Ghana - Community Secondary Schools Construction Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 15212 IMPLEMENTATION COMPLETION REPORT REPUBLIC OF GHANA COMMUNITY SECONDARY SCHOOLS CONSTRUCTION PROJECT (CREDIT 2278-GH) DECEMBER 20, 1995 Population and Human Resources Operations Division West Central Africa Department Africa Region This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENT Currency Unit = Cedi US$1 = 365 Cedis (May 1991) US$I = I,100 Cedis (June 1995) WEIGHTS AND MEASURES Metric System FISCAL YEAR OF BORROWER SCHOOL YEAR January - December January - December (Senior Secondary) September - June (Basic and Tertiary) ABBREVIATIONS AND ACRONYMS CIDA Canadian International Development Agency CIP Community Initiative Program CSSCP Community Secondary Schools Construction Project ICR Implementation Completion Report IDA International Development Association JSS Junior Secondary Schools MOE Ministry of Education NORAD Norwegian Development Agency ODA Overseas Development Agency (United Kingdom) OPEC Organization of Petroleum Exporting Countries PAMSCAD Program of Action to Mitigate the Social Costs of Adjustment PMU Projects Management Unit PTA Parent/Teachers' Association SAR Staff Appraisal Report SDR Special Drawing Rights SSS Senior Secondary Schools UNICEF United Nations Children's Fund UNDP United Nations Development Programme USAID United States Agency for International Development  FOR OFFICIAL USE ONLY TABLE OF CONTENTS PREFACE EVALUATION SUMMARY ......................................................i PART I: PROJECT IMPLEMENTATION ASSESSMENT ......................................................I L INTRODUCTION ......................................................1 IL PROJECT OBJECTIVES AND DESCRIPTION ......................................................2 A. Project Objectives ......................................................2 B. Project Description and Components ......................................................2 C. Credit Covenants and Special Agreements ......................................................4 D. Evaluation of Objectives ......................................................4 E. Achievement of Objectives ......................................................5 I IM PLEMENTATION EXPERIENCE AND RESULTS ...................................................6 A. Assessment of Project's Success and Sustainability ......................................................6 B. Sununary of Costs ......................................................6 C. Financing Arrangements ......................................................7 D. Implementation Schedule ......................................................7 E. Analysis of Key Factors Affecting Major Objectives ......................................................8 F. Bank Performance ......................................................8 G. Borrower Performance ......................................................8 H. Assessment of Project's Outcome ......................................................8 IV. SUMMARY OF FINDINGS AND KEY LESSONS LEARNED ......................................................9 A. Important Findings from Project Implementation Experience ....................................................9 B. Lessons for Future Projects in the Sector in Ghana ...................................................... 10 PART II: STATISTICAL ANNEXES ......................................................11 Table 1: Summary of Assessments ..................................................... 12 Table 2: Related Bank Credits ..................................................... 13 Table 3: Project Timetable ..................................................... . 13 Table 4: Loan Disbursements: Cumulative Estimated and Actual (US$ m) .............................................. 14 Table 5: Project Costs and Financing ..................................................... 14 A. Project Costs (in SDR Equivalent) ..................................................... 14 B. Project Financing Estimates (US$ Thousands) ..................................................... 14 Table 6: Status of Legal Covenants ..................................................... 15 Table 7: Bank Resources: Staff Inputs ..................................................... 17 Table 8: Project Output Indicators ..................................................... 18 Table 9: Bank Resources: Missions ..................................................... 19 Table 10: Technical Assistance Financed by Project ..................................................... 20 Table 11: Training of Personnel Financed by Project ..................................................... 20 APPENDIXES ..................................................... 21 A: Missions' Aide-Memoire B: Borrower's Contribution to ICR C: Map Tls document has a restricted distribution and may be used by recipients only in the performance of their lofficial duties. Its contents may not otherwise be disclosed without World Bank authorization. l   IMPLEMENTATION COMPLETION REPORT REPUBLIC OF GHANA COMMUNITY SECONDARY SCHOOLS CONSTRUCTION PROJECT (Credit 2278-GH) PREFACE This is the Implementation Completion Report (ICR) for the Community Secondary Schools Construction Project (CSSCP) for which Credit 2278-GH was approved on June 25, 1991 in the amount of SDR 11.1 million (US$14.7 million equivalent in 1991) to the Republic of Ghana, and which became effective on August 7, 1991. A Special Account in the amount of US$4 million was established on September 17, 1991. The Credit was closed on June 30, 1995 after three extensions from the original closing date of June 30, 1993. It was disbursed up to the amount of SDR 10.6 million (US$15.18 million equivalent in 1995). Disbursements were made until August 31, 1995 when the undisbursed amount of SDR 446,223 was canceled. This ICR was prepared by Ms. Margaret Kilo and Ms. Rebekah Kirubaidoss of the Population and Human Resources Division, West Central Africa Department. It was reviewed by Mr. Nicholas Bennett (Principal Education Planner, AF4PH), Mrs. Irene Xenakis (Portfolio Manager, AF4PH), Mr. Ian Porter (Division Chief, AF4PH) and Mr. Franz Kaps (Operations Adviser, AF4DR). The Borrower provided comments on October 8, 1995 which have been integrated into the report and can be found in Appendix B of the ICR. Preparation of the ICR was begun during the Bank's completion mission in April 1995 and completed during final discussions with the Borrower in October 1995. The ICR is based on the Staff Appraisal Report, the Credit Agreement, supervision reports, correspondence between the Bank and the Borrower, and other relevant records. The Borrower contributed to the preparation of the ICR by providing relevant data, preparing its own evaluation of the project, and commenting on the draft ICR. As part of the preparation of the ICR, the Bank's Client Consultation Fund supported a beneficiary assessment of the project during the months following the closing of the project. A copy of the beneficiary assessment is available from the Africa Information Services Center.   IMPLEMENTATION COMPLETION REPORT REPUBLIC OF GHANA COMMUNITY SECONDARY SCHOOLS CONSTRUCTION PROJECT (Credit 2278-GH) EVALUATION SUMMARY 1. Bank's Role in the Sector. By the mid- 1980s, the Government, together with the Bank and other donors, started a long-term plan of educational reform by restructuring the curriculum, reducing Ghana's pre-university program from 17 to 12 years, and restructuring the financing of the system. The first phase of the reform program (1987- 1989) focused on primary and junior secondary education, supported by the Education Sector Adjustment Credit (Cr. 1744-GH). The second phase of the reform (1990-1994) was supported by the Second Education Sector Adjustment Credit (Cr. 2140-GH) and entailed the reform of the senior secondary school (SSS) system. This phase focused mainly on structural reforms, controlling school system staffing levels, and keeping expenditures in line with budgets (para. 1.1). 2. In 1991, the Community Secondary Schools Construction Project (CSSCP) (Cr. 2278-GH) was developed to further support the Education Reform Programme by providing secondary school facilities in deprived areas using the community initiative approach tested under the Program of Action to Mitigate the Social Costs of Adjustment (PAMSCAD). Under PAMSCAD, the Government provided matching funds to communities' contributions made in cash, materials, or labor, for construction of school facilities (para. 1.2). 3. The Bank is continuing its support to the formal education sector through the Literacy and Functional Skills Project (Cr. 2349-GH); the Tertiary Education Project (Cr. 2428-GH); the Primary School Development Project (Cr. 2508); and the Vocational Skills and Informal Sector Support Project (Cr. 2695-GH). A Basic Education Sector Investment Project (GH-PA-975) is under preparation, and should be appraised early in 1996. 4. Project Objectives. The overall objectives of the project were to ensure that (a) qualified Junior Secondary School (JSS) graduates from underserved areas of the country had opportunities to enter effective and relevant day SSSs in their neighborhood; (b) essential physical facilities were constructed at about 140 new SSSs, and ready to be used by the beginning of the 1992 academic year; and (c) local communities and district assemblies took responsibility for the development and operation of the new secondary schools, ensuring community ownership. In addition, the project aimed at securing the development, at each new senior secondary school, of a library which would serve the school and the local community, as well as hygienic toilets for the use of staff and students (para. 2.1).  ii 5. Project Description and Components. The project provided support to about 140 communities to construct SSS facilities, following the PAMSCAD approach. The new facilities would: (a) provide a vocationally-oriented curriculum in deprived parts of the country; (b) bring schools closer to potential students; (c) increase the enrollment rate of girls in secondary technical schools; and (d) increase overall SSS enrollment by 18,000 in 1993, resulting in a 68 percent increase in the enrollment capacity of the three northern regions (Upper West, Upper East, and Northern Regions). The project components include: construction of classrooms, workshops/laboratories, offices, staffrooms and staffhouses; provision of library books; construction of generator shelters where needed; and for the construction of student and staff toilets using their own resources, each school would be rewarded with a pick-up truck (paras. 2.2; 2.4 - 2.7). 6. Credit Covenants and Special Agreements. Agreements with the Borrower (Republic of Ghana) were reached on the following issues, namely, that the Republic of Ghana would: (a) enter into Building Construction Agreements with communities concerned which shall have been approved by the Association; (b) pursuant to the Building Construction Agreements, obtain rights adequate to protect the interests of the Borrower and the Association; (c) through the Project Management Unit, prepare quarterly progress of project execution and submit such reports to the Association; (d) maintain records and keep accounts adequate to reflect the operations, resources and expenditures of the project; (e) furnish to the Association not later than six months after the end of each fiscal year, a certified report of the Special Account prepared by an independent auditor acceptable to the Association. The credit covenants and special agreements were complied with during project implementation (paras. 2.8; 2.9) 7. Evaluation of Project Objectives. The main objectives of the project were to increase access to SSS, particularly to students from underserved areas, and to involve districts and communities in the development and operation of the new schools. These objectives were consistent with the priorities of the Education Reform Programme, remained relevant to the most pressing needs of the SSS system, and were reconfirmed during the informal mid-term review in August 1993. In addition to furnishing essential school facilities, the project would reward each school with a vehicle for providing hygienic toilets for the use of staff and students. To accommodate the new students, the facilities were to be quickly constructed by the local communities with the support of the Government following the PAMSCAD approach which had produced many Community Initiative Projects (CIPs) using Ghanaian self-help traditions. By involving the local communities in school construction, it was envisaged that they would have access to the library books and share the facilities to which they had contributed (paras. 2.10-2.13). IMPLEMENTATION EXPERIENCE AND RESULTS 8. Assessment of Project's Success and Sustainability. The new schools increased enrollments by 30,712 intakes (of which 25.5 percent were girls) in 1992, as compared to the original SAR target of 18,000 intakes by 1993, in all project schools. The total intake in 1995 is 45,279 students, of which 37.2 percent are girls. The project provided multi-  iii purpose units including classrooms and workshops or laboratories (621 compared to 581 in the DCA); libraries and bookstores (151 compared to 132); staffrooms and offices (152 compared to 132); staff houses (370 compared to 286); and generator shelters (102 compared to 90). It must be noted that in 100 of the 157 project schools, facilities were added to existing schools, hence the small number of classrooms relative to intakes. In total, 16.7 percent more square meters were added to the originally estimated quantities and the construction cost was about 70 percent of other Bank-supported projects. All project schools were rewarded with vehicles for constructing toilets using their own funds. All communities contributed labor in varying degrees, but the project was less successful in accomplishing the objective of involving the communities and district assemblies in the management, maintenance, and utilization of the facilities to which they contributed. Most communities and district assemblies did not clearly understand their respective roles, and as a result delays in community construction led to private contractors being engaged to take over the entire construction component of the project. The school management committees, established in 1989 to support community initiatives, have been more effective in some communities than others. Today, district assemblies have responsibility for school management, however, the role of the community remains less clear (paras. 3.1 and 3.9). 9. Summary of Costs. The project was estimated to cost US$19.6 million with the Government contributing US$4.9 million, comprised mainly of community and district assembly contribution in cash, kind, or labor, and an IDA financing of US$14.7 million (see Table SB). The Credit disbursed US$15.18 million, an amount higher than the appraisal estimate due to changes in the US dollar/SDR exchange rate. A detailed breakdown of the cost of project components in SDRs is provided in Table 5A (para. 3.2). 10. Implementation Schedule. At the post-appraisal stage of EdSAC II, the Government identified the need for additional SSS facilities to cater to the increase in JSS enrollments due to the Education Reform Programme. Together the Government and the Bank prepared the project in March 1991. The project went to the Board on June 25, 1991 and became effective on August 7, 1991. Although originally estimated to be a quick-disbursing project and to close after two and one-half years, the project was extended three times and fully implemented over a four-year period (para. 3.5). This is still remarkable for the Africa Region. Originally due to close on June 30, 1993, the Credit closed on June 30, 1995. Disbursements were made until August 31, 1995 and an outstanding balance of SDR 446,223 was canceled (para. 3.4). 11. Project implementation during the early years was slow for several reasons. Firstly, the project was designed as a CIP with required facilities varying from school to school, but the Government tried to standardize construction so some communities were asked to contribute toward buildings they did not want or to support fewer buildings than they needed. Secondly, the project was launched during the period coinciding with the harvest season and its high labor demands. Thirdly, this was also at a time when district administrations had spent most of their annual budgets and could not help communities with funds to start the buildings according to the CIP tradition. Fourthly, by insisting that  iv all project funds pass through their accounts, district assemblies delayed construction due to their weak accounting practices. Finally, community enthusiasm waned during the time it took for the technical auditors to certify that they had completed a particular phase of work and the time they received payment (para. 3.5). 12. Analysis of Key Factors Affecting Major Objectives. PAMSCAD had succeeded in reaching out and involving local communities throughout the 110 district assemblies in the 10 regions of Ghana, and as a result, it provided a favorable environment for launching the community participation initiatives of the CSSCP. During project implementation, however, the factors identified in paragraph 11 above, weakened these initiatives. Further, communities involved in the CSSCP were not equally endowed in terms of leadership and material resources. Others lacked water supply, adequate roads to transport building materials and equipment, and communication facilities (such as a telephone). These inequities led to inequality in the rate of construction among the different sites (para. 3.6). 13. Assessment of the Bank's and Borrower's Performance. The Bank interacted efficiently with the client and responded in a timely manner. The Task Manager outposted to the field was available for direct consultation throughout the entire reform process and for most of the life of this project. This role was critical, particularly for the procurement process in school construction. For example, the Task Manager established a system of expediting the process of screening and approving all withdrawal applications, which included the verification of the level of construction work as ascertained by the technical auditors for all schools, before they were sent to the Bank for replenishment of the Special Account (para. 3.7). 14. The Government demonstrated its strong commitment to the Education Reform Programme by maintaining the same experienced and competent staff in the PMU throughout the life of the project. When it became clear that some sites would not be completed before the project closing date, the Government made timely arrangements for counterpart funds to be set aside for this purpose (para. 3.8). 15. Assessment of Project's Outcome. The CSSCP expanded access to secondary schools for students in underserved parts of Ghana by furnishing school facilities at a lower cost per unit than similar Bank-supported projects, provided more facilities, and enrolled more students than estimated at appraisal, contributing significantly to the absorptive capacity of the sector. In this respect, the project is classified as satisfactory. The community participation outcome was partially successful since communities did contribute to school construction but have not taken responsibility for school management (para. 3.9). IMPORTANT FINDINGS OF PROJECT IMPLEMENTATION EXPERIENCE 16. The procurement for civil works was handled directly by the district assemblies, providing support to local communities in their construction activities, with each phase of  v work certified by independent technical auditors. The unit costs of construction were around 30 percent lower than construction of similar quality procured under other Bank- supported projects. The modular approach simplified construction sufficiently for communities and small-scale local contractors to manage. Simplicity of design, and the location of the task manager in the field helped ensure rapid initial processing, and implementation over four years, less than half the regional average for PHR projects (para. 4.1). 17. Bank pressures for rapid appraisal of the project meant that the complexities of mobilizing community support was not fully appraised. Implementation experience showed that the key factor in the mobilization of community participation was local level leadership, but little attempt was made to find out whether such leadership existed or not. The project assumed that there was a commonality of interests between the newly established but cash-strapped districts and the communities surrounding the schools. In fact, the district assemblies wanted to control this financially significant activity, and also often wanted to delay passing on eligible funds to communities or contractors, but lacked the capacity to manage construction activities or project accounts (para. 4.2). 18. When community participation is to be relied upon it is very important that enthusiasm, once created, is maintained. If, through delays in payment or other problems, initial enthusiasm is lost it is very difficult to resuscitate. Therefore, it is important to launch a community construction project only when all systems are fully in place so there will be no delays. Even without delays, it is uncertain that community construction activities will not be interrupted. Consequently, in countries with significant inflation rates, some automatic allowance for inflation should be built into construction projects (para. 4.3). 19. The land tenure situation on some of the new sites on which construction was to take place should have been clarified before the agreements were signed with the local communities. The final six-month extension of the project was largely due to problems on sites where there were conflicts between the schools and the landlords (para. 4.4). 20. Finally, it should be emphasized that the efficiency in the development and appraisal of this project demonstrates the advantages of a strong field presence in the processing and implementing of projects. This should have significant impact on the eventual location of responsibility for the vast proportion of simple PHR projects (para. 4.5). LESSONS FOR FUTURE PROJECTS IN THE SECTOR IN GHANA 21. The CSSCP has provided a number of key lessons for future projects within the education sector: (a) the advantages of field presence for project preparation and supervision; (b) the efficiency of using technical auditors to monitor the progress of construction prior to payment; (c) the need to spend a great deal of time and effort in preparing community participation activities; (d) the necessity to properly time project  vi launch in order to ensure maximum community participation and the necessity to continuously maintain this motivation; (e) the need to pay communities and local contractors directly and not to pass through an intermediary such as a local government authority; (f) the advantage of using a modular, standardized approach for civil works; and (g) the difficulty that the Ministry of Education has in effectively communicating and controlling activities at the district and community levels (para. 4.7).  IMPLEMENTATION COMPLETION REPORT REPUBLIC OF GHANA COMMUNITY SECONDARY SCHOOLS CONSTRUCTION PROJECT (Credit 2278-GH) PART I: PROJECT IMPLEMENTATION ASSESSMENT I. INTRODUCTION 1.1 By the mid-1980s, the Government, together with the Bank and other donors, started on a long-term plan of educational reform by restructuring the curriculum, reducing Ghana's pre-university program from 17 to 12 years, and improving the financing of the system. The first phase of the reform program focused on primary and junior secondary education and was supported by the Education Sector Adjustment Credit (Cr. 1744-GH) for US$38.1 million from 1987 to 1989. The second phase of the reform was supported by the Second Education Sector Adjustment Credit (Cr. 2140-GH) from 1990 to 1994 for US$50 million, and entailed the reform of the senior secondary school (SSS) system. This phase focused mainly on structural reforms, controlling school system staffing levels, and keeping expenditures in line with budgets. 1.2 In 1991, the Community Secondary Schools Construction Project (CSSCP) (Cr. 2278-GH) was developed to further support the program by providing secondary school facilities in underserved areas using the community initiative approach tested under the Program of Action to Mitigate the Social Costs of Adjustment (PAMSCAD). Through the project, the Government would provide matching funds to communities' contributions made in cash, materials, and labor, for the construction of school facilities. The IDA credit of SDR 11.1 million became effective on August 7, 1991 and was originally scheduled to close on June 30, 1993. 1.3 Community Participation in School Construction and Maintenance. Under PAMSCAD's Community Initiative Project (CIP) the Government had developed a system of providing matching grants based on community performance which were used to undertake the development of schools, markets, health centers, etc. The level of support was usually between one-half and two-thirds of the estimated total cost of a project and was designed to cover the cost of imported materials and highly-skilled labor. Fifty percent of the Government's contribution was paid when the foundation was completed and the remainder upon completion of the project. CIP also successfully adapted Ghanaian traditions of self-help to mobilize substantial labor and material support from local communities while reducing the bureaucracy of normal contracting procedures. In August 1990, a review mission consisting of the Bank and other donors in Ghana (CIDA, ODA, UNDP, UNICEF, USAID) recommended that: (a) the contribution of poorer communities be reduced; and (b) communities be allowed to make a greater part of their contribution in labor.  2 1.4 Using the "Self-Help Building Guidelines" prepared by the Ministry for Local Government and PAMSCAD, groups of technical staff who would assist local communities in building the SSSs were trained under this project. It was further expected that according to CIP tradition there would be sufficient community ownership and interaction between the schools and communities to assure maintenance of the buildings and vehicles thus acquired. II. PROJECT OBJECTIVES AND DESCRIPTION A. Project Objectives. 2.1 The overall objectives of the project were to ensure that (a) qualified JSS graduates from underserved areas of the country had opportunities to enter effective and relevant day SSSs in their neighborhood; (b) essential physical facilities were constructed at about 140 new SSSs, and ready to be used by the beginning of the 1992 academic year; and (c) local communities and district assemblies took responsibility for the development and operation of the new secondary schools, and ensure community ownership. In addition, the project aimed at securing the development at each new SSS, of a library which would serve the school and the local community, as well as the provision of hygienic toilets for the use of staff and students. B. Project Description and Components. 2.2 Community Involvement and Regional Equity. At its onset the project involved providing support to about 140 communities to construct SSSs. Following the PAMSCAD approach, the Government would provide a two-thirds matching grant in two installments to the communities after they completed the building's foundation using their own resources, and committed to provide one-third of an agreed fixed cost of a particular type of building in cash, materials, or labor. The new facilities constructed would: (a) provide a vocationally-oriented curriculum in deprived parts of the country; (b) bring schools closer to potential students; (c) increase the enrollment rate of girls in secondary technical schools; and (d) increase SSS enrollments by 18,000 in 1993 resulting in a 68 percent increase in the enrollment capacity of the three northern regions (Upper West, Upper East and Northern Regions). The project would ensure that the new schools did not become academic islands but real community schools serving the communities' interests. 2.3 Choosing the Number of New Schools and their Sites. Early in 1990 a series of meetings involving all headteachers of the existing 254 SSSs were held to determine the number of JSS graduates they would be able to enroll in the first year of the new SSS system. The enrollments in the old schools were to be deternined in terms of: (a) boarding students based on the regional and national selection process; and (b) day students from the JSSs in the local catchment area. Taking into account the existing 5,136 JSSs, the Ministry of Education (MOE) was able to identify those districts or sub- districts where new SSSs were needed to ensure that places were available for 30 percent  3 of JSS 3 students in SSS 1. On the basis of this rudimentary school mapping exercise, it was decided that the new SSSs would be utilized to reduce existing regional disparities. In consultation with district assemblies and communities, 140 sites were selected for the new SSSs. In about 100 of these, well-endowed JSSs would be converted to SSSs with some additional facilities, and for the remainder, entirely new schools would be needed. In both cases, commitments were obtained from the district assemblies and the local communities to assist with cash, materials, and labor in the construction and furnishing of the facilities needed. In addition, 97 communities agreed to assist in housing their teachers. Each of the selected sites was visited by high level and technical staff of MOE. In January 1991 when the schools opened, additional information was obtained from the regional MOE staff on the numbers of students, staff, and physical facilities actually in place. 2.4 Classrooms, Workshops, and Offices. A space utilization factor of 85 percent was used to calculate the teaching space requirements for classrooms, laboratories, and workshops for the new schools. Each standard multi-purpose block was designed for one workshop or laboratory, or one staffroom and office, or one library and bookstore, or for two classrooms such that whenever an odd number of classrooms were required, an additional classroom was added. A total of 575 additional teaching spaces were needed. Each school would have a library and a bookstore for textbooks that would be loaned to students for a user fee. A staffroom, where teachers could sit to mark papers and hold meetings would also be included. In the new rural schools there was often no other meeting place in the vicinity. Finally, each school would also have a small office facility which included a general office, the headteacher's office, and a store. 2.5 Staff Houses and the Electrical Supply Problem. It was considered crucial for some teachers to live on-site so that the library could be kept open to students and the community after hours, in order for extracurricular activities to take place in the compound, and generally for the security of the school buildings and equipment. Two staff houses, each with two bedrooms designed to accommodate two bachelor teachers or national servicemen, were to be constructed on each school compound. Since most schools were not connected to the national electricity grid, generators would be procured for the schools with EdSAC II funds and the schools provided with generator shelters using the same construction process as the classroom construction. MOE drilling rigs would be used to drill boreholes at school sites where water supply was a problem. 2.6 Library Books. The project would also finance the procurement of at least 1,000 books for the library at each school. These books would include literature books, textbooks, reference materials, and encyclopedias in two or three multiple copies. In addition, educational charts would be procured and posted in the classrooms. 2.7 Transportation. Once the school had provided hygienic toilets for staff and students from its own resources, it would be rewarded with a single-cabin pick-up truck. These vehicles would be used in: transporting textbooks and equipment from regional stores to the schools; transporting school sports teams; collecting monthly salaries; and  4 transporting farm produce from the school farm. This form of transportation was essential given the remote location of the schools. Almost all the existing secondary schools had a mummy wagon, bus, or truck donated by benefactors, charitable trusts, or in some cases purchased by the Government. C. Credit Covenants and Special Agreements. 2.8 During Credit negotiations, the following agreements were reached, namely, that the Borrower (Republic of Ghana) would: (a) enter into Building Construction Agreements with communities concerned which shall have been approved by the Association; (b) pursuant to the Building Construction Agreements, obtain rights adequate to protect the interests of the Borrower and the Association; (c) through the PMU, prepare quarterly progress reports of project execution and submit such reports to the Association; (d) maintain records and keep accounts adequate to reflect the operations, resources and expenditures of the project; (e) furnish to the Association not later than six months after the end of each fiscal year, a certified report of the Special Account prepared by an independent auditor acceptable to the Association. 2.9 Category (1) of Schedule 1 of the Development Credit Agreement was amended on September 13, 1993 to increase IDA funding for building construction from 67 percent to 90 percent and the Credit was extended three times, due to delays in construction. There were no specific conditions for Credit effectiveness and these credit covenants and special agreements were complied with during project implementation. D. Evaluation of Objectives. 2.10 The main objectives of the project were to increase access to SSS, particularly to students from underserved areas, and to involve districts and communities in the development and operation on the new schools. These objectives were consistent with the priorities of the Education Reform Programme, remained relevant to the most pressing needs of the SSS system, and were reconfirmed during the informal mid-term review in August 1993. To this end, the project would provide new schools in about 140 communities, to enable JSS graduates in remote parts of the country to enter SSS in their localities. The new schools would increase enrollments for both boys and girls but particularly for students in the northern sector of the country. In addition to furnishing essential facilities, the project would reward schools with a vehicle for using their own resources to provide hygienic toilet facilities for the use of staff and students. To accommodate the new students, the facilities were to be quickly constructed by the local communities with the support of the Government following the PAMSCAD approach which had been previously successful and produced many CIPs. By involving the local communities in school construction, it was envisaged that they would have access to the library books and share the facilities to which they had contributed. 2.11 Buildings. This was the largest component of the project and was expected to be quick-disbursing. The contracts negotiated with the local communities were small  5 enough to be completed in about four months so that inflation was never factored in. Unfortunately, this component was responsible for delays in project implementation. The project launch workshop took place in August 1991, which inadvertently coincided with the peak of the harvest season and its problems (high labor demands; tarrying cocoa payments; school re-opening and its financial demands on families; etc.). The district assemblies were also at that point in the financial year when their own resources were depleted and, as a result, they could make little contribution to the project. Hence, the attraction for them to hijack the project whose funds were more than their annual budgets. Delays in appointing the technical auditors further delayed project implementation, and as payments to communities were delayed, construction fell behind and inflation affected the fixed price contracts that had been signed between the local communities and MOE. As a result, the project was unable to take off with the kind of momentum that had been expected. However, construction work did eventually speed-up, and the third extension of the project allowed the slower contractors and the conflict ridden area of the Northern Region sufficient time to complete school construction. 2.12 Library books. This component of the project has not been as successful as was anticipated. Although all schools did receive the 1,000 books, in some schools the books were not displayed in the library for lack of furniture. At the time of the preparation of this report, some students did not have access to the library books. The Government will utilize counterparts funds reserved for project completion to provide the missing furniture. It was also envisaged that members of the community would have access to the books. Unfortunately, this has not been the case, due to factors identified in paragraph 2.16 below. 2.13 Vehicles. The project procured simple pick-up trucks for schools as a reward for their constructing locally designed toilets (called the Kumasi Ventilated Improved Pit Latrine or KVIPs) for the use of staff and students. The vehicles are being used by the schools in a wide variety of activities and have tremendously improved transportation, notably for schools in remote areas. E. Achievement of Objectives. 2.14 The objective to increase access to SSSs in deprived areas has been substantially achieved. The project succeeded in providing new school buildings and helped increase SSS enrollments by 30,712 in 1992, including an increase of 7,820 (25.5 percent) girls compared to the original SAR target of 18,000 intakes by 1993. By 1994, enrollments in project schools had attained 46,569 and about 12,306 (37.2 percent) for girls. The total student intake in 1995 is 45,279, of which 12,149 are girls. The project provided multi- purpose units including classrooms and workshops or laboratories (621 compared to 581 in the DCA); libraries and bookstores (151 compared to 132); staftrooms and offices (152 compared to 132); staff houses (370 compared to 286); and generator shelters (102 compared to 90). In total, 16.7 percent more square meters were added to the originally estimated quantities and the buildings cost 70 percent of other Bank-supported projects. All project schools constructed toilets for their staff and students, and received vehicles as  6 a reward. All communities contributed labor in varying degrees, but the project was less successful in accomplishing the objective of involving communities and district assemblies in the management, maintenance and utilization of the facilities to which they had contributed. 2.15 Community Participation. There was an assumption at the onset of the project that CIP would function as in the past. All communities contributed their labor toward construction, but community contribution was below the 30 percent estimated at appraisal. Further, delays in community construction led eventually to private contractors being called in to take over the entire construction component of the project. 2.16 While all schools now have the 1,000 books mentioned in the SAR, some schools did have furniture (bookshelves, tables and chairs) in their libraries and the books were still packed in cartons. There are few trained personnel to oper- the libraries. By involving the local communities in school construction, it was envisaged that they would have access to the library books and share the facilities they had contributed in providing. However, apart from financial contribution to PTAs, most parents shy away from participation in school activities. III. IMPLEMENTATION EXPERIENCE AND RESULTS A. Assessment of Project's Success and Sustainability. 3.1 The project succeeded in providing new school buildings and helped to expand access to SSS (see Table 8). All communities contributed their labor to the projects to varying degrees, but the project was less success in accomplishing the objective of involving the communities and district assemblies in the management, maintenance, and utilization of the facilities. The extent to which the communities were expected to be involved in the schools was initially ambitious. Most communities and district assemblies did not clearly understand thei respective roles and as a result, delays in community construction eventually led to private contractors being engaged to take over the entire construction component of the project. The school management committees established in 1989 to suppL,,A community initiatives, have been more effective in some communities than others. Today, district assemblies have responsibility for school management, but the role of the community remains less clear. B. Summary of Costs. 3.2 The project was estimated to cost US$19.6 million with the Government contributing US$4.9 million, comprised mainly of community and district assembly contributions in cash, materials or labor, and an IDA financing of US$14.7 million (see Table 5B). Total disbursements at project closure of US$15.18 million were higher than the appraisal estimate due to changes in the US dollar/SDR exchange rate. The Credit was reallocated on September 13, 1993 to cover the increased cost of construction,  7 technical auditors, and vehicles. A detailed breakdown of the cost of project components in SDRs is provided in Table 5A. C. Financing Arrangements. 3.3 The project was financed according to the CIP approach. The Bank increased its contribution from 66 percent to 90 percent during an informal review of the project in August 1993, when it became clear that local communities were unable to contribute their share of 30 percent in cash or kind. The 90 percent was expected to cover about 80 percent of increased costs of building materials due to inflation. Toward the end of the project, the Government readily set aside counterpart funds for the completion of buildings for which advances had been paid and which, for one reason or another, were not completed. It is rather ironic that in the Upper West and Upper East Regions (normally considered to be deprived areas), the local communities were more eager to contribute their share than in more metropolitan and endowed areas. D. Implementation Schedule. 3.4 In December 1989, at the post-appraisal stage of EdSAC II, the Governnent identified the need for additional SSS facilities to cater to the increase in JSS enrollments as a result of the Education Reform Programme, and requested help from the Bank in obtaining assistance from potential co-financiers in November 1990. Early in 1991, when it became evident that no co-financing was forthcoming, the Government requested Bank assistance. Together, the Government and the Bank prepared and appraised the project in April 1991. The project went to the Board on June 25, 1991 and became effective August 7, 1991. About four months from Preparation to Board was a record for the Bank, and this was only possible because the Task Manager was present in the field and worked closely with Government officials (see Tables 3 and 7) during this phase. Originally due to close on June 30, 1993, the Credit closed on June 30, 1995 after three extensions totaling twenty-one months, due to delays imposed by the involvement of local communities in project implementation. The last disbursements were made on August 31, 1995 when the balance of SDR 446,223 was canceled. 3.5 Project implementation during the early years was slow for several reasons. Firstly, the project was designed as a CIP with required facilities varying from school to school. In spite of the wide degree of differentiation provided for in the SAR, the Government tried to standardize the early phase of construction so some communities were asked to contribute toward buildings they did not want while others were supporting fewer buildings than they needed. Secondly, the project was launched with a series of workshops throughout the country during August and September 1991, the period coinciding with the harvest season and its high labor demands. Thirdly, it was also at a time when district assemblies had spent most of their annual budgets and could not help communities with funds to start the buildings according to the CIP tradition. Fourthly, by insisting that all project funds pass through their accounts, district assemblies delayed construction due to their very weak accounting practices. Finally, community enthusiasm  8 waned during the time it took for the technical auditors to certify that they had completed a particular phase of work and the time they received payment. E. Analysis of Key Factors Affecting Major Objectives. 3.6 PAMSCAD, which was already in operation before the CSSCP was developed, succeeded in reaching out and involving local communities throughout the 110 district assemblies in the 10 regions of Ghana, and as a result, provided a favorable environment for launching the community participation initiatives of the CSSCP. During implementation, however, the factors identified in paragraph 3.5 above, weakened these initiatives. Further, not all communities involved in the CSSCP were equally endowed in terms of leadership and material resources. Others lacked water supply, adequate roads to transport building materials and equipment, and communication facilities (such as a telephone). These inequities led to inequality in the rate of construction among the different sites. F. Bank Performance. 3.7 The Bank interacted efficiently with the client and responded in a timely manner. The Task Manager outposted to the field was available for direct consultation throughout the entire reform process and for most of the life of this project. This role was critical, particularly for the procurement process in school construction. For example, the Task Manager established a system of expediting the process of screening and approving all withdrawal applications, which included the verification of the level of construction work as ascertained by the technical auditors for all schools, before they were sent to the Bank for replenishment of the Special Account. G. Borrower Performance. 3.8 The Government demonstrated its strong commitment to the Education Reform Programme by maintaining the same experienced and competent staff in the PMU throughout the life of the project. When it became clear that some sites would not be completed before the project finally closed on June 30, 1995, the Government made timely arrangements for counterpart funds to be set aside for this purpose. H. Assessment of Project's Outcome. 3.9 The CSSCP expanded access to secondary schools for students in underserved parts of Ghana by furnishing school facilities at a lower cost per unit than similar Bank- supported projects, provided more facilities, and enrolled more students than estimated at appraisal, contributing significantly to the absorptive capacity of the sector. In this respect, the project is classified as satisfactory. The community participation objective  9 was partially successful since communities did contribute to school construction but have not taken responsibility for school management. IV. SUMMARY OF FINDINGS AND KEY LESSONS LEARNED A. Important Findings from Project Implementation Experience. 4.1 The procurement for civil works handled directly by the district assemblies, providing support to local communities in their construction activities, with each phase of work certified by independent technical auditors, proved to be effective. The unit costs of construction were around 30 percent lower than construction of similar quality procured under other Bank-supported projects. The modular approach whereby most of the facilities constructed at the schools were fitted into a standard module of 1,000 sq. ft. simplified construction sufficiently for communities and small-scale local contractors to manage. Simplicity of design, and the location of the task manager in the field helped ensure rapid initial processing, and implementation over four years, less than half the regional average for PHR projects. 4.2 Pressures for rapid appraisal of the project in order to meet Bank lending targets and access additional resources meant that the complexities of mobilizing community support were not fully appraised. Implementation experience showed that the key factor in the mobilization of community participation was local level leadership, but during the initial field survey no attempt was made to ascertain whether such leadership existed or not. The project assumed that there was a commonality of interests between the newly established but cash-strapped districts and the communities surrounding the schools. In fact, the district assemblies wanted to control this financially significant activity, and often wanted to delay passing on eligible funds to communities or contractors. The district assemblies which had only just been established did not have the capacity to manage construction activities or project accounts. 4.3 When community participation is to be relied upon it is very important that enthusiasm, once created, is maintained. If, through delays in payment or other problems, initial enthusiasm is lost, it is very difficult to resuscitate. Therefore, it is important to launch a community construction project only when all systems are fully in place so there will be no delays. Even without delays, it is uncertain that community construction activities will not be interrupted. Consequently, in countries with significant inflation rates, some automatic allowance for inflation should be built into construction projects. 4.4 The land tenure situation on some of the new sites on which construction was to take place should have been clarified before the agreements were signed with the local communities. The final six-month extension of the project was largely due to problems on sites where there were conflicts between the schools and the landlords.  10 4.5 Finally, it should be emphasized that the efficiency in the development and appraisal of this project demonstrates the advantages of a strong field presence in the processing and implementing of projects. This should have significant impact on the eventual location of responsibility for the vast proportion of simple PHR projects. B. Lessons for Future Projects in the Sector in Ghana. 4.6 The Community Secondary Schools Construction Project was the first of a series of investment operations (following two sector adjustment operations) designed to support infrastructural developments in various parts of the education sector in Ghana. The projects that followed were: (a) Literacy and Functional Skills Project (Cr. 2349- GH); (b) Tertiary Education Project (Cr. 2428-GH); (c) Primary School Development Project (Cr. 2508); and (d) the Vocational Skills and Informal Sector Support Project (Cr. 2695-GH). A Basic Education Sector Investment Project (GH-PA-975) is under preparation, and should be appraised early in 1996. 4.7 The CSSCP has provided a number of key lessons for future projects within the education sector: (a) the advantages of field presence for project preparation and supervision; (b) the efficiency of using technical auditors in monitoring the progress of construction prior to payment; (c) the need to spend a great deal of time and effort in preparing community participation activities; (d) the necessity to properly time project launch in order to ensure maximum community participation and the necessity to continuously maintain this motivation; (e) the need to pay communities and local contractors directly and not to pass through an intermediary such as a local government authority; (f) the advantage of using a modular, standardized approach for civil works; and (g) the difficulty that the Ministry of Education has in effectively communicating and controlling activities at the district and community levels.  11 PART II: STATISTICAL ANNEXES  12 TABLE 1: SUMMARY OF ASSESSMENTS Not A. Achievement of objectives Substantial Partial Negligible applicable Macroeconomic policies X Sector policies X Financial objectives X Institutional development X Physical objectives X Povertv reduction X Gender concerns X Other social objectives X Environmental objectives X Public sector management X Private sector development X B. Project sustainability Likely Unlikely Uncertain x Highly C. Bank perfonnance satisfactory Satisfactory Deficient Identification X Preparation assistance X Appraisal X Supervision X Highly D. Borrower performance satisfactory Satisfactory Deficient Preparation X Implementation X Covenant compliance X Highly Highly E. Assessment of outcome satisfactory Satisfactory Unsatisfactory unsatisfactolr x  13 TABLE 2: RELATED BANK CREDITS Loan/Credit Title Purpose Year of Status Approval 1. Cr. 1744-GH To support education system reform 1987 Closed December 31, 1991 Education Sector Adjustment Credit l 2. Cr. 2140-GH To support education system reform 1990 Closed December 31, 1994 Second Education Sector Adjustment Credit 3. Cr. 2349-GH To support Government's 1992 Ongoing Literacy and Functional mass literacy program Skills 4. Cr. 2428-GH To support extending reform in 1993 Ongoing Tertiary Education tertiary education Project l 5. Cr. 2508-GH To assist increasing learning 1993 Ongoing Primary School achievement and enrollments Development Project 6. Cr.2695-GH To improve informal sector 1995 Ongoing Vocational Skills and productivity and reorient the Informal Sector Support vocational training to a demand- Project driven system 7. Basic Education To provide support to improve 1996 Under preparation Sector Investment learning outcomes, increase access, Project and strengthen management TABLE 3: PROJECT TIMETABLE Steps in Project Cycle Date Planned Actual date Identification Not available December 1989 Preparation Not available March 1991 Appraisal Not available May 10, 1991 Negotiations Not available May 21, 1991 Board presentation Not available June 25, 1991 Signing Not available July 8, 1991 Effectiveness Not available August 7, 1991 Project completion December 31, 1992 December 31, 1994 Credit Closing June 30, 1993 June 30, 1995  14 TABLE 4: LOAN DISBURSEMENTS: CUMULATIVE ESTIMATED AND ACTUAL (US$ M) FY1992 FY1993 FY1994 FY1995 Appraisal estimates 10.00 14.70 Actual 5.41 9.72 12.98 15.18 [Actual as % of estimate 54% 66% 88% 103% Date of final disbursement: August 24, 1995 TABLE 5: PROJECT COSTS AND FINANCING A. PROJECT COSTS (IN SDR EQUIVALENT) Items - Estimated Cost Actual Cost 1. Building Construction 8,190,000 9,187,493 2. Teaching Materials 690,000 517,006 3. Vehicles 860,000 1,231,582 4. Consultants' Services 220,000 343,436 5. Unallocated 1,140,000 0 TOTAL PROJECT COST 11,100,000 1 10,349,517 B. PROJECT FINANCING ESTIMATES (US$ THOUSANDS) Source Local J Foreign Total Community/District Assembly 3,449 1,479 4,928 Contribution IDA 8,494 6,203 14,696 TOTAL 11,943 7,682 19,624  TABLE 6: STATUS OF LEGAL COVENANTS Section Covenant Status Original Revised Description of covenant Comments class fulfillment fulfillment date date 2.02 (b) I C Borrower to open and maintain a special deposit account in a The account of US$4.0 million was opened in a I__ _ _ _ J_commercial banik. commercial bank on September 1991. 2.03 10 06/30/93 06/30/95 Closing Date shall be 6/30/1993, or later date as IDA shall The project closing date was extended from establish. 12/31/94 to 6/30/95. 3.03 9 The Borrower, through the PMU, shall prepare quarterly progress None of project execution, such reports to be submitted to IDA. 3.04(a) 10 C The Borrower shall enter into the Building Construction None Agreements with communities concemed which shall have been approved by IDA and which shall specify the fixed unit costs per building as set forth in this agreement. 3.04(b) 10 C Pursuant to the Building Construction Agreements, the Borrower None shall obtain the rights adequate to protect the interests of the Borrower and IDA, including the right to: 4 CP Stages of completion varies throughout the (i) require the community concemed to carryout the school schools targeted under the project. v construction with due diligence and efficiency and in accordance with sound technical, financial, and educational standards following the design specifications developed for this project; 4 CP DCA has been amended to increase IDA (ii) require the community to make a contribution to the school financing in Category I (Building Construction) construction costs of at least 33 pereent of the estimated costs from 67 percent to 90 percent require from time either in the form of labor services, building materials or in terms to time. 9 C of cash contributions; None (iii) inspect, by itself or jointly with representatives of IDA, if IDA 9 C shall so request, the school construction works carried out by the community; and None (iv) require the community to furnish the Borrower all such informalion as the Borrower or IDA shall require from time to ________ t~~ ~~~~~~~~~im e. __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 4.01 (a) (i) I C The Borrower shall maintain records and accounts adequate to None .reflect the operations, resources and expenditures of the project.  Section Covenant Status Original Revised Description of covenant Comments class fulfillment fulfillment date date 4.01 (b) s ul(i) The Borrower shall have the records and accounts including None those for the Special Account for each fiscal year auditcd by independent auditors acceptable to IDA; I C (ii) The Borrower shall fumish to IDA, as soon as available, but None not later than six months after the end of each such year, a certified copy of the report; 4.01 (c) (iv) I C For all expenditures with respect to which withdrawals from the None Credit Account were made on the basis of statements of Expenditures, the Borrower shall ensure that records and accounts are included in the annual report which should contain scparate opinion by said auditors. I = Accotiats/atidits 7 = lnvoluntary Resetlleaineiit C = Complied with 2 = Financial Perltrnnance/revenue 8 = indigenous people CD = Compliance after delay generation from beneficiaries 9 = Monitoring, review, and reporting NC = Not complied with 3 = Flow and utilization ot'project funds 10 = Implementation SOON = Compliance expected