Burundi - The First, Second, and Third Education Projects

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 10982 PERFORMANCE AUDIT REPORT BURUNDI FIRST EDUCATION PROJECT (CREDIT 679-BU) SECOND EDUCATION PROJECT (CREDIT 976-BU) THIRD EDUCATION PROJECT (CREDIT 1358-BU) JULY 28, 1992 MICROFICHE COPY TypoPAR Co"' F Report 10982 (CRC069 1, & 3 EDUCATION PR05/ OEDD1 ALEGRE, 1 / 1 727 By: TS Date Ent. - Operations Evaluation Department 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. Country Exchange Rates Name of Currency = Franc Burundian (F Bu) Year of Appraisal (Cr. 679) - 1976 US$1.00 = F Bu 90.0 1977 F Bu 90.0 1978 F Bu 90.0 Year of Appraisal (Cr. 976) - 1979 = F Bu 90.0 1980 F Bu 90.0 1981 F Bu 90.0 Year of Completion (Cr. 679)- 1982 = F Bu 90.0 Year of Appraisal (Cr. 1358)- 1983 = F Bu 90.0 1984 F Bu 116.9 1985 F Bu 121.4 1986 F Bu 107.7 Year of Completion (Cr. 976)- 1987 = F Bu 124.7 Year of Completion (Cr. 1358)- 1988 = F Bu 136.7 Intervening Years Averages: Cr. 679 US$1 = F Bu 90.0 Cr. 976 US$1 = F Bu 109.7 Cr. 1358 US$1 = F Bu 121.9 GLOSSARY AfDF - African Development Fund BCR - Borrower Completion Report BEET - Bureau d'Etudes de I'Enseignment Technique (Techical Education Bureau) BEPES - Bureau d'Etudes des Programmes de I'Enseignement Secondaire (Secondary Education Bureau) BER - Bureau d'Education Rurale (Primary or Rural Education Bureau) BPE - Bureau des Projets d'Education (Project Imple-nentation Unit) CFMP - Centre de Formation des Maitres Pratiques (Practical Subject Teacher Training Centre) CFP - Centre de Formation Polyvalente (Multi-purpose Learning Cefitre) (Primary School equipped for practical activities and non-formal education and training for youth and adults) CO - Cycle d'Orientation (Lower Secondary Cycle School) CSE - Centre Socio-Educatif !Social and Educational Center) (also known originally as Foyer Social) EFI - Ecole de Formation d'Instituteur (Primary Teacher Training School) ESTA - Ecole Secondaire des Techniques Administratives (Commercial Secondary School) ETMM - Ecole Technique Moyenne pour le Main d'Oeuvres (Middle Level Vocational School) ETP - Ecole des Travaux Publics (Public Works School) ETS - Ecole Technique Secondaire Jechnical Secondary School) FBu - Franc Burundian (Burundian Franc) FYDP - Five Year Development Plan IDA - International Development Association IIEP - International Institute for Educational Planning LP - Lyc6e Pddagogi4ue (Upper and Lower Cycle Secondary School with Primary Teacher Training Program) MNE - Minist6re de I'Education Nationale (Ministry of National Education) MSA - Ministry of Social Affairs PCR - Project Completion Report PAR - Performance Audit Report RPP - Regie des Productions Pddagogiques (Textbook Production Unit) SAR - Staff Appraisal Report TA - Technical Assistance UNDP - United Nations Development Program UNESCO - United National Educational Scientific and Cultural Organization UNICEF - United Nations International Children's Emergency Fund FOR OIICIAL IB ONLY THE WORLD BANK Washington, D.C. 20433 U.S.A. Office of Director-General Operations Evaluation July 28, 1992 HENORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Performance Audit Report on Burundi First Education Project (Credit 679-BU) Second Education Project (Credit 976-8U) Third Education Proiect (Credit 1358-BU) Attached, for information, is a copy of a report entitled "Performance Audit Report on Burundi - First Education Project (Credit 679-BU); Second Education Project (Credit 976-BU); and Third Education Project (Credit 1358-BU)" prepared by the Operation. Evaluation Department. Attachment This document has a restricted distribution and may be used by recipients only in the performance of tV Ar official duties. Its contents may not otherwise be disclosed without World Bank authodsation. FOR OFFICIAL USE ONLY PERFORMANCE AUDIT REPORT BURUNDI FIRST EDUCATION PROJECT (CREDIT 679-BU) SECOND EDUCATION PROJECT (CREDIT 976-BU) THIRD EDUCATION PROJECT (CREDIT 135&U) TABLE OF CONTENTS PREFACE ................................................... ........ BASIC DATA SHEET .................................................. EVALUATION SUMMARY ..............................................xv L PROJECT BACKGROUND ................................................ 1 Introduction .......................................................... 1 The Development Strategy ............................................... 1 Project Objectives and Scope ................ ............................ 6 Implementation Plan .................................................. 8 Project Cost and Financing Plans ....................................... 8 II. IMPLEMENTATION EXPERIENCE ...................................... 9 Project Management ................................................... 9 Physical Implementation.................................................9 Changes in Project Scope During Implementation ........................... 11 Use of Force Account .............................................. 12 Costs and Disbursements ............................................. 16 III. PROJECT OUTCOMES .............................................. 21 Expansion ....................................................... 21 Improvement in Quality ............................................. 25 Human Resource and Institutional Development in the Sector .................. 28 IV. FINDINGS AND ISSUES ................................................ 36 IDA Performance..................................................... 37 Borrower Performance ................................................. 39 Sustainability of Project Benefits ....................................... 40 Conclusions and Lessons ............................................... 42 Appendix ..... ........................................................ 45 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. PERFORMANCE AUDIT REPORT BURUNDI FIRST EDUCATION PROJECT (CREDIT 679-BU) SECOND EDUCATION PROJECT (CREDIT 976-BU) THIRD EDUCATION PROJECT (CREDIT 1358-BU) PREFACE This is a performance audit of the first three education projects in Burundi. For the First Education Project, Credit 679-BU in the amount of US$10.0 million equivalent was approved on February 8, 1977 and signed on April 28, 1977; the proceeds were fully disbursed as of April 13, 1982 and the Account was closed on March 31, 1983. For the Second Education Project, Credit 976-BU in the amount of US$15.0 million equivalent was approved on January 29, 1980 and signed on April 23, 1980; the proceeds were fully disbursed on April 14, 1986 and the Account was closed on June 30, 1986. For the Third Education Project, Credit 1358-BU in the amount of US$15.8 million equivalent (SDR14.5 million) was approved on May 10, 1983 and signed on June 9, 1983. The Account was closed on June 30 1988 and the last disbursement was made on September 13, 1988 when a balance of US$95 (or SDR 73.56) remaining in the Account was cancelled. The Performance Audit Report (PAR) is based on a review of the Project Completion Report (PCR),I/ Staff Appraisal Report (SAR);2/ President's Report,/ Development Credit Agreement (DCA), the record of Board discussions and IDA correspondence files, documents and reports relating to each project Discussions were also held with IDA staff associated with the projects. An audit visit to Burundi was undertaken in October 1991 and discussions were held with Government officials (who also made -vailable various reports of relevance to the projects) and visits were made to a sample of project schools. UNESCO officials in Paris also kindly made the files on the technical assistance components of the first and third projects available to the audit mission. The assistance of all these persons and of the Bank Group's Resident Mission in Bujumbura is gratefully acknowledged. The PCRs prepared for the three projects were generally useful. To the extent that errors of fact have been found in them, the PAR attempts to correct the record. The PAR takes a synoptic view of the investments made over the period 1978-1988 and this provides some insights (and lessons) not readily available from the analysis of isolated projects. This has enabled the PAR to be more critical than the PCRs of IDA and Borrower performance. Special attention is paid in the audit to the efforts included in these projects to build up human resources and management capacity. Y The relevant PCRs are Report No. 6713, dated April 10, 1987; 7518, dated December 6, 1988; and 9543, dated April 26, 199L af The relevant SARs are Report Nos. 1262-BU of Januay 21, 1977, 2659-BU of January 7,1980 and 4220-BU oApril 20, 198 g? 'lwe Prealdent's Reports concerned are No. P-1983-BU of January 27, 1977, P-2684-BU of January 9,1980, P-WS26-BU of April 1983. - ii - Following customary procedures, a copy of the draft audit report was sent to the Government and its agencies for comment on April 17, 1992. Comments have been received and have been incorporated as appropriate in the PAR. The letter from the Borrower has been reproduced as an Appendix with the deletion of only the trade names of the vehicles mentioned. UU9MMRZNC AMDIT REMOR b BURUNDI FIRST EDUCATION PROJECT (CREDIT 679-BU) BASIC DATA SHEET EY PROJECT DATA Appraisal Item Expectation Actgai Total Project Cost (US$ million equiv.) , 12.51 12.461_a Underrun (1) 0.04 Credit Amount (US$ million equiv.) 10.0 10.0 Disbursed (to 04/13/82) - 10.0 Cancelled - - Repaid (to 11/91) - 0.5 Outstanding (as of 11/91) - 9.5 Date Physical Components Completed July 84 Dec. 82 No. months since Credit Signature 63 68 Proportion completed by above date 95 100.& Time Overrun (Z) 7.9 Date of Overall Project Completion Sept. 30, 82 Dec. 82 No. months since Credit Signature 65 68 Proportion completed by above date 100 100Lk Time Overrun (Z) 4.6 Institutional Performance Good Very good CUMULATIVE ESTIMATED AND ACTUAL DISBURSEMENTS (US$ million) BZA YZ9 8Q FY81 M Y3 FY84 (i) Appraisal Eat. 0.05 1.53 4.23 7.06 8.94 9.99 10.0 (ii) Actual 2.0 5.4 7.2 10.0 10.0 10.0 10.0 (iii) as 2 of (i) 2000 353 170 142 112 100 100 - iv - STAF INPUTS (staff weeks) IZ IZO II2 IIl IZ IIZ II80 IIII Ill FY38 A I8s l A Preappraleal .3 22.7 66.7 89.7 Appraisal 60.3 22.7 83.0 Negotiations 26.0 26*0 Supervision 10.7 24.1 7.9 13.5 15.0 3.6 3.7 5.6 2.5 .7 87.3 Other .8 1.4 2.0 9.3 13.M Total 1.1 24.1 129.0 68.7 24.1 7.9 13.5 15.0 3.6 3.7 5.6 2.5 .7 299.5 PROJECT DATES Item Planned Date Actual Date First Mention in Files January 10, 1974& Negotiations November 10, 1976 Nov.-Dec., 1976Ld Board Approval February 8, 1977 Credit Signature April 28, 1977 Credit Effectiveness September 1, 1977 Closing Date March 31, 1983 OTHER PROJECT DATA Borrowers The Republic of Burundi Executing Agency: The Ministr.- of National Education Follow-on Proiects Name Second Education Third Education Education Sector Dev. Credit No. 976-BU 1358-BU 1881-BU Amount of Credit (US$m) 15.0 14.5 31.5 Credit Agreement (Date) 04/23/80 06/09/83 03/28/88 -v * MISSION DATA /e Type of Date No. of Specialization Performance Type of Mission persons Represented Rating Trend Problems if La h Sector Study/ Project Identification 04-05/74 6 Reconnaissance 07/75* 2 ,0E Preparation 11-12/75 5 AE Appraisal 02-03/76* A,E,GE,Ev,S Post Appraisal 09-10i76 1 8 Post Appraisal 04/77 2 E,GE Supervision 1 06/77 1 A - Supervision 2 10/77 3* A,E,SM* 1 2 - Supervision 3 02/78 1 E 1 2 - Supervision 4 03/78* 1 Ace - -- Supervision 5 05/78 1 E* 1 2 - Supervision 6 10/78 1 A 1 2 - Supervision 7 05/79 2* A,GE 2 2 m,0 Supervision 8 09/79 1 A 2 1 0 Supervision 9 01/80* 1 Acc - - - Supervision 10 03/80 2 A,GE 2 1 F Supervision 11 09/80 1 GE 2 2 F Supervision 12 11/80 1 Acc - - Supervision 13 04/81 3 A,GE,TE 2 2 .F Supervision 14 12/81 1 GE 2 2 F Supervision 15 07/82 3 A,GE,E a.a. n.a. n.a. Supervision 16 11/82 1 GE n.a. n.a. n.a. Completion 05/84 2 A,E - - vi- ALLOCATION 01 CREDIT PROCEEDS (US$) Original Allocation Actual (Credit Agreement) Disbursement (US$ n) (as of 04/13/82)LI 1. Civil Works 5.75 2,742,926.69 2. Materials, Furniture, 1.5 5,741,644.66 Equipment, Paper & Ve'iclse 3. Technical Assistanza.L 1.2 873,491.69 4. Salaries and other 0.15 641,936.96 operating expendituresa of the Project Unit 5. Unallocated 1.4 - Total 10.0 10,000,000.00 /g The PCR Basic Data table gives a total of 13.6 ml however, in the text a table on p. 17 gives 13.41. The Borrower's Completion Report gives a total of 13.56 m. The final accounts of the Project Unit give the total shown here and in Table 2.1. 1b, This project vae reduced in scope. Of the largest component (100 primary schools) only 602 vas completed at the end of the implementation period. This 100% therefore relates only to the reduced project. Ig The PCR mention* December 20, 1973. As early as October 1963 a delegation from Burundi had held discussions with the Bank on possible financing for education and a UNESCO staff member on the economic mission of 1971 identified a possible project. However, specific reference to this First Education Project did not appear until January 1974. /d The PCR gives a date of November 12, 1976. Negotiations were held November 29-December 3, 1976. Is This table differs in several respects from that given in the PCR which contains errors regarding dates of missions, composition, type of mission, etc. An asterisk appears to indicate a correction of what appears in PCR table. Lf A * Architect; Acc - Accounting staff (Controllers, RNEA); AE = Agricultural Educator; E Economist; Ev - Evaluation Specialist; GE a General Educator; 8M * School Mapping Specialist; TE - Technical Educator. 1 - Problem-Free or Minor Problems; 2 - Moderate Pro- vmes. 1 a Improving; 2 - Stationary; 3 w Deteriorating. F * Financial; M - Management; 0 - Other. Lst disbursement date. No formal reallocation was made for this Credit although there were extensive de facto reallocations among the categories. - vil - PERFORMACI AUpT REPORT BURUNDI SECOND EDUCATION PROJECT (CREpIT 97J-pU) BASIC DATA SHEET KEY PROJECT DATA Appraisal Item Expectation Actual Total Project Cost (US$ million) 18.0 18.0L& Underrun (Z) . Credit Amount (US$ million) 15.0 15.0 Disbursed (to 04/14/82) - 15.0 Cancelled - - Repaid (to 11/91) - 0.3 Outstanding (as of 11/91) - 14.7 Date Physical Components Completed Sept. 83 Oct. 85.L No. months since Credit Signature 41 66 Proportion completed by above date (Z) 90 100 Time Overrun (Z) 61 Date of Overall Project Completion June 85 Sept. 86 No. months since Credit Signature 62 77 Proportion completed by above date 98 100 Time Overrun (M) 24 Institutional Performance Good Good CUMULATIVE ESTIMATED AND ACTUAL DISBURSEMENTS (US$ million) Xs 81 82 es 84 85 86 (i) Appraisal'Eat. 0.7 5.1 11.0 14.0 15.0 - (ii) Actual 0.8 6.0 10.6 14.1 14.6 15.0 (iii) as 2 of (i) 114 118 96 101 97 100 - vii (staff veeks) X ZA ZZ ZA lt 2. U t L ft Iya Preappraisal 0.1 5.1 45.7 0.7 51.7 Appraisal 21.9 48.6 0.4 71.3 Negotiations 18.4 7.3 25.7 Supervision 22.3 30.4 19.3 28.0 11.0 12.3 17.7 141.0 Other 0.6 Total 0.1 5.1 67.6 67.7 30.2 30.4 19.3 28.0 11.0 12.3 17.7 290.3 PROJ9C DATES lsm klannidDate Atugal Date First Mention in Files May 1978 Negotiations Nov. 29-Dec. 3, 1979 Board Approval January 29, 1980 Credit Signature April 23, 1980 Credit Effentiveness August 6, 1980 Closing Date June 30, 1986 OTHER PROJECT DATA Borroveri The Republic of Burundi Executing Agency: The Ministry of National Education Follow-on Progects Name Third Education Education Sector Development Credit No. 1358-8U 1881-BU Amount of Credit (US$m) 14.5 31.5 Credit Agreement (Date) 06/09/83 03/28/88 - ix Type of Date No. of Specialization Performance Type of mission persons Represented Rating Trend Problems Sector Study/ Identification 02/78* 4 GE,TZE,,AE Project Identification 07-08/78 2 AE,GE Preparation 02/19 5 A,AE,GE,,TE Appraisal 05/79 5 A,AE,,GE,TE Post Appraisal 09/79 1 GE Supervision 1 03/80 2 A,GE 1 1 - Supervision 2 09/80* 2 GE,TE 1 1 - Supervision 3 04/81* 3 A,GE,TE 2 1 - Supervision 4 12/81 1 GE n.a. .a. - Supervision 5 07/82 3 A,E,GE 1 1 - Supervision 6 11/82 2 GE,TE 1 1 - Supervision 7 10/83* 2 A,E n.a. n.a. - Supervision 8 05/84 2 A,E 1 1 - Supervision 9 02/85 1 A 1 1 - Supervision 10 09/85* 2 A,GE 1 1 - Supervision 11 01/86 2 A,GE .a. n.a. - Completion 02/87 2 A,TE - - - ALLOCATION OF CREDIT PROCEEDS (US$) Original Revised Revised Revised Actual Allocation Allocation Allocation Allocation Disbursements (DCA) (as of (as of (as of (as of 06/08/81) 03/02/83) 06/27/85) 04/14/86) (US$ m) (US$ M) (US$ M) (US$ m) 1. Civil Works 6.280 6.98 2.7 2.3 2,387,226.81 2. Construction 5.150 4.55 7.8 9.5951 9,6Z5,441.40 Materials, furniture, equipment and vehicles 3. Technical 1.145 1.045 2.0 2.1 1,951,692.30 Assistance 4. Project Administration (a) Salaries 0.130 0.130 0.8 0.593 592,721.65 of Project Unit staff and other. carrying out the basic ed. study (b) other 0.160 0.160 0.7 0.320 321,064.89 operating costs of the Project Unit and the basic educ. study 5. Refunding of 0.135 0.135 0.092 0.919 91,852.95 Project Preparation Advance 6. Unallocated 2.000 2.000 0.408 - 7. Disbursement 0.5 - to Special Account** Total 15.0 15.0 15.0 15.0 15,000.000 La This Includes items added to the original project. The cost of the original project components is given in the PCR as US$13.05 million. This excludes the additions to the original project; these additions were physically completed by April 1986. A *Architect; AE - Agricultural Educator; E * Economist; GE - General Educator; TE * Technical Educator. I a Problem-Free or Minor Problems; 2 - Moderate Problems; 3 * Major Problems. I - Improving; 2 - Stationary; 3 - Deteriorating. f 7 - Financial; N - Managementl T - Technical; P - Political; 0 * Other. * Differs from PCR record. ** New disbursement head added on March 2, 1983. - xi - PERFORMANCE AUDIT REPORT BURUNDI THIRD EDUCATION PROJECT (CREDIT 1358-BU) BASIC DATA SHEET KEY PROJECT DATA Appraisal IUnm Expectation Actual Total Project Cost (US$ million equiv.) 19.30 16.61L, Underrun (Z) 32 Credit Amount (US$ million) 15.8 (14.5 SDR) 15.8 (14.5 SDR) Disbursed (to 09/88) - 15.8 Cancelled (as of 09/13/88) - (SDR 73.56) or US$95.00 Repaid (to 11/91) - - Outstanding (as of 11/91) - 20.24 Date Physical Components Completed June 87 March 88 No. months since Credit Signature 49 58 Proportion completed by above date (Z) 90 100 Time Overrun (%) 18 Date of Overall Project Completion Dec. 89 March 88 No. months since Credit Signature 55 58 Proportion completed by above date - 100 Time Overrun (M) 5 Institutional Performance Good Good CUMULATIVE ESTIHATED AND ACTUAL DISBURSEMENTS (US$ aillion)/b FY 84 85 86 87 88 (i) Appraisal Est. 1.4 4.6 9.0 13.8 15.8 (ii) Actual 4.23 8.61 12.77 14.98 15.80 (iii) as 2 of (i) 323 187 142 109 100 - xit - sF IUTS (staff veeks)Ic 86 87 8A Z9 Total Preappraisal 25.9 3.2 29.1 Appraisal 5.6 5.6 Negotiations n.* n.e. Supervision .4 17.9 9.9 9.8 8.3 1.8 0.2 483.0 Other Total 25.9 9.2 17.9 9.9 9. . 3 1.8 0.2 517.7 PROJECT DATES Item Planned Date Actual Date First Mention in Files February 1980 Negotiations March 23-26, 1983 Board Approval May 10, 1983 Credit Signature June 9, 1983 Credit Effectiveness September 9, 1983 November 16, 1983 Closing Date June 30, 1988 June 30, 1988 OTHER PROJECT DATA Borrower: The Republic of Burundi Executing Agencys The Ministry of National Education Follow-on Projects Name Education Sector Development Credit No. 1881-BU Amount of Credit (US$m) 31.5 Credit Agreement (Date) March 28, 1988 - xiii - ALLOCATION OF CREDIT PROCEEDS (SDRs)ld Original Allocation Actual Disbursement (as per Credit Agreement) (as of Sep. 1988) (in SDRs) SDR million US$ million (1) Civil Works 2.3 2.5 3,308,342.31 (2) Construction Material furniture, equipment and vehicles 6.7 7.3 7,660,753.01 (3) Consultants and specialists services 2.1 2.29 2,414,816.11 (4) Fellowships 0.07 0.08 112,778.46 (5) Salaries and operating costs of PIU 0.43 0.47 835,992.41 (6) Initial Disbursement to special account 0.7 0.76 167,244.14 (7) Unallocated 2.2 2.4 Cancelled 73.56 Total disbursed (14,499,926.44) Total 14.5 15.8 14,500.000.00 - Xiv - MISSION DATA Type of Mission Date No. of Specialization Performance Trends Type of persons Represented Rating Problems Reconnaissance 11-12/81* 1 Media Specialist Reconnaissance 12/81* 1 GE Preparation 03/82* 6 A,E,GE,GE,TE,TP Appraisal 06/828 4 A,E,GE Post Appraisal 11/82 1 GE Supervision 1 12/84 2 A,E 1 1 Supervision 2 05/84* 2 A,E, 1 1 Supervision 3 02/85* 1 A 2 2 F Supervision 4 09/84 2 A,GE 1 Supervision 5 01/86 2 A,GE I I Supervision 6 06/86 2 E,GE 1 1 Supervision 7 11/86 2 A,GE I I Supervision 8 09/87 2 A,GE No report seen J- Completion 1* 09/88 1 E Completion 2*Lj 03/89 2 A,GE / Actual cost of original Project components. The expansion agreed to during project implementation raised the actual total cost to US$18.53 million. lb The first FT (FY84) of the appraisal estimate and all the years of the actual disbursement are wrongly stated in the PCR table. The calculation of percentage in the PCR shows an error in FY84 as well. These figures are adapted from the PCR which did not use the prescribed readings. & Data regarding actual disbursement not Iamediately available in US dollars. e A a Architect; E - Economist; GE - General Educator; TE - Technical Educator; TP - Textbook Production Specialist ft 1 * Problem-Free or Minor Problems; 2 * Moderate Problems 1 - Improving i F - Financial; M - Management; 0 - Other; T - Technical /I The only reference to this in the files is the terms of reference: there is no mission report in the files. The PCR records it as a mission. Ll A reconnaissance mission in September 1988 was undertaken to prepare for the PCR. Subsequently, UNESCO sent a mission to prepare the PCR around March 1989, but the report of this mission was not used by IDA. IDA staff then prepared the PCR which was published. * Differs from PCR table. -xv- PERFORMANCE AUDIT REPORT BURUNDI FIRST EDUCATION PROJECT (CREDIT 679-BU) SECOND EDUCATION PROJECT (CREDIT 976-BU) THIRD EDUCATION PROJECT (CREDIT 1358-BU) EVALUATION SUMMARY Introduction included a lower secondary school component and new nonformal centers. Of the total cost of 1. The first three education projects in these projects, about 41% was allocated to Burundidi were generated over the decade 1974- primary education (and related teacher training), 83. The first (Credit 679-BU for US$10 32% to technical/vocational education, 6% to million), was signed in April 1977, the second nonformal education, less than 5% to secondary, (Credit 976-BU for US$15 million), in April about 4% to support administration and plan- 1980, and the third (Credit 1358-BU for ning and just under 12% to prject adninis- US$15.8 million), in June 1983. All credits have tration. been fully disbursed, except for a sum of US$95 from the third that was canceled. The projects bplementation Exnence have been successfully implemented. 4. The projects were implemented by a unit Objectives and Content under the Minister of Education. Time and cost performance was generally within appraisal 2. Significant differences emerged during estimates and savings realized in the Second and project generation between the country and IDA Third Projects were used to enlarge project on development priorities. Enrollment ratios at scope. all levels were very low. While the country was interested in a broad range of assistance to 5. Changes occurred during the implementa- primary, secondary, technical and higher educa- tion of all the projects. For the First: (i) only tion, IDA was primarily concerned with basic 60% of the primary/nonformal education centers (primary and adult nonformal) education and (CFPs) were built; (n) instead of minor reluctant to invest in secondary and higher extension of an existing institute to provide for education. training practical subject teachers, a new institute was constructed, and (Mi) at CTPs, 3. The ist project aimed at expanding and increases of 100% in workshops and 66% in improving primary education, textbook produc- teachers occurred. The textbook production tion and curriculum development services, nonf- unit and curriculum unit were strengthened as ormal education for adults and youth, and train- planned. ing for teachers of practical subjects. The see- gn sought to expand and improve technical/ 6. In the Second Project all the planned items vocational education and render logistical were completed: (i) one new technical and one support to Ministry of Education services. The new vocational school; (H) extension of two third reainforced the objectives of the first and technical schools to add new courses; (il) studies -xvi - of basic education and technical education, and tion, staff, supervision, supplies, public (iv) logistical support (mainly vehicles) for MNE interest and awareness and links between offices. The savings of US$5 million provided 16 the learning process and earing capacity; CFPs, rehabilitated 11 secondary schools and in 1991, grossly under-enrolled, they were increased the technical assistance services. suspended and the Government is planning a reduced scale and new style of operation. 7. The Third Project achieved its targets: (i) expanding primary teacher training (320 places) (b) The practical subject teacher training insti- and lower secondary education (1,920 places); tution, after some years of training both (ii) providing 20 rural social (nonformal educa- practical subject and general primary tion) centers (CSEs); (iii) expanding textbook teachers, has been closed down pending the production capacity and (iv) strengthening outcome of the review of nonformal educa- school mapping. The savings (US$2.7 million), tion and training. helped to improve technical school libraries, provided for secondary school maintenance and 11. In the area of institution building, the increased boarding, provided new offices for projects have helped to provide a competent BER, extended the printshop, built more CFPs primary curriculum, unit (BER) and to and increased technical assistance expert services strengthen textbook production. However, some (largely for project management) and project components included insufficient fellowships. measures to build local capacity and, after imple- mentation of the projects, institution building 8. The standard of building design and needs remained in educational planning, project construction was good. The project unit management, general administration and competently undertook most of the construction supervision of the system, and, for a few years, by force account (for want of a strong local in textbook production as wen. construction industry). Thus the projects have helped very little to build -; local construction ustainability capacity outside the project unit itself. 12. Sustainability in the case of these projects Results varies according to subsector or activity. 9. Project outcomes have been mixed. The (a) The formal pma1y school program is expansion has been achieved in primary, generally sustainable. The practical sub- secondary and technical/vocational education. jects program is working and sustainable in The primary teacher training component of the some schools while in others it is not Third Project increased secondary enrollment working. but, as students do not choose the pedagogical stream to the extent expected, these enrollments (b) The nonformal groffam for adults (and have been below target. Textbook production is out-of-school children) both in the CFPs prc ceeding, with some funding problems and in the CSEs is not sustainable for the attnbutable to shortfalls in Government moment as they have been poorly attended subver tions and book fees. and supported and the program has been suspended and is being redesigned. 10. The results in nonformal education have been negative: (c) The seconda1y schools and EFIs are sus- tainable, even though there is need to (a) The CFP and the CSE programs suffered reduce over-crowding to derive optimum deficiencies in prior planning and prepara- benefit. - xvau- (d) The technical/vocational schools are sus- (d) The selection of reform goals in primary or tainable. technical education that recognize the difficulty of changing attitudes (paras. 3.17, (e) The textbook production program is sus- 3.21) and allow adequate support to units tainable at its current level, but will need piloting the reforms--this was done in part more assured financing arrangements to for primary education (para. 3.31) but not provide greater benefits. for technical education (paras. 3.21, 3.35). (f) The curriculum unit (BER) for primary (e) The need for greater care in preparing education and the project management unit Credit Agreements, and in maintaining (BPE) are both sustainable. However the project tiles than was noted in these strong and effective construction enterprise projects (paras. 4.8-4.13). within BPE may need to be reviewed as it - depends heavily on IDA-financed projects (f) The need to provide for supervision of the and credit financing for its existence. use Rnd maintenance of buildings and equipment, in this case to guide principals (g) Overall, the declining share of the national of schools and support village communit budget going to Education is an issue in maintenance and undertake urgent relevant to sustainability that needs to be repairs (paras. 2.9, 3.20, 4.23). addressed. (g) The value of more rigorous project Conclusions and Lessons appraisal, in particular: 13. A few conclusions and lessons are drawn 9 the use of the services of a nonformal from the experience: education specialist to appraise the non- formal education components of Educa- (a) The importance of planning as a basis for tion I and M (paras. 3.12-3.14); project identification and design (paras. 1.7, 3.18, 3.39). 0 closer attention to cost estimating at appraisal (paras. 2.27-2.28); (b) The need for projects to address the crucial problems of the sector and not be based on 0 proper assessment - through reappraisal - pre-determined positions on priorities; of the major changes proposed in the where secondary and higher education project during the implementation of enrollment ratios are very low, as they were Education I (paras. 2.14, 3.10-3.14). in this case, a crisis can develop regarding the production of the middle and higher (h) I view of the extensive use of force level manpower required for key leadership account in construction in these projects, and product.ve roles in the economy (paras. the need to consider promoting greater 1.20-1.21). participation by the construction industry as a whole, while developing the expertise (c) The importance, in addressing basic needs already built up in BPE construction (primary and literacy classes), of ensuring services (paras. 2.15-2.23, 4.23). that relevant administrative services are equipped to conduct these activities (i) The need for supervision missions to moni- efficiently (paras. 1.21, 3.14, 3.47-3o48). tor the required maintenance of separate accounts for each project (paras. 2.24-2.25). (j) The importance of detailed prior agree- 0 staff training as a means of building ment on the details of the recurrent permanent capacity, especially through financing of textbook production (para. fellowship programs as the experience of 3.23). these projects shows to have been successful (paras. 3.31, 3.33, 3.35, 3.39). (k) In the design of technical assistance compo- nents, the need for close attention to: (1) The advisability of placing the custody of stores of materials in the Project Unit * genuine IDA/Borrower consensus in under a department such as Finance and advance on the program aims and con- Administration so that it is independent of tent (paras. 3.25-3.28); the Technical Services (para. 4.16). * measures to eliminate future dependence (i) So as to promote sustainability of project on foreign expert services (para. 3.40)--in benefits, the importance of protwIng the particular, by requiring counterpart train- sector share of the national budget during ing (paras. 3.40, 4.11) and mandatory implementation of the projects, which, in limits on the term over which any this case, IDA did not address (paras. 1.19, individual expert can be financed out of 4.18-4.19) under these projects. loan/credit proceeds (para. 3.42); and PERFORMANCE AUDIT REPORT BURUNDI FIRST EDUCATION PROJECT (CREDIT 679-BU) SECOND EDUCATION PROJECT (CREDIT 976-BU) THIRD EDUCATION PROJECT (CREDIT 1358-BU) I. PROJECT BACKGROUND Introduction 1.1 Burundi (which gained independence in 1962) is one of the smallest and poorest African countries and in the 1960s had one of the lowest literacy rates on the continent. It was (and remains) almost entirely dependent on agriculture. Its population was fairly evenly distributed. Bank Group financing of educational development in Burundi started relatively late. By the time that the first Education Credit was approved for Burundi in FY77, neighboring Tanzania had already had five credits in the sector. 1.2 In 1969 Burundi obtained UNESCO assistance to carry out a major reform of education. This involved restructuring the system; "ruralizing" the primary curriculum, adding practical subjects to it; using Kirundi as the language of instruction in primary schools; providing literacy and basic skill training at primary (community) schools for youth and adults, reorganizing secondary level education (including technical/vocational) and consolidating higher education. 1.3 A Bank Group Economic Mission of late 1971, which included a UNESCO staff member to review the education sector, found that there was need to improve the efficiency of primary education and the external productivity of technical education. Serious disturbances occurred in the country in 1972 and this appeared to dispose the Bank Group to move urgently on education assistance and project generation efforts began in 1974. 1.4 The First Education Project included 100 primary schools--multi-purpose learning centers (centres de formation pratique-CFPs)--to provide formal primary schooling to children and non- formal education and training for adults and youth. Other components were for textbook production, modest facilities for training workshop teachers for the CFPs, and technical assistance for primary education curriculum development. 1.5 Following approval of the First Education Credit in February 1977, the Second and Third were approved rapidly - in January 1980 and May 1983, respectively. The Second provided for technical/vocational education while the Third provided for primary, primary teacher training and general secondary education. The Development Strateey 1.6 A cursory look at the content of the three projects would suggest that these are fairly typical and, therefore, that the development objectives of IDA and Borrower were in harmony - expansion and improvement of educational opportunity at primary, lower secondary, primary teacher training, technical/vocational and nonformal adult education levels and the strengthening of various -2- administrative services related to the projects. However, within each subsector there were differences of objectives and variations in tempo and riorities and in one important area - higher education - there was no agreement. The strengthening of administration also showed discrepancies between the Government's needs and the very narrowly circumscribed areas of IDA interest. 1.7 The overriding problem in generating all of these projects was that there was no planning basis for them. During the eight years of the project generation period - from June 1974 (when the first real stirrings occurred in the generation of Education I) to June 1982 (when Education I was presented to the Board) - there was no educational planning to speak of. . However, the education programs which appeared in all the national Five Year Development Plans (FYDP) - from Plan I (1968-72) to Plan V (1988-92) - provide a fair idea of Government thinkii.g, which, although in some respects unrealistic (especially in terms of implementability) was quite clear and consistent on certin essentials and on the problems faced. One can therefore compare the Government and Bank strategies. 1.8 Government Strategy. Under FYDP I (1968-72), the Government recognized a need to improve the productivity of the education and training system, hoping thereby to achieve much needed growth without increasing the 23% sector share of the national budget. Yet, against the stark picture of only 100 Burundian males and two females graduating from secondary general education and a 16% enrollment ratio in rural areas at primary level (26% nationally), the implications of enrollment stagnation were serious. The lack of local engineers, for example, was cited as a reminder of the severity of the human resource problem. The financial burden on the Government of boarding in secondary schools was noted as the basis for the idea that parents should share in the cost and that such schools should undertake income producing activities. 1.9 FYDP II (1973-77) noted that the primary enrollment ratio, far from rising at the projected rate of one percentage point per year under FYDP I, had fallen from 28% in 1967-68 to 19% in 1972-73. Hence this Plan aimed at increasing this to 23% and, in order to better cope with manpower needs, diversifying upper cycle secondary education. In 1977, education was still receiving 23% of total government expenditure. 1.10 FYDP IH (1978-82) showed considerably more awareness (than the previous two plans) of the importance of education in shaping society and promoting economic development. Reviewing the unsatisfactory results of previous plans, it emphasized, much more than those plans had done, that the shortage of trained persons was a more significant stranglehold on development than the shortage of money. Some of the main concerns included making primary education polyvalent (and so appropriate to rural needs), expanding secondary education, promoting adult literacy, and planning higher education needs in a more precise manner. The preparation of this plan coincided roughly with the generation of Education I and the Plan was in the course of implementation when Education II was generated. By 1982, the education share of total government expenditure had dropped to 19%. 1.11 FYDP IV (1983-87) expressed satisfaction with the results of the Third Plan (which was the first to have benefitted from Bank Group financing of Education). The lack of secondary graduates was considered the prime reason for the feebleness of national production. The development aims continued to encompass expansion of primary and secondary education as well as higher education. Preparation of this Plan and the generation of Education II, which assisted lower -3- secondary education, were concurrent to some extent. In 1987, education still received 19% of the total government expenditure. 1.12 FYDP V (1988-92) stressed the need for localities to share in providing and maintaining primary schools, the hope that secondary schools would help to change the public attitude of waiting for the state to do everything and the frank recognition that nonformal education and training was disorganized. Of the total 1989 Government expenditure, at the end of Education III, only 16% was spent on education. Thus, over the period of association with the three projects, the sector share of the national budget had declined by seven percentage points. 1.13 In general, therefore, the Government aims were clear. Consistently it sought to step up the transition rate between primary and secondary by expanding secondary education. National development required not only a literate population, in which the Government was strongly interested, but also higher and middle level staff suitably educated and trained; hence. the development of the system had to be multi-pronged. At the same time, while there were trends in the national budget that were not entirely favorable to education, these should not be seen as deliberate lack of Government interest in the sector. As the SAR for Education III points out, the Government in 1983 was prepared to raise the education share of the budget to 25% by the late 1980s. 1.14 Bank Group Strategy. The Bank Group strategy on the other hand had a different rhythm and focus in both project generation and sector work. First of all, the Bank Group in the 1960s was hesitant to become involved in the sector for reasons that are unclear. The Government asked informally in 1963 (on the occasion of the Annual Meetings) for aid to technical education and in 1969 (on the occasion of a visit by a reconnaissance mission) for aid to secondary education. Nothing resulted from these requests. The department concerned in 1969 was.disinclined to give priority "to any operation for small countries like Rwanda and Burundi"--a position that was surprising in regard to a country of over three million people when smaller countries like Jamaica had already received loans in education and other sectors. The 1971 economic mission proposed nonetheless that financing be considered for primary teacher training. Throughout the 1960s the Bank Group policy was not to lend for primary education; the 1971 Sector Working Paper changed this. 1.15 Once a specific project began to be considered, the Bank Group aversion to investment in the subsectors above basic education began to manifest itself. The Bank Group staff member on the 1974 Bank/UNESCO Project Identification Mission emphasized that, after 1972, with the inequitable representation of the two major ethnic groups in secondary level education, it would be inequitable to finance secondary education; hence, the focus should be on primary and adult basic education. Such a position, ironically, did more harm to the disadvantaged because it was only through expanding the secondary system that opportunities for them could have been significantly widened. A second influence that was never mentioned specifically was the wave of interest shown in the Bank in basic education, following the McNamara poverty alleviation speech in Nairobi at the Annual Meetings of Governors in 1974. In this post-Nairobi climate, secondary and higher education were often seen in the Bank Group, not in the context of their crucial contribution to development, but rather in terms of promoting elitism and social inequity; hence they were de-emphasized. This trend away from higher education and toward basic education was reflected in the lending program in the next decaCle as internal Bank reviews of Bank Group lending in the primary and higher subsectors showed. -4 1.16 Direct project generation efforts began in mid-1974 with a UNESCO project identification mission in which a Bank Group staff member participated. The mission originally considered a balanced project with 20 primary schools, a technical, a vocational and a para-medical school, textbooks and support for educational radio. However, eventually the report proposed a much less balanced package: 50 primary community schools and a para-medical training school. An IDA July 1975 reconnaissance mission rejected involvement in both secondary education (pending its reform) and technical education. The reasons given were the lack of manpower data and the high unit costs which the mission attributed to boarding and a low student to teacher ratio. The mission did not pay attention to the high (40%) proportion of costly expatriate teachers and the alarmingly low enrollment ratio of only 2% at secondary level. The Minister of Education asked for reduced assistance to primary in favor of secondary, technical and higher education; nonetheless, this mission went further along the oppos:te route and proposed 80 primary schools and no other type. Eventually the First Education Project included 100 primary schools - multipurpose learning centers (Centres de Formation Pratique - CFPs) and some support for the curriculum development service for the primary subsector--BER. Nothing was included for secondary or technical education. In the light of this project generation background, it was inaccurate of the 1978 project brief concerning the proposed Education II project to have stated that the Bank Group was asked under Education I to essist primary education because other donors were helping secondary and higher. 1.17 For the proposed second project the Bank Group identification mission in 1978 suggested technical schools with little boarding, and technical teacher training along with studies of basic and technical education.1 The UNESCO preparation mission, at Government's request, (1) inLreased the boarding, (2) added new lower secondary schools and support for libraries and laboratories in existing general secondary schools, (3) added a post secondary institute for agricultural training and (4) proposed to strengthen planning. The 1979 appraisal mission stated that, because of its strong objection to a four year lower secondary cycle, it had deleted the lower secondary schools. This objection was inconsistent with the fact that the Bank Group had accepted the four year lower cycle secondary education in several francophone African countries that were receiving assistance to education at the time. The post-secondary agricultural institute was also deleted. In the end, the agreed project included nothing to strengthen planning, secondary education or, more to the point of the project, technical teacher training. 1.18 The Bank's Education Sector Memorandum of 1981 was important, coming as it did after the first two education credits had been approved. While this Memorandum asserted that Bank involvement in education in Burundi would assist the Government to make decisions and achieve objectives (and IDA had indeed argued its involvement in primary education in Burundi on the grounds that this would help to achieve the objectives of the reform), it -lso stated, (ironically enough) that higher education required reorganization before the Bank Group should consider investing in it [a position reminiscent of the 1975 stand on secondary education]. The report found that the education system was developing in random fashion without "actual planning" and observed that the modem sector labor market was flooded. This last observation (a) implicitly called into question the Education II investment in modem sector technical/vocational education that had just been approved, (b) ignored the serious issue of dependence on expatriate manpower and (c) steered away from the contemporary Government claim that lack of trained persons was the major constraint y/ For some reason, technicalNocational education did not have the same stigma attached to it as secondary and higher did in that era In the Bank. -5- on development Nevertheless, the Memorandum concluded with what had been the view even before the study began--that it was the traditional sector that required concentrated attention. It stressed the need to continue educational reform, support nonformal education, promote internal efficiency in the system and long term educational planning [which Education I supported lightly and Education I was not designed to support], promote geographical equity, cooperate with private agencies, relate enrollment to employment needs and adopt "new financial policies". These "financial" policies involved, for example, the levying of fees, employer sponsorship of technical training, and choice of curricula according to employment requirements. In fact, they had serious educational and social implications and were not simply financial Many of these concerns about "financial policies" became prominent in the context of generating Education IV which was approved in 1988. By that time, according to figures received by the audit mission from the Ministry of Finance, the education share of actual total government expenditure had declined to 17% and then 16% in 1989. 1.19 The identification mission of 1981 for a proposed third project suggested assistance toward basic education (primary and nonformal), related teacher training and educational planning and emphasized the need for efficiency and cost reduction. The 1982 Project Brief still objected to participation in secondary education unless the cycle were shortened to three years. Eventnally, however, with the AfDF ready to assist in a large expansion of secondary education, IDA agreed to include some lower secondary schools in Education m, even with the four year cycle, amounting to just 12% of project costs. On one significant point a difference in priority between IDA and the Government emerges. The SAR for Education I (para. 2.36) commenting on the Government's readiness to raise the education shar of the budget to 25% in the late 1980s, expressed doubt that the Ministry of Education "could [sic] or should fully absorb" this share, "given Burundi's need to finance other sectors" and its intention to reduce unit costs in education.' 1.20 Bank reluctance to finance secondary and higher education seems plausible in the light of low enrollment ratios in primary education. However, a country of three million people with such low secondary enrollment ratios in the late 1960s and 1970s was a cmuntry in deep crisis as far as development and the pursuit of economic growth are concerned. In all its d-mrsion of country strategy, the Bank Group never prominently addressed the excessive and costly dependence upon expatriate manpower even within the education sector. Graduates of any secondary or higher education program, reformed or unreformed, were urgently needed. Reform or reorganization at these levels would have been facilitated by Bank Group involvement. University development in Burundi was instead left to proceed on the basis of uncoordinated donor initiativc which have left in their wake a proliferation of underfinanced and uncoordinated programs. Witho,t the benefit of a master plan for development and careful programming (the need for which th,, government recognized in its Third FYDP), which Bank involvement could have helped to secure. he university now lacks the research and library facilities needed for optimal operation and fa-.es such basic problems as conducting a Faculty of Agriculture without a farm. 1.21 Females were a stated area of interest in the basic education components of Education I & II; but this was in no way a justification of the selection of basic education as the area of emphasis because the position of girls was most inequitable not at primary and nonformal lavels but V The Regional Office in commenting on the draft of this Report, stated (a) that th- declining share of education in total public expenditure in the mid 1980's reflected shifts in intersectoral priorities at a time when government embarked on structural adjustment, and (b) that by the beginning of 1987 when the Fourth Education Project was appraised, the share of education in total public etpenditure had risen to 20%. at the secondary, technical and higher levels. Nor was the pattern of other external aid to education a justification, since such aid was directed at all subsectors, including primary and non-formal education. The audit takes the position that the IDA concentration on primary and nonformal education did not simply create an imbalance; it was well beyond the absorptive capacity of the country--as expanded, the primary system could not be managed properly and the nonformal system could not be fully utilized or managed well. On the other hand, the minimal involvement of these projects in secondary education and the omission of higher education did littlo (i) to help the country to provide for its leadership and to render itself independent of large numbers of expatriates and (ii) to promote a proper relationship between programs at these expensive levels and the resources to conduct them efficiently.3 ProJect Objectives and Scope 1.22 The three projects can be taken together as a series that responded in a broad sense to the Government's wishes and priorities (with the exception of higher education which the Bank Group avoided), although the timing and mix did not always meet these priorities. 1.23 All projects supported aspects of change in the activities and subsectors which they touched and assisted project implementation by financing salaries of experts and operational expenses. 1.24 Education I was designed to help in the reform of primary education, expand the system and help Government to increase its control over the subsector. Thus it included 100 primary schools (CFPs), facilities to train practical subject teachers, support for textbook production and curriculum development, and training of one person in educational planning. 1.25 Education H was intended to provide facilities for the training of middle level technicians, skilled workers in various trades and in secretarial and administrative fields, evaluate the reform of primary education, review technical education and strengthen the school supervisory services in secondary education. Thus it included new and expanded technical/vocational schools and support for the relevant ministerial services. 1.26 Education W aimed at continued expansion and improvement of primary education (25% of project cost) and primary teacher training (16%), expanding lower secondary education (12% of cost) and non-formal rural education and training (16%), textbook production and logistical support to key ministerial services. The project therefore included primary, teacher training and general secondary schools, rural social centers, and assistance toward educational planning and evaluation. 1.27 The content of the projects at the time of Credit Signature can be summarized as follows: I/ In commenting on the draft of this report, the Regional Ofoe states that (a) the low prinary enrollment ratio justified the urgency to address the Issue of participation in basic education and assist Burundi In upgrading and expanding its human capital base, (b) moreover, other donors were already assisting Burundi in developing other levels of education and (c) it Is important to note that the audit focusses solely on IDA assistunce. -7- Subsector or Activity Education I Education II Education III Assisted A. (1) Primary Schools 100 24 (CFPs) (2) Primary Extension of one - 2 Ele (with Teacher Training teacher training attached lower Institution to train secondary practical teachers section.) (3) Lower Secondary - - 2 attached to Schools above BFle & 4 separate schools (4) Technical/ I now technical Vocational Schools school; 1 new trade schooll extensions to two existing tech. schools. Two experts (7 years) for two schools. (5) Ron formal (included in 100 CFPs - 20 rural social education/training above) centers; 1 or 2 centers specialists (2 years) in rural women9e education and training B. (1) Textbook Expansion of print - Printing equip- production shop facilities; ' ment; vehicles for printing equip-menti book distribution; vehicles for book 200 tons of paper; distribution; book 3 years expert storage facilities at services; 3 years 400 schoole; 250 tons fellow-ships for 4 of paper; expert for staff. 3 years (2) Curriculum 12 years services of vehicles and development 4 experts; 6.75 years equipment for BEET fellowehips for 9 (technical education stafF of BER (primary office); vehicles education) for BEPES (secondary education office) (3) Educational 1 9-month fellow-ship 6 months expert to 1 year of planning to IIEP (Paris) study basic consultancies in education; 6 months evaluation and expert to study planning equip- technical training; ment and operating vehicles for expenses for planning unit. school mapping. (4) Education Vehicles for administration Secondary inspectorate. (5) Project 8 years of expert 9 years of expert 19 years of expert management services; operating services; operating services and expenses. expenses. operating expenses. Implementation Plan 1.28 The implementation of the three projects depended upon a project implementation unit (BPE), responsible directly to the Minister of National Education. Project Cost and Flnancing Plans 1.29 The appraisal estimate of the cost of the three projects and the financing planned for these were as follows: Education Education Education Total __________________I II III _____ A. Appraisal Cost Estimate (US$m)a 12.511 18.000 19.27 49.781 B. Approximate share of total cost (2) 1. Primary Education (a) Primary School. 75.41 - 24.91 28.59 (b) Primary Teacher Training - - 16.50 6.39 (c) Primary Textbooks 6.07 - 12.71 6.45 (Total Primary) (81.48) - (54.12) (41.43) 2. Non-Formal 1.07 - 15.78 6.38 3. Secondary - - 12.30 4.76 4. Technical/Vocational - 89.25 - 32.27 5. Administration/ Planning 9.9 Lb 3.25c ) - 14 3.66 6. Project ) Administration 7.55 7.5 ) 17.80 11.50 TOTAL 2 100.0 100.0 100.0 100.0 C. Financing Plan (US$M) IDA 10.0 15.0 15.8 Government 2.511 3.0 3.5 a For details see Tables following para. 2.25 below. 1 For curriculum development (BER) and one education planning fellowship. c For BEET and BEPES, and studies of primary education and technical education. d The amount allocated for technical assistance for school mapping, non-formal education, teacher training and project administration cannot be der-rmined from the SAR. -9- 1.30 Of the total cost, therefore, the three projects taken together allocated at the outset some 41% to primary and related teacher training; 32% to technical/vocational education, 6% to nonformal, just under 5% to lower secondary, about 4% to support administration and planning and just under 12% to project administration. H. IMPLEMENTATION EXPERIENCE Project Management 2.1 The overall performance in implementing the three projects was good. BPE was generally well managed and its staff was on the whole competent and well motivated. IDA operational staff often cited two points in favor of BPE - the timely manner in which the projects were implemented and the actual cost which generally fell below appraisal estimates. Physical Implementation (a) Achievements 2.2 Physical implementation of the projects was generally satisfactory. 2.3 Education L Fifty, rather than 100, CFPs were built under force account by BPE.4 The program was halved for several reasons: the large additions to the construction program for teacher training and BPE absorbed much of the resources, and delays and price increases arising from the Tanzania-Uganda war also took their toll. The printshop facilities were provided, with the addition of a stock room. Overall, therefore, the construction program achieved was substantially different from that originally envisaged. 2.4 Education H. Construction of the two new technical/vocational schools was scheduled to be procured through ICB. When the bids received were rejected by Government, BPE was allowed to construct these schools under force account. The whole construction program, benefitting from an early start (thanks to funding of materials and project unit salaries out of Education I), was completed well ahead of schedule and well below the appraisal cost estimate. 2.5 Education W. The four COs, two EFI/COs, the 24 CFPs and the 20 CSEs were provided, also well below appraisal cost and well ahead of schedule. I/ The records provide no consistent story of progress in constructing CFPs or under which project they were provided. The construction program was continuous and at each point ihe picture of completion changed. BPE has adopted a sensible way of doing it: October 1979 - November 1983 - 60 were completed and attributed to Education I November 1983 - April 1986 - 16 more were done and attributed to Education II August 1984 - June 1987 - 29 more were done and counted under Ed. III The cut-off points correspond tc the Clsing Dates as BPE saw them. The Educ. I construction prog dm produced SO CFPs. However, construction continued without interruption under Education IL The Borrower PCR for Ed. II notes that nine CFPs actually built under Ed. II were accounted for in the Borrower PCR for Ed. I. The IDA PCR for Ed. 11 states (para. 4.09) that 50 were built under Ed. I and 10 more under Education II using funds from Education L The final record shows that 60 were provided under Ed. 1, 16 under II and 29 under IIL - 10 - (b) Quaft of desin and construction 2.6 With regard to the overall construction program, the audit mission found that the buildings were generally well designed to minimize maintenance and provided with good lighting and ventilation. Nevertheless, there were design problems noted in the Education M PCR, most of which were subsequently corrected.5 Thg draft UNESCO-prepared PCR for Education III also noted some design and construction flawso that suggested the need for better quality control on the part of BPE, and the audit mission also observed similar problems.7 In addition, the widespread practice has been to lay out the buildings on a campus in a scattered way iistead of economically placing them along a covered walkway. The omission of ceilings in some buildings, intended as an economy measure, resulted in noise levels in classrooms during rain showers that were high enough to curtail teaching. 2.7 Construction was solid and generally well done. Nevertheless, at he time of the audit visit there were signs of deterioration that stemmed from flaws in construction. The CFMP suffers from widespread damage to ceilings caused by roof leaks; one of the buildings on this campus also had an extensive set of cracks in the floor. 2.8 It is quite evident that, despite (a) the fact that the DCA for Education I specified that BPE should prepare a manual for school maintenance and brief school principals in proper maintenance procedures and (b) a design that generally minimizes maintenance, the schools are not being properly maintained. At Cibitoke EFI/CO. where the audit mission observed that mosquito netting in both the male and the female dormitories was damaged, the mission was informed that the incidence of malaria in the two schools was so high that it had a noticeable impact on the grade repetition rate. Yet, the inexpensive repairs needed to protect the students in the crowded dormitories from the consequences of the damaged netting were not being undertaken. The digesters provided at three schools visited by the mission--a fundamentally sound idea to provide both for sewage disposal and for gas for cooking--were not functioning in two of the cases. Another concern is that at the ETMM in Bubanza each dormitory is provided with two exit doors, one at each end of the building; one of these is kept locked so as to better control entry and egress; this is a serious hazard in the event of fire (and smoke) because escape through windows is impossible. Ironically, the ETMM teaches construction trades. 2.9 In general, therefore, a fundamentally sound design and good construction performance need careful evaluation and attention to quality control and more attention to assisting, coaching and requiring schools to adopt sound practices and procedures in preventive maintenance and in the use 5] See pages 16-17 of PCR (omission of (i) glass in the windows in CSEs, (l) ceilings in some buildings, (ill) mosquito netting in dorms, for erample, as economy measures). &I For instance, ablution areas built over septic tanks, inadequate roof overhang at the ends of the building, inadequate protection of wals against water penetration from the top. 7 For example, the kitchen compounds correctly include toilets for kitchen stafM however, in one case the audit mission observed that toilets opened directly onto a lobby, the floor of which was used to prepare food for cooking; the design permitted any drainage from the toilet space to flow across that floor. &f While these were not widespread, the problems included holes in concrete Noors (revealing in one case some thinness of the concrete) and rust appearing in some metal roof trusses and window and door frames. - 11 - of the buildings, in order to secure permanently the benefits of the investment. (See para. 4.21 and footnote.) 2.10 The buildings provided for BPE itself, BER and the print shop (RPP) present no special issues, nor does the provision of furniture and equipment for the projects as a whole. Changes in Project Scope During Implementation 2.11 All three projects experienced important changes in scope during implementation. Some changes in technical assistance programs are detailed in Table 3.1 below. Other changes are shown in Table 2.1. In Education I, (a) the extension intended at an existing teacher training institute to provide for the training of practical teachers was changed to the provision of a whole new institute comprising 3210 m2 by comparison with the 290 m2 originally envisaged; (b) premises for BPE were added, comprising not only staff offices but also warehousing and workshops for its construction enterprise, and (c) a large reduction was made in CFPs from 100 to 50 (10 more being provided under Education II with funding from Education I) and (d) instead of lockers at 400 schools, only closets were provided at the project schools. Table 2.1: CHANGES IN PROJECT SCOPE DURING IMPLEMENTATION Items Deleted or Dropped Items Added EDUCATION 1 1. About 50 CFPs & lockers at 400 schools. 1. Teacher Training last. for practical subjects. 2. Extension of existing Teacher Training 2. (..ices, warehousing for SPE. Institute to provide for practical subjects. 3. Specialist for 3 years In production of teaching aide (supplied by other donor) EDUCATION II 1. About 26 CFPs (10 of which theoretically belonged to Education 1). 2. Rehabilitation of 11 secondary schools. 3. Relabursement of UNESCO for Technical Assistance services under Education I. 4. Extension of expert services at two technical schools and addition of two brief fellowships. 5. Additional equipment for CFP.- EDUCATION III 1. Construction of BER offices. 2. Extension of printshopt equipment, fellowships. and audit study for printshop. 3. Material, expert services, fellowships for iSA. 4. Equipment for educational planning. 5. Library books for ETS. ET, ETW 6. Technical assistance for BTS, ETP. 7. Rehabilitation of ETS 8. Rehabilitation of 4 COs and 3 other educational Institutions. 9. Expansion of boarding facilities and libraries at COS and EFlS. 10. Construction of 5 more CFPs. 11. Vehicle natutenane*. 12. Extension of technical assistanco services for BPE. - 12 - 2.12 Under Education II and III, "savings" of US$5 million and 2.7 million, respectively, were applied to carry out what in each case was a long list of urgent needs, such as rehabilitation and maintenance, supplementary supplies, fellowships, extending expert service contracts and adding CFPs. Education II (like Education I) had not addressed urgently needed investment in secondary schools; the "savings" under Education II were used to rehabilitate 11 secondary schools as well as to add 16 primary schools (CFPs)--subsectors that had not been included in the original agreed project.9 Education III included in its extra program (i) library books for technical schools, (ii) four CFPs, (iii) rehabilitation of 10 secondary schools, (iv) expansion of BER offices and RPP failities, and (v) an expanded program of technical assistance (chiefly fellowships). In this way, the projects as completed have a much more "balanced" appearance than the narrow subsector focus of the projects as originally agreed, and this highlights how impractical it was to have ignored really urgent needs at the project design stage. 2.13 The IDA response to the need for change in project scope was generally positive. The area in which the most drastic change occurred and in which that positive response was unwise was in regard to the practical subject program of the CFP under Education I. 2.14 The Borrower changed the CFP program in three key respects: (a) from the concept of two small multi-purpose workshops at each CFP to four large workshops; (b) from the idea of all or some of the existing primary teachers being oriented toward practical subject teaching to the idea of four separate specialists at each school, in addition to the regular complement of six teachers, and (c) the tenfold increase in the size of the practical subject training facilities. IDA gave its implicit approval to the change in the workshops and teacher requirements and its explicit approval to the change in the teacher training component. From the point of view of (i) the scale of the changes (ii) the capital and recurrent costs involved (iii) the logistics of securing teachers, equipment and supplies under the new arrangement and (iv) the significant change in the teaching program that this implied, this was a proposal that should have been reappraised. It was not re- appraised and this proved disastrous (paras. 3.10-3.11; 3.13-3.14). Use of Force Account 2.15 One dominant characteristic of the three projects is the extensive role played by BPE under force account in constructing project buildings; this has developed a large construction enterprise within BPE. 2/ The concept and use of "savings" in this project are both extraordinary because the same technical/vocational institutions of Education II that had been declared completed when these "savings" of USS5 million were identified, were themselves two to three years later, the b-neficiaries of investment of the "savings" from Education III, which was originally not intended to cover technical/vocational education. In other words, during implementation these projects financed originally unintended subsectors. Education I did not originally include technical education; nevertheless, BPE used the resources of this project to gain an early start on the technical schools of Education II, with eventual reimbursement of Education I out of the Education II account. Education II was not designed to touch secondary or primary education. It ended up financing both out of "savings". Education III was not intended for technical/vocational education but did eventually invest in it. - 13 - 2.16 Under Education I, the construction of the 100 CFPs was originally (at project preparation) expected to be done by an external aid group. Later on (at appraisal), the church missions were expected to carry it out. After Board Approval of the Credit, these arrangements fll through and, because of the scattered, remote and not easily accessible sites involved and the difficulty of attracting private contractors at a reasonable price in a country where the local construction industry was weak, IDA agreed to construction by force account. The BPE unit was able to achieve only about one half of the target by the time of final disbursement (April 1982). This was a creditable performance in the circumstances.10 2.17 In presenting arguments to IDA in October 1981 against an award of a contract under Education II to the lowest bidder for two technical schools (and in favor of its award to BPE instead), the Borrower indicated that the bid was actually above the estimate and that its quoted prices wouid rise anyway during construction, because the bidder lacked the financial means to stockpile materials in an inflationary situation. The Borrower further stated that BPE, with seven expatriates on its staff (including four architects and including five persons on four year contracts paid by the project) had too large a staff simply to supervise work done by others 11. In short, BPE required direct construction work in large amounts to justify its large and costly staffing establishment. Such an argument applied with equal force to its large skilled labor force and to its unskilled workers as well. In the transition between two successive projects, the start up of the new was financed in part by the old. In Education II, this was covered by reimbursement of the Education I Credit Account out of the proceeds of the Education II Credit Account. In Education III there was retroactive financing provided to cover this advance expenditure. Such measures helped to avoid retrenchment of the labor force and gain time. Equally, as the unit ended its construction task under Education H and III it promptly sought and received IDA approval to undertake some more work with the "savings" and this also helped to avoid retrenchment. 2.18 Moreover, BPE was not simply concerned with Bank Group financed projects. It executed construction for AfDF financed projects as well and was not averse to undertaking other work. In October 1979, for example, BPE contracted with UNDP Bujumbura to provide a building for FBu 2.1 million. Upon completion of the job, BPE billed UNDP for FBu 4.0 million and in June 1981 justified the cost overrun to UNDP. Not having received payment even a year later, BPE raised the question with an IDA mission. IDA wrote to UNDP New York in June 1982, which in turn wrote to UNDP Bujumbura stating that the World Bank "is quite anxious to receive payment for the work carried out by their [sic] education projects on behalf of UNDP". The UNDP perception of BPE as a Bank unit may have been overdrawn, but it showed how easily IDA could be linked (in the mind of even informed observers) with a construction enterprise. Further, it was never clear whether, in fact, IDA Credit proceeds had been involved in financing this venture, for example by use of materials, vehicles and staff supplied through the Credit. The audit finds it difficult to see how they could not have been involved at least temporarily, even with the greatest care being taken, if the same staff and equipment were used, especially as all that BPE received in advance from UNDP was FBu 0.5 million. In the case of AfDF financed projects, BPE took reasonable care to separate that stock of materials from the IDA financed stock. .LO/ The shortfall occurred for reasons of a late start as BPE mobilized itself, the Tanzania-Uganda war which interrupted the flow of cement and inflated prices, the preoccupation with constructing a new teacher training institute, and the change to a larger CFP (with more workshops) than initially envisaged. iL/ BPE was allowed to do the job and the Credit Agreement was suitably amended to permit force account for that job. - 14 - 2.19 This extensive use of force account, especially in the context of successive overlapping projects, entailed risks of commingling of funds of more than one project as well as risks of BPE self interest influencing or determining the BPE recommendations on the allocation of construction tasks. Interestingly enough, at the same time that Education III was being prepared for Board presentation, IDA was engaged in a separate project to build up local construction capacity. 12 In order to support the aims of this latter project, it was agreed that seven or nine schools under Education III in and around Bujumbura would be let out to private local firms. In the event, only five were so let. 2.20 Because it was contrary to the spirit of the Bank Group Procurement Guidelines, this large use of force account provoked some internal discussion within the Bank Group. In this discussion, the arguments advanced in favor of use of force account were that: (i) BPE was delivering buildings more cheaply than private firms; (ii) BPE was delivering faster than other firms; (iii) there were no local firms large enough to do the jobs; (iv) the private firms that existed would not generally operate on such small jobs in remote areas (where many of the CFPs were located), at least not at a reasonable price. While the views of the private firms were not presented in this debate, the absent ingredient was evidence of an interest or purpose on the part of BPE in solving the critical problems noted as a rationale for use of force account. Furthermore, there was never any reference to reducing the high standards of construction so as to facilitate construction in remote areas. The Borrower has also indicated, in commenting on the draft PPAR, that dispersed sites, poor roads, the limited work in primary schools where the community provides materials, and the rehabilitation of schools during the school year are all constraints that are difficult for the contractors to accept. 2.21 The most important point in the internal debate, however, relates to the validity of the cost comparisons between BPE and private contractors. BPE completion reports always went to some length to show a wide disparity between BPE costs and those of private firms. In these comparisons, however, BPE did not include certain of its own costs and overheads: (a) salaries of technical assistance staff and other BPE "overheads" relevant to the construction enterprise; (b) depreciation of BPE vehicles, equipment, warehouse and office furniture, equipment and buildings used in the construction operations; (c) availability and extensive use of free and ready capital for purchasing construction supplies, materials and equipment and for paying salaries - a situation with certain large advantages: (i) avoidance of interest charges; I/ Local Construction Industry Project (Credit 1230-BU). - 15 - (ii) possibility of large scale stockpiling of materials as a hedge against inflation;13 (iii) the absence of a BPE need for reserves or profits, and (iv) the resources to eliminate the middleman--to undertake tasks ancillary to construction that mipht otherwise have been contracted out--thus reducing the flow of profits to others. In the view of the audit, full cost accounting would show only marginal differences between BPE costs and private contractor costs.15 2.22 The audit does not question, and indeed asserts, that BPE has by and large done an excellent job in a quick and efficient manner and that the staff have been vigilant to prevent individual misappropriation or misuse of resources entrusted to BPE. One is forced to admire the seriousness and competence of this well motivated staff. What the audit questions is whether it is a proper use of Credit proceeds to build up and maintain, on a semi permanent basis, a construction enterprise of this type, with a continuing prior claim on construction jobs financed under the projects (which has the effect of denying private firms a chance to develop). 2.23 It would appear that urgent consideration ought to be given to separating the construction operations of BPE from its project management role, as well as to further options that might preserve and even develop the capacity already built up in the BPE construction service and at the same time help to promote the private sector firms.16 1j The November 1986 supervision mission found that the stock was valued at US$1.1 million. The audit mission received an estimate of US$0.75 million. The total value of equipment, vehicles, buildings and materials in the possession of BPE construction service at any given time and the size of its payroll would probably make it the largest construction service in Burundi. L4/ For example, charges for transporting materials were found to be too high and so BPE promptly bought trucks to handle almost all of its transport needs. With a large fleet of trucks to maintain, it was able to obtain an expatriate fleet manager who was paid out of the technical assistance budget, to build and equip a vehicle maintenance service and to have its own store of fuel It could establish a well equipped workshop to manufacture furniture and another to fabricate metal doors and windows and roof trusses. All of this showed tremendous drive and initiative as well as good business sense, but widened the competitive gap vis-a-vis the private sector. jg The UNESCO team which prepared a draft PCR for Education III estimated that the costs attributed to BPE should be increased by 68% in order for the comparison to be fair. The UNESCO team that prepared the PCR for Education II also found that the comparisons of cost made by BPE unfairly left out some of the overheads of BPE. The Borrower, in commenting on the draft PPAR, has implied willingness to have an independent study of the relative costs of BPS and private sector construction. J/ The Fourth Credit for Education included some provision to separate the construction function from the project implementation role by creating a school construction department and a project management department This was commendable in the sense that it looked forward to a more permanent status for school construction activities; however, (a) it does not appear to have been carried out and (b) it amounts to little more than a new name for an old unit. There are clearly many ways in which to resolve this matter. The audit identifies one possible approach - that, upon separation from the project management function, the BPE construction service might function as a para statal body (with a mixture of force account and competitive bidding) for a transitional period of a few years and then become a fully independent corporation bidding for jobs in education and other spheres of competence. There are other approaches that might be adopted. The Borrower, in commenting on the draft PPAR points out, however, that privatization is not a miraculous solution for all problems. The Regional Office, also commenting on the draft PPAR, urges very careful consideration of a range of possibilities. - 16 - Costs and Disbursements 2.24 Csts. These three projects did not simply change in scope considerably during implementation; in several respects they became commingled; items were financed under one project that belonged to another project. Completely isolating the costs of each project is almost impossible (and to some extent meaningless) since there are overlaps for example in BPE expert services and use of construction materials that are almost impossible to disentangle. The standard DCA covenant requiring the keeping of accounts by the Borrower showing all expenditure on the project separately was generally respected. However, the complete separatio, of expenditure under Education I from that under Education II was never satisfactorily achieved. 2.25 The BPE accounting unit provided the audit mission with the final figures of actual costs of the three projicts. Despite the fact that the BPE estimates of project cost have been used generally by the author. of the PCRs, in all three projects the audit found differences 41ween the PCR costs and the BPE final accounts and, on occasion, problems with the BPE costs." Data given in the three PCRs do not correspond to the final statements in the BPE records in several small but not negligible particulars. Most of these differences in the Second and Third Projects are attributable to the fact that at the time of preparation of the PCRs the accounts had not as yet been finalized to reflect all of the supplementary works and services performed by BPE out of the substantial amount of savings. Consequently the audit provides the latest BPE picture of cost of each project (Tables 2.2, 2.3. & 2.4). This final figure has been carefully assembled by BPE and should be taken as authoritative. 1/ Under Education I a large part of the construction materials needed for the construction of 100 CFPs was purchased. Subsequently it was recognized that only about 50 CFPs would be constructed under Education I and so there would be ecess material that could be used for Education IL On the basis of staff services, materials and equipment procured under Education I, implementation work began on Education II before Education II became effective and before the necessary book transfer could be made. The confusion as to the extent of ependiture under Education I for Education II purposes is evident from the following estimates: September 1980 Supervision Mission: US$350,000 of Education I spent on staff; US$600,000 of Education I spent on facilities, materials and equipment to start Education IL November 1980 Controllers mission: US$403,000 of Educ. I spent on materials since January 1980. June 1981 IDA Staff estimate: From Education I Credit proceed US$300,000 for civil works, $600,000 for construction materials, furniture and equipment and $100,000 for Technical Assistance (BPE stafi) spent on Education II project items. 18 For cuample (a) Education I - the total actual cost given in the PCR Is 1207.5 million FBu; the BPE final accounts figure is 1122.1 million. This discrepancy arose because the PCR understated by 15 million the amount spent on CFPs and double counted a=penditure on project administration (1012 million) thus arceeding the real cost by about 85 million FBu; (b) Education II shows several figures in the PCR which are different from the BPE final accounts, although in this case the differences are not major. Two points need noting - the BPE figures show expenditure for civil works fo BPE buildings but this does not appear la the PCR costs. The BPE figures show the costs of tCe 10 Education I CFPs under the regular costs (where they belong) rather than, as in the PCR, under the use of 'avings"; (c) Education IIL the costs are not available in any of the Completion Reports in US dollar terms. The BPE tables contain discrepancies that the PCR does not address. For example, the CFPs are given as costing 341 m. FBu in one BPE table and 286 m. in another BPE table. The PCR incorporates both tables without clarification. - 17 - EDUCATION I Table 2.2: PROJECT COSTS (US$ M equivalent) Appraisal Actual Costs Difference (Z) By Activity Estimates /a CFPs 9.434 7.37 -21.9 BER (excl. TA) 0.759 0.62 -18.3 Teacher Training 0.134 1.80 1243 Institute BPE Furniture/Equipment 0.040 0.37 825 TA BER 1.239 TA BPE 0.584 TA Total 1.823 1.16Lb -36.4 Project Administration 0.321 1.14 255 Total 12.511 12.46 -0.4 By Category Civil Works 8.410) 10.10c -2.0 Furniture/Equipment 1.957) Technical Assistance 1.823 1.16 -36.3 Project Administration .321 1.14 255 Total 12.511 12.46 -0.4 /a Actual costs here are taken not from the PCR (whose figures differ somewhat from these) but from the BPE final accounts. lb Ir the BPE PCR the total given is $1.153 equivalent (34.8 m. F3u for BPE and US$0.765 m. for BER); however, BPE final accounts show changes in both elements (57.2 m. and 47.1 m. FBu respectively for BER and BPE) giving a total of $1.16 m. equivalent. /c The audit has not been able to break down the total actual costs of physical plant into buildings, furniture and equipment. This figure covers all three items. - 18 - Table 2.3: PROJECT COSTS (US$ 000 equivalent) Appraisal Actual Costs Difference patimatjee /a M% Technical/Vocational Schoole 16,065 10,047 -37.5 Studies 212 96 -54.7 LoSistical Support MNE 373 241 -35.4 Project Unit 1,350 2.668 +97.6 18,000 13,052 -27.5 Civil Works 11,061 5,839 -47.2 Furniture 969 468 -51.7 Instructional Equip- ment Vehicles 3,753 3,280 -12.6 Project Administration 458 1,464 219.7 Technical Assistance 1,587 1,899 19.7 Professional Fees ___172 102 -40.7 18,000 13,052 -27.5 /a For this table the PCR figures are used as they refer to the original project content. They correspond roughly (not exactly) to the final accounts of BPE. Additional coste were incurred to utilize the 0savinge"I these, shown below, bring the total to US$18,000,000. US$ million Completion of (10) Education I CFPs 0.964 16 Education II CYPs 1.985 Secondary Schools Maintenance 0.889 Reimbursing UNESCO for Education I task 0.227 Extension of technical assistance 0.240 Additional maintenance for schools 0.514 Additional equipment for CFPs ,l,29 4,*948 - 19 - EDUCATION III Table 2.4: PROJECT COSTS (US$ N equivalent) Appraisal Actual Costs Difference By Component Estimates /a (.) Primary Schools 4.80 2.34 -51.2 Lower Secondary Schools 2.37 2.35 -0.8 Teacher Training Inst. 3.18 2.89 -9.1 Adult Education 3.04 1.95 -35.9 Textbooks 2.45Lb 1.98 -19.2 Project Unit 3.43Lb 5.10 48.7 Total 19.27 16.61 -13.8 By Category Civil Works 10.781 Furniture 1.24) 14.17 -5.5 Equipment 2.98) Professional Services 0.09) Fellowships 0.09 0.12 33.3 Experts 2.82 3.08 9.2 Project Administration 1.27 1.16 -8.7 Total 19.27 18.53 -3.8 /a The PCR does not give actual costs in US$; these appear only in SDR and FBu. These figures have been arrived at by the audit by using an average exchange rate of 121.9 fBu to the US$. The actual cost by component in this table is the actual cost of the original components. The actual cost by category is the cost of the expanded project using the msavingeb for additional items. b Included in this total is the cost of expert services for school mapping, textbooks, teacher training, non-formal education, educational planning and project administration. The SAR does not provide sufficient information to permit the costs to be disaggregated. - 20 - 2.26 This suggests that much more attention needs to be paid to the keeping of accounts by project accounting staff, with appropriate monitoring of this by supervision missions, so as to ensure (a) due separation of accounts in the case of multiple projects being implemented by the same unit and (b) clarity and transparency in the attribution of costs to various components and categories. Similar care needs to be exercised by completion reporting staff, costs being an essential ingredient in the evaluation process. 2.27 Appraisal Cost Estimating. The cost of the average CFP was estimated at the appraisal of Education I at US$94,000 including contingencies. The actual cost was US$139,000 equivalent." While part of this difference can be explained by larger construction areas (more elaborate workshops), the rise in the price of cement and the change from the original intention that a non- profit agency would build the schools, the appraisal estimate appears to have been too low. 2.28 Under Education II, "savings" of almost US$5.0 million or 28% of the original estimated project cost were realized. These savings were attributed by IDA and BPE to the timely and efficient performance of the construction task by BPE. This conclusion is only partly valid. There was again a significant cost underrun of 14% under Education III, even though the capacity of the BPE construction service had been well established and known at the time of appraisal. In the view of the audit, by contrast with the underestimation of costs under Education I, there was, in the light of the known capacity of the BPE construction service, some overestimation of costs for Education II and III. 2.29 It is important, for the purposes of any study that attempts comparability of costs, to recognize that a large part of the costs shown under project administration in these three projects belongs properly under civil works, since it involves salaries for architects, engineers, site supervisors and other items used directly in the civil works program (see para. 2.21). 2.30 The most notable and consistent change in cost (between appraisal estimates and actual cost) over the three projects was the increase in project administration costs (in US$ million equivalent): Appraisal Actual Estimate Cost Difference (%) Education I 0.95 1.73 83 Education II 1.35 2.67 98 Education HI 3.43 5.10 49 5.73 9.50 66 A significant part of this increase was for extended use of expert services, (despite the absence of any major time overrun on any project), and for unforeseen construction of BPE premises. 19 Here the overall cost taken is the average for the frst 60 CFPs constructed under Education I and Education II. If the cost of BPE technical assistance and overheads related to construction are Included, the actual costs would be much higher (see para. 2.21). The later CFPs were less costly, being generally smaller. - 21 - 2.31 Disburseans. All Credits were fully disbursed, with the exception of a small amount (US$95) of the proceeds of Credit 1358-BU (Education III) which was canceled. With the concurrence of two projects in implementation most of the time, there were occasional disbursement problems. Because of the use of Education I construction materials and expert services to achieve an early start on implementing Education II, it was necessary for some US$1 million of the proceeds of Education II to be transferred to meet the shortage of funds that developed in Education I and an appropriate amendment was made in the DCA for Education H. Another unusual measure was the use of US$0.227 million if the proceeds of Credit 976-BU (Education II) to pay off the balance owing to UNESCO under its contrat signed with the Borrower under Education I, after the Closing of the Education I Credit Account, without amendment of the DCA. Under Education II, there was a case of double dthbursement for the same set of cement which was rectified by requiring the Borrower to buy an equivalent value of cement entirely out of its own resources for the project. Otherwise, disbursements were handled in a careful and expeditious manner, despite the unusuai extent of purchase of construction materials, trucks and so on. W. PROJECT OUTCOMES 3.1 The projects were designed to: (a) expand the size of the school system at various levels--primary, lower secondary, primary teacher training, technical/vocational and nonformal; (b) improve the quality of education and (c) help to develop the human resources of the country. While much was achieved, there were shortfalls in every aspect. Expansion 3.2 The projects succeeded in expanding the system. Nevertheless, the expansion in primary and lower secondary education was dwarfed by the rapid growth of these subsectors that was financed from other sources. The modest expansion of primary teacher training and technical/vocational programs was achieved. In the practical subject teacher training program, however, the project investment is yielding no benefit for the time being, as the school has been closed down. In nonformal education, the places were provided, but the enrollments never materialized to the expected extent and the programs have been effectively suspended. 3.3 Primary Education. The project provided 105 CFPs approximately - 28,350 places (Education I, 60 - 16,200 places; Education II, 16 - 4,320 places, and Education III, 29 - 7,830 places). Education I was initially expected to provide 100 CEP (27,000 places) which would have represented a 21% increase over the (130,000) places existing in 1974-75. The quantitati, j impact of the projects, set in the context of national evolution of enrollment, has been as follows: - 22 - 1976-77 1983-84 1988-89 Actual Growth 1976-89 National Enrollment20 131,338 302,611 570,438 439,100 Project contribution of places 1 16,200 28,350 28,350 Project school places as % of 9% 5% 6.5% national enrollment and growth Even if the 27,000 places had been provided by 1983 as originally scheduled, their impact would have been muted in a system which had already grown by 130%, largely through Government cooperation with the communities.21 In one sense, the actual impact of the project has been greater than appears from these figures - most of the schools have been placed in remote areas, which had been undersupplied or not supplied with places previously, or they have replaced unsatisfactory facilities. They have given a facelift to the system that perhaps is as important as the modest numerical increase that they have accommodated.22 3.4 Secondary General Schools. Education III included four lower secondary schools (COs) and two EFIs, each with a CO section, each CO (or CO section) to provide 320 places. During the period of project generation, the Government was seriously considering ways of avoiding the financial burden of boarding students, but recognized that this would require the rapid establishment of a network of small widely dispersed COs, each drawing students from a limited geographical area within which access to the school on a daily basis was reasonably easy. IDA staff, anxious to bring about an end to boarding, sought to minimize the amount in each school and requested a school location study to be done early--at first, before appraisal, and later, before Credit Negotiations. The study was carried out quickly by a Canadian firm and was based on two assumptions which eventually proved wrong. The first was that the secondary system would grow rapidly so that a promotion rate of 30% from primary to secondary would be achieved. At project completion this rate was only 11%. The second assumption was that the Government would adopt a regional examination (to replace the traditional national examination) to select students for secondary education - in effect, zoning students - which is what IDA had been urging it to do. The national exanY ation, which has public confidence, has endured. The project included some limited boarding. However, this proved to be quite inadequate since not only were the schools heavily overcrowded in response to the @f There is no available information oa the actual number of places in the system. The double shift system was introduced in 1982 and this has complicated the picture. In this section the enrollment is taken to reflect the national demand for places. 21/ The Regional Office, in commenting on the draft Report, expressed the view that this kind of comparison with the overall growth in the system (both in this paragraph and in para. 3.6 below) is unfair. The audit maintains the comparison as it formAd part of the appraisal goals. g Two CFPs visited by the audit mission had enrollments of 519 and 487 (41% femre in each school - slightly below the national aveuage off 44%). P3 Some 80 places for each CO (25% of enrollment) and 160 for the EFIs (50% of enrollment. During implementation, IDA and Borrower agreed to provide 50% boarding at certain COs. - 23 - overwhelming demand for places, but also the proportion needing boarding was much higher than IDA had estimated originally.24 3.5 It is probable that IDA did not fully appreciate the real practical possibilities of effecting change in the school system. For instance, in 1982 when Education III was being generated, IDA proposed to the Government that it change to a district examination as a condition of support for the inclusion of COs. Such a change would have been feasible to consider only after a sufficient network of COs was in place. Otherwise, public concern would have developed over how equitably (whether on the basis of locality or on the basis of ability) the scarce places were being awarded. Another example of unrealistic conclusions can be seen in the aide-memoire of the January 1986 supervision mission which proposed that, if the Government issued a ministerial decree (i) prohibiting the boarding of repeaters and boarding in urban secondary schools and (ii) mandating an annual rise in the proportion of day students in secondary school enrollment by 10 percentage points each year, IDA would agree to the provision of more boarding facilities at secondary schools (which could later be converted into classrooms). Such a decree would have been a dead letter from the start; change of this kind is brought about by careful planning, detailed programming and physical arrangements; it is not brought about either by a decree alone or with such speed. 3.6 The national evolution in secondary general enrollment (Grades 7-12) was as follows: 1976-77 (time of approval of Credit 679) - 6,663 students. 1982-83 (time of approval of Credit 1358) - 8,487 students. 1988-89 (completion of Credit 1358) - 20,108 students Education III provided 2,240 places (1,280 in four COs and 960 in two EFI/COs), representing about 19% of the national increase during project implementation. The appraisal mission expected that the secondary enrollment ratio would remain low (3.4%); it has, in fact, far exceeded these expectations and by 1988-89 was about double that figure. Hence, as in primary education, the quantitative impact of the project was much more modest than originally envisaged. 3.7 Technical Vocational Education. Education II expanded opportunities considerably in technical/vocational education. Two new schools were provided and were fully enrolled at the time of the audit visit: ETMM Bubanza - 393 students (with 380 capacity) ETP Gitega - 345 students (with 330 capacity) 3.8 The secretarial and accounting sections of ESTA and the Mechanics Department of ETS were also fully subscribed. Equipment is generally satisfactory (of a very high standard in the 24 Amo,,g the schools visited by the audit mission for example, the CO at Bubanza which was originally intended for 320 (with 80 or 25% boarding) had 492 students with 449 (or 91%) boarding; the EFI/CO at Cibitoke which was originally designed for 480 (with 240 boarding) had 658 students with 640 (97%) of them boarding. Both schools adopted drastic measures to cope with the excess enrllment and boarding, converting classrooms into dormitories and, in the case of Cibitoke, appropriating the nearby CSE classrooms for teaching, the library for staff rooms and the pedagogical bureau intended for the practice teaching activities for dormitories, and even unilateral Government provision of boarding facilities. - 24 - Mechanics Department of BTS). The audit mission observed a shortage of chairs and an almost total absence of library materials in ESTA. 3.9 The agreed specializations have been established at the technical/vocational schools. An evaluation of the effectiveness of the training in the context of the labor market will need to be done at some future date. The plumbers graduating from the ETMM--the first 25 were produced in 1990- are the first of that discipline produced in Burundi. The overall output of these institutions will help to reduce dependence on expatriates. It is encouraging to note that female students have enrolled in even the most traditional male-dominated technical fields, although in the industrial fields their numbers are still quite low.25 ETMM has constructed several buildings on its campus including a workshop to take in jobs to earn revenues for the school 3.10 Non formal Education. The provision of CFPs (with workshops) in the three projects was intended in part to meet the needs of adults (as well as out-of-school children and youth) for literacy and basic skill training. In this respect, the Social Centers or CSEs, which came under the Ministry of Social Affairs, were only marginally different. Because of the discovery by IDA, during the generation of Education III, that the workshops of existing CFPs were grossly underutilized, the Education III project excluded the provision of workshops in 22 of the 24 planned CFPs so that the new CFPs were more like regular primary schools. The 20 CSEs were provided and equipment procured. The audit mission found that uh of that equipment had been removed from the CSEs because their program is now suspended. Prior to the suspension of programs, the enrollment in the CSEs was very low - well below the expected average of 120 students per center which was compared at appraisal to an existing 90. In fact, the average of 52 at project completion given in the PCR27 appears to be wrong. In September 1988, according to a UNICEF survey o. CSEs, there was an average of 34 in 24 centers (including an average of 37 in four project CSEs) and this overall average declined to 27 by April 1989. In regard to expanding nonformal education, therefore, the project was a failure. 3.11 For the practical subject teacher training, the CFMP provided under Education I was closed at the time of the audit visit while the fate of the nonformal education program is being determined. 3.12 It was during the field appraisal of Education III that the Government presented to the mission the proposal for CSEs (then termed "foyers sociaux" or social centers). These were included in the project. In every technical sense, the appraisal of this item was inadequate and it is quite clear that its preparation was also inadequate. In the sense that nonformal activities require considerable PJ At ETP, for aample, there were 17 female and 328 male students on rol at the time of the audit visit. Of the female students, six were in the Gst year common course, one in surveying, four in the clerk of works (conducteur des travaux) course, none (surprisingly) in drafting (which usually attracts females) and six ln rural works (genie rurale). _W The Ministry of Social Affairs was abolished in February 1991 and the Ministry of Youth was given responsibility for nonformal education The entire program involving 140 CSEs (half of them private) and 112 CFPs - several of them without nonformal students -was reduced in May. The Ministry of Youth is now drawing up plans to focus on about 20 former CFPs and 20 former CSEs as pilot centers, all of which are expected to have been those provided under the three projects. At the time of the audit mission those pilot centers were being selected and no one appeared to know precisely where the equipment was stored. 7 Para. 15 of the PCR. - 25 - preparatory and exploratory work,28 it is therefore surprising that (especially in the light of the unfavorable experience of UNICEF with these centers in Burundi that was drawn to the attention of the appraisal team) the mission did not focus on the institutional strengthening of the department responsible for this activity and on an evaluation of experience in this subsector and limit any direct teaching to a small pilot project of about four centers in strategic parts of the country. In the event, 20 CSEs were included and these encountered difficulties (such as public apathy, poor staffing and indifferent results) to such an extent that the program has been suspended. 3.13 In this respect, the appraisal of the practical subject program of the CFPs was also unsatisfactory. The formal regular program of the primary schools was in the process of being reformed. The major challenge was the practical subjects and there were no schools adequately equipped for this. The rest of the program (the non-practical side) was already being introduced up to the first four grades. There was need for suitable facilities and teachers to proceed with the practical subjects in Grades 5 and 6 and the project aimed at providing these facilities for both the formal and the nonformal programs. As the experience showed, the appraisal of this component was deficient in accepting it for implementation on such a nation-wide scale without prior trial, and without some determination of interest and acceptance on the part of the students and communities, agreement on the precise program to be followed (which would have prescribed the teaching spaces and kind of teacher needed and the teacher training program) and identification of the necessary supervisory support. 3.14 Nonformal education and training is acknowledged to be a difficult area and one with which the Bank Group has not had much success especially in Africa. A review of the Bank Group operations in this area was not published until 1987. The main reason for failure was the lack of expertise among Bank Group staff. In the generation of the Burundi projects e is struck by the absence of nonformal education specialists in the composition of missions . A very senior Government official informed the audit mission that the failure in this area--the precipitate rush into providing the infrastructure for this nonformal program for which the necessary preparatory work had not been done--hurt the image of the Bank Group in the country. Improvement In Quality 3.15 Primary. The CFP program had two distinct facets: (i) the formal reformed primary curriculum which was being made more relevant to the needs of the rural areas and being given a more practical bent and (ii) the non formal program of literacy and practical skills (which is discussed below under non-formal programs). &/ Such tasks as the identification of target groups (beneficiaries), involvement of the beneficiaries in determining their educational and training needs and the teaching methods that best suit their circumstances, the design of the programs and related materials, the training of the instructors or leaders and the design of arrangements to help stimulate increases in earning capacity that should go along with many of the learning activities and to ensure involvement of the community in managing the programs and mobilizing community participation in the classes. ;J A community development specialist from UNESCO was on the project preparation mission for Education III; however, it was at the appraisal stage that the proposal for CSEs was presented. - 26 - 3.16 The implementation of the formal program has proceeded with the help of new textbooks and syllabi and the use of the mother tongue (Kirundi) as the language of instruction. In general, this has gone well under the leadership of BER. Some problems have been observed, such as a weakness in French, especially in students who proceed to the secondary level. In regard to the practical subjects, however, there were problems arising from the shortage of competent teachers, the shortage of instructional supplies and the double shift system (introduced in 1982 at the behest of IDA) which reduced the length of the school day by 25% and affected practical subjects most. The survival of the practical content of the curriculum will depend upon the enthusiasm of the principal and staff of the individual CFP. The audit mission saw some cases of apathy and very few cases of enthusiasm about practical activities among the schools visited. Enthusiasm is mainly in gardening, where some income is generated, and in sewing. 3.17 In the view of the audit, the reform of the formal program that was supported by these projects was too ambitious in scope, chiefly in regard to the practical subjects. In a system where there are serious problems of quality and quantity, weak physical infrastructure and major financial and human resource limitations, such multi-faceted goals requiring vast changes in attitude and establishment of new skills depend heavily for their achievement upon thorough planning and programming. This massive scale of change was attempted in other African countries in the 1970s especially with UNESCO assistance, and the results have been almost uniformly disappointing.3 The responsibility of IDA, with its international experience, was to have warned against so wide a range of objectives at the start and to have proposed concentration on basic primary skills. 1 3.18 The lack of the necessary planning, on which good IDA/Borrower agreement on targets and objectives should have been based, was nowhere better shown than in the practical subject teacher training item. One year after Credit Signature, IDA became aware that the Government wished to construct a new practical subject teacher training center, to have four practical teachers (in addition to the regular six primary teachers) and to provide four separate (rather than one or two multipurpose) workshops at each CFP. IDA agreed to the replacement of the planned modest practical subject training facilities by a new teacher training institution, without any real evaluation of the proposal. This new center from the outset trained practical teachers in a one year course and, by 1984, had produced some 229 such teachers, along with having provided short courses for headteachers and school inspecto.s. From 1984, when the policy of not constructing workshops at CFPs was proposed by IDA and accepted by the Government, the center operated like a regular primary teacher training institute and relegated its practical subject training to an upgrading in service course of one week per year. In 1991, the center was closed along with the suspension of the CSEs. 3.19 Secondary LeveL The EFIs provided under Education III have been transformed into upper cycle secondary schools with a pedagogical stream. The PCR (para.17) criticizes the Government decision to allow graduates of the pedagogical streams to enter University after serving 29 For acample in Sierra Leone (Bunumbu project) and UIberia (Kakata project). L/ A good crample of a Bank Group approach (in 1975) which was much more restrained was the econd Education Project in Liberia where a major reform-the comm aity schools program-was being developed by the Government and UNESCO. Originally the country apected about as much from the proposed reform as Burundi did in this case - new rural curricula, practical courses for children, adult use of workshops etc. The Bank Group response (and advice) was to (i) focus project objectives on the basic aspects of the primary curriculum. (i) avoid the extraneous and inappropriate elements of skill training, (ii) omit workshops, (v) separate adult education totally from the management of the primary programs and from the responsibilities of the primary school and (v) provide for strengthening the management of adult education. - 27 - as teachers for some time. IDA staff consider that this conveys the idea to the teachers that teaching is not a worthwhile career, and states that the attractiveness of the profession should be secured through improved conditions of service. The audit agrees with the PCR that securing persons to teach in primary school should be done through attractive conditions of service, bvt considers, however, that it is a positive measure to open University doors in this way, creating opportunities for individual growth. 3.20 The programs in the lower secondary school posed no major qualitative questions. Nevertheless, the underutilization of the science laboratories (and, in one of the schools visited by the audit mission, their non-use32) and the conversion of libraries to other purposes (a dormitory in one school and staff rooms in another) suggest that there is need to (a) strengthen supervision of the schools to provide the leadership in the use and maintenance of rooms and equipment and (b) relieve the overcrowding. The technical/vocational schools appear to be operating with less serious constraints. 3.21 Technical Education. The agreement between IDA and the Borrower was that a tracer system and an evaluation program for the technical/vocational schools would be organized and that employers would be involved in the organization and operation of the technical/vocational subsector. These measures were intended to promote greater relevance of the schools to the needs of the labor market. The school board of ESTA was established as early as 1982 and is functioning with employer representation. ETP instituted a tracer system by 1985. BEET itself was very slow to establish the (national) advisory committee for technical education. Thus these undertakings have only partly been met by the Borrower. Supervision missions noted that private employers were unwilling to cooperate (and were even opposed to cooperating) with the schools, since they saw no benefit to themselves in this. Employer tendency to look askance at formal technical/vocational education and training is quite commonplace in the developing countries--employers used to hiring semi-skilled workers at the gate or importing their skilled and even semi-skilled workers tend to see local formal training as irrelevant to their needs and as producing persons with unrealistic expectations and little skill. The failures on the institutional side suggest that more project support was needed for strengthening the management of the subsector (particularly BEET) to enable it to undertake the necessary groundwork and to provide the necessary leadership to employers and the schools in such innovations. Employers needed to be persuaded and incentives designed to encourage their participation and schools needed to be guided on the follow-up of graduates and on the analysis and assimilation of information derived from such studies. This would have required a well trained and mobile technical/vocational supervisory service. 3.22 Textbooks. Education I and III provided assistance to textbook production. Education I expanded and equipped the printshop of BER, provided 115 tons of paper (not 250 tons as intended) and, according to the BCR of March 1984, five vehicles (not six as intended) - two trucks for distribution of teaching materials and two station wagons and a car for use in staff training. The equipment supplied for the printshop was not entirely satisfactory. The goal was to produce about f At Cibitoke which is a combined CO/EFl Institution (Grades 7 - 13) while two science classes were in progress In regular classrooms at the time of the audit visit, the science lab was empty. There was no seating in the laboratory space provided, water was not available, the store room and preparation room had practically no equipment and the entire lab showed no sign of having been used in months. - 28 - 680,000 textbooks.33 Some 260,000 books were produced, 190,000 of which had been distributed by 1987. Education III profited from the experience gained under the previous investment and provided for RPP an expansion of the printshop, a more satisfactory set of equipment, 250 tons of paper and eight vehicles (six for book distribution and two for errands). The unit is functioning more smoothly and 61 titles have been produced. 3.23 The principal issue has been the recurrent funding of textbook production. Funding has depended upon (i) project contributions of material and printing supplies (chiefly paper), (ii) the proceeds of a book fee and (iii) Government subventions. In both the fees and the subventions there have been shortfalls. In some areas of the country, the children avoid school and so the fee; generally the whole fee is paid late and the recent increase in the fee might further stimulate evasion of payment.34 The Government included funding for RPP regularly in the budget up to 1986. However, it was in the habit of not paying the final 25% of the amount. For the four years 1984-87, the total operating budget of RPP was 175.5 million FBu. The book fee realized 113.7 million FBu and the Government subventions only 36.7 million, leaving a deficit of 25.1 million FBu. It is evident from this difficult situation that this project component required much more preparatory work on the financing arrangements than was devoted to it. Human Resource and Iastitutional Development in the Sector 3.24 There were two important aspects to human resource development within the education sector under these projects - the building of expertise and competence in specific fields (training) and the enhancement of the capacity of units to perform their assigned tasks efficiently (institutional development). Technical assistance was the principal means of achieving these objectives, with the occasional support of logistical or other assistance (vehicles, office equipment, operating expenses). 3.25 While the Borrower interest in technical assistance was in a mix of expert services, local training and fellowships, there was perhaps greater faith in fellowships to achieve individual growth. On the other hand, IDA manifested far more interest in experts (to get the job done) than in training. In the generation of Education II, for example, the Bank Group/UNESCO project identification report suggested 28 years of experts and 15 years of fellowships; the Credit Application requested 18 years of experts and 135 years of fellowships; the approved Credit included 17 years of experts and nothing for fellowships. 3.26 Table 3.1 sets out a summary of what was originally agreed in each project in terms of expert services and fellowships and what was actually implemented. 3f The estimate was 170,000 textbooks and 22,200 teacher guides per year, rising to 215,000 and 2,700 respectively in air/seven years' time. I 'lThe fee covers three elements: () books, (il) insurance and (ill) school supplies. Up to 1991 the fee was the equivalent of US$1.00 divided among the three purposes in the ratio 50*15:35; this was raised in 1991 to the equivalent of US$1.50 divided in the ratio 75:25:50. This 50% increase in the book fee should be helpful. - 29 - Table 3.1: TECHVICAL ASSISTANCI IN BURUNDI EDUCATION PROJBCTS I, II, III Initially Approved as Actually Implemented Comment in DCA Experts Fellovehips Experts Fellowships Education I Cr 679 BER Number 5 10 .5 81A *Instead of Months 180 90 118 84 accountant and BPE Number 2 7* procurement MontheL 9 _ 135 specialist, there Total Months 276 90 253 84 were 3 site supervisors, 2 architects, 1 engineer and I accountant. Education II Cr 976 Technical Schools Number 2 3 Months 84 98 Studies Number 2 2 Months 12 12 BPE Number 4 8 Monthsh 108 189* *architects 6.75 Total Months 204 299 years, site supervisors 3.08, accountant 2.75, procurement 3.08. Education III Cr 1358 ETP Number - 1 Months 4 MSA Number 1 or 2 - 2 3 Months 24 68 3 BER Number 2 - 2 9 Months 48 42 26 RPP Number 1 4 1 7 Months 36 36 60 33 EPU Number not known - Seminars organized Months 12 by IIEP in addition. ITS Number - - 1 3 months 48 10 BPE Number 6 - 7 - *Actual figure MontheT 228 M* obtained Total Monthe 348 36 504 76 from UNESCO draft I __I___PCR. /a It is difficult to find out exactly what happened regarding the number of fellows. The BCR gives seven names and then adds an eighth. At the end of the project implementation period an IDA supervision mission report included an attached list of achievements under technical assistance for this project which gave the seven names listed later in the BCR, but not the eighth, and three other names as well (a total of 10 names). The audit mission was informed that the correct list is the list of 8 in the BCR. Of these 8, one (not 8 as the PCR para. 4.20 states) went to study educational planning at IIEP, Paris. He returned to the educational planning unit and later vent to- UNDP Bujumbura. Seven were trained in Belgium (1) and France (6) where they covered various fields relating to curriculum development and supervision. Of these 7, all served in BER upon their return and three have since been promoted - one to the Inspection Generale of Education, two to the Ministry of Artisanal Training - one in charge of an artisanal center, and one to be director of Enseignoment de Mdtier. /b The total of months actually implemented in BPI is 610 in this table (which drawa information from the various PCRs. However, a detailed list of technical assistance experts paid under the Credits for the three projects gives their total period of tenure as 794 months. Thus the PCRs have understated the actual amount. - 30 - 3.27 As the Table 3.1 shows, increases occurred in expert services during project implementation, largely on account of BPE because of (a) the expansion of tho BPE role in construction and (b) the expansion of the construction programs themselves. Some of the savings in Education III were used to finance several fellowships, so that the actual use far exceeded the original project provision both in numbers of fellowships and in total staff months. Part of the reason for the changes during implementation was that there had been very limited consensus, if any, in advance between IDA and Borrower on the details of the technical assistance components. As early as March 1978, for example, following Credit Signature for Education I in April 1977, the Government sought to change the inservice teacher training adviser and community development posts into education planning and preservice training posts. Only one month after Credit approval, that is in July 1983, the Government requested significant chanies in the technical assistance program of Education III, partly so as to increase its training element.'5 3.28 In the case of each unit or activity, the efforts at institutional strengthening had their own particular strategy or mix of investments and atter >nt success or failure. These units or activities were (a) BER, (b) Textbook Production, (c) Technical Education, (d) Educational Planning and (e) Project Management. 3.29 (a) BER. Under Education I, BER was scheduled to receive 15 years of expert services in textbook production, teacher training, rural community development, itinerant pedagogical advisory services and evaluation, principally to help with the training of practical subject teachers (three experts), to assist with teaching aids and syllabus preparation (one expert) and to build up evaluation and monitoring of the reform (one expert). In addition, nine BER fellows were to be sent to study similar reforms in African countr.es and vehicles were to be supplied to help not only with distributing books but also with BER field work. The SAR specifically stated that building permanent capacity was an objective. This helps to explain the fellowships, since the expert services items clearly had the overwhelming aim of guaranteeing project implementation rather than the training of staff. Education III also originally included two primary teacher training experts but no fellowships. 3.30 The expert services component of Education I was implemented with minor changes.36 These expert services, executed by UNESCO, were deemed "less successful in supporting BER"37 although local staff were trained in evaluation. Under Education III, the experts performed their task and trained staff. 3.31 The fellowship program of Education I, intended for study tours in Africa, was modified during implementation to provide instead for training in France and Belgium in curriculum design and supervision for seven staff members. These all returned to BER and four were still there some seven & Instead of 10 years of experts and US$75,000 for fellowships as agreed, the Government wanted 9 years of experts and US$82,400 for fellowships. Furthermore, a change in the experts posts was requested from one on Education of rural women and one on teacher training to one covering curriculum development and inservice teacher training, and one on preservice teacher training along with a consultant to train MSA staff. & Three experts paid by the project served BER, for 76 months rather than five for 180 months. Two others (a printshop expert and a teaching aids production expert) were financed through a parallel UNESCO program. For the community development post, a nonformal education specialist was substituted and two teacher trainers (one in-service specialist and one pre-service specialist) were utilized. 7 PCR for Education I, para. 4.21. - 31 - years later at the time of the audit visit, one as Director. The three who ld left BER, after some post-fellowship service, were in senior Government positions elsewhere." On the basis of the "savings" in Education III, during implementation, seven BER staff were sent on short study tours in African countries and two went on nine month training programs to Europe. These are all serving in BER, except one who has become Director General of Basic Education. The study tours of African countries came at an appropriate time--that is after the staff had been trained and had acquired some post training experience. These modifications to Education I and III represented important improvements over the initial project provisions; they have helped to provide a core of competent staff in BER and have had spill over benefits for other sections of the education and training system. PER is now a competent, sustainable service. 3.32 (b) Textbook Production. The technical assistance support for textbook production under Education I was purely for expert services which were actually furnished instead by the separate UNESCO program noted above (para 3.30). Vehicles were supplied for book distribution. Under Education III, the textbook production unit (com.prising manufacture, distribution and accounting services) separated from BER and renamed RPP, was allocated funding for an expatriate printshop expert (as adviser to thc Director of RPP) and four nine-month fellowships. 3.33 Education I was deficient in that it did not provide specifically for training senior production management staff to take over from the experts. The Education III fellowship program, which was intended to remedy this somewhat, was executed by UNESCO which, in turn, left the placement of the fellows to the French National Commission for UNESCO. Seven (not four) persons were placed in France, three of them in attachments that proved somewhat irrelevant to the needs of RPP; a fourth (a key one) returned to live in France soon after her training. Thus, while the impact of this program was very limited, the wisdom of increasing the number of persons was demonstrated. While RPP was functioning reasonabb well at the end of implementation of Education III, it was still dependent upon external aid. However, as of 30 June, 1991, the last expatriate, paid under Ecucation IV, ended his term of service in RPP. A more purposeful and comprehensive training program from the outset (Education I) and more careful placement of trainees would clearly have helped to achieve self sufficiency earlier. In hindsight, it appears that this component lacked the aim of building a permanent self sustaining capacity at the project design stage of Education I and was devoted principally to guaranteeing the achievement of production targets. It was the expansion of the fellowship program under Education III that set things in motion towards serious local capacity development. 3.34 (c) Technical Education. Education II included an expert to serve in each of two project technical schools.4u It did not specifically include training. The experts' task was to organize and implement the technical/vocational training programs. The two experts carried out their duties and they were subsequently extended under a modification of Education III. One trained a counterpart, the other did not. This modification also provided two brief fellowships for staff of the schools and _W See footnote, Table 3.1. L9 Education IV included three years of expert services and two fellowships for the unit. IP/ On was the ETP and the other was the Mechanical Department of ETS. Each was known as "chef de travaux", responsible for organizing and implementing the training programs. - 32 - one of the ex-fellows has since been dismissed. Educatior H also provided office equipment and vehicles for BEET - modest logistical support that officials appreciated but considered insufficient. 3.35 In this project the emphasis was on technical assistance experts to execute the project. That there were weaknesses in BEET and the sub-sector was quite clear to IDA; indeed IDA included a study of technical education in the project. The shortage of Burundian technical teachers was also evident to IDA, but the fellowships proposed by the Government to train them were rejected by IDA on the ground that there were enough returning engineers. An Executive Board member rightly enquired, during Board consideration of the proposal for this Credit, whether salaries were such as ta attract them to teaching. The audit points out that these returning engineers were not persons prepared for technical school teaching and that some sort of pedagogical training program would have been necessary anyway. What was evidently needed for capacity bilding in this subsector therefore was: (i) administrative strengthening (particularly staff training) for BEET to enable it to provide leadership in developing the subsector (para. 3.21); (ii) pedagogical support for the schools (materials, training for teachers and supervision) and (iii) a staff development program for technical staff, including fellowships to prepare senior teachers, directors of schools and central administrative and technical specialists. The fact is that the focus in this project was not on building national capacity, no real institutional development occurred and dependence upon expatriate assistance in the subsector (including teaching) remains strong.41 3.36 (d) Educational Planning. Support for educAtional planning in these projects was very limited, although throughout the generation and implementation of these projects IDA harbored serious reservations about national development strategy. The total of support for planning in the thret. projects as designed was: (i) one fellowship to HEP Paris for nine months (Education I); (ii) two six month consultancies-one for a study of basic education and the other to study technical education (Education H); (iii) logistical support (vehicles and equipment) to the planning unit (Education III), and 1J/ The Burundian teachene in technicalocational education increased from 134 in 1979-80, when Education II was signed, to 248 in 1985-86. Over that same period the foreign teachera in the subsector increaed from 105 (or 44% of total staf) to 125 (or 34% of total). In 1989-90 foreignera comprised 35% of the teaching staff of the schools (123 foreign and 234 national). Thus there has been a recent dedine in the number of Burundian teachers. - 33 - (iv) consultancies in evaluation and planning to help with national school mapping and a special school mapping study to test the feasibility of reduced boarding in secondary schools (Education M). 3.37 All of these were carried out during project implementation. In addition, through modifications in the technical assistance component, (a) a planning expert for three years was supplied (Education I), and (b) IEP conducted local school mapping seminars (Education III). 3.38 What is clear from this pattern of small, a hoc items of support is that IDA was seeking specific answers to specific issues and was little interested (if at all) in building up a competent educational planning unit. Indeed, the ad hoc studies of the kind sponsored in these projects often distract a unit and distort the sector-wide an nation-wide perspective that it needs to cultivate through a more balanced program of studies. 3.39 Establishing a planning unit in 1975 was one of the aims announced in the Government's 1973-77 Second Five Year Plan. The unit was established in 1977 and this was an opportune moment for IDA to have provided strong initial support under Education I Such a fledgling unit would have required five or six persons, each trained in post-graduate courses of one, two, or three years' duration in planning disciplines (such as economics, education, school building design, social research, statistics, technical/vocational systems) along with support staff trained in s . The assistance actually afforded under Education I- was too fragmentary, limited and unsustained. The beneficiary of the one year of IIEP training financed by Education I served for a time in the unit as head of statistics, which functioned well, and then departed to work with UNDP Bujumbura. Neither a plan nor planning as such was established under these projects,4 although other donors assisted in training staff and the unit did participate in the design of the Fourth Education Credit. Under Education IV an expatriate expert was considered necessary and fellowships for seven relevant staff were also included. It is therefore only more recently that the effort has taken a serious institution building turn. 3.40 (e) Project Management. Building local project management capacity was one of the most notable areas of failure in this series of projects. 5 Despite the large allocations of staff years to expert services included in each project (in addition to bilaterally supplied experts for project management), there was nothing specifically provided to finance or mandate the training of local staff to perform the functions concerned. The lack of fellowships was paralleled by the absence of any 4V See, for example, pp. 12 - 13 of Auerhan, J. et al: Institutional development in Education and Trainina in Sub-Saharan African Countries. World Bank Discussion Paper, Education and Training Series No. EDT 22. g The PCR erred in stating that eight BER staff went to IIEP (pam. 4.20 of PCR). Only one person went (as planned) to IIEP. 44 In Bank's Education Sector Memorandum (Report No.3692 BU) of 20 November 1981 described the planning unit as "adequately staffed" and having developed a good system for collecting and processing statistical data but noted that it had not done "actual planning" (para. 3.19). 4 The PCR for Education I did not address this issue. The PCR for Education II criticized the lack of training of local staff and the PCR for Education III criticized but took note of the strides made in replacing axpatriates. -34- reference in the relevant experts' job descriptions (either in the DCAs for the three projects or in the SAR for Education IH) to the appointment of counterparts or to the training of local staff by the PIU experts. This is particularly surprising since the OED report Review of Training in Bank- financed Projects I was issued in March 1982, stressing the need to ensure that experts train local staff and the need for an explicit strategy to eliminate future dependence upon outside expertise wherever such expertise was needed in a project The field appraisal of Education IT took place three months later (June 1982). 3.41 The Project Director, at strong Government insistence, was always Burundian. IDA did not indicate any interest in having Burundian counterparts for the project management in technical, accounting and procurement operations until the January 1986 mission raised the issue," following which some funds were found to help train the local accountant and the Project Director on very short courses. After the observations of the supervision team, the Minister of Education promised to assign counterparts with a view to the replacement of the expatriates by 1988. Even with this notice, IDA was quite unprepared for the departure of three experts (including the technical and accounting heads) within a few months of one another in 1988-89, and telexed the Government expressing concern about the departure of "several" persons and enquiring whether suitable replacements could be found. These three persons had spent a total of 21 years in BPE and had trained no counterparts as such; yet they were all satisfactorily replaced by local staff. 3.42 Over the years, at least eight experts financed by the projects spent at least three years in BPE.' One has stayed 11 years in procurement and the audit mission saw no sign of his early replacement. In any event, according to the recommendations of the OED Training Study of 1982, an expert should be refused contract renewal precisely on the grounds of having failed to train a successor and, if necessary, a new expert recruited. In the view of the audit a limit needs to be placed on the number of years that an individual expert can be financed out of loan or credit proceeds. &J Review of Training in Bank-Financed Projects OED Report No. 3834, March 1, 1982. 4 l The next mission in June 1986 reinforced its position by observing that the proposed Education IV project should include provision to train counterparts. -4 The exact figures are No. of Persons Duration of Tenure 2 Les than 2 years 4 2- years 4 3-4 years 4 7, 8, 9, 11 years. V2 The success of counterpart training arrangements depends, Inter als, on (a) the selection of suitable persons as counterparts and (b) the active efforts of the expert to transfer expertise to the counterparts. The audit, after discussion with IDA and Borrower officials, concludes that in this case these two essential conditions have not co-existed at any time. Hence, it would have been more effective to have relled instead upon fellowship training as the principal vehicle for imparting the necessary expertise and for both IDA and the Borrower to have placed a ceiling on the number of years that any individual technical assistance expert would have been allowed to be financed out of the proceeds of any Credit or combination of Credits. -35 - 3.43 The evolution of professional staffing levels in BPE shows a steady expansion of the national share of management after 1985 (Table 3.2). 3.44 Table 3.2 reveals a dramatic watershed in 1988-89 when the preponderance switched from foreign to national staff. This change is explained by: (a) the output of several ov9peas scholarship programs that raised the total number of Burundian architects, (b) the output of engineers by the Faculty of Applied Sciences of the University of Burundi. 3.45 The projects did nothing to help create this situation, although there was one compelling reason why they should have sought to nationalize staff - cost. The ratio of national staff cost to expatriate staff costs ranged between 1:15 and 1:40. Salaries of expatriate staff (excluding housing, transport, travel, medical insurance and pension contributions) ranged from about US$32,000 to US$90,000 in 1991, whereas a local specialist of the same qualifications was salaried at US$2,000- 4,000, with vastly inferior fringe benefits. Table 3.2: NUMBERS OF SENIOR STAFF OF BPE (LOCAL AND FOREIGN) National Staff (excl. Proj. Dir.) Expatriate Staff La Total Tech. Proc. Acetg. Other Total Tech. Proc. Acctg. Other 1977 A - 3 3 1978 - 5 5 1979 - 6 5 1 1980 I - 8 6 1 1 1981 - 7 5 1 1 1982 1 1 6 4 1 1 1983 d 1 1 7 4 1 1 1 1984 e 1 1 7 4 1 1 1 1985 1 1 6 3 1 1 1 1986 3 2 1 6 3 1 1 1 1987 4 2 1 1 6 3 1 1 1 1988 " 4 2 1 1 6 3 1 1 1 1989 6 4 1 1 4 3 1 1990 6 4 1 1 3 2 1 1991 6 4 1 1 3 2 1 a These figuree relate only to persons financed by the projects. A Belgian aid expert spent 10.5 years 1980'-90 vith BPE and was replaced by another who vas there in 1991. b Start of Education I. * Start of Education II. Start of Education III. * End of Education I. End of Education II. End of Education III. 10J In 1983, for example, there were only two local architects In BurandL By 1991 there were between 40 and S0. - 36- 3.46 This enormous disparity in cost was not justified by performance. For example, problems arose in procurement, accounting and in building design and construction under the oversight of expatriates. The major errors in the design and construction of the CFMP (para. 2.7) occurred when no local technical staff were involved. Some expatriate staff have had to be removed for inefficiency or other performance problems. 3.47 Conclusion. On the whole, therefore, after some thirteen years of experience with IDA financed projects, there is still considerable dependence on outside experts and some gaps that will require time to fill. The successes (like BER) have been few and modest. One area that has been almost totally neglected is general administration - central administration and district and school supervision. The projects have not matched their attention to expanding the school system (and rendering it more complex to administer and support) by a commensurate program of assistance to administration. One of the objectives of Education I was "to help the Government increase its control over the sector" (SAR, para. 3.01), but very little was provided to strengthen management of the sector. Education II provided some logistical support (mainly vehicles) to the secondary level curriculum unit (BEPES), the corresponding technical/vocational unit (BEET) and the secondary inspectorate, and Education m lent the same type of support to planning. The officials concerned expressed to the audit mission their appreciation of this support. Even a rather small investment, as in the case of the secondary level, has helped BEPES and the inspectorate to go out to the schools and conduct teacher training in the districts instead of relying on the costly measure of bringing teachers into Bujumbura. Such support for the primary inspectors to help them to conduct inservice teacher training would have been even more valuable in Education I - III. 3.48 A Ministry of Education report in 1988 ("Politique Sectorielle") commenting on primary education, observed that teacher absenteeism and misuse of fees collected at the school and of the resources of school cooperatives were too common and were attributable to insufficiency of supervision - only 400 principals for 1307 schools (often four schools sharing one principal); 163 schools per regional inspector and 42 schools per cantonal inspector. Such fundamental school management issues urgently needed attention as the first concern of any reform effort, as the role of head teachers and school supervisors in bringing about any change within the school is crucial. The large scale waste associated with the CFP/CSE programs is an example of the high price paid when development is ill-planned or un-planned. In this series of projects, the planning and management weaknesses (which entail extensive administrative and development waste) and the urgency of helping the Borrower to be better equipped to address them were inadequately dealt with. IV. FINDINGS AND ISSUES 4.1 Several issues have been touched upon in the previous two chapters. These include: (a) the development strategy in project generation (paras. 1.8-1.22); (b) school design, construction and maintenance problems (paras. 2.6 - 2.9); (c) the extensive use of force account (paras. 2.15-2.23); (d) recording and reporting on project costs (paras. 2.24-2.26); -37 - (e) significant changes in project scope (paras. 2.11 - 2.13, 2.21); (f) appraisal deficiencies (paras. 2.14, 2.26-2.29, 3.J-3.14, 3.17-3.18, 3.21), and (g) the relative neglect of building up human resource and management capability in the sector (paras. 3.24-3.48). 4.2 In this chapter some comments are offered on IDA and Borrower performance and the sustainability of project benefits, and conclusions and lessons are drawn from the experience of the projects. IDA Performance 4.3 Several comments on the performance of IDA in generating and supervising these projects have been made in previous sections of this PAR. Overall, the Association performed very well in many other aspects. It was largely because of its flexibility during implementation (in contrast to its inflexibility during project generation) that amendments were made possible that on the whole led to a much more balanced pictuie of the projects at the completion stage than was evident at Credit Signature. Supervision missions generally did a thorough job. In Education I, it was refreshing to note that the missions reported on adherence to the statements made in the Government's Side Letter. IDA missions also rendered valuable advice on resolving problems of procurement and other matters. 4.4 This section of the PAR therefore limits itself to a few issues: (a) the extent of IDA involvement in management details concerning technical education; (b) the shortcomings regarding the preparation of DCAs, and (c) the incomplete state of the IDA correspondence files; 4.5 Management Details in Technical Education. In April 1983, IDA reviewed the booklists proposed for the ETP and telexed the Borrower expressing its strong reservations about the inclusion of certain French publications as culturally irrelevant to the needs of the African technical education student.51 In the view of the audit, the Association, having accepted the need for French in the curriculum, should have left it to the teaching staff to determine what textual and reference material was needed to achieve the objectives of the syllabus, and focused only substantive departures from the agreed curriculum. 1 M tcl read, inpart EXPRMONS FORTES RESERVES CONCERNANT TEXIES MOLIERE, CORNEIALB, RACINE ET LAGARDE Er MICHARD MOYEN AGE ET XVI SIECLE .[as.[ th we] TROP EWOIGNES DE LANGUE ET CONTEXTE SOCIO- CULTUREL DBEVES ENSBIONANT TECHNIQUE AFRICAINS" - 38 - 4.6 On the other hand, IDA was correct in querying the change to an increase in classrooms at ETP and ETMM. This brought to light that, without due consultation with IDA, the Borrower had changed the agreed common course from the first two years to the first year only, thus increasing the number of specialized classes. Nevertheless, eventually, after review of time tables, alloation of student time and practical and theory classes, the matter appears to have been Rcepted.5' In the view of the audit, however, this change fell within the bounds of the degree of curricular flexibility that is needed in a technical program. 4.7 A third issue was the firm objection by IDA in August 1983 to the sharing of the use of the excellent new facilities of the Mechanical Department of ETS with the University Faculty of Applied Sciences. These facilities were not only underutilized at the time, but they also contained equipment suited to a degree program. Moreover, the Faculty was producing engineers (among whom were potential teachers of technical schools) and lacked proper facilities. Indeed, this more intensive use of the facilities deserved to be encouraged, provided it did not frustrate the carrying out of the agreed technical education program. The curtailment of this collaboration appeared more in keeping with IDA objection to any form of support for higher education. 4.8 Development Credit Agreements. DCAs have traditionally been prepared with such care, precision and thoroughness that they seldom present a problem during implementation. In this series of projects, however, there were some shortcomings which complicated the tasks of disbursement or project management and are raised here to indicate the need for quality control to maintain the high standard of these documents. 4.9 The Education I DCA (Credit 679-BU) was found during implementation to lack: (a) the usual audit covenant, even though the SAR (para. 4.19) included a reference to the need for an annual audit; (b) any reference to fellowships in Schedule 1 (the schedule authorizing disbursement of proceeds for the various purposes), although Part E of the project description in Schedule 2 referred to fellowships--thus technically making it impossible to disburse for fellowships without first amending the DCA; (c) a list showing the location of the CFP workshops, which was the means by which disbursement officers could be assisted in determining whether withdrawal claims pertained to the agreed project.53 The audit also notes that there was no formal reallocation of Credit proceeds among the various disbursement categories for this project during project implementation despite the considerable actual change that occurred. SbThe files do not show the acceptance, but the one year common course was in operaticn at both schools at the time of the audit visit L3 In this case, the exact locations had not been agreed upon in advance of Credit Approval. However, at least some arrangement could have been made to have an updated agreed list as more localities were identified. -39.. 4.10 The Education II Credit Agreement (Credit 976-BU) contained no reference to a revolving account (although the Education I DCA had included such a reference) and eventually had to be amended to provide for a Special Account. 4.11 The DCAs for the first two projects contained the description of duties of various BPE experts and the Project Director. These did not include reference to counterparts or to the need to train staff. This was probably not an oversight, because the subject was not broached until 1986. 4.12 The audit also draws attention to the imprecise wording of the DCA for Education I on the recruitment and management of technical assistance specialists that could have led to unnecesspr complications in the administration of the project. The problem was rectified in the next DCA.54 4.13 Correspondence Files. Project files are expected to be and are normally very meticulously kept. It was surprising, therefore, to find that the Education I files contain very little dita on missions sent out by the Bank/Unesco Cooperative Program during project generation on such matters as mission composition, back-to-office reports and terms of reference. The Education II files often did not contain a copy of the final supervision report (for example, for the missions of April 1981, October, 1983 and May, 1984). For Education III, there were no Minutes of Credit Negotiations on file; only a progress report during the negotiations. Furthermore, the files shed very little light on the fate of the draft PCR prepared in 1989 by UNESCO - a report which, according to Bank staff commenting on the draft PPAR, was discarded because it did not conform to the Bank's new guidelines.55 Borrower Performance 4.14 The performance of the Borrower has been of a uniformly high standard over this series of projects. Physical implementation has to be judged very good according to criteria of time taken, cost and the quality of the physical plant provided. This performance is attributable to the high standard of leadership that BPE always enjoyed, which ensured good staff morale and accountability. The modifications made in the projects during implementation were usually made at the request of the Borrower and these, with one exception,' 6 have improved the projects in terms of institution building, developing staff capacity and expanding and improving the system especially at secondary and technical levels. Relations between IDA and the Borrower were always cordial and cooperative. 4.15 There were a few problems that deserve mention: 8/ The project unit was to be responsible for, "in collaboration with BER, the selection, briefing, coordination and supervision of the specialists provided under Part D of the Project" [Schedule 4, para. 3(h)]. Part D concerned specialists for both BER and BPB. Thus BPE would have had to consult BER on recruiting and managing BPE staff, whereas the intention was merely to give BPE some influence over the BER experts. The wording of the next DCA (Credit 976-BU) stated the responsibility more precisely. "the selection... of the specialists provided under Part F of the Project and, in collaboration with BEET and other appropriate services of MINEDUC, the specialists provided under Part C of the Project" [Schedule 4, para 3(h)]. Part F concerned BPE staff and Part C the technical specialists for the schools. jf This Report (Burundi Rapport d'Achevement du Troisibme Projet d'Education Credit 1358-BU. June 1989, UNESCO) was published by UNESCO but did not integrate the valuable comments that the Borrower offered on the draft. 15J The exception is the major change in Education I which added the teacher training institution and increased the workshops and staff at the CFPs. -40- (a) the Special Account was use0 as a regular bank account for some time, apparently through a misunderstanding;3 (b) the covenants on establishing a tracer system at the technical schools and ensuring employer participation in the organizing and conduct of technical/vocational education were largely not fulfilled, mainly because the Borrower could not provide the leadership necessary to bring about these innovations. The important point with these and other minor problems was that the Borrower did make every effort to correct them as they were encountered. The audit also considers that the Borrower should have acted more decisively in regard to counterpart training by ensuring that the persons selected as counterparts were all suitable and following this up by requiring all experts to train their counterparts according to an agreed timetable. 4.16 The audit mission observed one problem regarding the organization of BPE that should be addressed. The control of the stores (construction materials and so on) is vested in the Technical Services which is also in charge of the use of these stores. This is not in line with sound administrative and accounting practices. The custody of the stores should be under either an independent unit directly responsible to the Director General or (and this is preferable because of the heavy direct supervisory responsibility placed on the Director General by thi lack of a second Director) under a Director who is also in charge of Finance and Administration. 8 Sustainability of Protject Benefits 4.17 There are two significant aspects of the question of sustainability of the benefits of this project--financial and institutional--and theb, are closely intertwined. The project benefits are not immune to the problems faced by the sector as a whole. 4.18 The most striking phenomenon in educationa4financing in Burundi is the declining share of the national budget going to the education sector.5 While the Government has committed itself to increasing the share of the education budget going to primary education, this is of little importance if no measures are being taken to protect the sector share of the budget. IDA has focussed instead on the issue of the subsector sharing of the sector allocation. In the view of the audit, sustaining the education and training activities at their current level and providing for expansion will require an increase rather than a decline in the sector share of the national budget Furthermore, in the view of the audit, the most vulnerable area of expenditure during a decline in L7 It was found in 1985, that the Special Accounts required under Education II and III were being used by BPB like a regular bank account - the Government was not providing its share of the financing so that the apected amount of US$1.0 million in the accounts had fallen to US$170,000 by February of 1985. The Association was able to have the Borrower replenish the Accounts (under the threat of suspension of disbursements of Credit proceeds) by January 1986. & In commenting on this point after review of the draft PPAR, the Borrower indicated that the separation of the stores from the procurement service, under which it has been placed, and placing it under the Technical Service was done for greater efficiency and has led to much Improvement. However, the Borrower remains open to suggestions including the possible appointment of a Director of Finance and Administration. & The percentage has declined almost steadily from 23 In 1977, to 20.4 In 1980, to 17.6 In 1983, to 16 in 1989. From an average of 20% in the sik years 1977-8Z it has dropped to an average of 16.8% in the years 198449. - 41 - the sector share of the national budget is the level of primary teachers' salaries, the decline of which in real terms leads to declining morale, to the injury of educational quality throughout the system. 4.19 In order to supplement Government resources going to the sector, the Government is requiring communities to assume a greater role in maintaining and constructing primary schools and has raised the primary school fee. Secondary school cost to the family has reached a level where many children have to depend on community contributions and on the extended family to be able to accept the places; in the view of Borrower officials, further increases at this level could mean that some children will simply have to decline secondary places on grounds of poverty. Thus, in the view of the audit, the situation does not offer very much hope of large sums being raised from families. 4.20 On the whole, the primary school benefits appear sustainable. The Government is committed to universal primary enrollment. However, an improvement is needed in the management of the subsector, particularly through appointing principals for each school and providing a school supervisory staff of sufficient size, mobility and capability to perform inservice training of teachers and principals on a regular basis. 4.21 As the secondary system grows and is distributed more widely over the country, there is scope to lower unit costs by reductions in boarding; hence, there is no immediate apparent danger to the project benefits at that level. The technical education benefits are also sustainable. Both subsectors require some more strengthening of their management, especially as to the supervision and pedagogical support of the scbwls and guidance on the use of the specialized facilities and the maintenance of physical plant. 4.22 The temporary dormant state of the nonformal education and training activities means that the benefits in this area are not sustainable in the current setting. A review of the new policy on the subsector that is being prepared (para. 3.10) would be needed in order to determine how much of the project investment can be salvaged and in what form. 4.23 The vast expansion in the construction capacity under BPE is sustainable only under the prevailing system of force account and of advance project financing of construction activities. The challenge remains to reduce dependence on this special arrangement, while preserving the gains made so far and contributing to tht development of the Burundi construction industry. The maintenance of school buildings will need to be attended to on a regular basis. For the primary level, it is reasonable to establish some village responsibility in line with the increasing participation of the villagers in constructing the schools. For the secondary level, however, it would be necessary to supplement any such community resources with a central Government role, which so far has depended largely on the projects and particularly BPE; such dependence on the projects for maintenance is not sustainable. 4.24 Project management capacity can be maintained, but the process of nationalizing staff ought to continue. The failure to replace the procurement specialist serves as a reminder that determined action by the Government is needed to find and assign suitable staff (and obtain any necessary training abroad early). It also requires support for such a move by IDA staff. §/ In comnting on the draft PPAR, the Borrower indicated that a seminar was held by BPE on management and maintenance of school property in May 1992 and that the Maiatenance Guide is being prepared and will be issued by October 1992. - 42 - 4.25 With regard to textbook production, the existing level of activity is sustainable. However, securing more reliable financing arrangements requires closer attention and will be essential to further development. Conclusions and Lessons 4.26 A few conclusion and lessons can be drawn from the experience of these projects: 1. The development efforts regarding an education system should be based on sound planning and, where that planning is absent, IDA should accord high priority to providing assistance toward early establishment of plans and a planning capacity (paras. 1.7, 3.18, 3.39). 2. The development strategy pursued in the projects should respond to the crucial problems of the sector and not be based simply on pre-determined positions on priorities; in this context, where secondary and higher education enrollment ratios are very low, as they were in this case, a crisis can develop regarding the production of middle and higher level manpower required for key leadership and productive roles in the economy (paras. 1.20-1.21). 3. In addressing basic needs in the sector (primary and literacy classes), care should be taken to ensure that the relevant administrative services are adequately equipped to conduct these activities efficiently (paras. 1.21, 3.14, 3.47-3.48). 4. The setting of reform goals in primary or technical education should take into consideration the practical difficulties of effecting changes in traditional attitudes (paras. 3.17, 3.21) and appropriate strengthening of units to help them to pilot reforms should be undertaken as needed--in part under these projects this was done for primary education (paras. 3.31, 3.33) but not for technical education (paras. 3.21, 3.35). 5. Greater care was needed in the preparation of Development Credit Agreements and in the maintenance of project files in this series of projects (paras. 4.84.13). 6. When extensive expansion of a school system is undertaken attention should be paid to ensuring the viability of services for the proper use and care of the buildings and equipment, in this case to guide principals of schools and support village communities in the maintenance effort and to undertake repairs as needed (paras. 2.9, 3.20, 4.23). 7. The performance of IDA reflected insufficient rigor in appraisal in regard to: (a) failure to use the services of a nonformal education specialist to appraise the nonformal education components of Education I and W (paras. 3.12-3.14), (b) appraisal cost estimating (paras. 2.27-2.28) and - 43 - (c) failure to reappraise the major changes proposed in the project during the implementation of Education I (paras. 2.14, 3.10,3.14). & The use of force account in construction of buildings by the project implementation unit under these projects was extensive and consideration needs to be given to a system which promotes more participation by the construction industry as a whole while at the same time preserving and developing the expertise already built up in the present BPE construction services (paras. 2.15-2.23, 4.23). 9. Supervision missions needs to monitor the maintenance of separate accounts for each project, especially as cost is an essential part of the evaluation process (paras. 2.24- 2.25). 10. In these projects more detailed prior agreement was needed on the details of the recurrent financing of textbook production (para. 3.23). 11. In the design of technical assistance components, there should be (a) closer attention to ensuring genuine IDA/Borrower consensus in advance on the aims and content of the program (paras. 3.25-3.28); (b) arrangements in the project to eliminate future dependence upon foreign expert services wherever the need for such services is determined (para. 3.40)- -in particular, by always requiring the expert to train counterparts (paras. 3.40, 4.11) and mandatory limits on the term over which an expert can be financed out of loan/credit proceeds (para. 3.42); and (c) concentration on staff training as a means of building permanent capacity, especially through well designed fellowship training programs, as the successful experience of these projects indicates (paras. 3.31, 3.33, 3.35, 3.39). 12. It is advisable that in the Project Implementation Unit the custody of stores of materials be placed under a unit such as Finance and Administration, independent of the Technical Services (para. 4.16). 13. With regard to the sustainability of project benefits, the single most worrisome issue is the steady decline in the education share of the national budget, which, in this case, IDA did not address (paras. 1.19, 4.18-4.19). - 45 - Appendix Translation Republic of Burundi Ministry of Primary and Secondary Education Office of the Minister No. 620/748 Bujumbura, June 2, 1992 Mr. Graham Donaldson Chief, Agriculture and Human Development Division Operations Evaluation Department The World Bank Dear Sir: We have received the Project Performance Audit Report for the first three education projects, Credits 679-BU, 976-BU and 1358-BU, which we read with great interest. Mr. Ralph Romain, who wrote the report, has tried to show the strong points and the weak points of both the Bank and Borrower. We congratulate him for having been objective. We will seek to improve our performance in the years ahead and ask the Bank to be attentive to our concerns. Until such time as we engage an impartial consulting firm to compare the costs of BPE civil works with those in the private sector, we would ask you to kindly hold off on the proposal to change the current structure of the unit, which has always demonstrated its effectiveness. The change you recommend in para. 2.23 that would make BPE a private.company after a transitional period as a parastatal body is not appropriate if we wish to continue to minimize costs. Privatization is not a panacea for all problems, if any exist at alli The constraints posed by the dispersion of sites over impassable roads, the small (contract] size for the primary schools where the local communities provide local building materials, and rehabilitation of the schools during the school year require flexibility that is often not acceptable to a private company. In addition, BPE has made long-term investments and is amortizing over the same timeframe, whereas the private sector looks at the short run and bids double or triple costs to cover all risks. Wo have just opened the bids for 10 primary schools that the Bank had decided should be awarded to the private sector. No bids were received for four of them. A reasonable price of FBu 12,225,076 was proposed for only one school. The estimates for the five others range between FBu 24,403,181 - 46 - and FBu 29,157,036, even though the local communities will also provide quarry stone, gravel, sand and bricks for the six classrooms and principal's office. Control of the stores was placed under'the Technical Department in the interest of greater efficiency. They were previously attached to the Procurement Office and we have seen a big improvement since the function was split in two. Nevertheless, we will ask the next audit to recommend a better organization chart. One possibility is the appointment of an administrative and financial director, a position that had been eliminated at the recommendation of Bank experts who wanted to reduce administrative costs and unwieldiness. We are open to any sound suggestion. Kindly verify and indicate the exact number of CFPs to which para. 2.3 and Note 3 refer; BPE built 60 CFPs under the first project, 16 under the second and 29 under the third, representing a total of 105 for all three combined. Regarding the comment in para. 2.8 on school maintenance, we are now preparing a Maintenance Guide that will be issued before October 1992. BPE has also just held (May 20-22, 1992) an awareness-raising and orientation seminar on school administration and maintenance. The final report will be sent to you in the very near future. The differences in the costs established in parallel by BPE and the authors of the PCR (para. 2.25 and Note 17c) are readily explained. BPE realized substantial savings on the initial program for the second and third IDA projects. When the completion report was prepared, some small amounts still remained for supplementary works and services. RPP vehicles (para. 3.22) should be reflected under the third project and not the first. The reference to the vehicles should also be corrected to reflect the fact that RPP actually acquired eight vehicles instead of the five indicated in the report, these being six Mitsubishi trucks, five of which have special bodies for book distribution, a Ford automobile and a Toyota twin-cab pickup truck for official errands. We would also point out that RPP never had an expatriate director. It was assisted by two French technical advisors, who cqmpleted their work in 1987, and a UNESCO expert funded by the IDA Credits until June 30, 1991. Since that time, RPP has not had expatriate support. These few clarifications aside, we once again thoroughly congratulate Mr. Romain for his perspicacity and objectivity. Please accept, etc. /s/ Eug6ne Ndaro Minister of Primary and Secondary Education