Morocco - Rural Basic Education Development Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 19364 IMPLEMENTATION COMPLETION REPORT KINGDOM OF MOROCCO RURAL BASIC EDUCATION DEVELOPMENT PROJECT (Loan No. 3295-MOR) May 28, 1999 Human Development Group Middle East & North 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 EQUIVALENTS Currency Unit = Moroccan Dirham (MAD) as of February 26, 1991 (approval date) MAD 1.00 = US$ 0.13 US$1.00 = MAD 7.96 as of June 30, 1998 (closure date) MAD 1.00 = US$ 0.10 US$1.00 = MAD 9.75 SCHOOL SYSTEM Grades 1 through 6 Lower basic education (primary school) Grades 7 through 9 Upper basic education (college) Grades 10 through 12 Secondary education ("Iycee") ABBREVIATIONS AND ACRONYMS AfDB African Development Bank CGED Comptroller-General of Expenditure Commitments IBRD International Bank for Reconstruction and Development ICV "Study Assessing the Inipact of Certain Variables on Access to School and Schooling in Rural Areas" IREDU Education Economics Research Institute (University of Burgundy) MOE Ministry of Education MOF Ministry of Economy and Finance NAP National Assessment Program FISCAL YEAR July 1 - June 30 Vice President: Kemal Dervi,s Country Director: Christian Delvoie Sector Director: Jacques Baudouy Responsible Staff: Jeffrey Waite  FOR OFFICIAL USE ONLY IMPLEMENTATION COMPLETION REPORT KINGDOM OF MOROCCO RURAL BASIC EDUCATION DEVELOPMENT PROJECT (Loan No. 3295-MOR) TABLE OF CONTENTS Preface ..............i Evaluation Summary .............. iii Part One Projet Implementation Assessment ......................................1 A. Project Objectives .......................................1l B. Achievement of Project Objectives .....................................2 C. Implementation Record and Major Factors Affecting the Project ........................................6 D. Project Sustainability ........................................9 E. Bank Performance ........................................9 F. Borrower Performance ....................................... 10 G. Assessment of Outcome ....................................... 10 H. Future Operations ....................................... 11 l. Key Lessons Learned ....................................... 12 Part Two Statistical Tables ....................................... 15 1. Summary of Assessments ....................................... 16 2. Related Bank Loans/Credits ....................................... 18 3. Project Timetable ....................................... 22 4. Loan Disbursement: Cumulative Estimated and Actual .... 22 5. Key Indicators for Project Implementation ........................ 23 6. Key Indicators for Project Operation ................................. 24 7. Studies included in the Project ....................................... 26 8A. Project Costs ....................................... 27 8B. Project Financing ....................................... 27 9. Economic Costs and Benefits ....................................... 28 10. Status of Loan Covenants ....................................... 28 11. Compliance with Operations Manual Statements .............. 31 12. Bank Resources: Staff Inputs ....................................... 31 13. Bank Resources: Missions ....................................... 32 Annexes A. ICR Mission Aide-Memoire ....................................... 35 B. Borrower's Contribution ....................................... 41 C. Map ........................................ 55 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.  IMPLEMENTATION COMPLETION REPORT KINGDOM OF MOROCCO RURAL BASIC EDUCATION DEVELOPMENT PROJECT (Loan No. 3295-MOR) PREFACE This is the Implementation Completion Report (ICR) for the Rural Basic Education Development Project in the Kingdom of Morocco, for which Loan 3295-MOR in the amount of US$ 145 million equivalent was approved on February 26, 1991, and made effective on March 19, 1992. Following four cancellations on March 4, 1996, August 19, 1997, March 9, 1998 and September 23, 1998, the project loan was reduced to US$ 55 million equivalent. The loan was closed on June 30, 1998, following an extension (agreed to on June 23, 1997) from its original closing date of June 30, 1997. Final disbursement took place on, at which time a balance of US$ 472,460.47 was canceled. The ICR was prepared by Jeffrey Waite (HDNED) and reviewed by Jacques Baudouy (Director MNSHD) and Marisa Femandez-Palacios (Project Adviser). The borrower provided an official note that is included as appendixes to the ICR. Preparation of this ICR was begun during the Banks final supervision/completion mission, from July 20 to August 1, 1998. It is based on material in the project file. The borrower contributed to preparation of the ICR by preparing its own evaluation of the project's execution and commenting on the draft ICR. i  VI  IMPLEMENTATION COMPLETION REPORT KINGDOM OF MOROCCO RURAL BASIC EDUCATION DEVELOPMENT PROJECT (Loan No. 3295-MOR) EVALUATION SUMMARY Introduction In 1985, the Government launched a thoroughgoing reform of the educational system with the goal of expanding access to quality education while simultaneously containing costs. The Government's main objective was to improve basic education enrollment figures in rural areas, particularly among girls. The reforn had already received Bank support in the form of two loans. This project was designed as a follow-up to those earlier loans insofar as it sought, first, to increase the capacity of upper basic education in order to accommodate a growing number of students who complete the sixth grade as a result of expanded access to lower basic education, and, second, to motivate parents to enroll their children in primary school with the knowledge that children will henceforth have the opportunity to continue their studies through the ninth grade. Project Objectives Thus, the goal of this project was to support the Government's educational objectives through investments designed: a. To promote greater equality of educational opportunity and increased school attendance by improving college enrollment figures among disadvantaged populations such as rural children and girls; and b. To improve the educational environment through in-service teacher training, general inspections, curriculum development, assessment of student achievement and proper use of students' materials. The first objective was translated into quantifiable terms: (i) to increase rural enrollment from 51,600 (in 1989/1990) to 115,200 (at project completion, initially scheduled for 1995/1996); and (ii) to increase the rate of continuation from lower basic to upper basic education from 62.0% to 77.2% over the same period. Although it incorporated major components, such as in-service training and development of a National Assessment Program (NAP), the second objective was never quantified. Implementation Experience and Results Construction of colleges: Of the 208 colleges that were to be built in connection with the project, only 78 had been put into service with IBRD financing as of June 30, 1998. iii  Enrollment figures: The project was able to provide only a very modest contribution to the main goal of the 1985 reform package, i.e. expanded access to nine years of basic education. In fact, net upper basic enrollment rates have risen only very slightly nationwide since 1994/1995, when the first IBRD-financed colleges were put into service. Even though the proportion of young people enrolled in college progressed only slightly at the national level, the project nevertheless nearly reached its objective of increasing rural upper basic enrollment figures by 54,500 students during the three-year period initially projected for putting new colleges into service. During the four years of actual implementation, the project witnessed increased enrollment of nearly 53,000. Enrollment figures for girls: Enrollment at the 78 colleges financed through the IBRD loan and put into service by 1997/1998 came to approximately 26,000 students during this school year, including 24.0% girls. The objective of increasing the proportion of girls was not quantified at the time of the initial project appraisal. It is nonetheless justifiable to express disappointment over the fact that rural upper basic enrollment figures for girls have progressed so slowly, from 23% in 1994/1995 to 28% in 1998/1999. Quality of instruction: In-service training was provided to teachers in conjunction with overall MOE activities, but was not specifically adapted to rural education. Insofar as teachers are often young and inexperienced and have just recently graduated from teacher training institutes, this lack of appropriate training may diminish the quality of instruction in rural areas. With respect to technical assistance, the project was supposed to include two major studies. One, assessing the irmpact of certain variables on school enrollments and attendance in rural areas, which was submitted to MOE in 1997, has only recently been endorsed by the monitoring committee. The other, designed to help institute a National Assessment Program (NAP), has not yet been carried out and will be transferred to the Social Priorities Project (4024-MOR.) Assessment of Outcome and Key Lessons Learned On the Bank's rating system, the project was rated highly unsatisfactory. However, a social assessment, carried out independently of MOE by a team of Moroccan researchers at the same time as the implementation completion mission, demonstrated that the project should be seen in a better light, if only because the colleges are likely to improve over time, as they reach their "cruising speed" and the communities they serve adapt to the new context. This trend can be seen quite clearly in the occupancy rates, which continue to grow in the colleges that have been the longest in operation.. There are four lessons to be drawn from the project: * Support from senior government officials throughout the project is critical to its success; * The choice oiF a site - an issue that was neglected during project preparation - is essential to the success of a college in a rural setting; iv  * Community contributions must not be ignored, even in countries where traditions of local participation are not strong; * Putting additional resources into the building program cannot compensate for the lack of activities designed to improve the quality of instruction. v  IA  IMPLEMENTATION COMPLETION REPORT KINGDOM OF MOROCCO RURAL UPPER BASIC EDUCATION DEVELOPMENT PROJECT (Loan No. 3295-MOR) PART I PROJECT IMPLEMENTATION ASSESSMENT A. Project Objectives 1. The goal of this project was to support the Government's educational objectives through investments designed: a. To promote greater equality of educational opportunity and increased school attendance by improving college enrollment figures among disadvantaged populations such as rural children and girls; and b. To improve the educational environment through in-service teacher training, general inspections, curriculum development, assessment of student achievement and proper use of students' materials. 2. In 1985, the Government launched a thoroughgoing reform of the educational system with the goal of expanding access to quality education while simultaneously containing costs. To generate savings on wages paid to teaching and non-teaching staff,' the primary level (five years) and secondary level (seven years) were replaced by a system of basic education (nine years) and secondary education (three years). Basic education was then divided into two levels: six years of lower basic education, followed by three years of upper basic education. The Government's main objective was to improve basic education enrollment figures in rural areas, particularly among girls. 3. The reform had already received Bank support in the form of two loans (2664-MOR and 3026-MOR). The purpose of the first loan was to provide broad support for implementing the reform, while the second focused on expanding lower basic education enrollment figures in rural areas and improving the quality of instruction nationwide. Loan No. 3295-MOR was designed as a follow-up to the latter, in the sense that it addressed the following concerns in particular: a. First, to increase the capacity of upper basic education in order to accommodate the growing number of students completing the sixth grade as a result of expanded access to lower basic education; 'Primary/basic education teachers' salaries are lower than those paid to secondary education teachers. 1  b. Second, to motivate parents to enroll their children in primary school with the knowledge that children will henceforth have the opportunity to continue their studies through the ninth grade. 4. To achieve these objectives, the Government opted for the concept of a "rural college" that would operate like urban colleges - covering upper basic education (i.e. the last three grades of the nine-year system) - but in a predominantly rural setting. By reducing the distance between home and school, the Government expected to increase demand for education. 5. It is importaLnt to note that the "rural college" concept was neither tested on a pilot basis, nor compared with other options through a rigorous evaluation process. Indeed, the rural college optioll was explicitly chosen over building urban colleges with boarding facilities. This choice appears to have been justified by the Government's more general policy of slowing urban drift by creating small townships around a core of basic infrastructure: roads, water, electricity, health services, mosques and... schools. The Government also rejected the option of adding Grades 7, 8 and 9 to rural primary schools to create an "integrated school" covering all nine years of basic education -- on the grounds that it was too difficult to reconcile the major differences that separate the two levels of basic education. Upper basic education is in fact closer to secondary education with respect to teacher management and organizational structure. 6. With a loan of US$ 145 million, the project was large and probably fell just within the financial and operational capacity of the MOE. The risk was increased however by the fact that the Government had not indicated clear support for this level of the education system, that it was facing a number of challenging issues (most notably the need to make lower basic education universal), and that its budget was only increasing slightly. B. Achievement of Project Objectives 7. Construction of colleges. The initial program is being implemented much more slowly than originally anticipated. Of the 208 colleges that were to be built as part of the project, only 78 had been put into service with IBRD financing as of June 30, 1998. Although World Bank loan proceeds are no longer available in the case, MOE expects to put approximately fifty of the other colleges originally called for in the project into service at the beginning of the 1998/1999 and 1999/2000 school years. (Of the 42 colleges to be financed through the parallel AfDB loan, 35 were in operation on June 30, 1998. It is anticipated that all 42 will come on stream before this loan, likely to be extended through December 31, 1999., is closed.) 2  Number of Colleges in Service 250 - F 200 - AfrDB-pro -am med Forecast , IBRD-programmed 150 - *-ProjectTarget 100 -' 8. Enrollment figures. The project was able to provide only a very modest contribution to the main goal of the 1985 reform package, i.e. expanded access to nine years of basic education. In fact, net upper basic enrollment rates have risen only very slightly nationwide since 1994/1995, when the first IBRD-financed colleges were put into service. For the seventh grade, the net rate rose from 19% in 1994/1995 to 20% in 1997/1998 for boys, and from 16% to 19% for girls, while at the ninth grade level, over the same period, the rate remained at 13% for boys and climbed from 12% to 13% for girls. Net Enrollment Rates 25%- 20%- -4 Grade7 BOys -*-Grade 7 Girls 15% --i _ ' - Grade 9 Boys p /- . ->+ - : ! s - 2 - zA- Grade 9 Girls 10%- A-'F e -A- - d' 5% 9. Even though the proportion of young people enrolled in college progressed only slightly at the national level, the project nevertheless nearly reached its objective (albeit two years late) of increasing rural upper basic enrollment figures by 54,500 students during the three-year period initially projected for putting new colleges into service (1993/1994 to 1995/1996). During the four years of actual implementation, i.e. from 1994/1995 to 1997/1998, the project (IBRD and AfDB components combined) witnessed increased enrollment of 52,600. 3  Rural Upper Basic Education Enrollments 150000 =-Actual Enrollments . f 100000 - - Project Target 500001 C 4 LL 10. Girls' participation. The objective of increasing the proportion of girls was not quantified at the time of the initial project appraisal. It is nonetheless justifiable to express disappointment over the fact that rural upper basic enrollment figures for girls have progressed so slowly, from 23% in 1994/1995 to 28% in 1998/19992 -- especially as this increase appears to be the straightforward consequence of girls' increasing their share of enrollments in lower basic education, rather than as the result of policies specifically targeted at girls (see para 25). Girls' Rural Enrollment Share in Grades 6 and 7 35%- 30% - - -Grade 61 25%- =_<____ , ,f/ |~~-*Grade7| 25%- 20% 11. Promotion from lower to upper basic education. The initial appraisal envisioned a gradual increase in the rate of promotion from Grade 6 to Grade 7, rising from 62.7% in 1989/1990 to 77.2% in 1995/1996. In fact, the real rate has been falling over the last few years, from 84.2% in 1994/95 to 77.9% in 1998/99, probably because overall demand has been weaker than originally anticipated. 12. Occupancy rate. The staff appraisal report envisioned colleges of "minimal" size, i.e. schools comprising nine classrooms to be occupied by twelve sections (approximately 360 students). In 1997/1998, average enrollment in the 78 IBRD-financed colleges was 333 students. Although the number of students varies considerably from college to college, the breakdown of school enrollment figures, based on the year of being put into service, suggests a gradual increase in enrollment, ultimately approaching the standard college enrollment capacity. 2 Between 1989190 and 1998,'1999, urban upper basic enrollment figures for girls gradually increased from 42% of students to 45%. 4  Average Enrollment by Years of Service of 78 IBRD Colleges (1997198) 450 400- 350 - B Boys [ Girls| 300 - II r 250 -X 200 G 150 100 50 0 0-1 1-2 2-3 3-4 No. Years in Service 13. Pupil-teacher ratio. The student-teacher ratio was expected to rise and then fall over the life of the project, down to 15.1 students per teacher in the final year of the project. In fact, the real ratio was higher than projected (18.6 in 1995/1996), but fell to 18.1 in 1997/1998. This resulted in a lower than expected unit cost for upper basic education. Pupil-Teacher Ratio in Rural Upper Basic Education 20 18 a14- ' _ I - Actual 12- |L oEstimated 100- 14. Student achievement. It is impossible to know whether the project has had any impact on student achievement. An assessment of the learning achievement of the student population affected by the project could have been carried out as part of the National Assessment Program (NAP), but this part of the project was never implemented. Furthermore, the lack of standardized national exarns at the end of the upper basic level makes it impossible to compare the results of end-of-year testing for different groups of students. 15. Boarding and canteen scholarships: The number of boarding scholarships awarded to students from rural areas has not met expectations. Instead of the 9,500 boarding scholarships expected for 1995/1996, the figure at project completion in 1997/1998 was 6,090. Only 17.5% of these scholarships were awarded to girls, far below the figure of 40% projected at the time of the initial appraisal. Of the 634 canteen scholarships awarded in rural areas in 1997/1998, 120 (or 18.9%) went to girls. It should be noted that the Government has preferred to finance the scholarships through the national budget instead of the loan. 5  C. Impilementation Record and Key Factors Influencing the Project C]. Implementation Record 16. The project was overly large. When the project was first designed, both the unit cost of the colleges and the MOE's implementation capacity were overestimated. This overestimation, when combined with the MOE's changing priorities and the weakening of the Moroccan currenacy against the US dollar, led to four successive reductions in the loan amount that lowerecl it by almost two-thirds (from US$145 million to US$55 million). 17. The project started two years behind schedule because of protracted dealings between MOE and the Ministry of Finance and Economy (MOF) over the issue of the unit cost of construction. MOF insisted that contracts be negotiated at a lower cost range than MOE estimates had initially projected. This delay was partially offset by a one-year extension of the projlect (from June 30, 1997 to June 30, 1998). 18. School building. Of the 208 colleges projected for World Bank financing, 78 were in operation when the loan was closed; 57 were under construction (including six contracts to be terninated) and 18 were in the approval, start-up or study phase. The rest were dropped after the series of reductions in the amount of the World Bank loan. Implementation Progress of 208 IBRD-Programmed Colleges 55 cancelled,y-_ (in78 in service 18 in pre- construction phase 57 under construction 19. Sanitation, water, electicity. With a view to reducing the cost, sanitation and water and power hook-ups were not a condition of the choice of site, and were not included in the building specifications. However, toward the end of the project, the MOE made a particular effort to make up for this oversight: all 78 colleges already in operation have septic tanks and cesspits; almost all colleges have water supply -- some are connected to community mains but most have a water tower and a well, or just a well; most colleges that have been in operation for three or four years have electricity (they are either connected to the community grid or, in a few cases, possess a generator). Classroom and boarding fumiture and fittings have been delivered to the colleges already in operation. Textbooks have not yet been delivered. 6  IBRD Colleges Connected to Water Supply IBRD Colleges Connected to Electricity Supply (June 1998) (June 1998) 40 40 30 - t [ Unconnected |30 Si In Uncogresse D in progress 25- o2D - UnConneted | Completed o 15 _ 15 _-1 15 - | 5 - PE o '' 1994/95 1995r96 1996197 1997/98 1994/95 1995/96 1996/97 1997/98 Year of College Opening Year of College Opening 20. Teacher training. The "quality of services" component was less successful than the "physical achievements" component. In-service training was provided to teachers in conjunction with overall MOE activities, but was not specifically adapted to rural education. Insofar as teachers are often young and inexperienced and have just recently graduated from teacher training institutes, this lack of appropriate training may diminish the quality of instruction in rural areas. 21. Service vehicles. Although MOE purchased 46 service vehicles in March 1996 so that inspectors could more easily cover rural areas, the level of educational supervision is clearly inadequate to offset the additional difficulties faced by teachers in such areas (which include lack of training for inspectors, overly heavy workload given the distances that must be covered, and the intermittent availability of vehicles often used by the MOE's provincial office for other tasks). 22. Technical assistance. The Government did not take ownership of the technical assistance activities, which were either delayed or were not undertaken at all. The project was supposed to include two major studies. The study assessing the impact of certain variables on school enrollments and attendance in rural areas was submitted to MOE in 1997 and still awaits the endorsement of the monitoring committee. As for the other study (designed to help institute a National Assessment Program (NAP)), a call for tenders was issued on two occasions, with no successful outcome. This underscores not only the institutional weakness of MOE in procurement, but also the low priority accorded to this program by the MOF. (In contrast, the MOE appears eager to complete this study and has asked the Bank to finance it from another loan currently being implemented.) C2. Key Factors Influencing the Project 23. Several factors apparently led to weak performance in this project. The main ones are covered by just three themes: (i) weak Government ownership, especially on the part of the MOF; (ii) inadequate project preparation; and (iii) overestimation of the MOE's financial and operational capacity. 24. Weak Government ownership. The MOF showed little enthusiasm for the project from the start, and, once the loan was made, tried to limit its scope. MOF reservations seem to have become even more pronounced as a result of several years of difficult economic circumstances (partly caused by droughts), which obliged the Government to make drastic 7  cuts in investments. Following a reduction in the unit cost of rural colleges (and the resulting reduction in the level of facilities provided), as well as the delays imposed by the MOF, the MOE lost interest in a project that was no longer accorded budgetary resources. 25. In addition, whilst both the Bank and the MOE attached great importance to girls' education, the two diverged significantly as to what measures needed to be put into place. The State did not inmplement policies aimed specifically at compensating for the fact that many parents consider their sons' education more important than their daughters'. In addition to the opportunity costs of schooling their daughters (in terrms of loss of household labor for tasks such as childcare, housework, agricultural chores, fetching firewood and water, and carpet weaving), parents think of college as being ill-equipped to prepare their daughters for their future as housewives and mothers, and as a place where their daughters risk running wild. According to the social assessment undertaken for this report, there is nevertheless a demand for girls' education which could have been expanded if only more appropriate incentives had been put in place. 26. The assessment identified the following incentives as being the most important: a. an unbroken school day (that enables students to return home early); b. the presence of adults (especially parents) to accompany students home; c. a convenient transport system; d. the existence of appropriate facilities: latrines, boarding facilities, canteens, water and electricity; e. the siting of the college and boarding facilities in a small, well-equipped township (e.g. in the county's main town); f. the separation of sexes, with respect to latrines and sleeping quarters. 27. Inadequate preparation: Without taking away from the project design, it must be pointed out that no serious study was undertaken to ensure that the strategy finally adopted was the most cost-effective. In addition, more solid and painstaking project preparation, particularly as regards plans and manufacturing processes, could have helped the project start at an earlier date. As indicated above, the project started two years behind schedule because of lengthy dealings between MOE and MOF over the issue of the unit cost of construction. 28. Furthermore, the factors that were necessary to select sites that would guarantee the success of rural colleges were not identified beforehand. This meant that the availability of land (offered free of charge to MOE by local authorities, for example) sometimes weighed too heavily on the choice of college sites, and therefore, resulted in sites that had few facilities (e.g. water, power). Few colleges were built in rural townships, where they would have benefited from better facilities and transport without noticeably reducing most students' ease of access. 29. Overestimation of MOE's financial and operational capacity. It is unlikely that the MEN could have implemented such a large project without major adjustments to its procedures and without intensjive management training and other groundwork. The MOE is highly centralized and, as a result, has only a limited capacity in procurement and in 8  monitoring worksites throughout the country. This capacity was further weakened early on in the project with the elimination of the Funded Projects Division, initially established to manage this type of project, and a dispersal of its functions among the various MOE directorates. The resulting break-up of responsibility made project coordination and management more difficult. 30. The MOE had difficulty allocating a sufficient portion of its budget to the project for new investments to be made at the rate envisioned at the time of the initial appraisal. This difficulty is due to the fact that, however large the loan, the MOE budget remains at a fixed level, usually insufficient to meet all its priorities. The Moroccan Government's practice leaves little incentive to the MOE to use the loan, and led to successive cancellations of two-thirds of the loan amount. 31. Finally, the MOE has relatively little experience with studies, technical assistance and training. It is even weaker in this area than in school construction. The administrative procedures related to qualitative activities appear to have been even more difficult to follow efficiently than those related to physical outputs. D. Project Sustainability 32. Provided that the Government continues to provide necessary staffing, keeps the existing colleges in operation and sets up a reliable system of building and equipment maintenance, then it is highly probable that the project will maintain at least the level of activity observed at implementation completion. Evidence obtained in the course of the implementation completion mission suggests that the MOE intends in fact to extend the program to other rural communities - in a form that incorporates lessons from this project. E. Bank Performance 33. Identification, preparation and appraisal. A better contribution by the Bank with respect to project preparation, particularly in assessing MOE's capacity to manage a loan of US$145 million, would undoubtedly have produced a more modest operation with a greater likelihood of achieving its objectives. 34. Supervision. The Banks performance was inadequate. High turnover of project managers throughout the life of the project (one project manager during the preparation phase, six others during the supervision phase) certainly did not facilitate, on the part of the Bank, proper implementation. Insufficient familiarity with the project and its key players and a lack of accountability for project outcomes -- sentiments often attributable to a lack of continuity in project management -- also contributed to lax supervision and to the absence of restructuring in the face of persistent problems. 9  Supervision: Six Task Managers * First Manager o Second Manager * Third Manager o Fourth Manager A Fifth Manager o Sixth Manager E Ef 0[1*> @ < O *> AA Om 1991 1992 1993 1994 1995 1996 1997 1998 35. Although the Bank observed a decline in project performance as expressed in the implemtentation record3 -- indeed, a decline much more pronounced than is usually found in the region, as the following graph indicates -- it is not clear that, until nearly the end of the project, the Bank made any efforts to convince the Government to take project implementation more seriously. However, it must be said that impact assessment is made difficult by the fact that a college takes three years - from the time it opens - to reach its "cruising speed". Implementation Progress Ratings ,:;Trend1ine for MNA Educaoti- Projds S: 3 3295-MOR o- 4 ,. ,s0O 300