Senegal - Pilot Female Literacy Project

Document of The World Bank Report No. 15517-SE STAFF APPRAISAL REPORT REPUBLIC OF SENEGAL PILOT FEMALE LITERACY PROJECT May 9, 1996 Population and Human Resources Operations Division Western Africa Department Africa Region  CURRENCY EQUIVALENT Currency Unit = CFA Franc (CFAF)/ US$ 1 = 500 CFAF ABBREVIATIONS AND ACRONYMS AGETIP Agence d'execution des travaux d'interet public contre le sous emploi CCS Cellule de coordination et de suivi CIDA Canadian International Development Agency CNOA Coordination national des operateurs en alphabMtisation CROA Coordination regional des operateurs en alphabtsation GDP Gross Domestic Product GTZ German Agency for Technical Cooperation KfW Kreditanstalt fur Wiederaujbau IDA Intemational Development Association MCALN Ministere charge de I 'alphabetisation et des langues nationales NGO Non-governmental Organization OPEC Organization of Petroleum Producing Countries PAIS Projet d 'alphabetisation intensive au Senegal PPF Project Preparation Facility SOE Statement of Expenditures UNESCO United Nations Educational, Scientific and Cultural Organization UNICEF United Nations Children's Fund USAID United States Agency for International Development FISCAL YEAR ACADEMIC YEAR January I-December 31 October-June  REPUBLIC OF SENEGAL PILOT FEMALE LITERACY PROJECT TABLE OF CONTENTS CREDIT AND PROJECT SUMMARY ..................................................................i I. INTRODUCTION ..................................................................1 II. THE CONTEXT ..................................................................1 A. Development Context ..................................................................1 B. Sectoral Context ..................................................................3 1. Illiteracv Among Women ..................................................................3 2. Provision of Literacv Programs ..................................................................3 3. Government Strategy and Action Plan ..................................................................5 4. Issues in literacv ..................................................................6 5. IDA's Strategy for Literacv im Senegal .................................................................. 7 6. Lessons Learned ..................................................................8 a. IDA-Financed Education Projects in Senegal .................................................................8 b. Bank-Financed Literacy Projects or Components ...........................................................8 c. IDA's Experience with AGETIP ..................................................................9 III. THE PROJECT .................................................................. 0 A. Scope and Objectives ................................................................. 10 1. Literacy and Post-Literac Sub-Projects ................................................................. 10 2. Project Management and Capacity Building in Program Deliverv, Management, and Evaluation ................................................................. 12 B. Program and Project Cost and Financing Plan ................................................................. 12 iV. PROJECT IMPLEMENTATION ................................................................. 13 A. Status of Project Preparation . ................................................................. 3 B. Project Coordination and Management ................................................................. 14 C. Monitoring and Evaluation ................................................................. 14 D. Procurement ................................................................. 15 E. Disbursements ................................................................. 17 F. Accounting, Auditing and Reporting ........................... ...................................... 18 V. PROJECT BENEFITS AND RISKS ................................................................. 18 VI. AGREEMENTS TO BE REACHED AND RECOMMENDATIONS ................................. 19 This report is based on the findings of an appraisal mission conducted in December 1995 and a donor's meeting held in March 1996. The appraisal mission included Rosemary Bellew (Mission Leader), Serge Theunynck (Consultant, Implementation Specialist), David Kahler (Consultant, Evaluation Specialist) and Alassane Diawara (Operations Officer, Senegal Resident Mission). Amy Ba is the staff assistant for the project. Also involved in project preparation were William Cooper (Consultant), Wolfgang Chadab (Disbursement Officer), and Bernard Abeille (Implementation Specialist). Djamalddine Rouag, Robert Prouty, and Carmen Perez Castafio (AF5PH Education team) also provided input during the review process. Birger Fredriksen and Ok Pannenborg are the Acting Department Director and managing Division Chief, respectively, for the operation. Tom Eisemon and Nat Colletta are the peer reviewers.  TABLE OF CONTENTS (Cont'd) ANNEXES 1 - Basic Data 2 - Project Benefits 3 - Draft Policy Paper and Action Plan for the Literacv Sector 4 - Sub-Project Eligibility Criteria 5 - Estimated Cost Structure of Sub-Projects 6 - Project Cost and Financing 7 - Stakeholder Participation 8 - Evaluation Results of the Test Phase 9 - Assessment of Management Options 10 - Project and Sector Monitoring Indicators 11 - Work Plan and Procurement Schedule 12 - Selected Documents Available in the Project File 13 - Disbursement Profile MAP: IBRD 27781  i REPUBLIC OF SENEGAL PILOT FEMALE LITERACY PROJECT CREDIT AND PROJECT SUMMARY Borrower: Republic of Senegal Beneficiary: Ministry in Charge of Literacv and National Languages (MCALN) Credit Amount: SDR 8.7 million (US$ 12.6 million equivalent) Poverty: Program of targeted intervention Terms: Standard IDA terms, with 40 years maturitv. Program Objectives: The project would support the first phase of the Government's 10-year program which aims to reduce the illiteracy rate of the age group 10-39 years old from an estimated 50% in 1995 to 25% bv the vear 2005, and from 66% to 30% for women. The main objective of the first phase is to pilot a strategy which supports non-governmental literacy providers to expand their programs with the aim of lowering the illiteracy rate to about 40% overall and to 47% for women. This would require reaching about 300,000 beneficiaries, 75% women. Other objectives include: (i) ensuring the sustainabilitv of literacy skills, and (ii) strengthening the capacity of the private sector to deliver effective programs, and the capacity of the public sector to monitor, evaluate and coordinate sector investments. Project Description: IDA financing would support literacv and post-literacy sub-projects in the regions of Diourbel, Louga, Kolda, Fatick, and in peri-urban areas of Dakar to reach about 135,000 adults (75% women) over the five-year period 1996- 2001 -- or about 75% of the total required to meet the Government's targets for those regions. Parallel financing for about 75,000 beneficiaries in Tambacounda, Kaolack, St. Louis, Ziguinchor, and Niayes would be provided by CIDA, for 14,000 beneficiaries in Fatick and Kaolack bv the GTZ, and for 50,000 beneficiaries through on-going programs. The proposed project would also strengthen the capacity of providers in program development and delivery, and the capacitv of Government, providers and AGETIP in program monitoring, supervision, evaluation and coordination. Benefits: Given trends in primarv school expansion, if no adult education programs are provided, by the year 2015, only about 45% of young women in their peak years of childbearing and child-rearing will be literate, and much fewer in rural areas. The consequences are striking for productivitv, health, fertilitv, child survival and education. A recent anal!ysis from Senegal shows that children of illiterate mothers are 50% more likelv to die in childhood. illiterate women want 1.8 more children on the same income as literate mothers: and literate mothers are about 50% more likelv to send their  ii daughters to school, ceteris paribus. Under these assumptions, the expected health benefits of the literacy program are sufficiently high to make the program attractive on health grounds alone. The program is estimated to result in a 10% reduction in disease burden, equivalent to 8.5 million (disability adjusted) life years saved, at a cost of about US$ 2.00 per capita. In comparison, the most cost-effective health intervention -- a minimum package of clinical services -- at an expenditure of US$ 7.80 per capita is estimated to reduce the disease burden by 24% in low-income countries. The program is also expected to result in an estimated 10-14% decline in infant mortality, a 3-5% decline in the number of births, and a 23% increase in girl's enrollment. Risks: This is a pilot that tests a new approach to delivering literacy programs. The experimental phase, which began in four regions in August 1995, provided useful feedback on procedures and processes which were incorporated into the project's design. The risk that a larger pilot may not achieve its objectives are difficult to assess at the outset for three main reasons. First although literacy providers show significant interest in the program, capacity is widely believed to be weak among small providers. To minimize this risk, the project will provide capacity building services to assist the weaker providers and increase absorptive capacity. The second risk concerns the ability of Government and providers to maintain an effective partnership. The alliance established is innovative and recent. To be successful, providers will need to place less emphasis on protecting their turf and more on strengthening their technical capacity. And, the MCALN will need to guard against exercising undue control over providers, particularly in areas where they are more experienced. To assist the process, the project supports regular discussions between the MCALN and providers on program implementation and results, and training for both to improve technical competence. A third potential risk concems the sustainability of literacy skills. Some researchers argue that relapse into illiteracy may occur when the environment offers no support for the new literate to apply and improve her skills. However, there is no strong evidence to reject the hypothesis that when programs are well designed, and offer a minimum instruction of 300 hours, neo-literates retain their skills over time. To address this issue, the project will conduct longitudinal studies to assess skill retention and operational research to assess the cost-effectiveness of select post-literacy activities. Environment: This is a category C project. It is expected to have a beneficial environmental impact through improved knowledge transmission among women, essential stakeholders for environmental management.  iii ESTIMATED PROJECT COSTS AND FINANCING PLAN' (US$ million)2 ESTIMATED PROJECT COSTS LOCAL FOREIGN TOTAL (USS million) 1. Literacy Sub-Projects 10.3 0.0 10.3 2. Strengthen Program Delivery, Management, 1.9 0.9 2.8 Monitoring and Evaluation 3. PPF refinancing 0.6 0.2 0.8 Total Base Costs 12.7 1.1 13.8 Physical Contingencies 0.0 0.0 0.0 Price Contingencies 0.1 0.1 0.2 TOTAL PROJECT COSTS 12.8 1.2 14.0 Financing Plan (US$ million) IDA 12.6 Government 0.5 Beneficiaries and NGOs 0.9 TOTAL 14.0 Estimated IDA Disbursements: US$ Millions IDA FY 1997 1998 1999 2000 2001 2002 Annual 0.5 1.9 2.4 2.8 2.7 2.3 Cumulative 0.5 2.4 4.8 7.6 10.3 12.6 ' The estimated cost of the first phase of the program is USS30.0 million. IDA would provide USS10. I million, CIDA USS12.6 million (escluding bilateral technical assistance), the GTZ US$1.4 million, NGOs and beneficiaries US$4.4 million, and Government USSI.4 million leaving a financing gap of about USS4.0 million which is under discussion with other donors. Financing details for the full program, are in Annex 3. 2 Totals may not add up due to rounding.   REPUBLIC OF SENEGAL PILOT FEMALE LITERACY PROJECT I. INTRODUCTION 1.1 The Government of Senegal has requested IDA's assistance to support the first phase of a 10-year program which aims to reduce the illiteracy rate of the age group 10-39 years old from an estimated 50% in 1995 to 25% by the year 2005, and from 66% to 30% for women. To achieve this target for women, about 50,000 a year will need to be reached by adult education programs. The program has broad based support, having been developed through a series of consultations between the Ministry of National Education in charge of Basic Education and National Languages (MCALN) and a broad spectrum of representatives of Government, and literacy providers, including NGOs, language associations, academic institutions, enterprises and the more loosely knit women's and economic interest groups. 1.2 The main objective of the first phase (1995-2000) is to pilot nationwide a strategy which supports non-governmental literacy providers to expand their programs with the aim of lowering the illiteracy rate to about 40% overall and to 47% for women. This would require reaching about 300,000 beneficiaries, 75% women. Other objectives include ensuring the sustainability of literacy skills and strengthening the capacity of the private sector to deliver effective programs, and the capacity of the public sector to monitor, evaluate and coordinate sector investments. 1.3 The proposed project provides partial financing for this program Total program costs of the first phase for the country-wide pilot are estimated at about US$ 30 million, with IDA providing US$ 12.6. IDA financing would support the provision of literacy and post-literacy programs in the regions of Diourbel, Louga, Kolda, Fatick, and in peri-urban areas of Dakar for about 135,000 beneficiaries, as well as the capacity building objectives of the program. Total project costs are estimated at US$ 14.0 million with a foreign exchange component of US$ 1.2 million. Communities and/or NGOs would contribute about US$ 0.9 million equivalent, and the Government US$ 0.5 million. For the rest of the program, parallel financing is expected to be provided by Canada (CIDA, US$ 10.0 million equivalent, excluding bilateral technical assistance), Germany (GTZ, US$ 1.4 million equivalent), and NGOs and beneficiaries (US$ 2.9 equivalent). (See Annex 3 for the financing plan of the full program.) II. THE CONTEXT A. Development Context 2.1 Poverty. Senegal is a coastal semi-arid country situated in Africa's Sahelian zone with a per capita income of US$ 611 (1995). Its population of 8.3 million has been growing at a rate substantially above the real GDP growth rate (2.7% vs. 1.9%, 1985-95). The rnajority of the population (58%) lives in rural areas and half is under the age of 16. Despite a relatively high level of income for the Sahel region, poverty is widespread in rural areas and in Dakar. During the most prosperous time of the year for farmers, in 1992, 33% of the population did not spend enough on food to assure a minimum caloric intake. Poverty is highest in the regions of Fatick, Kaolack, Kolda, Louga, Tambacounda and Ziguinchor were the population receives a comparatively lower  2 level of public transfers (such as scholarships, retirement benefits, family allowances, etc.). Together, these regions account for two-thirds of the total poor population in Senegal. The high levels of poverty are linked to a low level of economic growth and skewed distribution of income (59% of income going to 20% of the population), human capital and access to public services. 2.2 Social indicators. Senegal has significantly lower indicators of development than what would be predicted on the basis of its per capita income. Child mortality averages 120 compared with 62 for low-middle income economies; fertility averages 6.0 compared with 2.9; and, primary education enrolls the equivalent of only 56% of school-aged population compared with 104% for lower-middle income economies and 68% for Sub-Saharan Africa. Widespread illiteracy among women is a significant factor. 2.3 Economic Growth. Between 1988-1991, real GDP per capita grew at an average annual rate of -2.0%, and key economic sectors deteriorated. Since 1991, fish processing declined in volume by 66%, food processing by 46%, phosphates by 24%, and textiles by 37%, while groundnut exports declined in value by 41%. As the tax base shrank, Government increased its reliance on exceptional and ad-hoc revenue measures which increased the costs of production and forced firms to close or join the informal sector. The fiscal deficit and balance of payments worsened; substantial domestic and external arrears accumulated; and foreign reserves declined sharply. To halt the deterioration, Senegal adopted several measures in 1993 to deepen deflationary policies and devaluated the CFAF. Since then, tourism, cement and textile production, and exports of phosphoric acid and fish show signs of recovery. However on the fiscal side, shortfalls in Government receipts are chronic, particularly from import duties on non-essential consumer goods. 2.4 The Country Assistance Strategy aims to assist Senegal to reduce poverty by restoring growth in per capita income and fostering a more equitable distribution of the benefits of growth. To relaunch economic growth, the strategy focuses on increasing opportunities for private sector expansion and improving productivity in agriculture, particularly in the production of traditional export, food, and non-traditional and high value-added crops. To support this growth, two elements are essential: (i) a more efficient and cost-effective provision of public goods such as power, water and transportation, and (ii) an improvement in the human capital base. To foster a more equitable distribution of benefits, the strategy focuses on improving public resource mobilization and the efficiency and cost-effectiveness of public expenditure, particularly in basic education and health, targeting girls, women and the poor to enable them to contribute to and benefit from economic growth.  3 B. Sectoral Context 1. Illiteracy Among Women 2.5 The majority (80%) of young rural Senegalese women aged 15-39, women in their peak years of productivity, childbearing and motherhood, are illiterate, with rates as high as 80- 90% in the regions of Diourbel, Tambacounda, Kolda and Louga, and 65-75% in Kaolack, Fatick, St. Louis and Thies. Comparable rates for their urban counterparts and men average between 40% and 45%, respectively. In the regions of Dakar and Ziguinchor, where gross primary enrollment rates are at or near 100%, illiteracy rates are significantly lower at 28% for men and 41% for women. The consequences of high illiteracy among women are striking. A recent analysis from Senegal shows that children of illiterate mothers are 50% more likely to die in childhood; illiterate women want 1.8 more children on the same income as literate mothers; and all other things being equal (income, geographic location, ethnicity, etc.), literate mothers are about 50% more likely to send their daughters to school than illiterate mothers (Annex 2). 2.6 Widespread illiteracy among women is directly related to an historical under- provision of primary education, to financial, cultural, and social constraints that limit girls' school enrollment, and limited education opportunities for teens or adults who missed the opportunity to attend formal school. To expand access to primary education, IDA, the KfW, France and Japan are supporting a five-year program to increase the (gross) primary enrollment ratio to at least 65% by the year 1998 (73% for boys and 57% for girls), mainly by expanding access in rural areas, improving quality and implementing a program to increase girls' enrollment. This effort alone, however, will be insufficient to significantly lower illiteracy among young women for at least another generation. Projections suggest that even assuming the objectives for primary education are attained, in the year 2015, still about 45% of teens and young women (aged 15-39) would be literate, and even less in rural areas. 2.7 The proposed project complements and the efforts to expand basic education by supporting the expansion of literacy programs for teenage girls and young women who dropped out, missed the opportunity to attend, or are not yet reached by formal schooling. The program is consistent with poverty reduction objectives and with the Country Assistance Strategy which was discussed with the Board in January 1995. It targets girls and women in the poorest parts of the country where the impact on social development indicators is likely to be greatest. 2. Provision of Literacy Programs 2.8 Demand. Evidence suggests there is strong demand for literacy training among women. In 1994, women comprised between 75%/o-90% of enrollment in literacy programs, depending on the region. A recent survey of women in seven regions (St. Louis, Thies, Diourbel, Tambacounda, Kolda, Louga, Fatick and Kaolack) shows that 35% of women had participated or were participating in a literacy program. About 75% of other women report that they would like to participate (MCALN, 1994). The majority do not attend due to the absence of a program in the community or to a lack of information on the availability of programs. 2.9 Supplv. Literacy programs in Senegal have historically been organized and provided by a diverse group of NGOs, language associations, enterprises, and youth, women's and  4 producer's groups, using their own resources and cost-sharing arrangements with beneficiaries. According to a comprehensive survey of literacy providers conducted in 1994, some 200 registered NGOs, and a larger number of associations claimed to offer some form of literacy education. In practice, about 100,000 individuals were attending classes offered by some 94 organizations (MCALN 1995). There is, however, a strong case to be made for public financing of literacy programs -- the social rate of return is substantial (see Annex 2); the resources of non- governmental organizations are insufficient to significantly expand their operations; and, it is unlikely that women can or would pay the full cost of the program. Based on experience, literacy providers in Senegal estimate the maximum average ability to pay at the equivalent of $5.00 (or about 13% of average unit costs). 2.10 Program content and duration. Literacy programs are offered in French (mainly in urban centers) and in national languages, in particular in Wolof, Pulaar, Sereer, Joola, Mandinka and Soninke. They vary widely in program content and structure depending on the provider's definition of literacy, its experience, and the needs of the target group. Most providers offer programs that can be referred to asffunctional literacy (reading, writing and arithmetic along with vocational skills), integrated literacy (comprehensive education that includes problem solving skills with the goal of making participants leaders in their communities), or adult basic education (basic literacy skill training combined with a general education in health, nutrition, farming, and other areas of interest to rural women). The latter may be independent or linked to development projects. (Refer to Annex 2 for definitions). The majority of programs offer between 200-300 hours of instruction spread over 12-18 months (i.e. 2 hours a day, 3 to 4 days a week). 2.11 Between 1993-1995, the Government, with CIDA assistance, tested a mass literacy campaign (Projet 1000 classes), which provided exposure to reading, writing and arithmetic during 140 hours of instruction spread out over four months, the main objectives being to sensitize the population to the importance of literacy and to learn more about what conditions contribute to a successful program. Recognizing that the program was insufficient to cultivate basic skills, a second phase of 150 hours of instruction with a particular focus on arithmetic was added (Projet PAIS), and a year later an additional 150 hours was added for instruction in thematic areas such as health, nutrition, environmental protection, etc. Evaluations of all phases of the campaign showed predictable weaknesses related to the acquisition and distribution of instructional materials, inadequate instructor training and on-the job support, and questionable learning outcomes. The evaluation results were fed-back into the program's design which allowed the MCALN to improve upon itsfaire-faire strategy and to delegate implementation of the third phase of the campaign to experienced literacy providers. 2.12 Characteristics of effective programs. Evaluative information is available only on a few progranms in Senegal. Only one provides reasonable measures of learning achievement, and none provide information on key variables that are likely to be related to program effectiveness. Nevertheless, this information supports the conclusion from international research on literacy that the following elements are characteristic of effective programs: (i) minimum of 250-300 hours of instruction which is necessary to attain a level of literacy (fourth grade equivalent) where skills are actually acqmured and, therefore, can be maintained, (u) meaningful content for adult learners, (iu) materials appropriate for adults, (iv) teachers knowledgeable about the content and trained in adult education approaches, (v) regular in-service teacher support, and (vi) participation of beneficiaries in progran design and implementation. A review of the literature is in Annex 2.  5 2.13 Program costs. Actual costs of on-going programs vary across providers, depending on program duration, level of fees paid to instructors, how much materials development is involved, the quantity of program materials, and the extent of supervision, in-service and pre- service instructor training. The information suggests that the average cost per participant for a 300-hour program averages US$ 65 (32,500 CFAF), ranging between US$ 48 (23,000 CFAF) and US$ 125 (62,500 CFAF) per participant when materials development and testing is involved. Drop out rates from adult literacy programs are high internationally with 30-40% being considered good. Drop-out rates reported by literacy providers in Senegal, like programs everywhere, vary between 100/6-60%. Programs with the lowest drop-out rates are generally those that contain the elements of effective programs (para. 2.11) and are also, therefore, more costly. 3. Government Strategy and Action Plan 2.14 Prior to 1991, the Ministry of Education was responsible for overseeing literacy activities, and had traditionally limited its role to providing technical support to donor-financed projects. In 1991, a Ministry responsible for literacy was created which later became the Ministry of Basic Education and National Languages (MCALN). The MCALN is responsible for developing public policy and publicly financed programs, mobilizing resources, coordinating donor support, and planning, programming, monitoring, and evaluating the effectiveness of public investments in the sector. 2.15 The MCALN's program for literacy (Annex 3) was developed in 1994 through a series of consultations with a broad spectrum of Government agencies, donors and literacy providers. Its priorities for literacy are to: (i) substantially reduce illiteracy among females, (ii) reduce the literacy gap between regions and between urban and rural areas, (iii) promote the use of national languages in functional and basic education programs to improve learning outcomes, (iv) promote a literate environment, and (v) coordinate the interventions of financial partners to work towards these objectives. Specific literacy objectives for the decade 1995-2005 are to reduce the illiteracy rate of the population 10-39 years old from an estimated 50% in 1995 to 25% by the year 2005, and from 66% to 30% for women. Intermediate objectives for the year 2000 are 38% overall and 46% for women. Although most of the improvement will be due to the expansion of primary education, about 50,000 women and 10,000 men will need to be reached each year by other education programs. Other objectives of the program are to: (i) improve the effectiveness of literacy programs; (ii) ensure the sustainability of literacy skills; and (iii) improve monitoring, evaluation and coordiation of sector activities. 2.16 The main strategies by which Government aims to achieve the targeted literacy rates are: (i) supporting a strategy of "fairefaire" by which Government channels financing to qualified local organizations to develop and deliver literacy programs, and (iu) developing a participatory approach involving communities and beneficiaries at all stages (design, implementation and evaluation). This strategy has so far resulted in a consensual policy framework and action plan for the sector, the creation of a coordinating mechanism among litacy providers, and a consultative mochanim linking government and providers to review and revise policies and programs.  6 4. Issues in literacy 2.17 The main questions surrounding literacy programs in general, as well as those in Senegal, are: (i) What is the best strategy to increase the availability of programs for women? (ii) Is the absorptive and technical capacity of literacy providers sufficient to provide the services desired? (iii) Can the public sector forge a sustainable partnership with literacy providers to enable this to happen? (iv) How can efficiency and quality be ensured in program delivery? And, (v) what are the odds that literacy skills will be retained once acquired? 2.18 Strategv to increase availability of programs. The availability of effective literacy programs, responsive to the needs of women, is a main constraint to increasing their level of literacy. How to improve the supply has been the subject of much discussion in recent years among donors, Government and literacy providers. Although there are no clear answers to this question in the absence of experimentation, experience in Senegal and elsewhere with government delivered programs has led all parties to conclude that a different approach was needed. A main objective of the proposed pilot project is to test a mechanism to operationalize a strategy of faire- faire par les partenaires sur le terrain, in other words, to support private sector literacy providers to expand their operations and encourage new providers to enter the market. 2.19 The absorptive and technical capacity of providers to expand their operations has proven difficult to assess through surveys and interviews. With the exception of the larger providers (e.g. Tostan, Aide-et-Action, SODIFITEX) financed by bilateral or multilateral donors, most providers operate on small budgets with few specialist staff and offer small-scale programs, targeting narrowly defined groups in specific geographic areas. Not much is known about how effective the programs are which smaller providers offer, nor if smaller providers have the professional capacity to develop and manage larger programs. Many providers maintain, however, that care should be taken to ensure technical support to smaller providers, particularly in the areas of instructor training, adult education methods, and instructional materials. 2.20 Maintaining an effective partnership. The faire-faire strategy is an appropriate one in the Senegalese context. However, its effectiveness depends importantly on maintaining an effective partnership between Government and suppliers. To develop this partnership, in 1993, the MCALN began consultations with organizations interested in literacy, with the aim of obtaining feedback on its development plan and defining modes of collaboration (See Annex 71. Trust and interest on the part of providers to participate in a government program have improved. In 1995, providers formed a professional association to represent them in consultations with Government and to provide technical support services for members to maintain standards while programs expand. Suregtheng and sustaining this partnership will be critical for the program to achieve its objectives. 2.21 Quality of program delivery. The absence of evaluations of the quality and effectiveness of on-going literacy programs could raise questions concerning the pertinence and the effectiveness of instruction and of instructional materials. Concerning program effectiveness, a system to qualify providers has been put in place. However, instructional materials deserve particular attention. A multiplicity of literacy publications exist in the six national languages. Although there is little difference among them in content, methodology and presentation, in 1994, at least 24 literacy operators were publishing their own materials. The reasons they do include: (i)  7 content in available materials is inconsistent with their training methodology or target group, particularly in the areas of health, nutrition, family planning, environment and women's rights, (ii) unawareness of the existence of other materials, and (iii) difficulty of acquiring materials even if they are known to exist. The latter two problems result mainly from insufficient operating capital for materials production and storage, and from insufficient publishing skills in planning, editing, design, illustration and economical production methods. As a result, publications are not advertised; printing costs are relatively high because print runs are small (1,000 copies or less); and material is poorly manufactured and has a short useful life. 2.22 Sustainability of Literacy Skills. A criticism often levied against literacy programs, as well as primary education in general, is that even if literacy programs are well designed and delivered, the beneficiaries will not retain their skills over time, particularly if the immediate environment offers no support for the new literate to apply and improve her/his skills. Other professionals, however, refute this hypothesis and argue that if basic literacy programs are of sufficient duration and use well-designed learning materials with content that is interesting and useful to adults, adults do acquire and retain the skills. Still others maintain that, even though skill retention is improved when literacy programs are well-designed, post-literacy activities (such as village libraries, local language newspapers, structured learning occasions, or the creation of locally-managed structures such as a cooperative run by new literates) to which reading materials can be attached should form part of the literacy program. 2.23 There have been no studies in Senegal on skill retention nor on the effectiveness of post-literacy activities, and very few in other countries, to strongly support or reject any of these hypotheses. The most that can be said is that: (i) there is no evidence to reject the hypodhesis that when programs are well-designed and delivered and offer an adequate duration of instruction (250- 300 hours) young adults retain their skills, and (ii) there is no evidence that post-literacy activities have been effective. However, there is stronger evidence to support the conclusion that: (i) the longer the period of instruction the higher the level of skill development and retention, and (ii) of all post-literacy activities, newspapers, particularly when reinforced by radio broadcasts, seem to be most effective. (See Annex 2 . 5. IDA's Strategy for Literacy in Senegal 2.24 The Bank's strategy for increasing the level of literacy in Senegal has focused exclusively on providing support for the expansion and improvement of primary education since it is widely believed that this is the most cost-effective strategy for developing the basic human capital required for development. There are at least four reasons for this view. First, it is supported by the experience of mass literacy campaigns and functional literacy programs popularized by UNESCO during the 1960s and 1970s which, by and large, were unsuccessful. Although literacy programs have continued to develop and expand, they still have not recovered from the bad reputation acquired from this time. Second, governments in Africa already have great difficulty organizing and delivering primary education. Support for literacy programs risks distracting Government from fulfiling this obligation. Third, literacy programs do not require significant investment, and they are difficult for Government to provide because they are usually small-scale operations that need to be responsive to needs of diverse groups. Finally, the Bank has little experience with literacy programs. 2.25 Since 1991, the Senegalese authorities have persistently requestd IDA support for a literacy program. However, for all the above reasons, particularly the specter of Government  8 becoming more involved in providing literacv programs when its capacity was already weak to fulfill its obligations of providing primary education, the Bank was reluctant to move ahead. Additional concerns were the limited experience of the MCALN in the area of literacy, and that increased Government involvement in the sub-sector would turn towards program delivery and regulation of private providers. Therefore, at the outset, IDA-Government discussions on this proposed project centered on defining the meaning of its faire-faire strategy, the appropriate roles for Government and literacv providers, and on an experimental approach to expand literacy programs. 6. Lessons Learned a. IDA-Financed Education Projects in Senegal 2.26 Since IDA's first project in the education sector in 1971, six education and training projects have been approved for a total of about US$ 87.0 million. Three projects supported vocational and technical education (US$ 39.0 million); one supported the establishment of a regional management training center (US$ 5.5 million); one covered a variety of objectives in nearly all the sub-sectors of the system (US$ 12.0 million); and the on-going Second Human Resources Development Project (US$ 30.0 million, 1993) focuses on expanding access and improving the quality of primary education. 2.27 Project performance in the education sector with respect to the Operations Evaluation Department's categories of outcomes, sustainability and institutional development, is poor to mediocre. Across all projects, key constraints have been: (i) lack of counterpart funds or insufficient operating costs for institutions; (ii) poor management of the sector; (iii) weak Government commitment to change certain policies, particularly in higher education; (iv) poor management and use of technical assistance; and (v) complexity of the operations with too many sub-sectors or institutions involved. However, some gains in capacity building and policy reform were achieved in primary education, and more recently in higher education. Following the concems raised about project complexity and weak sector management, the Bank's assistance strategy to the sector was to address the needs of each sub-sector in separate operations but within the context of an overall policy and financial framework. Operations were to be phased in over a period of 3-5 years, beginning with primnary education (Second Human Resources Development Project, Cr. 2473-SE), followed by higher education (FY 96), literacy, and secondary general and vocational and technical education. Combined, the operations support a policy framework which places priority on the expansion of primarv education, and the improvement of education at all levels, as well as on providing non-formal education for young women who missed the opportumity to attend formal schooling. b. Bank-Financed Literacy Projects or Components 2.28 The Bank does not have a long history of supporting literacy programs. With the notable exceptions of Indonesia (now on its fourth non-formal education project) and Ghana, where the Bank has financed free-standing literacy operations, small literacy programs have been financed mainly through agricultural (Mali, Senegal, Burkina Faso), Women in Development (The Gambia, C6te d'Ivoire) or basic education projects (Philippines, Burundi, Mali). In all cases, Govemment delivers the programs. A 1987 review of Bank-financed non-formal education  9 components3 concluded that the following aspects of non-formal activities need careful attention: (i) active participation of the borrower in preparation to ensure commitment; (ii) needs analyses of potential clients for large-scale non-formal education programs; (iii) close attention to management capacity and clear implementation arrangements, (iv) measures for effective monitoring and evaluation; (v) adequate resources for materials; and (vi) sufficient flexibility to respond to different needs. 2.29 Project completion, audit and supervision reports of later projects draw attention to the following common problems with literacy programs: (i) significant organizational problems in the provision of program inputs (materials, teacher training); (ii) inadequate advertisement of the program; and (iii) weak capacity of Government staff. In African countries in particular, such as Mali and Burundi, organizational problems were severe. Smaller-scale programs, however, in Senegal and Burkina Faso that targeted organized groups (extension workers for example) were more successful. c. IDA's Experience with AGETIP 2.30 The Agence d'excution des travaux d'interet public contre le sous emploi (AGETIP) was created by the Government in 1989. It has the status of an NGO and is governed by a Board of Directors consisting of a Director from the private sector, representatives of the National Council of Employers of Senegal, the Union of Construction and Public Works Workers, and the Association of Mayors of Senegal. It operations are govemed by a procedural manual which forms part of the Govermnent of Senegal-AGETIP agreement which established the Agency. 2.31 Since 1989, AGETIP has been involved in the management of several IDA financed projects, most notably two Public Works and Employment Projects, the Community Nutrition Project, and the classroom construction component of the Second Human Resources Development Project. The Public Works projects aimed to create employment, strengthen the capacity of small construction firms to respond to increased demand, and test procedures to enable Government to commission small construction projects. A large number of small contracts, and weak financial and managerial capacities of small enterprises in need of timely payments are main features of the projects. AGETIP has successfully carried out these projects and has developed a competence for efficient management of a large number of small contracts, effective monitoring, and rapid payments while keeping administrative costs at or below 5% of program costs. 2.32 Given this experience, social sector Ministries also became interested in using AGETIP's management experience. Since 1994, the Ministry of Health has used AGETIP in the preparation and execution of the Community Nutrition Project. Due to its demonstat expertise with transparent and efficient procedures for the evaluation of sub-projects and the management of numerous sub-project conacts, in 1995, the MCALN involved AGETIP in the preparation of the Implementation Manual for the proposed project and in the management of its experimental phase. AGETIP's performance thus far has been effective, efficient and professional. In April 1995, AGETIP's convention with the Govermnent of Senegal was modified to allow it to offer this assistance to support the delivery of basic public services in nutrition, health, education and literacy. I Ralph Romain and Lenor Arm sog. Review of World Bank Operamtio in Non-Fonial Education and Trning Education and Training Department Disussion Paper, Report No. EDT 63, 1987.  10 III. THE PROJECT A. Scope and Objectives 3.1 The MCALN's program for literacy is a 10-year program which aims to reduce the illiteracy rate of the age group 10-39 years old from an estimated 55% in 1995 to 25% by the year 2005, and from 66% to 30% for women. The program is divided into two phases. The first phase tests a strategy offaire-faire par les oprateurs throughout the country with the intention of expanding the program during the second phase, if successful. The main objective of the first phase (1995-2000) is to lower the illiteracy rate to about 40% overall, and to 47% for women, by supporting literacy providers to offer programs to a total of about 300,000 beneficiaries, 75% women. Other objectives include: (i) ensuring the sustainabilitv of literacy skills, and (ii) strengthening the capacity of the private sector to deliver effective programs, and the capacity of the MCALN to monitor, evaluate and coordinate public investments in literacy programs. The total estimated cost of the first phase is US$ 30.0 million, of which IDA would provide 12.6 million, CIDA US$ 10.0 million (excluding bilateral technical assistance), the GTZ US$ 1.4 million, NGOs and beneficiaries US$ 4.4 million, and Government US$ 1.4 million. Program costs and financing are in Annex 3. 3.2 The IDA financed project would support literacy and post-literacy sub-projects in the regions of Diourbel, Louga, Fatick, Kolda, and in peri-urban areas of Dakar to reach about 135,000 adults (75% women) over five years (1996-2001) -- or about 75% of the tota required to meet the Government's targets for those regions -- to lower the illiteracy rate from a combined average of 50% to 35%. Targets by region are in Annex 3 and shown in the table below. The project would also finance activities to strengthen capacities in program development, delivery, monitoring, evaluation and consultations with partners. Target Number of Beneficiaries Region Beneficiaries Diourbel 25,000 Louga 25,000 Fatick 25,000 Kolda 25,000 Dakar, peri-urban 35,000 3.3. Parallel financing for about 75,000 adults in Tambacounda, Kaolack, St. Louis, Ziguinchor and Niayes would be provided by CIDA. The GTZ will provide financing for about 14,000 beneficiaries in Fatick and Kaolack. On-going projects are expected to cover an additional 50,000 beneficiaries. Progress in achieving the IDA supported project would be assessed against performance indicators in Annex 10. A more complete set of indicators for monitoring and supervision are in the Project Implementation Manual. 1. Literacy and Post-Literacy Sub-Projects (US$ 9.5 million IDA) 3.4 Description. The project would provide US$ 8.2 million equivalent to finance eligible literacy sub-projects prepared and submitted by literacy providers. Literacy providers (see para 2.8) would mobilize participants, develop and provide the program, procure instructional materials, recruit and train instructors, and supervise program implementation. The project would  11 also make available a limited amount of financing (US$ 1.3 million equivalent) for pilot post- literacy activities developed and implemented by providers financed under the project. The project would finance 85% of the cost of the program; the provider would cover the remaining 15%. The post-literacy activities are intended to support skill retention of neo-literates. Procedures governing access to, and management of, the funds are described in an Implementation Manual for the project available in the project files and briefly summarized below. 3.5 Eligibility criteria. Sub-project eligibility criteria for literacy programs were established on the basis of economies of scale in supervision (at least 10 classes of 25-30 auditeurs per class for one sub-project), and characteristics of effective programs in terms of duration -- 250-300 hours of instruction spread out over a period of time agreeable to the beneficiaries, generally 12-18 months. The capacity of providers to deliver the program is assessed against their experience, the quality of the proposal and the commitment of beneficiaries to participate. Criteria are summarized in Annex 4. 3.6 For post-literacy activities, financing would be limited to sub-projects on the list of acceptable post-literacy activities in the Project Implementation Manual. At present, the list contains two priority areas: (i) developing structured learning activities for neo-literates directly linked to their social and economic activities, (ii) promoting the development of local newspapers, and (iii) supporting the development of publications in national languages. Implementation modalities and criteria for the evaluation of sub-projects was discussed during negotiations and will be evaluated at the mid-term review of the project. For all post-literacy sub-projects, providers will be required to set up an experimental design for the program, with treatment and non-treatment groups, in order to assess the cost-effectiveness of the program. The evaluation would be financed by the project and coordinated by the MCALN. 3.7 Proposal evaluation. Proposals would be prepared by the literacy provider, if necessary with assistance from the CROA (Regional Committee of Literacy Providers), and submitted to the MCALN for evaluation. A standard form for proposal submission is contained in the project's Implementation Manual. Proposals would be accepted in May and September each year. This schedule facilitates management and evaluation and corresponds to the peak preparation period for providers as programs are not offered during the peak agricultural months of June- October. Proposals received would be evaluated by an ad-hoc committee comprised of a representative of the MCALN, the Ministry of Women, Children and Families, AGETIP (paras. 4.5-4.6), and an external expert from the provider community. A representative of the CNOA and the chief of the Comite de Coordination et de Suivi (CCS) have observer status on the committee. 3.8 Financing. The project would provide financing up to a maximum of US$ 65 per participant. This cost covers the cost of materials, instructor fees and training, the marginal cost of the program's administration, each representing about one-third of the cost per participant. Beneficiaries (or NGOs) are required to contribute 2,500 CFAF per participant. This figure was selected because it was considered reasonable by literacy providers. A breakdown of estimated program costs is in Annex 5.  12 2. Project Management and Capacity Building in Program Delivery, Management, and Evaluation (USS 3.1 million) 3.9 This component would complement CIDA's capacity building support to the MCALN by financing activities to strengthen the capacity of: (i) providers to deliver effective literacy programs, focusing particularly on improving the quality of their human resources, and (ii) the MCALN to monitor progress in the sector and establish reliable and valid evaluation designs and procedures for sub-projects financed under the credit. The component will also finance client consultations to strengthen the partnership between Government and providers. 3.10 To strengthen provider capacitv, the project would provide support in three areas. Fir it would finance 9000 person days of training and workshops provided by the CNOA. Second, AGETIP, as part of its project responsibilities, would provide advice to providers on managing financial records and accounts to ensure that they meet the guidelines presented in the Implementation Manual for the project. Third, to improve the quality and availability of literacy materials, the project would finance (a) local training seminars for about 10 literacy publishers and MCALN staff responsible for publishing, marketing, sales and distribution of literacy materials, (b) 10 months of short-term international training for MCALN staff responsible for the evaluation of literacy materials, and (c) 3 months of foreign technical assistance to assist MCALN to evaluate the quality of materials used by providers. 3.11 To strengthen the capacity of the MCALN in monitoring and evaluation, the proposed project would finance: (i) the establishment of a geographic and management information system to map out and systematically track progress in the sector, and the cost and effectiveness of projects, (ii) an ex-post evaluation of a representative sample of literacy programs every two years, (iii) a longitudinal study of a sample of program participants to assess skill retention; and (iv) vehicles, equipment, furniture, supplies and training for project management staff involved in monitoring progress in the sector. The development of the ex-post evaluation design has been split into two phases. The first phase assessed procedures, processes and mechanisms and was developed and tested under the experimental phase of the project. The instruments and results are available in the project files. The second part, dealing with literacy achievement, is under preparation and will be tested between October - December 1996, after the programs financed under the experimental phase have been in operation for 12 months. At negotiations, the Government would be asked to provide assurances that: (i) the results of the evaluation will be discussed with IDA and the literacy providers, and the project will be adjusted accordingly; and (ii) the design for the longitudinal study will be developed during the first year of project implementation (not later than June 1997). B. Program and Project Cost and Financing Plan 3.12 Project Cost. Total project costs are estimated at US$ 14.0 million, including taxes, with a base cost of US$ 13.8 million and a foreign exchange component of US$ 1.2 million. Price contingencies (US$ 0.2 million) and physical contingencies (US$ 0.0 million) assume an annual inflation rate of 2.5% local and 2.4% foreign. The IDA credit would finance 90% of total project costs (US$ 12.6 million). Government would finance 4% (US$ 0 5 million). Minimum community and provider contributions arc estimated at 6% of total costs (US$ 0.9 million). Detailed cost estimates are given in Annex 6.  13 3.13 Recurrent Cost Implications. Total costs of the project to the Government of Senegal are estimated at CFAF 250 million (US$ 0.5 million); total recurrent costs of the program to the Government of Senegal are estimated at CFAF 700 million (US$ 1.4 million). The recurrent cost impact is low because: (i) the program and project are pilots and limited in scope, (ii) implementation is carried out by non-governmental organizations, and (iii) no new government staff are being recruited to execute or manage the project. However, the cost of the literacy programs themselves will need to be financed by donors for the foreseeable future. It is anticipated, however, that over time the project will leverage additional resources from the NGO community as an increasing number acquire experience in delivering literacy programs and incorporate it into their other development activities. IV. PROJECT IMPLEMENTATION A. Status of Project Preparation 4.1 The project has been under experimentation since June 1995. Discussions on the Government's approach to literacy, program design, financing and implementation arrangements began in earnest in 1993 with the MCALN and donors. Over time, the MCALN began a dialogue with providers to obtain feedback on its plans and develop a partnership that would enable providers to expand and improve their programs. In July 1995, providers organized a professional association (CNOA) to strengthen their professional capacity and to represent them in discussions with the MCALN. Representatives of this association as well as other providers provided useful feedback on the Implementation Manual for this project. Participation of stakeholders in project design and imnplementation is described in Annex 7. 4.2 An Implementation Manual which describes, inter alia, the role and responsibilities of the parties involved, and the funding process and mechanism was finalized following pre-appraisal and tested prior to appraisal. The test phase provided financing for 24 literacy sub-projects for about 7,200 beneficiaries in four target regions (Diourbel, Louga, Tambacounda, Kolda), 90% women, to: (i) assess provider interest and capacity, (ii) test a potential system of providing technical support to providers (operateurs relais), (iii) assess the efficiency and equity of the proposed processes and procedures, and the performance of the various actors, and (iv) develop and test a participatory evaluation design where results are fed- back into program and project design. 4.3 The response of providers to the test phase has exceeded expectations both in terms of interest and program quality. Seventy-two sub-project proposals were submitted for about 22,000 beneficiaries. An evaluation of the efficacy of the funding and award process and mechanisms, and the performance of the actors involved was conducted during October-December 1995. The results are summarized in Annex 9; the full report is available in the project files. The main changes proposed were: (i) replace the op&rateur relais strategy with technical support provided by the CNOA and the decentralized structures of the MCALN; (ii) enlarge the ad hoc evaluation committee to include a representative of AGETIP and a representative of the CNOA, the latter as an observer; (iii) reduce the number of eligible books providers could use to two core books per language to reduce the cost of publications; and (iv) improve the format of reports to enable better monitoring of sub-project progress. Project design was subsequently modified to incorporate these recomnmendations.  14 B. Project Coordination and Management 4.4 In June 1995, the MCALN created a coordinating unit (Cellule de Coordination et de Suivi, CCS) for the project consisting of a project coordinator, project officer, and two support staff. Under the proposed project, the unit would continue to be responsible for: (i) promotional campaigns and dissemination of information about the project to providers and communities, along with AGETIP and the CNOA; (ii) coordinating the evaluation and approval of sub-projects; (iii) ensuring that all reporting requirements are met; (iv) maintaining a computerized management information system to assess quantitative and qualitative progress in the sector; (v) coordinating evaluations of the sub-projects financed by the credit; and (vi) discussing results with all stakeholders to ensure high visibility and transparency. The CCS would also ensure appropriate coordination with: (i) the Human Resources Development Project (Cr. 2473) concerning the evolution of enrollment in primary education and the subsequent adjustment of literacy targets, (ii) the Community Nutrition Project concerning the literacy needs of the women's groups involved, and (iii) literacy projects funded by other donors. Detailed terms of reference for the CCS are in the Inplementation Manual for the project. 4.5 The MCALN has requested, and the Government of Senegal has designated, AGETIP (Agence d'Execution des Travaux d'lnteret Public) to act an executing agency to manage the contracts with literacy providers, and to procure the goods and services financed by the credit on behalf of the MCALN. Following an assessment of alternative management arrangements for the proposed project, which are summarized in Annex 9 AGETIP was deemed the most appropriate option, principally because AGETIP has relevant competence in the management of nurnerous small contracts and can ensure rapid payments to providers. 4.6 Under the proposed project, AGETIP's would: (i) manage the credit's accounts, including the preparation of withdrawal applications, (ii) manage the contracts with the literacy providers financed by the credit, (i.e. making payrnents and verifying the delivery of services according to contract provisions), (iii) conduct all procurement on behalf of the MCALN, and (iv) ensure that all reports and audits are provided to the MCALN and IDA as required. AGETIP would execute these responsibilities in conformity with its overall convention with the Government of Senegal, existing procurement and disbursement procedures as defined in its own Procedural Manual, the Development Credit Agreement to be signed between the Government of Senegal and IDA, the specific project agreement to be negotiated and signed between IDA and AGETIP, and the Project Implementation Manual. To assist in project implementation, AGETIP would recruit a project coordinator, two technical verifiers, an administrative staff, a chauffeur and accountant for the unit within AGETIP responsible for the project. As a condition of credit effectiveness AGETIP would recruit the coordinator and accountant. C. Monitoring and Evaluation 4.7 Monitoring and evaluation of sub-project inputs and outputs is the responsibility of AGETIP which would provide bi-annual reports of project performance based on disbursements, service delivery verification reports, and the monitoring indicators in the project's Implementation Manual. AGETIP would report on disbursements, and the status of sub-contracts execution relative to the terms of the sub-contract including the number of beneficiaries, the  15 number of female beneficiaries, hours of training for instructors, availability of teaching and learning materials, etc. To undertake reporting and analysis, AGETIP would establish a computerized monitoring system with baseline data on the characteristics of providers and programs drawn from the proposals submitted for financing, and data on implementation progress reported in trimestrial reports provided by the providers and in the verification reports completed by AGETIP. At negotiations, AGETIP would be asked to provide assurances that the computerized monitoring system will be in place not later than June 1997. In the meanwhile, AGETIP would record the data on a manual basis and ensure that all data will be retroactively input into the computerized system. 4.8. The MCALN is responsible for evaluating the quality of the literacy programs provided, progress in the sector taking into account all interventions, the impact of the project on the sector, and coordinating interventions in the sub-sector. The MCALN would: (i) submit annual reports of program quality and impact on the sector of a form and content acceptable to IDA, (ii) organize annual reviews of the project with AGETIP, IDA, literacy providers and other donors, (iii) organize annual reviews with IDA during the fourth quarter of each year to review progress reports and expenditure statements of the current year, and work programs and budgets for the following year; and (iv) not later than September 1999, organize a mid-term review of project implementation, the results of which will be incorporated in an action plan, acceptable to IDA for further implementation of the project. 4.9 To provide information necessary to evaluate progress and outcomes, the MCALN will put into place a Geographic information system to situate and track literacy interventions, beneficiaries and literacy rates, and will evaluate sub-project outcomes in terms of learning achievement and skill retention by carrying out: (a) every two years an evaluation of the outcomes of literacy sub-projects based on a representative sample of sub-projects every two years, and (b) a longitudinal study to assess literacy skill retention and the need for post-literacy activities. Key supervision / monitoring indicators are in Annex 10. D. Procurement 4.10 Table 1 summarizes project costs, including taxes and duties, by disbursement category and procurement method. Costs include all local taxes and import duties. Vouchers are provided by the Government of Senegal to suppliers to cover the tax component. Senegal's procurement laws and regulations conform to IDA procurement guidelines. No special exemptions, permits or licenses need to be specified in credit documents. Senegal's procurement regulations allow IDA procedures to take precedence over any contrary provisions in local regulations. 4.11 Service contracts, which include literacy and post-literacy sub-projects delivered by literacy providers, total an estimated US$ 10.3 million equivalent of which IDA finances 100 % of the amounts paid by AGETIP to the literacy providers - these amounts representing 93% of the cost of literacy programs (US$ 8.2 million) and 85% of the cost of post-literacy progams (US$ 1.3 million equivalent). Procurement procedures for the evaluation and financing of literacy and post-literacy sub-projects are detailed in the Implementation Manual for the project. Requests for proposals will be advertised locally and through information campaigns conducted by the MCALN, AGETEP and the CNOA. Evaluation of the proposals received will be conducted by a Committee consisting of a representative of AGETIP, the MCALN, the Ministry of Women,  16 Children and Families, and an external independent expert. To improve the technical quality and transparency of the process, the CNOA and the CCS have representation as observers. Proposed contract awards to providers esimated at more than US$ 100,000 are subject to prior approval by IDA. Selection and award of provider contracts estimated at under US$ 100,000 will examined a posteriori during IDA supervision missions. Table 1. Summary of Proposed Procurement Arrangements (USS million, including contingencies, taxes and duties) Project Element Methods for IDA financing ICB Other | Total 1. Goods Furniture, Equipment, Vehicles, Supplies 0.5 0.1 0.6 (0.4) (0. 1) (0.5) 2. Specialist Services Service contracts -- 10.3 10.3 (9.5) (9.5) Local and international consultants a/ -- 1.2 1.2 (1.2) (1.2) 3. Training Local 0.4 0.4 (0.4) (0.4) Foreign -- 0.1 0.1 (0.1) (0.1) 4. Incremental Operating Cos 0.6 0.6 (0.2) (0.2) 6. PPF Refund -- 0.75 0.75 (0.75) (0.75) TOTAL 0.5 13.5 14.0 TOTAL IDA (0.4) (2.2) (12.6) al Includes AGETIP fees. bf Includes contractual personnel, and expenses related to supervision such as gas, travel and other expenses. Note: Figures may not add up due to rounding. Figures in parentheses are amounts financed by the IDA credit. 4.12 Goods. Goods financed under the credit (US$ 0.6 million, including taxes) include furniture, equipment and vehicles, of which IDA finances 85% of local expenditure and 100% of foreign expenditure. Goods will be grouped into packages of at least US$ 0.15 million and would be procured through international competitive bidding (ICB) using the World Bank's Standard Bidding Documents (January 1995). Goods which cannot be grouped into these packages would be purchased locally through prudent local shopping based on price quotations obtained from at least three reliable suppliers, provided the aggregate amount of such procurement does not exceed US$ 0.1 million equivalent. IDA-financed contracts for goods above a threshold of US$ 150,000  17 equivalent would be subject to IDA's prior review procedures. The review process would cover 80% of the total value of the amount contracted for goods. 4.13 Consultant services. International and local consultant services financed by the credit (US$ 0.5 million) would be financed in accordance with the Bank's Guidelines for the Use of Consultants (August 198 1). The services include training and technical assistance for monitoring and evaluation, and literacy materials and auditing services. Prior IDA review would not apply to contracts estimated to cost less than US$ 100,000 equivalent for firms, and less that US$ 50,000 for individuals. However, this exception would not apply to the terms of reference for such contracts, single-source hiring of firms, assignments of a critical nature as determined by IDA, or to amendments of contracts raising the contract value to US$ 100,000 equivalent or more for firms and US$ 50,000 equivalent for individuals. E. Disbursements 4.14 The proposed credit of US$ 12.6 million equivalent would be disbursed over a period of about six years. The disbursement schedule based of the expected implementation schedule and the disbursement profile is similar to those of education projects in Senegal as shown in Annex 12. Table 2 shows disbursements by category and financing percentages. Table 2. Allocation and Disbursement of the IDA Credit (excluding taxes and duties) Proposed IDA Allocation Expenditures Category of Expenditure (USS million) Financed by IDA I . Goods (Furniture, equipment, vehicles 0.5 90% of local, 100% of & supplies) foreign expedire 2. Literacy sub-projects 8.8 95% of total sub-project 3. Consultants a/ 1.1 cost 4. Training 0.4 100% 5. Incremental Operating Costs 0.3 90% 6. PPF Refund 0.7 7. Unallocated 0.8 TOTAL 12.6 a! Includes paper and other supplies, gas, tavel and other expenses related to supevision 4.15. Disbursemnents of the Credit would be fully documented except for expenditures valued at less than US$ 50,000 equivalent which would be made against Statements of Expenditures (SOEs). Documentation for withdrawals under SOEs would be retained by AGETIP for review by IDA supervision missions and for semiannual audits. To facilitate disbursements, AGETIP would open a Special Account in a commercial bank to cover IDA's share of eligible expenditures. AGETIP will submit invoices for its own services to MCALN for approval prior seeking the subsequent payment for its own account, and approval of AGETIP fees will be attached to the said withdrawal applications The authorized allocation for the Special Account would be 500,000,000 CFAF (US$ 1.0 million equivalent). IDA would make an initial deposit of  18 that amount from the proposed Credit upon credit effectiveness and would replenish the Special Account upon receipt of satisfactory proof of incurred eligible expenditures. Replenishment requests would be accompanied by up-to-date bank statements and reconciliation of the Special Account. Applications for direct payments and reimbursement or requests for special commitments would apply to contracts above US$ 50,000. F. Accounting, Auditing and Reporting 4.16 AGETIP will be responsible for the project's financial management. It would maintain consolidated project accounts. A computerized accounting system developed for the project was put in place under the experimental phase. All project accounts would be audited annually by independent auditors acceptable to IDA; all disbursements under SOEs would be audited semi-annually. Audit reports of reasonable scope and detail would be submitted to IDA within four months of the end of each auditing period. As conditions of credit effectiveness, recruit an independent auditor under a multi-year contract acceptable to IDA. V. PROJECT BENEFITS AND RISKS 5.1 Benefits. Given trends in primary school expansion, if no adult education programs are provided, by the year 2015, only about 45% of young women in their peak years of childbearing and child-rearing will be literate, and much fewer in rural areas. The consequences are striking for productivity, health, fertility, child survival and education. A recent analysis from Senegal shows that children of illiterate mothers are 50% more likely to die in childhood; illiterate women want 1.8 more children on the same income as literate mothers; and literate mothers are about 50% more likely to send their daughters to school, ceteris paribus. Under these assumptions, the expected health benefits of the literacy program are sufficiently high to make the program attractive on health grounds alone. The program is estimated to result in a 10% reduction in disease burden, equivalent to 8.5 million (disability adjusted) life years saved, at a cost of about US$ 2.00 per capita. In comparison, the most cost-effective health intervention -- minimum package of clinical services -- at an expenditure of US$ 7.80 per capita is estimated to reduce the disease burden by 24% in low-income countries. The program is also expected to result in an estimated 10-14% decline in infant mortality, a 3-5% decline in the number of births, and a 23% increase in girl's enrollment. 5.2 Risks. This is a pilot that tests a new approach to delivering literacy programs. The experimental phase, which began in four regions in August 1995, provided useful feedback on procedures and processes which were incorporated into the project's design. The risk that a larger pilot may not achieve its objectives are difficult to assess at the outset for three main reasons. First, although literacy providers show significant interest in the program, capacity is widely believed to be weak among small providers. To minimize this risk, the project will provide capacity building services to assist the weaker providers and increase absorptive capacity. The second risk concerns the ability of Government and providers to maintain an effective partnership. The alliance established is innovative and recent. To be successful, providers will need to place less emphasis on protecting their turf and more on strengthening their technical capacity. And, the MCALN will need to guard against exercising undue control over providers, particularly in areas where they are more experienced. To assist the process, the project supports regular discussions between the MCALN and providers on program implementation and results, and training for both  19 to improve technical competence. A third potential nsk concerns the sustainability of literacy skills. Some researchers argue that relapse into illiteracy may occur when the enviromnent offers no support for the new literate to apply and imnprove her skills. However, there is no strong evidence to reject the hypothesis that when programs are well designed, and offer a minimum instruction of 300 hours, neo-literates retain their skills over time. To address this issue, the project will conduct longitudinal studies to assess skill retention and operational research to assess the cost-effectiveness of select post-literacy activities. VI. AGREEMENTS TO BE REACHED AND RECOMMENDATIONS 6.2 During negotiations, the Government gave assurances that it would: (a) discuss with IDA and the literacy providers the results of the evaluation, and adjust the project accordingly (para. 3.11); and (b) develop the design for a longitudinal study of skill retention during the first year of project implementation, i.e. not later than June 1997 (para. 3.11). During negotiations AGETIP provided assurances that it would have a computerized monitoring system in place not later than June 1997 (para. 4.7). 6.4 As conditions of Credit effectiveness. AGETIP would: (a) recruit a project coordinator and an accountant for the unit in AGETIP responsible for the management of the proposed project (para. 4.6); and (b) recruit an independent auditor, with qualifications and experience acceptable to IDA, under a multi-year contract to conduct audits of the project's accounts (para. 4.16). Recommendation. Subject to the above terms and conditions, the proposed project would be suitable for an IDA credit of US$ 12.6 million equivalent on standard IDA terms.   Annex I Page 1 of I REPUBLIC OF SENEGAL PILOT FEMALE LITERACY PROJECT Basic Data 1995, or latest year Lower-middle Indicator Senegal Sub-Saharan Africa Income Economies GNP per capita (IUSS) 611 510 1590 Total population (million) 8.3 565 1097 Percent urban 42 28 55 Population growth rate 2.7 2.9 1.6 Adult literacy rate T otal 38 50 80 Female 25 38 Total fertility rate 5.9 6.2 2.9 Infant mortality 68 94 39 Under five mortality 120 172 62 Life expectancy at birth (years) 49 52 67 Primary gross enrollment ratio (%) Total 56 68 104 Female 50 62 Sources: Countrv and Sector Operations, Western Africa Department. Development Diamond Lte expectancy GNP per arosr pnmemy Caplta, emrolhent Serfegd - wer-rdide-irtc , Access to sate water  Annex 2 Page 1 of 4 REPUBLIC OF SENEGAL PILOT FEMALE LITERACY PROJECT Project Benefits Effectiveness of Literacy Programs 1. Literacy programs have a checkered past. Although there are many examples of successful programs, there have also been many failures with high drop-out and low acquisition of literacy skills. Poor program design and implementation are generally the cause of inefficiency. Our review of the literature shows that ineffective programs are characterized by: (i) a duration of instruction and target skill too low to allow the development of skills that can be retained over time, (ii) Government organization, development and delivery of programs, (iii) the use of materials and methods designed for children, (iv) uninteresting and uninformative content, and (v) poorly trained teachers with little education themselves. 2. These characteristics are found especially in mass literacy carnpaigns of the 1960s and 1 970s. It is widely recognized that such campaigns have largely failed. They targeted a level of literacy that may be more accurately called pre-literacy, providing usually 100 hours of instruction, or one-third the number generally considered necessary for the acquisition of skils that can be applied and retained. Instructional and learning materials were frequently lacking. Teachers were poorly trained. And, programs were implemented under unrealistic time constraints and with inadequate resources. As a result, drop out rates were exceedingly high and skill acquisition and retention low because skills were never acquired in the first place. 3. Recognizing the failures of mass literacy campaigns, UNESCO popularized the idea of functional literacy and tested such programs in 11 countries between 1967 and 1972. Functional literacy was defined as skills that enable the individual to engage in activities required for the effective functioning of one's group or community and to use reading, writing, and calculation for is own and the community's development. However, in program terms, functionality was linited to vocational skills and work-oriented program content. Although there may have been some cases where the programs were effective (e.g. Bangladesh), on the whole, there is little evidence to show success. And, due to the vocational content of the programs, they were also costly (UNESCO 1989). 4. Other programs offer what is sometimes referred to as comprehensive, or basic, education. Such programs are longer in duration, provide a range of important information of interest to beneficiaries, and develop critical thinking and problem solving skills, at times with the goal of making participants leaders in their communities or for women's empowerment. In the context of this project, the following definition of literacy programs is understood: to provide participants with a usable level of reading, writing and math skills, teach a range of important information, cultivate critical thinking and problem solving skills, and bring participants to a level where they can retain their skill and improve it over time. This is usually a fourth grade equivalent.  Annex 2 Page 2 of 4 Benefits of Literacy 5. The general literature. Most studies on the benefits to literacy relate the number of years of fomial schooling an individual completed as a child to development indicators such as income, fertility, child mortality, malnutrition, use of health services, etc. when they are adults. This evidence suggests that investments in female education probably yield the highest return of any other investment available to the developing world today. In Africa, estimates of increased income due to a year of primary schooling average 26%. Estimates of the gain in agricultural productivity due to 3-4 years of primary education average 8-10%, and higher when complementary inputs are available. The social benefits are even more significant. Fertility and child mortality are estimated to fall by 5-10% for each year of primary education. The direct health benefits are so significant that the World Development Report on Health concludes that: The cost per year of life saved by educating girls is sufficiently low to make them attractive on health grounds alone, other benefits only strengthen the case. And, mothers who attended school are more likely to send their daughters to school, passing the benefits on to succeeding generations. 6. The evidence above deals with years of formal schooling acquired in childhood and not the effect of literacy skills per se. It is generally assumed that primary education is a proxy for these skills. Although there is a paucity of research that estimates the benefits to literacy skills themselves or to the benefits of literacy acquired as an adult, the few studies available provide evidence to support the hypotheses that: (i) the positive effects of formal education on health and fertility are related to the level of literacy skills retained into adulthood (Levine, Robert et al 1994); and (ii) literacy education for women leads to at least the same impact as primary schooling in the areas of health, family planning and children's education. A review of 43 case studies concluded that women's literacy increased the likelihood of using oral rehydration therapy and immunization services, and improved nutritional and birth spacing behaviors (Comings 1995). Indeed, it is reasonable to suspect that the impact of adult education would be stronger on health and fertility than that of primary education since the knowledge base of adult learners is broader, and literacy programs that target women generally offer health and family planning education at a time when the need and application are more immediate (Comings 1995). In other countries (Nepal, Botswana, South Africa and the US), studies of the effect of adult literacy education on the education of children conclude that women who complete adult literacy classes are more likely to send their children to school, and that literacy level and reading practices are predictors of their daughters' educational attainment. 7. Adult literacy programs are also shown to have important empowerment effects. They increase self-confidence and self-efficacy of women, and provide an opportunity for women to meet, discuss and formulate solutions to an array of common problems. Beneficiaries themselves often describe a more wide ranging set of benefits that motivate them to participate. Among the most common are: the desire to learn reading, writing and calculation for its own sake, to gain more independence and greater control over their personal life, and to acquire new knowledge useful in their daily lives (Wagner, 1994).  Annex 2 Page 3 of 4 8. In Senegal. Although not supported by empirical analyses, literacy providers in Senegal also contend that school enrollment, particularly for girls, has increased significantly in the villages where programs are operating. An analysis from Senegal of available household surveys (Sadio 1994) also confirns many of the social benefits cited above. It shows that children of illiterate mothers are 50% more likely to die in childhood; illiterate women want 1.8 more children on the same income as literate mothers; and literate mothers are about 50% more likely to send their daughters to school, ceteris paribus. With these assumptions, below we estimate the reduction in disease burden, infant mortality and fertility, and the increase in girls' primary school enrollment that could be expected from the Government's literacy program. 9. Table I shows the estimated impact of the program on reducing the disease burden and compares the cost effectiveness of the literacy intervention with the most cost-effective health intervention -- a minimum essential package of clinical services. At US$7.80 per capita, for low- income countries, an essential package of clinical services would reduce the disease burden by an estimated 24%. The second best cost-effective health intervention -- an essential public health package -- is estimated to reduce the burden of disease by 8% at a cost of US$4.20 per capita. The GOS's literacy program is estimated to reduce the burden of disease by 10-14% at a cost of US$2.0 per capita, saving an estimated 8.5 (disability adjusted) life years saved, the equivalent of about 170,000 child deaths. Table 1. Reduction in Disease Burden */ Intervention Cost per Reduction in disease burden capita (percent) b/ (US$) Literacy Program 2.0 10-14 Minimum package of clinical services C/ 7.8 24 Essential public health package d/ 4.2 8 a/ Data on healft interventions is the average for low-income countries from the World Development Report, 1993. b/ Marginal benefitg. c/ Includes tuberculosis treatment; management of the sick child; prenatal and delivery care; family planning, STD treatment; treatrnent of infection and minor trauma; assessment, advice and pain alleviation. dt Includes expanded program on immunization; school health programs; tobacco and alcohol control; health, nutrition and family planning infonration; vector control; STD prevention; monitoring and suiveillance. 9. Table 2 shows the estimated impact of the literacy programs on infant mortality, the birthrate, and girls' primary school enrollment . Infant mortality for Senegal is predicted to fall by 10-14%, the number of births by 3-5%, while girls' enrollment is expected to receive a boost of 23%. Table 2. Impact on Social Indicators % Reduction in infant mortality % Reduction in % Increase in girls' enrollment births 10-14 3-5 23  Annex 2 Page 4 of 4 Cost Effectiveness Relative to Primary Education II. Evidence suggests that schooling requires a much higher level of resources per student than do literacy programs for the same literacy outcomes. In Senegal, over 1,500 hours of instruction in language and mathematics are required to achieve a fourth grade education, i.e. the level assumed to be required to acquire a level of skill sufficient to use and retain over time. This excludes the number of hours invested in students who repeat and those who drop-out before fourth grade. For adults, 200-300 hours of instruction are generally assumed to be needed to acquire an equivalent level of skills mainly because adults acquire the skills with considerably less time on task. Due to the comparatively long duration of primary school, and because teachers' salaries (excluding benefits) are more than triple the earnings of literacy instructors, the cost of producing a graduate from an adult literacy program is significantly less that the investment required to obtain a fourth grade completer. Table 3 shows estimates of the average cost of literacy program completers at increasing rates of program drop-out. With 0% drop-out, the cost per completer of an adult literacy program is one-tenth the investment required to obtain a fourth grade completer. Even at a high drop-out rate of 50%, the cost of an adult literacy program completer is one-third the investment required to produce a fourth grade graduate. Table 3. Cost Per Literacy Program Completer Compared with the Investment Required to Obtain a Fourth Grade Graduate from Primary School Cost per graduate Drop-out rate from as % of cost of Number of literacy literacy program US$ 4th grade completer participants required 0% 65 10% 280,000 10% 72 11% 308,000 30% 93 14% 364,000 50% 130 19% 420,000 70% 217 32% 476,000 Sustainability of benefits 12. The literacy programs supported by the project address a short and medium-term need. Unlike prinary education which must become a permanent feature of all countries, the need for literacy programs should diminish as primary education reaches a larger share of school-aged children. But like primary education, questions have been raised about the sustainability of the literacy and numeracy skills acquired. However, experts do not agree that this is a widespread problem. Some researchers and practitioners argue that relapse into illiteracy may occur when the immediate environment offers no support for the new literate to apply and improve her/his skills. Others maintain that the sustainability of literacy skills is a function of the quality of the literacy program itself. They argue that the skills acquired are maintained in the absence of special programs to create a literate environment when literacy programs themselves are effective, a key factor being meaningful content. Still others argue that, although skill retention is enhanced when programs have meaningful content, post-literacy activities should form part of the literacy program. Such activities might include village libraries, newspapers, structured group learning, further training, or the creation of community structures managed by new literates to which reading materials can be attached. Information on which to assess the effectiveness of such activities is extremely scarce.  Annex 3 Page 1 of 5 REPUBLIC OF SENEGAL PILOT FEMALE LITERACY PROJECT Draft Policy Paper and Action Plan for the Literacy Sector (1995-2000) 1. CONTEXT AND JUSTIFICATION Senegal has one of the world's highest illiteracy rates, estimated in 1988 at 68% of the population age 6 and over, 59% for males and 77% for females (compared the Sub-Saharan average of 65% for women). Illiteracy rates are even higher in the rural areas, reaching 95% for rural women in Diourbel, Tambacounda, Kolda, Louga and 85% for rural women in St Louis, Kaolack and Fatick, compared with 65% of urban women. The disparities among regions and between urban and rural areas are due mainly to unequal access to primary schooling, lack of school buildings or teachers, limited access to adult literacy programs and, in some cases, to the population's reluctance to enroll their children, particularly girls, in the formal school system. In the short and medium run, even with the ambitious program (Programme de Developpement des Ressources Humaines, PDRH) that the Government has embarked upon to expand access to primary school and promote girls' education, the literacy rate will not increase noticeably unless greater efforts are made to promote literacy among adults and adolescents who have never had access to school or have dropped out early. Projections show that, assuming PDRH objectives are attained by 1998 and the enrollment growth rate continues at the same level after 1998, only 55% of the female population in the 9-39 age group will be literate in the year 2000, and only 45% in 2005, with literacy rates of 75% and 65% for rural females in 2000 and 2005 respectively. The advantages of literacy in terms of increasing workers' productivity, family management capacity and income, improving family health and fmnily planning, and promoting democracy are well known. Experience in Senegal shows that from the point of view of costs, literacy programs could be more cost-effective than formal schooling. The average cost for a literacy learner is esfimated at about 40 000 CFAF or 10-15% of the cost for four years of formal primary education for children. II. GENERAL ORIENTATIONS The Government's literacy policy is in line with the conclusions of the Etats Generaux de I 'Education et de la Formation (EGEF), the recommendations of the Commission Nationale de Reforme de I 'Education et de la Formation (CNREF), the Jomtien World Conference on Universal Basic Education, the Loi d'Orientation n