized when SPP-federal shifted its focus from project analysis to program design, while the screening of appraisal reports is carried out in SPP state delegations. Seventh, the role of COPLADEs was enlarged with formulation of the POA that encompasses resources flowing through the Programa Normal of the federal agencies, the Programa de Solidaridad Estatal (PSE), and the state investment programs." This, for the first time, will permit monitoring the poverty impact of the entire investmnent program of each state. 2.44 In the past, serious inefficiency resulted when most federal government funds were disbursed only at the start of the rainy season. This problem has affected all Bank investment projects. In the course of project preparation, new directives were issued by SPP ensuring a smoother flow of funds to the states. All work in progress is now automatically reprogrammed into the next fiscal year (oficio de secas) and thus is not interrupted because of a need to wait for the new budget to be released; the closing of the fiscal year was extended to March. 2.45 Another serious inefficiency resulted from processing expenditure authorizations at SPP in Mexico City. To ensure a smoother flow of funds, the CUD system now requires only that expenditures be verified by the SPP delegate in the state. Once approved, each program is immediately funded by SPP through a revolving fund amounting to 25 percent of the total cost. After expenditures have been verified by Statements of Expenditures (SOEs), the fund is automatically replenished. These reforms represent important steps toward improving project performance (see Policy Letter, Annex 1, page 2). 2. Incrasing Flexibility in Resource Channeling Municipal Solidarity Funds 2.46 To strengthen communities' social demands, emphasize beneficiary participation, and introduce competition among agents that provide services to those communities, two additional channels of resources have been created: (a) the Municipal Solidarity Funds (FMS); and (b) the Development Funds for Indigenous Populations (discussed in the next section). In 1990 about 1,400 poor municipalities have benefitted from the FMS, whose funds will total about US$120 million. For 1991, the FMS will be extended to 2,000 municipalities with the doubling of the budget. These funds 10. 1r are 31 planning regions in he four maw. 11. Progranu Nonnal Estatal.  - 14 - are accounts created at the municipal and community levels to provide capital grant for small community projets with budget fiaming. States and municipalities have been selected on the basis of their willingness to participate and specific povert indices. The fiour poorest states would receive about US$40 million. Te project would support this component through the financing of investment programs; thus the creation of a special program of the social fund type is not warranted. 2.47 The FMS support microprojects for social and community infrastructure (with emphasis on rehabilitation and maintenance), and microprograms to meet basic and social needs of the community and to enhance employment. No individual project or program can exceed US$25,000. The FMS would also finance technical assistance for implementing such programs. The technical assistance would be provided by state or line agencies, nongovernment organizations (NGOs) or private firms. On average, each municipality received US$100,000 equivalent in 1990 and the double is planned for 1991. The FMS are replenihed as the money is spent, rewarding successfil implementation and satisfactory annual auditinglmonitoring/evaluation. This intoduces a mechanism for competition among municipalities. In 1990 about 15,000 projects implemented through the FMS have been completed. 2.48 The FMS aim at furthering the process of decentralization to municipalities and encouraging local communities to participate actively in the identification, design, execution, operation, and evaluation of government programs. The basic operational mechanisms are: (a) setting up of the FMS by PNS,3 through agreement wIth the states and municipalities; (b) establishment and operation of municipal councils within the organic municipal law; (c) creation of local committees to participate fully in the programs funded by the FMS; and (d) coordination with organized social groups that agree to participate in identifying, executing, maintaining, and operating infrastructure and social programs. 2.49 States are requested to participate with 25 percent in the funding of the FMS. Beneficiaries are required to contribute at least 20 percent of the total cost of individual projects, which may be paid in kind with labor or materials. Project standards and operating procedures are flexible enough to take into account existing municipal capacity and to help strengthen the planning and administrative capacity of the weaker municipalities. FMS projects would be selected according to policy priorities, efficiency, impact, and participation criteria developed by SPP, reviewed and approved by the Bank, and already implemented in 1990. eeloEmn Funds of IndigeouM L nm 2.50 Indigenous communities present special problems because of their cultural traits and their experience with discrimination. Indian communities should benefit from investment programs like all other population groups in each municipality. However, to avoid possible exploitation of these programs by cacigues and other traditional power brokers, special funds have been created in the last 10 years targeted to the Indian communities and handled by the National Institute of Indian Affairs (INI), which maintains a network of specialists working with native communities. 2.51 Development Funds for Indigenous Populations have been strengthened and enlarged, following rules similar to those governing the FMS. They provide recoverable capital to Indian communities, landholding groups (ejidos), cooperatives, and civil associations for local production and development activities. Each fund is administered by a board of directors composed of representatives of the participating indigenous communities and organizations. The regional INI 12. See discussion of PNS in Section D, below.  - ~ ~~~ lS- cordinating center and other line agencies cooperating with INI wilt provide technic assistance. For FY90, SPP has committed Mex$500 million (approximately US$185,000) per development fund; there will be 44 funds administered in the four project states and a total of 89 funds throughout the country, amounting to about USS8 million. The norms agreed upon between INI and the SPP allow for about 10 percent of these funds to be used for technical assistance and 5 percent for project design support; the remainder would be used to finance capital grants for social and productive projects. 2.52 INI expects .it eventually each Development Funds for lndigenous Populations would be able to meet its own technical assistance and institutional support needs without Govenmment help. During the first three to four years, however, INI will need to provide strong technical and institutional support to the Funds. For this reason, the project includes a technical assistance and institutional support program for those Funds. DugrUg V2", INI agreed that specific investment plans proposed for indigenous populations would be submitted to the COPLADEs in the states and included in their annual investment plans. To ensure provision of resources, INI is represented at the state level in specific sectoral subcommittees as well as in the COPLADEs. 3. Adninstrative Dexentralization 2.53 lThe aim of Mexico's decentralization program is to strengthen and speed up the process of decentralizing administrative functions from the federal to the state and municipal levels, consistent with implementation capacity. For all ministries, quality control and coordination are to remain responsibilities of the federal administration. Priority areas for decentralization are education, health, rural roads, agriculture services, and water supply systems. To prepare the decentralization program, a working gr%up at the federal level, with representatives from all ministries and led by SPP, carried out a detailed analysis of 14 main administrative functions. The agreements reached are contained in the Decentralization Matrix presented as an attachment to the Policy Letter.Y The agreements, to be monitored under this project, provide the basis for the project's instiutional d-welopment component. The operational details of the agreements, as well as their corresponding budgetary implications, have already been worked out between SPP, the ministry in charge of coordinating public administration, and federal line agencies. The following decisions have been made: o Agricultural Sectr. Parastatal agroindustries, livestock, marketing, and storage firms are being transferred from the Secretariat of Agriculture and Water Resources (SARH) to producer organizations and the private sector, and the privatization program will be monitored under the proposed AGSAL II. Most of the services of the former Mexican Coffee Institute (INMECAFE) and fruit agency (CONAFRUT) have been transferred to the association of coffee producers. Operation and maintenance of water wells are being transferred from the Trust Fund for Risk-Sharing (FIRCO) to producer organizations. Centers for livestock extension and marketing organizations of livestock products will be decentralized to producer organizations. Control of forest fires and natural resource management will be transferred from the Integrated Rural Development District (DDRI) to state governments, and administrative control of producer organizations will be transferred to municipalities. 13. See attachmns to Poicy Leter in An=ex 1.  16 o SItatend _rl Ro_0 Feeder road responsibilities have been decentralized successfully to state road agenicies in all project states except Hidalgo. In 1991 construction, rehabilitaLioui, and maintenance of swe roads will be cozupletell decentralized, witl the Secretariat of Communications and Transport (SCT) maintaining a supervisory role. In 1992 construction and rehabilitation of rural roads will be decentralized from the 'rust Fund for the National Commission of Feeder Roads and Highways (CONACAL) to state agencies, with the former retaining a supervisory and technical assistance role. Decentralization of rural road maintenance began in 1990, with Chiapas assuming responsibility for about 1,200 km supported by a rehabilitation program financed under the CUD. Rural roads being built under the project will be transferred to municipalities and their maintenance financed under the FMS. Transfer of the rural road network to the states will be completed by 1992, and in 1993 routine maintenance wilt start to devolve to municipalities. In addition the following actions are being pursued under separate Bank projects: o Educatig. Management and operation of the primary and secondary school systems within each state are already being transferred to the state secretariats and the process will continue, monitored under the proposed Primary Education Loan. o Healt. The health system has already been decentralized to the state of Guerrero with good results and the process will be completed in Chiapas, Hidalgo, and Oaxaca, monitored -under the Basic Health Project (Report No. 8927-ME). O Water Supply. Construction, operation, and maintenance of water systems have been decentralized to state water companies. This process will be monitored and strengthened under the Water Supply and Sanitary Sector Project (Loan No. 3271-ME). 2.54 Within the four states, municipalities have been ranked according to their levels of development. An all-agency working group has drawn up a program for strengthening state agencies and municipalities. During the course of the project, functions would be transferred from states to municipalities, in accordance with the constitution and laws, as local conditions and capacity allow. For functions that require a minimum size to have sufficient technical and financial resources for efficient operation, subregional centers would be established linking groups of municipalities. These subregional centers already exist for road construction and agricultural extension; additional functions would include technical and administrative support for municipalities, ope.ation and major maintenance of water systems, and major maintenance of state and rural roads. D. Government Strategy 2.55 The Salinas administration has accorded high priority to reducing poverty. Its debt reduction and public sector restructuring policies have created conditions for resuming economic growth. Its measures to reduce protection, restore an open exchange rate regime, and allow agricultural prices to respond to international markets have significantly reduced the economy's urban bias. Tax and interest policy reforms have sharply reduced subsidies to capital-intensive techniques. More labor-intensive growth should result. The administration significantly increased public expenditures in the social sectors in 1990 and sought the support of the Bank to improve efficiency in those sectors.  - 17 2.56 Tli administration has assigned a centrai policy and implementation rotle in poverty alleviation programs to the PNS. an operating agency of SPP. All mninistries in the prodiuctive and social areas participate in a steering committee. Aided by state and municipal governmenv:, the program seeks to channel health, education, nutrition, housing, employment, and productive resources to the poorest groups and introduce a more equitable expenditure policy strontgly favorable to the poor. In rural areas, the program focuses on indigenous populations and the poorer agricultural zones, and in urban areas it concentrates on the urban fringes. PNS is a nationwide program. The proposed project is not supporting the PNS program in its entitety. Inctead, it finances selected investments in the four poorest states. These investments are also not part of the overall PNS program. PNS also supports other programs, such as the rural stores which are excluded from the proposed project. As in most of PNS programs, a large part of the subprojects under this proposed loan are demand driven. However, in order to be eligible for Bank financing, subprojects would have to conform to the conditionality described in this staff appraisal report. 2.57 In close collaboration with the Bank, the Salinas administration has initiated policy reforms intended to improve the situation of marginal groups. The Baik report Strategy Pronosal for the Development of the Disadvantaged States (Report No. 7786-ME) has been influential in bringing about these reforms. The new strategy, pursued largely through the PNS, aims at channeling social and infrastructure public expenditures more directly to marginal populations. Under the present project, the Bank would assist the Government in carrving out these policy reforms and institutional changes, and in implementing the investment programs against poverty, with special emphasis on the four poorest states. 2.58 The Government has already substantially increased investment in poverty alleviation programs, reversing the tertdency of the last decade. PNS resources were increased from US$400 million in 1989 to aimost US$2 billion in 1990. A significant part of the increase was slated for the social sectors and poverty alleviation. The Government has accorded high priority to improving conditions of the rural population through the Programa de Modernizacion del Campo. The proposed AGSAL II will support elimination of remaining anti-agricultural distortions and expansion, and better targeting of food subsidies, including a nutrition component in the poorest states. Meanwhile, the Basic Helth Caue Projct (Report No. 8927-ME) and a T.roposed Primary Education Project support reform and improvement in those subsectors. 2.59 The Government strategy for decentralization has been defined in the Plan Nacional de Desarrollo 1989-1994. In each of the main administrative subsectors, the Salinas government has continued the policy initiated by the previous administration in decentralizing state and local administrative functions. Supported by the Bank under the Financial Sector Adjustment Loan (Loan No. 3085-ME), further decentralization of spending has taken place, but fiscal decentralization has been subordinated to the more pressing aim of maintaining a tight tax administration for purposes of macroeconomic stabilization. E. Rationale for Bank Invol&gment 2.60 While the Bank has supported rural development, water supply, and low-income housing in Mexico for a long time, a broad dialogue on poverty alleviation has only emerged since late 1988. The strategy advocated by the Bank is fully consistent with the 1990 World Development report: (a) sound macroeconomic management as a basis for restoring growth; (b) elimination of anti- emplovinent biases in trade policies (as supported by Trade Policy Loans - TPL I and II), interest policies (supported by General Interest Rat- Agreement - GIRA), ad tax policies; (C) elimination of  - 18 - urban bias in agricultural, food, and hutrition policies (as supported by the agricultural sector adjustmcnt loans, AGSAL 1, and proposed AGSAL 11); and (d) better targeting of public expenditures and greater effort in human resource development. 2.61 Within this overall strategy, and at the request of the Government, the Bank will concentrate regional development and social sector projects heavily on the four poorest states. (Concentration on these four states does not, of course, imply that poverty is confined to these states or that the Government is not attending to poverty issues elsewhere.) The Basic Health Project will strengthen the Public Social Assistance Program (PASSPA) for delivering primary health care and nutrition programs in the four states and in the Federal District. The planned Primary Education Project will cover the four states. The Agricultural Extension and Applied Research Project (PROCATI) will add rural development districts in each of the four states to those it covers in other states. The Chiapas Rural Development Project will continue developing regions in Chiapas with relatively high agricultural potential. And the four states will participate in nationwide sectral investment programs supported by the Water Supply Project ard the proposed Irrigation Sector Loan. The regional development project provides an umbrella for these actions by strengthening state capacity to plan, finance, execute and monitor the state-wide investment efforts. The concentration of action of a number of World Bank projects in the same states under the same strategy will enable the components oi the development program to be supervised and monitored consistently by the appropriate sector specialists of the Bank. This greatly reduces complexity of each supervisory task and adds flexibility not previously feasible in integrated rural development projects. However, the synergistic impact of multiple components which earlier rural development projects hoped to achieve can still come about. 2.62 Within this concentrated effort on the four poor states, the specific rationale for the proposed project is that: (a) the Bank can build on past experiences with regional and rural development in Mexico; (b) in addition to providing investment resources for the poor states, the project will allow the Bank to monitor the administrative and fiscal decentralization process; (c) Bank involvement would maintain a poverty focus for the total investment program of each state and ensure better targeting to Indian communities; and (d) Bank involvement would create better conditions for balanced and sustainable development through preservadon of the environment and the cultural patrimony.  - 19 - A. Origin of the Project 3.01 At. > close of the PIDER III project in 1987, the Mexican Government asked the Bank to continue its involvement in regional and rural development with special emphasis on the poorest states. The Bank insisted that, prior to any further participation, both government and the Bank needed to review past experience. These reviews lasted from July 1988 to about May 1989. Incorporating lessons of the r.views, the incoming government launched PNS as a poverty alleviation program and implemented marny changes in its regional development programs, as described in the previous chapter. Preparation of the present project started with a workshop held in Oaxaca in July 1989, during which the reviews and ensuing strategy were discussed. Three more project preparation workshops were held, one in each of the other three states. Appraisal of the project was concluded in May 1990. 3,. PoectW ObWectir 3.02 The overall objective of the project is to increase the access of poor and indigenous populations in the four poorest states of Mexico to basic infrastructure, social services, markets, and technologies. The specific project objectives are: (a) to assist these states in developing, executing, and financing investment programs with a strong poverty alleviation focus; (b) to assist the federal and state planning and budgeting ministries in making project selection, environmental assessment, monitoring, and evaluation more efficient and financing more flexible; and (c) to enable the decentralized state and municipal institutions to identify, prepare, build, operate, and maintain the investments in a more participatory manner. 3.03 To achieve these objectives the project would: (a) build poverty weights into revenue- sharing formulas, project selection criteria, grant eligibility rules, cost recovery requirements, and monitoring criteria for state investment programs; (b) support the gradual decentralization process already initiated by the Mexican Government; and (c) improve the sustainability of investment through increased financing of operation and maintenance expenditures from revenue generation by the states, from enhanced federal and state budget allocations, and from beneficiary contributions in cash and kind. 3.04 The strategy under the proposed project assumes that: (a) basic needs in poor regions will not be met in the next 20-30 years without significantly increasing investments in infrastructure and social services; (b) the required investment is well within the financial capability of Mexico, despite budgetary constraints imposed by the present stabilization program; (c) regional development programs shusu ,uport and strengthen fiscal, financial, and administrative decentralization; and (d) the proposed state inN.- ftment programs already have a strong commitment from state governments, use state plans to promote equity, and incorporate strong community and beneficiary participation.  20 - C. PMn%jt e tsciition and CPwj"uo 3.05 The project would address three areas: (a) Inystment. It would finance a part of the investment programs for poverty alleviation in the tour poorest states, based on regional and sectoral deve'-ment strategies; (b) Environmental and cultural protection. It would help preserve the environmental and cultural patrimony in these states; and (c) Technical assistance. It would strengthen state and municipal institutions to assume additional responsibilities through decentralization; increase local fiscal effort and improve the revenue-sharing system between states and municipalities; and support more decentralized and flexible funding mechanisms through Municipal and Indian Development funds. 3.06 The project will finance an investment program for poverty alleviation in the four states, to be coordinated by a single ministry: Secretary of Programming and Budgeting (SPP), although it will be implemented by several line ministries operating in the four states, state institutions and municipalities. One single administrative mechanism would be used for planning, selecting, and controlling the execution of investment subprojects: the Development Agreements (CUD) between federal and state govermments. Through this mechanism of regional planning and budget control, an annual investment program is agreed, sources of financing committed, and implementing agencies defined. Operational Manuals regulate the flow of funds, the selection and execution processes as well as procurement, disbursement and auditing. An extension of the cur '-as been created, with a special Operating Manual, for operating Municipal Solidarity Funds that explicitly recognize community participation and flexibility in planning and budgeting procedures. The project would finance subprojects of the investment programs agreed under the CUD for the four poorest states that satisfy eligibility criteria defined below. 3 )7 The investment component of the project (US$274 million) would help finance investment subprojects with a major impact on poverty alleviation. These are part of the core Investment program defined by the following categories of investment: agricultural promotion and extension - section 3F1 (agricultural promotion and extension including forestation, plant nursery, soil conservation, watershed protection, drainage, horticulture, animal husbandry, plant and animal disease control and coffee plantation rehabilitation); irrigation - section EA (construction and rehabilitation of small scale irrigation schemes); agroindustry - section FD (processing of fruits, vegetables, fisheries and produce); education - section SF (construction and rehabilitation of public primary and secondary schools); rural roads - section EK (construction, rehabilitation, modernization and extension of rural, state and feeder roads and bridges; and maintenance of such roads); water supply and sanitation - section EC/ED (construction, extension and rehabilitation of water supply, drainage and sewerage systems and water treatnent plants); and rural electrification - section EU (construction of electricity supply networks and transmission lines, installation of transmission lines, and connection of rural communities to the electricity supply network). 14. Section numbers refer l thc CUD Operational Manuel.  -21 3.08 The enviroament and cultural site protection (US$43 million) would support: (a) protection of natural reserve areas including the preparation of management plans, upgrading of patrol services, and the carrying oult tif the Lacandona Protection Plan; (b) strengthening of the institutional capacity of SEDUE delegations in the project states and of environmental agencies of the proiect states and municipalities therein in order to improve project. assessment techniques and environmental policy formulation; and (c) restoration of selected archeological sites, and preparation of an inventory of archeological sites. 3.09 The institutional development component (US$18 million) would strengthen executing agencies (mainly in the agricultural and roads sectors)," state and municipal planning and budgeting systems, and prepare them to assume additional administrative functions received as part of the gradual decentralization process agreed under the Decentralization Matrix (see Annex 1, Attachment 1). The key actions, under this component, required to increase implementation capacity of executing agencies are listed in Annex 1, Attachment 2, pages 3 to 9, and would be part of a side letter. 3.10 The proposed Bank loan would contribute to the PNS program in Chiapas, Guerrero, Hidalgo, and Oaxaca for the 1990-94 period, starting in approximately May 1990. The proposed loan would be made to NAFIN, which would transfer the proceeds to SPP. The signing of contractual arrangements between those two institutions, satisfactory to the Bank, would be a condition of effectiveness. SPP would use the resources to finance the federal grant' contributions to subprojects executed by state and federal sectoral agencies, municipalities, and communities. The project uses grants rather than loans because it emphasizes social and access infrastructure"' and targets its effects on the poorest states and regions; it would not finance credit or recurrent costs. I. The Development Agreements (CUDs) and the Regional Planning Mechanism 3.11 The Development Agreement (CUD) is an annual agreement between the federal and each state govermment that establishes the overall framework for the matching grant investment program. The federation contributes from 50 percent to 75 percent of program cost, the remaining being contributed by the state and beneficiaries. The investment program agreed under the CUD is called the POA. The formulation, selection, execution and control of all subprojects covered by the POA is regulated by the CUD Operational Manual. By signing the CUD, the state undertakes the responsibility to contribute with its financing and follow the Operational Manual. In addition, and as a result of the project, the states would also undertake the responsibility of improving their resource mobilization and distributing grants equitablv. The federal government (SPP) and each of the project states would have to enter into a CUD and a project state agreement, annexed tc the CUD, to assure 15. For a Ust of Executing Agencies see Annex 10. 16. As described in Annexes 2 and 3 most of the taxes are collected by the fedeal governmn, and thc revew-harn system as well as the matching grant sytem for investments is a way to redistibute fiscalrsc to t and muticipalities. TIhs, contry to other Bank projects where loan poceedsc are used by the oetral govenment to nmke loans to states for invesanert progammS, the proceeds are used for the matching grnt sysem among the fedald and te govenments. 17. The issue of providing grunts for some productive projects for poor communities would be addressd under fitur operations, in view of the small share of thoe projects, and the fact tht they ar provided in a package with mixed credit.  - 22 - that all Bank and project rules will be respected, including the Bank's right to suspend disbursements in case of noncompliance, as a condition of effectiveness. 3.12 Project implementation will take place through the established regional planning mechanism described in the previous paragraph. The annual investment program (POA) results from proposals by line agencies, state institutions and communities, properly screened and evaluated by the respective technical ministries working in the state, coordinated through the COPLADES. The POAs are then negotiated with SPP/federal. The Bank would review the POAs. All subprojects financed by the project would have to be part of a Bank approved POA and satisfy the rules specified in the Operating Manual. The CUD Operational Manual specifies rules for appraisal and selection of projects, implementation, flow-of-funds, financing, cost recovery, and operation and maintenance. The Manual has been thoroughly reviewed ar.d simplified in collaboration with the Bank, and has been amended to reflect all the rules defined herewith for subprojects. Tle Manual cannot be changed without Bank consultation. Annual reviews will also provide the opportunity to redirect institutional development efforts to those agencies which reveal institutional weaknesses and investment resources to those agencies which have shown greater implementation capacity. The Bank has the right to stop disbursements to a noncompliant state and may, upon request of the Borrower and Guarantor, reassign loan resources from a noncomplying to a complying state. 3.13 Within the CUD mechanism, a system of Municipal Funds has been created to allow municipalities to undertake small community projects that will mobilize local labor resources and would allow a more flexible and faster project selection and implementation. Rehabilitation and repair of social infrastructure are already mobilizing a large effort and enthusiasm in the more than 800 municipalities in the four states, that have started to benefit from those Funds last year. 3.14 The following sections describe the specific mechanisms of the project. The first and second section describe what criteria and rules are being used to focus the project in poverty alleviation and maintain the sustainability and efficiency of subprojects, as well as the targeting mechanisms. Since this is a regional operation that would complement some sectoral Bank operations already underway in the Project States, Section 4 specifies the coordinating mechanism among those projects, and Section 5 describes the content of the Investment Programs that have been drawn up through a widely participatory process during project preparation. In view of the large investment effort required by poverty alleviation and to assure good implementation and operation capacity, Section 6 describes the institutional development effort to strengthen the agencies involved in project execution. 2. Providijng Resources for Poverty Alleviation throug h the Investment Prgams of the Four Poorest States 3.15 The national poverty alleviation strategy is being defined by PNS. To ensure consistency of community preferences with national efficiency and distribution objectives, various mechanisms have been specified in the project which include: (a) rules to strengthen and monitor the poverty focus of the program and its subprojects; (b) an inproved appraisal and selection system for subprojects; (c) a matching grant system that specifies how much federal, state, and municipal sources will contribute to each type of subproject; (d) a schedule of beneficiary contributions in the form of cash, kind, and/or credit; (e) cost recovery rules for operation and maintenance which, in the case of utilities, require 100 percent recovery via tariffs; and (f) proper budgeting of recurrent costs of the investment plans and of individual subprojects via financing plans. The poverty alleviation objectives ef this project are summarized in Ta! '. '.1 in the form of performance indicators to be used for monitoring progress in the annual in ;stment programs.  -23 - Table 3.1. PERFORMAMCE INDICATORS PRQGOFUINCCATOR | . _. W~~~~~~~~~~~~~~G6HEGATE HAPAS kUEFMMR HU AGO S CA| cuvcn PROoUCnTVE INESTMENT DEVS.OPMENT OF WIUATED AREAS h . OF RIGATIONPOWETIALTOREDEVELOPD 6 2 3 13 a > MRIATED AfEAS WICH WOULD BE S4A3UBATED OR EXPOAI. THOUSAND HA (%I 143 (701 18164) 9175) 7(170) 43 10s) . AGROWOUSTRIAL PROOUCTON AND TRAWSFORMATOM WUMBER OF SENEFICUUES (THOUSANDI 472 125 117 176 6S SFRASTRUCTURE INVESTMENTS K. FEEDER AND RURAL ROADS I. % OF PLANNf ROAD NETWORK WHCH WOULD BE BUILT . % OF REHAMUTATION OF ROAD NETWORK 33 30 69 so 21 3. % OF THE ROAD NETWORK WHICH WOULD RECEIVE MAhNTENANCE 77 76 78 62 73 EU. TRANSMSSION AND DISTRIBUTION UNES % OF POPULATION LACKING ELECTRFICATION WHICH WOULD BE 73 72 74 78 70 SERVICED 36 27 35 61 47 SOCAL INVESTMENTS FF. EXPANSOIN AND IhROVEENT OF THE PHYSICA PLANT FOR PRIMARY EDUCATION * . % OF POPULATION WITHOUT PMARY EDUCATION WHICH WOULD BE SERVED 40 33 33 33 68 2 % OF POPULATION LACKING SECONDARY EDUCATION WHfCH WOULD BE SERVED 29 26 27 27 42 EC. CONSTRUCTION AND RIHABILITATION OF POTABLE WATER SYSTEIS % OF POPULATION WITHOUT POTABLE WATER WHCH WOULD BE SERVED 61 63 so 80 48 ED. DRAINAGE AND SEWERAGE TREATMENT % OF POPULATION LACKING DRAINAGE WHCH WOULD BE SERVED 30 30 23 s0 27 mNMRowEffT AND CULTJRAL COMPOT _ REFORESTED HECTARES 82.400 42,00 18.0oo 4.700 17.200 PROTECTION SYSTEMS IN THE LACANDONA: PRESERVING FOREST MASS * DECREASE/PROH18TION OF LOGGING M%) 100 DECREASE IN AREA SUBJECT TO SLASH/BURN 70 * REFOfESTATION HECTAfES 35.000 fHABIUTATIONlPROTICTION OF CULTURAL PATRIMONY S OF RECUfERATION OF ARCHEOLOGICAL MUONUhENTS 31 34 36 37 28 Source: Annex 1, Attachment 2. See the attachment for definitions end more details. (a) Maintaining the Poverty Focus of Investent Subhrograms 3.16 The project wifl use five mechanisms to maintain its povert focus:' (a) SPP and the Bank will review the POAs to ensure a minimum average investment per capita and per poor person in each region of each state;" (b) they will review the sectoral composition of the states' overdl investment programs to ensure poverty focus, using such criteria as dominant share of investment allocated to primary education, rural roads, potable water, etc.; (c) selection criteria will favor projects that provide benefits to poorer groups, based on rough evaluation of beneficiaries" income levels; (d) actual investments will be reviewed on completion to ensure that programs have been implemented according to the above criteria; and (e) matching grant ru.es wiUl provide larger 18. Thb u in major coat to PIDER and other smaller niag deveopmnt pmojecu. 19. Poor mgion defuned by the COPLAMARICONAPO poverty indica.  - 24 - grant components for projects in more marginal regions, as measured by the marginality index developed by COPLAMAR/CONAPO. (b) Project Appraisal and Selection 3.17 The system of project appraisal builds on long Bank and SPP experience in regional planning, as developed by the PIDER project and the CUD regulations and as refined during project preparation. Subprojects are classified as small-class A (under US$25,000) with local impact; medium-class B (between US$25,000-US$200,000); and large-cl-ss C (over US$200,000) with statewide impact. Subprojects of the first type, comprising about 40 percent of the total investment program and 80 percent of the number of expected projects, are evaluated from a community development perspective by a municipal commnittee and supervised by COPLADEs, using a scoring method summarized in the next paragraph and described in Annex 4.2 Ctrtain exclusion criteria also apply."' Class B and class C projects are subject to technical, financial, and economic analysis by the state and federal line agencies. Large projects receive more thorough analysis and medium projects an intermediate analysis. Rate-of-return criteria will be a2lied to all productive and large infrastructure projects. In the case of road projects, the required eligibility criheria is specified in Annex 4, Attachment 2. These criteria have been included in the CUD Operational Manual and evidence has been provided to the Bank that instructions have been issued. During negotiations the Government provided assurances that SPP will submit to the Bank, on a quarterly basis, a list, satisfactory to the Bank, of all investment subprojects authorized during the prior three months by SPP for financing under this project. 3.18 The system, described in Annex 4, permits municipalities and states to select projects of high priority through a simple, systematic, and objective technique. Five quantifiable criteria are aggregated into a project score: (a) technical feasibility and efficiency; (b) benefits relative to cost; (c) minimum beneficiary contribution (cost recovery); (d) targeting to the poor; and (e) beneficiary commitment to operation and maintenance. Projects must satisfy minimum requirements for each criterion, and have a minimum overall score. Among competing projects those with the highest scores are selected. 3.19 The proposed project cycle and rapid appraisal system constitute one of the most innovative features introduced by this project. As explained in Attachment 1 of Annex 4, project proposals are generally presented by community groups to municipalities. Th,e municipalities fill out the basic project formats and apply the screening process. Once projects are selected, municipalities aggregate them into proposals and forward them to the state planning agencies. These agencies review the scores and aggregate the lists of all municipalities with the estimated funding required. These state programs are then reviewed by COPLADEs and the SPP delegations and negotiated with SPP federal. 20. Scoring is based on a ystan of weighb for severel key objectives. The sysen is simple (decision-makem have low educational level and information is limited); it has alrady been implemented and experience so far demonsttas its effectivencu. 21. Among othem, projects tht have an advese envirmn impact or can damage the cultural patinony, or lead to involuntary rlesment of populations, ar excluded. 22. The form required for its presnin is contaimd in SPP, Otiws 2ea la -elabomcion dc edientes tenicos simolfcados, Mexico, December 1989.  -~~~~~~~~ -2S - 3.20 Technical feasibility and efficiency Is subject to the aorms and stdards established by the federal and state agencies. For example, the CNA would follow the same technical rules for fesibility studies and project screening for all projec, independently of the source of financing. Every project has to be reviewed and approved by the televant sector agency (in Guias Kara I jlaboracion de Expedientes Tecnicos Simnliflcados, Annex to the CUD Operational Manual) and the sectoral committee of the COPLADE. The Bank reviewed the corresponding technical manuals and found then. satisfactory. 3.21 The economic efficiency criteria require at least a 12 percent rate of return for productive and large infrastructure projects and that social sector and small infrastructure projecs must minimize costs and respond to revealed social demand, as well as being appropriately targeted to the poor. In the case of social sector projects, basic technical norms, as defined by the Secretariat of Public Education (SEP) or CNA, seek to minimize cost per social demand. These minimum costs are incorporated into the guidelines for project selection and into the scoring system for small projects. 3.22 Operational manuals issued for the Municipal Funds already incorporate the system for appraisal and selection of small projects described in Annex 4. The final draft of the CUD and Municipal Funds Operational Manuals has been prepared in close collaboration with the Bank. Formal approval of such documents has been provided to the Bank. The experience with the small projects selection system of the FMS will be reviewed with the Bank no later than October 31, 1991. TMe revised system for small project appraisal, agreed with the Bank, would be incorporated in the CUD Operational Manual by November 1991, to start to be applied to the POA of 1992. By that date the two systems for channelling monev (CUD and FMS) to small projects would have similar rules for appraising small projects. In addition, and in order to monitor improvements in analyzing all medium and large projects, a satisfactory appraisal report of two subprojects (one medium and one large) in each of the eight program categories to be financed by the Bank, with at least four subprojects per state, would have to be submitted to the Bank in order to comply with conditions for disbursement in calendar years 1991, 1992 and 1993. 3.23 Environmental screening and monitoring of investment projects would be carried out within the framework established by the national environmental law (Ley General de Equilibrio Ecologico y Proteccion al Ambiente, 1988) and its regulations, as well as the Ley de Obra Publica (art. 28, in particular). Projects below US$200,000 will be subjected to a simple environmental screening process. Larger projects would require an envirownental assessment. Projects with ex-ante favorable impact (e.g., sewerage, reforestation ) or without environmental impact (e.g., education, training) would be exempt from this process.' Funds will be provided throughout implementation to monitor compliance with environmental safeguards established during enviromnental screening and assessment. SEDUE will be responsible for clearing the annual invesunent programs of each state and individual subprojects with regard to compliance. The rules to be used for environmental screening of subprojects are regulated by the above law. During 1991, SEDUE will prepare guidelines to simplify and improve the application of the above law (see Annex 4, Attachment 3 for the agreed rules). 23. See Annex 7. Atuhainemnt 5 for a broader desription.  -26 - (c) EmWnbg ad u ".IiigKimiuL 3.24 The matching grant system used in the CUD has induced a better use of resources and is an important instrument for ensuring the consistency of national and state objectives. Investment grants from the federal government are given as matching grants to each project. According to the CUD Manual, the state share varies from about 75 percent for a productive project in a rich state to 25 percent for a social project in a poor state. Ihis system is applied to the entire PNS (see Annex 5 for the cost sharing in the four states). Ihese rules would also apply under the proposed project. fii) Own-state and municinal ro 3.25 1hrough the matching grant system, states are required to contribute at least 25 percent of their own resources collected through general taxation within their jurisdiction (Annexes 2 and 3). During project execution it is expected that a similar system, t ing into account the poverty level of the municipality, will be designed and implemented between each state and its municipalities for projects with local impact. According to projections presented in Annex 3, Attachment 1, states will be able to finance the proposed investment program. The 1980 tax reform transferred to the Federation most of the state and local taxes and introduced a revenue-sharing system through which about 18 percent of federal taxes are devolved to the states in the form of a block grant. The block grants represent 70 percent to 80 percent of total revenues of the four states. The remaining 20-30 percent will come from property taxes, water charges, and other minor service charges (derechos and productos). Urban infrastructure projects also obtain credits from BANOBRAS. 3.26 The project states are being encouraged to improve state tax collection, as referred to in the Policy Letter. The 1989 revenue-sharing formula between the federation and the states introduced local taxes as an element in calculating each state's share. Last year, in anticipation of this change, there was a significant increase in property tax collection (in Guerrero, collections increased by about 800 percent, and in Hidalgo 600 percent). The project would further help enable the states to collect revenues required for the investment program's matching grants and recurrent expenditures. Te project would assist all four states to: (a) complete updating their cadastres by the end of 1991; and (b) reassess property values by the end of 1992. Several studies have been initiated to codify and modernize their fiscal systems; a major study on fiscal laws and state budgets was conducted in the course of project preparation. No new taxes are being proposed, but major shortfalls in the operation and collection of existing taxes and levies have been identified. A commission to review state fiscal laws and draft proposals will be established by June 1991 in each of the four states. New state riscal codes will be drafted before the end of 1992 and the respective bills presented to the state legislative branches. The Policy Letter (Annex 1) addresses the aims regarding these issues, and the action programs agreed with each of the four states would be incorporated in the Execution Annexes of the CUD Agreement. 3.27 As municipalities increase their institutional capacity, a larger part of the state revenue-sharing (gartici=acionemw is passed on to them. Under the new Hidalgo Fiscal Coordination Law at least 20 percent of the participaciones are transferred to the municipalities. When municipalities are not yet capable of performing a special function as established by law, states retain this function and charge municipalities for administrative services. It is estimated that, due to decentralization, transfers to municipalities will increase by about 15 percent (Chiapas) to 100 percent (Hidalgo) in real terms.  -27 - 3.28 Reform of the formulas for allocating revenue-shasing passed on by the state to municipalities will be promoted during the project. In some states the formula is unclear; in others it, poses problems of inefficiency and inequity. A reform is impotant since almost 80 percent of the - revenue of municipalities in these states derives from those transfers. Hidalgo has taken the lead and enacted a law containing a new formula based on population, local tax effort, and a marginality index, in close agreement with Bank proposals. The other three states have initiated studies required for moderniztion of their revenue-sharing formulas. It is expected that new draft laws would be compieled by June 1991 and presented to state congresses in the second half of that year. This action would also be reflected in the action programs of the project states as referred to in para. 3.11. Lack of sufficient progress under the monitorable actions in this paragraph and para. 3.26, cggld constitute a rean for suspending disbursements to a particular state. In such cases, funds could be canceled or reallocated to other states. (d) Benefrciary Contributions to Investment Cost 3.29 The CUD Operational Manual incorporates rules for beneficiary contribution. Beneficiary contributions to investment costs are regulated through the CUD/PNS Manual Unico de Operacion, table "Apertura Programatica y Estructura Financiera." Productive projects are financed through a mix of grants (from 75 percent to a poor state down to 50 percent to a well-off state) and loans (from 50 percent in a well-off state down to 25 percent in a poor state) that have to be secured by the beneficiaries, prior to project execution. Part of the grant is federal (maximum of 50 percent to a poor state), and the rest is a matching grant from the state. States also require municipal and/or beneficiary contributions for such that are incorporated in the state matching grant. In a survey carried out of subprojects during 1989, cash contributions represented from 1 to 9 percent of project costs and materials, and community labor (provided free) from 10 to 15 percent of project costs (see Annex 5). Social projects are financed by grants from the federal government (from 75 percent in a poor state down to 50 percent in a well-off state) and the rest has to be covered by a state matching grant. States also request municipal and/or beneficiary contributions. To build a prinury school or a rural road, the beneficiaries are required to contribute local materials and construction work (as free community work-1