Tanzania - Human Resources Development Pilot Project

Document of The World Bank Report No. 16519-TA STAFF APPRAISAL REPORT TANZANIA HUMAN RESOURCES DEVELOPMENT PILOT PROJECT August 26, 1997 Human Development Technical Group 1 Country Department 4 Tanzania and Uganda Africa Region   CURRENCY EQUIVALENTS Currency Unit = Tanzanian Shillings I SDR= 1.39408 USD 1 SDR = 869.91 Tsh 1 USD = 624 Tsh As of June 30, 1997 GOVERNMENT FISCAL YEAR July 1 - June 30 (FY97 or FY96/97 = July 1, 1996 to June 30, 1997) ABBREVIATIONS AND ACRONYMS Acronym Definition BEMP Basic Education Master Plan CAS Country Assistance Strategy CEF Community Education Fund CSSC Christian Social Services Commission Danida Danish International Development Assistance DC District of Columbia DCA Development Credit Agreement DEO District Education Officer DflD Department for International Development (formerly ODA) DGIS Netherlands International Development Assistance EPRP Education Planning and Rehabilitation Project (Cr. 2137-TA) ESCC Education Sector Coordinating Committee EU European Union FY Fiscal Year GDP Gross Domestic Product GNP Gross National Product GSES Girls' Secondary Education Support GTZ Deutsche Gesellschaft fur Technische Zusammenarbeit (German Technical Cooperation) HRDP Human Resource Development Pilot Project (the subject of this report) HQ Headquarters IAPSO United Nations Procurement Agency (Inter-Agency Procurement Services Office) IAS Intemational Accounting Standards Vice President: Callisto Madavo Country Director: James W. Adams Technical Manager: Ruth Kagia Task Team Leader: Charles C. Criffin  Acronym Definition IBRD International Bank for Reconstruction and Development ICB International Competitive Bidding IDA International Development Association IRR Internal Rate of Return M&E Monitoring and Evaluation MOEC Ministry of Education and Culture MOU Memorandum of Understanding NCB National Competitive Bidding NECTA National Examination Council of Tanzania NETF National Education Trust Fund NGO Non-Government Organization NIF Not Financed by IDA Credit NORAD Norwegian Agency for International Development PIP Project Implementation Plan PMO Prime Minister's Office POC Program Objective Category (highest level objective) PSG Pilot Support Group PSLE Primary School Leaving Examination RM Resident Mission SA Special Account SDP (Education) Sector Development Program SDR Special Drawing Rights SIDA Swedish International Development Agency SOE Statement of Expenditures SMG (Education) Sector Management Group SSR Social Sector Review TFR Total Fertility Rate TOT Training of Trainers Tsh Tanzanian Shillings UDSM University of Dar es Salaam UNDP United Nations Development Program UNICEF United Nations Children's Fund USD United States Dollar VETA Vocational Education and Training Authority  iii PROJECT PREPARATION Tanzanian Team: Ministry of Education and Culture (S. Odunga, Principal Secretary) and the Social Sector Steering Committee (J. Kipokola, M. Malale, T. Magere, I. Mtulia, R. Shirima), plus many others too numerous to list but including D. Bagenda, Kahitwa, C. Kalugula, S. Kanyelele, N. Kessy, M. Lupatu, M.W. Macharia, Madata, S. Mahai, J. Marango, D. Mbilinyi, A. Mduda, W. Mfuko, J. Mapunda, S. Mlawi, B. Mndeme, Mpanda, J. Mrutu, N. Mugogo, F. Mukyanuzi, M. Mwanukuzi, H. Mwenisongole, D. Ndabi, C. Omari, I. Omari, K. O-Saki, L. Shirima, P. Siwingwa, S. Sumra, Tegisa, E. Temu, and E. Yona. Bank Team: Core Team: C. Griffin (AFTH1, Team Leader); L. Ferreira, A. Follmer, U. Panizza, M. Ravicz; H. Taddese (AFTHI); E. Malangalila, B. Laseko, G. Minja (AFMTZ); A. Dar (HDDSP), E. Adu, M. Kane (LEGAF); P. Vandenheede (LOAAF); R. Levine (now at the Inter-American Development Bank) Extended Team: M. Day, G. Gopal, G. Ransom, Y. Seium, E. Wylie (AFTHI); M. O'Rourke, S. Chakraborty, K. Marble, (HDDED); A. Deolalikar, J. Gill, J. Irving, K. Slenning, A. Staszewicz, L. Tsoflias; D. Whittington (consultants); F. Kranz, C. Magnus, U. Mbfane, J. de Regt (AFTS1); R. Brigish, M. Konishi, M. Mwakangale, E. Sakaya (AFMTZ) Collaborating Agencies: Donor representatives or technical professionals who participated at different stages: 0. Therkildsen, M. Kiernan (Danida), J. Hough (DfID), and S. Manda (UNICEF). Peer Reviewers: N. Burnett (HDDED); D. Steele, B. King (PRDPH) Managers: Over the period this project was prepared, it was guided by J. van Lutsenburg Maas (Division Chief, AF2PH), R. Kagia (Technical Manager, AFTH 1), and F. Colaco and J. Adams (Director)  iv TANZANIA HUMAN RESOURCES DEVELOPMENT PILOT PROJECT CREDIT AND PROJECT SUMMARY Borrower: United Republic of Tanzania Implementing Agency: Ministry of Education and Culture Beneficiary: Not applicable Poverty: Program of Targeted Interventions Amount: SDR 15.0 million (US$20.9 million equivalent) Terms: Standard IDA terms: 40 years' maturity with a 10-year grace period Commitment Fee: Standard Financing Plan: IDA US$20.9 million Government of Tanzania US$ 1.0 million Communities US$ 2.1 million Total US$24.0 million Economic Rate of Return: 12 percent average, conservatively estimated for a matching grant program for primary schools and a scholarship program for secondary school girls. Project Appraisal Document: Report No. 16519-TA Map: Not applicable Project Identification Number: TZ-PE-02789  V CONTENTS COUNT YR AND SECTOR BACKGROUND ........................................................... 1 CURRENT SITUATION IN THE SOCIAL SECTORS ............................................................ 1 Substantial Growth and Improvement in Poverty Under Structural Adjustment ........1 Failure to Improve Social Sector Outcomes ............................................................ 1 MAIN SECTORAL ISSUES .............................................................2 Low Enrollments and Poor Quality of Primary Schooling ....................... ...................3 Low Secondary School Enrollments ............................................................3 Institutional and Governance Issues: Inadequate School-Level Management and Little Accountability to Parents and Communities ...................................................... 5 SOCIAL SECTOR STRATEGY .............................................................5 THE PROJCT ..............................................................6 PROJECT OBJECTIVES ............................................................6 PROJECT DESCRIPTION ..............................................................7 Overview ............................................................7 Community Education Fund Pilot ............................................................8 Girls' Secondary Education Support Pilot ........................................................... 10 Capacity Building ........................................................... 12 Policy Development, Planning, and Research ........................................................... 14 PROJECT COST AND FINANCING ........................................................... 16 PROJECT IMPLEMENTATION ........................................................... 18 Duration of the Project ........................................................... 18 Design and Pre-Test ........................................................... 18 Institutional Responsibilities ........................................................... 19 Project Implementation ........................................................... 19 Procurement ........................................................... 20 Community Education Fund and Girls' Secondary Education Support Program.. 21 Procurement of Goods and Works for Other Aspects of the Project ........... .......... 24 Consultants' Services ........................................................... 26 Disbursement ........................................................... 27 Accounting and Auditing ........................................................... 29 Monitoring and Evaluation ........................................................... 30 LEssoNs LEARNED FROM PREVIOUS BANK/IDA INVOLVEMENT .................................. 31 Low Disbursement Levels ........................................................... 31 Administrative Complexity ........................................................... 31 Inadequate Participation ........................................................... 32 Unmanageable Recurrent Costs .................... ....................................... 32 RATIONALE FOR BANK INVOLVEMENT ................................... .......................... 32 Country Assistance Strategy ........................................................... 32 Economic and Sector Work: Analytical Underpinnings ........................................... 33 Unique IDA Contributions ........................................................... 33 Economic Rationale ........................................................... 34 Complementarity with Other Donor Agencies' Involvement .................................... 35 ASSESSMENTS ........................................................... 36  vi TECHNICAL ASSESSMENT ................................................................... 36 INSTITUTIONAL ASSESSMENT ....................... ............................................ 37 Community Education Fund ........................ ........................................... 37 Girls' Secondary Education Support Program ........................................................... 37 Overall ................................................................... 38 ECONOMIC ASSESSMENT ................................................................... 38 IDA EXIT STRATEGY ................................................................... 39 FINANCIAL ASSESSMENT ................................................................... 40 Funding Required to Support Primary Education and the Project ............................. 40 Sustainability of Primary Education and Project Costs .................. ........................... 40 ENVIRONMENTAL ASSESSMENT ........................... ........................................ 41 PARTICIPATORY APPROACH ................................................................... 41 BENEFITS ................................................................... 42 Access and Quality of Primary and Secondary Education ......................................... 42 Inequality ................................................................... 42 Gender ................................................................... 42 Reduced Population Growth, Improved Family Health, Higher Investments in Children ................................................................... 42 Institutional Strengthening, Capacity Building, and Policy Development ................ 42 RISKS ........................................................... 43 Community Education Fund ......................... .......................................... 43 Girls' Bursary Component ................................................................... 44 AGREEMENTS REACHED AT NEGOTIATIONS ................................................ 44 CONDITIONS OF EFFECTIVENESS ................................................................... 44 IMPLEMENTATION ................................................................... 45 ANNEX A: SUMMARY IMPLEMENTATION PLAN .......................................... 47 ANNEX B: SUMMARY SUPERVISION PLAN ..................................................... 49 ANNEX C: REVIEW OF PRE-TESTS ........................................................... 51 COMMUNITY EDUCATION FUND ........................... ........................................ 51 GIRLS' EDUCATION SUPPORT PROGRAM ................................................................... 53 ANNEX D: ECONOMIC ASSESSMENT ........................................................... 55 BENEFIT-COST ANALYSIS: INTERNAL RATES OF RETURN .................. ......................... 55 Overview of the Range of Returns for the W hole Project ................ ......................... 55 Benefits ................................................................... 55 Costs ................................................................... 57 Community Education Fund: Least-Cost Approach? .................... ........................... 57 GSES Program: Impact on the Distribution of Subsidies .................... ..................... 58 Enrollment Impact ........................................................... 58 Incidence Analysis ........................................................... 59 ANNEX E: FINANCIAL ASSESSMENT ............................... 61 FUNDING REQuIRED TO SUPPORT PRIMARY EDUCATION WITH NATIONAL EXPANSION OF THE CEF AND GSES ............................... 61 Scenario 1: No Increase in the Gross Enrollment Rate ....................................... 61  vu Scenario 2: Gross Enrollment Rate of 90 Percent ................................................... 61 Scenario 3: Gross Enrollment Rate of 105 Percent ................................................... 62 SUSTAINABILITY OF PRIMARY EDUCATION AND PROJECT COSTS .. 62 Scenario 1: No Increase in Primary School Gross Enrollment Rates ...................... 62 Scenario 2: Gross Enrollment Rate of 90 Percent ................................................... 63 Scenario 3: Gross Enrollment Rate of 105 Percent ................................................... 63 ANNEX F: CHRONOLOGY OF PILOT DEVELOPMENT ................................. 65 ANNEX G: SELECTED BIBLIOGRAPHY ................................................... 67 ANNEX H: SUMMARY OF OTHER DONOR PROGRAMS IN EDUCATION 69 ANNEX I: LOGICAL FRAMEWORK . ................................................. 71  viii FIGURES FIGURE 1. GROSS PRIMARY SCHOOL ENROLLMENT RATES ........................................... I FIGURE 2. GRoss SECONDARY SCHOOL ENROLLMENT RATES ...................................... 2 FIGURE 3. PRIMARY EDUCATION: RECURRENT SPENDING 1995 ................................... 3 FIGURE 4. PROPORTION OF FEMALES AT EACH EDUCATIONAL LEVEL (PERCENT) ........ 4 TABLES TABLE 1. CORRELATION BETWEEN EDUCATION AND FERTILITY BEHAVIOR IN TANZANIA, 1994 ............................................................ 4 TABLE 2. OVERVIEW OF PROJECT WITH IDA FINANCING .............................................. 7 TABLE 3. ESTIMATED PROJECT COSTS IN US$ MILLIONS .............................................. 8 TABLE 4. SCOPE OF THE CEF PILOT ............................................................ 9 TABLE 5. SCOPE OF THE GSES PILOT ............................................................ 11 TABLE 6. PROJECT FINANCING PLAN ............................................................ 17 TABLE 7. PROJECT TIMING ............................................................ 18 TABLE 8. PROCUREMENT ARRANGEMENTS ............................................................ 21 TABLE 9. IDA DISBURSEMENT PLAN ............................................................ 27 TABLE 10. KEY PERFORMANCE INDICATORS FOR THE COMMUNITY EDUCATION FuND30 TABLE 11. KEY PERFORMANCE INDICATORS FOR THE GIRLS'BURSARY PROGRAM.... 30 TABLE 12. ESTIMATED RATES OF RETURN TO ADDITIONAL YEARS OF SCHOOLING BY LEVEL OF EDUCATION (1991) ............................................................ 34 TABLE 13. DISTRIBUTION OF BENEFITS OF SPENDING ON EDUCATION: RECURRENT AND DEVELOPMENT BUDGETS BY EXPENDITURE QUINTILE, FY93/94 .................. 35 TABLE 14. ESTIMATED INTERNAL RATE OF RETURN (PERCENT) ................................. 39 ANNEX TABLE D. 1. ESTIMATED INTERNAL RATE OF RETURN (PERCENT) .................. 55 ANNEX TABLE D.2. PECUNIARY AND NON-PECUNIARY PROJECT BENEFITS FOR HRDP WITH ESTIMATES OF MONETARY VALUE ............................................................ 56 ANNEX TABLE D.3. ESTIMATED TOTAL COSTS FOR HRDP ......................................... 57 ANNEX TABLE D.4. ESTIMATED ENROLLMENT IMPACT OF GSES ............................... 59 ANNEX TABLE D.5. DECOMPOSITION OF EDUCATION SUBSIDY DIFFERENTIAL- RICHEST VERSUS POOREST QUINTILE ............................................................ 60 ANNEX TABLE D.6. DISTRIBUTION OF GOVERNMENT SUBSIDIES FOR SECONDARY EDUCATION BEFORE AND AFTER THE GSES PROGRAM ........................................ 60 ANNEX TABLE E. 1. ESTIMATED EDUCATION SECTOR OUTLAYS WITH A GROSS ENROLLMENT RATE OF 74 PERCENT IN 2007 ......................................................... 61 ANNEX TABLE E.2. ESTIMATED EDUCATION SECTOR OUTLAYS WITH A GROSS ENROLLMENT RATE OF 90 PERCENT IN 2007 ......................................................... 62 ANNEX TABLE E.3. ESTIMATED EDUCATION SECTOR OUTLAYS WITH A GROSS ENROLLMENT RATE OF 105 PERCENT IN 2007 ....................................................... 62 ANNEX TABLE E.4. EDUCATION BUDGET CHANGES TO MAINTAIN PER-PUPIL EXPENDITURES, CEF, AND GSES (GROSS ENROLLMENT RATE 74%) ................... 63 ANNEX TABLE E.5. EDUCATION BUDGET CHANGES TO MAINTAIN PER-PUPIL EXPENDITURES, CEF, AND GSES (GROSS ENROLLMENT RATE 90%) ................... 63 ANNEX TABLE E.6. EDUCATION BUDGET CHANGES TO MAINTAIN PER-PUPIL EXPENDITURES, CEF, AND GSES (GROSS ENROLLMENT RATE 105%) ................. 64  Country and Sector Background CURRENT SITUATION IN THE SOCIAL SECTORS Substantial Growth and Improvement in Poverty Under Structural Adjustment 1. Although Tanzania remains one of the poorest countries in the world, with estimated per capita GNP in 1995 of only $120 (World Bank, 1997), there has been steady positive growth in the economy since adjustment began in 1984. Growth was fast enough that, despite a population growth rate of about 2.8 percent per year, even in per capita terms incomes grew rapidly in the second half of the 1980s, with slower growth prevailing in the 1990s. 2. Tanzania's success in rural poverty reduction through growth is an impressive achievement. A comparison of 1991 and 1983 data on rural household incomes shows that the resumption of economic growth in the 1980s substantially improved the lot of the rural poor. In 1983, 65 percent of the population was living on less than $1 a day equivalent; this percentage dropped by 14 points to 51 percent by 1991. The percent of the population in severe poverty (less than $0.75 per day) fell from 51 percent in 1983 to 42 percent in 1991. As a result, there are actually fewer people living below $0.75 per day today than in 1983: nearly 11 million in 1983, and about 9 million in 1991. Without economic growth, population growth alone would have swelled the ranks of the poor from 11 million in 1983 to at least 14 million in 1991. Failure to Improve Social Sector Outcomes 3. Growth has been accompanied by greater inequality. The very poor appear to have fallen behind, most likely due to the skewed Figure 1. Gross Primary School distribution of human capital and Enrollment Rates rural infrastructure. Moreover, despite steady income growth, indicators of basic human welfare 120 have been stagnant or worsening 100 * compared to levels achieved in the 80 [Kenya early 1980s. The gross primary 60 . l I_ . *Tanzania school enrollment rate, which 40 I | t OUganda should have been well over 100 20 11 X1.1 percent by now if previous 0 . l improvements had been sustained, 1980 1985 1990 1995 dropped from 93 percent in 1980 to 74 percent in the early 1990s and has thereafter risen only slightly to 78 percent in 1995 (United Republic of  2 Staff Appraisal Repofl Tanzania, 1996). It remains persistently lower than Kenya's falling rate and is likely to be matched or exceeded soon by Uganda (Figure 1). Figure 2. Gross Secondary School 4. Tanzania's secondary Enrollment Rates school enrollment rate is one of the two lowest in the world, at 26 only 5 percent as recently as 1995 20 (comparedto 11 percentin * 1* * 1 Uganda and 25 percent in Kenya; 15 see Figure 2), and is increasing 10 '.TanzI only marginally year by year. 5 The infant mortality rate, estimated at 144 in 1960 and 98 in 0 1980 1985 1990 1995 1982, dropped to 82 per thousand live births in 1995. Life expectancy at birth was estimated to be 51 in 1995, compared to 52 in 1982. This frustrating lack of progress on key indicators of human capital investment and welfare persists despite substantial investments to improve the supply of social services (which may have been due to inappropriate design of the interventions) and despite income growth (which should have contributed independently to improved social indicators). The Tanzania Social Sector Review (SSR, World Bank 1995) concluded that increasing enrollments and improving the poor quality of schooling at both the primary and secondary levels, coupled with improved opportunities for girls, are the highest priority areas for intervention to reverse this stagnation in human capital and welfare outcomes. This project takes aim at these problems, which are discussed in more detail in the next few paragraphs. MAIN SECTORAL ISSUES 5. The SSR-which surveyed education, health, family planning, water and sanitation, and nutrition-concluded that the highest priority areas of attention are as follows: The secondaiy enrollment rate is sensitive to what is included in the numerator and the denominator, and there has been some disagreement in Tanzania over the actual rate. Using official statistics for 1995 and census data, we estimate the 1995 gross enrollment rate for lower secondary at 6.75 percent and for upper secondary at 1.09 percent. These figures exclude enrollments in teacher training colleges. If Grade A teacher training college enrollments are added to upper secondary, the gross enrollment rate at that level would increase to 2.02 percent. The age ranges for the denominator are 14-17 for Forms I-IV and 18-19 for Forms V-VI.  Tanania Human Resource Development Pilot Project 3 Low Enrollments and Poor Quality of Primary Schooling 6. In 1993/94, the average girl starting primary school in Tanzania was 9 years old, and the average boy was nearly 10. By the time these Figure 3. Primary Education: Recurrent children reach ages 14 to 17, only Spending 1995 5 to 8 years later, about 50 percent have already left secondary school 4% (or had never even started). How to get more children, especially girls, into school earlier is the first [ili] problem (the quantity problem). Materials How to make sure schools are effective in keeping them there, advancing them, and providing a 87% good education is the second problem (the quality problem). 7. Low spending on primary education is an contributing factor to this situation. Public spending in 1995 of $23.96 per primary school pupil was allocated almost entirely to teachers' salaries and emoluments, plus examination costs (Figure 3). Parents added an estimated $7.76 to government spending, for a total of $31.72 per pupil. About 42 percent of the parental contribution was for uniforms; for the poorest fifth of the parents, uniforms constituted 59 percent of the contribution. The result is low spending on non-salary and non-uniform inputs at the primary level, and conditions in the schools reflect it. School buildings are in terrible condition: 70 percent of the 3.7 million primary students do not have desks; 30 percent of the 105,280 teachers do not have a chair or a desk; and learning materials are scarce. A recent survey found that schools in 27 out of 30 districts appeared to have attained the government's objective of 1 textbook for every 3 students, yet the books were not in the classrooms. While learning "can take place under a tree," it is done most effectively when the teacher can devote time to the students and has materials to work with, which is typically not the case in Tanzania today. Low Secondary School Enrollments 8. Entry to secondary school in Tanzania has historically been rationed for manpower planning purposes, at least up to 1986, when non-government schools were allowed to enter the market. Rationing has deprived the country of a large stock of human capital in comparison with all other countries, has unintentionally channeled public subsidies for post-primary schooling to the better off, and has distorted the country's income distribution by raising the returns to schooling for the select few who advance through the system. Failing to get more girls through secondary school has reduced the impact of other costly investments Tanzania has  4 Staff Apprisal Report made in health, nutrition, family planning, water, and sanitation. The key issue is how to increase substantially the Table 1. Correlation between Education and Fertility transition rate from Behavior in Tanzania, 1994 primary to lower Characteristic No Primary Secondary Education Complete or Higher secondary, and to do Girls 15-19: Already a 39.9 30.1 9.6 this in a way that mother or pregnant e ph, ze ........................................................I................................................................................. Women 20-24: Mean 5.8 4.9 3.7 education for girls, number of children desired _ , ~~~~~~~~~~~~................................................................................................................................................. 9. Education Women 15-49: Use a 6.0 13.8 31.1 for girls is an modern contraceptive conce at the (percent) .... .......................................................................... ..................... Percent of the sample 29.1 45.6 4.0 secondary level. Source: United Republic of Tanzania and Macro Intemational 1995 First and foremost, the social impacts of completing primary and secondary school for girls in Tanzania are extremely high. Apart from improving the welfare of the girls themselves, education for girls provides benefits beyond the individual herself in the form of better health, nutritional, and educational outcomes for her family. Just one indicator of these benefits is the difference in various dimensions of fertility-related behavior for girls who complete primary or higher levels of education, as summarized in Table 1. The gains shown for each level of education are large, and analysis of the data has suggested that this is a causal relationship. Yet, as indicated by the bottom row of the table, the task of getting a larger proportion of girls through primary and secondary schools is far from finished: only 4 percent of the sample of married women between ages 15 and 45 attended secondary school. 10. Girls' participation in education in Tanzania has improved steadily, and they have maintained parity with boys at the primary level for ten Figure 4. Proportion of Females at Each years (Std IV in Figure 4). Educational Level (Percent) However, girls remain slightly less likely than boys to enter 6 secondary school (Form I in Figure 4). Unfortunately, girls so who do enter secondary school do 40 fStd Iv so with a deficit-the cut-off point 30 t _ _ IForm I on the Primary School Leaving 20 (3 Fr VI Examination for selection of girls 10 1Unjvorsity into government secondary school o is approximately 10 points lower 1972 1985 1990 1995 than for boys. Further, after entering secondary school, they do not perform as well as boys and are more likely to drop out. Only about 25  Tawdia Human Resource Development Pdot Project 5 percent of the students continuing to upper secondary school (Form VI in Figure 4) are female, and the percentage of women at the university level has at best stagnated over the past quarter century. Because of the low enrollment rates, these percentages are based on tiny absolute numbers. For example, at the upper secondary level in 1995, of only 5,841 students, just 1,556 were female (out of an estimated 683,696 19 year-olds). 11. Raising general secondary enrollment rates is a high priority, as Tanzania falls farther and farther behind the rest of the world in producing secondary school leavers. Pursuing this goal by targeting girls, as this project does, will raise the social return to the investment. Institutional and Governance Issues: Inadequate School-Level Management and Little Accountability to Parents and Communities 12. One option for improving outcomes in education is simply to increase the volume of resources available in the current system of public primary and secondary schooling. However, it is widely acknowledged in Tanzania and elsewhere that the impact of additional expenditures and the sustainability of these improvements can be increased by adding resources in a manner that increases the power of teachers to manage the schools, and to increase their accountability to parents and to objective standards of performance. In Tanzania, while local governments have responsibility for primary schools, they are dependent on the central government for all inputs, including policy, management, capital, personnel, money, supervision, books, and supplies. Responsibility for inputs is fragmented across bureaucracies at the central, district, and local levels with the result that there is almost no accountability at any point in the system for the results that are being delivered. Parents are required to make contributions to the school, but until recently, even these school fees were sent to the district.2 Primary school committees have not functioned well and are used primarily to enforce compliance with school policies. However, as a result of a number of interventions (including training for head teachers and the Primary Education Support Program) school committees are being strengthened. SOCIAL SECTOR STRATEGY 13. The Government recognizes these problems and the importance of correcting them. It has proposed a framework for the social sectors, the Social Sector Strategy (United Republic of Tanzania 1994b), which contains the following six basic principles: 2While this practice was officially changed in the Budget Speech of 1995, it is still apparently widespread.  6 Staff Appraisal Report a. Concentrate public sector resources on core activities of Government. b. Balance personnel and other inputs within the social sectors. c. Decentralize authority to the local level. d. Eliminate constraints to private sector participation in provision of social services. e. Promote high quality standards. f. Move resources closer to the household and promote household investment in human capital. 14. The Ministry of Education and Culture (MOEC) prepared an Education and Training Policy (United Republic of Tanzania 1995) that emphasizes liberalizing the education sector in step with the rest of the economy and increasing the accountability of service providers to clients. A draft Basic Education Master Plan (BEMP, United Republic of Tanzania 1997) was produced in February 1997 as part of the design and management process for an Education Sector Development Program (SDP). These documents recognize the need for an adequately funded, pluralistic school system that benefits from greater involvement of parents and local government in management. They propose additional donor investments in basic education. The BEMP builds on innovations in school-based management to be pilot tested under this project and decentralized district-based management that is being developed under Danida-, DGIS-, and Irish-financed projects. The Project PROJECT OBJECTIVES 15. At the Country Assistance Strategy (CAS) level, the overall program rationale for the project is to efficiently increase investment in human capital to raise incomes, reduce inequality, and improve non-market outcomes. This is but one among an array of IDA-financed activities intended to assist the Government to achieve this goal.  Tanzania Human Resource Development Pilot Project 7 16. This project will Table 2. Overview of Project with IDA Financing contribute to the CAS Component Total* objective by seeking to A. Community Education Component 5.01 strengthen the Borrower's Matching Grants 4.71 education sector in the Academic Improvement Contestable Grants 0.15 Nutrition & Health Contestable Grants 0.15 following areas: (a) raising ''B Girls' Bursary'Component 6.93 enrollments and Bursaries/Scholarships 5.82 quality/leazning outcomes of Science Kits 1.00 Academic Improvement Contestable Grants 0.06 primary education through Gender Improvement Contestable Grants 0.06 increased parental Z ei~~............................................................................................................... 2 increased parental c. Capacity Building 2.22 participation and financing, CEF-Related Training 0.38 GSES-Related Training 0.29 school-based planning and MOEC - Pilot Management 1.10 management of resources, MOEC - Training Program 0.25 school-based quality ..... MOEC - Conferences & Disseminaton 0.19 enhancement initiatives, and D. Policy Developments Planning, and Research 4.49 Strengthening NECTA 0.80 improved support for schools Strengthening MOEC Statistics Section 0.37 at the district level; (b) Pilot Monitoring & Evaluation 0.65 Pilot Monitoring & Evaluation Survey Costs 0.87 expandmig educational Welfare Monitoring Surveys 1.20 opportunities and improving.Policy and Strategy 0.60 =...................................................................... R c.. . . ...................... 9............... quality at the secondary Contingencies for prices (4 percent) 0.45 level, articularl for orer Contingencies for Contributions (5 percent) 0.24 Contingencies for Enrollments (5 percent) 0.24 girls; (c) building capacity Refund of PPF 1.20 through these programs at Total IDA Financing 20.86 the district and community Note: *In millions of US Dollars levels and through selective training and project implementation at the central level; and (d) improving policy development, planning, and research in the education sector through modernized testing and statistics systems, operations research, policy studies, and strategic plans for basic and secondary education. The project will function within the institutional framework of the Government's Education Sector Development Program. PROJECT DESCRIPTION Overview 17. Table 2 shows the four major components of the project with IDA financing. Table 3 shows details of project financing, including local contributions. Each component is discussed separately below.  8 Staff Appraiad Repoil Table 3. Estimated Project Costs in US$ Millions Local Foreign Total Foreign Total ............................ ............................................. Project Component US Dollars Exchange Base Cost (Millions) (Percent) (Percent) A. Community Education Fund Matching Grants 7.84 - 7.84 0 36 Academic Improvement Contestable Grants 0.15 - 0.15 0 1 Nutrition & Health Contestable Grants 0.15 - 0.15 0 1 Component Sub-Total 8.17 - 8.17 0 37 ......... ..... __................................................................................................. .................... ................... .............................................. .............................................._ B. Girls' Secondary Education Support Bursaies/Scholarships 5.82 - 5.82 0 27 Science Kits - 1.00 1.00 100 5 Academic Improvement Contestable Grants 0.06 - 0.06 0 0 Gender Improvement Contestable Grants 0.06 - 0.06 0 0 Component Sub-Total 5.93 1.00 6.93 14 32 C. Capacity Building CEF-Related Training 0.34 0.04 0.38 10 2 GSES-Related Training 0.26 0.03 0.29 10 2 MOEC - Pilot Management 0.88 0.22 1.10 20 5 MOEC - Training Program - 0.25 0.25 100 1 MOEC - Conferences & Dissemination - 0.19 0.19 100 1 Component Sub-Total 1.49 0.73 2.22 33 10 ........ .... _ .................. .................... ............................ .................... ........................................... ........................... D. Policy Development, Planning, and Research Strengthening NECTA 0.08 0.72 0.80 90 4 Strengthening MOEC Statistics Section 0.04 0.33 0.37 90 2 Pilot Monitorng & Evaluation 0.39 0.26 0.65 40 3 Pilot Monitonng & Evaluation Survey Costs 0.79 0.09 0.87 10 4 Welfare Monitoring Surveys 1.08 0.12 1.20 10 6 Policy and Strategy 0.36 0.24 0.60 40 3 Component Sub-Total 2.73 1.76 4.49 39 21 Total.. giiWd. ............................................................. .......1... 8.3- ...... ..........3....4...9... ............. 21..... 8- .... ................16 ............................ I'",.................................... Total Base Cost 18.32 3.49 21.81 16 100 Contingencies Contingencies for prices (4 percent) 0.05 0.49 0.54 90 2 Contingencies for Contributions (5 percent) 0.24 - 0.24 0 1 Contingencies for Enrollments (5 percent) 0.24 - 0.24 0 1 Refund of PPF 1.20 - 1.20 0 6 ....... ....................................................... ........................................... . .................... ............................................... ...................... .................... ............. ..................... Total Project Costs 20.05 3.97 24.02 17 112 Community Education Fund Pilot 18. Objective: to raise enrollments and the quality ofprimary education with increased participation of the community and teachers in managing their schools (IDA financing: $5.0 million equivalent). 19. The Community Education Fund (CEF) pilot is a matching grant program for primary schools designed to increase enrollment, improve the quality of schooling, and increase parental and community involvement in monitoring school performance. Planning of the CEF, begun in March 1995, included pre-tests of the design and consultations with beneficiaries over a two- ear period, financed by the Education Planning and Rehabilitation Project (EPRP).  Tan_aia Human Resource Development Pilot Project 9 20. During the pilot phase to be financed under this project, districts and/or schools will be invited to join the CEF program. For these schools, the government's contribution from the recurrent budget will remain unchanged to avoid disrupting the operation of the system. The pilot will add funds at the margin while helping to develop an institutional framework to enable schools ultimately to manage most of their own affairs. The CEF program requires four basic steps, as follows:3 a. The primary school (typically the teachers and possibly the school committee, led by the head teacher) draws up a three-year plan. b. Parents and the local government agree to the plan, democratically choose a per-pupil contribution to support activities for the current year, and pay in half of their planned contributions. c. MOEC, through the CEF program, matches parental contributions and could in the future match other community and district contributions, all according to a formula that favors poorer communities/schools (a sliding scale with a cap on the per-pupil match). d. Schools account to parents and to the project for their expenditures and performance against the school plan. The accounting is expected to take place at the midpoint (when parents, if satisfied, provide the rest of their contributions, which are also matched) and at the end of the school year. e. This cycle is repeated annually. 21. From Table 4. Scope of the CEF Pilot calendar year Year New Schools Total Schools Source of Financing 1997 through 1996 30 30 EPRP (Design & Pre-test) 1999 (classes 1997 105 135 EPRP (Design & Pre-test) begin in 1998 115 250 HRDP Year I 1999 105 355 HRDP Year 2 January), the 2000 645 1000 HRDP Year 3 CEF program 3The details of the operation of the CEF program are specified in several project implementation documents: Community Education Fund: Draft Program Implementation Document (full document and a shorter version for schools); Community Education Fund: Memorandum of Understanding and Three Year Plan; Community Education Fund: Handbook for Primary Schools. These documents are available in the project file and on the Bank's intranet (http://afr.worldbank.org/aft2/educ/tanzcef.htm). They are intended to be reviewed and updated each year during the pilot project to reflect an improved understanding of how to implement the project in the simplest possible way.  10 Staff Apprasal Repod will gradually expand to cover all schools in the districts of Kibaha, Kilosa (these two districts were fully covered in 1997), Lindi Urban, Mtwara Rural, and Sumbawanga Urban (a total of 355 schools, see Table 4).4 The gradual expansion of the CEF will be used by the monitoring and evaluation component to track its impact. No later than the third and final year, the project will be expanded to cover an additional 650 schools on the assumption that the program design and implementation experience will at that point be adequate for such an expansion. The additional schools (or districts) will be chosen nationally on a competitive basis to test the ability of the program to be extended nationwide. 22. The project will finance the matching grants plus development costs for school-level governance, management, and accountability, including the necessary materials and training, plus incremental administrative costs at the district level. In addition, for schools that have successfully participated in the CEF program, following the first year, the project will solicit academic improvement plans and school nutrition/health plans from the schools, to be funded on a competitive basis with the limited funds available for this purpose. In addition, CEF schools will be eligible to compete for additional grants to implement school health and performance improvement programs. Girls' Secondary Education Support Pilot 23. Objective: to expand educational opportunitiesforgirlsfrom poorer families and to improve quality at the lower secondary level (IDAfinancing: $6.9 million equivalent). 24. As with the CEF, development of the Girls' Secondary Education Support (GSES) component has been financed under EPRP, with pre-tests and client consultation to improve the design.5 The GSES component provides for the following: a. Selection of one poor girl per primary school annually (which will usually translate into one girl from each 4 The number of districts and schools in Table 4 will be subject to revision each year following discussions between the Government and IDA. The goal will be to accelerate the expansion, if feasible, so that project funds can benefit more schools. However, performance criteria defined in the Project's Operational Guidelines will have to be met to justify expansion to more schools. The details of the operation of the GSES program are specified in: Girls' Secondary Education Support Program: Draft Program Implementation Document (full document and a shorter versions for schools and villages). These documents are available in the project file and on the Bank's intranet (http://afr.worldbank.org/aft2/educ/tanzgses.htm). They are intended to be reviewed and updated each year during the pilot project to reflect an improved understanding of how to implement the project in the simplest way.  Tanzania Human Resource Development Pilot Project 11 village) who is academically capable but will not be able to attend secondary school without a bursary.6 b. Payment of a bursary by the project to a pre-qualified secondary school in which the girl enrolls. c. Monitoring of the girl and the secondary school to make sure that agreements are kept and the girl is doing well. 25. Girls are eligible for support during all four years of lower secondary school. They may interrupt their schooling for up to one year for any reason or repeat a year of school and retain eligibility. During the first year of the project, the Government will carry out a review of existing policies governing the interruption of school attendance by girls in primary and secondary schools and prepare an action plan in consultation with IDA, based on the recommendations of that review. Twenty percent of the girls with the highest scores on the Certificate of Secondary Education Examination each year will be eligible for an additional two years of bursary support to complete Forms V and VI, with no provision for interrupting their studies or for repeating a year of school. Schools and girls are required to keep the "sending" community informed of each girl's progress. 26. Under the pre-test, 392 girls were placed in 1996 in a total of 23 public Table 5. Scope of the GSES Pilot Year* New Lower Active Lower New Upper Source of Financing Secondary Secondary Secondary Bursaries Bursaries Bursaries 1996 392 392 EPRP (Design & Pre-test) 1997 364 733 EPRP (Design & Pre-test) 1998 894 1,627 HRDP Year 1 1999 894 2,521 HRDP Year 2 2000 894 3,046 74 HRDP Year 3 2001 0 2,682 73 HRDP Year 4 2002 0 1,788 179 HRDP Year 5 2003 0 894 179 HRDP Year 6 2004 0 0 179 HRDP Year 7 2005 0 0 0 HRDP Year 8 3,415 684 Total Number of Girls Note: Of 392 girls entering in 1996, 369 continued to the second year. *Calendar Year 6 Selecting girls has been a problematic area during the pre-test, and experimentation will continue. The goal is to finance at least one new girl from each village for secondary school annually. Another suggestion has been to finance one girl per stream, so as not to favor smaller schools. In the second year of the pre-test, villages could select up to three girls from each school and split the bursary among them (requiring in some cases partial support from the parents). This did not work at all and had to be stopped because it was too complex and villages resisted making decisions about providing only partial support for some girls. In addition, not all selected girls have been able to find a secondary school that would take them. It may take several years to discover a workable solution that allows the program to  12 Staff Appraisal Repoif and private secondary schools. In January, 1997, that number was expanded to 733, after 23 girls dropped out of the program and 364 new girls entered. Additional girls will be added each year of this pilot project, which will require that financing for this component be continued through calendar year 2005 (see Table 5), until all of the girls qualifying for upper secondary school bursaries have finished (see Table 7 for a more detailed explanation of project timing). 27. IDA will finance the bursaries plus the costs of setting up and managing the GSES program. In addition, for schools that have successfully participated in the GSES program, following the first year, the project will solicit academic improvement plans and plans to improve the opportunities and care of girls in the school. These plans will be generated by the schools, and grants will be awarded on a competitive basis with the limited funds available for this purpose. Finally, IDA will finance the purchase of science kits to improve the teaching of science at the secondary level in Tanzania.7 Capacity Building 28. Objective: to help develop capacity to implement CEF and GSES at all levels and to help MOECformulate and implement innovative solutions to problems in primary and secondary education (IDA financing: $2.2 million equivalent). 29. CEF and GSES Training: The CEF and GSES are vehicles for capacity building in primary and secondary schools, districts, private firms, and NGOs to support school-based management at the primary level and development of a primary school-based selection mechanism for the bursary program. During the CEF pre-test, there has been a large investment in training materials and seminars to familiarize educators and parents with the mechanics of the program. Once operating, though, the CEF program facilitates learning-by- doing at the community level in school planning, democratically discussing options and agreeing to an appropriate level of contributions to support the school, procurement, supervision of contractors, and review of implementation activities. The pre-test of the GSES program has also required training of communities on the value and mechanics of the program, of secondary schools on the opportunities and challenges presented by the program, and of district and regional education officials to irnplement aspects of it. Through learning-by- doing, the GSES pre-test has built capacity by a private contractor to develop and manage the bursary program in coordination with district and regional achieve its objectives, given the myriad problems parents face in getting children into secondary school in Tanzania. 7These science kits are a carry-over from EPRP, which may have insufficient funds to finance a procurement that is at the award stage in FY98. Improving the teaching of science in secondary schools is a high priority for MOEC. Assistance for the teaching of science to be funded under this pilot project will be targeted partially to benefit girls in the GSES program.  Tanzania Human Resource Development Pilot Project 13 education officials and secondary schools; capacity and interest at the village level in the selection of girls for the award; and capacity in secondary schools to meet the needs of these special girls. A particular feature of the GSES program has been its contribution to improving knowledge of the secondary school system in some villages that have not sent a child to secondary school in many years, which is essential for better functioning of this market. 30. The pilot project will finance training in CEF schools and training for expansion of the GSES program to new primary and secondary schools. The initial training may be undertaken by firms, NGOs, and district education officials under an open, competitive bidding process. Follow-up after the initial training is expected to be the responsibility of the district education team. Funds and materials made available after the first year for "just-in-time" assistance to district teams and schools and for "help desks" to assist with specific problems are expected to address problems that arise. 31. Each year the training materials, program documentation, forms, and procedures will be reviewed to improve and simplify the program. Learning-by- doing will be enhanced through the development of twinning arrangements, dissemination of model school plans and best practice, and improved documentation for the program, with the ultimate goal that many primary schools should be able to implement the program without special training. The training element of the program is expected to be expensive, so the MOEC is committed to finding ways during the pilot period to minimize these costs to increase the potential for national implementation. Incorporating CEF and GSES activities into the day-to-day work of district and school officials is a key element of planned efforts to reduce administrative costs. 32. MOEC/Pilot Management: During the pre-test, the MOEC contracted out most of development and implementation of both the CEF and the GSES. It may continue to do so for the pilot, but to expand the programs, it has become clear that MOEC needs to develop greater capacity to develop, manage, and monitor pilot programs of this nature as an element of policy development for the sector. During the life of the project, the credit will finance a pilot director, a coordinator for each of the components (CEF, GSES, and M&E/Policy/Capacity Building), an accountant, a procurement specialist, an internal auditor (with skills as a financial analyst and Management Information Systems expert if possible), and a small office staff (two secretaries and two drivers8) to support them. These personnel are expected to help the Ministry develop the capacity to manage recruitment and supervision of outside contractors to assist in training and implementation, to develop and manage contracts for complex tasks such as data collection and evaluation, to monitor progress of the pilots, to implement a communications strategy, and to pilot within the ministry performance-based 8 In fact the project will purchase 3 vehicles, but the MOEC will supply I driver on its own payroll.  14 Staff Apprasoa Report employment contracts for personnel. As changes in public sector employment practices are made under civil service reform, it is expected that this management group will be absorbed into the regular staff of the MOEC under normal terms and conditions of civil service employment, consistent with coordinated management of donor-financed activities under the SDP. 33. MOEC/Training: The credit will also finance a small training program to improve skills in the MOEC in areas where there is a shortage of well trained local professionals, including the economics of education, analysis/evaluation of education policy, impact evaluation of education programs, finance, and educational management. The project will support approximately 5 candidates to be selected competitively from among education sector staff and approximately 5 candidates selected through a national competition, to seek masters degrees in high priority areas explicitly defined in the training program. A fixed dollar maximum has been set aside for this program, so the number to be trained depends on the cost of each program. IDA-financed training will be planned by the Government to be complementary to and consistent with other donor-financed training under the SDP. 34. MOEC/Conference/Dissemination: The credit will also finance participation in a small number of international conferences annually for education sector staff to present results of these pilot programs and of Tanzania's educational reforms to professional audiences, to gain from greater exposure to international practices, and to get feedback on the Tanzanian program. Policy Development, Planning, and Research 35. Objective: to strengthen policy development, planning, and research in the education sector through improved testing and statistics systems, operations research, policy studies, and strategic sector development plans (IDA financing: $4.5 million equivalent). 36. A key part of the Government's economic reforms and of its Social Sector Strategy has been to re-orient government to its core responsibilities and to reorient central ministries to an output and policy orientation. In a decentralized system, it is widely agreed that the central ministry responsible for education needs to perform a number of core educational tasks to support the system, including: (a) a national policy framework; (b) maintenance of the core curriculum; (c) teacher certification/standards; (d) technical assistance to lower levels of the system; (e) national performance evaluations and national education statistics with feedback to educators, policy makers, analysts, parents, and students; and (f) monitoring and evaluation of educational programs. As a complement to on-going activities of the ministry and other donors iP (a), (b), and (c), this project through the capacity building component will provide  Tanzania Human Resource Development Plot Project is support for (d) and under this policy component will provide support for (e) and (f), as described below. In addition, the project will assist in high priority elements of (a) not currently covered by the government or other donors. 37. NECTA (National Examination Council of Tanzania): The project will support the mechanization of the national exam system through the use of modem scanning and computer technology. This sub-component will support implementation of Tanzania's policy to improve the transparency of the examination system and provide timely, high quality, reliable feedback to teachers, students, policy makers, researchers, and parents. Other donors, particularly DffD, are expected to assist in improving the quality of the exams themselves, through parallel financing coordinated by MOEC under the SDP. 38. MOEC (MOEC Statistics Section): The project will also support the mechanization and modernization of MOEC's statistical system. Currently, information is hand tabulated at the district level (so school-level data are lost), and it is nearly impossible to merge exam data with school data to understand better the connection between inputs and outcomes, as measured by exams. As with the exam system, support to MOEC's statistical section is intended to improve the feedback to schools and parents, and to improve the usefulness and timeliness of data to be used by researchers. Other donors, particularly the EU, are expected to assist in improving the quality of the statistics, through parallel financing coordinated by MOEC under the SDP. 39. Pilot Monitoring and Evaluation: A natural element of any school system is local-level innovation. In Tanzania, donors, districts, and schools have historically supported specific innovations or interventions aimed at improving educational outcomes. This project explicitly is testing two innovations on a pilot basis. Yet the MOEC does not have an adequate monitoring and evaluation capacity to draw lessons for policy and to disseminate best practice based on such innovations. This project will sponsor monitoring and evaluation of the CEF and GSES components to produce lessons for education policy. In doing so, it aims to build this capacity within MOEC as a regular part of its operations. DfID, EU, SIDA, and Danida are expected to provide support for complementary activities, through parallel financing coordinated by MOEC. 40. Pilot Surveys: Feeding into the monitoring and evaluation effort will be household, community, and school surveys specifically focused on the CEF and GSES pilots under this project, but which may also be of use to evaluate other interventions. The project will finance surveys necessary to evaluate the pilots and will make them widely available for analysis by education researchers. DGIS, DffD, EU, SIDA, and Danida are expected to provide support for complementary activities, through parallel financing coordinated by MOEC under the SDP.  16 Staff Appraisal Report 41. Welfare Monitoring: A key function of the central government in a decentralized system is to monitor who gains benefits of government-financed programs and to offset problems to the degree possible through equalization grants or similar devices. Under this project, the CEF program will employ a sliding scale for matching grants (with a cap on grants), so that poorer schools get a higher match than do richer schools, and the GSES program will target girls from poor families. Yet Tanzania currently has no capability to distinguish poverty indicators below the regional level, and certainly not at the school or household level. Therefore, the project will finance a welfare monitoring capability with the narrow focus of creating a database to guide, and to evaluate the impact of, the targeting criteria to be used in the CEF and GSES pilots. Even with this narrow focus, the effort will naturally provide all of the data necessary for welfare monitoring, and the data will be made widely available for that purpose. 42. Education Sector Development Program: As an element of the SDP, the project will provide resources to support high priority studies needed for further development and implementation of the Basic Education Master Plan and the Secondary Education Master Plan (United Republic of Tanzania, 1997a and 1997b). Financing by IDA is intended for work for which financing is not available from the Borrower or other donors. Use of IDA funds for these purposes will be reviewed and agreed annually by agencies involved in the SDP, using the mechanisms proposed in the MOEC's proposals in April 1997 on SDP management and design (or as subsequently modified by the Government). PROJECT COST AND FINANCING 43. A breakdown of project costs and the financing plan are summarized in Table 6. The total cost of the project is estimated at $24.0 million equivalent, composed of IDA's contribution, the government's incremental contribution for primary schooling, and parents' (or beneficiaries') cash contributions required by the project. Parental contributions to support CEF schools are voluntary, but in the context of this program, they are a substitute for the lack of local taxes to support primary education that otherwise might be tapped. 44. Project costs include contingency allowances over the life of the project of $0.5 million for inflation (estimated at 4 percent annually in dollar terms) applied to base costs after the first year to all elements other than to the CEF matching grants; $0.2 million equivalent to compensate for possible underestimation of community contributions (this is arbitrarily set at 5 percent of the estimated value of the grants); and $0.2 million equivalent for 5 percent higher enrollments than are anticipated at the primary school level. The number of CEF schools and girls supported each year, and the size of the matching grants and bursaries, can also be adjusted to match actual expenditures to available funds, based on yearly revised estimates of the costs.  Tanzania Human Resource Development Piot Project 17 45. Project costs include an estimated $1.5 million equivalent for duties and taxes. Foreign exchange costs are estimated at $4.0 million equivalent, or about 19 percent of total project cost, including contingencies. 46. The proposed credit of $20.9 million will finance about 87 percent of total project costs, including 100 Table 6. Project Financing Plan percent of foreign Source US$ Percent of Total exchange and 78 (millions) costs percent of local IDA 20.9 87 costs. IDA Contributions from Beneficiaries 2.1 9 Central and District Government 1.0 4 financing of such a Contributions high percentage of Co-Financing 0.00 0 local costs is Total 24.0 100 justified to support community development (specifically, the strengthening of local primary schools), district-based training to support local schools, targeting of poorer beneficiaries who primarily live in rural areas, and increasing gender equity in access to secondary education. There is simply not much scope for foreign costs in any of these areas. 47. In addition, the project will be financed by: (a) a government contribution estimated at $1.0 million; (b) beneficiaries' cash contributions of $2.1 million, in addition to unmeasured opportunity costs borne by parents and students, due to incremental increases in enrollments and retention rates caused by the project. No co-financing is anticipated, although as indicated in the project description, coordinated parallel financing by other donors is anticipated (see Table 6). The Government contribution is primarily the estimated contribution of local authorities to the CEF and is assumed to offset dollar-for- dollar parental contributions. These are expected to be minimal estimates of local financing, based on the matching formula and estimated coverage of the CEF.  18 Staff Appraisal Repoil PROJECT IMPLEMENTATION Duration of the Project 48. This is estimated to be an 8-year project, to be implemented over 9 fiscal years (starting by December 31, 1997, with the project accounts closing by December 31, 2005). The project will span 8 school years, from 1998 through 2005. Table 7 shows the timing. The core of the project will be completed Table 7. Project Timing Timetable Project School & Fiscal Fiscal Year Calendar Year Year Year Start End Planned Effectiveness Date (12/31/97) 0 1997 FY97198 FY97t98 First Pilot Year 1 1998 FY97/98 FY98/99 Second Pilot Year 2 1999 FY98/99 FY99/00 Third and Last Pilot Year 3 2000 FY99/00 FYOO/O Girls Finish 4 2001 FY0O/Ol FY01/02 Girls Finish 5 2002 FY01/02 FY02/03 Girls Finish 6 2003 FY02/03 FY03/04 Girls Finish 7 2004 FY03/04 FY04/05 Girls Finish & Project Ends (6/30/05) 8 2005 FY04/05 FY04/05 Girls Finish & Project Closes (12/31/05) 8 2005 FY05/06 FY05/06 during the first three years, but the GSES component will continue until the intake of girls beginning secondary school in January 2000 complete their "O" levels and the 20 percent of those girls going on to upper secondary school complete their "A" levels by December 2005 (which would be halfway through their final school year). Major decisions about the results of the pilot and follow-on steps will be made in the context of the midterm review near the end of the second year. Design and Pre-Test 49. In January 1995, MOEC advertised locally for consultants to prepare the design of the CEF and the GSES components. The work began in March 1995 with technical assistance by IDA staff and short-term consultants. By June 1995, feasible designs had been prepared and vetted. The consultants set out to pre-test the CEF proposal in 4 schools beginning in August 1995, to find out whether it could be implemented successfully. This pre-test was expanded to 30 schools in 1996 and to an additional 135 schools in 1997. Similarly, the GSES proposal was pre-tested beginning with 392 girls in 1996 and expanded to 733 girls in 1997 (netting out 23 who dropped out of the program in the second year). The design, documentation, training program, and procedures have been revised in light of the experience each year of the pre-test. 50. Both pre-tests were reviewed by a professional educator, Professor S. Sumra of the University of Dar es Salaam (UDSM), in 1996. Professor Surnra's reviews suggested that the designs were substantially reaching the goals of the  Tanzania Human Resource Development Pilot Ptoject 19 programs but that a number of changes were merited to either simplify them or to heighten their impacts.9 The recommended changes will be considered in mid-1997 for the 1998 school year (beginning in January), which will be the first year supported by the proposed credit. Institutional Responsibilities 51. As part of the MOEC-led SDP, these pilot activities will be overseen by the inter-ministerial Education Sector Coordinating Committee (ESCC) and a Special Girls' Education Advisory Group. The pilots will be implemented by a MOEC-based Pilot Support Group (PSG) composed of a Project Director; coordinators for the CEF, GSES, and capacity building/policy components; and support staff within the MOEC. Administrative arrangements will follow Government of Tanzania regulations and policies within the framework of donor coordination under the SDP. 52. Technical support will be provided by the Directorates of Primary and Secondary Education. Primary schools, under the supervision of village councils and school committees, will be responsible for implementing the CEF in their communities and for some monitoring and evaluation responsibilities. Primary school teachers, school committees, and village councils will be responsible for the selection of GSES scholarship candidates. Secondary schools (government and non-government) will manage each girl's schooling once she is enrolled, will take some responsibility for non-tuition funds made available for each girl, and will have responsibilities for monitoring and evaluation. District Education Officers will be responsible for oversight of all pilot activities in their districts. Ward Education Coordinators will be responsible for supporting pilot activities in their wards. Parents will be fully engaged in all elements of both pilots. Project Implementation 53. Project Implementation Plans have been prepared for the CEF, GSES, and Monitoring and Evaluation. Procurement of the science kits is at the award stage. A needs assessment of proposed procurements for strengthening NECTA and the MOEC statistics section have been prepared. Studies for policy and strategy have been proposed as part of the Basic Education Master Plan. A work program for the welfare monitoring survey has not yet been developed. Following negotiations, when all elements of the project are agreed, these plans will be compiled and made consistent with one another, along with operational 9 The reviews are available in the project file. 0 The design phase has been financed by the EPRP first as a research/project preparation activity. The credit was restructured in June 1996, in part to finance the expansion of the pre-tests through February, 1997.  20 Staff Apprmisal Report procedures for the project, into the project's Operations Manual, to be completed by credit effectiveness, expected by December 31, 1997. 54. Implementation will be the responsibility of the MOEC's PSG, led by the pilot director and the individuals responsible for each component, with support from other staff in the MOEC and PMO. The Operations Manual and its component implementation plans will be revisited each year and revised to reflect best practice, ideas for simplifying implementation, and problems that have been overcome. At the PSG's choice, more than one approach to implementation for each part of the project will be encouraged, and the team is expected to constantly search for more cost-effective ways to implement the programs. The PSG's primary goal will be to maximize the amount of money directly benefiting primary schools and the bursary girls; in other words, minimizing the proportion of funds spent on overhead and implementation costs. 55. Within this context, (a) development objectives and a limited set of key performance indicators will provide the overall framework for measuring project performance; (b) annual work plans by project managers will provide the basis for annual resource allocations and monitoring; (c) the central monitoring mechanism will be the review of progress reports covering enrollments, pupil performance, client satisfaction and expenditures; (d) interim reviews will assess progress against work plans, budgets, and recommendations of supervision missions; and (e) the pilot will entail a phased incremental expansion on the basis of project performance. Performance reviews will be built up from the school level, in which school performance will be monitored against school plans by parents and project managers. Schools will be held to performance targets, with memoranda of understanding in the CEF drawn up between the school and MOEC, and in GSES, between the girl, schools, parents, and MOEC. Disbursements may proceed slower or faster than planned, depending on the program's performance during the previous year. Procurement 56. Procurement arrangements are summarized in Table 8. About 53 percent of the procurement under the project will be at the school/community level for the CEF and GSES programs. Employment of consultants to undertake training in support of these two programs as well as to perform policy studies, monitoring and evaluation, and other services in support of the capacity building and policy components is expected to account for 23 percent. Procurement of goods, including science equipment, vehicles, and computers at the central government level, will account for about 13 percent.  Tanzania Human Resource Development Piot Project 21 Table 8. Procurement Arrangements Category ICB NCB Other NIF Total 1. Consultants' Services 0 0 4.39 0 4.39 (0) (0) (4.39) (0) (4.39) ..... ... .....................................I................................... ............................................ .............................. ................................. ................................. .............. ..................... 2. Goods 2.58 0.48 0.31 0 3.37 (2.58) (0.48) (0.31) (0) (3.37) ..... ... ............................................................................ .............................................................................. ...................................... .................... .................. ..................... 3. Training 0 0 0.25 0 0.25 (0) (0) (0.25) (0) (0.25) 4. Incremental Operating Costs 0 0 0.33 0 0.33 (0) (0) (0.33) (0) (0.33) ..... ... .............................................................................. ................................................................................ ............................................................. .......... ... ......... ........... 5. Matching Grants to Schools under 0 0 8.57 0 8.57 Part A of the Project (0) (0) (5.41) (0) (5.41) ..... ....................... .................................................. ............................................ . ..................... ................... ....................... 6. Girls' Secondary School Bursaries 0 0 5.91 0 5.91 under Part B of the Project (0) (0) (5.91) (0) (5.91) ..... ... .............................................................................. ................................................................................ ...................................... ..................... .............. ..................... 7. Refunding of the Project 0 0 1.20 0 1.20 Preparation Advance (0) (0) (1.20) (0) (1.20) .................................................................................................................................................................................................. Total 2.58 0.48 20.97 0 24.02 (2.58) (0.48) (17.8) (0) (20.86) Notes: ICB: International Competitive Bidding NCB: National Competitive Bidding Other: Other methods includes consultant contracts following the rules for hirng consultants, community-based procurement following the Operational Guidelines, international and national shopping, and purchasing from IAPSO NIF: not financed by IDA The top number in each cell is the total estimated amount; the bottom number in parentheses is the amount financed by IDA. Community Education Fund and Girls' Secondary Education Support Program 57. Operational Guidelines have been developed for both components including rules for procurement and disbursement. These rules have already been revised twice as part of the pre-tests and for the most part have resulted in successful procurement for the primary schools and the girls supported by the scholarship program. Contracts for implementation of the CEF and the GSES program will be many and of small value. For the CEF, the average annual grant to each school is estimated to be less than US$4,000. This sum will be used by the school to purchase a number of goods and services including school supplies, furniture, equipment, construction materials, and small contracts for works." The nature of the goods or works that will be procured will be determined during implementation. In the GSES, some funds, considerably less than US$100 per girl, may be used for the purchase of books, uniforms, transportation, tutoring, and other education inputs. Parents may choose to use some of the CEF to finance salary supplements for teachers. For accounting purposes, we will assume that such payments come from parental contributions and not from the IDA- financed contribution. Salary supplements are expected to be a minor part of what schools finance with the CEF and will be monitored during supervision.  22 Staff Appyusol Report 58. Given the nature of the procurement, the small sizes of the contracts, and the wide dispersion of the ultimate users of the funds, packaging of procurement contracts will be difficult and may have a negative impact on community participation and ownership. Therefore procurement will be handled and managed by head teachers and school committees to enhance community ownership; to ensure that the goods, works, and services are procured in a timely fashion; to permit the use of funds for diverse needs as determined by the teachers and parents themselves; and to permit development of grassroots artisans. However, a number of measures will be taken to ensure economic and efficient procurement. a. FinancialAccountability for Procurement. CEF schools and GSES girls/schools will follow a basic accounting system detailed in the Handbook for Primary Schools. b. Procurement Planning at the SchooL In the CEF, the management of each participating school (typically led by the head teacher and the School Committee) will prepare a school plan that will be discussed with the School Committee and endorsed by the Village/Ward council. As part of preparing the school plan, a procurement plan will be developed in which the goods or works to be procured and the estimated costs for each item will be listed. This procurement plan will be attached to the CEF grant application to be forwarded to the MOEC (or its agent). The MOEC and the school will enter into a standard Memorandum of Understanding, of which the school plan and procurement plan will be a part, governing the use of IDA-financed matching funds. The MOEC will check on the eligibility of the proposed expenditures before transferring funds to the school account. Schools will be required to post the procurement plans as part of the process of seeking competitive bids locally. c. Information Assistance on Procurement. To ensure economy of costs, the MOEC has prepared a catalogue with reference prices for a list of goods that primary schools are likely to procure. It contains prices, terms, and conditions applicable to purchasing from the quoted suppliers. The catalogue will be distributed to each school participating in the CEF to assist the schools to understand what goods are available and at what prices. The catalogue will be updated and distributed annually.  Tanzanua Humam Resource Development Pilot Project 23 d. Transparency in the Process of Procurement. Under the CEF component, beneficiaries will contribute no less than about 33 percent of the cost of the items being procured, depending on the poverty rating of the community. The MOEC will transfer matching funds to schools only after the DEO has verified documentary evidence that household contributions have been deposited in an account with a commercial bank (or another satisfactory arrangement on an exceptional basis when such banking of funds is impossible, such as managing funds at the district level by the DEO). The parents' personal stake in ensuring that the funds are used properly is high, because the IDA-financed matching grant is commingled with parental contributions, and the pre-tests show so far that they monitor the use of these funds closely. This built-in assurance and pressure from the parents does not necessarily guarantee economic use of funds, but it is expected to be a strong deterrent to the un-economic use of funds. Moreover, parents will be required initially to contribute only half of their annual pledge. When these initial funds have been disbursed, the school must come back to the parents and report progress with the initial funds relative to the posted procurement plan. If parents are not happy with the use of their contributions or with progress against the plan, they have the option at that point not to risk more of their funds, at which point IDA's financing of the matching grants will also stop, as there will be nothing to match. The MOEC will provide assistance to schools to solve the problems that caused the cessation of parental contributions. 59. For the GSES, procurement includes selection of a school for each girl and responsibility for procuring necessary additional items for the girls, such as textbooks and school supplies. Detailed procedures and simplified documents governing procurement are part of the Operations Manual and are reflected in the implementation plan for this component. Most procurement takes place when the girls begin school, in January/February each year, when 75 percent of the fees under the school's joining instructions is paid. A separate financial log is maintained by the school for each girl. This log is reviewed every six months to determine whether expenditures are eligible. At the beginning of the second term, the remaining 25 percent of the bursary is transferred to the school if eligibility criteria are met. 60. Procurement under the GSES has two components. First, for tuition, the value of the scholarship is determined by the "joining instructions" that are  24 Staff Appraisal Report customarily sent to all prospective students just prior to the start of the school year. Tuition and fees are thus determined outside the program and will generally be the same for all students. The bursary program will only finance the fees that students normally pay. Second, savings from tuition can be used through agreement between the girls and their schools to purchase other necessary inputs for the girl on the local market. These will be extremely small retail purchases. A Memorandum of Understanding between the school and the GSES program and between the girl and her village govern the use of these funds. Experience with this arrangement during the pre-test has been generally good, although mixed. Continued participation by a secondary school in the program will be partially dependent on how well these funds are managed on behalf of the bursary girl. 61. Recognizing the school- and community-based nature of procurement under these two programs, the Borrower will conduct procurement audits and assessments at each school each year and will provide feedback to the CEF and GSES beneficiaries on best practice for improving procurement. These audits will also be used during the annual review and revision of the programs' guidelines. Procurement of Goods and Works for Other Aspects of the Project 62. Goods and works will be procured in accordance with the provisions of Section I of the "Guidelines for Procurement under IBRD Loans and IDA Credits" published by the Bank in January 1995 and revised in January and August 1996 (the Guidelines). Because no large-scale works are expected to be completed during this project, most of the procurement outside of the CEF and GSES will be for goods. 63. Procedures. It is expected that of $3.37 million for goods to be financed by IDA will be procured under contracts awarded through International Competitive Bidding, in accordance with the provisions of Section 11 of the Guidelines and paragraph 5 of Appendix I thereto, primarily for science equipment, computers, and scanning devices. Exceptions are as follows: a. National Competitive Bidding may be considered for goods and works estimated to cost less than $100,000 equivalent per contract, up to an aggregate amount not to exceed $500,000 equivalent, which may be procured under contracts awarded in accordance with the provisions of paragraphs 3.3 and 3.4 of the Guidelines. b. Computers, office equipment and vehicles estimated to cost less than fifty thousand dollars ($50,000) equivalent per contract, up to an aggregate amount not to exceed three hundred thousand dollars ($300,000) equivalent may  Twaizana Human Resource Development Pdlot Project 25 be procured under contracts awarded on the basis of international shopping procedures in accordance with the provisions of paragraphs 3.5 and 3.6 of the Guidelines. c. Computers, office equipment and vehicles estimated to cost less than twenty thousand dollars ($20,000) equivalent per contract, up to an aggregate amount not to exceed two hundred thousand dollars ($200,000) equivalent may be procured under contracts awarded on the basis of national shopping procedures in accordance with the provisions of paragraphs 3.5 and 3.6 of the Guidelines. d. Computers, office equipment and vehicles estimated to cost less than fifty thousand dollars ($50,000) equivalent per contract, up to an aggregate amount not to exceed three hundred thousand dollars ($300,000) equivalent may be procured from the United Nations Inter-Agency Procurement Services Office (IAPSO) in accordance with the provisions of paragraph 3.9 of the Guidelines. Procurement from IAPSO should be limited to those instances when items are urgently needed. e. If the need for a limited number of small force account works (which is not anticipated at this time) arises due to a lack of local contractors (e.g., if CEF schools must contract with the Ministry of Works for construction), these works will be carried out in accordance with the provisions of paragraph 3.8 of the Guidelines. The aggregate cost of force account works is not expected to exceed $200,000. 64. Planning. Prior to initiating any procurement activities, the Government will prepare a procurement plan for the project that will be furnished to IDA for its review and approval.12 Procurement of all goods will follow this procurement plan as approved by IDA, and with the provisions of paragraph 1 of Appendix I to the Guidelines. A General Procurement Notice listing all procurement shown in the plan will be published in accordance with paragraph 2.7 of the Guidelines. 12 Apart from small scale procurement for the CEF and GSES, and some equipment for the PSG, procurement in the first year will be limited to science kits for secondary schools, which is currently at the award stage. Procurement in subsequent years by the central government to strengthen NECTA and the MOEC statistics section will be the result of small scale testing of the planned technologies during the first year, following which MOEC's proposal for proceeding (including the procurement plan) will be produced.  26 Staff Appraisal Report 65. Prior Review. With respect to each contract for International Competitive Bidding estimated to cost the equivalent of $100,000 or more, prior review by IDA will be required, and the procedures set forth in paragraphs 2 and 3 of Appendix I to the Guidelines will apply. For each contract to be awarded under Shopping estimated to cost the equivalent of $25,000 or more, the following procedures will apply: (a) prior to the selection of any supplier under shopping procedures the Government will provide to IDA a report on the comparison and evaluation of quotations received; (b) prior to the execution of any contract procured under direct contracting or shopping procedures, the Government will provide to IDA a copy of the draft contract; (c) in addition, the procedures set forth in paragraphs 2 (f) and 2 (g) and 3 of Appendix I to the Guidelines will apply. The first three contracts for goods or works, irrespective of contract value or purpose, will be subject to prior review by IDA. 66. Post Review. All other contracts will be subject to post review during supervision missions. Consultants' Services 67. Quality- and Cost-Based Selection. An estimated $4.39 million will be disbursed for consultants' services. For consulting contracts in excess of $100,000 equivalent, consultants will be selected in accordance with the "Guidelines: Selection and Employment of Consultants by World Bank Borrowers" published by the Bank in January 1997 (the Consultant Guidelines). The fundamental procedure to be followed is called Quality-and Cost-Based Selection, which requires the Government to solicit proposals composed of a technical proposal and a financial proposal. The technical proposals are evaluated first according to criteria set forth in the letter of invitation to bid, then the financial proposals are opened and incorporated into the rating process as set forth in the letter of invitation. Cost is always a factor in the award. Short lists for consultancy services estimated to cost less than $100,000 equivalent per contract may comprise entirely national consultants. 68. Other Proceduresfor the Selection of Consultants. The following selection methods may be used in specific cases, as indicated: (a) Selection Under a Fixed Budget Services for assistance to districts, schools, and communities in training to implement the CEF and GSES programs may be procured under contracts awarded in accordance with the provisions of paragraphs 3.1 and 3.5 of the Consultant Guidelines; (b) Least-Cost Selection Services for standard, well defined auditing and accounting tasks for all parts of the project estimated to cost less than $50,000 equivalent per contract may be procured under contracts awarded in accordance with the provisions of paragraphs 3.1 and 3.6 of the Consultant Guidelines; (c) Individual Consultants Services for training, monitoring, evaluation, policy studies, financial management, project management, and surveys under all parts of the Project that  Tanania Human Resource Development Piot Project 27 meet the requirements set forth in paragraph 5.01 of the Consultant Guidelines may be procured under contracts awarded to individual consultants in accordance with the provisions of paragraphs 5.1 through 5.3 of the Consultant Guidelines. 69. Prior Review. Prior review by IDA will be required for consulting contracts with firms of $50,000 or more and with individuals of $20,000 or more. For contracts with firms between $50,000 and $100,000, IDA must be notified of the results of the technical evaluation prior to opening the financial proposals. For contracts with firms above $100,000, IDA must be furnished with the technical evaluation report and given sufficient time to review it before opening the financial proposals. These procedures are explained in Appendix 1 of the Guidelines. 70. Post Review. All contracts for consultants' services that are not subject to prior review will be subject to post review during supervision missions. Disbursement Table 9. IDA Disbursement Plan 71. The disbursement plan for Fiscal Year Annual Cumulative IDA financing is shown in Table 9. FY978199 4.523 27.5 The project is expected to be FY99900 4.89 11.94 disbursed over a period of eight FYOO/01 4.33 16.27 years, as shown above in Table 7. FY01/02 1.44 17.71 Estimates are based on credit FY02/03 1.13 18.84 effectiveness by December 1997. FY03/04 0.89 19739 FY05/06 0.47 20.86 72. The Government will open, Total 20.86 and the Pilot Support Group (PSG) under MOEC will operate, two Special Accounts. The Project Coordinator and the Accountant will be the signatories to the Special Accounts. One Special Account will be used to fund matching grants and bursaries; the other will fund all other project expenses. IDA will deposit US $1.7 million in the grants and bursaries Special Account, estimated to be equivalent to at least six months of expected expenditures. The Special Account for all other expenses will be funded with $0.5 million-sufficient to fund approximately three months of these expenditures. 73. This project will use existing government systems to make timely disbursements. The PSG will provide an advance to the Ministry of Finance sufficient to fund six months of CEF matching grants to primary schools. If matching grant expenditures are captured in the districts' budgets, the Ministry of Finance will use these funds to provide advances to the DEOs in each participating district. If the matching grant expenditures are captured in  28 Staff Appraal Report MOEC's budget, the Ministry of Finance will provide an advance to MOEC which will provide advances to the DEOs in each participating district. DEOs will use their advances to provide matching grants to primary schools. They will keep track of the amount received by each school and account for expenditures to the PSG on a bi-annual basis or whenever they apply for an advance replenishment, whichever comes first. Districts can receive two CEF advances each year. 74. The districts will make matching grants to schools upon acceptance of a school plan, a memorandum of understanding, a bank statement indicating that funds from the parents have been collected and deposited in the school's bank account, and a statement of expenditures from previous grants. Districts will account for matching grants made to primary schools by providing Statements of Expenditure (SOEs) to the PSG that will include the amnount that each school received, the date funds were disbursed, and the telegraphic transfer number of each disbursement. Because schools have the opportunity to obtain six matching grants over the course of the program, parental willingness to make subsequent contributions will be the best indicator MOEC could have that previously-supplied matching funds were well spent. Based on year-end audits, this process will be re-evaluated and improved each year. 75. In the first year of the program, the PSG will transfer GSES bursaries from the Special Account to participating secondary schools. Beginning in the second year of the program, funds will flow from the Special Account through the Ministry of Finance to districts as they will flow for the CEF. Before the beginning of each school semester, districts will disburse appropriate funds to each secondary school that bursary recipients from their district are attending. Funds will be disbursed to secondary schools based on a Memorandum of Understanding in which secondary schools pledge to provide specified education-related services and supplies, and pocket money to participating girls. From the second school semester onwards, schools will also supply a summary certificate of performance. Districts will keep track of the amount paid to each secondary school and account for each semester's expenditures three months before the start of the following semester. Districts' SOEs will provide information to the PSG on the amount that each school received, the date funds were disbursed, the telegraphic transfer number of each disbursement, and the names of the girls attending each school. Based on year-end audits, this process will be re-evaluated and improved each year. 76. Expenditures for contracts for individual consultant services not exceeding US$10,000 and for all other contracts not exceeding US$50,000 will be claimed on the basis of statements of expenditures. All other expenditures in excess of these amounts will be fully documented when submitted to IDA for disbursement. Replenishment applications for the matching grant and bursary Special Account will be submitted at least every three months. Replenishment applications for the  Tanzania Human Resource Development Plot Project 29 Special Account funding all other expenditures will be submitted every month. Replenishment applications and supporting documentation will be provided to IDA irrespective of the volume of transactions during each month. All documentation supporting SOEs will be available at the MOEC to facilitate inspection by IDA supervision missions. Accounting and Auditing 77. The MOEC will maintain accounts for the project. It will produce quarterly and annual financial statements that will be submitted to the ESCC. Copies will also be sent to IDA. Accounts will be maintained in accordance with International Accounting Standards (IAS) on a PC-based accounting program. To facilitate production of financial information in a manner that will allow it to be linked to technical and operational data, a Management Information System will be developed in a manner that will permit full integration of accounting information. An effective accounting and management information system will be necessary for schools and districts that join the pilot, and systems both at the district and central levels will be improved over the life of the project. Part of the impact of the pilot will be the development of simple but effective systems to support decentralization of financial and management decisions, consistent with the overall SDP framework for donor-funded projects. 78. The annual financial statements of the project will be audited in accordance with International Standards of Auditing by independent auditors acceptable to IDA. The audited financial statements, with the auditor's report thereon, and the auditor's management letter, covering internal control and accounting procedures weaknesses discovered as a result of the audit, will be sent to IDA within six months of the end of the period audited. The MOEC will submit to IDA copies of its quarterly and annual financial reports. 79. At the community level, all accounts will be maintained by head teachers and school committees and available for inspection by MOEC or its authorized representatives. The training materials contain guidance on accounting and simple bookkeeping that will be cleared with accounting specialists within IDA, which will also be stated in the operational manual.  30 Staff Appraisal Report Monitoring and Evaluation 80. The logical framework for this project is included, by component, in Annex I of this document. The key performance indicators, as agreed upon with the government at negotiations, appear in Table 10 and Table 11. A detailed description of the monitoring and evaluation included in the project design is available in the Operations Manual. The Memoranda of Understanding for the Table 10. Key Performance Indicators for the Community Education Fund Objective Key Indicator How Measured Increased Village- or ward-level gross primary These indicators will be Enrollment school enrollment rate measured relative to previous improved Student Measures of school performance on performance in each CEF Performance Primary School Leaving Examination school, in CEF schools relative to non-CEF schools within a Improved School Changes in physical plant, availability district, and in CEF districts Environment for of books and other teaching inputs relative to non-CEF districts Learning (where CEF refers to Increased Parental Parental contributions, number of participation in Part A of the Involvement school committee meetings annually Project). Details of the methods to be used and ............................................................ ..................................................... Improved School Production and quality of school plans targets for the indicators are Management contained in the Monitoring Improved Available funds for primary school and Evaluation Section of the Financing from all sources Project Implementation Plan. CEF schools and for the GSES girls contain performance goals that reflect these performance indicators. Table 11. Key Performance Indicators for the Girls'Bursary Program Objective Key Indicator How Measured* Increased Enrollment of Girls from Primary Participating Primary Schools Enrollment of Girls Schools vs. Non-Participating Primary from Primary Schools School High Survival Rate Survival Rates of GSES Girls through Achieve a Rate of at Least of Supported Girls Form IV 75%, with Improvement for Each Cohort, to End of Form IV; 95% to Form VI for Those Who Receive the Additional Bursa ry. High Performance Performce e of GSES Girls on Achieve a Score in at Least the of Supported Girls Certificate of Secondary Education 75th Percentile of Those Examination Taking the Examination Gender- Improvement of School Environment Objective Criteria to be Friendliness of for Girls according to Objective Developed School Measures Environment *These indicators will be measured for girls or schools, as appropriate, relative to their previous performance; relative to comparators who are not beneficiaries of the GSES program; and relafive to GSES girls attending different schools. They will be applied separately to Lower and Upper Secondary Pupils)  Twanzwa Human Resource Development Pilot Project 31 81. The MOEC and IDA in collaboration with other donor agencies (through the Education Sector Management Group and the ES CC) will conduct joint project implementation reviews at least annually. These evaluation will determine whether program modifications should be made each year and what form they will take. A comprehensive midterm review will take place within 20 months of the start of the project (approximately September 1999). IDA plans two supervision missions annually whose timing coordinated with the production of interim monitoring reports. LESSONs LEARNED FROM PREVIOUS BANK/IDA INVOLVEMENT 82. The Bank Group has financed eight projects in the education sector in Tanzania, beginning in the 1960s. A review of this portfolio provided lessons for design and implementation of this project: Low Disbursement Levels 83. Despite large projects (up to US$38 million), the amount of annual disbursements to education was small, on a maximum scale of $2-3 million annually. In contrast, HRDP is designed to make it feasible to reach a higher level of IDA disbursements by unbundling bulky central disbursements for works and equipment, attracting other funds into primary schools, moving smaller decisions down through the chain to the school level (where there is an incentive to identify the most important investments, undertake them without delay, and minimize costs, as the parents are also contributing to the resources). IDA's previous annual disbursement performance in support of education is expected to be exceeded under HRDP by just the direct grants to primary schools and girls. IDA disbursements will leverage additional community resources for primary education through the device of the matching grant. Administrative Complexity 84. All of the projects were too complex, with too many implementing agencies and too many components. Despite the fact that each project evaluation noted that the complexity had hampered implementation, each subsequent project was even more complex. The current Bank-supported operation (Cr. 2137-TA) is perhaps the most complex, with three loosely related components, 13 sub-components, 3 management units, and at least 9 separate implementing organizations or units. It supports centrally-planned primary and secondary education with inputs of training, textbooks, civil works and technical assistance. In the design of HRDP, the Government and IDA have responded by reducing the number of components, assuring that they are complementary, and focusing on a limited number of high-impact activities. Feasibility of implementing the pilots has been confirmed through pre-testing, and high priority areas for capacity building have been identified.  32 Staff Apprisal Report Inadequate Participation 85. Project design and implementation was centralized, with little opportunity for the ultimate beneficiaries to influence the way the funds were spent. HRDP explicitly attempts to push decision-making authority and resources as close to the final user as possible and places much of the responsibility for monitoring performance at that level. The project supports instruments to increase the information available at all levels to make better decisions-about school plans, school procurement, prices and availability of inputs, secondary schools, test results, and school performance. Implementation of the project will include repeated consultations with beneficiaries and local implementers. Unmanageable Recurrent Costs 86. The World Bank and the government were unsuccessful in responding to the recurrent cost demands associated with the project investments. Many of the capital investments made in the past have depreciated dramatically due to lack of maintenance, lack of recurrent budget support, and lack of local ownership of assets. In the CEF, IDA investments are not made unless localities identify their education needs and put up some of their own funds to co-finance the investment. For girls, the investment is in their human capital, with a long- lasting stream of benefits generated for the girls and for society. Although there is no cost-sharing with parents under the bursary program, the parents incur substantial losses of non-market productivity from the girls to send them to school. There are no future recurrent costs to government generated from the investments in the girls; in fact, in the long run, these investments should reduce demands on the recurrent budget by reducing population pressure on social expenditures. RATIONALE FOR BANK INVOLVEMENT Country Assistance Strategy 87. The central theme of the Bank's Country Assistance Strategy for Tanzania (World Bank 1997b, dated May 7, 1997, and discussed by the Board on June 20, 1997) is poverty alleviation through economic growth and human development. One of the five principal objectives of the CAS is to "Stabilize and then reverse the downward trend in the social indicators through measures that decentralize responsibility, increase local resource mobilization and rationalize expenditures. In the education sector, there is a need for containing expenditure on staff and raising the outlay on other inputs; increasing the efficiency of expenditure of public sector secondary education; and providing added incentives for greater involvement of the private sector in secondary as well as tertiary education. The target gross enrollment rate for primary education should be at least 85 percent by the year 2000." This project is  Tanzwu Human Resource Development Pilot Project 33 specifically oriented to help the government find an effective community-based mechanism to achieve these goals in primary and secondary education. Coupled with education sector conditionalities in the Structural Adjustment Credit (World Bank 1997a) and another operation currently being prepared to strengthen and rationalize higher education (TZ-PE-2806), IDA will soon have three powerful instruments to help the government achieve this CAS objective. Economic and Sector Work: Analytical Underpinnings 88. Public Expenditure Reviews in 1993 and 1997 found substantial under- spending by the government in primary education relative to the cost of a minimally effective combination of inputs. Despite the government's best efforts to protect basic education, the severity of budget cuts precipitated by shortfalls in revenue and debt obligations during the last two budget cycles have led to a further deterioration in the quality of basic education. Current estimates suggest that the central government will have great difficulty keeping per capita education expenditures from falling for the foreseeable future because of limited prospects for revenue growth coupled with a high population growth rate. In addition, the Social Sector Review demonstrated that the incidence of public expenditures on health and education are skewed toward the richer segments of the population in Tanzania, except for basic services (only about half of the education budget has gone to primary education in the 1 990s). The Poverty Profile (World Bank 1994) and the Country Economic Memorandum/Poverty Assessment (World Bank 1996) demonstrate that one of the main factors implicated in the increasingly skewed distribution of income in Tanzania is ownership of human capital in the form of primary education and, especially, secondary and higher education. This situation is due largely to the high returns to education coupled with the highly variable quality of primary education, and to past policies that rationed access to secondary and higher education. Unique IDA Contributions 89. One serious problem the Government has faced in trying to address long- standing problems in the education sector has been the tight overall budget constraint and lack of room for maneuver in budgets that are dominated by salaries (91 percent for primary and 91 percent of non-boarding costs for secondary). IDA's unique contribution in this situation is the ability to reduce risks and improve the quality of possible policy changes by supplying low-cost funds to test potential solutions to Tanzania's most pressing education sector problems. The IDA funds will allow MOEC to avoid disturbing the existing system, other donor programs, or its sector-wide initiative, recognizing that if the pilots turn out to be impractical, no harm should be done to the existing system. In addition, IDA support will allow the Government to test, then to purchase and implement, new technology to improve two fundamental building blocks of an effective national educational system, the examinations and  34 Staff Apprasal Repori statistics systems. In addition to the analytical work cited above, IDA's technical resources have been instrumental in assisting Tanzania to design both the pilots and the monitoring and evaluation program. Economic Rationale 90. Increasing Tanzanian children's educational attainment is fundamental to the country's effort to accelerate economic growth and distribute the benefits of that growth throughout the population. Additional schooling for both men and women has a significant effect on welfare. For example, simulations done for the Poverty Assessment indicate that achieving a rise in the stock of schooling to the primary school level for all adults would mean an increase of 22 percent in their average adult equivalent expenditure (a proxy for consumption). 91. Returns to primary and Table 12. Estimated Rates of Retum to Additional Years secondary of Schooling by Level of Education (1991) education in Group Primary Secondary University Vocational On te Job Tanzania Private Rates estimated from All 3.6 6.9 9.0 19.4 35.2 the 1992 Labor ~~~~(2,113) (609) (41) (814) (514) '.he 1992 Labor ................................ ....... (.i ...... 3 (i .......... i 9.(.?..)...................... (.14 .......... Forc Suvey Male 1. . .9 17.8 33.0 orce Survey (1,612) (360) (28) (523) (416) suggest high Female 10.8 9.0 11.4 20.2 35.0 private returns (501) (249) (13) (291) (98) in wage sector Quasi-Social Rates employment in All 3.6 1.5 0.0 0.0 Tanzania for Note: Numbers in parentheses are the number of observations. both men and Source: World Bank 1997c, based on the 1992 Labor Force Survey. women, but especially for women. The per-year estimates in Table 12 correspond to approximately a 25 percent return for completing primary and 28 percent for completing secondary school. Quasi-social rates of return are also substantial for primary and lower secondary (shown in the bottom row of Table 12). The social returns to educating girls are expected to be even more striking when non- market social benefits are included, such as the impacts on education, health, nutritional status, and fertility.  Tanzania Human Resource Development Pilot Project 35 92. The project is expected to improve Table 13. Distribution of Benefits of Spending equity in the public on Education: Recurrent and Development financing of education as Budgets by Expenditure Quintile, FY93194 well. The richest 20 Poorest <<< Quintile >>> Richest percent of the population 1 2 3 4 5 captures about 38 percent Primary 19% 22% 21% 20% 18% of government subsidies Secondary 8% 14% 17% 24% 36% for education, and the University 0% 0% 0% 0% 100% poorest 20 percent Overall 14% 16% 16% 17% 38% captures only about 14 Source: World Bank 1995 percent of the subsidlies Note: Each cell shows the estimated percentage of public (Table 13). Most of this spending captured by the quintile. If the benefits were equally inequity is caused by the distributed, each cell would be 20%. The distribution of way secondary and University spending is estimated using a very small sample. tertiary places are rationed and financed. One way to improve equity in the Tanzanian education system is to put more money into primary schools, where the benefits are about equally distributed across quintiles. A second way to improve equity is to target subsidies for secondary school to those who are too poor to attend, such as poor girls. These two interventions are exactly what the CEF and GSES are designed to do. For example, we estimate that in districts where the GSES will be piloted, the share of subsidies going to the poorest 40 percent of the girls in the population will rise from 24 percent today to 65 percent by the end of the pilot. Complementarity with Other Donor Agencies' Involvement 93. The major donors in education include Denmark (Danida), the European Union (EU), Finland (Finida), France, Germany (KFW, GTZ), Ireland (Irish Aid), the Netherlands (DGIS), Sweden (SIDA), UNDP, UNESCO, UNICEF, the United Kingdom (DfID), World Bank (IDA), and Norway (NORAD). Donors are a small factor in the financing of education: they were budgeted to spend Tsh 4,952 billion, only about 4 percent of all education expenditures in FY94/95. 41 percent of these donor funds went to primary education, 44 percent to higher and technical education, and only 3 percent to general secondary education. In general, donor funds are allocated to technical assistance, construction and rehabilitation of facilities, textbooks, equipment, materials, assistance for district planning, and training of managers and teachers. Specific donor programs are summarized in Annex H. 94. The intention under this pilot project is to provide funds to schools and to girls in a manner that will be complementary to the technical and financial support provided by MOEC and donors to improve teacher quality, textbook availability, school maintenance, district education planning, and decentralized management. Crudely put, IDA funds for the CEF and GSES represent "demand-side"  36 Staff Appiwisal Reporl interventions; our focus on demand has made feasible by the complementary focus of Government and donors on strengthening elements of the supply side. Both types of interventions are necessary. 95. Supply-side support from IDA under this project for policy studies, strategy development, and monitoring and evaluation will be coordinated annually by MOEC with other sources of assistance in these areas within the framework and institutional setup of the SDP. Allocations for these components have been made on the assumption that parallel financing from other donors will become available over the life of the project, and IDA's support can be modified annually in consultation between the MOEC and its partners. As IDA funds for these elements become redundant, the Borrower and IDA agreed at negotiations that freed-up funds would be reallocated to the CEF and GSES programs. 96. IDA intends to participate in joint donor supervision missions and to be sure that all CEF and GSES inputs financed by IDA are complementary to other sources of assistance. For example, the CEF program should complement Danida/DGIS-funded assistance for district education planning. In a number of districts the CEF program already overlaps with assistance from Irish Aid, UTNICEF, and Danida/DGIS. Coordination has been far from perfect, but IDA will continually encourage the MOEC to coordinate IDA-financed inputs with these other complementary programs. SIDA has supplied textbooks, has improved the distribution of textbooks and supplies to districts; and plans to focus in the future more on what happens in the school. SIDA's inputs can only improve the efficacy of the CEF program and vice versa. The BEMP has a broad reform agenda in which the CEF or CEF-style school-based programs play a major role, with a clear connection to the strengthening of the districts. For the GSES, GTZ has a new program to support science education in secondary school, which can be coordinated with the bursary program to improve science training for poor girls. Assessments TECHNICAL ASSESSMENT 97. This section will assess whether the design can work and whether alternative approaches might be preferred. The CEF design is consistent with the worldwide movement towards greater decentralization and accountability to beneficiaries in education. Matching grants and equalization grants are well developed tools to influence local spending and to improve equity in decentralized systems. The GSES design uses another well accepted and understood financial instrument, scholarships, to target subsidies to a specific, disadvantaged group (poor girls). Many alternatives to the CEF and GSES designs were considered and rejected because they did not (a) adequately encourage community involvement or (b) create substantially different alternatives to approaches currently used in Tanzania. Pilot projects should help  Ta_Atia Human Resource Development Pilot Project 37 find new ways to solve persistent problems. Thus this project utilizes approaches that differ from current practice, and yet are consistent with the government's objective to decentralize management and responsibility. 98. During development of the pilots, many concerns have been raised. The aide memoire of the appraisal mission, in the project files, examines each of the concerns about the design and responds to them. In addition, Professor Suleman Sumra (1996) reviewed the CEF pre-test in the first four schools that participated in the program and the first year of the GSES (Sumnra et al. 1997). Professor Sumra's findings and the conclusions of Merit International Limited, the consulting firm contracted to implement the pre-tests, are summarized in Annex C and are included in full in the project file. INSTITUTIONAL ASSESSMENT Community Education Fund 99. At the political level, self reliance and cost-sharing for social services is politically accepted, so the matching fund approach will have political legitimacy. The Third Term Government is committed to self-reliance, which this project promotes. The Parliamentary Committee for Social Services is committed to decentralization and community participation in decision making. At the sector level, the project is consistent with the Ministry of Education's BEMP and is framed within the SDP. The MOEC commitment to finance the design and pre-test of the project is a good indication that it owns the project and will continue to support it. 100. At the district level, DEOs have the skills to inspect schools and to deal with the project's primary education supply side issues. This project will provide DEOs with additional resources for primary education in their districts, and they will be responsible for the program in their districts. At the community level, the ward or village government has a major role to play in approving the primary school plan and in choosing girls for the bursary. During the pre-test, they have shown considerable flexibility to make the program work but have also exercised a fiduciary role in reviewing choices made at the school level. At the school level, teachers and school committees have been enthusiastic about the project in pre-test districts. This project will encourage school committees to become more active because the project augments school-based management and community empowerment. Girls' Secondary Education Support Program 101. The project fits in the existing education system and will be executed by it. At the political level, parliamentarians are ready to assist in promoting this project, and the President has made several statements highlighting his concern for the education of girls. At the sector level, the MOEC has Directorates of  38 Staff Apraisal Repoit Secondary Education, Planning, and Inspectorate. The ESCC will provide oversight, and a PSG will be established to manage the bursaries within the MOEC. At the district level, the current staff of the DEOs assist primary schools to nominate girls for scholarships. They also can circulate necessary information to the selected girls, and ensure that girls report to the chosen secondary schools. 102. At the community level, the existing ward and village committees are adequate for recommending girls for bursaries. At the school level, primary schools need strengthening. The CEF is addressing this problem. It is anticipated that as the quality of primary education improves, more girls will become eligible to compete for scholarships. Secondary schools will assist the chosen girls to purchase necessary school inputs with their bursary funds. Overall 103. There are auditing capabilities at all levels to adequately monitor this project, and they will be improved under the project. With temporary, specialized assistance of consultants for training, monitoring and evaluation, and analytical work, the government has sufficient institutional capacity and personnel to implement the project. Most fundamentally, the pre-tests have shown that both the CEF and GSES can work, and communities improve substantially their ability to implement the programs within the first few years. ECONOMIC ASSESSMENT 104. This section summarizes the benefit-cost analysis elaborated in Annex D. Benefits used in the analysis are limited to those generated by the CEF and GSES, including the pecuniary benefits from future earnings accruing to individuals because of increased schooling, reduced repeating of grades, and increased school quality. A conservatively estimated monetary equivalent of non-pecuniary benefits, including reduced fertility among girls with more education, and improved schooling and health prospects for the children of primary and secondary school leavers, is also included. These benefits are associated with incremental improvements expected to be produced by the CEF and GSES. 105. Costs include the project's full costs, including those for capacity building, policy development, planning, and research. The incremental foregone opportunity cost to households of additional children attending school are also included.  Tanzania Human Resource Development Pilot Project 39 106. Under these extremely Table 14. Estimated Intemal Rate of conservative assumptions, and Retum (Percent) using equally conservative Cost Basis CEF GSES Total estimates for the benefits, the Program internal rate of return (IRR) for Full Project Costs 19 6 12 the CEF component is Direct Costs Only* 48 11 26 the CEF component ls * Household CEF Contributions, CEF Grants, and conservatively estimated at 19 GSES Bursaries percent; and for the GSES, 6 percent. The return for the project as a whole is estimated at 12 percent. If only the direct costs of parental contributions to the CEF, project contributions of matching funds for the CEF, and project funding of bursaries for the GSES are considered, the IRR for the Community Education Fund is estimated at 48 percent; and for the GSES, at 11 percent. With this lower cost structure, the return for the project as a whole is estimated at 26 percent (see Table 12). 107. Because this benefit-cost analysis assigns no independent benefits to the Capacity Building or Policy Development, Planning, and Research components, treating them only as costs or supporting activities of the CEF and GSES, the IRRs in Table 12 should be considered lower-bound estimates. Unless the unquantified, independent benefits for these other components are zero or negative, the returns to the project can be assumed to be substantially higher than estimated. 108. This analysis ignores the general positive externalities of a better educated population and the positive income distribution implications of the project. A large proportion of the new children likely to enroll in primary school as a consequence of the CEF will come from low-income, disadvantaged backgrounds. All of the estimated 3,415 girls participating in the GSES will come from households in the lower one-half of the income distribution. 109. This project is a relatively small, pilot effort. Thus the capacity building, policy development, planning, and research costs are distributed over a relatively small group of beneficiaries. However, the learning curve is steep, and the pre- tests have shown that after two years, most of the training costs can be eliminated. The IRRs can be increased during implementation if the Borrower is successful in moving resources out of administrative and training costs and into grants and scholarships, which it is committed to do if possible. IDA EXIT STRATEGY 110. The pilot nature of the project provides an exit for IDA. If the evaluation of the pilot is negative, either the Government or IDA can stop financing it. Any bureaucracy developed for the pilot will disappear with it. If successful, the financial instruments used for the CEF and GSES can be adapted to any chosen level of IDA support in the future. For the CEF, if the pilot project succeeds,  40 Staff AppR s Reoen IDA can put conditionalities on future support, such as: (a) the Government must move its subsidies out of the existing system and into the CEF and thus fully empower schools to manage themselves; and (b) IDA support will continue only on an agreed declining path based on district- or school-level poverty indices. For the GSES, if the program is a success and continues IDA's financing can continue on a declining basis. FINANCIAL ASSESSMENT 111. This section summarizes the financial assessment appearing in Annex E. This financial assessment examines the full annual cost of funding primary education and operating the project in 2007, when it is assumed that the project has a national scope and that all start-up expenses have already been incurred. The sustainability of the project under alternative assumptions about growth in real incomes is also considered. The analysis is performed under three gross enrollment rate scenarios. All figures are in 1996 US dollars. Funding Required to Support Primary Education and the Project 112. In 2007, if Tanzania maintains its current 74 percent primary school gross enrollment rate, the full cost of educating all the children in Tanzanian primary schools-including the full cost of operating the CEF and the GSES in all schools-will be approximately US$261 million. If Tanzania achieves a 90 percent primary school gross enrollment rate, this cost will rise to approximately US$317 million. Finally, if Tanzania achieves a 105 percent primary school gross enrollment rate, this cost will be approximately US$371 million. Under each scenario, approximately 77 percent of the cost can be attributed to just maintaining existing per-pupil expenditures; the remaining 23 percent represents the additional funding made available under the CEF and GSES. Sustainability of Primary Education and Project Costs 113. Tanzania's ability to support primary education and the CEF and GSES programs by 2007 depends critically on the pattern of real income growth over the next 10 years. If gross enrollment rates do not change, Tanzania should be able to support primary education and the nationwide roll-out of the CEF and GSES programs by 2007 assuming the country experiences at least 1 percent annual real growth in incomes over the next 10 years. If Tanzania's gross enrollment rate increases to 90 percent by 2007, the country should be able to support primary education and the nationwide roll-out of the CEF and GSES programs assuming the country experiences at least 3 percent annual real growth in incomes over the next 10 years. If Tanzania's gross enrollment rate increases to 105 percent by 2007, the country should be able to support primary education and the nationwide roll-out of the CEF and GSES programs assuming the country experiences at least 3 percent annual real growth in incomes over the  Tanzaua Human Resource Development Pilot Project 41 next 10 years and undertakes some education reforms.13 If these conditions are not met, this program, combined with continued government support to education at its current level of expenditures per pupil, will require either a significant increase in education's share of the government's net current expenditures and/or substantial additional donor support. ENVIRONMENTAL ASSESSMENT 114. The project is rated category C, though under the CEF component many schools will use the matching grant funds to construct classrooms or latrines. None of the activity is foreseen to have a negative impact on the environment in the short run. School construction will be monitored by the district authorities, who should ensure that national building guidelines for schools are followed. 115. In the long run, the project is expected to have a positive impact on the environment. Increased education levels will, among other benefits, contribute to: (a) lower fertility rates and thus reduce environmental pressures associated with rapid population growth; (b) increase the economic capacity of households to produce more educated farmers; and (c) produce better-off farners, who will be more receptive to learn and undertake land conservation. PARICIPATORY APPROACH 116. The design of this project was developed through a participatory process involving stakeholders at all levels, supervised by a Permanent Secretary-level Steering Committee. Identification of the project components is based on participatory sector work, the Tanzania Social Sector Review, which included a nationally representative household survey, focus groups, and other qualitative survey techniques. Project preparation included two rounds of systematic client consultation activities with households, implementers, students, teachers, local government, and donors. The chronology of participatory activities during project development are detailed in Annex F. Local consultants took primary responsibility for the design of the pilot project and for client consultations. As part of the design process, both the CEF and GSES were pre-tested to get more feedback before entering the pilot phase. The design has been repeatedly revised in response to the consultations and pre-tests. In addition, participation is fully built into the project by empowering clients to control resources, by requiring a cash contribution of the beneficiaries, and by strengthening local planning, consultation, and feedback mechanisms to improve the design of the project and the functioning of the schools. 13 From 1985-1995, per capita GNP grew by 1 percent annually and GDP grew by about 3.5 percent annually. If this growth continues or accelerates, the high-cost scenario could be financed.  42 Staff Appd isal Repoit BENEFITS 117. The economic assessment quantifies the main pecuniary and non- pecuniary benefits of the CEF and GSES. This section highlights the range of benefits that are expected, including some that are ignored in the economic assessment. Access and Quality of Primary and Secondary Education 118.- Direct project beneficiaries are estimated to include 400,000 primary school students, 3,415 girls who receive bursaries, and other secondary school students in the schools that will receive bursary girls. The direct beneficiaries include many children who would otherwise receive little or no education. As we expect parents to increase their efforts to enroll boys in secondary school in response to the improved opportunities for girls, we anticipate that the secondary school beneficiaries will include a significant number of boys who would not otherwise have gone to secondary school. Inequality 119. For an average rural household, we estimate that bringing everyone's educatioh to the primary level would result in a 14 percentage point reduction in the number of households living below the poverty line (World Bank 1996). Principally by widening access, improving quality, and explicitly targeting the poor, this project will contribute to poverty reduction and raise the earnings capacity of affected households. Gender 120. The project will reduce gender imbalances by facilitating access of girls to secondary education and by supporting mechanisms to keep girls in school. The GSES will benefit the girls already in the secondary education system by making the system more friendly to them through gender awareness activities. The bursary prograrn is expected to motivate girls in primary school to improve their performance. Reduced Population Growth, Improved Family Health, Higher Investments in Children 121. A longer run benefit will be higher earnings capacity (especially for the bursary girls) reduced levels of fertility, better nutritional and health status for children, and greater educational opportunities for future generations. Institutional Strengthening, Capacity Building, and Policy Development 122. An intentional exclusion from the benefit side of the economic analysis was a valuation for the project's institutional benefits, which will include: (a)  Tazada Human Resource Development Pilot Project 43 building capacity in the statistics department of the MOEC in monitoring and evaluating the educational system; (b) strengthening the capacity of district governments; (c) strengthening the national examination system; (d) improving information about the market for secondary education; (e) improving Government's ability to monitor the impact of its subsidies; (f) strengthening district education teams to assist primary schools; and (g) involving communities in generating to their educational problems. The benefits from these interventions are expected to be considerable, but they are fairly intangible. RISKS 123. There are three general project-specific risks. First, the acceptable performance achieved so far under the pre-test may be due to relatively close supervision that cannot be replicated on a larger scale. Second, the hypotheses that govern the development of this pilot may be wrong. For the CEF, the main hypotheses are that better planning, more money, more transparency, and more teacher and parental control at the primary school level will lead to better outcomes in primary education. For the GSES, the main hypothesis is that poor girls are failing to attend secondary school primarily because of financial reasons and that financing education for these girls will raise enrollments of both boys and girls, resulting in large payoffs for society. If these hypotheses prove to be wrong, the project will not have the expected impacts. Third, Tanzania has poorly developed financial and banking institutions, which may create difficulties in some areas of the country, as this project requires that schools use bank accounts. Community Education Fund 124. With the CEF, two additional risks concern the possible behavior of parents and teachers. First, parents may not be able to make the required financial contributions, or their contributions may be so low that the corresponding matching funds from the CEF will be insufficient to make an appreciable improvement in primary school quality or enrollments. Second, without more supervision and control from the Ministry of Education than envisaged, parents and teachers may decide to spend their funds on inappropriate purchases. 125. Other risks concern the institutional and administrative capacity of village-level committees, district governments, and the central government to implement the project. There are also questions about the interest and ability of district governments to carry out their responsibilities for training, support, and assessment. Administrative costs may be so high as to prevent wider implementation. Higher levels of funding may not increase school enrollments or augment the quality of education. Parents may inadequately supervise their local schools to get good value for the money that will be made available. For  44 Staff Appndsal Report all of these performance risks, schools will put themselves on a distribution. The question to be answered is whether in the end the average performance is high enough and the variance in performance narrow enough to conclude that the risks have been managed adequately. Already, as mentioned earlier, the design has been altered several times to better manage some of the risks. It is in the nature of the pilot that the design will continue to be adjusted as problems are encountered and solved. However, an innovation in this design is that school-level risks are small (not more than $5,000 annually), the CEF and GSES create incentives and mechanisms for self-policing of risks, there will be many good and bad examples to learn from, and the whole project is set up as a learning-by-doing exercise. Girls' Bursary Component 126. For the GSES, potential risks include: (a) the village council may not fulfill its role of providing a check to be sure poorer girls are selected; (b) some girls may receive the scholarship with such low scores on the Primary School Leaving Examination that they may fail without remedial training or may fail to fmd a secondary school that will accept them; (c) for rural communities, the number of players in the scholarship program may be too many to manage; (d) schools may fail in their fiduciary role to adequately care for the girls. As with the CEF, part of the reason for the pilot is to find out if the risks can be managed adequately so that the considerable potential benefits of the GSES program can be realized at reasonable cost. Agreements Reached At Negotiations 127. This section summarizes the main agreements reached during negotiations that have been incorporated into the Development Credit Agreement (DCA). For the details, the DCA should be consulted. CONDITIONS OF EFFECTIVENESS 128. Employment of the Pilot Support Group staff, as outlined on page 19. 129. Appointment of the Special Girls' Education Advisory Group, as outlined on page 19 130. Revised Operational Guidelines and Project Implementation Plan for the project, including revised implementation plans for all sub-components.  Tanzwna Huuma Resource Development Pilot Project 45 IMPLEMENTATION 131. To carry out the project in accordance with the Implementation Program as outlined in Schedule 4 to the DCA and the Project Implementation Plan, which will be produced as a condition of effectiveness and updated annually. 132. To have the records and accounts for the project, including the Special Accounts and all statements of expenditures, audited each year and to furnish the report of those audits within six months after the end of each year. The Government will also provide periodic financial statements, as detailed in this document. 133. To maintain, until completion of the project, its Education Sector Coordinating Committee, which will coordinate and oversee project activities, and the Special Girls' Education Advisory Group, as outlined on page 19. 134. To maintain per-pupil central government subsidies in the recurrent budget in Districts participating in the CEF at the same level it provides in other Districts in its territory, to be verified by a review each September 30. 135. To provide to IDA by March 31, 1999, an action plan for the expansion of the CEF in the third year of the pilot. 136. To adopt an improved sliding scale to replace the one currently used for calculating matching grants under the CEF by June 30, 1999. 137. To carry out a review of the existing selection system for government secondary school places by December 31, 1998, and prepare an implementation plan in consultation with IDA based on the recommendations of such review by March 31, 1999. 138. To carry out a review of existing policies governing the interruption of school attendance by girls in primary and secondary schools by June 30, 1998, and prepare an implementation plan in consultation with IDA based on the recommendations of such review. 139. To develop a database of secondary schools in the country by March 31, 1998, and produce and distribute an annual catalog of secondary schools, with the first version distributed by September 30, 1998, and subsequent versions distributed by September 30 of subsequent years. 140. To carry out an evaluation of constraints to the expansion of secondary school places within the country by June 30, 1998, and to furnish the recommendations of such report, including an action plan for any proposed policy changes, to IDA by September 30, 1998.  46 Staff ApphsvA Repu 141. To firnish to IDA by March 31, 1998, an action plan to test on a pilot basis the type of computer technology to be provided to NECTA and the Statistics Section of MOEC under the Policy Development, Planning and Research component of the project, as detailed in this document. 142. To prepare, within six months of the closing date or such later date as may be agreed with IDA, a plan for the future operation of the project.  Annex A: Summary Implementation Plan Dates Action Responsibility Issuesitasks July Revise PiPs MOEC PIPs should be revised annually at mid-year based on lessons learned, to be ready for the next round of new schools and girls, which starts in September .............................................................................. . .................. ..................... .. .. .. .. ..... .................................................. ............................... ..................................... July Preparations for Implementing Revise training to be consistent with Training Agencies and new PIPs and to reduce costs if MOEC possible based on previous experience ............................................................................. . ........... ............................... ... M t q! ................................ ........... ............... September- TOT and village- Implementing This should allow plenty of time for October level training Agencies GSES selection in the villages, and should allow enough time after training for CEF communifies to have School Plans for the following year completed and approved before end of the current school.... ............................y September - Monitoring Pilot MOEC, World Weekly progress report should be January Preparations for Bank, partners received from PSG that summarizes new school year progress by the various implementing (GSES agencies. selections; CEF training and supo.rt).. .... . .. . . .. .....! -----------------I....... . ....__.__......_._.-.-._.-.-... . ................ . ..... ..................... January - Monitoring Districts and PSG Information should be provided in December schools and girls weekly progress reports to MOEC to make sure (SMG), copied to IDA commitments are followed and to provide assistance ....................................... ............................................................................. _............................................. .................................................................................... December End of CEF pilot, Project continues for 5 more years to 31, 2001 no new GSES allow final-year GSES recipient to g,iris added .......- finish secondary school June 30, End of GSES 2005 December Close Project 31, 2005 Accounts   Annex B: Summary Supervision Plan Dates Action Responsibility Key Tasks and Skill Requirements November- Kick-Off Mission MOEC, World Should be coordinated with December Bank HQ and production of new reviews of the 1997 Resident Mission pretests by Sumra, visits to schools, Staff discussions of findings, joint donor mission for SDP ................. i; ~ i i .............. ......... ........................................................................................................................................................... j February- Supervision MOEC, World Review village-level audits; review March 1998, Mission Bank HQ and progress of distributing annually Resident Mission results/reporting back to villages. thereafter Staff Visit villages to see if they are receiving the support they require on CEF. Visit secondary schools to ensure welfare of GSES recipients be attended to by implementing . .. ~~~~~~~~~~~~~agencies. ............................................................................................................................... q..~?Ri s .................................................................................................. ...................................... June-July Mini-Supervision World Bank HQ Monitor progress since MOEC take 1998 Mission and Resident over implementation on April 1 Mission Staff September- Supervision World Bank HQ Monitor MOEC village-level training October Mission and Resident of new pilot sites. 1998, Mission Staff Review progress of distributing annually results/reporting back to villages. thereafter Visit villages to see if they are receiving the support they require on CEF. Visit secondary schools to ensure welfare of GSES recipients be attended to by implementing agencies. ............................................................................. .................................................... . . .............. ............................................ ............................... June 30, Project Ends 2005 December. . . ................................ .....................................s..... ...e......................... ........................................... ..................................................................................... December Accounts Close 31,2005 Supervision Core Team: Skill Mix: Economist (HQ and RM if possible) Education Specialist (HQ and RM) Monitoring and Evaluation Specialist (HQ and RM if possible) NGO/Participation Specialist (RM) Operations/Financial/Procurement Specialist (HQ and RM) Ancillary Team Skills (mix of non-Tanzanian and Tanzania consultants): Accountant Education Finance Specialist Education Statistics Specialist Examinations Specialist Financial Analyst Gender Specialist Household and Facility Survey Specialists (Quantitative and Qualitative) Project Implementation Specialist Managers of Similar Programs in Other Countries Municipal Finance Specialist School Architect School Health and Nutrition Specialist Schooling Planning Specialist  )k I  Annex C: Review of Pre-Tests COMMUNITY EDUCATION FUND 1. Professor Suleman Sumra (1996) reviewed the CEF pre-test in the first four schools that participated in the program. The paper includes the following major observations: a. Information and Knowledge. Parents were well aware of the program. All were willing to help finance school activities if they knew how the money would be spent and were confident that there would be adequate places to allow them to enroll their children in school on time. b. Emnpowerment and Participation. The CEF was a fundamental departure from education management in Tanzania because it gave communities and schools the flexibility to influence educational decisions. Indeed, it was the first time Professor Sumra had observed in Tanzania that teachers were asked to take responsibility for, and to help decide what, was happening in their schools. Further, it was the first time that teachers and parents were required to work together on these tasks. c. Financing. Professor Surnra was impressed by the willingness of parents to contribute to program financing. Payment of the Tsh 200 school fee (less than $0.50) was about 60 percent before the CEF program. In the CEF schools, pledges ranged from Tsh 2,000 to Tsh 4,000. In one village, parents borrowed the money from a wealthy farmer and paid him back after the harvest. Professor Sumra suggested that the willingness of parents to contribute might be due not to the existence of a matching grant, but rather to the transparent process that had been set up. Parents knew there was a plan for using the money, they could observe how it was going to be spent, and they would receive an accounting of expenditures before they were asked to pay again. d. Priorities. Parents' and teachers' highest priority was to construct additional classrooms and teacher housing. This is probably appropriate because of the lack of investment in these facilities over many years.  52 Annex C 2. Professor Sumra noted some potential issues: a. Participant Fatigue. This was a new program, and everyone was enthusiastic. But it creates a lot of work for teachers and parents. Would they (especially teachers) continue to undertake these activities if there was no remuneration for the extra work? b. Financing. The Project is predicated on schools having a bank account and is based on an urban concept of paying money. In the rural areas, people go for long stretches of time with no money. The pre-test was implemented by a consulting firm that paid the matching grant promptly and according to the rules of the program. Does the government have the capacity to implement this activity country-wide? The amount of money that the project was able to mobilize in the communities was a modest sum compared to what is needed. C. Construction Quality. Given that the program resulted in a focus on infrastructure, Professor Sumra was concerned about the ability of the participating communities to meet reasonable standards of construction quality. Yet he thought that some of the communities had made reasonable construction compromises, such as reducing the size of the classrooms to what was actually needed (smaller than the standards required). 3. In the terminal report of its consultancy, the pre-test implementing contractor (Merit International 1997) noted a number of problems with the program including: challenges in training at the village level; the voluntary nature of the CEF contribution by parents; inadequate marketing; the district government's inability to supervise schools due to a lack of transport; the need for better guidelines for classroom and desk designs for CEF schools; a lack of management training for head teachers; poor school-level statistics; the existence of multiple donors implementing different programs in several districts; and the need for commitment at each level (school, village, district, center) for the program to work. 4. Further, a key problem in the pre-test is the high cost of training required due to the participatory/transparency elements of the CEF and GSES. These costs can easily amount to several times the value of the matching grant and the girls' bursary, so they can only be justified if they are a one-time investment leading to high and sustained improvements in school-level funding. Thus a key element of pilot monitoring, evaluation, and annual improvements to the design  Review of Pre-Tests 53 will be how to minimize overhead costs so that 95 percent or more of the funds actually reach the school and the girls. GIRLS' EDUCATION SUPPORT PROGRAM 5. In addition, Professors S. Sumra, G. Malekela, A. Kilindo, and A. Dachi (1997) reviewed the Girls' Secondary Education Support Pre-Test. Generally this review is positive on the selection mechanism (it seems to work well in selected the best poor girls with few leakages to better-off families) and the girls' ability to succeed in secondary school. The report raises the following concerns, however: a. The marketing of the project in 1995 was done hurriedly. Not many of the girls who were eventually selected knew that they were being considered for the bursary. The overall result was that many of the bursary recipients were late in joining their secondary schools. There were also problems stemming from the selection process. In the second year, the program had better marketing, but more complex and confusing selection procedures were used. b. In a number of schools, the girls were not provided with all the items they were intended to receive, e.g., uniforms, textbooks, or money for transportation. It was unclear to parents and students what expenses would be covered by the bursary. Closer follow-up of the recipient girls would help address these problems. C. Parents and girls were not aware of the quality of secondary school in their zone and usually chose schools which were near their homes. A greater effort to convey the range of choices open to these families would be helpful. d. The performance of bursary girls was comparable to others in school. However, one problem that most face is that they join secondary schools with low PSLE scores. Unless they get help to overcome this disadvantage, it is unlikely that they will progress enough to proceed beyond lower secondary school. e. Girls studying at day schools face particular problems. Some stay with their relatives and some live in rented rooms. Some of those who live with their relatives get treated as house girls and get burdened with domestic chores. Those who live in rented rooms live in poor  54 Annex C conditions, without electricity or running water and have to face unwanted attention from men. Some girls had to walk long distances to school. There is a need for the project to look at ways of providing hostel facilities for these girls. f. Although the number was not large, there were cases of drop outs, most due to pregnancy. These girls, especially those away from their parents, get little or no guidance or counseling. There is a need for the project to ensure that all the girls have someone in the school to whom they can turn for help when in trouble.  Annex D: Economic Assessment BENEFIT-COST ANALYSIS: INTERNAL RATES OF RETURN Overview of the Range of Returns for the Whole Project 1. If the project's full costs are considered (including Annex Table D. 1. Estimated Internal capacity building, policy Rate of Return (Percent) development, planning, research, Cost Basis CEF GSES Total and the opportunity cost to C Program OPPOtY COS 0 ~~Full Project Css 19 6 12 households of additional Direct Costs Only* 48 11 26 children attending school) the * Household CEF Contributions, CEF Grants, and internal rate of return (IRR) for GSES Bursaries the CEF component is conservatively estimated at 19 percent; and for the GSES, 6 percent. The return for the project as a whole is estimated at 12 percent. If only the direct costs of parental contributions to the CEF, project contributions of matching funds for the CEF, and project funding of bursaries for the GSES are considered, the IRR for the Community Education Fund is estimated at 48 percent; and for the GSES, at 11 percent. With this lower cost structure, the return for the project as a whole is estimated at 26 percent (see Annex Table D.1). 2. Because this benefit-cost analysis assigns no independent benefits to the Capacity Building or Policy Development, Planning, and Research components, treating them only as costs or supporting activities of the CEF and GSES, the IRRs in Annex Table D. 1 should be considered lower-bound estimates. Unless the unquantified benefits for these other components are zero or negative, the returns to the project can be assumed to be substantially higher than estimated. Benefits 3. The IRR estimates for both programs are based on pecuniary, private benefits to individuals from increased schooling and increased school quality, and a conservatively estimated monetary equivalent of non-pecuniary benefits (such as reduced fertility among girls with more education, and improved schooling and health prospects for the children of primary and secondary school leavers). The benefits used in the analysis and the estimated value per pupil are shown in Annex Table D.2). 4. This analysis ignores the general positive externalities of a better educated population and the positive income distribution implications of the project. A disproportionately large proportion of the new children likely to enroll in primary school as a consequence of the CEF will come from low- income, disadvantaged backgrounds. All of the estimated 3,415 girls  56 Annex D participating in the GSES will come from households in the lower one-half of Annex Table D.2. Pecuniary and Non-Pecuniary Project Benefits for HRDP with Estimates of Monetary Value Component Value of Time Notes Undiscounted Unit Benefit, US$ Equivalent CEF: increment in annual 76 Per Estimated difference in income income from completing work between someone with no primary school year education and a primary school .. .... ... ............ 9raduate CEF: increment in annual 38 Per Assumed to be 50 percent of the income for not dropping out of work wage benefit of attending and school year completing primary school CEF: income in the extra year 270 One The average wage of a primary a person is able to work year school graduate minus the because of not repeating a average wage of a 13 to 15 year year of school old child attending school .. ...................................................... ..................................................................................... ...................... ........... .......... ............................................. . ........................... C^"EF: increase in income from 8 Per Assumed to be 10% of the improved primary school work income benefit that a person quality year receives from completing primary school CEF: Social benefit of a girl 231 Per One half the vaiue of the soc.al attending 7 years of school lifetime benefits of education in Kenya estimated in Summers (1994) GSES: Increment in annual 160 Per The d.iference in wages between income from attending and work a woman with a primary completing secondary school year education and a woman with a full lower-secondary education. G-SES i;ncrement in annual 91 Per Thedifferenceinwages etwe income from not dropping out work a Form II and a Form IV of lower secondary school year graduate. ...................................1