Document of The World Bank FOR OFFICIAL USE ONLY Report No. 12529-ME STArF APPRAISAL REPORT MEXICO SECOND PRIMARY EDUCATION PROJECT MARCH 11, 1994 MICROGRAPHICS Report No: 12529 ME Type: SAR Country Department II HL'man Resources Operarions Division Latin America and the Cribbean Regional Office This document has a resicted disMtibution and may be used by recipients only in the performance of their offlcial duties. Its content may no otherwise be disclosed without Wodd Bank authoroizaon CURRENCY EQUIVALENTS Currency Unit = New Peso (N$) USS1.00 = 3.12 New Pesos (Feb. 1993) FISCAL YEAR January 1 - December 31 ACADEMIC YEAR September 1 - June 30 ACRONYMS AND ABBREVIATIONS ANAM National Agreement for Basic Education Modernization (Acuerdo Nacional para la Modernizacion de la Educaci6n Bdsica) CAPFCE Federal School Construction Agency (Comite Administrador del Programa Federal de Construcci6n de Escuelas) CAS Country Assistance Strategy CONAFE National Council for Educationai Development (Consejo Nacional de Fomento Educativo) at SEP CONALITEG National Free Textbook Commission (Comisi6n Nacional del Libro de Texto Gratuito) CONAPO National Population Council (Consejo Nacional de Poblaci6n) COPLADE State Planning Council (Consejo de Planificaci6n del Estado) DGEI General Directorate for Indigenous Education (Direcci6n General de Eaucaci6n Indigena) at SEP DGEIR General Directorate for Evaluation (Direcci6n General de Evaiuaci6n, Incorporaci6n y Revalidaci6n) at SEP DGPPP General Directorate for Planning, Programming and Budgeting (Direcci6n General de Planeaci6n, Programaci6n, y Presupuesto) at SEP LGE General Education Law (Ley General de Educaci6n) MIS Management Information Systew NAFIN National Financing Agency (Nccional Financiera, S.N.C.) PAM Teacher Training Program (Programa de Actualizaci6n del Magisterio) at SEP PARE Bank's first Primary Education Project, Ln. 3407-ME, 1991 (Programa para Abatir el Rezago Educativo) PAREB Second Primary Education Project (Programa para Abatir el Rezago en Educaci6n Bdsica) PCU Project Coordinating Unit at the Federal level in CONAFE PME Education Modernization Program (Programa para la Modernizaci6n Educativa, 1989- 1994) PRODEI Bank's Initial Education Project, Ln. 3518-ME, 1992 (Proyecto para el Desarrollo de la Educaci6n Inicial) SCEP Servicios Coordinados de Educaci6n Pzdblica, former SEP delegation in the states essentially replaced by SEE SEB Undersecretariat of Basic Education (Subsecretaria de Educaci6n Bdsica) at SEP SEE State Education Entity (State Secretariat of Public Education or equivalent organization) SEP Federal Education Secretariat (Secretaria de Educaci6n P7blica) SHCP Federal Finance and Public Credit Secretariat (Secretaria de Hacienda y Credito Pablico) SNTE National Teachers Union (Sindicato Nacional de Trabajadores de la Educaci6n) SOLIDARIDAD National Solidarity Program (Programa Nacional de Solidaridad) SOPE State Secretariat of Public Works (Secretaria de Obras Pfiblicas del Estado) SPCU Project Coordinating Unit at the State level located in SEEs TAC Technical Advisory Committee TWP Teaching Career Program (Programa de la Carrera Magisterial) at SEP UPE Education Publications Unit (Unidad de Publicaci6n Educativa) at SEP FOR OFFICIAL USE ONLY MEXICO SECOND PRPMARY EDUCATION PROJECT STAFF APPRAISAL REPORr TABLE OF CONTENS LOAN AND PROJECT SUMMARY ................................ I. THE ECONOMIC AND SOCIAL CONTEXT .1 H. THE PRIMARY EDUCATION SYSTEM . ............3............... 3 A. Overview. 3 R Issues in Primaiy Education. 4 Quality ..... 4 Efficiency. 9 EquitY .12 C. Goverment Strategy for the Education Sector .13 D. Bank Role and Experience in the Education Sector .15 E. Lessons Learned from Past Operations .16 m THE PROJECT .20 A. Project Concept and Objectives .20 B. Project Areas and Target Population .20 C. Project Components .21 D. Project Description .23 Component A. Human Resoumces Development .23 Component B. Educational Material Resources .25 Subcomponent B 1. Teaching Materials and Student Educational Packages .25 Subcomponent R2. Reading Corners .. ............... 25 Subcomponent B.3. Textbooks and Educational Materials for Indigenous Education .................. : ..... 26 Subcomponent B.4. Improvement of Physical Infrastnucture .27 Component C. Institutionl Strengthening .29 Subcomponent C. 1. Management Capacity .29 Subcomponent C.2. Teacher Incentives .32 Subcomponent C.3. Improvement of the Supervision System 32 Subcomponent C.4. Books and Materials Distnbution System 33 Subcomponent C.S. Project Administration .34 This report is based on the findings of the appraisal mission that visited Mexico in November, 1993. The ndssion was composed of Mmes./Messrs. Antonio Gomes Pereira (mission leader), Eduardo Wlez. Manuel Vra, John Innes, Rosaria Iroza, Robert Etheredge "LA2HR), Silvia Elena Arevalo, Carlos Heymans and Maria Elena Anderson (consultangs). Messrs/Mme. ithn Nguyen, Rosita V Esnuda, KWin Rost (L42HR), and Rebecca Cary (consultant) contributed to the report at headquarters. Peer reviewers are Mwchelle Riboud (EC4IIR) and Luis Pisani (LA4HR). Messrs. Edilberro L. Segura and Kye Rbo Lee are the Department Director and Division Chief respectively, for this operation. Ths document has a restricted distnbution and m be used by recdpients only in the perfornwn of icir official duties. its contnts maw not othevws be disclosedwithout Word Ba authoizadion. - ii'- IV. PROJECT COST, PROCUREMENT, DISBURSEMENT, AND AUDITS . ....... 35 A. Project Costs and Finaning ............................... 35 B. Procurement ......................................... 37 C. Disbursements ....................................... 40 D. Accounts and Audits .................................... 41 V. PROJECT MANAGEMENT AND SUPERVISION ..................... 42 A. Project Implementation ............................ ..... 42 Project Preparation .............................. 42 Project Readiness and Implementation ........... ............ 42 Project Monitoring .............................. 43 Annual Implementation Reviews ........ ............. . 43 Mid-Term Review ................................. 44 B. Project Supervision ...................... 44 C. Project Launch Seminar ...... ........................ 44 VI. PROJECT BENEFITS AND RISKS .. 45 A. Project Benefits . ..................................... 45 B. Impact on Women ...................................... 45 C. Environmental Inpact ................................... 45 D. Project Risks ......................................... 46 E. Program Objective Category ......... ...................... 46 VII. AGREEMENTS REACHED AND RECOMMENDATION ................ 46 ANNEXES 1. Comparative Education Indicators 12. Strengthening of Planning and Policy 2. Primary Education Efficiency Analysis Indicators 13. Terms of Reference for Process 3. Educational Expenditures Evaluation & Information Systems 4. Government Policy Letter 14. Terms of Reference for Studies 5. Decentralization of Education Budget 15. Teacher Incentives by Project State 6. Marginality Ranking and Project 16. Project Management and Flow of Areas Funds 7. Indigenous Populations 17. Costs and Financing 8. Participating Municipalities and 18. Project Disbursements Beneficiaries by Project State 19. Monitoring Indicators 9. Financial Implications of Extension of 20. Amnnal and Mid-Term Reviews Basic Education 21. Supervision Plan 10. Teacher Training 22. Project Implementation Schedule 11. Physical Infrastructure and 23. List of Documents in Project File Distribution Network MAP - IBRD No. 25451 MEXICO SECOND PRIMARY EDUCATION PROJECT LAN AND PROJECT SUMMARY Borrower: Nacional Financiera, S.N. C (NAFIN) Gumrantor: United Mexican States Beneficiaries: The Federal Education Secretariat (SEP) through the National Council for Educational Development at SEP (CONAFE), and State Education Entities (SEEs) in ten Mexican states (Campeche, Durango, Guanajuato, Jalisco, Michoacan, Puebla, San Luis Potosi, Tabasco, Veracruz, and Yucathin). In addition, the four states (Chiapas, Oaxaca, Guerrero, and Hidalgo) assisted under the first Primary Education Project (Ln. 3407-ME, 1991) would also benefit from the management training program under the proposed project's Institutional Strengthening component. Amount: US$412.0 million equivalent. Terms: Repayment in 15 years, including 5 years of grace at the standard variable interest rate. Project The project would improve primary school students' academic Objectives: achievement levels and reduce the high repetition and dropout rates within ten of the country's poorest states, contribute to raising the human capital level, redressing some of the country's social and economic imbalances as part of the Government's social compensatory program. These objectives would be achieved through: (a) training teachers and administrators; (b) providing classroom-tested educational materials for both teachers and students, including specific materials for indigenous populations; and (c) strengthening the institutional capacity of the education system, particularly at the state level, to carry out its mandate within a decentralized system. The project would further benefit the four states being assisted under the first Primary Education Project by including their senior-level officials under the project's management training program. Project The project consists of three major components: the Human Description: Resources Development Component (17.5 percent of total project costs) would provide in-service training and assistance to upgrade the skills of primary school teachers, principals, and supervisors, emphasizing the role of principals and supervisors in providing pedagogical assistance to teachers in classroom teaching. The Educational M ial Resources Component (52.5 percent of total - iv - project costs) would: (a) provide educational materials for both teachers and students; (b) supply classroom library and reference books and promote their use; (c) design and deliver bilingual textbooks and materials for indigenous schools; and (d) rehabilitate, replace, or build urgently needed educational facilities to replace inadequate schools and meet new enrollment demand. The Institutonal Strengthening Component (30.0 percent of total project costs) would: (a) strengthen education management capacity at both central and state levels by (i) providing management training for planning and policy analysis, (ii) installing project monitoring, evaluation, and information systems, (iii) conducting education studies to prepare future investments at the preschool and secondary education levels, and (iv) supporting project promotion and diffuision activities; (b) provide incentives for teachers to work in remote, indigenous, or hardship areas; (c) improve the supervision system; and (d) strengthen the textbooks and materials distribution system. The project would be coordinated by CONAFE at the central level and implemented by SEEs, which would be responsible for the delivery, operation, and management of educational services at the state level. Project By raising children's cognitive achievement as well as their Benefits: probability of completing the primary education cycle, the project benefits would consist of a higher rate of human capital accumulation, contributing to economic development. Due to the project's targeted nature within the Government's social compensatory program, project benefits would also include a more equal distribution of economic opportunities amOng Mexican children, especially for underserved indigenous populations. The project would improve the quality of primary education through in- service training of teachers and administrators and enhance the system's efficiency through investments in institutional strengthening, thus contributing to better allocation and increased productivity of human resources in the sector and helping meet the challenges of a recently decentralized education system. Altogether, benefits would accrue to approximately 1.5 million children, 53,000 teachers, and 1,140 administators. Because half of the students would be girls, secondary benefits would include lower ferdtlity and improved health status and educational attainment of their children. Project The main project risks are associated with possible delays in Risks: project execution due to the slow improvement of institutional capacity of implementing agents at the state and, where applicable, community levels. The risk of weak institutional capacities is addressed under the project through technical assistance in key areas of management such as planning, information, and evaluation. The project would benefit from CONAPE's experience at the central level, which is curny implementing the first Primaty Education - v - Project as well as the Initial Education Project (Ln. 3518-ME, 1992) and would easily tasfer successfl operational expertise as well as familiarity with Bank processes and procedures to the proposed project. At the state level, SEEs would have had two years of operating a decentralized educational system, by the project's effectiveness. At the local level, participation of School Councils (formed under the 1993 General Education Law) would benefit from the experience of similar community organizations already operating in different sectors and other Bank-financed projects. The risks identified dunng early project preparation posed by the disparity in salaries for federal and state teachers (who would now all belong to the state system) have been practically eliminated: salanes have been leveled in all but three of the participating staves, and these three states are scheduled to complete this process in 1994. Pbverty Catgory: The project would support a program of targeted interventions. The project would support the national policy of compensation for regional disparities in education. Project interventions would be targeted not only to ten of the poorest states but also to the poorest regions and most indigenors areas therein. The proposed project and first Primary Education Project together would cover the 14 poorest states in the county (out of a total of 32 states). Major urban areas within the selected regions have been excluded. - vi - Project Cost Summary Local Foreign Total ---US$ million- Human Resources Development 99.8 2.6 102.4 Educational Material Resources Teaching Materials & Student Packages 85.6 85.6 171.2 Reading Corners 4.8 2.1 6.9 Indigenous Education 3.9 2.0 5.9 Infrastructure Improvement 93.8 24.7 118.5 188.1 114.4 302.5 Institutional Strengthening Management Capacity 11.6 5.3 16.9 Teacher Incentives 97.5 0.0 97.5 Improvement of Supervision System 16.6 3.4 20.0 Books & Materials Distribution System 7.6 2.8 10.4 Project Administration 26.6 1.0 27.6 159.9 12.5 172.4 BASE COST 447.8 129.5 577.3 Physical Contingencies 5.0 1.4 6.4 Price Contingencies 26.3 6.7 33.0 TOTAL PROJECT COST' 479.1 616.7 Financing Plan LoAl Foreign Total -TS$ million- Federal Govenunent 204.7 0.0 204.7 IBRD 274.4 137.6 412.0 Total Estimated Disbursements Tlmetable IBRI) F1scal Year 1994 199S 1996 1997 1998 1999 Annual 30.0' 52.0 88.0 112.0 80.0 50.0 Cumulative 30.0 82.0 170.0 282.0 362.0 412.0 Economic Rate of Return: Not Applicable Y Includes taxes and duties estimated at about US$44.3 million. ' Includes Special Account deposit of US$25.0 million and retroactive financing of up to US$6.0 million for eligible expenditures incurred on or after Cctober 31, 1993. - vii - BASIC DATA SHEET Mexico Year A. General Country Data GNP Per Capita (US$) 3,030 1991 Area (Thousand Square Kilometers) 1,958 1993 Population Estimate (Millions) 83.3 1991 Urban Population (% of Total) 73 1991 Population Projection, Year 2000 99.0 1992 B. Social Indicators Crude death rate 5.4 1990 Average Annual Rate of Population Growth (%) 1.9 1990 Total Ferility Pate 3.2 1991 Infant Mortality Rate (per 1,000 live births) 36.0 1991 Life Expectancy at Birth (years) 70.0 1991 Adult Illiteracy Rate (% of persons 15 and over) 11.8 1991 C Education Statisfics Primary (Net)(Grades 1-6)(%) 98.0 1991 Lower Secondary (Grades 7-9)(%) 58.0 1991 Total Primary Enrollment (millions) 14.4 1991 Federal (%) 71.i 1991 State (%) 22.5 1991 Private (%) 6.4 1991 Primary School and Efficiency Data Number of Teachers (thousands) 479.6 1991 Student/Teacher Ratio 30.0 1991 Incomplete Schools (% of all primary schools) 14.9 1991 Unitary Schools (% of all primary schools) 22.7 1991 Repetition Rate (%) 9.8 1992 Dropout Rate (%) 4.6 1991 Completion Rate (%) 58.1 1991 Education Expenditure Data Total Public Education Spending (% of GDP) 4.3 1991 Federal Education Spending (% of Total Public Education Expenditures) 82.8 1991 Federal Education Spending (% of Total Federal Exrenditures) 9.4 1991 Federal Pri- -,' Education Spending (% of Federal Education Expenditures Among Levels) 39.8 1991 Sources: Social Indicators of Development, 1991-92; INEGI Censo de 1990; World Development Report, 1993; SEP and Bank staff stimaes, 1993. viii - ElUCATION GLOSSARY Completion rate For six-ye-r primary education, the ratio between the number of students r sJing grade 6 in a given year T and the number of students entering grade 1 in year T-6, as a close approximation of the percentage of students of a given cohort to complete six- year primary school. Dropout rate Percentage of students who abandon the system without completing the education cycle, e.g., primary school. Educational attainment Number of formal school years achieved by a student. Educational achievement Magnitude achievement in any subject area, generally measured on the basis of national curriculum learning objectives. Enrollment ratio Ratio of total enrollment in a given school cycle (e.g., primary) to the population in the corresponding age group. Incomplete school Primary school with fewer than the six official primary education grades. Net enrollment rate Percentage of children of a given age group enrolled in their corresponding grade level. Promotion rate Percentage of the students enrolled in a given grade who enroll in the next higher grade the following year. Repetition rate Percentage of the students enrolled in a certain grade who are not promoted to the next grade and who therefore enroll again in the same grade the following year. Transition rate Percentage of the students completing one school cycle (e.g., primary) who enroll in the first grade of the next upper cycle (e.g., secondary) the following year. Unitary school Multigrade school with a single teacher (see Incomplete School). MEXICO SECOND PRIMARY EDUCATION PROJECT STAFF APPRAISAL REPORr I. THE ECONOMIC AND SOCIAL CONTE 1.1. Over the past seven years, Mexico has consistently implemented a highly successful economic reform program buttressed by macroeconomic stabilization policies which reduced inflation from over 105 percent in 1986 to less than 10 percent in November 1993, reversing the decline in economic growth from a negative 3.8 percent to a positive 2.6 percent over the same period, thereby restoring confidence and extnml creditworthiness in the Mexican economy. A far-reaching trade liberalization process has opened the economy to internatonal competition. The ongoing process of privatization, aiming to enhance domesuc economic efficiency, has resulted in the sale or closure of over two-thirds of state enterprises. More recently, reforms have been initiated in the agricultua sector, and even the communal farming ownership system, the ejido, is being restructured. Further liberalization of the Mexican economy through the North American Free Trade Agreement (NAFTA) with the United States and Canada will represent an unprecedented integration of large economies at very different stages of development. This rapid assimilation into the world marketplace has increased the competitive pressure on domestic industrial sectors, and Mexico appears determir- I to upgrde its labor force level as a prerequisite to take full advantage of these econo, dc opporuities for sustined growth. 1.2. The stabilization program has highlighted issues in social development Despite its macroeconomic achievements, Mexico's potential growth is stdl constained by relatively low levels of human capital accumulation and widespread poverty. The illiteracy rate for the adult population is still at 12.4 percent, higher than the rates for Argentina, Costa Rica, Chile, and Venezuela. The average number of years of schooling for persons aged 25 and over is 6.5, which is lower than those of Chile (7.5) and Colombia (7.1). The average primary education repetition rate for Mexico is 9.1 percent (Annexes 1 & 2). The bulk of unemployment is composed of young individuals with a low level of education, and about 20 percent of unemployed males have not completed primary education. An analysis of the determinants of unemployment in Mexico reveals that the probability of being unemployed decreases significantly with age and with education level. The impact of education as an explanatory variable for workers' income is twice as high as that of any other examined variable. Consequently, Mexico's capacity to compete in the international market will hinge on its ability to harness an educated and skilled labor force, increase the productivity of its workers, adopt modern production processes and technologies, and develop new skldls as a more diversified structure of labor demand unfolds The economics of education have repeatedly suggested that primary education is the cornerstone for a modernized human capital stock. -2 - 1.3. The continuing inadequacies in human resources are caused in part by poverty, and in turn they contribute to the perpetuation of poverty.Y' A 1991 Bank study summarizes the incidence of poverty in Mexico as follows: (a) 19.5 percent of the population live't in extreme poverty (i.e., with a per capita income below US$360 per annum in 1984 prices); (b) 76.6 percent of the extreme poor are located in rural areas; (c) 37 percent of the rural population in Mexico are classified as extremely poor; and (d) the extreme poor have large families, the highest dependency ratios, and the lowest educational levels.2' Real per capita incomes in the richer (mainly northern) states are more than twice those in the poorer (mainly southern) states. Given that labor is the principal asset of the poor, improvement in the provision and quality of education represents a key mechanism for reducing inequality and lowering the number of individuals living in absoiute poverty. The retums to schooling, however measured, have demonstrated that more and better schooling is an important policy tool for poverty alleviation, through increased productivity and income, and for non- economic success. Basic education, by conveying the skills of reasoning, communication, and numeracy, enables workers to interact more efficiently, follow instructions better, and complete increasingly complex tasks, which are crucial for job performance, workplace and community participation, and overall social advancement. 1.4. The severe curtailment of public investment in the 1980s and the recent liberalization process have created a need for investment in human resources to support continued economic growth. The Government has a clear perception of social needs and is assigning high priority to the development of human resources through investments in education, particularly at the primary level. Full success of the adjustment program requires policies L.at are effective in promoting a more equal distribution of gains and opportunities of macroeconomic reforms. Through improved basic schooling, a more capable labor force will be developed, which will contribute to offset regional imbalances and make more Mexicans active beneficiaries of development. In pursuit of this strategy, the Govermnent of Mexico asked for, and the Bank supported, a Primary Education Project (PRogmma para Abatir el Rezago Educativo, PARE, Ln. 3407-MIE, 1991) with a total investnent of US$352 million in four of the poorest states. In 1992, the Government sought and the Bank approved an Initial Education Project (PRODEI, Ln. 3518-ME, 1992) for the ten poorest states, including the four in PARE, with an investment of US$115 million. In 1993, the Government asked for a second major follow-up investment in primary education for the rnext ten poorest and educationally least developed states.3' The Bank started project preparation in February and appraised it in October/November. This report analyzes and supports the proposed Second Primary Education Project, Programa para Abatir el Rezago en Educaci6n Basica, PAREB. Behrman, J. The Action of Human Resources and Poverty on One Another. Washington, D.C.: The World Bank, 1990. 21 World Bank, Mexico in TIansition: Towards a New Role for the Public Setor, Washington, D.C.: The World Bank, 1991. 31 The first Primary Education Project, PARE, covered the states of Chiapas, Oaxaca, Guerrero, and Hidalgo. According to the composite index of marginality used for targeting purposes (Annex 6), Hidago is actually the I0th most disadvanaged state but was selected under PARE (1991) to maximize previous social sector imNestments in the same four states. The report will refer to the ten proposed proect states as the "next poorest" as a matter of convention. -3 - H. THE PRIMARY EDUCATION SYSTEM A. Overview of Sectoral Organization and Achievements 2.1. Since 1950, Mexico has made great strides in increasing both the access to and the quality of public education: adult illiteracy has decreased from 40 percent to 12 percent; the number of schools has tripled, the number of teachers has quintupled, and school enrollments have increased sevenfold as the total population increased only threefold. By 1980, 25 percent of children aged 4-5 attended preschool, 91 percent of children aged 6-14 attended primary school, and the enrollment rate in lower secondary school reached 48 percent. By 1992, 68 percent of children aged 4-5 had access to preschool education and net enrollment rates at the primary and lower secondary levels reached 98 percent and 58 percent, respectively. In 1992, total primary school enrollment (grades 1-6) was 14.8 million children, attended to by a network of 88,918 primary schools and about 510,800 teachers (about half the 1.1 million teachers), or an average of 29 students per teacher (Table 1), a substantial improvement over the 45 students per teacher ratio in 1970. In July 1993, the length of the school year was increas,d from 187 to 200 days, to be enforced by all states as part of the General Education Law (LGE) (paras. 2.18 & 2.32-2.38). Ahble 1. Mexican School System - 1992/1993 Students Teachers No. (000s} | % (OOOs) Institutions IInitial _ r 117.9 0.5 3.3 _1,495 Preschool 1,858.9 11.2 114.3 51,554 Basic 19,031.3 74.7 748.5 108,950 Primary 14,828.2 58.2 510.8 88,918 Secondary 4,203.1 | _16.5 __237.7 20_032 Upper Secondary 2,177.2 8.5 151.0 6,833 Higher Education 1,306.7 5.1 138.8 2,239 btal 25,492.0 100.0 1,155.9 171,071 2.2. The formal education system in Mexico includes two-year preschool, nine years of mandatory basic education (recently established by integrating the six-year primary education and three-year lower secondary education cycles, para. 2.36), three or four additional years of secondary education, and undergraduate and graduate university programs. A well- designed and mainly non-formal initial education (ages 0 through 3) program precedes preschool. Of a total enrollment of 25.5 million students at all education levels in 1992/93, 57 percent were enrolled at the primary level (gades 1-6). Post-basic education starts at the upper secondary level, leading to either the liberal arts baccalaureate (80 percent of level enrollment) or the vocational track. Higher education lasts from three to six years, depending on the career pursued, and the universities have a semi-autonomous relationship with the Federal Education Secretariat (SEP). The public sector provides educational services for 90 percent of total enrollment. -4 - 2.3. In May 1992, the Government tansferred the management and operation of all initial (ages 0-3), preschool (ages 4-5), primary (grades 1-6), special, and lower secondary education (grades 7-9) and of teacher training institutions to the states (para. 2.34). All the facilities along with technical and administrative staff were transferred to the states with a corresponding budget of N$20.3 billion (US$6.6 billion) for 1993. Currently, SEP maintains direct responsibility only for the education system in the Federal District and for upper secondary education nationwide. SEP's leadership role w}'l focus on: (a) preparation and enforcement of educational policy norms; (b) curriculum development and textbook production; (c) design and financing of compensatory and special programs for an equitable distribution of education services; (d) support of educational research; and (e) operation of the evaluation systems. The total SEP budget for 1993 was N$41.7 billion (US$13.4 billion) and, for 1994, is N$44.8 million (US$14.3 million). 24. Primary education is provided in three modalities (enrollment shares in parentheses): general primary education (92.9 percent); bilingual or indigenous education addressing the needs of 46 ethnic groups in 23 states (6.3 percent);-l and the CONAFE (National Council for Educational Development, Consejo Nacional de Fomento Educativol model, designed to overcome the problem of building and staffing schools in the most remot rural areas (0.8 percent).§' Only the first two modalities would be supported by the proposed project. B ssues in Primary Education 2.5. Despite recent educational achievements, key problems remain to be addressed under three sets of issues: quality, efficiency, and equity. Quality 2.6. Low ducational Achievement. Although no nationwide achievement tests are systematically conducted in Mexico, data from periodic testing suggest that effective learning in school requires substantial improvement. A 1988 SEP evaluation on Spanish, mathematics, and the natural and social sciences indicated that students' average scores in grades 4-6 were dismal, only slightly above 20 out of a possible score of 100.7' A 1992 study found that Mexican students mastered only 53 percent and 30 percent of the national core curriculum in Spanish and mathematics, respectively Al These low educational achievements were associated with inappropriate curriculum, poor teaching quality, high V During the past 12 years, enrollment in the bilingual modality has increased at an annal rate of 5.6 percent, indicative of stronger demand and faster population growt rate in indigenous comnunities. SEP's General Directorate for Indigenous Education (DGEI) is responsible for the sector. 5 CONAFE is a semi-autonomous institution of SEP and is curently coordinat PARE and PRODEI. §' This community-oriented program relies on young teachers recruited among basic education graduates, who receive teacher training, texbooks, and mateials designed by CONAFE. The 18,798 CONAFE schools were built and are maintained by the community For one year of service, teachs receive accommodations provided by the community, a monthly stipend, and a three-year scholarship to purse upper secondary studies. In 1991/92, total enrollment in this program was 380,000. Y' World Bank. Mexico: Primarv Education ProieM, Report No. 9770-ME, Washington, D.C.: 1991. et Palafox, J.C., Prawda, J. and V6eez, E., PriM, School Ouality in Mexico - A View from LAkT* No. 33, The World Bank: 1992. teacher absenteeism, inadequate supervisory capacity, insuficient didactic materials, dilapidated physical facilities, deficient management practices, and several nonschool factors such as parents' lack of involvement in school issues. 2.7. Inappropriate Curriculum. The primary school curriculum had not been revised for 20 years until 1992, when the National Agreement for Basic Education Modernization (ANAM) introduced a new integrated curriculum stressing basic skills (para. 2.35). Implementation of the new primary education curriculum stared in "992/93 with the lowest grades; the proposed project will support it with teacher training (paras. 3.8-3.13) and updated educational materials (pamns. 3.14-3.19). A curriculum plan for grades 7-9, published in September 1993, will be implemented in a gradual process through 1995. The proposed project would support four studies to assess the quality and appropriateness of the basic education curricula (para. 3.25(c)). 2.8. Pbor Teaching Quaity. Only 30 percent of Mexican teachers, and less than ten percent of the indigenous education teachers, possess the required college-level teaching qualification or licenciaturm. Thus, many teachers lack the theoretical background and methodological skills needed to foster good interaction with the stdents, teach children effectively in the crucial lower grades, make the best use of textbooks and other instructional materials, and identify and orient chldren with learning disabilities. These deficiencies are exacerbated by: (a) the low quality and old-fashioned curricula of the teacher truning schools (Escuelas Normales); (b) the insufficient length and frequency of in-service training progams, which currently cover too many topics with little relevance to classroom teaching; and (c) the lack of financial resources to assist rural teachers to attend tramng progams offered in urban areas. Rural teachers are those most in need of training because of their relative inexperience, lack of supervision, and the added difficulty of handling multigrade teaching of children from deprived communities. The case of unitary (22.7 percent of all primay schools) and incomplete schools (14.9 percent) deserves special attention because effective management of multigrade classrooms (almost the rule in rural schools) is one of the key ways to improve teaching quality, requiring coordinated action between training, provision of educational materials, and supervision. Disadvantaged socioeconomic conditions add to the learning difficulties of children in rural communities, whose families have little or no educational background. Teachers must leam how to deal with these situations on the job. The project would provide better-designed and more comprehensive in-service teacher training offered in convenient locations (paras. 3.8-3.13), which would be followed up by a continuum of technical supervision with improved pedagogical inflection (para. 3.29). 2.9. TIeacher Absenteeism and Misallocation. Teacher absenteeism in nrual areas stems from: (a) difficult working and living conditions; (b) the long disunces teachers have to commute from their residences; (c) the amount of time teachers must spend (an average of four working days a month) just to cash their bimonthly checks; and (d) the paucity of incentives (e.g., training opporunities. hardship pay, or faster promotion) to accept rural assignments. As a result, teacher absenteeism is high, and many students lose a lot of class days each year (an average of 56 days or 28 percent of the school year in the worst cases). In addition, a high turnover rate among rural teachers (approximately 17 percent each yea.) leads to the assignment of inexpenenced entry-level teachers to the most crucial inital grades and difficult locations. Difficulties are further compounded by hiring and teacher assignment practices which are influenced by the National Tbachers Union (SNTE), which supports -6- teacher tansfers to urban areas. resulting in overstaffing of many urban schools and understaffing in rural schools. Low teacher salaries have been a demotivational factor, having deteriorated in real terms by more than 33 percent from 1983 to 1990. Since 1990, however, a concerted effort has been made to redress this situation and teacher salaries have risen by 18 percent in real terms. Teacher salary levels have doubled from the equivalent of 0.9 times the 1985 per capita GDP to 1.8 times the 1993 per capita GDP, and have risen 36 percent more rapidly than the minimum wage over the same period. Providing incentives for teaching in remote nural areas under PARE has yielded good preliminary results of teacher attendance as measured by the number of effective school days. Similar teacher incentives would be provided for the ten project states, relying on community support for monitoring the presence of teachers in classrooms (paras. 3.26-3.28). 2.10. Inadequate Supervision. Teacher supervision is intequent and highly inadequate in the poorest states. This results in: (a) little or no feedback on teacher performance; (b) no systematic follow-up on school improvement in student achievements; (c) weak control over teacher attendance; and (d) insufficient supervisors' input to timely textbook and materials distribution. The problem is especially acute in rural schools, which typically lack a principal, have the least experienced teachers, and are most in need of supervision. Inadequate supervision is a multifaiceted administrative problem, strongly influenced by the SNTE, and involving supervisors' training and difficult work conditions such as lack of transportation and adequate offices. The number of schools allocated to each supervisor (between 10 and 24) and their location in remote areas make frequent visits almost impossible. Shortages of transport and budgetry allocations for per diem costs further exacerbate the problem, as supervisors must use public transportation and pay their own expenses. As a consequence, isolated mrual schools are for the most part left unattended. There are no supervision offices, and supervisors must borrow working space in schools or other public buildings or work out of their own homes. The project would improve the supervision system thmugh specific traing, the creation of a broader supervision framework, the financing of traewl incentives, and the provision of vehicles and supervision offices (pam. 3.29). 2. 11. Shortages of Library Books and Other lUaching Materials. Support materials such as library books, maps, models, posters, school supplies, audiovisual aids and basic equipment for the natural sciences demonstrably increase educational achievement and attainment. The lack of school library books is problematic, since few mrral Mexican schoolchildren have access - at school or at home - to any reading material besides their schoolbooks. The lack of early reading stimulation can retard children's progress in developing reading and writing skills. The absence of additional educational materials is particularly serious among indigenous children who often do not speak Spanish, have less qualified teachers, and exhibit the worst education indicators. A recent preliminary assessment of the impact of PARE indicates that the Reading Corners program2' has not been sufficiently incorporated into classroom strategies, mainly because most teachers do not 2' The Reading Corns Program (P,gnma de Rincones de Lectura) has been oped since 1986 by SEP's Educauional Publications Unit (UPE) to provide studems and teachers with both an easy-access hlbra and the motivational tools to promote its regular use. The program sarted on a small scale but its concept is national in scope; the program is being expanded with support provided by Bank projects. -7 - know how to use them. ' The project would provide: (a) sets of educational materials for classroom and school use, includirg instructions for appropriate use of classroom materials, textbooks, and Reading Corners (paras. 3.15-3.17); (b) reference books for teachers (para. 3.17); and (c) textbooks and reading materials for indigenous schools (paras. 3.18-3.19). 2.12. Deficiencies in Textbook Production and Distribution. For over 30 years, Mexican children have benefitted from a national federally-funded textbook production system. Since 1959, the National Free Textbook Commission (CONAUTEG) has asserted the right of each student to own a full and free set of textbooks for all subjects of prnmary education. Production of textbooks (not under the project) has been impressive, with over 90 million low-cost, well-designed and -illustrated educational books being produced in 1992/93 alone. Textbook distribution, however, is a major problem, particularly in rural regions, since many poorer states lack the resources and administrative capability for timely delivery of textbooks. Since textbooks and teacher guides are all too often the only didactic materials students and teachers can rely on, deficient distribution diminishes the quality of education. To help correct these deficiencies, the project would strengthen the distribution system by providirg for the construction of regional warehouses in strategic locations and the devising of state-specific distribution plans and mechanisms (para. 3.30). 2.13. Dilapidated Physical Facilities. In 1992, a total of about 81,000 classrooms needed rehabilitation. In addition, there is a current shortage of classrooms, estimated at nearly 23,460 at the primary education level based on enrollments and projected demand, which represents a total investment cost of over US$660 million. While multi-shift schools could mitigate this type of shortage to a certain extent, such programs have proved difficult to implement in rural areas due to infrastructure limitations. In addition, many schools lack toilets, many classrooms lack sufficient desks, chairs, and other furniture and equipment, and schools often have no funds for basic physical maintenance. 2.14. For decades now, school construction has been under the responsibility of the Federal School Construction Agency (CAPFCE), a SEP body with strong technical capacity and extensive bidding experience. CAPFCE also manages and supervises most federally funded school construction. Under PARE, difficulty of access to mrual communities resulted in high unit costs of construction. The experience with the "Municipal Funds" approach sponsored by the National Solidarity Program (SOLIDARIDAD) under the Decentralization and Regional Development Project (Ln. 3310-ME, 1991) has led to reduced costs and increased local participation with only a small loss in quality. Under the proposed project, the diversity of geographic location would make it necessary to take advantage of both CAPFCE and participatory approaches to school construction (paras. 3.20-3.23). 2.15. Need for Greater Community hrtlcipation. Community participation in school- related activities enhances the community's responsibility for bettering academic performance and fostering a positive learning environment. Historically, a centralized bureaucracy has coexisted in Mexico with a strong tradition of community organization. However, two key issues remain: (a) the lack of structures for community participation in many areas; and (b) the failure to make full use of them in others. One of the geat successes of 101 Epeleta, J. et al. Phgrama wa Abaur el Reago_ Fucatiio. Evaldad6n Cuolitatva del Imn"ao. Mexico D.F.: Cent de Investigacid6n y de Esmwos Avanzados del Instito Polit*nico Nacional, 1993. -8 - SOLIDARIDAD has been to meld these two tendencies into a process of actual decentralization and effective community participation. Community participation, via community committees, Municipal Education Councils, parents associations, or School Councils (Consejos Escolares de Participaci6n SociaOU is an effective vehicle for social commitment and a catalyst for practical and regular involvement with the school. The creation of School Councils in each school with the joint participation of parents, teachers, and local leaders is expected to foster closer cooperation between local communities and their schools. The different types of entities participating in the project (authorities at the national, state and local levels, professionals in national and state institutions, teachers, and the communitty at large) need to be adequately and timely informed about project activities to develop interest and start participation in the project. The proposed project would enhance participation through: (a) enlisting Schools Councils to support and monitor teacher attendance; (b) involving communities in school construction (para. 3.20); (c) transferring resources for school maintenance to municipalities; (para. 3.22); and (d) implementing an intensive project promotion effort among authorities and communities (para. 3.25(d)). 2.16. Lack of Evaluation. Limited resources and technical capacity contribute to difficulty in monitoring and evaluating academic achievements and teaching standards across the system. SEP's General Directorate for Evaluation (DGEIR) is the only public institution undertalking evaluation research on qualitative and quantitative aspects of the education sector. However, such evaluation research is typically limited in its sample coverage, periodicity, scope, depth of analysis, and accessibility. At the state level, practically no capacity for evaluation exists. This reduces the eli-iency and jeopardizes the sustainability of educational investment programs. The result is a system in which the selection of the most productive investments and most effective r'fonrms are clouded by scarce understanding of the impact brought about by previous programs and policies. The project would foster the capacity for sustainable measurement of the quality of education both at the central and state levels, including a project process evaluation (para. 3.25('b) 2.17. Inadequate Managerial Capacity. State governments have assumed full responsibility for managing their basic education systems, which until 1992 were under the federal goverunent. Existing managerial abilities in the states, although generally satisfactory, vary from one state to another, and weaknesses are still expected durng the early years of decentralization. However, the state administrative structures are not entirely new, since for decades the federal government developed field offices at the state level (Serucios Coordinados de Educaci6n P(dlica, SCEPs), which gathered considerable experience in operating the state education systems. 2.18. The 1992 ANAMi formally transferred the administrative apparatus, the schools, their assets and staff to state administrations; redefined SEP's normative functions; and required a new SEP structure to comply with those functions. The 1993 LGE confirmed the distribution of responsibilities between the federal and state govenmnents. SEP's internal regulations and procedures (Reglamentos Internos) are still being revised to conform with the LGE, but the operation of the system in general and, in particular, the delivery of educational services in the states have not been hampered by their absence. LI School Counwils and Municipal Eduaion Councils were established by the 1993 WE. -9 - 2.19. Effective integration of the transferred primary education systems into the existing state systems presented practical difficulties at first. Differences in teacher salaries and benefits between the state and federal systems, separately negotiated with differtnt labor union chapters, made it very hard to unify the systems in most states. Nevertieless, the gradual equalization of state and previously-federal teacher salaries, started in December 1992, is nearly completed. Only three states have not finished the process, but are due to do so in 1994. Altogether, 427,000 teachers and education workers will have benefitted from approximately $160 million of salary supplements from federal and state funds. 2.20. Before May 1992, the SCEPs had been in charge of translating central policies into concrete actions in the field. Since that time, the SEEs (State Secretariats of Public Education or equivalent organizations) have been created to take over the educational systems transferred by the federal government and integrate them with the previously existing state systems. In most cases, the structure and staffing of the former federal SCEPs in the states formed the basis of the SEEs. These SEEs will need to devote more time to policy analysis and strategic planning. Federal authorities are providing technical assistance to the states to help them build up their institutional capacities through the SEP's General Coordination for Education Decentralization. SEP intends to provide this service on a permanent basis at the request of state governments and is establishing a resident representative office in each of the states. SEP, however, does not have sufficient qualified human resources to provide all the required adequate technical assistance, particularly in key areas such as policy analysis, educational planning, information management, distribution systems, and evaluation. The project would finance an integrated management training program in educational planning and policy analysis (para. 3.25(a)) and the strengthening of project monitoring, evaluation, and information systems at both federal and state levels (para. 3.25(b)). Efficiency4V 2.21. High Repetition Rates. Mexico is beset by problems of high repetition rates - a typical Latin Amencan proxy for school failure - leading to a costly and inefficient education system. In 1993, official estimates for primary education repetition rates in Mexico were 9.1 percent. Independent estimates, using a different methodology, place nationwide repetition rates at 16.5 percent for the first grade and at 50 percent in the poorer states.!3) Recent estimates suggest that these inefficiency costs amount to nearly 12 percent of total expenditures in primary education.Al The problem is further aggravated by wide repetition rate disparities between and within states. SEP data for 1993 reveal that repetition rates vary from 4.6 percent in the Federal District to 14.9 percent in Oaxaca (Annex 2), and 14 states have rates of 10 percent or more. Other sources suggest that repetition rates higher than 30 percent can be observed for the first two grades in many rumral schools. A 12 Within this section, there wil be mention of several alternative measurements of school efficiency levels, which likely provide a more objective view against an international and comparative backdrop. Nevertheless, official Mexican statistics are being used for the project, particularly for monitoring and other indicators purposes. 13/ Schiefelbein, E., Output of the SMMG Model with Data from the GDPPB. Santiago, Chile: UNESCO-OREALC, Siri System, June 1991. W' Carlson, S.: Transforming the Vicious Circle: The Costs and Savings of School lnefficiency in Mexico, A View from LATHR," No. 8: The World Bank, 1991. - 10- preparatory case study on two rural schools of Oaxaca revealed that the percentage of students who repeated at least one year over a seven-year period were 48 percent and 63 percent, respectively.1Lv 2.22. High Dropout Rates. Low internal efficiency in Mexico is also characterized by high rates of primary school dropout. Every year, about 0.7 million children (or 4.6 percent of enrollments) drop out from schools, half of them before completing the third grade, i.e., before achieving functional literacy. Again, there are wide disparities among the states, from 0.6 percent in Baja California to 13.9 percent in Chiapas; 11 states exhibit rates of 5 percent or higher (Annex 2). The Government is addressing this issue by: (a) improving the quality of teaching through upgrading of teacher skills, strengthening the supervision process, and increasing the supply of didactic materials; and (b) granting scholarships (Becas de SOLIDANDAD) in the poorest areas to a limited number of students (up to 8 per class, or 24 per school) based upon continued attendance. Since the latter approach has a high cost per beneficiary (about US$600 per student), the proposed project would support only the first element of this strategy (paras. 3.8-3.13), which has more widespread benefits. 2.23. Low Completion Rates. The net effect of these high repetition and dropout rates is a low rate of students completing the full six years of primary education. In 1991/92, only 58 percent of the students entering the first grade were likely to complete the sixth grade. Wide disparities in completion rates exist between states, ranging from 31 percent in Chiapas to 82.4 percent in the Federal District. This implies that access to the upper level of basic education (grades 7-9) also varies greatly between states. Children from the poorer southern states are only half as likely as those from the northern states to gain access to grades 7-9. Many of the factors responsible for low quality, such as teacher training and absenteeism, insufficient didactic materials, and incomplete schools (paras. 2.6-2.20) are responsible for these low efficiency rates. 2.24. High rates of grade repetition, as well as substantial numbers of students dropping out of school, reflect the low efficiency rates. Approximately 50 percent of the reasons can be attributed to socioeconomic factors that affect the lives of children in the project area, such as poor nutrition, the need for children to participate at an early age as members of the labor force, or migradon of the family in search of seasonal work for the heads of household. Other factors can be traced to educational conditions such as poor training of teachers, lack of adequate educational materials, teachers not speaking students' native languages, materials not available in native languages, and deficient technical supervision.L6' The lowest quality and efficiency indicators are observed in states with higher proportions of rural and indigenous populations. The completion rate of indigenous children for primary schooling is only 27 percent compared with 58 percent for all children, and their repetition rate is 42 percent compared with 9 percent for all children. Research has shown that improving the quality of instructional inputs can increase student learning and motivate parents to keep their children in school longer, thereby increasing the efficiency of the system. The project would finance investments in improving the efficiency of education through teacher training (paras. ' Myers, R., School Repetition and Dropout - Case Studies of Two Rual Schools in the State of Oaxaca. Mexico. Report prepared for The World Bank, 1993. 16' Ibid. - 11 - 3.8-3.13), updated educational materials (paras. 3.15-3.19), and improved supervision strategies (paras. 3.8-3.13 & 3.29). 2.25. 2 inancial Constraints and Patterns of Federal Education Spending. Both efficiency and quality in the education system are heavily affected by limitations in the availability of financial resources and by the way the federal government allocates resources to the states. Federal education spending declined sharply in the 1980s, but has recovered rapidly since 1991, evidence of the heightened importance the Government is placing on education as a core element of future economic growth. The federal education budget grew rapidly during the 1970s from 1.6 percent of GDP in 1970 to 3.1 percent in 1980, a period when real GDP growth averaged about 7 percent per annum. Since 1982, federal spending on education dropped from 3.7 percent of GDP (14 percent of the total federal budget) to 2.6 percent of GDP (12.1 percent of total federal budget) in 1986 as a result of the economic crisis, affecting all sectors. Real federal spending in the 1990s has regained pre-crisis levels, reaching an estimated 4.1 percent of GDP in 1993 (Annex 3). Compared with many other Latin American countries, Mexico has maintained low levels of education spending in relation to GDP, particularly in per pupil spending on primary education. Mexico's education expenditures as a percentage of GDP per capita averaged 4.1 percent in the 1980s, whereas those of Chile, Costa Rica, and Venezuela averaged 4.5 percent, 5.1 percent, and 5.5 percent, respectively. Although the situation is improving, financial constraints still hamper the education sector, particularly in the provision of sufficient teaching materials and adequate salaries for teachers. Given the high level of Government commitment to education, it is anticipated that there will further, albeit small, increases in the percentage of GDP and of the federal budget allocated to education over the next few years, plateauing to a level of about 4.5 percent and 13.0 percent, respectively. Such expenditure increases, along with the positive policies adopted by the Government to improve quality (para. 2.35), are anticipated to make a major positive impact on school achievement. 2.26. Unbalanced Educational Expenditures. The repercussions of the deep budget cuts in the education sector in the mid-1980s will remain for the medium-term. These budget declines were accompanied by an increase in total enrollments, teaching staff, and number of schools, resulting in drastic reductions in real per pupil expenditures. Between 1982 and 1990, federal spending for primary education decreased in real terms at an average annual rate of 1.2 percent, while enrollments increased at an average annual rate of about 5 percent. In censtant 1980 pesos, spending per pupil decreased by 47.4 percent in that period, resulting in a scarcity of teaching materials and a reduction in the reMal value of teachers' salaries of nearly 40 percent. The detenoration of the education budget is also reflected in the inbalance between investment and recurrent expenditures. Almost 95 percent of federal spending now goes to recurrent costs (personnel and operating costs), up from 81 percent in 1980. Only about 5 percent of the budget is invested in buildings, teacher training, education materials, and classroom aids (Annex 3). 2.27. Inadequate Intra-Sectoral Allocation. There has been a long-term and increasing bias in favor of higher educaton expendiures. In fact, the index of public subsidizationLy 12' The index of public subsidization meaures the percentage by which the private rate of rturn exceeds the social rate of return. h G. Psacharopoulos and Y. Ng, Earnings and Education in Latin America. Assessing Priorities for School Imestent. World Bank Policy Research Worldng Paper 1056, 1992. - 12 - shows that higher education is the most heavily subsidized level of schooling (41 percent) and primary education the least subsidized (12 percent). Government policies addressing educational financing and allocation of spending are outlined in its Policy Letter (Annex 4). Equity 2.28. Allocation of Federal Government Expenditure. Public educational resources have traditionally been allocated by the federal government, accounting for over three-quarters of the actual educational expenditures, with private financing and contributions from states' own resources providing the balance. The percentage of expenditures from the federal government dropped from 86.3 percent in 1985 to 78.2 percent in 1991, whereas the state expenditures increased from 8.7 percent in 1985 to 16 percent in 1991 (Annex 3). Because of the national salary policy, which divides the country intO three zones to reflect cost-of- living differences, unit costs have been higher in the richer states: in 1992, they were N$1,154 per student in the non-project states compared with N$948 in the project states. This explains why the ten proposed project states, which account for 41.0 percent of national primary education enrollment in federally-funded schools, received only 38.2 percent of the federal allocation for primary education in 1993. Since these poorer states are less able to supplement federal allocations from their own resources (Annex 5), the Government has introduced compensatory education programs as a key element in reducing inequity in the education system by channeling more resources to the poorest states. The project states were chosen based on selected educational and poverty indicators, which make up a marginality index (Annex 6), and include San Luis Potosi, Veracruz, Michoacan, Puebla, Jalisco, Guanajuato, YucatAn, Durango, lTbasco, and Campeche (para. 3.3); the states of Chiapas, Oaxaca, Guerrero, and Hidalgo (covered under PARE) would also benefit directly from management training programs for SEEs (para. 3.25(a)) and indirectly from other institutional strengthening programs at CONAFE (paras. 3.25(b)-(d)). Primary education statistics for the project states are included as Annex 2. To ensure that federal allocations are maintained at a level in line with the compensatory scheme, monitoring of both federal and state allocations to primary education would be tracked on a regular basis and reviewed on an annul basis (paras. 6.1 & 7.2(v)). 2.29. Difficult School Access. Poverty not only causes low educational efficiency, but also limits access to school for children in rural and indigenous areas. In primary education, the net enrollment rate has remained at 98 percent for the past five years, but this nationwide figure disguises regional pockets of low matriculation. In the state of Campeche, for example, 30 percent of the schools are incomplete, i.e., they cover only the first three or four grades of instruction. Moreover, on a national basis, 0.4 million children still have no access to primary education at all, and a further 0.7 million children drop out of school each year. The population still without access to primary education is characterized by extreme poverty, geographic isolation, and early entry into low-paying positions in the labor market. The proposed project would increase access to prinary education for poor children living in the most deprived and remote municipalities within the ten states, particularly those children in indigenous schools (e.g., 91 percent of indigenous schools in the state of Veracruz would be covered), and schools temporarily without teachers (paras. 3.3-3.6). 2.30. Lack of Initial and Preschool Education. Both initial and preschool education increase the reurn on pnmary and secondary school investments and raise participant - 13 - productivity and income levels. In 1991, the total number of children benefitting from initial education in Mexico represented 2.8 percent of this age group. Overall, preschool services were offered to 2.8 million children in 1991-92, only 10 percent of whom lived in rural or indigenous areas. To respond to these needs, in 1992 the Bank supported PRODEI to help Mexico develop non-formal education for 1.2 million poor children by training 910,000 parents (mostly mothers) in the ten poorest states in the care and education of their young (aged 0-3) children. The Bank is supporting preschool education in the proposed project through a pre-investment study in preschool education (para. 3.25(c) & Annex 14, Part A). The study would be completed within the first two years of project implementation and would be reviewed jointly with the Bank. 2.31. Shortage of lextbooks for Indigenous Schools. Even though Mexico has the greatest mnuber of indigenous population in Latin American (between 7 and 12 million, according to estimates), quality education for these groups has received only superficial attention. The bilingual approach applied in indigenous schools has not been adequately supported by national programs. Because of their language and cultural characteristics, indigenous children have found themselves in a disadvantageous learning situation whenever they have been exposed to books and other reading materials designed to be used by Spanish- speaking children. Annex 7 provides statistics on Mexico's indigenous population and on educational services currently being provided to them. The proposed project would finance the production of textbooks and reading materials for the indigenous primary schools of 17 ethnic groups scattered throughout the ten project states (paras. 3.18-3.19). C. Government Strategy for the Education Sector 2.32. To address these issues, the Government has implemented over the last five years a comprehensive reform of the education sector. A national plan was launched in 1989 to ransform the education system; in 1992, the system was decentralized to the states; and in 1993, a national teaching career program was instituted, mandatory nine-year basic education established, and the LGE passed. These measures have been buttressed by a budget recovery for the sector (para. 2.25). The Government's sectoral policies are outlined in a Policy Letter, which is in Annex 4. At negotiations. the Govenment presented a final Policy Letter for the education sector (para. 7.1(a)). 2.33. An overall policy statement on education was made in 1989 by the incoming administration to embody the commitment to build up Mexico's human capital. This document, the Education Modernization Program 1989-94 (PME), was later incorporated into the National Development Plan 1989-1994 and identified and set the basic framework for taclding pressing issues of national education: decentralization, need for quality and modernization, linkages between schooling and production, community participation, scientific and technological advances, and need to increase the education budget. The PME had four overriding objectives: (a) to decentralize the system; (b) to improve the quality of education; (c) to raise the level of schooling throughout the population; and (d) to strengthen community participation. Additionally, the need to increase resources for education and the commitment of federal and state governments to commit resources thereto, within available resources, nav been clearly expressed dhrough both the PME and the LGE. - 14- 2.34. Decentralization. Basic education was decentralized to the states in May 1992 through the ANAM, which transferred from the federal government to the states the management of initial, preschool, primary, lower secondary, and special education and teacher training. To carry out their new.mandate, the states received all related facilities and equipment and technical and administrative elements of the educational system as well as corresponding budgetary allocations. An integrated model %-as implemented to address not only the structural framework of the system, but also issues of curricular contents (para. 2.7), re.-aluation of the teaching function (paras. 2.8-2.9), community participation (para. 2.15), and the nonnative (para. 2.3) and compensatory (paras. 2.3 & 2.28) roles of the central Government. 2.35. Quality of Education. The Goverunent's strategy to improve the quality of education incluaes three elements: (a) upgrading of teachers; (b) updating of curricula; and (c) provision of instructional materials. To support excellence in classroom teaching, foster teacher interest in professional improvement, and offer better salaries based on refined performance evaluation, the national Teaching Career Program (ICP) was launched, later to be enhanced by an in-service teacher training strategy widening performance standards for teachers to include class planning and evaluation activities, teaching strategies, and community interaction. A second quality-related input involves the redefinition and updating of the stuidy programs. The new curriculum, which was confirmed by the 1993 LGE, stresses a back-to-basics approach, emphasizing reading, writing, and practical math, with history and geography replacing the previous integrated 'social sciences" discipline. The third quality enhancement element is the maintenance and upgrding of the nationwide free textbook program, strengthened by the "Reading Corners" program (paras. 2.49 & 3.17), and the provision of teaching materials and students' packages (paras. 3.15-3.16) and indigenous textbooks and materials (paras. 3.18-3.19). 2.36. Schooling Level in Society. The Government is seeking to raise the general schooling level by: (a) facilitating school access to wider population segments; (b) prolonging mandatory school services; and (c) increasing students' retention and their learning achievements. To reach broader social segments, the SEP is extending a wide range of compensatory education progrms targeted at underserved populations through PARE (paras. 1.14, 2.42-2.50 & 2.59) and PRODEI (paras. 1.14, 2.51 & 2.59), bilingual education programs, and the CONAFE model (para. 2.4). The proposed project would expand the Government's compensatory program to reach an additional 1.5 million children in disadvantaged communities. To augment the length of learning experience, the Government (i) increased the school year from 187 to 200 days and (ii) established a new nine-year basic education mandate. This commitment to higher levels of basic education stems from results of econometric studies which suggest that length of schooling plays a significant role in explaining cross-country growth differentials.A8' The cost implications of the proposed reform would be analyzed under the project through a study (para. 3.25(c) and Annexes 9 & 14), but the Goverrnent does not expect any significant additional budget commitments due to existing unused capacity in secondary schools. To increase retention of students in the La' In January 1993, compulsory and free nine-year basic education was enacted by integrating the six- year primary education and the three-year lower secondary cycle, consistent with with UNESCO recommendations and the policies followed by other Latn American counties at a similar developmental - 15 - school system and raise their academic achievements, the Government is undertaking necessary measures to enhance the quality of the teaching-learning experience through teacher and administrator training (paras. 3.8-3.13) and provision of updated books and educational materials (paras. 3.14-3.19). 2.37. Community Participation. To make national education more democratic and relevant to social promotion, the Government is reorganizing the education system to involve communities in various aspects of school life, drawing the commitment of teachers, parents, students, education authorities, and community members and leaders. Since the PME, and mainly through the ANAM and the LGE, principles were stated, operational strategies recommended, and mechanisms created to provide local communities with sufficient structure and support for effective participation in school matters. "Councils of social participation" were created at the community, municipal, and state levels, respectively, and are starting to contribute valued input into the decision-making process. 2.38. Increasing Education Resources. Both the PME and the LGE express the Government's commitment to increase financial resources for primary education and provide special assistance to keep rural children in school. Recent education expenditures have confirned this commitment, and it is expected that this level of commitment will at least be maintained in the foreseeable future (para. 2.25 & Annex 3). Review of federal and state budget allocations would occur during the proposed project's annual and mid-term reviews (paras. 5.4-5.5 & Annex 20). I). Bank Role and Experience in the Education Sector 2.39. Prior to the 1990s, the primary focus of Bank assistance to the education sector was on technical training, which reduced the shortages of skilled technicians, long considered a major constraint to industrial growth. Through the First, Second, and Third Technical Training Projects (Lns. 2042-ME (1981), 2559-ME (1984), and 3358-ME (1991), respectively), the National System for Vocational and Technical Education with 252 vocational training centers and an enrollment of approximately 300,000 students was established, and about 40 mobile training units were constructed to provide practical training to conmnunities in marginal areas. The Manpower Training Project (Ln. 2876-ME, 1987) supported retraining for the unemployed and in-service training for workers in small enterprises. These projects have met their objectives and targets, and there have been no notable implementation problems. These activities are being consolidated by the Labor Market and Productivity Enhancement Project (Ln. 3542-ME, 1992). 2.40. Following the 1985 earthquake, the Bank financed the reconstruction, rehabilitation, and reinforcement of public schools in five states (Colima, Guerrero, Michoacan, Chiapas, and Oaxaca) and in Mexico City through the Earthquake Rehabilitation and Reconstruction Project (Ln. 2665-ME, 1986). Under the Decentralization and Regional Development Project (Ln. 3310-ME, 1991), the Bank is supporting rehabilitation of 14,328 schools and reconstruction of 2,006 classrooms and 2,540 annexes for poor and indigenous populations in the rural areas of the same four states covered by PARE, also supported by the Bank. - 16 - 2.41. In the 1990s, the Goverunent requested major Bank support for primary education. The 1991 PARE and the 1992 PRODEI, currently under implementation, seek to increase education quality through a variety of targeted interventions in the states with the worst educational indicators. With recent emphasis placed on Government policies for basic education, the Bank can make a significant contribution to human resources development by addressing the critical issues of quality, efficiency, and equity in primary education. E. Lessons Learned from Past Operations 2.42. Relevant lessons from the implementation of ed ication projects in Mexico, chiefly PARE, but also PRODEI, are discussed in this section. 2.43. Compensatory Projects and Education Quality. The compensatory PARE led to an increase in primary education expenditures for four of the poorest states by 2.2 percent in 1991, 3.7 percent in 1992, and 24.3 percent in 1993 when project implementation accelerated substantially. In particular, non-personnel expenditures per student increased by 6 percent in 1992 as a result of the compensatory programs, and by 305 percent in 1993, contributing to a subsequent quality improvement of primary education in those states.)9 While the education impact on students cannot be measured reliably over the short-term, there have been reductions in repetition and dropout rates in the PARE states. In Hidalgo, for example, the repetition rate declined from 12.4 percent in 1990/91 to 9.4 percent in 1992/93 and the dropout rate from 4.1 percent to 3.6 percent, resulting in an increase in the terminal efficiency rate from 58.7 percent to 67.0 percent over this period. Based on these positive results, the proposed project would cover the next ten poorest states in Mexico, channeling the resources mostly to non-personnel costs, which would be inshumental in improving the quality of primary education in those impoverished states; the four PARE states would also benefit from the management capacity building subcomponent. By 1998/99 the project states' share of all SEP primary education expenditures would increase fim 38.3 percent in 1993 to 40.3 percent, and the expenditure per student in the project states would increase rapidly, thereby reducing the gap between richer and poorer states. 2.44. ITrgeting. The first project targeted four of the economically and educationally poorest states. However, since even the poorest states were found to have municipalities at different levels of development, the efficiency of investment is being further maximized by focusing most project activities not only on the next ten poorest states, but also on the 511 poorest municipalities (out of a total of 934) within these ten states ("project areas") (paras. 3.3-3.6). 2.45. Decentralization. PARE has been carried out withdthe support of state Coordinating Units, which have played a successful role in implementation. Based on that experience, the proposed project would expand support to decentralization by: (a) involving the participant states' SEEs in the project preparation process; (b) creating state-level Coordinating Units (SPCUs, Unidades de Coordinaci6n Estatal) for project implementation and monitoring (para. 3.32); (c) providing technical assistance and human resources development to 12' The non-personnel federal expenditures per student per year increased from US$14.26 without the project to US$22.98 with the project in 1991, and most dramatically from US$13.99 without the project to US$56.76 with the project in 1993. - 17 - strengthen the SEES in the areas of planning and policy analysis, educational management information, and evaluation systems (para. 3.25); and (d) empowering School Councils to carry out school construction and maintenance (paras. 3.20-3.23). 2.46. Teacher Training. The PARE has shown that its teacher training courses were too short and needed to be reinforced with practical in-service supervision. Consequently, under the proposed project, the courses would be expanded in length and content and follow-up technical assistance in classroom performance has been designed to provide for a more active role by supervisors and principals after the formal training period (paras. 3.8-3.13). 2.47. Teacher Incentives. The evidence from PARE suggests that financial incentives for teachers serving in the most remote rural schools has been successful in many ways. Teacher absenteeism has been significantly reduced, the number of teachers willing to fill teaching positions in isolated schools has increased, and teacher turnover has been reduced to a minimum. Attaching the incentive to the school rather than to the teachers has succeeded in promoting teachcr stability. Community cooperation in helping oversee teacher attendance has increased parental commitment to other parts of school life, such as transportation of educational materials to the school and the use of the Reading Corners. Under the proposed project, the teacher incentives program would be extended to the project areas, focusing on schools without teachers and on multirade and indigenous schools (paras. 3.26-3.28). 2.48. Instructional Materials. The provision and use of instructional materials (such as paper, pencils, maps, games, and globes) under PARE has contributed to: (a) more active learning in those poor schools; (b) more cooperative classroom work; and (c) more productive use of teachers' time in the classroom. Some materials of poor quality had to be replaced, and teachers had to be trained to use the materials adequately. The proposed project would: (a) expand the provision of instmctional materials to the project areas; (b) carry out a more careful selection and design of didactic materials; and (c) train teachers on how to get the best educational results from these materials (paras. 3.15 & 3.17). 2.49. Reading Corners. From PARE, it was learned that the Reading Corners program can: (a) complement teachers' strategies; (b) stimulate students' interests and reading habits; (c) expand the benefits of selected reading to poor families; and (d) support teachers' continuing education through the provision of reference materials. The proposed project would: (a) strengthen the Reading Corners in the project areas; (b) increase the number and selection of titles, including books with regional content; and (c) introduce reference and technical materials to support teachers' classroom and community activities (para. 3.17). 2.50. Indigenous Texbooks. The development and production of new textbooks for indigenous schools under PARE yields two main lessons: (a) the participatory strategy used by the General Directorate for Indigenous Education (DGE) in developing these materials through close dialogue with indigenous leaders, teachers, schoiars and parents has been highly successful and unanimously supported; and (b) the visual improvement of the textbooks by using indigenous colors and iconography has increased communication effectiveness. The proposed project has incorporated these two features in the design and production of textbooks and other instructional materials for 17 additional ethnic groups in the project areas (paras. 3.18-3.19). - 18 - 2.51. Environmental Education. PRODEI has indicated that environmental education can and should be promoted from the early years of a child's life through: (a) the development of good health and hygienic habits; (b) consideration for the micro-enviromnent, particularly the domestic environment; and (c) the formation of a positive attitude toward natural resources and public facilities. The proposed project would finance an assessment of enviromnental education at the basic education level in Mexico and offer recommendations that could lead to further action (para. 3.25(c) & Annex 14, Part C}. 2.52. Educational Materials Distribution System. A preparatory study concluded that textbook distribution has been hampered by the absence of: (a) a comprehensive and participatory plan of delivery to schools; and (b) adequate transportation from warehouses to schools. In addition, the design of regional warehouses built under the first project needed some adjustments to better perform their required functions. Accordingly, the proposed project would: (a) require a logistical distribution plan to be developed by each state, including the use of state or municipal transportation along with community-based initiatives; and (b) provide vehicles for supervisors to use for both supervision (para. 3.29(d)) and distribution of educational materials (para. 3.30). 2.53. School Construction Models. A preparatory study has indicated that construction costs have been higher in isolated rural communities and that technical supervision has been insufficient, thereby undermining the quality of construction. Furthermore, construction costs have been higher than expected and construction of schools in remote rural areas under the Decentralization and Regional Development Project via the SOLIDARIDAD municipal funds (a strategy not applied under PARE) was able to mobilize community contributions and reduce construction costs. The proposed project would therefore: (a) maintain PARE's strategy for urban areas and more readily accessible rural areas, but with closer cost-control instruments; (b) introduce a decentalized and participatory approach modeled after the SOLIDARIDAD experience for more remote rural areas; and (c) strengthen the technical supervision of construction by local CAPFCE and private supervisors (paras. 3.20-3.23). 2.54. Project Implementation Experience. The General Directorate for Planning, Programming and Budgeting (DGPPP) under which the Project Coordinating Unit (PCU) operated had limit2d experience in working with Bank projects, thus there were some delays in PARE start-up. Now, the coordinating team, operating under CONAFE's direction, has learned to successfully coordinate, monitor and evaluate the activities of PARE. For the proposed project, the experienced PCU team would be retained and complemented. In addition, specific staff training has been planned to furtier improve their skills (paras. 3.25(a) & 3.31-3.33). Bank-Wide Education Projects 2.55. Project Design. Feedback from several education projects in Latin America indicates that a proper policy framework and adequate institutional capacity are major factors determining the success of project implementation. The proposed project, based on experience gained from PARE, focuses on improving the quality and efficiency of primary education on the basis of (a) a clearly stated Govermment policy framework and (b) improvement of the institutional capacity of central and state education entities. - 19 - 2.56. Project Management. When a project implementation unit is created separately from the existing organizational units, existing units often miss the opportunity to acquire technical expertise during the process of project implementation and to strengthen their institutional capacity. Moreover, externally-financed projects and the overall national program can develop simultaneously without proper synchronization. On the basis of such lessons, the proposed project has been explicitly designed to fit within the overall national program, and most project activities would be implemented and supervised by the SEEs through their existing operational structures, in coordination with the PCU and some central SEP units. F. Rationale for Bank Involvement 2.57. The Bank's most recent Country Assistance Strategy (CAS) for Mexico dated May 18, 1993 focuses on strengthening human resources and enviromnental protection and stimulating pnvate sector-led growth. The CAS was presented to the Board with the Highway Rehabilitation and Traffic Safety Project (Ln. 3628-ME, 1993) on June 24, 1993. The proposed Second Primary Education Project, PAREB, is consistent with the CAS. After the successful implementation of Mexico's economic adjustment program, the Bank has shifted its emphasis to investment lending geared towards assisting sustained growth and poverty alleviation programs. To support growth, the Bank will help the Government enhance the country's institutional capacity for efficient allocation and use of public resources. To support poverty alleviation programs, the Bank will assist the Government's efforts in targeting the poor for public resources. The Bank is supporting major initiatives in the human resources sector aiming at improving the quality and efficiency of social services, the labor market, and nutrition (in the near future) through projects and well-targeted sector work. Future proposed projects in education would focus on preschool and secondary education and on the modernization of technical training, backed in all cases by sector work. This proposed project represents the analytical foundation for this additional support in education. The Bank, with its long involvement in Mexico and with its experience in other countries, is in a unique position to play a catalytic role in: (a) increasing investments in primary education, thus helping to accumulate human capital, the lack of which is a serious constraint to growth; and (b) focusing on the most disadvantaged states and areas within these states, thus addressing the needs of the poor and contributing to a more equitable distribution of educational and economic opportnities. 2.58. The Bank has acquired significant experience in assisting primary education worldwide and has become the single most important source of external financing for the primary, education sector in Mexico. The Government has shown interest in involving the Bank more closely in policy and program development in the human resources sector in general and has requested the Bank's support for the basic education sector in particular. 2.59. By focusing on the poorest states, the proposed project would complement a set of Bank-financed projects targeted on the poor, including: (a) the 1990 Basic Health Care Project (Ln. 3272-ME) to extend health care services to 13 million uninsured poor in four of the poorest states (Chiapas, Oaxaca, Guerrero, and Hidalgo) and in the Federal District; (b) the 1990 Low-Income Housing Project (Ln. 3140-ME); (c) the 1990 Third Technical Traninmg Project to provide vocational training to adolescents from poor families, and unemployed; (d) the 1991 Decentralization and Regional Development Project to improve the - 20 - irdhastructure and productive capacity of rural communities in the four poorest states; (e) the 1991 Water Supply and Sanitation Sector Project (Ln. 3271-ME), to improve the living conditions of 24 million people, most of whom are in the lowest segments of society; (f) the 1991 Primary Education Project (PARE), to improve the quality and efficiency of primary education in the four poorest states; (g) the 1992 Initial Education Project (PRODEI), to support the early development of 1.2 million children in the ten poorest states of Mexico; and (h) the 1992 Labor Market and Productivity Enhancement Project to retrain unemployed workers, upgrade unskilled workers, and improve employment services for the poor. M. THE PROJECr A. Project Concept and Objectives 3.1. The objective of PAREB is to assist the Governmern in improving the quality and efficiency of primary education in ten southern states with a high incidence of poverty and low educational indicators (Campeche, Durango, Guanojuato, Jalisco, MichoacAn, Puebla, San Luis Potosi, Tabasco, Veracruz, and Yucatfin) in order to reduce their educational deficiencies and bring them closer to the national average. It would also reinforce management capacity in the four states (Chiapas, Oaxaca, Guerrero, and Hidalgo) covered by PARE. The proposed project draws on the experience of PARE and would support the Government's policy of providing compensatory assistance to the less developed states. 3.2. The expected improvement in quality and efficiency of primary education would be reflected in: (a) a decrease in repetition and dropout rates; (b) an increase in the level of children's cognitive achievement; and (c) a more effective management of the primary education system. These are high-priority objectives of the Government (paras. 2.32-2.38 & Policy Letter (Annex 4)), to be pursued by the states in the context of the decentralization policy (paras. 2.33-2.34). B. Project Areas and Tlrget Populaion 3.3. The ten states included in PAREB were selected based on three indicators: (a) the CONAPO (National Population Council) poverty index;20' (b) the primary education completion rate as a proxy for internal efficiency;A and (c) the indigenous education rates. Out of the ten project states, all are below the median and the mean of the CONAPO index (Annex 6) and all but one have primary education completion rates lower than the national 20/ The CONAPO index is a poverty index based on the results of 'principal components analysis", using 1990 census data and composed of several municipal indicators: (a) population; (b) percentage Of population 15 years or more who are illiterate; (c) percentage of population over age 15 without primary school education; (d) percentage of households without electricity; (e) percentage of overcrowded households; (f) percentage of occupants in households without a constructed floor; (g) percentage of population in localities with fewer than 5,000 habitants; and (h) percentage of population with fewer than two minimum salaries (about N$14 on countrywide average per minimum daily salary). 211 The school completion rate for primary education was chosen because it results from a number of factors affecting educational performance, including high repetition and dropout rates, a large number of incomplete and single-teacher schools in remote rural areas, inadequacy and shortage of educational materials, and poorly trained teachers. These educational deficiencies largely parallel weak economies and inability of the states to channel the resources necessary to improve education. -21 - average (Annex 2). In 1991, the ten states made up 39 percent of the population aged 6-11 in the country, 39 percent of the total primary enrollment, and 38 percent of enrollment in then federally operated (and now state) primary schools. lb reenforce the management capacity of the four states under PARE (which encountered some start-up difficulties due to institutional weakmesses, see para. 2.54), their managers would also benefit from the management training subcomponent. 22 3.4. Within the ten project states, project areas were selected on the basis of the municipality's poverty and educational indicators. In line with the national policy of basic education decentralization, each participating state conducted a diagnostic study to determine which areas within the municipalities should be included in the project. As a result, the project targets the neediest people living in areas where economic conditions are most severe and educational marginality most critical. A total of 511 municipalities (54.7 percent) w--re thus selected out of 934. The selected municipalities comprise 56 geographic regions, including 29 regions of high sierras and deep tropical valleys or semi-arid plateaus where indigenous peoples predominany live. State capitals and most major urban areas were excluded. Approximately 97 percent of project areas are poor, rural, and isolated villages of fewer than 1,000 inhabitants, lacking basic services like water, sewerage and electricity, and where the average family income is less than the minimum wage. By targeting the poorest areas, the project would also indirectly concentrate project activities in areas of high indigenous concentration (see Box next page). 3.5. The project would cover 13,858 schools, or 38 percent of the primary schools in the ten states. Approximately 1.5 million children are enrolled in the project schools, representing 27.5 percent of primary school enrollment in the project states, or 10.4 percent of the national primary school enrollment. The selected schools are served by 52,859 teachers, representing 29 percent of the teaching force in the ten states, or 11 percent of the national teaching force at the primary education level (Annex 8). 3.6. Poverty, usually associated with poor health and nutrition, also correlates with a low level of education. Average adult illiteracy in the project states is estimated at 13.7 percent, higher than the national rate of 11.8 percent. Average primary school completion rate for the ten states is 52.6 percent; dropout rates average 5.5 percent; and repetition rates stand at 18.7 percent, all worse than the national averages of 58.1 percent, 4.6 percent and 17.1 percent, respectively (Annex 2). C Project Components 3.7. The project has been organized into three major components, as follows: A. Human Resources Development (US$107.9 million or 17.5 percent of total costs including contingencies), to provide in-service training and upgrading for classroom teachers, principals, supervisors, and selected technical personnel; B. Educational Material Resources (US$324.2 million or 52.5 percent of total costs including contingencies), to provide good educational materials and facilities, 221 "Project staes" in this report will refr strictly to the ten staes previously identified (pama. 3.1). - 22 - INDIGENOUS PEOPLE AND THE PROJECT Mexico has the largest indigenous population in Latin America, approximately 8.7 million people (over three times the popuation of Costa Rica), corresponding to 10.4 percent of the national population. The indigenous population is massively occupied in the primary sector (76 percent compared to 22.6 percent counrywide) and geographically concentrated, with 78 percent living in one-third of the country's 2,403 municipalities. Being of indigenous origin is synonymous with backwardness and poverty. Educational indicators have remained significantly lower in indigenous than in non-indigenous areas of Mexico as shown below: Education Indicators - General vs. Indigenous Populations (%) Primary Education (PE) Adult Educ. Attainent Preschool Enrollment Enrollnent Repetition Completion % With 6- Adult Rate Rate Rate year PE Illiteracy General 67 98 9 58 68 12 Population _ Indigenous 41 72 42 27 14 46 ,Population ____ _ Indigenous schools suffer from specific shortcomings: (a) the absence of culturally- oriented textbooks and appropriate educational materials, a particularly serious problem among indigenous children who do not speak Spanish; and (b) 90 percent of indigenous teachers do not possess the officially required licenciatura. The project wtuld devote US$125 million (or approximately 22 percent of total project cost) to support the education of indigenous children by providing for teacher training and incentives, textbooks, didactic materials, Reading Corners, and school construction/rehabilitation and maintenance. Indigenous populations are targeted for project interventions in the ten project states which, together with the four PARE states, comprise 81 percent of Mexico's indigenous population. Using a more explicit targeting mechanism weighted heavily with poverty and indigenous population indicators, the project would benefit, on a relative scale, indigenous areas more. The teacher training courses would enroll an estimated 31,900 indigenous teachers, or 10.5 percent of all project teacher training instances. All indigenous schools in the project areas will receive the teacher incentives program, and 43 percent of the incentive recipients would be indigenous teachers. These same schools would also benefit from construction/rehabilitation and maintenance programs, and DGEI would develop native educational materials for the Reading Corners in indigenous schools. Textbooks and reading materials would be developed in the 17 main native languages spoken in the project areas: Chol, Chontal, Huasteco, Huichol, Maya, Mazahua, Mexicanero, Nahuatl, Purepecha, Tepehuano, Totonaco, three variants of Nahuatl, and one each of Huasteco, Maya, and Totonaco, covering about 90 percent of all ethnic groups in the project states. Book contents would be gathered through ethno-linguistic field research conducted through a participatory strategy successfully developed under PARE; about 126,000 textbooks, 126,000 student workbooks, and 5,000 guidebooks wrould be printed. The project would thus add to the eight languages for which materials were developed under PARE. Indirecdy, by developing technical know-how at DGEI, the project would also assist the production of textbooks and materials in 21 other languages that DGEI is working on in other states; eventually, indigenous textbooks will have been produced for 46 of the country's 56 main indigenous groups. - 23 - namely: (i) teaching materials, school supplies, and school library books both Spanish and bilingual (US$190.3 million); and (ii) urgently needed rehabilitation or replacement of approximately 6,900 classroom and other school facilities (US$131.7 million); and C. Institutional Strengthening (US$184.5 million or 30.0 percent of total costs including contingencies), to strengdten the management capabilities and skills of educational organizations at the national, state, and local levels. It would: (i) strengthen the management capacity of SEP and of SEEs by supporting improvements in planning and policy analysis and in the monitoring, evaluation, and information systems, conducting selected studies to help design future education programs, and designing and carrying out project promotion; (ii) provide work incentives for approximately 7,000 teachers serving in isolated areas; (iii) support a more effective educational supervision system; (iv) strengthen the textbook delivery and materials distribution system; and (v) finance the project administration at the central and state levels. D. Project Description Component A. Human Resources Development (US$107.9 million including contingencies) 3.8. The objective of the human resources development component is to produce more effective teaching and learning conditions in rural schools by enhancing specific teacher skills and tying teacher career advancement to objective performance criteria and by introducing an integrated pedagogical foundation to educational supervision. The contents of the training courses have been selected to fill areas of expertise not covered by the teacher education curricula in colleges and Normal Schools, but deemed critical to teaching under the special circumstances of project schools. The training program has borrowed from the experience of PARE and its components have been accepted by the PAM program, which intends to apply similar training concepts to other states. Teachers who graduate from the training courses would be eligible for the '1CP career advancement program (Annex 10 provides an in-depth description of the teaching training strategy). 3.9. Training Program. The training program would be administered to all project teachers, principals, supervisors, and assistant supervisors. Each program participant would be required to take a five-day summer program (40 hours) consisting of the three core courses - Educational Resources, Educational Problem Prevention, and Remedial Education - plus a number of other courses over seven one-day workshops (a total of 42 hours), approximately once a month on a Saturday over the following school year. Teachers would be required to take one state-specific course, either Written Language or Mathematics (depending on their specialty), and, where applicable, Multigrade Class Teaching (for new teachers likely to be assigned to a multigrade school) and a choice of one or more indigenous teaching courses (for teachers in indigenous areas): Bilingualism in Indigenous Education, Spanish as a Second Language, or Use of Indigenous Texts and Materials. Emnphasis would be on the teaching requirements of the first and second grades where the highest repetition and dropout rates are found. For administrators, a course in Pedagogical Counseling would be required of supervisors, principals, and technical assistants, while supervisors would take an additional course in Supervision. Educational supervision would be pushed beyond the -24 - present concern with administrative control. To this end, an innovative "pedagogical network" concept would be applied throughout the training courses and follow-up activities, thereby training principals and supervisors to provide teachers with guidance on sound educational principles and effective classroom techniques. Teacher trainers would be selected primarily from well-performing graduates of this group, principals, and supervisors. 3.10. Each year, the training program participants would be selected according to the priority needs of the area in which they work, using the same index used to target the 511 project municipalities (paras. 3.3-3.4); it is expected that all project participants would be trained by the fifth year of the project. This approach provides further incentives for teachers and administrators to work in the most disadvantaged localities by providing them the means for earlier career advancement under the TCP. Training centers would be housed in either an existing Normal School or in a secondary or technical vocational school with a wide geographical distribution to minimize travel and lodging costs. No additional physical infrastructure would be required for the training centers, but the project would finance required computers and audiovisual materials and equipment. The Govermnent's list of selected training centers was reviewed at appraisal and found satisfactory. The project would also finance the costs of transportation and meals, training materials, and remuneration (equivalent to overtime pay) for trainers, supervisors, and school principals for participation in the Saturday workshops. 3.11. The training program would operate under the same steering unit in the Undersecretariat for Basic Education (Subsecretaria de Educaci6n Bdsica, SEB) which oversees the PAM to avoid course content duplications and to ensure that credit points for successful program completion are properly recorded under the TCP. The summer courses would be integrated with a regular PAM one-week summer workshop to update curricular contents. A Central Technical Group attached to SEPs normative units for general and indigenous primary education would be responsible for overall planning of training activities, designing the courses, and developing training materials. This group would, in turn, be responsible for training the State Technical Groups and for following up and evaluating all project training activities in the states. State Technical Groups would similarly be organized in project SEEs for general and indigenous primary education, respectively. The State Technical Groups would be responsible for organizing the training courses for state trainers, teachers, and supervisors, and for evaluating the performance of trainers. Both the central and state Technical Groups would be made up of technical staff members of SEP and SEE, respectively, serving on a full-time basis. The State Technical Groups, together with the state trainers, would conduct follow-up activities to the Saturd workshops to encourage teachers to pursue further training through continued use of the Reading Corners and other self-training opportunities. 3.12. Printed training materials would be specifically developed for each training course. They would include study guides, a bibliography, a trainer's manual, evaluation instruments, and an anthology of selected readings. Posters and promotional materials would also be developed and distributed to encourage teacher participation in the training program. Materials developed for PARE have been revised for use in this project. 3.13. Beneficiaries would include all of the estimated 52,859 existing teachers in the project municipalities, who would participate in the required three general primary and the - 25 - one state-specific courses; of this number, approximately half would also each take either Written Language or Mathematics, depending on their concentration. Over the project life, an estimated 6,780 new teachers would attend the Multigrade Class Teaching course, while the indigenous education training courses would enroll an estimated 30,390 participants per course in the 12-day (summer and workshops) courses, and 1,520 in the 5-day (summer only) courses. Altogether, a total of approximately 302,985 teacher training instances would have been carried out. An estimated 8,730 participants would take the Pedagogical Counseling course, and 945 would take the Supervision course, for an aggregated total of about 9,680 participants. Component B. Educational Material Resources (US$324.2 million including contingencies) 3.14. The objective of the educational material resources component is to improve the quality of learning, student performance, and school attendance and thereby reduce student dropout and repetition rates through the provision of: (a) school and classroom teaching and learning materials; (b) student educaticnal supplies; (c) books for the Reading Corners; (d) texts and educational materials for indigenous education; and (e) urgently needed rehabilitation or replacement of inadequate school facilities and construction of new classrooms to meet expanding enrollment. 3.15. Subcomponent B.1. Teaching Materials and Student Educational Packages (US$179.2 million including contingencies). Packages of teaching materials containing maps, charts, posters, anatomic models, and other items would be distributed to each project school. Schools having access to electricity would receive an additional package of audiovisual equipment, including audio cassettes, a radio, a tape recorder, and an overhead projector. The package for schools without electricity would cost, on average, approximately US$3,850, and for those with electricity, about US$4,600. Each classroom would receive a package of grade-specific materials, including rulers, dictionaries, games, geometric solids, and chalkboards. Multigrade classes would have appropriately designed packages servicing all grades taught. These packages would cost approximately US$276 each per year for multigrade classes and US$160 for each of the six primary grades. 3.16. All students would receive a package of school supplies, including notebooks, pencils, crayons, dictionary, and rulers to encourage more active participation in learning both in school and at home, and to relieve the financial burden on their families. Packages would be differentiated in the number of items included according to the type and level of marginality of the community involved. The cost of a package would range from US$8 to US$13 per student per year. The Bank would finance non-durable student educational packages on a declining basis. The project would support an improved distribution system for both the national textbooks and the project materials (para. 3.30). 3.17. Subcomponent B.2. Reading Corners (US$7.1 million including contingencies). The proposed project would support the production and distribution of Reading Corners packages, through SEP's Education Publications Unit (Unidad de Publicaci6n Educativa, UPE) to all project classrooms. The Reading Corners program was included in PARE and was requested by the proposed project's participating states. During project preparation, it was observed that existing Reading Corners were often underused due to (i) insufficient mumbers of books and/or (ii) placement of the Reading Corners within a central and often - 26 - bureaucratized library system. Under the project, more books would be available, schools would be encouraged to provide books in the classrooms themselves, and teachers and principals would receive training on how to use and promote the Reading Corners. There are two versions of the Reading Corners package, one with at least 200 books (for schools with five or less classrooms) and the other with at least 350 books (for schools with six or more classrooms). These are higher numbers of books than under PARE or other Reading Corners elsewhere, which would make the books more accessible to students through teachers or directly in each classroom. Each package consists of books on various subjects, selected to enrich the educational experience of students, broaden their cultural outlook, and reinforce their reading skills. The books are used to supplement textbook materials, or they are borrowed by teachers, students, and the community for side reading. A further improvement to the program under the proposed project consists of a supplementary package of at least 15 educational reference books for teacher self-training and at least 30 books with regional content for each classroom. Training in the regular use of Reading Corners would be provided as part of the teacher training curriculum and reinforced through the pedagogical network. SEP's DGEI would also develop a special package of reading materials with native educational contents for Reading Corners in indigenous schools (paras. 3.18-3.19). 3.18. Subcomponent B.3. Textbooks and Educational Materials for Indigenous Education (US$6.2 million including contingencies). Language textbooks and reading materials for the Reading Corners would be developed in the 17 native languages spoken by the major ethnic groups in the project areas, which would cover the languages of about 90 percent of the indigenous population in the project states.L3' Content material for the books would be gathered through ethno-linguistic field research conducted under DGEI. A participatory methodology for text design and various visual presentation and communication modalities has already been developed with PARE support by a DGEI team, which would remain to develop indigenous textbooks under the project and other DGEI-sponsored programs. 3.19. Indigenous language textbooks and Reading Corners materials would be developed for the six primary grades. A technical group, with a DGEI coordinator and a state-level representative for each of the 17 languages, would be in charge of coordinating the selection and preparation of the native language texts. Each indigenous group representative would, in turn, head a four-member working group with direct responsibility for developing the respective textbooks and other instructional materials. The DGEI coordinator would supervise the work and ensure that national educadonal norms are observed. An outside consultant would advise the working groups and provide technical support during the first project year. Preliminary versions of textbooks, student workbooks, and teacher guidebooks for grades 1 and 2 would be designed, produced, and field tested on a pilot scale during the first project year; results and recommendations of the field tests would be incorporated into the final version which would then be printed and distnbuted. This process would be repeated for grades 3 and 4 during the second year, and those for grades 5 and 6 during the third year. This slow start-up period is required due to the large number and complexity of 23 The languages are Chol, Chontal, Huasteco, Huichol, Maya, Mazahua, Mexicanero, N6huatl, Puarpecha, Tepehuano, Totonaco, three variants of Nahuad, and one each of Huasteco, Maya, and Totonaco. For a broader perspective of the project's impact on indigenous education, see the Box Indieenous People and The Project on page 22. - 27 - indigenous languages, the need for appropriate treatment of cultural differences, and the small number of experts available. About 126,000 textbooks, 126,000 student workbooks, and 5,000 guidebooks for bilingual teachers would be evenually printed. 3.20. Subcomponent BE4. Improvement of Physical hnfrastructure (US$131.7 million including contingencies). School rehabilitation and/or construction subprojects would be implemented by the SEEs, following one of two strategies depending on the educational needs, location, access, and productive capacity of the project municipality. The first is a community participation strategy modeled after the successful SOLIDARIDAD program, which would be used for schools located in remote areas unlikely to generate contractor interests beyond the immediate vicinity and where the community is willing and able to contribute the required labor and time. It is expected that about 50 percent of construction/ rehabilitation under the project would fall under this modality. A basic design from the CAPFCE menu of school prototypes would be selected by in consultation with local School and Municipal Education Councils, with variations as applicable to accommodate local and regional diversity in climate, ecology, and available materials. This building modality does not use public bidding and operates through School Councils under the oversight of municipalities and under technical supervision provided by State Secretariats of Public Works (Secretarfa de Obras Pgiblicas del Estado, SOPE) (para. 3.23), with guidance from a self- help construction handbook. This handbook is being designed based on existing SOLIDARIDAD material, with necessary modifications to reflect the type of civil works to be carried out and the local or regional conditions. Contracting for small works such as rural schools would be done directly with local contractors for al or part of the work, using regionalized reference prices, and the community would contribute labor and locally available materials. The second strategy would be to transfer management, proement, and supervision responsibility to CAPFCE for the estimated remaining 50 percent of the imfrastructure works (located mainly in more urbanized areas or with easy access), which are considered '.arge enough to attract a number of contractors through a public bidding process. Annex 11 provides a more detailed description of the civil works to be carried out under the project. 3.21. School construction needs were determined by SEEs in each state through a survey of schools in the project areas. Civil works requirements were based on the need to: (a) improve existing school facilities, replacing rented or dilapidated classrooms or those that cannot conform to functional pedagogical or architectural standards; and (b) expand schools to accommodate current and expanding enrollments. Prioritization of infrastuure subprojects would be made by SEEs in consultation with State Planing Councils (Consejo de Panificaci&n del Estado, COPLADE) according to an Operational Manual (pam. 3.23) which would clearly state selection criteria including, er alia, overrowed conditions, grossly inadequate facilities, and consultation with supervisors. The Operational Mamnal would furher ensure that the following procedures, inter alia, be followed: (a) the CAPFCE prototype designs would include technical drawings, budgets, and technical specifications for at least three designs using different constuction materials for classrooms, sanitary services, and latrines; (b) evidence of rightful ownership of land by SEP or state governments through donation or purchase would be a condition for subproject approval; (c) in the case of civil works to be carried out through the community model, financial resources would have been transferred to the respective municipal government (ayunmentos). Teachers for new classrooms would be provided and financed by state governments, but this number is -28 - expected to be small (approximately 350) and sustainable under state education budgets. Assurance of teacher availability would be obtained prior to authorizing new classroom construction under the project; and (d) School Councils would have been created and fully operational to provide consultation and maintenance (para. 3.22). At negotiations, the Government arovided assurances that bv October 31 of each Year. starting in 1994. the Government would present: i) a list of schools and/or classrooms to be constructed/rehabilitated in the next year: (ii) ewdence that lnd is available to the SEEs for civl works: (iii) ewdence that adequa¢e staffing plans. includinf budgetarv allocations thereto have been pnalized- and Civ) evidence that local School Councils have been created and ar or erational rpaa. 7. 1(b)). 'lb meet urgent enrollment needs for the schoolyear 1994195, a small number of classrooms (approximately 315) would be constructed/ rehabilitated according to the CAPFCE model, subject to (a) all prior conditions related to this subcomponent and (b) an ex-ante review by the Bank scheduled for August 1994. 3.22. New and rehabilitated classrooms would be provided with appropriate classroom furniture, including student desks and benches and teacher desks and chairs. All schools in the project would be provided with bookcases and shelves as needed. The cost of classroom funiture and equipment has been included in the rehabilitation and construction costs. The project would support an innovative maintenance program for schools based on the operational experience (but not the financing mechanism) of the successful three-year-old Escuela Digna program. 4/ Municipalities would receive each year a fixed amount based on the number and type of facilities in the municipality (an estimated average of US$320 per facility) for infrastructure and furniture maintenance. The ayuntamientos, in coordination with the School Councils, would ensure that maintenance is carried out either by obtaining the services of a local contractor or by purchasing materials and carrying out the maintenance work themselves with the help of the community. In a relatively few cases (for about 5,500 existing classrooms), maintenance activities are more broadly defined to include some rehabilitation services necessitated by years of lack of maintenance. Because there is no previous local experience with systematic maintenance programs, the project would phase in implementation of the maintenance program in selected areas to test mechanisms tOr trnsf*r of maintenance funds to the ayuntamientos and to monitor their performance. Based on the results of this maintenance modality, the program would be expanded to cover the remaning project areas. The project would also finance printing of a maintenance handbook (currentlv being developed with technical assistance from international architects) which would be widely disseminated in the project areas, and the promotion program (para. 3.25(d)) would include stressing the importance of maintaning school buildings and furniture. 3.23. An Operational Manual for the implementation of the physical infucture subcomponent is being prepared by CONAFE in consultation with CAPFCE, with technical assistance from iemtional experts. Technical supervision of on-going civil works will be carried out by supervisors at CAPFCE and at SOPEs for the CAPFCE and community 24 The Escuela Digna ("Pride in Schools") is a SOLIDARIDAD-supported program for school repair and maintenance. Using a participatory approach, part of the costs is funded by the Federal Government (50 percent in the richer states, and 75 percent in the poorer states), part by the state (40 percent and 15 percent, respectively), and 10 percent by the local community. Since 1990, 80,000 schools have been assisted by the program. Under the project, the proposed maintenance would draw from federal funds and contnbutions (monetary, labor, or inkind) from local communities. - 29 - participation models, respectively, under terms of reference satistactory to the Bank. Each year, technical supervision reports would be presented and discussed during the annual and mid-term reviews (paras. 5.4-5.5) as part of the ex-ante review of infrastructure subprojects for the coming year, and audits of project accounts (para. 4.16) would include a specific review of a sample of infrastructure subprojects to verify that they were procured in a manner consistent with Bank guidelines, meet the minimum quality standards, and were priced within prvailing regionalized standar unit prices. At negotiations. CONAFE presented to the arnk a satisfactory drft Operational Manual for inrastructure subproiects (pam. 7.1(c)). As a condition of loan effectiveness, the Government will present. satisfactory to the Bank: (i) a inalized Operational Manual for istructure improvement: (ii) a hanbook for self-help construction: and Niii) a handbook for buidn and furniture maintenance (para. 7.2(a)). As a condition of disbursement for irastructure subproiects. the Government wll present terms of reference for technical supervision of infrastructure subprojects under both the -CAPFCE and communitv participation models. satfactora to the Bank pam. 7.3(a)). Component C. Istitutional Strengthening (US$184.5 miDion including contingencies) 3.24. The main purpose of the Institutional Strengthening Component is to assist state governments m developing manageial capacity commensurate with their newly acquired resonsibility for operating a decentralized basic education system. The project would support: (a) development of the management capacity at SEP, mainly CONAPE, and in project SEEs in the areas of (i) planning and policy analysis, (ii) monitoring, evaluation, and information systems, (iii) education studies, and (iv) project promotion and diffusion; (b) implementation of a teacher incentives program; (c) improvement of the supervision system; (d) improvement of the texbook and materials distribution system; and (e) administration of the project. The four states under PARE would also receive assistance directly thuough management training for planing and policy analysis (para. 3.25(a)) and indirectly through institutional strengthening activities at CONAFE aras. 3.25(b)-(d)). 3.25. Subcomponent C1. Manageent Capacity (US$17.6 million including contingencies). In the context of the 1992 decentralization agreement, SEEs are facing new administrative challenges which require an improved set of management skills and tools to carry out tasks previously conducted by SEP and SCEPs. The following activities would be financed by the project: (a) Maagement rining for Plnning and olicy Analysis. The project would strengthen the capacity of the state education planning bodies, enhancing skills among high-level officials through an integrated management training and technical assistance program in educational policy analysis and planning (Annex 12). The training would cover the following topics: (i) education quality and efficiency; (ii) methodologies for policy analysis and sector performance assessment; (iii) options to reform initial, basic, secondary, and hiu - education; (iv) the relationship between education, work and employment; (v) education financing; (vi) resource allocation modeling; (vii) strategic use of data collected from the education management information system; and (viii) evaluation and measurement of educational achievements. Other education issues which would be included in the trainng include decentralization, curriculum reform, and language of instruction. The project would provide 90 staff-months for overseas training fellowships and 380 staff-months of local training. Local training - 30- would be provided by educational and training institutions with recognized expertise in the relevant topics. The overseas training would be provided through workshops and internships in internationally recognized educational and research institutions. Participants would be nominated by the SEEs and selected according to predetermined criteria. Education officials in the four PARE states would be included under this program. At negotiations, the Government presented the selection criteria for the management training component. s ecifically for the overseas training fellowships. satisfactor to the Bank (para. 7.1(d)). (b) Monitoring, Evaluation and Information Systems. The project would support the strengthening of CONAFE's institutional capacity by developing its monitoring, evaluation, and infornation processing and analysis capabilities. The project would provide technical assistance to help develop an integrated project management informalion system (MIS) to be implemented at both central and state levels which would allow close monitoring and evaluation of project activities as well as sustainable measurement of educational quality through (i) reviewing the content of data registers, (ii) improving data collection processing and data analysis, and (iii) supporting the diffusion of information to all levels of the educational system. Project data collected and processed through the p