Djibouti - Education Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 11284 PROJECT COMPLETION REPORT DJIBOUTI FIRST EDUCATION PROJECT (CREDIT 1543-DJI) OCTOBER 16, 1992 Population and Human Resources Division South-Central and Indian Ocean Department Africa Regional Office 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 EQUIVALENTS Currency Unit = Djibouti Franc (DF) US$1.00 = DF 177.7 DF 1000 = US$5.6 MEASURES 1 meter = 3.28 feet 1 meter = 10.76 square feet 1 kilometer (1km) = 0.62 mile 1 square kilometer = 0.39 square mile LIST OF ACRONYMS AfDF - African Development Fund BEPE - Education Projects Execution Office CRIPEN - Education Information and Production Center FAC - Aid and Cooperation Fund IDA - International Development Association IREDU - Education Economics Research Institute PCR - Project Completion Report TOR - Terms of Reference UNDP - United Nations Development Programme UNESCO - United Nations Educational, Scientific and Cultural Organization Fiscal Year School Year January 1 - December 31 October - June  FOR OFFICIAL USE ONLY THE WORLD BANK Washington, D.C. 20433 U.S.A. Office of Director-General Operations Evaluation October 16, 1992 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Completion Report on Djibouti First Education Proiect (Credit 1543-DJI) Attached, for information, is a copy of a report entitled "Project Completion Report on Djibouti First Education Project (Credit 1543-DJI)" prepared by the Africa Regional Office with Part II of the report contributed by the Borrower. No audit of this project has been made by the Operations Evaluation Department at this time. Attachment 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.   FOR OmCIAL USE ONLY PROJECT COMPLETION REPORT DJIBOUTI FIRST EDUCATION PROJECT (CREDIT 1543-DJI) Table of Contents Pages Preface i Evaluation Summary iii PART I PROJECT REVIEW FROM IDA's PERSPECTIVE 1 A. Project Identity 1 B. Socioeconomic Context 1 C. Project Objectives and Description 2 D. Project Design and Organization 3 E. Project Implementation 3 F. Project Results 4 G. Project Sustainability 8 H. IDA Performance 8 I. Borrower Performance 9 J. Project Relationships 10 K. Consulting Services 10 L. Project Documentation and Data 10 PART II PROJECT REVIEW FROM BORROWER'S PERSPECTIVE 12 PART III STATISTICAL INFORMATION 33 A. Related IDA Credits (Table 1) 34 B. Project Timetable (Table 2) 34 C. Credit Disbursements Table 3: Cumulative Estimated and Actual Disbursements 35 Table 4: Graph of Planned and Actual Disbursement Schedules 35 D. Project Implementation (V. Annex 1): Table 5: Planned and Actual School Delivery Dates 36 Table 6: Planned and Actual Technical Assistance Dates 36 Table 7: Planned and Actual Training Fellowship Dates 36 E. Project Costs and Financing Table 8: Project Costs 37 Table 9: Project Financing 37 Table 10: Allocation of Credit Proceeds (SDRs) 37 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.  TABLE OF CONTENTS (Continued) F. Project Results Table 11: Direct Benefits. Number and Capacity of Project Schools 38 Table 12: Studies 38 Table 13: Assignment and Responsibilities of Djibouti Counterpart Personnel 39 G. Status of Covenant (Table 14) 40 H. Utilization of IDA Staff Resources Table 15: Man-Weeks per Stage of Project 42 Table 16: IDA Project Missions 42 ANNEX I: Comparative Schedule of Project Implementation 43 ANNEX II: Comparative Table of Estimated and Actual Project Costs 44 ANNEX III: Comparative Table of Project School Indicators 45  PROJECT COMPLETION REPORT DJIBOUTI FIRST EDUCATION PROJECT (CREDIT 1543-DJI) Preface This is the Project Completion Report (PCR) for the First Education Project in Djibouti, for which Credit 1543-DJI in the amount of SDR 5.1 million was approved on March 21, 1985. The credit was closed on June 30, 1991, two years behind schedule. The last disbursement was on December 2, 1991. An undisbursed balance of SDR 122,690, was canceled on the same date, reducing the total proceeds utilized to SDR 4,977,310. The PCR was prepared jointly by the Population and Human Resources Division of the South-Central and Indian Ocean Department (Preface, Evaluation Summary, Parts I and III) and the Borrower (Part II). Preparation of the PCR was begun after two missions to Djibouti, by an architect (January 1992) and an education specialist (February 1992), and is based on the Staff Appraisal Report, Credit Agreement, supervision reports, progress reports from the Borrower, documents prepared by the Education Projects Execution Office (BEPE), correspondence between the Borrower and IDA, and internal IDA memoranda.  6  - iii - PROJECT COMPLETION REPORT DJIBOUTI FIRST EDUCATION PROJECT 'Credit 1543-DJI) Evaluation Summary Proiect Objectives 1. In order to respond to demand for schooling and pave the way for development of the education sector by improving its effectiveness, the Government requested that IDA finance a project through which access to education in disadvantaged urban areas and in rural communities could be expanded, management of the education system as a whole reevaluated and reviewed, and development of the system organized on a sound basis of financial planning and cost control. More specifically, in addition to its proposal for the building or extension of two urban schools and eight rural schools, the Government requested technical assistance and specialization training for Djibouti nationals so as to be able to set up an Education Planning Service within the Ministry of Education. Proiect Implementation 2. The Government issued instructions for the creation of an Education Projects Execution Office (BEPE) as part of the Directorate-General of Education, designating the Director-General as its director and vesting it with responsibility for implementation of the project. It was to be supported by the regular departments of the Directorate itself, a French technical assistant to supervise pedagogical matters, and an administrator/accountant. The construction component of the project was made the responsibility of the technical departments of the Ministry of Public Works. This scheme of organization proved to be very effective. 3. With regard to construction of schools, project objectives were not only achieved but largely exceeded, as 84 classrooms were built and 19 schools either newly built or extended, instead of the 43 classrooms and 10 schools aimed for initially. These additional achievements, following two amendments to the Credit Agreement (January 30. 1987 and July 11, 1990), were made possible by: (a) sound management of the project; (b) major savings resulting from favorable variations in the SDR with respect to the Djibouti franc, which is tied to the dollar, and in the Djibouti franc with respect to the currencies used to pay for imported inputs; (c) competitiveness in the construction sector; and (d) architectural and technical simplifications in construction of the 41 second-tranche classrooms, which reduced per m2 built costs appreciably.  - iv - 4. The buildings put up were of good overall quality and very functional, while the furniture and equipment provided afforded working conditions and a degree of comfort well above those found elsewhere in the region. One matter for regret, however, is that the architectural and technical solutions opted for originally were not less complex, and that community participation was not better organized, with a view to reducing construction and maintenance costs. 5. The education planning and evaluation component, for reasons internal to IDA, was financed under Credit Agreement 1232-DJI (Technical Assistance), the funds in question being administered by UNDP. Despite a relatively slow start (the planning expert had to be replaced after 12 months instead of 24; a Djibouti counterpart planning specialist was recruited only seven months prior to completion of the experts' term of appointment and was immediately absent abroad for four months' training), this component did accomplish the goals set for it. An Education Development Plan was formulated for the 1989-93 period, on the basis of technical and pedagogical studies and data on education-system functioning and costs. The Education Planning Service was also organized, and its current staff of three specialists (planner, evaluator, statistician) have kept the Five-Year Plan updated and produced a statistical yearbook annually. 6. Following an amendment of the Credit Agreement on February 9, 1988, certain funds were reallocated to finance preparation of a second education project (Credit 2080-DJI). This meant that a consultant could be brought to Djibouti for two weeks to assist with launching of the activities of the Education Planning Service, and training obtained for four specialists (a teacher of French as a second language, a schools inspector/supervisor, a designer/model-maker, and a bookkeeper). 7. From the qualitative standpoint, the project has had effects which, although not among its initial specific objectives, are far from negligible. In addition to the results flowing from establishment of the Education Planning Service and provision of training fellowships in primary- school instructional methods, it is evident that the improved conditions created for primary-school pupils cannot help but have an advantageous qualitative impact, and that the foundations now exist for extending it. What still remains to be done is to ensure that other requirements are met: well- trained teachers, appropriate educational strategies, school textbooks that meet pupils' real needs, and a library for teachers and pupils at each school. The second education project has been identified and designed to achieve these ends. Prolect Sustainability 8. With regard to project buildings, their durability depends on the possibilities of maintaining both premises and equipment. Unfortunately, the architectural solutions chosen mean that maintenance activities cannot be left to local communities; neither can the district authorities in rural areas or the technical service departments of the Ministry of Education in Djibouti City be relied upon. Although the Ministry is operating efficiently at the present time, the same is not true of the district authorities, whose  - v - resources are limited. In the circumstances, users must become disciplined, especially where sanitary facilities are concerned, since these could otherwise soon become a source of infectious diseases. 9. Sustainability of the quality effects of the project can be counted upon. Apart from the fact that the school premises constructed will go on being used by more and more pupils, who will benefit from the very favorable working conditions they afford, development of the education system will be planned with increasing effectiveness thanks to the Education Planning Service, particularly as additional assistance from technical and training specialists is provided for under the second project. Conclusions and Lessons Learned 10. The quality of the dialogue that existed between the Government and IDA throughout the project not only enabled both parties to be very precise in identifying problems likely to delay quantitative and qualitative development of the education sector, but also paved the way for the second project now being implemented. The second education project benefitted from assets acquired during the first, in particular with the soundly organized BEPE and the Education Planning Service, which possess capabilities lacking at the time Credit 1543-DJI was launched. However, there are two major lessons to be learned from this first project, related to school construction and to the writing and dissemination of textbooks. On the subject of the architectural and technical solutions chosen for construction of schools and classrooms, greater care should have been taken to adapt design and specifications to local environment and to better involve local communities to construction/maintenance. Although these were partially improved when the 41 classrooms funded from project savings were built, it appears that certain architectural studies actually led the Ministry of Public Works to make recommendations - in respect of school building works only - that have never been taken into consideration. Unfortunately, the Ministry of Education possesses neither the personnel qualified in this area nor the carefully formulated specifications that would enable it to assess the worth of recommendations made to it. Steps therefore need to be taken that will lead to the adoption of solutions valid for all future building programs. On the subject of textbooks, measures should have been included to supplement actions planned or underway for the writing and dissemination of new textbooks of the type begun and pursued in relation to the second project.   PROJECT COMPLETION REPORT DJIBOUTI FIRST EDUCATION PROJECT PART I - PROJECT REVIEW FROM IDA's PERSPECTIVE A. Prolect Identity Project name: First Education Project Credit: 1543-DJI Country: Djibouti Sector: Education Subsector: Primary Education Amount of the Credit: SDR 5.1 million Board approval: June 22, 1985 Anticipated closing date: June 30, 1989 Actual closing date: June 30, 1991 B. Socioeconomic Context 1. This First Education Project was prepared, negotiated and signed at a time when Djibouti's general economic situation was deteriorating after a comparatively favorable period. GDP had begun to decline in real terms by 2% a year in 1984, despite an annual 3% increase over the 1978-82 period. The main reason for this deterioration was a drop in construction and service activities, which account for the major part of the country's GDP, since the primary and secondary sectors account for only 20%. Inevitably, this trend increased the economy's dependence on external aid, which covers over 40% of government expenditure and represents close to 25% of GDP. In order to improve the economic situation and achieve more equitable income distribution, the Parliament decided in 1982 that the country's economic priority should be to enhance the value of its human resources, increase agricultural and industrial production, and strengthen transport and communications services to meet international demand. 2. The role of education and training in achievement of these objectives was a crucial one. Obviously, the key problems affecting the sector had to be dealt with as soon as possible. The first step was to increase the primary-school enrollment rate by 41% in 1983/84, by quickly reducing the inadequacy or lack of access to basic education in disadvantaged urban districts and in villages in the interior. The second step would be to improve the effectiveness of education, adapting it better to the needs of the country but maintaining its quality. The final step would be to set up an education planning and evaluation service able to ensure harmonious development of the sector as a whole, given prevailing financial constraints and the need to sustain quality. In 1984, the proportion of the national budget allocated to education was 12%, a figure the Government proposed to  - 2 - increase only at the same pace as the rate of growth in the national budget, roughly 1% per annum. With this restricted budget, and unit costs that were very high in comparison to those in the region (approximately US$372 per pupil-year in primary education in 1985), any palliative measures would clearly have to meet the demands of strict planning and control. C. Prolect Obiectives and Description 3. Given the range of problems to be dealt with and the limited financial resources available for the purpose, the Government formulated a number of high-priority requests for presentation to the Donors' Conference held in Djibouti in November 1983. Following bilateral discussions between representatives of the Government and various IDA missions, it was agreed that the project would have a twofold objective: to respond to the increasingly pressing demand for education, and to lay the foundations for a long-term management and planning system designed to lead to eventual qualitative and quantitative improvements. Besides seeking to increase the possibilities of access to primary education in disadvantaged urban areas and in rural communities in the interior, the project would also introduce mechanisms for formulating short-, medium- and long-term development plans, to be based on continuous evaluation of the prevailing situation and to include formulation of priorities, estimation of the resources required, assessment of the cost of these resources, and mobilization of the financing to cover them. 4. More specifically, the project was to allow for: (a) construction and equipping with teaching materials and furniture of (i) two school complexes, each with 12 classrooms, in disadvantaged urban areas, plus housing for the principal of each school, and (ii) eight rural schools, four of them new and four to be extended, plus 20 housing units for principals and teachers and facilities for 650 boarders; (b) technical assistance with: (i) evaluation of results obtained through existing education programs; (ii) planning of the education/training system as a whole; (iii) training of Djibouti nationals as specialists in the fields of education planning and evaluation. 5. Subsequently, three amendments to the Credit Agreement expanded the specific objectives of the project as follows: (a) January 30. 1987: Financing for the construction and equipping with teaching materials and furniture of three additional schools, one urban and two rural, plus two housing units for teachers.  -3- (b) February 9. 1988: The allocation of Credit proceeds for technical assistance was altered to include preparation of a future second education project. (c) July 11. 1990: In order to utilize the undisbursed balance of the Credit proceeds, six additional rural schools were to be extended, five teacher housing units constructed in addition to the 22 already provided for, and accommodation for 50 additional boarders provided. D. Prolect Design and Organization 6. Design of the project began after receipt of a request from the Government in January 1984 which emphasized the need to go ahead rapidly with (1) strengthening of education-system organization and management, and (2) expansion and improvement of primary education. In actual fact, only financing for expansion of primary education was requested from IDA, as the qualitative components of the project were to be financed under cooperation arrangements with France. However, the request noted that improved working conditions, in the form of appropriately designed and equipped schools, would inevitably affect the quality of the education and training provided. 7. Before this final concept of the project was arrived at, a sustained and productive dialogue had taken place between the Government and the various IDA missions. By 1979, roughly two years after the country achieved independence, such enormous needs in the fields of education and training had been identified that an IDA Credit alone would clearly not be sufficient to finance a project targeting all of them. Prior to the Donors' Conference in November 1983 (para. 3), the initial IDA missions, in agreement with Djibouti officials, had considered a project to strengthen vocational and technical training as well as to improve the quality of primary education. But other donors, specifically France's Fonds d'aide et de coop6ration (FAC) and the African Development Fund (AfDF), had indicated their own interest in these two arenas; in the circumstances, the Government asked IDA to concentrate its assistance on achieving a major increase in the number of places available for pupils in primary education and on instituting sector development structures and strategies. 8. During the project preparation period, the funding planned by IDA was cut from US$6 million to 5 million. At roughly the same time, Saudi Arabia, which had agreed to provide approximately US$4 million for the building of primary schools, announced that it was forced to postpone this financing. Given these difficulties and in order to minimize their impact, IDA decided to add back approximately US$500,000 to its contribution by financing the evaluation/planning component through its Technical Assistance Project (Credit 1232-DJI). Accordingly, this component is dealt with in Schedule 2 to Credit Agreement 1543 under the heading "Technical Assistance," but is not mentioned in Schedule 1, which deals with expenditure categories. E. Prolect Implementation 9. As the results achieved demonstrate, the project as a whole was carried out satisfactorily. Its implementation was entrusted from June 1984  to the Education Projects Execution Office (BEPE), the director and alternate director of which were the Director-General of Education and the Chief of Administrative and Financial Services, respectively. Subsequently, in 1986, BEPE personnel was completed by the addition of a team of education and architectural specialists (drawn from the Ministry itself or assigned by the French Cooperation authorities) to the group already appointed. Also added were an accountant financed by FAC, and a secretary and driver, both paid from Government counterpart funds. The fact that the Directorate-General of Education was responsible for major project decisions proved very advantageous, since it meant they were reached quickly, that construction works could go ahead more rapidly, and that substantial savings on Credit proceeds were possible (later reallocated for additional works). 10. Construction works were the responsibility of the technical branches of the Ministry of Public Works, which also attended to architectural and technical studies, preparation of tender documents and award of contracts, and supervision of operations in the field. These services were not billed to BEPE, because the use of the Credit was limited in this instance to purchase vehicles for worksite supervision and a small lot of office equipment. However, in putting up the first 10 schools the Ministry simply recycled existing plans, the specifications for which were written to a higher standard than necessary. Although rectifications were made later when the additional project facilities were built, they were not carried far enough. 11. Since the Government had no experience in the selection, recruitment and administration of specialists, consultants and fellowship holders, implementation of the technical assistance component was entrusted to the Programme Operations Department of UNDP, under an execution agency contract between the Government and UNDP signed on June 11, 1985. IDA transferred a sum of US$490,000 to UNDP. These arrangements gave rise to no particular procurement problems or difficulties, but did not meet the expected cost/benefit (III.3). F. Proiect Results 12. Buildings and equipment: The initial quantitative goals of the project were far exceeded, since 84 new classrooms were built instead of the 43 proposed at the appraisal stage. Buildings, whether classrooms or ancillary facilities (staff housing, boarder accommodation, canteens, sanitary facilities), are soundly constructed on the whole, while the materials and fittings used guarantee working conditions and standards of comfort clearly superior to those found elsewhere in the region. However, a number of reservations need to be voiced. For one thing, the architectural and technical solutions initially chosen did not always guarantee that subsequent maintenance costs would be low: roofs and eaves with too little slope, classrooms opening directly onto corridors, poor layouts/ground plans, and types of sanitary facility not suited to local practices. Opportunities to lower construction costs by writing specifications to more economical standards were not investigated closely enough: for instance, considerable savings would have been possible on number and types of windows, ceiling materials used, superfluous electrical fittings and water-supply points, and alignment and characteristics of fencing/enclosures. Finally, although  -5- anticipated in the Appraisal Report (paras. 2.21(c), 4.01 and 6.06), there has been little or no participation by local communities in the building and maintenance of their own schools. The design specifications followed (since the authorities were not willing initially to revise their standard plans) called for materials and skills not readily available, particularly in rural areas, and therefore are likely in the short term to mean that even limited participation by beneficiary communities cannot be expected. 13. The project allocated funds for the services of a consulting architect and a specialist in construction engineering (Category 3 in the Credit Agreement) to undertake civil engineering studies. These resources were not utilized, the technical branches of the Ministry of Public Works, despite the FAC technical assistance available to them, being content initially to rely on existing designs and specifications, although during the second phase of the project - at IDA's request - they were modified in the interests of cost reduction and better adaptation to the local environment. It would be advantageous, as far as future building programs are concerned, if this effort and the studies accompanying it were carried further. 14. All school furniture supplied through the project was imported. Particular care was taken to see that it was suited to children of different sizes and ages. It is of acceptable quality, although with variations in this regard from one supplier to another. There were certain instances of over- equipping also: for instance, the number of fans provided in each classroom; and installation of electrical fittings and artificial lighting in some localities where power supply is not available during school hours. In contrast, the schools are without libraries. 15. Cost: Total project cost, including contingencies and taxes, was estimated at US$7.35 million, broken down as US$3.2 million in local currency costs and US$4.2 million in foreign exchange costs. By the closing date of the project, the actual cost figure was approximately US$7.9 million, or 7% more than anticipated. This difference is attributable first of all to significant overruns (+349%) in project administration expenses, mainly as a result of the longer duration of the project (5.5 instead of the anticipated 3.5 years), and then to the fact that the proceeds of this Credit were used to cover technical assistance and fellowship expenses initially planned as chargeable to the second education project (Credit 2080-DJI). The project closing date was twice postponed (to April 11, 1989 and April 5, 1990) to allow time for the investment of major undisbursed balances of Credit proceeds accumulated as a result of favorable coinciding variations in the SDR in relation to the Djibouti franc (tied to the U.S. dollar) and in the Djibouti franc in relation to currencies used to pay for imported products (procured mainly in the franc zone). The proceeds of the Credit covered 82% of project cost. 16. Unit civil engineering costs fell in the range US$486-680 per m on new construction by private contractors (average: US$564 m2), and in the range US$350-588 per m2on works executed on force account (average: US$473 per mi2). These figures may be compared to the US$600 per m figure anticipated at the project appraisal stage. These good results were achieved thanks to favorable exchange rates (para. 15), competition in the construction sector,  - 6 - and certain simplifications and improvements in building plans and specifications at the time additional classrooms were built. 17. Technical assistance and training fellowships: This component, financed as indicated above (para. 8) from Credit 1232-DJI proceeds administered by UNDP, was designed to improve management of the education system through effective planning, careful evaluation of operations, and cost control. Two stages were foreseen: initially, assignment of two experts, one in education planning and the other in education evaluation, plus training of Djibouti counterpart personnel; subsequently, creation of a permanent unit as part of the Ministry, to be responsible for system monitoring. A number of events impeded implementation of this component: the education planner stayed only one year in Djibouti instead of two, because of poor performance, the period of assignment of the evaluation specialist being extended to enable him to attend to both the evaluation and planning functions; recruitment, toward the end of the project, of a planning consultant for a period of only four months, and late appointment of a Djibouti counterpart to take over his functions on departure of the technical assistance team. In all, 46 man- months of technical assistance and 4 man-months of training were financed under this component, instead of the 36 man-months and 24 man-months, respectively, initially planned. 18. Despite these difficulties, the results achieved through this component may be considered acceptable: As well as a series of studies on costs, the Education Ministry budget, and certain educational aspects of primary-school instruction (V. Table 12, Part III), a draft Five-Year Education Plan (1989-93) was also developed. In addition, the Education Planning Service was set up as part of the Directorate-General; it is at present staffed by three professionals, namely a planner, who heads it, a statistician, and an evaluator. All have undergone or are scheduled to undergo specialized training. 19. Not included in the original project description were 0.5 man- months of consulting services and 30 man-months of training fellowships, funded from undisbursed Credit proceeds. This consultant's mandate was to bring the Planning Service into operation and monitor the performance of its head officer, who had received part of his training from the same consultant through the Education Economics Research Institute (IREDU). The fellowships went to four individuals: two education specialists (one in teaching French as a second language, the other in teacher training), a designer/model-maker, and a bookkeeper. Award of these fellowships, which were originally to have been funded under the second education project, enabled the first project to be implemented more rapidly. 20. Quantitative results (V. Table 11 and Annex III. Part III): In quantitative terms, the Appraisal Report set the following goal: 2,350 new pupil places, plus 1,800 additional places to have been created with financing from the Saudi Fund. Although this contribution was finally not available, it was partially replaced by financing an extension of the IDA project from Credit proceeds that had been saved. The total number of new pupil places made available to the Ministry was 5,226, or 25% more than anticipated under the two combined IDA and Saudi arrangements. While only 93% of this new capacity is  currently being used, the enrollment ratio appears to have risen from 41Z to 46%, although the reliability of these figures is no more than relative, as the census of Djibouti's school-age population is very approximate. However, enrollment rates for girls, which had been 45% in urban areas and 33% in rural areas, do not appear to have changed appreciably. Finally, each project school also provides an adult literacy course; the total initial enrollment figure here is 625 adults, which can be expected to increase. 21. Qualitative results: At the time the project was being appraised, it was one of a number of actions planned to deal not only with qualitative problems in the education field but also the gap between the kinds of training provided and the country's real needs in this regard. As funding for this latter area had already been committed by other donors (para. 7), it seemed logical for the project to focus mainly on quantitative development of the primary education subsector. All the same, there is reason to regret that, for example, no measures were included to supplement actions planned or under way for the writing and dissemination of new textbooks, since this would have facilitated a continuing dialogue on qualitative issues, of the type begun and pursued in relation to the second project. The impact of this first project on qualitative improvements in the education system in general and the primary education subsector in particular was, in any event, not negligible. 22. In the first place, the quality of school and classroom construction, furniture and equipment funded through the first project has an evident impact on the value of the education provided. All pupil and teacher requirements are now met: agreeable premises that are well lit and well ventilated, suitable furniture, a place and a desk for every child, and a reader for every pupil. Fittings in the form of cabinets, shelving and the like mean that textbooks are kept in good condition, as are the few library books available, although they are often of the wrong level for the pupils who might otherwise use them. Unfortunately, the plans made with IDA to begin collections of library books in the project schools came to nothing, as a result of procedural delays; they would have provided the right kind of impetus and laid the foundations for larger libraries, particularly as the French Cooperation authorities had promised financing to further this effort. 23. In addition, the training fellowships funded for an inspector/advisor to work with primary school teachers and a specialist in the teaching of French (para. 19) reinforced project impact in improving primary education in the qualitative sense - an impact that should come to be fully felt during implementation of the second education project (Credit 2080-DJI). 24. Creation of the Education Planning Service within the Ministry has proved to be the most worthwhile of the project's qualitative results. The experts and consultants in education planning and evaluation examined the structures, operation and costs of the education system and proposed guidelines for its development. These provided the foundations for an Education Sector Development Plan - formulated partly by a consultant and partly by his Djibouti counterpart - which the Planning Service updates each year. The Government regards this document as the master plan for education system development and as a reference basis not only for national officials but also for donors.  - 8 - 25. In conclusion, it may be noted that the education evaluation specialist, in addition to his other tasks in Djibouti in planning and in compilation of baseline data for the sector Development Plan, also devised and tested (in conjunction with educational advisors assigned by the French Cooperation authorities) a battery of achievement tests for first-, third- and fifth-graders at eight Djibouti-City schools. One of their most pertinent findings - that a trainee teacher whose work is well supervised by inspectors obtains better results than a fully qualified teacher who has undergone conventional teacher training - pointed up the need to make greater use of teacher supervision and to train Djibouti nationals in this field, concerns which have been taken up in the second education project (Credit 2080-DJI). G. Proiect Sustainabilitv 26. As far as project buildings are concerned, their durability is a question of maintenance of both premises and equipment. In general, considerable effort has gone into making school surroundings and playgrounds agreeable and keeping them in good order - thanks to the availability of water and tools, and the motivation shown by school principals in this regard. However, shortcomings in school design (para. 12) mean that maintenance problems cannot be attended to in rural areas, given their derisory budgets and parents' low incomes. The situation is better in Djibouti-City, where the Ministry of Education has a mobile team of skilled workers and parents contribute more often to school cooperatives. Nevertheless, in urban and rural areas alike, urgent action is needed to clean up sanitary facilities, which threaten to cause serious epidemics if left in their present state. 27. The sustainability of the education planning/evaluation component is both evident and promising: the organization chart of the Directorate- General of Education now includes the Education Planning Service; the Service has been assigned a high-quality staff group whose skills can be expected to increase under the supplementary training arrangements included in the second project; and plans exist to commission experts and consultants to advise the Service on necessary groundwork in the areas of school mapping, cost analysis, statistical data collection, organization and administration, and examination arrangements. It is to be hoped, however, that this initial impetus is not undermined by delays in recruiting technical assistance personnel and national counterpart staff. Finally, the framework set up to generate the data used in formulating the Education Development Plan is extensively used in updating the Plan annually, with every reason to expect this will continue in the coming years. H. IDA Performance 28. The main strengths in IDA performance were: gradual creation of a situation of open dialogue which enabled the various pre-preparation missions to gain a better understanding of the Government's needs in education, and to change their initial recommendations significantly in light of other donors' intervention programs. The comparatively high number of supervision missions, two or three annually, together with the almost continuous presence on the scene of a project education officer, prolonged and expanded this dialogue to the extent that a new project was identified as a logical extension of the  - 9 - first. Mission frequency and continuity also accounted for the rapid acceptance of various Credit Agreement amendments authorizing additional works, recruitment of consultants, procurement of equipment not originally provided for, and training opportunities for Djibouti counterpart personnel. 29. On the other hand, one important aspect of the building program merited closer attention from IDA, in the sense that greater pressure should have been brought to bear on the Government to carry out the architectural studies provided for and thus arrive at better and less costly building designs. Recommendations to this effect were made in memoranda from the IDA architects but were not acted upon until the plans for the additional project schools were finally redrawn. Along the same lines, although IDA showed great flexibility when it came to reallocation of undisbursed Credit proceeds, it unfortunately failed - because of delays in the drawing up of booklists (pedagogical arguments as to what should be included) and in awarding contracts - to meet the Government's requests for library books for new schools, an item not originally included in the project. School libraries still leave much to be desired in terms of both number of books and their quality, and an additional shipment (based on a list agreed on by the Government and IDA) would be a great help. However, this gap could probably still be filled: just as certain funds were committed under Credit 1543-DJI to the second project, the same type of arrangement might be used in reverse when equipment is being procured for the new schools built under 2080-DJI. I. Borrower Performance 30. As noted briefly already, the Government's administration of the project through BEPE proved effective: continuous contact with IDA and UNDP headquarters, and successful although difficult supervision of work sites (sometimes remote and hard to reach) and of deliveries and installation of equipment; careful keeping of accounting records; regular dispatch of half- yearly progress reports and annual audit reports; and inclusion in the budget of local counterpart funding as scheduled. The Covernment also complied promptly with Credit Agreement conditions. Another point to the credit of the Borrower was its consistent availability during mission periods and through exchanges of correspondence, factors that accounted for the rapid advance with identification and appraisal of the second education project. 31. There are two areas in which the Borrower could improve, however. In the first place, a concerted effort should be made to improve school infrastructure by issuing specifications and instructions for architects that reformulate objectives and requirements to be met. Obviously, these objectives should not be limited to educational concerns, but should extend to cost containment and greater reliance on local resources; similarly, requirements should include that of limiting maintenance needs, with a view to transfer of this responsibility to the local level. Secondly, it would be advisable to recruit national counterparts early enough to enable them to obtain instruction in theory abroad and practical training locally from project technical assistance personnel.  - 10 - J. Proiect Relationships 32. Dialogue between the Government and IDA was ongoing, frank and direct, and carried over as well into preparation, appraisal and launching of the second project. Relationships with other organizations, both bilateral and multilateral, were also excellent. With UNDP, complete understanding was reached on all aspects of the education planning/evaluation component. The same was true in the case of the technical cooperation provided by France, which during the course of the project agreed to finance the accountant position and authorized the technical advisor to the Director-General of Education to assist in implementing and monitoring the technical assistance and training component. This collaboration continued where the second project was concerned - under similar arrangements with France, and under parallel financing arrangements with UNDP in respect of the education planning component. Very good contacts were also maintained with the African Development Fund. K. Consulting Services 33. Construction works carried out by the private sector gave generally satisfactory results as regards scheduling, cost, and quality of final product, so much so that contracts for additional works in rural areas were also awarded to the private sector. Procurement procedures were in accordance with IDA guidelines. 34. The project allocated funding for the services of specialists to carry out architectural studies for the improvement of school buildings through cost reduction, incorporation of techniques for utilization of village manpower contributions, and facilitation of maintenance and repairs. However, specialists of this type were never recruited, although the studies were eventually conducted by the technical branches of the Ministry of Public Works in connection with the second phase of the project. Unfortunately, the Education Ministry has never been able to develop appropriately detailed specifications, and is without personnel qualified and experienced enough to assess the feasibility of proposals received and to arrive at their own recommendations. L. Project Documentation and Data 35. The project was carried out on the basis of the usual documents: Credit Agreement (March 21, 1985), Staff Appraisal Report (December 3, 1984), and President's Report (January 1985). These documents had been preceded by an "Application from the Government of the Republic of Djibouti to the World Bank" (January 1984), which contained a detailed analysis of the education sector and proposals for its development. Project documentation also included a detailed working document prepared by an IDA architect with particulars of built areas, costs, and equipment, building by building. In addition, a four-page undated document set out terms of reference for experts and fellowship awardees, in addition to a detailed cost estimate on the planning/evaluation component. These materials were utilized during project execution.  - 11 - 36. Information incorporated in the Project Completion Report was obtained from the Education Planning Service, BEPE, or IDA archives. Supervision reports were clear and complete and included voluminous technical information. As far as future supervision missions are concerned, it would be preferable to deal with the problems of only one project in each report and not to combine recommendations affecting various projects under way or contemplated. In addition, BEPE's regular half-yearly progress reports throughout the project period provided excellent reference materials.   - 12 - PART II - PROJECT REVIEW FROM BORROWER'S PERSPECTIVE Diibouti Ministry of Education Education Projects Execution Office (BEPE) PROJECT COMPLETION REPORT DJIBOUTI FIRST EDUCATION PROJECT (CREDIT 1543-DJI) I. PROJECT DESCRIPTION AND START-UP 14 1. Background 14 2. Project Identification and Objectives 14 3. Project Components: 15 (a) Urban primary schools 15 (b) Rural primary schools 15 (c) Technical assistance 15 4. Project Organization and Implementation: Ministry of Education 16 II. PROJECT DESIGN, EXECUTION AND IMPACT 16 1. Project Design and Execution 16 2. Major Results: 16 (a) Infrastructure 16 (b) Furniture and equipment 17 (c) Training 17 (d) Institutional development 17 (e) Expansion of primary education 17 (f) Study of the education sector 18 3. Project Coordination and Administration 18 (a) Responsibilities of coordinator 18 (b) Relationships between project coordinator and component managers 19 (c) Relationships with different agencies and ministries 19 (d) Lessons learned 19 4. Main Project Impact on Education System 19  - 13 - III. REVIEW OF BANK AND GOVERNMENT PERFORMANCE, AND TECHNICAL ASSISTANCE 20 1. Evaluation of Bank Performance during Project Execution 20 2. Action of the Government during Project Execution 21 3. Evaluation of Performance of Institutions and Consulting Firms 21 IV. PROJECT ECONOMIC AND FINANCIAL EVALUATION 22 1. IDA Participation in First Education Project (table) 22 2. Project Financing (table) 23 3. Project Implementation (table) 24 4. Procurement of Goods and Services (tables) 25 5. Recurrent Project Costs (table) 29 6. Foreseeable Project Impact on Education System 30 7. Lessons Learned 30 V. PROSPECTS FOR A NEW PROJECT 30 1. Primary Level 31 2. Secondary Level 31 3. Private Education 31  - 14 - I. PROJECT DESCRIPTION AND START-UP 1. Background Situated in a hot, arid region, the Republic of Djibouti possesses extremely limited natural resources. Its chief assets are strategic location, 370 km of coastal fishing grounds, and geothermal resources now in the course of being evaluated. From the economic viewpoint, a general distinction may be drawn between: (a) Djibouti City (with a population of approximately 200,000 at project start-up), with its sea/rail terminal, international airport, administrative services and almost all the country's commercial activities; (b) and the extremely poor districts of the interior, where the majority of the rural population (estimated at 160,000), nomadic or semi-nomadic, is spread over four rural districts (Ali-Sabieh, Dikhil, Obock, and Tadjourah). 2. Project Identification and Objectives On accession to independence in 1977, the Republic of Djibouti immediately set about enhancing the value of its human resources, indispensable if a service economy was to be created and general living conditions improved in the interests of social justice. Its first step in this direction was to promote access to education for all children of primary- school age, particularly girls, since an educated female population would be effective in curtailing population growth. In order to increase the capacity of the country's schools, the Government began a two-shift system of classes, relying for additional teaching staff on assistants for whom vocational training was organized simultaneously. These measures, however, proved insufficient and short-lived, since it soon became apparent that the basic school infrastructure would have to be developed to respond to social pressures for education created by the rural exodus, stabilize the rural population, and improve distribution of schooling opportunities throughout the country. Logically enough, these objectives led to a need for education planning and for management of the process already instituted to improve the cost-effectiveness and quality of the education system. A working group of representatives of the Ministries of Education and Public Works, assisted by a UNESCO mission (economist, educator, architect, and mission secretary), formulated a primary education development project as part of general education development policy, one of the Government's primary concerns.  - 15 - This project was presented, among others, to the 1983 Donor's Conference in Djibouti, where it attracted the attention of the World Bank, which agreed to finance it through IDA Credits 1543-DJI and 1234-DJI, the latter focused on technical assistance. 3. Project Components The First Education Project was made up of three segments: (a) Urban primary schools: Construction and equipping of two school complexes in the capital, as part of a plan to increase the number of places for urban pupils in response to heavy concentration of the population in major towns, particularly Djibouti City. These two schools, located in Gabode and Balbala, each consisting of 12 classrooms and a teacher housing unit, would allow enrollment or relocation of 1,200 pupils and a very marked reduction in double-shifting. (b) Rural primary schools: Construction or expansion and equipping of eight rural schools (plus housing for 20 teachers and accommodation for approximately 650 boarders), as follows: - construction and equipping of four new schools, at Bondara, Kontali, Khor-Angar and Ardo: extension and equipping of four existing schools, at Dammerjog, Mouloud, Ali-Add6 and Dorra. By making it possible to enroll or relocate 840 pupils, this component would contribute appreciably to reducing rural illiteracy and curbing the exodus toward larger population centers and urban areas. All rural schools were to be provided with refectory facilities, as a means of introducing children to group life and concepts of hygiene, and as a step in the campaign against malnutrition. In addition, since school premises were to be designed so that girls could be taught in separate classes, an increase in their enrollment rates was expected. (c) Technical assistance: The goal of this component was to make specialist planning services available to the Ministry of Education, to assist it in setting priorities and identifying goals and objectives for coordinated development of the different levels and types of schooling and training that would be consistent with the country's development needs and prevailing financial and other constraints. To this end, provision was made for recruiting specialists to organize education-system planning and evaluation, and for training fellowships for the Djibouti counterparts to these specialists.  - 16 - 4. Proiect Organization and Implementation: Ministry of Education An Executive Order dated 7 June, 1984 created within the Ministry of Education an Education Projects Execution Office (BEPE) consisting of the Ministry's Director-General of Education and Head of Administrative and Financial Services, as well as representatives of the Ministries of Finance and Public Works to provide financial and technical monitoring. Day-to-day project management was delegated to existing departments of the Directorate- General of Education. In 1986, BEPE was strengthened by the addition of a special team made up of Ministry of Education staff, technical cooperation personnel, and an architect. A bookkeeper, secretary and driver were also recruited for project purposes. In September 1988, project administration and accounting were assigned to a member of the French technical assistance group. II. PROJECT DESIGN. EXECUTION AND IMPACT 1. Proiect Design and Execution The design of each of the project components proved to be satisfactory, and their implementation was generally in line with initial cost estimates. Substantial savings were achieved thanks to strict management of Credit proceeds. These, in addition to sums generated by favorable coinciding exchange-rate movements during the project period, resulted, on completion of implementation, in an undisbursed balance of approximately SDR 2,300,000, or 45% of the original amount of the Credit. This remainder made it possible to fund a component identified initially as part of the project, but which the Saudi Development Fund could unexpectedly not finance. This involved the construction and equipping of three school complexes, one in Balbala, part of Djibouti City, and two in main district towns, namely Dikhil and Tadjourah. Even after this component was implemented, a further undisbursed balance equivalent to SDR 600,000 made it possible to extend five existing rural schools (in Kalfa, Sagallou, Moudo, Guirori, and Waddi) and advance funds to cover certain urgent preparatory operations for the second education project. 2. Malor Results (a) Infrastructure - Construction of five urban school complexes, each with 12 classrooms plus housing unit, office, storehouse, sanitary block, maintenance shop and enclosure/fence.  - 17 - Construction of five rural schools, each with three classrooms plus three housing units, refectory, dormitory, office, sanitary block, maintenance shop, and enclosure/fence. Extension or rehabilitation of nine existing schools through the building or renovation of nine classrooms, 15 housing units, three sanitary blocks, one office, one storehouse, two refectories, two dormitories, and two enclosures/fences. In all, these works provided a total of 84 new classrooms. (b) Furniture and equipment: The project financed the full furnishing and equipping of all buildings put up, except for teacher housing units (equipped and furnished at Government expense). The project also financed the purchase of 12 vehicles (for use on job sites and by Government officials), job-site equipment (including six cement mixers), and office furniture and requirements (including computer equipment) for BEPE. (c) Training: This component called for organizational and administrative strengthening of the education system. The training provisions it included covered 24 man-months of fellowships in project design, appraisal and execution for two Djibouti counterparts. Difficulties in recruiting a planning expert who met the TOR criteria, and also Djibouti counterpart personnel, led to partial and delayed achievement of goals in this arena. As a result, a suitable candidate for the evaluation function could not be identified, while recruitment and training of a planner were completed only after the experts had wound up their missions. The planning specialist was trained at the Education Economics Research Institute (IREDU) during the last three months of 1989; this was followed by bringing an expert from that institute to Djibouti for two weeks, mainly to assist in setting up and computerizing the new planning service. (d) Institutional development: The main thrust of this component was to increase Ministry of Education management capability in medium- and long- term forecasting, project design and execution, and education planning and programming. Inclusion in the Ministry's organization chart of the Education Planning Service achieved this goal at the institutional level. This unit, incorporating two divisions (one concerned with schooling organization and private education, and the other with studies and evaluations), was gradually provided with the staff and materials it needed to carry out these tasks. (e) Expansion of primary education: The building and rehabilitation works financed through the project resulted in the following increases in enrollments and staffing:  - 18 - PUPILS TEACHERS SCHOOLS Pre-Project Post-Project Pre-Project Post Project 1986-1987 1991-1992 1986-1987 1991-1992 Gabode II 0 940 0 21 Balbala II 0 1,193 0 24 Balbala III 0 1,131 0 24 Dikhil 1 0 869 0 19 Tadjourah II 0 388 0 9 Ardo 0 97 0 3 Dorra 91 121 3 5 Bondara 0 70 0 2 Kontali 0 65 0 2 Dammerjog 240 270 6 7 Khor-Angar 0 59 0 2 Mouloud 211 199 7 6 Ali-Add6 165 155 6 4 Moudo 102 29 2 2 Guirori 16 61 1 3 Waddi 45 72 2 3 Kalaf 0 124 0 3 Sagallou 184 231 6 6 TOTALS 1,054 6,074 33 145 To these figures, add the following job positions: 5 school principals, 9 Arabic teachers, and 31 service jobs (nightwatchmen and maintenance workers). (f) Study of the education sector: Through the studies launched by the evaluation and planning experts in conjunction with Ministry of Education staff, and their continuation by the new Education Planning Service set up as part of the Directorate-General of Education, an overall structural and quantitative concept of the education system, as well as the rudiments of a mechanism to evaluate its internal productivity, were developed. In addition to examining scenarios for quantitative development of the system based on various working hypotheses, these studies also led to inclusion in the Education Sector Plan of a concern with qualitative development of the system through better adaptation of curriculums and teaching methods to Djibouti's particular circumstances. 3. Project Coordination and Administration (a) Responsibilities of coordinator: The project coordinator was the Director-General of Education, who oversaw implementation and monitoring of the project in conjunction with a group of his deputies assigned to manage its various components.  - 19 - (b) Relationships between project coordinator and component managers: Architectural design studies, preparation of construction contract bidding documents, and monitoring of job-sites were matters assigned, at the request of the Government, to the technical branches of the Ministry of Public Works, whose Director of Housing and Urban Development, in his capacity as Alternate Director of BEPE, oversaw the technical side of project construction works. Procurement of equipment, reception of goods supplied, and their installation were supervised by the Chief, Financial Services Department, Directorate-General of Education, who in his capacity as assistant to the Director of BEPE oversaw administrative and financial matters arising out of the project. The Technical Advisor to the Director-General of Education, as the official responsible for strictly educational matters, oversaw implementation and monitoring of project technical assistance and training components. Project accounts were kept, invoices paid, and progress reports issued by BEPE personnel. On the whole, these operations proceeded without any significant problems affecting the relationships involved. (c) Relationships with different agencies and ministries: The various ministries and government agencies associated with the project made valuable contributions to its sound implementation, despite some administrative delays and red tape, mainly in the Ministry of Finance, which supervised administration of Credit proceeds and settlement of suppliers' invoices. On occasion, proceeds were temporarily blocked and accounting vouchers redeemed only with difficulty, since the Director of Finance, a member of BEPE, did not play his part fully. (d) Lessons learned: In order to improve the administration of contracts whose provisions have already been negotiated precisely and in detail and been approved beforehand by the Government, simplification of Ministry of Finance administrative channels would facilitate adherence to implementation timetables. 4. Main Project Impact on Education System Through its first component, the project provided the means by which a clearer picture of the organization, operation, and cost of the education system can be obtained. The instruments and procedures recommended by the project experts and developed by the Djibouti planning counterpart led to formulation of the draft Education Sector Plan adopted by the Government as its master plan for education-system development. Updated every year to reflect both demand for schooling and physical and budgetary constraints, it provides a framework giving a perspective on projects submitted for decision to the country's policy authorities, while also incorporating those currently in the course of execution. Not only an instrument for management for the status quo but an aid in decision-making as well, the Plan will serve from now on as a permanent reference point for national education policies.  - 20 - The second component of the project, under which primary-school complexes in urban areas and rural regional schools were built and equipped. allowed a response to increasing demand for primary schooling but not at the expense of the material conditions of education. More particularly, the savings achieved on construction and equipping of the classrooms provided for in the original project meant that undisbursed proceeds from the Credit were available to finance other facilities rendered necessary by the increase in the school-age population. Generally speaking, these buildings, in combination with the move by the Government to recruit and train teachers and inspection/supervision personnel, have resulted in a significant improvement in enrollment rates, despite the rise in population growth. The project regional schools allowed a satisfactory response to the problem of nomadic pastoral groups, by providing schooling opportunities in the vicinity of water-supply points. Because of their proximity to the most frequented camping sites, they encourage the schooling of girls. Finally, the school canteens established are not only a major factor in the campaign against malnutrition but also a means of providing social training and instruction in the rudiments of hygiene. III. REVIEW OF BANK AND GOVERNMENT PERFORMANCE. AND TECHNICAL ASSISTANCE 1. Evaluation of Bank Performance during Project Execution The Bank facilitated implementation of the project by agreeing rapidly to the two applications for amendments submitted to it, one for utilization of undisbursed proceeds and the other for reallocation of funds. Responses to most applications for non-objection were rapid and the observations accompanying them instructive. However, the purchase of books from undisbursed proceeds could not be organized in the time available owing to Bank prescriptions regarding choices of textbooks and structure of libraries. BEPE had to revise its bidding documents in the matter completely, although they had been drawn up on data provided by educators and teacher-training advisers, both local and attached to technical assistance groups. The Bank monitored the project on a regular basis, as may be seen from the following list of missions: Missions by IDA Representatives Nov 26-Dec 5, 1966: Mr. A. Stam March 15-20, 1987: Mr. D. Rouag Oct 23-30, 1987: Mr. D. Rouag Feb 18-March 10, 1988, to preappraise second and supervise first project: Messrs. Rouag, de Gaiffier, Bourgeois, Medus, Mezel and Mrs. Kovatcheva, the last four from UNESCO  - 21 - May 22-26, 1988: Mr. Rouag Oct 16-23, 1988: Mr. Rouag May 18-June 4, 1989, to appraise second and supervise first project: Messrs. Rouag, de Gaiffier, Papineau, and Berruer (UNESCO) Nov 20-23, 1989: Mr. Papineau Feb 8-10, 1990: Messrs. Rouag and Colliou Oct 29-Nov 4, 1990: Mr. Rouag May 12-15, 1991: Mr. Rouag Missions to Washington by BEPE Personnel April 9-15, 1988: Mr. Ali Ismael Hassan, BEPE accountant (Seminar on disbursement proceedings) May 7-11, 1990: Mr. Ibrahim Ahmed Moussa, BEPE Alternate Director (Seminar on disbursement proceedings) 2. Action of the Government during Project Execution The Government honored its commitments. Ministerial reshufflings and replacements of the Minister of Education had no adverse impact on the project, since the Government had adopted the Five-Year Plan proposed by the Ministry of Education as its master plan. All decisions concerning BEPE development and reorganization were taken at the appropriate moments. The Government's broadly calculated counterpart obligations were carefully incorporated in the national budget in accordance with the due dates for their disbursement. The Minister of Finance agreed to authorize utilization of undisbursed proceeds for the 100% funding of school buildings not approved [sic] by the Bank. No problems impaired relations between the Bank and the Government, as both shared a common interest in ensuring the success of the project. 3. Evaluation of Performance of Institutions and Consulting Firms The first education project called for the intervention of two experts. One of them, with a two-year appointment, was to advise on sector administration and forward planning. However, the first person recruited remained with the project only six months, as his work was not on a par with his terms of reference. Credit proceeds allocated to this position were therefore used partially to fund a replacement, although no other planning specialist could be found to accept appointment for more than four months. The replacement planner proved highly effective, since he was able to complete the draft Five-Year Education Plan within this period.  - 22 - The second expert - in project design, appraisal, and execution - was recruited initially for a term of one year, later extended by a further 12 months. As a result of his study, certain changes were introduced in the education system, mainly as regards double-shifting. Although the material he produced constituted an initial approach to evaluation of the education system, it was essentially descriptive in nature, providing no means of relating pertinent variables with evolvement of the system. A more scientific type of study was therefore still required when his appointment came to an end. From the overall perspective, the results of project technical assistance arrangements were not commensurate with their cost. The fact that the services of a planning expert were available for such a short time was certainly an obstacle to introduction of an effective education planning system. UNDP, in its capacity as the agency responsible for financial execution of the technical assistance component, did everything that could be expected of it. The consulting firm commissioned to audit the project accounts performed this task with all due professionalism. IV. PROJECT ECONOMIC AND FINANCIAL EVALUATION 1. IDA Participation in First Education Project The situation of the Credit as of December 2, 1991 was as follows (in US$): EXPENDITURE ALLOCATED DISBURSED BALANCE I Civil engineering works, including salaries of force account personnel 3,020,000.00 3,295,838.73 -275,838.73 2A Building materials, including transport 970,000.00 867,217.23 102,782.77 2B Other equipment and supplies 375,000.00 414,696.00 -39,696.00 III Specialist services 130,000.00 -96,786.95 33,213.05 IV Administrative costs, excluding 95,000.00 75,293.58 19,706.42 salaries V Special Account 510,000.00 119,588.88 390,411.12 TOTAL 5,100,000.00 4,869,421.37 230,578.63  2. Project Financing (in DF) 1985 1986 1987 1988 1989 1990 1991 TOTAL WORLD BANK 88,860,500 366,649,330 158,289,949 87,981,138 226,110,917 193,340,394 1,121,232,228 GOVERNMENT 7,742,210 38,897,516 70,345,692 35,138,512 9,220,346 47,152,001 79,531,337 288,027,614 TOTAL 7,742,210 127,758,016 436,995,022 193,428,461 97,201,484 273,262,918 272,871,731 1,409,259,842 l  - 24 - 3. Proiect Imnlementation (in DF) (A) EXPENDITURE PROJECTED INITIALLY _ ITEM ESTIMATED AT ACTUAL BALANCE APPRAISAL Construction and equipping of 2 urban schools 416,491,940 283,290,931 133,201,009 (Gabode II, Balbala II) Construction and equipping of 4 rural schools 543,533,290 306,986,099 236,547,191 (Bondara, Kontali, Khor-Angar, Ardo) Extension and equipping of 4 existing schools 340,680,490 193,999,992 146,680,498 (Ali-Add6, Dammerjog, Mouloud, Dorra) Equipping of BEPE and operation over 2 years 13,212,180 12,104,863 1,107,317 (incLuding audit) TOTAL AS OF 12/31/88 1,313,917,900 796,381,885 517,536,015 (B) USE OF UNDISBURSED BALANCE Construction and equipping of 3 additional 414,723,432 schools (BalbaLa III, Dikhil, Tadjourah) Extension and equipping of 5 ruraL schools, 132,644,407 repair of schools damaged by bad weather Operation of Projects Office over 3 years and 65,510,118 advance on account of second project TOTAL EXPENDITURE 1,409,259,842  4. Procurement of Goods And Services DESCRIPTION SUPPLIER DATE/PERIOD TOTAL COSTS (DF) MODE OF PROCUREMENT OBSERVATIONS TAXES EXCLUDED I. CONSTRUCTION A) Construction on Contract - Balbala II La Concorde Aug. 86 123,393,650 International Bid - Gabode Eger Verzi Aug. 86 122,558,166 InternationaL Bid - Balbala III La Concorde Jan/Dec 90 106,342,871 International Bid - Dikhil E.T.P.P Jan/Dec 90 126,149,633 International Bid - Tadjourah E.T.P.P Jan/Dec 90 130,541,012 International Bid - Moudo ZECCA Feb/July 91 20,590,585 Local Bid - Guirori ZECCA Feb/July 91 22,246,798 Local Bid - Waddi ZECCA Feb/July 91 22,246,798 Local Bid - Kalaf Abdou. Md. Is. Feb/July 91 12,954,560 Local Bid - SagatLou Abdou. Md. Is. Feb/July 91 17,557,600 Local Bid - Dammerjog M.A. CHILI May 88 6,650,465 Quoted Price (enclosure/fence) Daher Aber Om Jan/June 91 7,000,000 Negotiated Contract 100% State - Chebelley B) Construction on Force Account (District Commissioners) 10,500,605 (a) Purchase of materials: 46,058,998 - Dammeriog 50,182,250 International Bid - Mouloud & Ali-Adde Le Point Dec 86 49,273,349 - Bondara & Kontali 24.039.059 - Ardo & Dorra 180,054,261 - Khor-Angar 898,286,399 CARRIED FORWARD (b) Salaries and other costs of construction personnel: - Dammeriog May 87/March 88 6,891,902 - Mouloud & Ali-Adde Local workmen and April 87/May 88 27,671,714 - Bondara & Kontali jobbers July 87/July 88 24,839,660 Recruited Locally 106,038,785 - Ardo & Dorra June 87/Aug 88 29,859,867 - Khor-Angar Aug 87/July 88 16,775,642 - Misc. travel April/May 91 514,500 (c) Miscellaneous additional Miscellaneous 1986-1991 40,649,269 Consult. with suppliers materials for force- or quote account construction and repairs to project schools  DESCRIPTION SUPPLIER DATE/PERIOD TOTAL COSTS (DF) MODE OF PROCUREMENT OBSERVATIONS TAXES EXCLUDED (d) Transport of building Miscetlaneous 1986-1991 3,208,600 Quote materiats (e) Fuel and maintenance of Mobit Oil 1986-1991 13,743,683 Quote job-site vehicles C) UtiLity services - power EDD 1987-1990 18,303,560 Quote - water ONED 1987-1990 9,618,813 Quote - telephone OPT 1987 152,000 ouote D) Access roads. Drotection - Gabode (access road) Public works June 91 3,171,498 Quote - Moutoud (embankment) 1988 660,000 Quote - Ardo (gates) Garage Intern 1989 193,536 Quote CARRIED FORWARD 1,094,540,643  DESCRIPTION SUPPLIER DATE/PERIOD TOTAL COSTS (DF) MODE OF PROCUREMENT OBSERVATIONS TAXES EXCLLDED E) Driilina operations water G6nie Rural Jure 87-88 82,260,000 Contract (on quote) search II. EQUIPMENT A) Purchase of building Miscellaneous 1987-1988 16,919,500 Consult. with suppliers materiats B) Furniture for first group Eqt Rationnel March 1987 52,444.274 International bid of buildings C) Furniture for second group Scolex Jurw 1989 39,203,566 International bid of buildings D) Furniture for dditional Techno-Pro 1991 4,629,290 Consult. with suppliers classroom E) Nisctllaneos additionl Niscella*neous 1967-1991 6,160,614 Consult. with Suppliers setll Iten_ of equipmnt F) Fuwniture and compaters, piseIlIaneous 1967-1991 14,562.063 Consult. with suppliets etc., for mPE 6) Msellsneou vehicles [ts U rill Sept as 2,517.5S0 Consult. with suppiIer 1001 State Ets Mnill Dec aS 9,920,000 Local bidding 100 state U) Job-site vehicles Fte dr1ss Sept eS 2 40,000 Consult. with supplier 1001 State Its Idrnic Sept aS 2,450,000 Cosult, with supplier 100X State Dec a5 19,560,000 Local biddine 100S State 196 4,990,000 Corwult. with supplier A CMIE FOMIND 1,3S2e,607360 | _  DESCRIPTION SUPPLIER DATE/PERIOD TOTAL COSTS (DF) MODE OF PROCUREMENT OBSERVATIONS TAXES EXCLUDED III. TECHNICAL ASSISTANCE AND FELLOWSHIPS A) Audit of accounts Cab. Robleh 1985-1991 5,200,000 Restricted bidding B) Cost of mission to Acct. BEPE 1988 881,000 Invit. by Bank Washington Acct. BEPE 1989 718,551 Invit. by Bank C) Cost of mission to Washington CIES 1990 15,951,122 Restricted bidding D) Teacher training (advance on second project ) Public Works 1990 1,112,000 Quote E) Soil study, CRIPEN (advance on second Marc Richard 1990 1,282,353 Chosen from group of project) experts F) Computerization of Education Planning Service OD IV. MISCELLANEOUS ADMINISTRATIVE COSTS A) BEPE operations Miscellaneous 1985-1991 14,118,521 Quote and order form B) Salaries and atlowances 1 driver Aug 85 17,388,935 Recruitment of contract 100X State BEPE personnel 1 bookkeeper Dec 90 workers 1 secretary 1,409,259,842 TOTAL I_I  5. Recurrent Proiect Costs (DF) CATEGORY ESTIMATED COSTS ACTUAL ADDITIONAL COSTS NO. MONTHLY UNIT COST ANNUAL TOTAL COST NO. UNIT COST TOTAL COST PERSONNEL - School principal 5 120,000 7,200,000 5 120,000 7,200,000 - Urban teachers 60 116,000 83,520,000 97 116,000 135,024,000 - Rural teachers 11 86,000 11,352,000 15 90,000 16,200,000 - Teachers of Arabic 5 116,000 6,960,000 9 150,000 16,200,000 - Custodial mntnce. staff 12 20,000 2,880,000 31 20,000 7,440,000 - Planning speciatist 1 215,000 2,580,000 1 270,000 3,240,000 - Specialist in project 1 215,000 2,580,000 1 170,000 2,040,000 design/apprsl/exctn. SUBTOTAL 117,072,000 187,344,000 EQUIPMENT - School complexes 5 1,800,000 9,000,000 84 cl. 122,000/class 10,248,000 - Schools (DF 3,000/pupit) 840 2,250,000 43 cl. 122,000/class 5,246,000 MNTNC. & REPAIRS - School complexes 1.5% of invest. 13,538,000 60 rooms 90,000/room 5,400,000 - Schools cost 8.508.000 48 rooms 90,000/room 4,320,000 TOTAL 150,638,000 212,558,000  - 30 - 6. Foreseeable Proiect ImDact on Education System The project was extended because of a significant undisbursed balance on Credit proceeds. Utilization of these funds was programmed and executed in agreement with the World Bank and in accordance with the spirit of this whole undertaking. By the time it was completed, therefore, the first education project had increased primary-school enrollment capacity consider- ably, matching demand growth in the areas where the new facilities were provided. Identification, preparation and start-up of a second education project then grew out of the first project, as marked population growth in such areas as Balbala and Ali-Sabieh meant that existing school facilities that could no longer meet resulting demand under the customary conditions. The benefits of the first project were thus not only preserved but extended further by the existence of the second. 7. Lessons Learned From the economic and financial standpoints, the design of any similar project should take account of the recurrent charges it will give rise to, especially where physical/material operation of the education system is concerned. Care should be taken to incorporate such costs fully into the national budget on completion of the project, since this is a necessary condition for successful continuation of the actions initiated through it. V. PROSPECTS FOR A NEW PROJECT The technical assistance component of the first education project resulted in formulation of a master plan for development of Djibouti's education system focused on some of the needs arising from pressure to bring education supply up to demand level. As a logical further step, prompted by the results of a UNESCO/World Bank appraisal mission, the Government of Djibouti applied for funding for a second education project, thus ensuring continuity at a moment when the needs of an education system hardly out of its infancy were urgent. This second education project, estimated to cost US$7.5 million and currently at an advanced stage of execution, was designed as a vehicle for pressing ahead with actions already instituted under the first project, in particular: creation of additional capacity in the primary-school subsector so that enrollment rates among children in the 6-12 age group can be improved; introduction of a series of measures to increase the internal effectiveness of the education system through research, production of educational materials, strengthening of primary-level supervision structures, and improved education planning and administration.  - 31 - In addition, the second project includes a component focused on strengthening coordination between the education system and labor market needs, thereby broadening the field of action of the first. However, despite an improvement in enrollment rates as a result of the material achievements of the first and second projects and the rational utilization of existing infrastructure (double shifting, a longer school day), unmet demand for education is still heavy at both primary and secondary levels. Although this can be attributed in part to a sharply increased population growth rate, there is still insufficient primary-level capacity and secondary-level facilities are unable to accommodate a great many children who complete primary school. Primary level: At the commencement of the 1991-92 school year, the number of enrollments at this level rose to 30,157. However, the admission rate, a barometer of the literacy of the population, was 46%, placing the Republic of Djibouti among those countries with such low schooling rates that they need to double them before the end of the century (according to the Ministry of Education and UNESCO). In the circumstances, the next Education Plan (1994-99) should aim for an admission rate of 60Z. SecondarX level: Capacity at this level is very limited in relation to the marked growth in primary-level enrollments. One of the goals of the next Education Plan should be to increase the primary-to-secondary progression rate, which is now 24%. Secondary-school enrollments rose to 7,319 at the beginning of the 1991-92 school year, including 5,688 in the junior secondary cycle. It is clear, given the rate of development in primary education, that accommodation capacity at the secondary level, particularly in the junior cycle, must be expanded. Private education: In view of the demand for education today and budgetary limitations that are increasingly severe, the Government must act over the next few years to promote offers of education from private enterprises. Reorganization of the private education subsector, with all its present anomalies (questions of legal status, language, etc.), should be one of the objectives of the next Five-Year Plan. At the same time, that Plan should emphasize the need to improve the capacity of the Djibouti authorities to manage the country's education system. A policy designed to foster ongoing training should lead to the creation of a pool of Djibouti professionals qualified to manage, administer and evaluate all aspects of the education system. From this perspective, completion of the first education project provides the foundation for identification of future system needs that must be met if actions already instituted in the primary-education subsector are to be continued successfully and carried over into the entire secondary subsector. A third education project would contribute largely to ongoing improvement of the country's education system, by raising the basic-education enrollment rate and increasing both internal and external system productivity.   - 33 - PART III STATISTICAL INFORMATION A. Related IDA Credit 34 B. Project Timetable 34 C. Credit Disbursements 35 D. Project Implementation 36 E. Project Costs and Financing 37 F. Project Results 38 G. Status of Covenant 40 H. Use of IDA Staff Resources 42 ANNEX I: Comparative Schedule of Project Implementation 43 ANNEX II: Comparative Table of Estimated and Actual Project Costs 44 ANNEX III: Comparative Table of Project School Indicators 45  - 34 - A. Related IDA Credit Table 1: IDA Credit to the Education Sector Credit Title Purposes Status and Observations Manpower and 1. To improve vocational (a) Components 2 and 4 Education coordination with labor- take up and amplify First Development Project market structure. Education Project (Credit (Credit 2080-DJI) 1543) objectives. 2. To increase access to primary education. (b) Component 3 now under way, progressing 3. To improve relevance satisfactorily. Three and quality of primary fellowships financed from education and junior proceeds of Credit 1543. secondary education. (c) Preparation of 4. To improve admin. bidding documents and capabilities of Ministry contracts for Component 2 of Education. are nearing completion. (d) Component 4. with UNDP parallel financing. being implemented by UNESCO. Experts now _ being recruited. B. PROJECT TIMETABLE Table 2: Planned and Actual Dates of Project Timetable Item Planned Date Actual Date Sector Memorandum August 1983 August 31, 1983 Identification November 1984 Oct 30-Nov 6, 1983 Preparation (UNESCO) Jan 15-Feb 25, 1984 Government Application February 26, 1984 Appraisal July 1983 Feb 29-March 22, 1984 Credit Negotiations October 1984 Nov 12-16, 1984 Board Approval December 1984 January 22, 1985 Credit Signature February 1985 March 21, 1985 Credit Effectiveness April 1985 July 16. 1985 Credit Closing June 30, 1989 June 30, 1991 Final Disbursement December 30, 1989 December 2, 1991  - 35 - C. CREDIT DISBURSEMENTS Table 3: Cumulative Estimated and Actual Disbursements (USS 000, SDRs 000) IDA Fiscal Year 1985 1986 1987 1988 1989 1990 1991 1992 Appraisal Estimate (USS 000) 900 2,100 3,700 4,700 5,000 Appraisal Estimate (SDR 000) 918 2,142 3,774 4,794 5,100 Actual (SDR 000) 489.8 1,456.4 2,475 3,024.2 3,271.2 4,661.4 4,977.3 Actual as X of Estimate 23.3 39.4 52.7 60.5 65.4 93.2 99.5 Tgble 4: Grach of Planned and Actual Disbursent Schedulea * Appraisal Estimte 0 Actusl Schedble 6000 - 5000 SDR 000 4000 - 3000 - / - _ 2000 - 1000 0~ 0 1985 1986 1987 1988 1989 1990 1991 1992 IDA Fiscal Years  - 36 - D. PROJECT IMPLEMENTATION Table 5: Planned and Actual School Delivery Dates School DeLivery Works Comoletion Date Plamned Actuau Urban Schools Djibouti/Gabode March 31, 1987 September 1987 Djibouti/BaLbala East March 31, 1987 June 1987 Djibouti/Balbala West - January 1991 RuraL SchooLs Danumerjog, Ali Adde, Bondara, Ardo June 30, 1986 Septenbher 1968 KontaLi, HouLoud, Dorra, Khor-Angar June 30, 1987 September 1988 Dikhit - February 19