Madagascar - Food Security and Nutrition Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 11301-NAG STAFF APPRAISAL REPORT DEMOCRATIC REPUBLIC OF MADAGASCAR FOOD SECURITY AND NUTRITION PROJECT FEBRUARY 18, 1993 Population and Human Resources Division Department III Africa Region This document has a restricted distribution and may be used by recipients only in the perfornance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EOUIVALENT Currency Unit = Malagasy Francs (FMG) U.S.S = FMG 1,800 MEASURES Metric System FISCAL YEAR January 1 to December 31 GLOSSARY OF ABBREVIATIONS ACN Community Nutrition Worker (Agent Communautaire de Nutrition) BPS SECALINE Project Office (Bureau d'Excution du Projet SECALINE) CICS Interministerial Coordination Committee (Comite Interministeriele de Coordination du Projet SECALINE) CRT Nutrition Rehabilitation Center (Centre de Rehabilitation Thirapeutique) EMSAP/PASAGE Economic Management and Social Action Project (Projet des Actions Sociaks et Appui a la Gestion Economique) EWS Early Warning System (Systlne d 'Akrte Pricoce) FARITANY Province (groups about 20 fivondrononana) FID Social Fund (Fonds d'intervention pour le Diveloppement) FIRAISANA Local Govemment Grouping of Villages or Neighborhoods FIVONDRONONANA Districts (includes everal Firaisana) FOKONTANY Village or Neighborhood ICB Intemational Competitive Bidding IEC Information, Education and Communication (Information, Education et Communications) IDD/ TDCI Iodine Deficiency Disorders/ Troubls Das aax Carences en lode ILO/ BIT Intemational Labor Office/ Bureau International du Travail KAP Knowledge, Attitude and Praetice LCB Local Competitive Bidding MSF Doctors without Borders (Midecins sans Frontieres) NGO Non Govemmental Organization (Organisation Non Gouvernmentak) PCN Community Nutrition Program (Programme Communaulaire de Nutrition) PNSAN National Food Security and Nutrition Monitoring System (Programme National de SurveiUlance Alimentaire et Natritionelke) SECALINE Food Security and Nutrition Project (Sicuriti Alimenatire et Nutrition Elargie) UNICEF United Nations International Childrens Emergency Fund WFP/PAM World Food Program (Programme Albnentaire Mondial) This report is based on the findings of a World Bank appraisal mission composed of Messrsn.Mss. Qaiser Khan (Mission Leader and Human Resource Economist), Paul Geli (Consultanttlmplementation Specialist and Deputy Task Manager), Michele Lioy (IEC Specialist), Tonia Marek (Nutritionist), Eileen Murray (Financial Analyst) and Alf Persson (ConsultantJMIS Specialist). Further assistance was provided by Ms. Trudy Bower-Pirinis of the World Food Program. Messrs. Alan Berg and Harry Walters are the Lead Advisors while Mr. Jim Greene is the Peer Reviewer. Mr. Francisco Aguirrre-Sacasa is the Department Director, Mr. Alain Colliou is the managing Division Chief and Mr. Jerome Chevallier is the Department Operations Advisor. The report benefited strongly from the advice and support received from Dr. Jacques Baudouy who was Acting Division Chief, AF3PH for an extended period before and after the appraisal mission. Ms. Hilda Emeruwa provided secretarial support with support and guidance from Ms. Silvia Marquina-Loon.  FOR OFFICIAL USE ONLY DEMOCRATIC REPUBLIC OF MADAGASCAR FOOD SECUErrY AND NUTRmON PROJECT TABLE OF CONTENTS Credit and Projet Summary .................. i I. BACKGROUND ..1.............. I A. Introduction ................ 1 B. Food Insecurity and Malnutrition in Madagascar ................... 2 C. Government Policies and Actions and the Bank's Experience .... ....... 4 D. Role of Non-Governmental Organizations (NGOs) .................. 5 E. Rationale for IDA Involvement .............................. 6 II. THE PROPOSED PROJECT ...... ..................... 6 A. Project Objective and Design ............................... 6 B. Project Description ..................................... 7 C. Project Costs and Financing .............................. 11 HI. PROJECT IMPLEMENTATION ............. .. ................. 14 A. Status of Project Preparation .............................. 14 B. Project Management ................................... 14 C. Project Monitoring and Evaluation .......................... 17 D. Procurement ........................................ 19 E. Disbursement ........................................ 23 IV. BENEFITS AND RISKS ....... .......... .................... 25 A. Benefits ........................................... 25 B. Risks ............................................. 26 V. AGREEMENTS. ASSURANCES ANDRECOMMENDATION .... ......... 27 ANNEXES ANNEX I: Fonds d'Intervention pour le Develoopement (FID) ................ 30 A. General . .......................................... 30 B. Legal set up, Organization and Management .................... 30 C. Types of Projects, and Selection and Evaluation Criteria .... ......... 31 D. Beneficiaries of FID .......... ......................... 32 E. Operation of FID ........... .......................... 32 ANNEX II: URBAN FOOD-FOR-WORK PROGRAM ....................... 35 A. Objective ........................................... 35 B. Types of works ............ .......................... 35 C. Management and Implementation ........................... 35 D. Contracting ......................................... 36 E. Co-financing . ........................................ 36 This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  ANNEX IH: COMMUNITY NUTRMTION PROGRAMME (PCN) .... .......... 37 A. Background .. 37 B. Component Description ............. .. .................. 37 C. Component Objectives ................ .................. 41 D. NGO Involvement ....... ........... ................... 42 E. Overview of the Manual for the Community Nutrition Program ... ...... 43 ANNEX IV: IODINE DEFICIENCY DISORDERS aDD) CONTROL PROGRAM .... 46 A. Background ......................................... 46 B. Proposed strategy ..................................... 46 C. Institutional arrangements ................................ 47 D. Estimated Cost ....................................... 47 E. Project Financed Activities ............................... 48 ANNEX V: INFORMATION. EDUCATION. AND COMMUNICATION (IEC) ... 50 A. Introduction ....... .......... .. ..................... 50 B. Management of IEC Activities ............................. 50 C. IEC Target Audiences ................ .................. 51 D. IEC Strategy ........................................ 51 ANNEX VI: PROJECT MANAGEMENT ASPECTS ...................... 53 A. General ........................................... 53 B. Project characteristics ................................... 53 C. Project management and coordination ......................... 53 D. Project structure ...................................... 54 E. Distribution of supervision and coordination roles ................. 55 F. Management information system ............................ 58 G. Service functions ..................................... 58 H. Training and support ................................... 58 Attachment : Terms of Reference for Key Project Management Staff ... ..... 59 ANNEX VIII: MONITORING AND EVALUATION ....... .. ............. 61 A. General ........................................... 61 B. Social Fund (FID) ..................................... 62 C. Urban Food-for-Work .................................. 63 D. Community nutrition program (PCN) ......................... 64 E. Iodine deficiency disorder (IDD) control program ................. 65 F. IEC Component ...................................... 66 G. National Food Security Strategy ............................ 67 ANNEX VIII: LETTER OF FOOD SECURITY POLICY ................... 69 ANNEX IX: DETAILED COST TABLES ............................. 71 ANNEX X: ESTIMATED SCHEDULE OF DISBURSEMENT ................ 85 ANNEX XI: SUPERVISION PLAN ................................. 86 ANNEX XII: LIST OF DOCUMENTS IN PROJECT FILES .................. 88  DEMOCRATIC REPUBLIC OF MADAGASCAR FOOD SECURITY AND NUTRITION PROJECT Credit and Projet Summary Borrower: Government of the Democratic Republic of Madagascar Beneficiaries: The Fonds d'Intervention pour le Developpement (FID) and Community Nutrition Centers. Amount: SDR 15.5 million (US $ 21.3 million equivalent) Terms: Standard IDA terms with 40 year maturity and a 10 year grace period. Project Description: The project which would be primarily active in two of Madagascar's six provinces has three components: (a) Income Generatinp Activities to be carried out through (i) the Fonds d'Intervention pour le Dgveloppement (FID) financing labor intensive works and other income generation activities and (ii) a Food for Work program targeted to the urban poor in Greater Antananarivo; (b) Nutrition Interventions including (i) a Community Nutrition Program in the provinces of Toliary and Antananarivo working through mothers groups organized by a community nutrition agent supported by NGOs to provide nutrition education, arrange supplemental feeding and refer severe cases of malnutrition to nutrition rehabilitation centers operating under medical supervision and (ii) a national Iodine Deficiency Disorders (IDD) control program; and (c) Institutional Strengthening to support the development and execution of an Information, Education and Communication (EC) program and support the development of a National Food Security Strategy for the long term. Estimated Project Costs. Financinf Plan and Disbursements: Summary Project Costs Estimates in US $ Millions ComponUVC(ory L1cal Forein Total A. Income GacCnsaim Activities 1. FID project 3.0 2.8 10.3 2. Food for Work 3.7 3.5 7.3 B. Nutrition 1. Counuity Nuriom Tobty 2.2 1.3 3.6 2. Community Nuition Aa_uriw 1.1 0.8 1.9 3. Iodine Derfciency Disorder 0.1 0.6 0.7 C. IEC and Food Scurty Siuteg 0.6 0.4 1.1 Pjet Pemsnt 0.6 0.5 1.1 PPF d apmOmt 0.5 1.5 2.0 Total BASE COSTS 16.9 11.5 23.5 Physical C _ezmgmciss .3 0.4 1.2 Price Continpaci 2.2 0.5 2.7 Totl PROJECT COMTS 20.0 12.4 32.41/ I/ Excluding taxes of US$0.9 million and duties of US$1.1 million.  - ii -. Financing Plan: Source Local Foreign Total IDA 14.4 6.9 21.3 World Food Program 2.6 4.7 7.3 Japan 0.1 0.4 0.5 UNICEF - 0.4 0.4 Beneficiaries 1.2 - 1.2 Government 1.7 - 1.7 TOTAL 20.0 12.4 32.4 Estimated IDA Disbursements [Year FY94 FY95 FY96 FY97 FY98 Annual 3.4 5.6 4.6 5.0 2.7 Cumulative 3.4 9.0 13.6 13.6 21.3 .igures mnay not add up due to rounding. Benefits and Risks: The proposed project would alleviate poverty and reduce malnutrition. By providing the poor with access to income earning opportunities, it would not only help them escape from poverty but also contribute to national output. Labor intensive works financed by the FID would also have development benefits over and beyond immediate job creation. The project would help improve the capacity of NGOs and community groups. Component specific benefits are described below. FID projects are expected to: (a) through labor-based works, create 1.9 million person-days of employment and provide 91 contracts to small enterprises to help their development; (b) support 25 NGO income-generating activities annually targeted to the poorest; and (c) help create about 2,000 permanent jobs through micro-enterprises development. These estimates are based on a certain assumption of the distribution of FID projects, however as FID will be demand-driven, this distribution may change. Food-for- Work is designed to be a temporary transfer program to address the acute poverty in Greater Antananarivo but it would help create some community infrastructure such as solid-waste bins. The program would create 4.2 million person/days of employment over five years targeted to the poorest workers (the majority women) in the poorest neighborhoods. Community Nutrition Program. It will take a preventive/public health approach to reducing malnutrition through emphasizing growth monitoring, nutrition education and supplemental feeding to keep nutrition problems under control and to help households better use existing resources. This means that households will be better able to prevent malnutrition even without the aid of the project. By the end of the project 91,200 children will be monitored in Toliary which is almost universal coverage for the 8 target districts and another  - Hii - 30,600 children will have undergone supplemental feeding. The objectives for the province of Antananarivo include growth monitoring for 60,800 children in the five target districts which gives a coverage rate of over 60% and 15,300 children 0-3 in Antananarivo will benefit from supplemental feeding. These figures are indicative and may change as the project intends to follow a beneficiary led demand driven approach. Iodine Deficiency Disorders Control Program would reduce at a very low cost ($0.11/person) a problem that is prevalent in Madagascar and is the cause of much morbidity. By reducing the prevalence of iodine deficiency related disorders, significant savings in public and private health care expenditures would be achieved. IEC program will help to communicate messages designed to improve the food security and nutritional status of the population. The development and adoption of a National Food Security Strategy would help the country reduce its food insecurity in the long-run. The main risk is implementation delay that could result from inadequate management. This will be minimized by setting up a project management structure outside the administration and following private sector rules. There is a risk resulting from the political transition process which may have a blocking effect on key activities. Several pre-conditions for the project emphasize steps to be taken with respect to project management. Component specific risks are described below. The political risk political is reduced by the fact that this type of poverty program enjoys support across the political spectrum. There is a risk of conflict between two potentially strong institutional entities: the Food Security and Nutrition project office (BPS) and the FID. This risk will be reduced by the National Coordinator of SECALINE (Food Security and Nutrition Project) being appointed as the Prime Minister's representative on the Board of FID. FID. Specific risks for this component are its unsustainability in the long-run and the potential of FID becoming a non-performing public sector agency at the end of the project. These risks are minimized by: (a) targeting the poor by using strict project selection criteria; (b) supporting the development of grass-roots development groups which focus on self-help; (c) financing training and technical assistance to develop sustainable income earning opportunities such as microenterprises; and (d) setting up the FID not as a public enterprise but as a private non-profit association. Food-for-Work. Specific risks include: (a) logistical problems in moving food from port to work-sites; (b) limited capacity of local authorities; (c) leakage of benefits to the non-poor; and (d) effect of importing about 9,000 tons of rice (about 2,000 tons per year) on domestic producer prices. These risks will be minimized by: (a) contracting the private sector to store and transport the material; (b) hiring the International Labor Office (ILO) as an executing agency for this component; (c) by (i) targeting the poorest neighborhoods and the most unappealing work (cleaning drains etc.); and (d) importing 2,000 tons a year represents less than 0.2% of total rice supplies and by targeting the rice to the poorest parts of the population who would otherwise not have the purchasing power to buy the rice. Community Nutrition Pro2ram. Specific risks include: (a) non- responsiveness to the needs of the beneficiary communities; (b) creating dependence on food aid; and (c) logistical problems of running a program in vulnerable and sometimes inaccessible areas. These risks are minirmized by: (a) requiring community participation from the beginning; (b) emphasizing (i) nutrition surveillance, (ii) preventive feeding supplementation as necessary rather than regular meals, and (iii) nutrition education focusing on more efficient use of materials already available to the households; and (c) (i) using an NGO familiar with the area to  - iv - execute the program, (ii) relying as much as possible on locally purchased materials, and (iii) giving as much decision making independence as possible to the ACNs and their immediate supervisors. Iodine Deficiency Disorders (IDD) Control Program. Specific risks are: (a) private salt producers may become dependent on subsidies to iodize salt; (b) public may be unwilling to pay slightly more for iodized salt; and (c) some may be allergic to iodine. These risks will be minimized by phasing subsidies out over time as more and more of the salt in the market is iodized and by initially subsidizing iodized salt so that it costs the same as non-iodized salt. By the time the subsidies on iodized salt are phased out, the health benefits should be evident. Institutional Development. Specific risks are: (a) inappropriate IEC messages; and (b) a long term food security strategy that does not reflect a consensus. These risks are minimized by use of Knowledge, Attitudes and Practice (KAP) and beneficiary assessment surveys to develop IEC messages and by supporting a national seminar to discuss the recommendations of the working group on food security strategy. Map: IBRD No. 20035R Internal Rate of Return: Not Applicable  MADAGASCAR F'OOD SECURTlBY AND NUTRMTON PROJECT STAFF APPRAISAL REPORT L IAC A. In hion 1.1 With a population of 11.6 million, growing at 3 % per year, and a per capita income of US$220 distributed unequally, Madagascar is among the least developed countries in the world. The agriculture dominated economy, which had grown modestly in the late 1980's following a decade of decline, is now stagnant and severely disrupted by political developments since mid-1991. Starting in July 1991, widespread demonstrations and strikes paralyzed public administration and severely disrupted economic activity. A transitional power-sharing arrangement was negotiated out of the political stalemate and a new coalidon government was put in place in January 1992. A national referendum approved a new constitution in August; the first round of the Presidential Election was held in November, 1992 and the second round was held in February, 1993; parliamentary elc,tions are expected in April, 1993 following which a new government is expected to be in place by mid- 1993. These political developments worsened the economic situation of the poor and a survey carried out by the government and UNICEF found that 83 % of the urban poor lost part or all of their income sources during the strikes. Public administration is inefficient and its capacity extremely limited. Much of the development effort is carried out by NGOs, the private sector and some foreign-assisted projects. Despite the country's significant development potential, its social conditions are below sub- Saharan African standards: chronic malnutrition rates are about 40%, child mortality claims 20% of live births before age 5; endemic malaria has become a major health problem; tuberculosis and other communicable diseases are growing in importance; and education standards are low with enrolment rates falling, though from a relatively high level. 1.2 The government started adjustment operations in 1985 with support from the Bank and the IMF, focusing first on liberalizing agriculture, industry and trade; and moving on to a broader public sector adjustment program. The government also requested Bank support in the social sectors and sectoral operations in education (FY90) and Health (FY91) have been approved by the Bank. To help the poorer segments of the population during the period of adjustment, the government developed an Economic Management and Social Action Project (EMSAP/PASAGE) which included pilot targeted programs to reduce food insecurity among the poor. This project received Bank support but the pilot programs have not had the expected results because of managerial weakness. Nevertheless, that experience has provided useful lessons in developing the proposed food security project which was requested by the government in 1990. These lessons, which have been incorporated into the design of this project, include the need to: (a) involve communities and beneficiaries in project design and execution; (b) work outside traditional government structures; (c) have clearly outlined decision making structures and lines of authority in multi-component projects; and (d) have a good monitoring, financial management and accounting system.  B. Food Jlnecuribt and M alutrition in Madar Food -kasecun 1.3 Madagascar's economy grew modestly from its independence in 1960 through 1970; stagnated during 1970 to 1980; deteriorated sharply from 1980 through 1982; and experienced stabilization with modest economic growth from 1983 through 1987. Between 1988 and 1990, the economy showed encouraging signs with positive per capita income growth (about 1 % per year) for the first time in almost two decades. Private investment increased, particularly in labor intensive export industries, but the political turmoil that swept the country in the second half of 1991 wiped out much of the gain made during the 1988 to 1990 period. 1.4 Productive assets are distributed unequally and access to economic opportunities is inequitable in both urban and rural areas. A study found that between 1962 and 1980, income inequality increased with the Gini coefficient of income inequality rising from 0.391 to 0.489 (Gini=0 denotes perfect equality and Gini= 1 denotes perfect inequality i.e. the higher the value the more unequal the income distribution). The increased inequality was more pronounced in rural areas where the Gini coefficient changed from a more or less equitable 0.290 to an inequitable 0.435. The Gini coefficient in the six largest towns in 1962 was 0.500 and while no corresponding measure is available for 1980, a survey of the income distribution in Antananarivo in 1989 found the Gini coefficient close to 0.800 (extreme inequality). According to this survey, the top 10% of the urban population earned over 80% of urban income. If this level of inequality is representative of the whole population, it would be unprecedented in the world. Whether this survey is representative or not, it is evident that inequality in Madagascar has worsened during the last two decades. 1.5 Even though food production appears to exceed estimated needs by 20% or more, almost 40% of the population is food insecure because their own production and/or purchasing power is insufficient. Furthermore, markets are inefficient and segmented leading to pronounced seasonal malnutrition even in areas producing substantial food surpluses. Bank sector work shows that food insecurity in Madagascar is linked to poverty and unequal access to economic opportunities . The Bank food security study found that in 1989, food insecurity was estimated to affect 34% of rural households and 42% of urban households. Urban food insecurity was much more intense, partly because many of the rural poor who do not have access to land or other productive assets move to urban centers (particularly the Greater Antananarivo area). Food insecurity was found to vary regionally with the highest levels being in the provinces of Toliary (54% of households) and Antananarivo (46% of households). The study found that food insecurity could not be reduced through distribution neutral economic growth alone. To reduce food insecurity, a poverty reducing growth strategy has to be followed and the poor have to be provided with the means with which they could participate as full partners in Madagascar's economic life and share in the benefits of growth. 1.6 The sector study on food security was done in 1990 and the situation has worsened since, largely due to the economic stagnation brought about by the political crisis. A survey 2/ of the impact of the political crisis on the poorest households in major urban areas showed that 83 % of 2/ UNICEF/PNSAN/Ministere de la Recherche Scientifique Enquete sur l 'Impact de la Crise Socio-Politique de mal a novembre 1991 d Madagascar, Antananarivo, July, 1992  the poor households in urban areas lost part or all of their income sources with 27% of poor households losing all their income sources and another 56% suffering forced reductions due to the strike. Even after the political crisis was over, 40% of these poor households were unable to regain their original sources of livelihood. During the crisis 72 % of these households were forced to reduce rice consumption (an important signal of economic distress in Madagascar), while after the crisis 61% of these households were consuming less rice than before the political crisis. These recent developments have increased the need for the type of targeted interventions which this proposed project would support. 1.7 The southern rim of Madagascar which has a semi-arid agro-climatic zone has long been neglected and remains at the periphery of the national economy. Frequent droughts leading to famines or near-famines have affected the region even during those years when there were food surpluses nationally. In 1992, famine threatened almost a million people even as the country was producing a very good rice harvest in the highlands and the west coast. The long-term food security of the region requires: (a) research and extension efforts to promote drought resistant crops; (b) improved infrastructure to better link the region with national markets; (c) diversification of the income sources of the population; and (d) increased efforts to develop the region's ground-water resources. Before the benefits from these long-term efforts begin to bear fruit, a series of short-term targeted measures are necessary to ease the vulnerable population through the transition period. The proposed project includes some of these short-term measures. Malnutrition 1.8 Chronic food insecurity combined with poor health conditions has led to very high rates of malnutrition in Madagascar. The Programme National de la Surveillance Alimentaire et Nutritionelle (PNSAN) found that chronic malnutrition affected between 33% and 37% of children under five rising as high as 60% in some areas while acute malnutrition affects between 6% and 7% of children rising as high as 13% in some surveys 4/. Seasonal malnutrition was high even in the agriculture surplus region around Lake Alaotra. 1.9 The problem of protein-energy malnutrition which is directly correlated with food insecurity and endemic diseases such as diarrhoea, is compounded by the existence of various micronutrient deficiencies. The most common deficiency is iodine, but iron deficiency, calcium deficiency and vitamin A deficiency are also prevalent. These micronutrient deficiencies are directly attributable to the Malagasy diet which is dominated by rice (cassava among poorer people and in certain regions). Some of the micronutrient deficiencies are likely to be reduced as incomes grow and households are able to afford more vegetables and protein. 1.10 Iodine deficiency disorders (IDD) are endemic in Madagascar particularly in the Highlands due to the lack of iodine in the diet and the consumption of goitrogenic food such as cassava, certain green leaves and cabbage. The most common symptom of IDD are goiters but IDD also causes birth of (i) mentally and physically retarded children, (ii) deaf-mutes and (iii) mildly retarded children. IDD also retards the mental development of children and is unlikely to be reduced by income growth. Even though the problem has been known for many years, it has not been possible to develop an IDD program until now because the following pre-conditions for such a program were only met in 1992: (a) an IDD prevalence map was published; (b) a salt distribution study was done; 4/ PNSAN Documents de l 'Atelier National de Surveillance Alimentaire et Nutritionnelle - Synthese des donnkes sur 1'etat nutritionnel, Antananarivo, July, 1989  -4- and (c) a feasibility study for salt iodination was carried out. Based on these data, the Ministry of Health with support from UNICEF developed a national IDD control program in November 1992, which will be partly financed by this project. 1.11 Weaning foods tend to be a variation of the adult diet with the bulk of rural infants receiving watery rice (vary sosoa). In urban areas, there is a little more variation with muffins (mofogasy) being predominant among the urban poor. These foods do not have the nutritional density recommended for weaning foods. However, mothers breast-feed their infants as long as possible, usually stopping only when the next one is born. Nutrition programs have to focus on educating mothers as low-cost weaning foods with high nutrient density can be prepared with materials found in most Malagasy homes, though access to supplementary resources will have to be made available to the poorest households. C. Government Policies and Actions and the Bank's Experience Policies 1.12 To reduce food insecurity a set of policies with the following objectives are needed: (a) increase economic growth; (b) improve the efficiency of markets; (c) reduce the opportunity to engage in rent-seeking activities; (d) increase income earning opportunities for the poor (by improving the human resource endowment of the poor and giving them equal access to opportunities); and (e) integrate less favored regions into the national economy. These are all among the stated objectives of the government though progress is slow. 1.13 The transition government is officially committed to building a competitive market economy and to reduce rent-seeking opportunities by liberalizing and integrating markets. However, policy reform has been stalled because a multiplicity of interim government institutions limits the government's capacity to act decisively. The regulatory environment which allowed the rent seeking activities and contributed to increasing inequality during the past two decades remains largely untouched. An example of this comes from the vanilla sector which is Madagascar's primary export crop. A Bank study found that the regulatory framework is such that much of the benefits accrue to a few private traders and the government. The small-holders who grow vanilla receive little for their effort giving them no opportunity to improve their economic position and almost no incentive to increase the output or improve the quality of their vanilla. 1.14 Despite the encouraging results on the economic front during 1988-90, the supply response to structural adjustment has been inadequate and has not attained its potential with traditional export crops and industries stagnating. The supply response has been constrained by infrastructure bottlenecks, poorly functioning factor and commodity markets, and adverse terms of trade. The stated commitment to human resource development has not been reflected in budgetary allocations to the Ministries responsible for education and health. Actions 1.15 With support from the Bank and other donors, Madagascar has embarked on a series of operations in agriculture, infrastructure and human resources which would help improve the country's food security situation. Specific actions include: (a) improvement of infrastructure; (b) removal of remaining constraints on agriculture markets; (c) investment in agriculture research and extension; (d) irrigation development; (e) education sector development; and (f) health sector development. However, the execution of many of these programs has been hampered by the  weakness of the central administration combined with an unwillingness to delegate control of resources outside the capital. 1.16 An early warning system which had been set up by the government to provide advance information on food emergencies is largely dysfunctional; in any case even functioning early warning systems have proven to be ineffective unless they are linked up to an emergency response management system. The recent famine in southern Madagascar illustrates the problem clearly. A non-governmental early warning system put in place by the Medecins-Sans-Frontires (MSF) warned of the crisis more than a year before the famine actually occurred but no action was taken until the World Food Program decided to set up an emergency response network with help from other donors. 1.17 Various NGOs and churches operate nutrition programs on an ad-hoc basis. These are primarily designed to relieve the symptoms of malnutrition and do not aim to support actions which would help solve the problem in the long-term. A systematic approach which combines nutrition activities such as surveillance, supplementation, rehabilitation and education with support to develop income earning opportunities is needed. The nutrition component of the proposed project would support such a program. Lessons from Previous IDA Involvement 1.18 Execution of projects in Madagascar has been hampered by bureaucratic delays and procurement problems because of cumbersome procedures and weak managerial capacity in the public sector. The proposed operation will address these problems by building private sector and NGO capacity to deliver social services through: (a) setting up a project management structure outside the central administration to be staffed by personnel recruited on a contractual basis from the private sector; (b) involving NGOs, local communities and private entities in project execution; and (c) setting up FID as a private non-profit association operating under private sector rules. Separate special accounts would be set up for the project unit and for FID. Key project staff have been trained in procurement, accounting and financial reporting procedures and Bank approved standard documents will be used. The audit and financial management system is being prepared by an international audit firm. A project launch workshop and a mid-term review are planned. The project has also benefited from Bank experience in social funds all over the world and the following has been incorporated into the design of FID: (a) complete operational independence; (b) a detailed operations manual clearly defining procedures; (c) requirement of some minimal counterpart funding in cash or kind from beneficiaries; and (d) frequent audits. D. Role of Non-Governmental Organizations (NGOs) 1.19 NGOs (broadly defined to include churches and their local parishes and missions) are active in programs targeted to the poor; outside the capital and other principal cities, much of the development effort is carried out by NGOs. Most NGOs active in the field are churches and their parishes and in many cases, the poorest segments of the population appear to have more confidence in NGOs than in the authorities, though local services of national ministries are better accepted than their central counterparts. It is likely that much of project implementation in the field will be done by church based NGOs, though non-religious NGOs will also be important in areas where they are active. The proposed project expects to rely heavily on NGOs to execute the major components. The community nutrition program will be executed by the communities with support from local NGOs contracted by the project. These NGOs will sensitize the communities and help them set up the program. NGOs will be eligible to receive a major share of funding from the FID (see paragraph 2.8).  - 6 - 1.20 NGOs played an active role by participating in the project preparation units for Toliary and Antananarivo. They were also consulted by the team preparing the FID. The FID itself will be set up as a non-profit association of which interested NGOs could become members. E. Rationale for IDA Involvement 1.21 Given Madagascar's very low level of development, alleviating poverty is a priority of the Bank's assistance program. Sustained growth is a necessary condition for providing a durable and effective response to this problem, and IDA-financed operations in agriculture and infrastructure are addressing two important elements of this response: growth of agricultural output and improved infrastructure to integrate markets. However, sustained growth will not in itself be sufficient to alleviate poverty. Some of the poorest households cannot be adequately reached through agricultural growth; many more need also to receive more targeted protection, whether in the form of increased access to economic opportunities or through transfers. The Bank strategy is to rely increasingly on NGOs and private organizations with a proven record for efficient delivery of programs reaching the most vulnerable groups. The proposed Food Security and Nutrition Project is a key component of this assistance strategy; it will address the problems of food insecurity and malnutrition affecting the poorest groups in the country, and will do so by integrating NGOs and local communities more firmly into the process of executing and delivering services. By alleviating the worst forms of poverty, it would also help the country recommence its adjustment efforts and complement on-going social sector projects in Education (FY90), Health (FY91) and Vocational Training (FY92). II. THE PROPOSED PROJECT A. Project Objective and Design 2.1 The project objective is to reduce food insecurity and malnutrition in the two most food- insecure provinces through income generating projects and targeted nutrition programs. The project will operate outside the central administration and its field activities will be mostly executed by NGOs, community groups and the private sector. A comprehensive IEC program to create awareness and educate the population of the project objective would be carried out while beneficiary assessment surveys would provide feedback from the target population. 2.2 The project has been designed to address those aspects of food insecurity that would not be tackled by adjustment operations or investment operations in agriculture, infrastructure and the social sectors. It has also been designed to integrate NGOs more firmly in the execution and delivery of social and economic services. The government has been preparing a national program called SECALINE (LECurite ALImentaire et Nutrition Elargie) to address these objectives and the proposed project will support this national program. SECALINE will include income generation activities including food-for-work specifically targeted to address the acute food insecurity problem in Greater Antananarivo and a nutrition program emphasizing activities to prevent malnutrition complemented by nutrition education designed to help households use available resources more efficiently. These actions would be complemented by an IEC campaign, and by support for a national Iodine Deficiency Disorders (IDD) Control Program. 2.4 Experience elsewhere in Africa shows that successful nutrition programs are those that are combined with income generating activities. To assure synergy between the nutrition activities and the income generation activities included in the project the following links have been developed: (a) both activities target the same provinces (i.e. Antananarivo and Toliary); (b) nutrition workers would be tools including a manual to help them support the community in developing income  - 7 - generation activities proposals for the FID; and (c) the regional coordinator for the nutrition program will maintain close liaison with the Regional Director of the FID and also serve on the regional projects advisory committee. 2.5 Progress in achieving project objectives will be assessed against monitoring indicators that have been developed and agreed upon with the government and FID. These indicators are given in Table 3.1 and include: (a) number of contracts with small enterprises for labor intensive infrastructure projects financed by FID; (b) number of contracts with NGOs for sustainable income generating projects financed by FID; (c) person-days of temporary employment created by the Food- for-Work program; (d) number of children, by province, served by community nutrition centers; and (e) number of children and mothers of child-bearing age receiving iodine capsules. B. Project Description Income Generating Activities 2.6 Background and Justification. As already indicated, the principal underlying cause of Madagascar's food insecurity is unequal access to income earning opportunities by the poor. This component has the dual objective of providing short-term employment to the poor and also helping poor households and communities develop sustainable income generating activities. The income generating activities will be divided into two major sub-components: FID and Urban Food-for-Work. 2.7 FID. ($10.8 million) The Fonds d'Intervention pour le Developpement (FID) will finance on a grant basis small projects and activities sponsored and undertaken mostly by NGOs and local communities. FID will be a private non-profit association operating under a Board of Directors elected or appointed by its members. It will receive project funds under a financing agreement with the government. FID would specifically target the most vulnerable segments of society and would initially fund projects in the provinces of Antananarivo and Toliary. The projects would be selected according to specific criteria, and the internal administrative and financial management structures and procedures would reflect those of the private sector and not those of the government. Annex I of this report provides details on project selection criteria, legal structure and internal financial management and administrative procedures to be used by FID. 2.8 FID will finance construction and rehabilitation of basic infrastructure, income- generating activities and private sector support progranis, including the development of grass-root community groups and local NGOs. No projects over 100 million FMG (about US$55,000 equivalent) will be financed nor will FID finance recurrent activities, emergency projects, projects already receiving funding from other sources, projects eliminated from government or local authorities investment programs, projects that have a negative impact on the environment, and projects that are too sophisticated or heavily mechanized. As a general rule, the FID will not implement projects itself, but rely on "operators", although the modalities of implementation will vary depending on the types and costs of projects. The NGOs, local grassroots groups and communities requesting funding for income-generating activities, private sector support programs and small infrastructure projects (costing less than FMG 15 million), will be responsible for implementation and will enter into contracts, if required, with consultants, artisans, small enterprises and suppliers. Their rights and responsibilities as operators (including the schedule of disbursement of the grant, their counterpart contribution, if any, and the procurement procedures that they will have to follow) will be clearly defined in their contract with FID (protocole d'accord). For those types of projects, FID will have a financing and monitoring role. For all other infrastructure projects, the FID will be "Maitre d'Ouvrage Delegue", and will in all cases enter into contracts with small and medium scale  enterprises for the works. The FID will also contract with "operators", such as NGOs, technical assistance projects and regional and local units of technical ministries to act as "MaItre d'Oeuvre"; however, for infrastructure projects costing more than FMG 50 million (about US$28,000 equivalent) for which no operators can be identified, the FID will use local consultants as "Maitre d'Oeuvre". FID will provide grants only and will not be a lending agency. However, for income-generating activities, the operator will set up and manage a credit scheme for all investments benefitting individuals and households, in order to have a revolving fund and ensure the sustainability of the activities and programs. During the project period it is estimated that FID will help to create: (a) about 2 million person-days of temporary employment on infrastructure projects; (b) provide about 100 contracts to small construction enterprises on infrastructure projects; (c) support 100 different NGO-sponsored income-generating activities; and (d) help create permanent jobs through the private sector development program. These estimates are tentative as FID will be demand-driven and it is difficult to predict in advance the distribution among various types of projects. Year by year quantitative targets for this component are given in Table 3.1. 2.9 Urban Food-for-Work. ($7.3 million) This program would address the acute food insecurity in Greater Antananarivo by providing short-term food transfers to the poorest population groups who will be recruited to maintain, rehabilitate community infrastructure in their own neighborhoods. The program will be an expansion of the World Food Program's (WFP) existing urban project and will intervene in 90 of the capital area's poorest fokontany (neighborhood/village) during the first three years and may be extended to other fokontany in Antananarivo and other cities during the remaining two years of the project. The works will: (a) be simple to execute, (b) require non-skilled labor, and (c) be executed by the respective fokontany. WFP will be providing most of the resources for the program along with some management assistance (IDA will not finance food). An IEC campaign focusing on nutrition education will be targeted to the workers. The poorest groups have been identified by a 1990-91 survey and further surveys to target the population will be carried out during execution. The program would create over 3 million person days of employment benefiting over 60,000 workers. The experience of the initial pilot project indicates that unemployed women workers will be the major beneficiaries of this employment. Year by year quantitative targets for this component are given in Table 3.1. Annex II provides further details on the urban Food-for- Work program. 2.10 Role of IEC in Income-Generating Activities. IEC will be used to promote the FID and will have a role at three levels: (a) to inform the target populations about the project and on how to prepare proposals which will be eligible for FID financing; (b) to inform potential operators/groups of the role they can play in assisting the target populations to get financing; and (c) to "sell" the FID to policy-makers, and national and local opinion leaders. In order to achieve these objectives, the project will have to develop a multi-media communication strategy which will not be based on the mass media which in any case do not reach the target populations. The details of the proposed strategy are included in Annex V. IEC will also be used to promote the Food-for-Work program. A three pronged communication strategy will also be developed in collaboration with WFP. Nutrition Program 2.11 Background and Justification. Madagascar's high child malnutrition rates which are compounded by seasonal variations even in food surplus areas provide strong justification for a nutrition program that focuses on prevention through timely food supplements and nutrition education. It is also apparent that a pure feeding program, while it would prevent malnutrition in the short run, would have little long-run impact. The Programme Communautaire de Nutrition (PCN) or Community Nutrition Program is designed to help households make more efficient use of the  -9- resources available to them to prevent chronic malnutrition among their children. Feeding supplementation would be a secondary objective of the program and cases of acute malnutrition will need to be referred to nutrition rehabilitation centers. These programs would be run by the community with support from an NGO contracted by the project's regional office. Nutrition activities also include an Iodine Deficiency Disorders (IDD) Control Program or Lutte Contre les Troubles DL2s aux Carences en lode (TDCI). 2.12 Community Nutrition Program. ($5.5 million) The community nutrition agents or Agents Communautaires de Nutrition (ACN) are at the heart of the program and are chosen by the community groups. They would organize mothers from the community with the support of a community nutrition committee. The ACN would provide the following services to the community: (a) nutrition surveillance of children 0 to 3 years of age to (i) identify malnourished children and (ii) gather information for community analysis and assessment of the problem; (b) nutrition education using participative training methods; (c) help the community become more organized; (d) provide basic information on the FID and refer the community to the FID; (e) help moderately malnourished children with a supplementary feeding program; and (f) refer severely malnourished children to the nutrition rehabilitation centers run by the Ministry of Health. An agreement with the Ministry of Health will be reached for monthly medical supervision of moderately malnourished children. 150 community nutrition centers will be set up in the province of Toliary, and food will be provided to 7 existing nutrition rehabilitation centers. In Antananarivo Province 100 community nutrition centers will be set up and food will be provided to 2 existing nutrition rehabilitation centers. A manual detailing how the program would be run (Livret du PCN) has been prepared for the ACN and another one has been prepared for the supervisors. Another booklet will be given to the ACN on the FID (with examples of projects that could have an impact on household food security and nutrition, and basic information on the Fund). Supervisors have been identified by NGOs, and a small regional project staff would be employed to supervise performance of the NGOs and make spot checks of the community nutrition program. Year-by-year quantitative targets for this component are given in Table 3.1. More details on this component is provided in Annex III. 2.13 Iodine Deficiency Disorders (DD) Control Program. ($0.7 million) The short term strategy is to provide iodized oil capsules through the health system to children 0-4 years of age and to women of child-bearing age while the long-term strategy is to iodize locally produced salt. The program includes an operation research component to help keep it on track. The program has been designed by the Ministries of Health and of Scientific Research and the Ministry of Health has appointed a National IDD Coordinator. 876,000 capsules will be distributed to children and 374,000 to pregnant women. Furthermore, all other women of child-bearing age will be targeted whenever they come into contact with the health services - specially during clinic visits or when they bring their children to vaccinated. The targets for Year by year quantitative targets are in Table 3.1. The project will finance iodine control activities through an executing agency arrangement with UNICEF. A summary of the IDD control program is provided in Annex IV. 2.14 Role of IEC in support of Nutrition Activities. IEC will support nutrition activities by: (a) promoting PCN among the target population; (b) supporting a program to increase awareness of infant malnutrition; (c) supporting nutrition education through "social marketing"; and (d) promoting (i) awareness of Iodine Deficiency Disorders and (ii) use of iodized salt. Results from the Knowledge, Attitude and Practice (KAP) surveys and the beneficiary assessment surveys will be used to develop IEC messages. Before being used in IEC campaigns, these messages will be tested among the target population to ensure that they are well understood and that they convey the appropriate concepts. The linkage between income-generating activities supported by FID and the community  - 10- nutrition program will be addressed in IEC messages in order to create synergy between FID and Community Nutrition Program. Institutional Development Support 2.15 The project will support: (a) execution of an IEC program which would be underpinned by beneficiary assessment surveys; (b) preparation of a long-term National Food Security Strategy; and (c) coordination of nutrition and poverty reduction efforts at regional and local levels by setting up formal and informal structures. 2.16 IEC Program. ($0.8 million) The IEC program will support all the components of the project providing the links between the various components to ensure that various activities are complementary and that messages disseminated by the various components are harmonious. To ensure this, the SECALINE IEC specialists (one in each province and one in the central office) will develop a communication strategy for the project as a whole with provincially tailored activities. The objectives of this communication strategy will be to: (a) inform decision makers about the project and its objectives in order to motivate them to promote and support the project; (b) inform NGOs about the project and their expected role in order to motivate them to assist community groups to develop projects for the FID; and (c) mobilize potential beneficiaries in order to encourage them to participate in and benefit from the various programs included in the project (FID, PCN, Food-for-Work). In implementing this strategy care will be taken to ensure that: (a) materials used for communications activities will be developed according to a proven methodology centered on the target audiences; (b) messages concerning the various components of the project are harmonious and complementary; (c) communication activities are monitored on a regular basis so that rumors of misinformation are dealt with rapidly and that the project continues to be perceived positively; and (d) communication among the personnel working on various components are harmonious and help to promote project objectives. The IEC component will not only provide support to each of the project component, but also will be an useful project management tool. Further details are in in Annex V. 2.17 National Food Security Strategy. ($0.3 million) This would help Madagascar design a program of policies and actions to improve the country's food security in the long term. This strategy should cover: (a) macroeconomic policies as they affect growth and poverty; (b) agriculture and other sectoral development policies to assure that sectoral growth objectives are combined with poverty reduction; (c) policies and actions to improve the basic health and nutrition status of the poor; and (d) policies and actions designed to improve access by the poor to income earning opportunities. The development of this strategy will be the responsibility of the National Coordinator and the Comite Inter-ministeriel de Coordination du Projet SECALINE (CICS) with support from international and national experts financed by the project. A draft strategy would be discussed in a national seminar and the seminar recommendations would be presented to the authorities for adoption by the government. Prior to Board presentation, a letter of development on food security policy dated February 4, 1992 was received from the government (paragraph 5.2). The text of this letter is given in Annex VII. 2.18 Local Coordination. The community nutrition program has been prepared in each of the two target regions by working groups including NGOs, private sector and govermnent representatives. These groups would serve as advisory groups which would: (a) advise the Regional Coordinator; (b) help avoid duplication of activities; and (c) help the project management interpret results from the beneficiary assessment surveys and reorient the project accordingly.  - 11 - C. Project Costs and Financin Costs 2.19 The project will be implemented over a four-year period. Total costs for the proposed project are estimated at FMG 58,297 million (US$32.4 million equivalent). Base costs are estimated at US$28.5 million equivalent and contingencies amount to US$3.9 million equivalent (12 percent of total project costs). In addition, taxes amount to US$ 0.9 and import duties amount to US$ 1.1 (not included in project costs above). Foreign exchange costs account for US$12.4 million or 38 percent of total project costs. Project costs by category of expenditure are summarized in Table 2.1 below and detailed cost tables are provided in Annex IX. Table 2.1 Accounts Costs Summary (in US$ million) Local Foreign Total I. INVESTMENT COSTS A. Equipment, Vehicles, Furniture, Material 1.1 2.3 3.3 B. TA, training and Studies 3.1 1.5 4.6 C. FID projects 6.4 1.6 8.0 D. Infrastructure Rehabilitation 2.0 - 2.0 E. Food 1.1 4.5 5.6 F. Transportation, storage and Handling 0.9 - 0.9 G. Project Preparation 0.5 1.5 2.0 Total INVESTMENT COSTS 15.1 11.3 26.3 Il. RECURRENT COSTS A. Operating Costs 2.0 0.2 2.2 Total BASELINE COSTS 16.9 11.5 28.5 Physical contingencies 0.8 0.4 1.2 Price Contingencies 2.2 0.5 2.7 Total PROJECTS COSTS 20.0 12.4 32.4 Source: Appraisal Mission November 1992 * Figures may not add up due to rounding 2.20 Contingency Allowances. Physical contingencies have been calculated at 10 percent for equipment, vehicles and materials. The following price contingencies have been included: (i) on foreign exchange expenditures 3 % per annum and (ii) on local cost expenditures 10% in 1994 and 8% per annum thereafter. 2.21 Estimates for equipment, training, studies, and vehicles are based on previous experience with other IDA-financed projects; estimates for food (including transportation and storage) were prepared by WFP. Local salaries and travel allowances are based on current UNDP scales. 2.22 Project costs by component are shown in Table 2.2 below. The FID accounts for 38 percent of base costs, the Food-for-Work for 25 percent of base costs, the nutrition program for 22  - 12 - percent of base cost, the IEC and food security strategy for 4 percent of base costs, project management for 4 percent, and the PPF and Japan Grant for 7 percent of base costs. Table 2.2 Project Cost Summary (in US$ million) Local Foreign Total A. Income Generation Activities 1. FID Projects 8.0 2.8 10.8 2. Food-for-Work 3.7 3.5 7.3 B. Nutrition 1. Community Nutrition Toliary 2.2 1.3 3.6 2. Community Nutrition Antananarivo 1.1 0.8 1.9 3. Iodine Deficiency 0.1 0.6 0.7 C. Institutional Development 1. IEC and Food Security Strategy 0.6 0.4 1.1 2. Project Management 0.6 0.5 1.1 D. PPF and Japan Grant 0.5 1.5 2.0 Total BASELINE COSTS 16.9 11.5 28.5 Physical Contingencies 0.8 0.4 1.2 Price Contingencies 2.2 0.5 2.7 Total PROJECT COSTS 20.0 12.4 32.4 Source: Appraisal Mission (November 1992) * Figures may not add up due to rounding 2.23 Incremental Recurrent Costs: During the project period recurrent costs average $500,000 per year for a total of $2.2 million. The long-run recurrent cost impact of the project will be minimized because: (a) the FID projects will support (i) development of sustainable income earning opportunities, (ii) infrastructure rehabilitation as a priority over new infrastructure, and (iii) new infrastructure only when community groups have demonstrated their willingness to organize maintenance; (b) the Community Nutrition Program will emphasize preventive approaches to malnutrition and nutrition education programs would focus on how to improve the efficiency with which existing resources can be used by the households; (c) the Iodine Deficiency Control Program will stop subsidizing salt iodination once the health benefits have become apparent and households are willing to pay extra for iodized salt; and (d) the Food-for-Work program will support maintenance of community infrastructure. Furthermore, all the food costs in the program are covered by WFP grants. 2.24 Prgect Sustainabiliqy. The project will help develop NGO and private sector capacity to deliver social services and strengthen institutional capacity of local communities and grass-roots groups to carry out development projects. The Community Nutrition program will provide nutrition education directed at improving the efficiency with which households use available resources. The  - 13 - recurrent costs of the community nutrition program are low and it is anticipated that eventually the executing NGOs will be able to run the program themselves. The FID will help the vulnerable households secure access to training and equipment to develop sustainable income generation activities. The FID will also fund labor intensive works such as water supply if the communities agree to provide for maintenance. Overall, the project is designed to help the poor acquire productive assets and skills so that they may become more productive and benefit from economic growth. It should be noted that parts of the project such as Food-for-Work are transfer programs targeted to the most vulnerable and are not sustainable. Nevertheless, in view of Madagascar's deep- rooted poverty, these are necessary to sustain the poor until strong economic growth removes the need for these types of programs. Financing 2.25 Of the total project costs of US $32.4 million, IDA would finance US$21.3 million. Only parallel financing is envisaged under the project. WFP is expected to contribute US$7.3 million which will finance: (a) all foodstuffs under the project; (b) 50% of the internal food transport costs; (c) tools for the food-for-work program; and (d) cooking equipment for the community nutrition program. The government is expected to contribute US$1.7 million and the beneficiaries US$1.2 million. UNICEF will contribute US$400,000 for the iodine component. A Japanese Grant (US$500,000) has been made available to finance local and international consultants for the preparatory work for the project. In addition, the government will ensure that taxes (US$900,000) and import duties (US$1.1 million) are covered. Table 2.3: Financing Plan by Disbursement Category in US $ imillion IDA World Fd__ UNICEF J L| B nifJ Government j Total I Pfr_ I I Am_oun % _ A=zA Ammt % A Amunt % Amount % Amotnt % Amount S A.Eqpt..Veoi., 3.0 73 0.8 12 0.4 9 0 0 0 0 0.0 0 4.1 13 Furn. ,Supp. I B. Nuthi(tm 1.2 70 0 0 0 0 0 0 0.8 30 0 0 2.6 S Ctr. Rehab. C.Food 0 0 6.1 100 0 0 0 0 0 0 0 0 6.1 19 D.Trmng, 5.2 100 0 0 0 0 0 0 0 0 0 0 6.1 19 TA & Studies EFood 0.4 50 0.4 5D 0 0 0 0 0 0 0 0 0.9 3 F.FID 6.8 85 0 0 0 0 0 0 0.4 5 0.8 10 2.0 25 Projedt G.Opering 1.9 68 0 0 0 0 0 0 0 0 0.9 32 2.6 l Cots_ H.Poect 1.5 75 0 0 0 0 0.5 25 0 0 0 0 2.0 6 ToWI 21.3 66 7.3 2 0.4 1 0.5 2 1.2 4 1.7 5 32.4 100 Disbursement I I - ourte: AppraisS Mission, 11 orer, 1992  - 14 - m. PROJECT IMPLEMENTATION A. Status of Project Preparation 3.1 The proposed operation was identified in July, 1990 and project preparation was carried out with financial support from a PPF advance which was provided in February 1991, while the Japanese government provided two grants to support project preparation. The project was cleared for appraisal in June 1991 but the appraisal mission was interrupted by political developments and project preparation was stopped. In January 1992, the interim government abolished the government directorate responsible for the project and decided to set up a project structure outside the central administration reporting directly to the Prime Minister. The new structure was approved in June 1992 and the National Coordinator was named. The new project management team has been effective and a Bank mission worked in close collaboration with project management to help define the final phase of preparation. Project preparation is now almost complete. Key staff have been recruited and are performing satisfactorily. Appraisal took place in November 1992 and a supplemental PPF advance has been provided to support pilot activities and training until credit effectiveness. The transitional authorities have shown strong support for this project and for independent project execution. Support for this type of poverty alleviation project is widespread across the political spectrum. B. Project Management National Level 3.2 While many of the projects's activities are to be managed at the provincial level, the National Coordinator, who reports to the Prime Minister and the inter-ministerial coordination committee or Comite Inter-ministeriel de Coordination du Projet SECALINE (CICS), will have several key responsibilities including coordination between the FID and other project components. The National Coordinator will be responsible for: (a) coordinating all project activities; (b) supervising the regional coordinators; (c) liaising with FID; (d) liaising with WFP on Food-for-Work Program and supervising the executing agency recruited to manage the program; (e) coordinating the development of a national food security strategy; (f) organizing the project financial management, reporting and auditing system; (g) handling international procurement as well as large scale domestic procurement; and (h) recruiting and supervising personnel. The National Coordinator will delegate other responsibilities to the Regional Coordinators particularly for executing the Programme Communautaire de Nutrition (PCN). The National Coordinator's office or Bureau d'Ex&ution du Projet SECALINE (BPS) will be staffed by an Office Director responsible for procurement, day-to-day management and backing up the coordinator as necessary; by an administrative and financial officer; by an Accounting Officer; by an IEC specialist and a Monitoring and Evaluation (M&E) specialist; and necessary support staff. In order to assure that the most competent staff are recruited and to ensure a high level of productivity, staff will be recruited on one-year renewable contracts in most cases, and their compensation will be in line with the UNDP scales for Malagasy staff. Terms of reference for key staff are in Annex VI which addresses project management aspects. During negotiations, assurances were received that the positions of National Coordinator, Office Director, Administrative and Financial Officer, Accounting Officer, and IEC and M & E Specialists in the BPS, will be filled at all times with personnel whose qualifications and experience are satisfactory to IDA (paragraph 5.4). 3.3 The FID has been set up as a non-profit association operating in accordance with the provisions of its Articles of Agreement, By-laws and Manual of Procedures, and of the Framework  - 15 - Agreement between FID and the government. FID will operate under a Board of Directors appointed by its members. The Board of Directors will appoint an Executive Director from a short-list to be prepared by a management consulting firm after an international search among Malagasy with strong management skills living either in Madagascar or outside the country. The Executive Director will appoint Regional Directors for the regions of Antananarivo and Toliary who would be responsible for promoting, appraising and implementing projects within their region. In accordance with the provisions of the Manual of Procedures, projects submitted for financing would have to be approved by the Board of Directors, though projects costing less than 50 million FMG may be approved by the Executive Director and those costing less than 10 million FMG by the Regional Director. The operators (i.e. the NGOs and other entities requesting funding) would be supervised by FID staff. FID staff would also be on hand to provide technical assistance, training and other support necessary to the operators. During negotiations, it was agreed that at all times during project implementation: (i) the articles of agreement, the manual of procedures and the by-laws of FID, and the Framework Agreement between the government and FID will be acceptable to IDA, and (ii) the positions of Executive Director and Regional Directors of FID will be filled at all times with personnel whose qualifications and experience are satisfactory to IDA (paragraph 5.4). Prior to negotiations, the government and FID provided satisfactory evidence of the constitution of the Association with articles of agreement and by-laws acceptable to IDA (paragraph 5.1). The draft of the Framework Agreement and the the Manual of Procedures were discussed and agreed to during negotiations (paragraph 5.3). Prior to Board presentation the decree recognizing FID as a public service association (association reconnue d 'utilite publique) was signed (paragraph 5.2). The signing of the Framework Agreement, the formal adoption of the Manual of Procedures by FID and the appointment of an Executive Director and two Regional Directors for FID would be conditions of effectiveness of the proposed credit (paragraph 5.5). 3.4 The Food-for-Work program will be executed in cooperation with the target fokontany (village or neighborhood local government) with technical and logistical support from an International Labor Office (ILO) technical assistance team which would be recruited to manage the program. The ILO team will plan, execute and manage all the logistics with support from WFP Project Officer and the BPS. The actual transportation will be contracted to private firms. 3.5 The Iodine Deficiency Disorders control program would be executed by UNICEF working in collaboration with the Ministries of Health and Scientific Research. Private sector salt manufacturers and distributors have been involved in preparation and will receive support enabling them to iodize salt. 3.6 The IEC campaign will be managed in partnership between the National Coordinator and the Regional Coordinators with the IEC specialist in the BPS being responsible for assuring overall coordination. The beneficiary assessment surveys which will play an important role in determining the contents of the IEC messages will be carried out by teams working under the Regional Coordinators. The National IEC specialist, in addition to being responsible for coordination, will focus on the national aspects of the IEC campaign and help the National Coordinator communicate project objectives to decision makers and to the population at large. The Regional IEC specialists will focus on developing and communicating messages appropriate to the area working closely with the NGOs executing the PCN and the regional offices of FID. IEC communication materials will be produced under contract by Malagasy entities, public or private, under supervision of project IEC staff. 3.7 The National Coordinator and the Comite Inter-ministeriel de Coordination du Projet SECALINE (CICS) will be responsible for coordinating the development of a long-term Food Security  - 16 - Strategy for the country. He will recruit an international research institute or consulting firm with expertise in the area and attach the group to a Malagasy research institute. The international experts and CICS will develop a draft strategy for presentation and eventual adoption by the govermment. Regional and Community Level 3.8 The Regional Coordinators have been appointed by the Prime Minister and report to the National Coordinator who delegate to them: (a) execution of PCN in the region; (b) management of the regional IEC campaign; and (c) supervision of the beneficiary assessment surveys. The Regional Coordinator will provide all necessary support for FID operations within the province and assure coordination between FID and the PCN; he will also be a member of the Regional Advisory Committee of FID. The Regional Coordinator will be assisted by a Regional PCN Specialist, a Regional IEC Specialist and an Accountant, in addition to necessary support staff. During negotiations, assurances were received that the positions of Regional Coordinators, PCN and IEC Specialists and Accountants, will be filled at all times with personnel whose qualifications and experience are satisfactory to IDA (paragraph 5.4). 3.9 The Regional Coordinator's primary responsibility will be to assure that the PCN is well implemented within the province. He/She will identify potential NGOs and draw-up PCN execution arrangements with them. The NGOs will be responsible for supporting PCN activit es in the field which would be implemented according to the manual (Livret du PCN). The NGOs will visit communities to explain the objectives of PCN and help groups of interested mothers in the communities to organize and run the program. A lead volunteer would be selected by the mothers from among themselves. This lead volunteer would be trained by the project to be a Community Nutrition agent or Agent Communautaire de Nutrition (ACN). The ACN would be given food rations (financed by WFP) in compensation and would be the program's focal point in the community. The cash value of these rations would be more than the prevailing wages in the area and would serve as a strong incentive. The other mothers in the group would also receive some compensation in food for the time they spend doing tasks like cooking, cleaning etc.. The ACNs would be supervised by staff from the NGO who would be provided with motor-cycles. The Regional PCN Specialist would make random inspections of the nutrition centers and supervise the NGOs executing the program. If the work-load becomes too heavy a second Regional PCN Specialist would be recruited after consultations with IDA. A draft community Nutrition Manual was presented before negotiations (paragraph 5.1). 3.10 The ACN, with help from the NGO Supervisor, would mobilize the community to prepare proposals for income generation projects and local infrastructure projects to the FID for funding. The NGO could also become the "operator' (executing agency) for the project and work with the community to execute it. All projects would have be approved by the FID according to the manual of procedures. 3.11 The Nutritional Rehabilitation Centers (CRTs) are managed by the Ministry of Health and are located at or near medical facilities because nutritional rehabilitation requires medical supervision. They would take referrals of acutely malnourished children from the community nutrition centers and send them back to their communities once rehabilitated. Project Implementation Manuals 3.12 A detailed implementation manual has been developed for FID which includes: (a) project selection procedures; (b) project execution guidelines; (c) sample executing agency  - 17 - agreements; (d) sample bidding documents; and (e) internal administrative procedures for FID. This manual, which has been discussed in detail with the government and with FID, will facilitate execution of this component. 3.13 The Livret du PCN also provides detailed guidance on how to execute the community nutrition programs. It is supplemented by model contracts between: (a) the project and the NGOs executing the PCN; (b) the NGOs and the communities requesting nutrition centers; and (c) the NGOs and their supervisors. These documents, which clearly delineate the responsibilities of all involved, were agreed upon during negotiations and will facilitate the execution of the PCN component. 3.14 Guidelines delineating the responsibilities of various parties (Bank, SECALINE project, WFP and communities) involved in the Food-for-Work have also been developed and agreed upon during appraisal. C. Project Monitorine and Evaluation 3.15 A monitoring system will be set up to provide the National Coordinator with prompt and timely answers to the following questions concerning project execution:(a) are activities implemented as planned, on schedule and within budget? (b) are activities reaching the intended beneficiaries? (c) are activities producing the expected outputs? (d) are activities leading to exr xted (immediate) effects? and (e) what is (probably) causing delays or unexpected effects? Monitoring is an essential part of any good management system geared towards management by objectives, which will commonly include the following information-related elements, all of which contribute to monitoring: (a) general accounting; (b) cost accounting; (c) audits (financial, management and technical); (d) periodic management reporting; (e) supervision visits; (f) diagnostic 'problem-solving' studies; (g) beneficiary feedback; and (h) communication (internal and external). It is evident that a monitoring system is closely tied to a regularly audited well functioning accounting and financial management system. The details of the monitoring system are discussed in Annex VII. 3.16 The monitoring system which provides timely information has to be complemented by an evaluation system which examines the impact of the project vis-a-vis its objectives in order to allow informed adjustment of policies and objectives during the course of the project. Evaluation is monitoring at a higher 'strategic level' but with a time-frame of a year or so and since the emphasis is much more on the analysis of beneficiary characteristics, behavior and interaction, the data needs are more complex. Two types of evaluations are planned to allow smooth implementation of this project: (a) on-going spot evaluations of certain activities such as FID projects or PCN impact within a community, and (b) annual reviews, and a detailed mid-term evaluation. 3.17 Table 3.1 shows the targets which should be monitored. These targets are tentaLive particularly for the FID because it is expected to respond to demand for different types of projects and it is therefore difficult to predict the final distribution among them. The targets for the Community Nutrition Program are also likely to change particularly in the out-years as the program will respond to demand from communities rather than rely on vertical planning and execution. The percent coverage for growth monitoring may exceed 100% due to anticipated leakage to older children though individual centers will be evaluated on their performance among the 0-3 year old children. The coverage rate within the immediate catchment area of each PCN site is likely to be over 80% of children 0-3.  - 18 - Table 3.1 Project Monitoring Wdicators Co11pontCRlnM 1993-94 1994-95 1995-96 1996-97 Total MID 1. Small Scale Infmudruct a. Persn Days of Temp.Emplnyment 388,000 446,000 513,000 593,000 1,904,000 b. Coractb witb Snll Enterpris is 21 24 28 91 2. Income Generating Activities Contact with NGOs 25 25 25 25 125 3. Petmunent Jobs Created brough Privte Sector upport projects 300 400 600 600 1,900 Urban-Food-for-Word I. Numberof Beneficiaries 10,4S4 11,368 14,066 16,889 62,807 2. Person Days of Employment 629,062 682,100 843,975 1,013,350 3,168,487 Community Nutuition Toliary 1. Populatin of Trget Fnvondronona 556,253 571,272 586,696 602,537 2. No. of New Caute to be eablihed 30 30 40 50 150 3. No. of Chi;dren in Growth Monioring 18,240 36,840 60,800 91,200 207,080 4. No. of Children in Supplmetal Feeding 2,700 5,400 9,000 13,500 30,600 5. No. of Children Referred for Nutritional Rehabilitation 912 1,824 3,040 4,560 10,336 6.Growth Surveillance Coverage Rate - Percent of Children 0-3 in Taret Fivondronana 27% 53% 86% 126% 7. Percent Reduction in Malnutriion Rate in target fivodrononw 10% 19% 31% Community Nutrition Antananrivo 1. Population of Target Fivonronon 1,038,00 1,066,026 1,094,809 1,124,369 2. No. of New Cente to be etebrib 5 10 35 50 100 3. No. of Children in Gowth Montoring 3,040 9,120 30,400 60,800 103,360 4. No. of Children in Supplemental Feding 450 1,350 4,500 9,000 15,300 5. No. of Children Referred for Nutritiol Rehabilitation 152 456 1,520 3,040 5,168 6.Growth Surveillance Coverage Rate as Percent of Children 0-3 in Taget Fivondronana 2% 7% 23% 45% 7. Percent Reduction an Malnutrition Rate in target fivondronone IX 3 % 8 % Iodine Deficiency Disordes Control 1. No. of Children Receiving Capsle 336,000 120,000 240,000 180,000 876,000 2. No. of Pregnt Women Receiving 93,600 93,600 93,600 93,600 374,400 Infonnation Educatro ad Communication KAP urveys nd Launch natona Second Round (IEC) benediciay IEC campaign to Benefcicuy _ssesab to develop comnunicate Assesment nutrition nessges project objectives Survey _ Natonal Food Security Strategy Identify experts to CICS preent Government support CICS draft Sreat to adopt sategy government 3.18 As part of the supervision of the project, joint IDA/government reviews of project implementation would be carried out on an annual basis to examine progress, discuss the results of the previous year's work program and agree on the following year's work program and budget. During negotiations, it was agreed that the annual review of project implementation will be held not later than September 30 of each year on the basis of appropriate documentation to be prepared by the BPS and FID (paragraph 5.4). A more detailed mid-term review will examine progress on the various project components comparing actual results with anticipated outcomes, in order to decide - - - - - - - - - - - - - - - - - - - -  - 19 - whether and how to continue, modify or eliminate project components or parts thereof. The most important issues to be covered during the mid-term review are discussed in Annex VII. During negotiations, it was agreed that the government will: (i) carry out, jointly with IDA, a mid-term review of the progress of the project not later than June 30, 1996, (ii) submit to IDA a report evaluating the project's progress not later than four weeks prior to such review, and (iii) implement the recommendations of the mid-term review (paragraph 5.4). The recruitment of a M & E Specialist in the BPS is a condition of credit effectiveness (paragraph 5.5). D. Procurement 3.19 Table 3.2 summarizes the project elements and their estimated costs and proposed methods of procurement. Table 3.2: Procurement Arrangements (US$ million) Project Element Total I.C.B. L.C.B. Other N.B.F. Cost 1. Works 1.1 FID Infrastructure 3.7 1.1 4.8 Projects (3.2) (0.9) (4.1) 2.6 2.6 1.2 Nutrition Center (1.8) (1.8) Rehabilitation 2. Goods 2.2 1.0 0.9 4.1 2.1 Equipment/vehicles (1.1) (1.0) (0.9) (3.0) and materials 0.6 1.0 1.6 2.2 FID Equipment (0.5) (0.8) (1.3) 3. Service Contracts 0.9 0.9 3.1 Food Transportation (0.5) (0.5) 4. Services 4.8 4.8 4.1 Technical Assistance (4.8) (4.8) 1.0 1.0 4.2 Training and studies (1.0) (1.0) 1.6 1.6 4.3 FID Training (1.4) (1.4) 5. NiscelLareous 2.9 2.9 5.1 Incremental Operating Costs (1.9) (1.9) 6.1 6.1 5.2 Food 2.0 2.0 5.3 PPF Advance and Japan Grant (1.5) (1.5) TOTAL Baink financed 2.2 8.8 15.3 6.1 32.4 (1.1) (7.0) (13.2) - (21.3) Figures in parentheses indicate amounts financed by 1DA 3.20 The main objective of this project is to alleviate poverty by: (a) reducing food insecurity and malnutrition; (b) providing food in exchange for work; and (c) by putting in place a mechanism whereby NGOs and other grassroots groups can undertake small scale development projects (such as small scale labor intensive public works, training for grass roots groups and development of micro-enterprises). Procurement would be responsive to these objectives. Amounts  - 20 - for civil works, equipment and training for FID projects are shown separately. As mentioned earlier, the maximum cost of a FID project is 100 million FMG or about US $ 55,000 equivalent. 3.21 Country Procurement Assessment. Madagascar's cumbersome procurement regulations were a major constraint to the implementation of Bank-financed projects. This was due to the fact that many ministries performed numerous reviews. In January 1991, Government issued new regulations on public procurement following a comprehensive review of the country's procurement regulations (Country Implementation Review (CIR), December 1990). A follow-up CIR occurred in March 1992, resulting in the Government's decision to use Bank Sample Bidding Documents (Prime Minister's decree No. 892/92, March 6, 1992). This is expected to speed up procurement and project implementation considerably. 3.22 Civil Works. The target group of firms expected to participate in procurement of civil works comprises small and micro-enterprises since they are more likely than larger firms to bid for labor-intensive work projects. Civil Works undertaken through the FID will include: rehabilitation of small roads, reforestation, communal water supply systems, and community infrastructures such as markets and health centers. In order to respond to the poverty alleviation objective of this project, the works will: (a) be undertaken in the poorest areas of the country which are food insecure; (b) be labor-intensive so as to employ as many workers as possible; and (c) be executed by small and micro-enterprises, and artisans. Consequently, ICB is not considered an option for contracting these works because they are small, labor-based, scattered and will be executed by many different entities in remote areas. 3.23 For works procured through LCB, tenders will be advertised locally, bidders would be given 30 days for submission of bids, evaluation criteria will be specified, all bids will be opened in the presence of bidders' representatives, no preference margin will be allowed for national contractors, eligible foreign contractors will be allowed to participate in the bidding process, and standard bidding documents will be used. Contracts for works awarded through LCB for FID projects will range in size from US$10,000 to US$55,000 with a total aggregate amount not to exceed US$3.7 million (IDA US$3.2 million). Smaller FID contracts ranging in size from US$3,000 to US$10,000 will be awarded to local craftsmen on the basis of price quotations with an aggregate amount not to exceed US$1.1 million (IDA $0.9 million). The grouping of these contracts will not be possible because it will involve different operators, different beneficiaries in different locations. For the rehabilitation of nutrition centers, contracts for LCB will range in size from US$10,000 to US$50,000 with a total aggregate amount not to exceed US$2.6 million (IDA US$1.8 million). 3.24 QGods. Equipment, vehicles and materials will be purchased by ICB with an aggregate amount estimated at US$2.2 million (IDA $1.1 million). To the extent possible, contracts will be grouped into bid packages estimated to cost US$300,000 or more. Qualifying domestic suppliers will be given a 15 percent margin of preference, or applicable customs duty (whichever is less) over the c.i.f. prices of competing goods. Contracts ranging in size from US$50,000 to US$300,000 will be procured through LCB with an aggregate amount not to exceed US$1 million (IDA only). Contracts below US$50,000 for furniture, equipment and supplies for the nutrition centers and for the Food-for-Work component (food will not be financed by IDA) will be procured by local shopping procedures acceptable to IDA with a minimum of 3 price quotations up to an aggregate amount not to exceed US$900,000 (IDA only). Regarding equipment for the FID projects, contracts ranging in size from US$10,000 to US$55,000 will be procured through LCB procedures with an aggregate amount not to exceed US$600,000 (IDA US$500,000). Contracts below US$10,000 will be diverse in size and in type of goods (sewing machines, carpentry tools, etc) and  - 21 - will be procured through local shopping procedures acceptable to IDA with an aggregate amount not to exceed US$1.0 million (IDA US$0.8 million). 3.25 Expert Services. Studies. and Training. Technical Assistance (US$ 4.8 million), training (US$ 1 million), FID training (USS 1.6 million) consultants will be selected in accordance with World Bank Guidelines: Use of Consultants by World Bank Borrowers and the World Bank as Executing Agency (Washington, D.C. August 1981). Local training per diems not exceeding US$ 200,000 will be paid in accordance with UNDP local scales. The ILO will be recruited to manage the Food-for-Work program as a single source due to their previous experience with labor intensive works in Madagascar. 3.26 Incremental Operating Costs. US$2.9 million, IDA (US$1.9 million). Whenever possible local shopping will be used for the purchase of office supplieswith an aggregate amount not exceeding US$ 100,000. Each contract will not exceed US$5,000. Local shopping will not be possible for fuel and utilities as their prices are set by government controlled parastatals. 3.27 Bank's Procurement Review. During project supervision, Bank-financed works contracts above a threshold of US$40,000 will be subject to the Bank's prior review procedures. Goods contracts or packages above US$40,000 will also be subject to prior review. Selective post review of awarded contracts below the threshold levels will be carried out on about 1 in 4 contracts civil works and 1 in 10 goods contracts. During negotiations, it was agreed that the BPS and FID will use IDA's standard bidding documents for goods and services and the standard letter for the recruitment of consultants (paragraph 5.4). 3.28 Procurement Management. The BPS will be responsible for overseeing all administrative, managerial and financial aspects of the project including all procurement with the exception of the FID which will handle its own procurement. Local staff have already received formal Bank procurement training abroad and a local procurement specialist has been hired. Bidding documents, technical specifications and quantities will either be prepared by the procurement personnel in the FID or by the procurement specialist in the BPS in collaboration with the respective executing units. 3.29 The implementation schedule is provided in Table 3.3 below, from which the estimated disbursement schedule (Annex X! has been derived.  - 22 - Table 3.3 Implementation Schedule (Amounts in US $ million) Project Element Prr- Project Year Total Remarks project Payment 1 2 3 4 Loan Timlng SigriEffec/Close _ * Civil Works Prequal/Bid/Award ** ** ** FID Infrastr. Project 1.1 1.1 1.2 1.4 4.8 80% LCB; 20% IS/LS Nutrition Center Rehabilitation 0.4 0.4 0.7 1.1 2.6 100% LCB Goods Equipment/Vehicles 1.8 0.6 0.8 0.9 4.1 55% ICB; 25% LCB; 20% IS/LS FID Equipment 0.3 0.4 0.4 0.5 1.6 100% LCB Service contracts Food Transportation 0.1 0.2 0.3 0.4 1.0 100% LCB Consultancies Compl. Std. Bidding Documentb ** Technical Assistance 1.2 1.2 1.2 1.2 4.8 Bank Guidelines for consultants Training 0.3 0.3 0.3 0.3 1.0 FID Training 0.3 0.3 0.4 0.4 1.4 Miscellaneous Incremental Oper. cost 0.7 0.7 0.7 0.8 2.9 100% LS Food 1.1 1.3 1.7 2.1 6.2 Financed by W.F.P. ** ** ** ** 0.0 PPF Advance Refinancing 2.0 2.0 TOTALS 9.3 6.5 7.6 9.1 32.4 Sou rce: AIppraisal Mission, November, 1992 3.30 Procurement information will be collected and recorded as follows: (a) prompt reporting of contract award information by the borrower; (b) comprehensive quarterly reports to the Bank by the BPS and the FID indicating: (i) revised cost estimates for individual contracts and the total project costs, (ii) revised timing of procurement actions, including advertising, bidding, contract award, and completion time for individual contracts, (iii) compliance with aggregate limits on specified methods of procurement, and (iv) compliance of procurement with the manual of procedures as well as systematic review of procurement procedures; and (c) a completion report by the government within six months of the credit closing date.  - 23 - E. Disbursement 3.31 The proposed IDA credit of SDR 15.5 million (US$ 21.3 million) will be disbursed in accordance with Table 3.4 below. Table 3.4: Disbursement Categories (Percentages below refer to amounts without taxes) US $ Percentage 1. Equipment, Vehicles, furniture and other materials 2.5 100% 2. TA, Trg and studies 3.2 100% 3. FID A. FID projects 7.5 85% of total costs B. TA, Training studies 1.5 100% of total cosa C. Equipment, Vehicles, furniture and other materials 0.5 100% of total costs D. Operating costs 0.7 100% of total costs 4. Infrastructure rehab. 1.5 70% of total costs 5. Transportation, Shipping & Handling 0.5 100% of total costa 6. Incremental Operating Expenditures 0.8 70% of total costs 7. PPF Advance 1.5 8. UnaHlocated. 1.1 TOTAL 21.3 3.32 Category 3.D. and category 6 (incremental operating expenses) includes operating expenditures for FID, the community nutrition programs, the food for work program, and the National Coordinator's Office. 3.33 Contracts for goods, civil works and services or single purchases of less than US$20,000 will be submitted against statements of expenditure (SOE). All disbursements will be fully documented by the BPS and FID respectively in a manner acceptable to IDA. The SOEs and all records such as contracts, orders, invoices, and payroll vouchers will be retained by the National Coordinator's Office and the FID who will be responsible for maintaining this information in a manner that would facilitate inspection by supervision missions and for review by the external audit. An accounting structure is in the process of being set up under PPF financing (a financial officer and an accounting officer have already been hired). They will be responsible for monitoring disbursements and compiling the financial information (on a quarterly basis) obtained from the various executing agencies. It is anticipated that the FID will also be recruiting a financial manager and an accountant to perform the same tasks. The accounting officer of the BPS will be the primary liaison between IDA and the borrower for all issues pertaining to disbursements and accounting under this project. The credit closing date will be July 31, 1998. The estimated quarterly schedule of disbursement is in Annex X.  - 24 - 3.34 Special Accounts and Project Accounts. To ensure that funds for this project are readily available and to facilitate disbursements, two special accounts denominated in US$, one for FID and one for the BPS, will be opened in a commercial bank acceptable to IDA. The authorized allocations of the Special Accounts would be US$700,000 for the FID Special Account and US$500,000 for the BPS Special Account, representing IDA's anticipated share of eligible expenditures for three months. IDA would make an initial deposit of those amounts from the proposed credit immediately upon credit effectiveness. Replenishment of the Special Accounts would be made on the basis of full documentation except for contracts valued at less than US$20,000 equivalent, which would be against Statements of Expenditure (SOEs). The documentation for withdrawals under SOEs would be retained at the BPS and FID for review by IDA staff during supervision missions and for regular audits. The Government will open in a financial institution two project accounts (one for FID, and one for BPS for all other project components) to cover its counterpart contribution to the project. Those accounts will be replenished quarterly. During negotiations, it was agreed that those two project accounts will be in a financial institution acceptable to IDA and replenished quarterly based on justification of expenditures (paragraph 5.4). Initial deposits by the Government of the equivalent in Malagasy francs of US$75,000 in each of the two project accounts will be a condition of credit effectiveness (paragraph 5.5). 3.35 Financial Reporting. Accounting and Auditing Arrangements. BPS and FID staff will establish and maintain project related accounts for their respective interventions. They will be responsible for compiling and then consolidating the necessary accounting documents from the various executing agencies. Such accounts shall be maintained in accordance with sound and internationally recognized accounting principles and practices that are satisfactory to IDA. These two entities will provide annual financial statements to reflect the financial performance and position of the project. An auditor's opinion with special emphasis on the audit of SOEs and report satisfactory to IDA on such statements will be provided. The auditor's report will include a statement on the inadequacy or otherwise of the accounting systems and internal controls, the reliability of statements of expenditures as a basis for credit disbursements, and compliance with financial covenants. In the case of FID, there will also be a management audit to ensure compliance with the manual of procedures and a technical audit for quality control. It was agreed during negotiations that the frequency of the audit and the period for the submission of audit reports will be as follows: (a) for the BPS: annual audit, with audit report to be submitted within six months of the close of the fiscal year; (b) for FID: (i) financial audits - quarterly for the first two quarters, then semi-annually, with audit reports to be submitted within three months of the end of the period, and (ii) technical and management audits - every year with audit reports to be submitted within three months of the close of the fiscal year (paragraph 5.4). The foregoing accounting, financial reporting and auditing arrangements should provide adequate and timely information to IDA for project supervision. During negotiations, it was agreed that the government and FID will submit the audit reports with the above mentioned frequency. 3.36 An accounting/audit firm to design the financial management and accounting system was selected before negotiations. This will allow the accounting systems to be fully operational for the BPS and the FID by the time this credit becomes effective. The accounting firm recruited under PPF financing will design a computerized accounting system for all project components taking into account the decentralized nature of project activities. During negotiations, agreement was reached that the proposed accounting system will be responsive to IDA's accounting requirements. The adoption of an accounting and financial management system acceptable to IDA and the appointment of auditors acceptable to the Bank for the first three years would be conditions for effectiveness.  - 25 - IV. BENEFITS AND RISKS A. Benefits 4.1 The proposed project would alleviate poverty and reduce malnutrition. By providing the poor with access to income earning opportunities, it would not only help them escape from poverty but also contribute to national output. Labor intensive works financed by the FID would also have development benefits over and beyond immediate job creation. The project would help improve the capacity of NGOs and community groups. Component specific benefits are described below. 4.2 The FID would benefit the poorest segments of the population by financing: (a) labor intensive works to generate employment in the poorest areas of the country; (b) long term sustainable income generation activities; and (c) training and other support to help develop grass-roots development groups. The direct job creation effect is going to be highest for labor intensive works though the jobs would be temporary. Labor intensive works would also help develop local infrastructure and promote development. FID projects are expected to: (a) through labor-based works, create 1.9 million person-days of employment and provide 91 contracts to small enterprises to help their development; (b) support 25 NGO-run income-generating activities annually targeted to the poorest; and (c) help create permanent jobs through micro-enterprises development. These estimates are based on a certain assumption of the distribution of FID projects, however as FID will be demand-driven, this distribution may change. 4.3 Food-for-Work is designed to be a temporary transfer program to address the acute poverty in Greater Antananarivo but it would help create some community infrastructure such as solid-waste bins. The program would create 4.2 million person/days of employment over five years targeted to the poorest workers (the majority women) in the poorest fokontany. 4.4 Community Nutrition Program. It will take a preventive/public health approach to reducing malnutrition through emphasizing growth monitoring, nutrition education and supplemental feeding to keep nutrition problems under control and to help households better use existing resources. This means that households will be better able to prevent malnutrition even without the aid of the project. By the end of the project 91,200 children will be monitored in Toliary which is almost universal coverage for the 8 target fivondronona and another 30,600 children will have undergone supplemental feeding. Through these activities, the malnutrition rates in the target fivonodronona are expected to be reduced from about 30% to 21%. The objectives for the province of Antananarivo include growth monitoring for 60,800 children in the five target fivondronona which gives a coverage rate of over 60% and 15,300 children 0-3 in Antananarivo will benefit from supplemental feeding. In view of the denser population and lower coverage rate for Antananarivo malnutrition rates are expected to be reduced somewhat less than in Toliary. These figures are indicative and may change as the project intends to follow a beneficiary-led demand-driven approach. 4.5 Iodine Deficiency Disorders Control Program would reduce at a very low cost ($0.11/person) a problem that is prevalent in Madagascar and is the cause of much morbidity. By reducing the prevalence of iodine deficiency related disorders, significant savings in public and private health care expenditures would be achieved. 4.6 IEC program will help to communicate messages designed to improve the food security and nutritional status of the population. The development and adoption of a National Food Security Strategy would help the country reduce its food insecurity in the long-run.  - 26 - B. Risks 4.7 The main risk is implementation delay that could result from inadequate management. This will be minimized by setting up a project management structure outside the administration and following private sector rules. There is a risk resulting from the political transition process which may have a blocking effect on key activities. Several pre-conditions for the project emphasize steps to be taken with respect to project management. Component specific risks are described below. The political risk political is reduced by the fact that this type of poverty program enjoys support across the political spectrum. 4.8 There is a risk of conflict between two potentially strong institutional entities: the BPS and the FID. This risk will be reduced by the National Coordinator of SECALINE being appointed as the Prime Minister's representative on the Board of FID. The execution, supervision and coordination responsibilities described in Annex VI are also designed to avoid such conflict by clearly delineating decision authority and reporting responsibility. 4.9 FID. In addition to implementation delay mentioned earlier, the major risk of this component is its unsustainability in the long-run. The funding for the types of projects the FID would support, particularly the labor intensive works, is heavily dependent on donor support and is unlikely to be financed by the government budget. Another potential risk is that the FID becomes a non-performing public sector agency at the end of the project. These risks are minimized by: (a) targeting the poor by using strict project selection criteria; (b) supporting the development of grass- roots development groups which focus on self-help; (c) financing training and technical assistance to develop sustainable income earning opportunities such as microenterprises; and (d) by setting up the FID not as a public enterprise but as a private non-profit association. 4.10 Food-for-Work. Specific risks includes (a) logistical problems in moving food from port to work-sites; (b) limited capacity of fokontany authorities; (c) poor coordination between the BPS, WFP, the consulting firm/NGO and local authorities; (d) leakage of benefits to the non-poor; and (e) effect of importing about 9,000 tons of rice i.e. about 2,000 tons a year, on domestic producer prices. The logistical risks are minimized by contracting the private sector to store and transport the material. The risk from limited fokontany capacity is minimized by hiring a consulting firm/NGO to provide logistical and technical support. The risk of lack of coordination is minimized by elaborating procedural guidelines setting out the responsibilities of each of the parties. The risk of leakage is minimized by (i) targeting the poorest fokontany and (ii) the difficult and unappealing (cleaning drains etc.) nature of the work. The risk of distorting rice markets is not great because : (i) 2,000 tons represents less than 0.2% of total rice supplies in the market and (ii) the rice will be targeted to the poorest parts of the population who would otherwise not have the purchasing power to buy the rice. 4.11 Community Nutrition Program. Specific risks include: (a) non-responsiveness to the needs of the beneficiary communities; (b) creating dependence on food aid; and (c) logistical problems of running a program in vulnerable and sometimes inaccessible areas. The first risk will be minimized by (i) running the program through volunteer mothers recruited from the community, (ii) recruiting community nutrition worker (ACN) from the community, and (iii) beneficiary assessment surveys. The second risk will be minimized by emphasizing (i) nutrition surveillance, (ii) preventive feeding supplementation as necessary rather than regular meals, and (iii) nutrition education focusing on more efficient use of materials already available to the households. The logistical risk will be minimized by (i) using an NGO familiar with the area to execute the program,  - 27 - (ii) relying as much as possible on locally purchased materials, and (iii) giving as much decision making independence as possible to the ACNs and their immediate supervisors. 4.12 Iodine Deficiency Disorders aDD) Control Program. Specific risks are: (a) private salt producers may become dependent on subsidies to iodize salt; (b) public may be unwilling to pay slightly more for iodized salt; and (c) some may be allergic to iodine. The first would be minimized by phasing subsidies out over time as more and more of the salt in the market is iodized. The second would be minimized by initially subsidizing iodized salt so that it costs the same as non-iodized salt. By the time the subsidies on iodized salt are phased out, the health benefits should be evident. 4.13 Institutional Development. Specific risks are: (a) inappropriate IEC messages; and (b) a long term food security strategy that does not reflect a consensus. These risks are minimized by use of KAP and beneficiary assessment surveys to develop IEC messages and by supporting a national seminar to discuss the recommendations of the working group on food security strategy. V. AGREEMENTS, ASSURANCES AND RECOMMENDATION 5.1 Before negotiations the government and FID provided evidence of the following: (a) Constitution of the Association (FID) with articles of agreement and by-laws acceptable to IDA (paragraph 3.3); (b) Preparation of a Community Nutrition Manual - Livret du PCN (paragraph 3.9); (c) Selection of an accounting/auditing firm for the design and implementation of accounting and financial management systems (paragraph 3.36). 5.2 Prior to Board presentation the government: (a) submitted to IDA a letter on food security policy dated February 4, 1993 (paragraph 2.17); and (b) signed a decree recognizing FID as a public service association (paragraph 3.3). 5.3 During negotiations, the draft of the Framework Agreement between government and the FID and the Manual of Procedures were discussed and agreed upon (paragraph 3.3). 5.4 During negotiations, the following agreements were reached: (a) The positions of National Coordinator, Office Director, Administrative and Financial Officer, Accounting Officer, and IEC and M & E specialists in the BPS, will be filled at all times with personnel whose qualifications and experience are satisfactory to IDA (paragraph 3.2); (b) The articles of agreement, the by-laws and the manual of procedures of FID, and the Framework Agreement between the government and the FID will be at all times acceptable to IDA (paragraph 3.3);  - 28 - (c) The positions of Executive Director and Regional Directors of FID will be filled at all times with personnel having qualifications and experience and satisfactory to IDA (paragraph 3.3); (d) The positions of Regional Coordinators and regional PCN and IEC specialists and accountants will be filled at all times with personnel whose qualifications and experience are satisfactory to IDA (paragraph 3.8); (e) The borrower will carry out, jointly with IDA, an annual review of project implementation not later than September 30 of each year, on the basis of an appropriate documentation that would also include the proposed work program and budget for the following year (paragraph 3.18); (f) The borrower will: (i) carry out, jointly with IDA, a mid-term review of the progress of the project not later than June 30, 1996; (ii) submit to IDA a report by consultants evaluating the project's progress not later than four weeks prior to such review; and (iii) implement the recommendations of the mid-term review (paragraph 3.18); (g) The borrower will use IDA's standard bidding documents for goods and services; and the standard letter for the recruitment of consultants (paragraph 3.27); (h) The borrower will maintain in a financial institution acceptable to IDA project accounts (one for FID and one for BPS) to cover its counterpart contribution to the project and replenish those accounts quarterly (paragraph 3.34); and (i) The borrower will submit to IDA, audit reports as follows: (a) for the BPS: annual audit, with audit report to be submitted within six months of the close of the fiscal year; (b) for FID: (i) financial audits - quarterly for the first two quarters, then semi-annually, with audit reports to be submitted within three months of the end of the period, and (ii) technical and management audits - every year with audit reports to be submitted within three months of the close of the fiscal year (paragraph 3.35). 5.5 The following would be conditions of effectiveness of the proposed credit: (a) the appointment of an Executive Director and two regional directors of FID, the signing of the Framework Agreement between the government and FID, and the adoption by FID of a Manual of Procedures acceptable to IDA (paragraph 3.3); (b) the recruitment of a Monitoring and Evaluation Specialist in the BPS (paragraph 3.18); (c) deposit by the govenment of the equivalent in Malagasy francs of US$75,000 in each of the two project accounts (one for the FID, and one for all other components of the project) for the government counterpart contribution (paragraph 3.34). (d) the adoption by the BPS and the FID of accounting and financial management systems acceptable to IDA (paragraph 3.36); and (e) the appointment of external auditors acceptable to IDA for the audit of the project accounts (including the FID) for the first three years of the project (paragraph 3.36).  - 29 - 5.6 Reommendation. Subject to the above terms and conditions, the proposed project would be suitable for an IDA credit of SDR 15.5 million (US$21.3 million equivalent) to the Democratic Republic of Madagascar on standard IDA terms with 40 years maturity.  -30- ANNEX I ANNEX I: Fonds d'Intervention pour le Developpement (FID) A. General 1. The "Fonds d'Intervention pour le Developpement (FID)" is an independent agency that promotes and finances small community projects and activities for economic and social development that are proposed, initiated and implemented by community groups, NGOs and local authorities. Its aim is to implement, with rapidity and efficiency and in a flexible way, small projects and programs that are responsive to the needs of the poorest groups, and to support the establishment and development of community organizations. The FID is independent of the government structure, and is managed and operated as a private sector organization and in a decentralized manner. B. Legal set up, Organization and Management 2. The FID is a non-profit Association, officially declared as serving the public interest, but managed and operated as a private sector company. In addition to ordinance N