Tunisia - Northwest Agricultural Production Project

Document of The World Bank FOR OMCIAL USE ONLY Report No. 5142-TUN STAFF APPRAISAL REPORT TUNISIA NORTHWEST AGRICULTURAL PRODUCTION PROJECT February 6, 1985 Europe, Middle East and North Africa Projects Department This dooment has a resticted distnrbution and may be used by recipients ily in the performaune Of J teir OUcil. duies. Its contents may not othrwise be dislosed without World Bank authoriation.  CURRENCY EQUrVAIEXNTS Tunisian Dinar (D) 1.00 = US$1.37 US$l.0o = D 0.73 FISCAL YEAR January 1 to December 31 WEIGHTS AND MEASURES 1 kilometer (km) = 0.6214 mile (mi) 1 hectare (ha) = 2.4711 acres (ac) 1 liter (1) = 1.0567 quarts (qt) 1 metric ton (ton) = 2,204.6226 pounds (lb) GLOSSARY OF ABBREVIATIONS APMANE : Small Farmer-Supervised Credit Project in the Northeast (Assistance aux Petits et Moyens Agriculteurs du Nord-Est) BNT : National Bank of Tunisia (Banque Nationale de Tunisie) BNDA : National Bank for Agricultural Development (Banque Nationale pour le Developpement Agricole) CCDA : Central Agricultural Development Conimission (Counissariat central de developpement agricole) CCGC : Central Cooperative for Field Crops (Cooperative Centrale des Grandes Cultures) CIMMYT : International Center for Maize and Wheat Improvement (Centre International pour l'Am6lioration du Ma-is et du Ble) CRDA : Regional Agricultural Development Commission (Commissariat regional de d6veloppement agricole) COCEBLE : Central Cooperative for Wheat (Cooperative Centrale du B1) CTV : Local Extension Center (Cellule territoriale de vulgarisation) DAAF : Department of Administration and Finance (of MOA) (Direction des Affaires Administratives et Financieres) DAPMEA : Department of Assistance to Small and Medium Size Farmers (Direction de l'Assistance aux Petits et Moyens Exploitants Agricoles) DERV : Department of Training, Research and Extension (Direction de l'Enseignement, de la Recherche et de la Vulgarisation)  FOR OFFICIL USE ONLY DGR : Department of Rural Engineering (Direction du Genie Rural) DPA : Department of Animal Production (Direction de la Production Animale) DGPC : Highway Department of the Ministry of Public Works (Direction G&enrale des Ponts et Chaussees) DPSAE : Department of Planning, Statistics and Economic Analyses (Direction de la Planification, des Statistiques et des Analyses Economiques) DPV : Department of Crop Production (Direction de la Production Vegetale) DRGR : Center for Rural Engineering Research (Centre de Recberche dui G6nie Rural) IFAD : International Fund for Agricultural Development (Fonds International pour le D6veloppement Agricole) INRAT : National Institute of Agricultural Research (Institut National de Recherche Agronomique Tunisien) MOA : Ministry of Agriculture (Ministere de 1'Agriculture) MOE : Ministry of Public Works (Minis tere de 1' Equipement) MOF : Ministry of Finance (Minist6re des Finances) MOP : Ministry of Planning (Ministere du Plen) oC : National Cereal Board (Office des Cereales) OEP : National Livestock and Range Agency (Office d'Elevage et du Paturage) UCP : Production Cooperative (Unite Cooperative de Production) UJSAID : United States Agency for International Development (Agence des Etats-Unis pour le Developpement International) This document has a resticd 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.  STAFF APPRAISAL REPORT TUNISIA NORTHWEST AGRICULTURAL PRODUCTION PROJECT Table of Contents Page No. I. INRODUCTION .......... ...................................... 1 II. SECTOR BACKGROUND ........................................... I A. Agricultural Planning and Performance ..... .............. 2 B. Sector Issues ........................................... 3 C. Bank Involvement in Agricultural Projects .... ........... 5 III. THE PROJECT AREA. 6 A. Main Features ........................................... 6 B. Productive Potential and Constraints ..... ............... 6 C. Other Projects in the Area .............................. 9 IV. THE PROJECT ........... ...................................... 9 A. Rationale ............................................... 9 B. Objectives .............................................. 9 C. Detailed Description ........... ................. 10 D. Status of Engineering ............. ............... 13 V. PROJECT COSTS AND FINANCING ............................. 13 A. Cost Estimates ................. 13 B. Financing Plan .14 C. Procurement .15 D. Disbursement .16 VI. ORGANIZATION AND IMPLEMENTATION ............................. 18 A. Organization ............................................ 18 B. Implementation .................. ....... 19 C. Credit ....... 21 D. Monitoring and Evaluation ............................... 22 E. Accounts, Audits and Reporting .......................... 22 This report is based on the findings of an appraisal mission to Tunisia in January/February 1984. Mission members included D. Lister (Mission Leader), J.B. Doolette (Agriculturalist), M. Ballesteros (Economist/ Pricing), and B. Baudet (Credit Consultant). The appraisal team was assisted at various stages by Messrs. J-P. Chausse (Economist/Credit), B. Dussert (Financial Analyst), and J. Hall (Livestock Specialist).  -ii- Page No. VII. PROJECT JUSTIFICATION AND RISKS ............................. 23 A. Production, Yields and Technological Chang3 . . ........... 23 B. Markets and Prices ... 24 C. Farmers' Incentives ..................................... 24 D. Cost Recovery .......................................... 25 E. Economic Analysis ....................................... 26 F. Risk and Sensitivity Analysis ........................... 27 VIII. RECOMMENDATIONS .......................................... 27 ANNEX: Plan of Action for the Organization of the Project Extension Services SUPPORTING TABLES Table 1: Project Component by Year Table 2: Summary Account by Project Component Table 3: Estimated Schedule of Disbursements of Bank Loan Table 4: Rural Roads Segments Table 5: Selected Documents and Data Available in the Project File CHARTS 1. Ministry of Agriculture - Organization Chart (World Bank - 26161) 2. Implementation Schedule (World Bank - 26200) MAP: IBRD 17994 - Project Location IMPLEME=NTATION VOLUME - Table of Contents Annex 1: The Project Area Annex 2: Crop Production and Cropping Systems Annex 3: Livestock Development Annex 4: Meat Marketing Annex 5: Agricultural Extension Services Annex 6: Agricultural Finance Annex 7: Wheat, Beef and Milk Pricing Annex 8: Farm Models Annex 9: Detailed Cost Estimates Annex 10: Indicators for Monitoring and Evaluation 2441E/p50  REPUBLIC OF TUNISIA NORTHWEST AGRICULTURAL PRODUCTION PROJECT LOAN AND PROJECT SUMMARY Borrower: Republic of Tunisia Beneficiary- National Bank of Tunisia (BNT) Amount: US$15.0 million equivalent Terms: 17 years, including 4 years of grace, at the standard variable interest rate Onlendinir The Government would onlend US$8.0 million of the Bank loan to BNT on terms and conditions satisfactory to the Bank. These terms and conditions dould be determined on the basis of an action plan, satisfactory to the Bank, covering national credit issues, to be adopted by the Government by June 30, 1985. Disbursement of these funds would be subject to adoption of this action plan. Project Description The proposed project would help alleviate the constraints to increased production of cereals and livestock products in Northwest Tunisia by strengthening extension and research in the area and improving supportive infrastructure. The project would have an import substitution impact and increase farm income. The project would consist of provision of extension, research and livestock services, farm machinery testing, rural roads upgrading, project management, training, technical assistance, studies and credit for on-farm investments, seasonal inputs and input resale centers. The project would raise the incomes of about 4,500 farmers and 24 production cooperatives. The risk that farmers might adopt the new project technology more slowly than expected is minimized by projection of realistic adoption rates and by ensuring the adequacy of incentives and .he availability of inputs on credit.  -ii- Estimated Project Cost.v Local Foreign Total US$ Million -- Extension 1.7 1.2 2.8 Research 0.3 0.3 0.7 Livestock Services 1.1 0.7 1.8 Farm Machinery Testing 0.2 0.2 0.4 Rural Roads 2.4 1.9 4.3 Project Management 0.2 0.0 0.2 Training, Technical Assistance, Studies 0.5 1.3 1.7 On-farm Investments 5.5 5.2 10.7 Seasonal Inputs 2.1 1.1 3.1 Input Resale Centers 0.7 0.3 1.0 Base Cost 14.6 12.3 26.9 Physical Contingencies 1.5 1.3 2.7 Price Contingencies 5.0 3.3 8.3 Total Project Cost 21.1'k 16.8 37.9' Financing Plan Local ForeiRn Total US$ Million - World Bank - 15.0 15.0 Belgium - 0.2 0.2 BNT 5.4 1.6 7.0 Beneficiaries 3.2 - 3.2 Government 12.5 - 12.5 Total 21.1'k 16.8 37, gob Estimated Disbursements: Bank FY 1986 1987 1988 1989 1990 1991 US$ Million --- - Annual 0.7 1.8 3.5 4.0 3.5 1.5 Cumulative 0.7 2.5 6.0 10.0 13.5 15.0 Economic Rate of Return: about 30 percent Staff Appraisal Report: No. 5142-TUN, dated February 6, 1985 Map No. 17994 /a Figures may not add due to rounding. /b Including US$9.7 million of taxes and duties. (SUPERCOPY OF 827Q/p3&4)  STAFF APPRAISAL REPORT TUNISIA NORTHWEST AGRICULTURAL PRODUCTION PROJECT I. INTRODUCTION 1.01 The Northwest Agricultural Production Project would support the development of rainfed agriculture in one of the major production zones of Tunisia as part of the Government's drive to increase food production. It would achieve its objectives primarily through the introduction in the Project area of an effective agricultural extension service, organized along the lines of the Training and Visit (T&V) System, to transfer known improved agricultural technologies to farmers, and other support measures. This would be the first experience in Tunisia of the introduction of a unified professional extension service; similar arrangements are planned to be introduced nationwide after the introduction and testing of the extension system introduced under this Project. 1.02 Following publication of the Bank Agricultural Sector Survey of Tunisia in September 1982,-' the Project was identified in mid-1983 by a joint Bank/FA0-CP mission. The Project aims at contributing to the priority identified in the sector study of improving production from Tunisia's extensive rainfed agricultural areas which have been neglected in the past compared with the irrigated areas. I, SECTOR BACKGROUND 2.01 Almost half of Tunisia's 6.7 million population is rural. Agriculture's share in Tunisia's GDP declined from 24% in 1960 to 12% in 1983. Similarly, agriculture's share in total employment declined from 56% in 1960 to about one third presently. The relative decline of agriculture has been caused mainly by the more rapid growth of other sectors, particularly petroleum, manufacturing and tourism. Over the past fifteen years, however, the value of agricultural production has grown substantially. 2.02 Tunisia has a total land area of 16.4 million ha of which 8.4 million ha are suitable for agriculture and grazing. Forests and esparto grass cover 1.2 million ha, and 2.2 million ha are used for grazing. Of the 5 million ha of cultivable area, 34% is planted in cereals, 35% in fruit trees, 5% in forage crops, 2% in grain legumes, 3% in vegetables and 1% in industrial crops; 21% is normally left fallow. Tunisia's ratio of irrigable to arable land is very low (3.3% in 1977-80), reflecting scarcity of water resources. Tunisia has an irrigation potential of about 250,000 ha of which 205,000 ha are presently irrigable and about 160,000 ha effectively irrigated. Underutilization of irrigation water is a major issue in the irrigation subsector. I/ Report No. 3876-TUN, September 29, 1982.  - 2 - 2.03 Tunisia can be divided into three agro-climatic zones. The northern part (25% of Tunisia's land area) is the most fertile, receiving considerable rainfall (400-1,000 mm). The central zone (15Z of Tunisia) receives between 200 mm and 400 mm of rainfall. Rainfed agriculture in the central zone is limited to fruit tree plantations (mostly olives), low yield cereals, and range. The southern part (60% of Tunisia) is a pre-desert zone receiving less than 200 mm of rainfall with extensive grazing and some irrigated agriculture. The North is Tunisia's predominant agricultural region both for rainfed and irrigated farming, and farmers of the region are generally responsive to innovations. Variations in soil, topography and climate produce a range of potentials and variability between years. Development has been uneven, favoring mainly the development of irrigated agriculture, so that most rainfed areas including the Project area have sizeable unfulfilled potential. A. Agricultural Planning and Performance 2.04 Under the Fifth Development Plan (1977-81), Government's objectives for the agricultural sector included the pursuit of food self-sufficiency (defined as a balanced commodity trade in agricultural products), increased rural employment and income generation, foreign exchange earnings and savings, and economic growth. The investment effort in agriculture as a percentage of total investment corresponded to agriculture's share of value added (13Z). Investments were concentrated in irrigation (442), farm machinery (171), and livestock (121). Overall performance of the sector has been satisfactory, although the pace of growth in recent years (0.92 per annum for 1976-81) has slowed down considerably compared to earlier years (8.21 per annum for 1971-76), strongly affected by drought in recent years. Until the severe drought of 1982-83, growth in cereals (about 41 per annum) and fruits and vegetables (61) exceeded population growth (about 2.6Z per annum) and the 1980-81 cereal harvest of 1.2 million tons was the best ever recorded. 2.05 The Government's Fifth Plan objectives in employment and food self-sufficiency, however, were not attained: the impact of employment- generating investments such as irrigation, livestock, and fruit trees was offset by employment replacing investments in farm mechanization. The agricultural trade balance worsened, because of relative stagnation of exports, a rapid growth of imports, especially cereals, and a deterioration in the terms of trade of agricultural goods. Tunisia's rapid per capita income and population growth, combined with a shift of tastes to higher value crops, created a growth of demand in food, diverting exports to local consumption and increasing imports. 2.06 The basic issues for agriculture under the Sixth Plan (1982-86) have not changed though the need to resolve key issues such as employment and regional income distortions, as well as increased efficiency of investments, has become greater due to a projected decline in the rate of economic growth and the necessity to adjust the economy to a post-hydrocarbon situation of lower investments. The Government is attempting to achieve this adjustment through, inter alia, (a) shifting investments from large, capital-intensive projects to the promotion of employment opportunities in small- and medium- scale enterprises such as in agriculture; and (b) the creation of an appropriate policy environment for removing constraints for farmers, including the introduction of realistic prices for agricultural products. The Government's investment strategy is shown by an increase in the total  investments allocated to agriculture from 13% (1977-81) to 19% (1982-86); and within agriculture, a decline in the share of investment in capital- intensive subsectors such as farm machinery (17 to llZ) and an increase in the share of labor-intensive subsectors such as livestock (12 to 15X) and fruit tree plantations (8 to 172). In pursuit of increased food production, more attention is being given to the improvement of production from the rainfed areas, particularly for grain and livestock production. The proposed resetting of priorities is dependent on increasing absorptive capacity in the more labor-intensive subsectors. The Bank-financed Technical Assistance Project (Loan 2197-TUN) is facilitating the preparation of high priority projects and subsector strategies including policy reforms. B. Sector Issues Land Tenure 2.07 Land distribution is highly skewed. Most farms are small (681 have less than 10 ha), and fragmented (707 have two or more separate parcels). Much of the farm land is owned by absentee landlords who in many cases have little interest in intensification, and by cooperatives which sometimes exploit the land poorly due to lack of managerial skills. Few farmers have land titles. There is no legal protection for tenant farmers which discourages them from investing in agriculture. Land reform has been discussed for many years, but action has been slow and generally confined to the public irrigation areas where land consolidation has begun. In order to accelerate the current program of land titling and provide a better balance of land distribution, the Government is preparing a National Cadastre and Cartography Project for implementation during the Seventh Plan period. Following discussion with the Government of a preliminary subsector report prepared by the Bank, a more detailed study of land issues is being carried out under the Technical Assistance Project (Loan 2L97-TUN). Some land consolidation has also taken place under Bank-financed irrigation projects. Extension and Research 2.08 At present there is no effective national extension service in Tunisia: various projects and commodity organizations have successfully introduced autonomous or semi-autonomous extension systems for individual crops or regions, often relying on special organizational arrangements which are not necessarily permanent. Few activities are based on properly formulated extension messages which respond to the needs of the farmers, or conflicting advice is provided to farmers by competing extension agents. The management systems in place seldom ensure the most efficient use of extension personnel. In addition, the agents perform multiple functions including credit analysis and the collection of agricultural statistics besides their extension work, and are inadequately trained. While some good progress has been made in agricultural research, insufficient emphasis is given to adaptive research. Research is too centralized and has little contact with farmers, thus lacking the critical feedback from farmers needed to establish appropriate priorities. The proposed Project would help reform and strengthen extension and research services in the Project area as the first stage of a national program to be developed under the Technical Assistance Project referred to above for implementation during the Seventh Plan period (1987-91).  - 4- Input Supply and Output Marketing 2.09 Public marketing and input supply enterprises exist for cereals, grain legumes. olives, wine, meat, fertilizer, seed, and farm machinery. Several Government development authorities (Offices) participate in the collection of milk, fruit and vegetables. All these institutions require improvements in their efficiency. They are slow to respond to market signals and to introduce cost-cutting efficiencies. Farm inputs are frequently not available in the quantities and at the time required. Private marketing and input suppliers are not permitted to compete legally with public enterprises for certain commodities (cereals, olive oil, certified seed of high yielding cereal varieties). The proposed Project would improve the collection of milk and the supply of inputs in the Project area, especially seed and fertilizer, partly by encouraging private sector initiative in these activities. An input supply study financed under the Technical Assistance Project has already led to urgently needed increases in fertilizer distribution margins, and an agricultural marketing study, also funded from the Technical Assistance Project, is expected to lead to a series of output marketing measures for implementation during the VIIth Plan period. Agricultural Credit 2.10 The institutional structure of formal agricultural credit in Tunisia is complex. The principal vehicle for medium- and long-term investment credit is the National Bank of Tunisia (BNT). The main sources of short-term credit are BNT (in the case of larger farmers, generally clients of long standing) and the Mutual Guarantee Societies (SCMs). SCMs grant seasonal credit, using BNT funds but with Government guarantee, to small- and medium-size farmers. Supervised credit is also offered by a number of agricultural development projects financed by external donors. Some Government-owned enterprises operating outside the financial sector (e.g., Offices) provide credit in kind. A new agricultural bank (BNDA) was recently created to finance investments by large farmers or agricultural enterprises. Credit demand in the Project area ic currently constrained by the farmers' lack of appreciation of investment opportunities, primarily due to the ineffective extension system referred to above. The ongoing Third Agricultural Credit Project (Loan 1885-TUN) is addressing priority credit problems. Prices and Subsidies 2.11 In an effort to provide food at low cost, most consumer prices are fixed by Government. Traditionally, farmgate prices in Tunisia have also been fixed artificially low compared to world prices, discouraging the use of more productive but higher cost techniques and investment. The exception was for irrigated crops (most fruit and vegetables) for which output prices are free. Farm input subsidies, which the Government has seen as a means of compensating farmers for low producer prices, benefit only a minority of farmers and tend to be provided for the most modern inputs (purebred cows, feed concentrate, irrigation water, fertilizer, pesticide, etc). While this encourages use of these inputs, it also causes excessive use by farmers fortunate enough to obtain the subsidies. The Government recognizes the negative resource allocation, budgetary and income distribution effects of its present pricing  and subsidy policies and one of its objectives is to move towards a pricing system that more closely reflects the real costs and benefits of inputs and outputs to the economy. Some agricultural producer prices (milk and cereals) have thus been increased more rapidly than average price inflation in the recent past. As a result, current producer prices are now for the most part reasonably in line with world prices in nominal terms. Consumer prices for milk and beef have also been increased recently. C. Bank Involvement in Agricultural Proiects 2.12 Bank Group lending for agriculture in Tunisia started in 1967, and to date 17 projects (including two highway projects and one technical assistance operation) have been approved for a total of US$343.2 million of Bank/IDA funds. Of these, seven have been closed. Those remaining consist of three irrigation projects: the Southern Irrigation Project (Loan 1796-TUN), the Medjerda/Nebhana Irrigation Development Project (Loan 2157-IUN), and the Central Tunisia Irrigation Project (Loan 2234-TUN); one Fisheries Project (Loan 1746-TUN); one Agricultural Credit Project (Loan 1885-TUN); one Rural Development Project for the Northwest Region (Loan 1997-TUN); and one Grain Distribution and Storage Project (Loan 2052-TUN). The Bank is also supporting infrastructure and on-farm investments through the Third and Fifth Highway Projects (Loans 1601-TUN and 2108-TUN). The Technical Assistance Project (Loan 2197-TUN) also has a substantial agricultural component including the studies mentioned above. 2.13 Performance under these projects has been mixed. Implementation of the irrigation projects has been the most satisfactory, the irrigation Offices having demonstrated their ability to implement construction works efficiently. On-farm development and the implementation of efficient farm support services have been slower to materialize, partly due to the priority given to construction work, and the non-involvement of farmers in the decisions concerning water distribution and operation of the systems. The First Fisheries Project was completed at the end of 1979 with considerable delays and low loan recoveries for boats. These problems were addressed under the Second Fisheries Project after agreement was reached on a strategy for improving recoveries. The First Agricultural Credit Project financed BNT lending for farm machinery, date palm plantations and dairy development and processing, but was not fully disbursed until 34 months after the original closing date because Bank Group funds were onlent at a higher interest rate than Government funds. The PPAR (No. 2497 of May 11, 1979) indicates that while the project achieved a good rate of return, mechanization did not yield the benefits expected, and dairy processing suffered major cost overruns. Under the Second Agricultural Credit Project, the continuing problem of higher interest rates on Bank funds than on Government funds, and the lack of decentralization of BNT's subloan approval process, caused disbursements to be slower than anticipated. The project closed on December 31, 1982, two years after the initial closing date. The Third Agricultural Credit Project is addressing priority credit problems: it calls for uniformity of lending terms regardless of the source of financing, and requires BNT to increase agricultural staff, establish a staff training program, decentralize appraisal and approval of small subloans to branch offices, undertake annual reviews of the adequacy of its agricultural interest rates compared to price intlation  and make adequate proposals for their modification as necessary, and improve recoveries. Progress so far has been slow but in the right direction. Action has been taken to decentralize BNT's operations, establish training programs, raise interest rates for larger farmers and improve loan recoveries, the last by granting BNT the "privilege du Tresor" allowing overdue loans to be recovered without lengthy legal proceedings. Following discussions of the Bank's Financial Sector Survey and Agricultural Strategy Paper during the first half of CY85, the Government intends to prepare a plan for further action on interest rates and measures to improve loan recoveries (para. 6.13). The physical implementation of the Northwest Rural Development Project has made a good start. The Grain Storage Project is also now making a good start after initial delays and the recruitment of new consulting engineers. Under the Technical Assistance Project, studies are being carried out for several subsectors. The Government also completed a study on milk collection. The proposed Project reflects Drogress on several important issues which have emerged from the preparation of these studies, including milk marketing and fertilizer distribution. _L THE PROJECT AREA A. Main Features 3.01 The Project area (see Map), located in the plains of Northwestern Tunisia, includes some 400,000 ha of some of the best agricultural land in the country. This land is situated in some 39,000 private farms (86% of the land area), 68 production cooperatives (11%) and 11 agro-combinats (3%). The present land use pattern is simple, basically cereals grown in rotation with fallow, with some grain legumes for family consumption. Sheep and dairy cattle in varying proportions are included in the farming enterprise to use crop residues and the natural growth on the land left fallow. Livestock is of secondary importance in terms of land use but currently provides some 25% of the farmers' incomes. B. Productive Potential and Constraints 3.02 The productive potential of the climatic zone within which the Project area is situated has been under field evaluation by Tunisian research agencies since the late sixties. Research recommendations have been tested under farming conditions on the Fretissa experimental and demonstration farm operated -with the assistance of the Belgian aid program since 1969. Technical specialists from the farm and from INRAT have shown that major productivity increases can be obtained by a better integration of livestock and cereal production, as has also already been demonstrated by some farmers in the Project area. Application of known and tested improved technologies can lead to farm level yield increases of some 75% for field crops. For example, farmers adopting recommended improved cultural practices have obtained long- term average yield increases, in the case of wheat of some 750 kg per hectare to reach 1.8 tons per hectare. The potential for improvement in the yield of animal products has also been shown tc be considerable. Farmers introducing improved milk herds have shown a milk yield per animal unit of some 2,500 liters depending on the production system; beef as a by-product to dairying  was doubled from 110 to 220 kg per animal unit; and sheep meat production from an average of 20 to 39 kg per animal unit. Animal numbers per hectare could increase slightly on the small farms but double on the large farms. 3.03 To achieve these increases requires: (i) rationalizing the use of fallow and combining this practice with other methods of soil fertility restoration; (ii) producing fodder and forage as a main source of animal feed, relegating crops by-products to a lesser role and allowing some to be returned to the soil as organic matter, and reducing the dependence on feedstuffs from outside the farm; (iii) using the forage components of the rotation to improve the soil fertility; (iv) capitalizing on the carryover effects of fertilizer application from one phase of the rotation to the next; and (v) further reducing the costs of crop disease and weed management by integrating it with the rotation system. 3.04 Fallow can be reduced to allow for intensification of fodder and forage production, but not eliminated because some fallow is needed to ensure some stability in cereal yields, given the variability of rainfall in the Project area. In the higher rainfall areas, fodder production increases can be obtained by the intensive production of cereal-legume mixtures to be fed as silage or hay. In the lower rainfall areas, this practice can be employed along with planted pastures for forage based on annual self-regenerating forage legumes. Appropriate dairy cattle or sheep production systems have been developed according to location and farm size. Cropping systems are described in detail in Annex 2 and animal production systems in Annex 3. 3.05 The principle constraints on the more widespread adoption by farmers of these improved practices have been: (a) lack of knowledge on how to apply the techniques because of inadequate extension services; (b) lack of means, principally seasonal inputs and machinery, and their timely availability; and (c) lack of consistent and attactive market opportunities for milk and meat. Extension Services 3.06 The general problems with the extension services in the country were described above in paragraph 2.08. Even though there are a number of agricultural technicians performing part-time extension, there are currently only 15 direct contact, multidisciplinary extension agents, operating from local extension centers in the Project area providing a coverage of one extension agent to some 1,000 farmers. This is inadequate even if efficient arrangements were in place for the management and operation of the extension service. Given the relatively large size of farms in the Project area, a ratio of one agent to 250 farmers would be more appropriate. Project area farmers also currently receive differing and conflicting advice from different departments of the Ministry of Agriculture (DPV, DPA, DAPMEA) and commodity agencies. By instituting an adequately staffed, unified extension service, consisting of multidisciplinary extension agents, supported by extension managers and subject-matter specialists, the Project would make it possible  - 8 - to transfer the known technology of cereals/livestock integration to farmers. The Project would also promote improved adaptive research to ensure that progress is maintained in disease control, fertilizer strategies, and integrated weed control programs, and to strengthen the links between research and extension (para. 4.04). Infrastructure and Input Supply 3.07 An adequate network of rural roads is essential to transport inputs (seeds, fertilizers, pesticides, etc.) to the farms, and to send the outputs (wheat, milk, meat, etc) to market. Inputs must also be available at appropriate times. While a basic network of roads exists in the Project area, access is generally restricted to the dry season and in some high potential areas access remains very poor. While ongoing road improvement programs have improved the situation, more needs to be done to provide an adequate level of road access for inputs and outputs. In addition, the inpuit distribution network is inadequate. At present, there are 16 distribution centers in the Project area, providing for only about half the projected demand in the Project area. The lack of centers is partly because fertilizer resale margins have not been set at high enough levels to encourage investment in distribution centers. Recently, however, the Government took action to restore incentives for fertilizer distribution. Fertilizer quantities delivered and projections of the underlying demand in the Project area were established by SWEDFARM in the National Master Plan for Agricultural Input Supply. In this study, it is projected that between now and 1990, fertilizer consumption in the four Project provinces could increase from around 42,000 to 90,000 tons per year, and improved crop seed demand from approximately 7,300 to 11,500 tons per year. Market Opportunities for Milk and Beef 3.08 The most secure market for Project area producers of beef and milk is Tunis, but this involves high transportation costs and few farmers are prepared to travel that far to market their produce. Little milk or beef is thus produced in the Project area beyond that required for home consumption. With no organized livestock markets, only one milk plant of 10,000 liters/day at Jendouba and no milk collection facilities outside the area of the ongoing Northwest Rural Development Project, compared to 25 purchase centers for grain (80% owned by the Office des C6r6ales) the marketing infrastructure in the Project area reflects the traditional grain producing orientation of the farmers. With a view, however, to improving marketing conditions for milk and beef in the Project area prior to the start-up of the proposed Project, the Government has already promoted two initiatives. First, to improve milk marketing, it has licensed the construction by private companies with Government participation of two new dairy plants in the Project area: one of 80,000 liters/day capacity at Bou Salem, the other of 50,000 liters/day capacity at Medjez El Bab. This will greatly facilitate milk marketing in the Project area. The plants are expected to be operational by late 1987 when a consumer price premium for local processed fresh milk is also expected to be instituted. This will provide an additional stimulus to fresh milk production in the Project area. Second, to improve meat marketing, the parastatal company El Louhoum has started construction of meat freezing facilities of 300-ton capacity and refrigeration facilities of 75-ton capacity in Tunisia's second largest city, Sfax. This will greatly increase demand for meat from the Project area. In addition to these measures which have already been taken, the Project would provide an additional milk collection center at El Kef.  -9- C. Other Projects in the Area 3.09 The following projects are currently ongoing in the Project area: (a) The USAID-financed APHANE Project, recently extended to March 1986, which is an assisted-credit project for small- and medium-size farmers covering parts of the provinces of Beja and Siliana, as well as three provinces outside the project area. This project provides some specialized extension advice for its clients, but has involved relatively few farmers in the project area. (b) A Tunisian-Belgian Project which provides technical assistance and equipment to demonstrate control of cereal crop weeds with herbicides. An important Tunisian-Belgian cooperation activity since 1969 has been the pilot farm at Fretissa (para. 3.02) which has provided some of the essential technical information for the design of the proposed' Project. Much of the Project technology has been pioneered at Fretissa for on-farm use and the consistent results have set the standards of yields that can be achieved. Cc) The Bank-financed Third and Fifth Highway Projects (para. 2.12) which are concerned with rural road construction under the management of the DGPC and the construction of some extension infrastructure in the area. The 650 km of roads in the Project area completed or planned under these projects have provided for urgently needed improvements. The facilities developed in the Project area would be fully incorporated into the proposed Project. IV. THE PROJECT A. kationale 4.01 The Project area has good rainfall and soils, capable and responsive farmers, but lacks support in agricultural extension and research, and has shortages of certain infrastructure such as rural roads and marketing facilities. The Project is designed to relieve these constraints, thereby enabli,ng farmers to realize more fully the productive potential of their land. B. Obiectives 4.02 The main objective of the Project is to assist Tunisia to increase production of cereals (mainly wheat) and livestock products (chiefly milk and. mutton/lamb) on private farms and UCPs in the plains of the four northwestern provinces of Beja, Jendouba, El Kef, and Siliana. Since Tunisia imports large quantities of wheat, feedgrains, milk and meat, additional Project production would assist the balance of payments. Farm income would also increase. These objectives would be achieved by strengthening extension and research in the area and improving supportive infrastructure. The Project would be managed by a Project Implementation Unit in the Ministry of Agriculture which would coordinate the Project actions.  - 10 - C. Detailed Description 4.03 Extension (US$4.0 million). The Project would provide for an increase in the ratio of MOA direct contact extension agents located in local extension centers from the current ratio of one agent for every 1,000 farmers to one agent for 250 farmers, the introduction of supporting staff and a management system along lines of the T&V system. This would require the reassignment or recruitment of 51 new extension agents in addition to the 15 agents already in the Project area, and 4 regional deputies for extension to operate in the 33 priority areas covering 200,000 ha selected on the basis of land resource characteristics. On the basis of two agents attached to each extension center (CTV), the-Project would provide 18 new and expand 15 existing CTVs. Standard plans for a new CTV consist of offices, a conference room and two staff houses totaling 200 mz, a storage facility of 50 m2, land and fencing. Additions to existing CTVs would consist of staff housing and fencing. Equipment consisting of one minitractor with implements'and one set of audio-visual material per new and existing CTV, as well as vehicles consisting of one light car for each new extension agent at both new and existing CTVs, two light cars for each of the CRDAs and four minibuses for the transport of farmers, would also be provided. Similar equipment is already in place in other areas and is being used efficiently. The minibuses in particular are needed to transport farmers over the often long distances which separate the farms in the sparsely populated Project area. Incremental vehicle and machinery operating expenses and building maintenance costs are also included. In addition to normal extension activities oriented towards smaller farmers, the Project would establish a farm management center at Beja to advise large (over 100 ha) farmers who would fall outside the focus of normal extension activities, including at least one technical adviser to be financed by Belgian aid. Approximately, 800 out of the expected 4,500 Project participants would fall into the category. Details of the proposed operation of the extension services are given in para. 6.05 below. Deployment of Extension Staff Total Project Year Number Categorv Required 1 2 3 4 CRDA Deputies for Extension 4 '' Extension Agents at CTV Level 66gb 15 17 15 4 /a To be recruited prior to loan effectiveness. lb Of which 15 already in place in 15 existing CTVs. 4.04 Research (US$0.9 million). To support the revitalized extension service and improve its linkage with research, the Project would complete the network of research facilities in the Project area. These consist mainly of INRAT's animal production research station at El Afareg in Beja province and INRAT's field crop research stations at El Kef and Beja. The Project would rehabilitate the El Afareg station by providing a new 600 m2 administration- laboratory building, two small houses for technicians, one new cattle stable  - 11 - for fattening and milk production in experimental animals, rehabilitation of a sheep stable for experimental sheep, construction of a milking shed for experimental sheep, development of 30 ha of land for precision experiments, small-scale farm equipment to conduct crop and pasture experiments, laboratory equipment for the preparation of materials for experiments and samples from experiments, and two sets of farm equipment for the general farm operations which supply animal feed requirements. Minor amounts of essential laboratory equipment to complete facilities for plant disease work at INRAT's El Kef and Beia stations would also be provided, as would funds to study the function, location and physical requirements for future research/extension support stations. 4.05 Livestock Services (US$2.5 million). In support of the objective of increasing production of milk, sheep meat and wool, the Project would provide: (a) construction and equipment of three animal health centers of 280 mz each at Dahmani, Testour and Teboursouk for the DPA; (b) vehicles and equipment for ten dairy cattle artificial insemination posts (3 each in the provinces of Jendouba and Siliana, two each in Beja and El Kef) also for the DPA; (c) construction of one milk collection center of 90 mz at El Kef, including a cooling vat of 2,000 liters, cans, nine can-carrying trucks, and four tank trucks for the OEP; and (d) construction of three animal purchase centers where the OEP would buy store (lean) animals for sale to Project farmers for fattening. Each center would include fencing, housing, offices and storage, as well as the purchase of six trucks for animal transport and three pick-ups. Incremental staff, building maintenance and vehicle operating costs would also be provided for the animal health, artificial insemination and buying center components. 4.06 Farm Machinery Testing (US$0.6 million). To ensure that Project farmers and UCPs can also be advised through the Department of Training, Research and Extension (DERV) of the most appropriate farm machinery to maximize productivity, the Project would set up the nucleus of a farm machinery testing facility under the control of DGR at the advanced school of rural engineering at Medjez El Bab. The Project would provide four man-months of short term and 12 man-months of long-term consultants services to help choose the type and source of farm implements to be tested and devise the methodology for on-farm comparative testing. The consultants would also supervise the implementation of the testing program. The Project would also provide for the purchase of the implements to be tested and small equipment required for the testing program. 4.07 Rural Roads (US$6.3 million). To ensure adequate access to input supply centers and output markets, the Project would provide for the following road improvements under the supervision of the DGPC: (a) upgrading to all weather standard (N2 or N3) of approximately 110 km of priority rural roads not covered by ongoing highways projects, and  - 12 - (b) repairs, culverts and fords for about another 20 km. In keeping with practice under the Third and Fifth Highway Projects, rural roads in the two northern provinces of Beja and Jendouba (58 km) would be upgraded to the higher N3 standard, due to the more difficult rainfall and soil conditions in these two provinces. The roads in El Kef and Siliana (52 km) where conditions are less demanding would be upgraded to the lower N2 standard. Details of the road segments involved are given in Table 4. 4.08 Training, Technical Assistance and Studies (US$2.1 million). In addition to routine training in technical messages, training of extension agents would take two forms. The first would be preparatory training in communications including screening for aptitude for extension. The second would be periodic technical refresher courses. Arrangements for these are established in Tunisia. Provision is also made for overseas training for extension dupervisors and subject-matter specialists. The farm management center (para. 4.03) represents a new approach in Tunisia to dealing with the special extension needs of large farmers for whom the normal extension services cannot cater. Technical assistance will be provided for its establishment and to develop suitable operating procedures. Belgian aid has agreed to provide at least one specialist for this purpose. The services of an extension management consultant would be provided to help make the reorganized extension service operational. 4.09 To ensure adequate research station development under the Project, a research station development expert from one of the international agricultural research centers would be financed using Project funds, and a Tunisian technician would undergo training at CIMMYT's research station development and management training course. Provision is also made under the Project for a study of the need in Tunisia for support stations which would be the stage beyond the regional research stations and where technologies are developed at the farm level and research and extension linkages developed. Provision is made for a research study of the treatment of straw to improve its value as animal feed. A study would also be provided to assist with the development of service cooperatives in the Project area, including pilot activities on two such cooperatives. Funds would also be provided to the DPSAE for the implemen- tation of a monitoring and evaluation system for the agricultural sector. In total, about 110 staff-months of consultant services would be provided. 4.10 Credit Investments (US$21.4 million). To adopt the improved technology to be extended under the Project, private farmers and UCPs would need to make certain incremental medium- and long-term investments (US$15.4 million). Given the expected level of on-farm investment and farmer and UCP adoption, these are expected to consist in the first four years of construction of dairies for approximately 925 cows, improvement of dairies for about 3,100 cows, construction of sheep shelters for approximately 10,700 sheep, lamb fattening corrals for about 10,800 lambs, steer fattening corrals for approximately 6,200 steers, about 260 tractors and related implements, establishment of approximately 5,000 ha of pastures, 1,700 ewes, 300 rams, 1,200 purebred cows and 750 crossbred cows. Increased quantities of seasonal inputs (US$4.6 million) including seeds, fertilizers, herbicides, pesticides, machinery services for crops and veterinary and reproductive services for animals would also be required. The Project would also provide fifteen input resale centers on credit, seven for the OC, five for the COCEBLE and three for the CCGC (US$1.4 million).  - 13 - D. Status of Engineering 4.11 Standard architectural plans and specifications are available for most Project buildings. For the rural roads component, design standards follow those established for the Bank-financed highway projects (Loans 1601-TUN and 2108-TUN). The studies division of the Ministry of Public Works has hired consultants to prepare, by June 30, 1985, substantially completed detailed engineering including any necessary topographic surveys, hydrological studies, and draft tender documents for the 29 km of roads included in the first year's program. Terms of reference have also been prepared for most Project-financed consultants. V. PROJECT COSCIS AND FINANCING A. Cost Estimates 5.01 Over the four-year implementation period (1986-89), the total cost of the Project is estimated to be US$37.9 million (D 27.7 million). The foreign exchange component is US$16.8 million (D 12.3 million) or 44% of the total. Of the total Project cost, US$29.1 million (D 21.2 million) or 77% represent investment costs, while US$8.8 million (D 6.5 million) or 23% consist of incremental recurrent costs (including farm inputs). Taxes and duties totaling US$9.7 million (D 7.1 million) or 26% of total Project costs are also included. The Project cost is summarized in the following table and shown by year in Table 1 and-by summary account in Table 2: TUNISIA NORTHWEST AGRICULTURAL PRODUCTION PROJECT PROJECT COST SUMMIARY (D Million) (USS Million) Z Total ------------------- Z Foreisr, Base Local Foreign Total Local Foreign Total Eychange Costs EYTENSICt 1.2 0.9 2.1 1.7 1.2 2.B 41 11 RESEARCH 0.3 0.2 0.5 0.3 0.3 0,7 46 2 LIVESTOCK SERMJICES 0.8 0.5 1.3 1.1 0.7 1.B 40 7 FARM ACHINERY TESTING 0.1 0.2 0.3 0.2 0.2 0.4 59 2 RURAL ROADS 1.7 1.4 3.2 2.4 1.9 4.3 45 16 PROJECT IMPLEMENTATION UNIT 0.1 0.0 0.1 0.2 0.0 0.2 23 1 TRAINING, TECH.ASST.Y STUIDIES 0.3 0.9 1.3 0.5 1.3 1.7 73 6 ON-FARM INVESTMENTS 4.0 3.8 7.8 5.5 5,2 10.7 48 40 SEASONAL INPUTS 1.5 0.8 2.3 2.1 1.1 3.1 34 12 INPUT RESALE CENTERS 0.5 0.2 0.7 0.7 0.3 1.0 31 4 Total BASELINE COSTS 10.7 8.9 19.6 14.6 12.3 26.9 46 100 Phfsical Continsencies 1.1 0.9 2.0 1.5 1.3 2.7 46 10 Price Continmencies 3.7 2.4 6.1 5.0 3.3 8.3 40 31 Total PROJECT COSTS 15.4 12.3 27.7 21.1 16.8 37.? 44 141  - 14 - 5.02 The above cost estimates are based on January 1984 prices updated to January 1985 levels, and are derived from recent quotations and appraisal estimates. Physical contingencies have been estimated at the rate of 15% for civil works and 10% for vehicles, machinery, equipment and furniture, and incremental recurrent costs (except salaries and wages and consultants services). Price contingencies have been calculated on the basis of the following annual percentage rates, applied to local and foreign costs including physical contingencies: 1986 1987 1988 1989 Local Costs 10.0 10.0 10.0 9.0 Foreign Costs 9.0 9.0 9.0 7.5 B. Financing Plan 5.03 A Bank loan of US$15.0 million to the Government of Tunisia is proposed to finance 100% of the foreign exchange component of the Project, excluding seasonal inputs (US$1.6 million) and technical assistance to be financed on grant terms by Belgium (US$0.2 million). The balance of funds required would be financed by the Government, BNT and Project beneficiaries. The Bank loan would thus finance 53% of the total Project cost net of taxes. Project costs would be shared in the following amounts and proportions: L of Net Local Foreign Total Project Costs (US$ million equivalent) - ----- World Bank - 15.0 15.0 53 Belgium - 0.2 0.2 1 BNT 2.4 1.6 4.0 14 Beneficiaries 1.8 - 1.8 6 Government 7.2 _ 7.2 26 Net Project Costs 11.4 16.8 28.2 100 Taxes and Duties 9.7 - 9.7 Total Project Costs 21.1 16.8 37.9 5.04 The proposed Bank loan would be for 17 years including 4 years grace. It would be utilized by the Ministry of Agriculture (for extension, research, farm machinery testing, PIU and training, technical assistance and studies), by the Ministry of Equipment (for rural roads), and passed on as a grant to the OEP (for livestock services) and to the BNT under a subsidiary loan agreement satisfactory to the Bank to be signed no later than November 30, 1985. During negotiations, the Government gave assurances that it would cause BNT to provide funding for the credit component of the Project as required.  - 15 - C. Procurement 5.05 Procurement arrangements are suimmarized in the following table: Procurement Method Total Cost Project Element ICB LCB Other NA (US$ million equivalent) -_ -_- Civil Works 9.2 0.8 10.0 (4.0) (0.3) (4.3) Goods 3.2 1.0 0.2 4.4 (1.4) (0.5) (0.1) (2.0) Credit 21.4 21.4 (8.0) (8.0) Consultants' Services 1.8 0.3 2.1 (0.7) (-) (0.7) Total 3.2 10.2 2.8 21.7 37.9 (1.4) (4.5) (1.1) (8.0) (15.0) Note: Figures in paren,theses are the respective amounts financed by the Bank. 5.06 Works. Project buildings (US$3.7 million), including both new construction and rehabilitation of CTVs, research facilities, and livestock service facilities for OEP, would be procured on the basis of local competitive bidding (LCB). The small and scattered nature of the works would make them of no interest to foreign contractors. No individual contract for these works is expected to have a value of more than US$250,000. Contracts for rural roads (US$6.3 million) would also be awarded on the basis of LCB procedures, as experience with similar works under the two Rural Roads Projects (Loans 1601-TUN and 2108--TUN) indicates that such works would fail to attract bids from foreign firms not already established in the country. Roads would be constructed under unit price contracts in four lots. Enough construction firms, both local and foreign, are established in Tunisia to assure competitive prices under LCB. Civil works such as buildings which fail to attract bids from contractors would be constructed under force account. Civil works and goods under the credit component would be procured through regular commercial or other local channels. 5.07 Goods. Project vehicles, machinery, equipment and furniture (US$4.4 million), including vehicles, tractors, implements, and office furniture for the extension service; research equipment; vehicles, machinery and equipment for the livestock services component and tractors for the farm machinery testing component, would be grouped wherever possible into packages and procured by international competitive bidding (ICB) in accordance with  - 16 - Bank guidelines. When bulking is not feasible, vehicles, machinery, equipment and furniture contracts not exceeding the equivalent of US$150,000 each, and totaling in the aggregate no more than US$1.0 million, may be awarded on the basis of local competitive bidding procedures acceptable to the Bank. Qualifying domestic manufacturers would receive a preference in bid evaluation of 15% or the prevailing import duty, whichever is less. Specialized research equipment totaling in the aggregate no more than US$150,000 would be procured on the basis of quotations from at least three qualified suppliers. 5.08 Bank-financed consultants for studies and technical assistance (US$1.8 million), with qualifications and terms and conditions of employment satisfactory to the Bank, would be appointed following Bank guidelines on the use of consultants published in August 1981. 5.09 Contract Review. All bidding packages for works estimated to cost the equivalent of US$300,000 or more and bidding packages for goods costing the equivalent of US$150,000 or more would be subject to the Bank's prior review of procurement documentation resulting in a coverage of about 75% of the total estimated value of contracts for both goods and works. The balance of contracts would be subject to random post-review by the Bank after contract award. The first bidding document of each agency would also be reviewed prior to the start of the bidding. Assurances that the above procedures would be followed were obtained at negotiations. D. Disbursement 5.10 The disbursement profiles for Bankwide area development projects, EMENA agricultural projects, Bankwide agricultural projects and all Tunisia projects show that full disbursement has taken an average of eight years. In the case of the present project, full disbursement is expected to take place over six years (Table 3). The reasons for deviating from the disbursement profiles are: (a) the existence of standard architectural plans and designs for most major disbursement items (buildings and roads); and Cb) the fact that a specific attempt has been made to design the Project in such a way that disbursement delays will be minimized, for example, by scheduling major investments early in the Project period, and using the time between negotiations and project start-up for recruitment and procurement. 5.11 Disbursements would be made after receipt of full documentation, except in the case of civil works by force account and goods and services provided under contracts valued at under US$100,000 for which disbursements would be made against certified statements of expenditure (SOE). The SOE would be certified by the Project implementing agency incurring the expenditure and by the Central Commissioner for Agricultural Development in his capacity as Project Manager. Supporting documentation would be retained by the implementing agency and the PIU and made available for inspection during Bank supervision missions and by external auditors. Except in the case of direct payments, no reimbursement application for less than US$20,000 would be accepted. Assurances to these effects were obtained at negotiations. Disbursement categories, amounts allocated and the share of expenditures to be financed would be as follows:  - 17 - Amount of the X of Expenditures Category Loan Allocated to Be Financed (usS Million) (1) Civil Works 3.6 100l of foreign and 45Z of local (2) Goods 1.7 1002 of foreign and 50Z of local (3) Medium- and Long-Term Credit 8.0 70% of BNT disbursements for subloans (4) Consultants' Services 0.5 1001 of foreign and 80% of local (5) Initial Deposit in Special Account 0.3 (6) Unallocated 0.9 Total 15.0 5.12 To ensure the efficient and timely implementation of the Project, the Bank would deposit after loan effectiveness an initial sum of US$300,000 to set up a revolving fund to finance eligible expenditures corresponding to the Bank's share of the first three months' Project needs. The deposit could later be increased by agreement between the Borrower and the Bank to meet Project needs. The revolving fund would be used to pay eligible expenditures under the Project for which Procedure I would be applied: civil works; goods; and consultants' services. No local salaries would be paid through the revolving fund. 5.13 Bank funds would be deposited in US dollars in a special account at the Central Bank of Tunisia (BCT), and would be channeled to the Project as needed through BNT. For this purpose, a project account would be opened by the CCDA at BNT in Tunis, and each CRDA would have a similar project subaccount at the local BNT branch in Beja, Jendouba, El Kef, and Siliana. Government counterpart funds would be secured through appropriate budget allocations. The Project's annual budget and financial plans, approved by the Project Coordinating Committee (para. 6.02) and by the Ministries of Planning and Finance would be provided to the Bank for information no later than August 31 of each year. 5.14 The Bank would replenish the revolving fund as requested upon receipt of satisfactory evidence that expenditures paid were eligible for financing out of the fund. A replenishment request would normally be sent to the Bank when 50Z of the revolving fund has been disbursed, i.e., about every two to three months. Should any disbursement made from the special account not be  - 18 - acceptable to the Bank, Government would be required to deposit the corresponding ineligible amount into the special account, or if the Bank so requests, refund to the Bank an amount equal to the amount not eligible. 5.15 No further deposits into the special account would be made by the Bank when the total unwithdrawn amount of the part of the loan that is expected to be disbursed at loan closure would be equal to US$0.6 million. If it is determined, at any date, that any amount outstanding in the special account is not required to cover further payments for eligible expenditures, the Government, unless otherwise agreed by the Bank, would refund to the Bank such amount outstanding in the special account. Assurances on the above points were obtained at negotiations. VI. ORGANIZATION AND IMPLEMENTATION A. Organization 6.01 The proposed Project would be implemented through existing institutions, strengthened as necessary to ensure effective project implementation. No new institutions would be created, and changes within existing units would be minimal, the aim being to strengthen rather than bypass existing line agencies. 6.02 Overall responsibility for implementing the Project would lie with the Ministry of Agriculture (Chart 1), through the Central Commission for Agricultural Development (CCDA) located at the Ministry's headquarters in Tunis. The head of the Commission, the Central Cormnissionner for Agricultural Development, would be the titular Project Manager. However, day-to-day management would be handled by a Deputy Project Manager leading a Project Implementation Unit within the CCDA comprising a Chief Accountant, an Agricultural Economist, a Secretary, and a Driver. The Central Commissionner would chair a Project Coordinating Committee (PCC) to be created by October 31, 1985 for the purpose of periodic liaison with other entities involved in Project execution, including under the Ministry of Agriculture, the Department of Crop Production (DPV), the Department of Animal Production (DPA), the Department of Assistance to Small and Medium Size Farmers (DAPMEA), the Department of Planning, Statistics and Economic Analyses (DPSEA), the National Institute of Agricultural Research (INRAT), the Office of Cereals (OC), and the National Livestock and Range Agency (OEP). Outside the Ministry of Agriculture, the CCDA would coordinate as necesenrf with the service cooperatives COCEBLE and CCGC, the National Bank of Tunisia (BNT), the National Bank for Agricultural Development (BNDA), the Department of Roads CDGPC) of the Ministry of Public Works (MOE), the Ministry of Planning (MOP), and the Ministry of Finance (MOF). 6.03 Field management of the Project would be handled by the regional branches of the CCDA, known as Regional Commissions for Agricultural Development (CRDA), located in each of the four Project provinces (Beja, Jendouba, El Kef, and Siliana). By October 31, 1985, the head of each of these units, the Regional Commissioner for Agricultural Development (CRDA), would be appointed subproject manager in his province. Thereafter, he would be responsible for implementing the various Project components in his province  - 19 - in accordance with a yearly action program agreed with the Project Manager. The CRDAs in each of the four Project provinces would also receive the incremental staff necessary to manage the Project at the provincial level. New staff in each province would consist of: a Deputy for extension and for Project management and execution and an Accountant for the DAPMEA unit. The Deputy would provide day-to-day management support to the Commissionner for the implementation of extension reforms. The Regional Commissioner would be responsible for liaising locally with other entities involved in Project execution such as the local research stations, local input supply and output marketing facilities of the OC, COCEBLE and CCGC and DEP, local branches of the BNT and local offices of the DGPC. During negotiations, assurances were obtained that extension staff would be deployed according to an acceptable plan (para. 4.03 and Annex). Progress in meeting the plan would be reviewed annually by the Government and the Bank. B. Implementation 6.04 The CRDA would be responsible for the Project extension component; INRAT for the researci facilities; the OEP for the livestock input supply and output marketing facilities; and the DGPC for upgrading and maintaining the rural roads. Individual farmers and UCPs would implement the on-farm components of the Project with the aid of seasonal and medium/long term credit from BNT. OC, COCEBLE, and CCGC would implenment the input supply component, also with the aid of credit from BNT. 6.05 Extension. Implementation of the extension component would be the responsibility of the individual Commissioners in each province whose plans and actions would be coordinated by the Central Commissioner (Project Manager). To assist the regional Commissioners to manage the extension service, the structure of each CRDA would be modified to include a new post of Deputy. He would coordinate the extension system from subject-matter specialists to field extension agents and set up a management system for agricultural extension within the CRDA modelled on the T&V system. Whereas at present the policy is to build one rural extension center (CTV) for each extension agent, under the Project for reasons of economy, two extension agents would share the facilities of a CTV. For farmer contact only one agent would deal with an individual group of farmers in place of the current system under which some farmers are contacted by several crop specific agents. This mult;purpose agent would be responsible for conveying all extension messages to groips of about 250 farmers. Subject-matter specialists from the DPV, DPA, DAPMEA, and OEP, located at the CRDA level would develop extension messages, help train extension agents and answer specialized questions. Extension agents would receive regular training in extension messages, and make regular visits to farmers. The general organization principles to be adopted under the Project are given in more details in the attached Annex. Agreement was obtained during negotiations that these principles would be adopted under the Project. 6.06 Research. Implementation of the research component would be the responsibility of INRAT which generally has sufficient staff to carry out the actions proposed under the Project. INRAT employs the DGR and the buildings section of the DAAF for all of the construction work. The candidate for training in research center development (para. 4.09) would be sent for training in Year 2 and his return would coincide with the consultancy for research station development, both of which would preceed the development of the 30 ha  - 20 - on the El Afareg station. Phasing of the implementation would also coincide with the training of the additional scientists to be assigned who would be required to work one year at the central research station before being assigned. Their assignment to El Afareg would follow immediately the completion of the basic infrastrleture. 6.07 Livestock Services. The four activities under this component would be implemented as follows: (a) Animal Health Centers: The three animal health centers to be constructed under the Project would be staffed and managed by the DPA. Construction would conform to standard plans that are already used by DPA for centers elsewhere in the country. Construction of the centers from bidding, letting the contracts to supervision would be the responsibility of the PIU. The bidding process would start immediately after loan signaing. Before starting this process, the PIU would agree with DPA on the location and suitability of the plans and get a formal commitment that the facilities would be staffed immediately upon their completion. (b) Artificial Insemination Posts: Ten posts already exist in the Project area and no further expansion is required. The posts would be managed by DPA who would be required to transfer trained inseminators in Years 2, 3, and 4 (4, 4, 2). The inseminators equipment which is specialized and estimated to cost approximately D 1,000 for each technician would be purchased locally by DPA. (c) Milk Collection: OEP would be responsible for this subcomponent, including procurement. Plans are available as OEP has had a number of other collection plants constructed. The specialized equipment has also been purchased previously. For the milk collection center to be constructed under the Project, consistent with the Government's policy of gradual withdrawal from milk collection, the Government has agreed that at the latest five years from the start of operations, OEP would sign and maintain a leasing agreement satisfactory to the Bank with a private party or service cooperative for the management of the center, OEP having assisted the tenant during the year prior to the signature of the agreement. (d) Animal Buying Centers: All three centers would be the responsibility of GEP to plan, procure, and manage. Plans would be prepared during 1985. Construction of the first would start in the first year of the Project. The experience of this center would therefore be available before commencing the second and third. 6.08 Farm Machinery Testing. Selecting the technical assistant who would advise on machines to be tested and testing procedures, importation of machinery and the conduct of tests would be the responsibility of the farm machinery unit in the DC-R. Actions for the purchase of silage-making equipment for demonstration would be the responsibility of each CRDA. Likewise, directing farmers or groups of farmers to appropriate tractors and machinery to purchase or rent would be the responsibility of the CRDAs through the extension services.  - 21 - 6.09 Rural Roads. Road upgrading and maintenance would be the responsibility of the DGPC, the same unit which is responsible for rural roads under the Third and Fifth Highway Projects. Before final approval, the road segments scheduled for upgrading (Table 4) would follow basically the same procedures and meet the same criteria as those applied under the Fifth Highway Project. By April 30 of each year (June 30 in 1985), the DGPC would submit to the Bank for review and comment a plan for the rural road upgrading to be carried out during the following calendar year. Each rural road included in the plan would be supported by a draft feasibility study employing methodology and selection criteria including an economic rate of return satisfactory to the Bank. After reviewing the Bank's comments, the DGPC would finalize the plan including the feasibility studies and submit it to the Bank for its approval by at latest September 30 of the year in which the draft plan had been submitted. Government would also maintain the entire network of all weather roads in the Project area. Assurances on these points were obtained at negotiations. Supervision of construction would be undertaken by the DGPC's regional offices which are experienced in this type of work. The aintenance requirements of the network of all weather roads in the Project area would be absorbed into the DGPC's general maintenance programs, the capacity of which has been built up under the Bank-financed rural roads projects (para. 3.10). 6.10 Project Implemeutation Unit. Rapid assignment of key staff, both in the CCDA and CRDAs, is essential for successful Project implementation. Establishment of the PIU in the CCDA, headed by a Deputy Project Manager and including a Chief Accountant, and Agricultural Economist, and assignment of the Deputy and Accountant for each of the CRDAs would therefore be conditions of effectiveness of the proposed Bank loan. Assurances to these effects were obtained at negotiations. These positions would be filled by transferring competent technical staff from within the Ministry of Agriculture and by hiring qualified accountants. 6.11 Training, Technical Assistance and Studies. The CCDA would be responsible for implementing this component including organizing training abroad, recruiting technical assistance and managing the studies financed under the Project. 6.12 Input Supply. The OC would be responsible for developing plans and specifications for all 15 input resale centers included in the Project but would only be responsible for the construction of seven. The PIU in the CCDA would assist the cooperatives (COCEBLE and CCGC) to develop subloan applications and develop a timetable for construction that would allow the first two of their eight outlets to begin functioning by Year 2 of the Project. C. Credit 6.13 Credit to farmers would be provided by BNT through its existing network of branches in the Project area which is adequate to cover the needs of the Project. The Accountant to be recruited for each CRDA would assist with the implementation of the credit component in his province. OC, COCEBLE, CCGC would borrow from BNT to finance the construction of input resale centers. In order to ensure that the terms and conditions of onlending of Project funds by BNT are appropriate, the Government agreed during negotiations to adopt before June 30, 1985 an action plan satisfactory to the Bank aimed at inter alia improving interest rate and cost recovery policies. Thereafter the Government  - 22 - would implement the action plan according to the timetable contained therein. The adoption of such a plan and the signature of a subsidiary loan agreement satisfactory to the Bank would be conditions of disbursement for the credit funds in the Bank loan. Subloans would be made by BNT on terms and conditions and according to eligibility criteria satisfactory to the Bank, to be defined in the context of the action plan. D. Monitoring and Evaluation 6.14 The Agricultural Economist to be aLipointed in the PIU would be responsible for monitoring Project implementation including both physical execution and achievement of production and policy objectives, according to the key indicators presented in Annex 10. An assurance to this effect was obtained at negotiations. Data on input use, cropped areas, production and yields are already collected by the CRDA. The PIU would also prepare the Project Completion Report, evaluating the results of the Project, within six months of the closing date, and an assurance to this effect was obtained at negotiations. E. Accounts, Audits and Reporting 6.15 Accounts and Audits. CCDA, which would have primary responsibility for the Project, would establish and maintain separate project-related accounts in Tunis, and at its regional branches (CRDA) in Beja, Jendouba, El Kef, and Siliana. The CCDA in Tunis would be responsible for consolidation of all accounts from the start of Project implementation. These accounts would be maintained in accordance with sound and acceptable accounting principles and practices that would enable CCDA to provide interim and annual financial statements to reflect the financial situation of the project. As a condition of effectiveness of the Bank loan, CCDA would complete the design of a Project accounting system, satisfactory to the Bank, with the assistance of a CPA firm of international repute. The accounting system would be installed by January 1, 1986 and be fully operational by March 31, 1986. Assurances on the above were obtained at negotiations. 6.16 Project accounts would be audited annually by indeoendent auditors acceptable to the Bank. An understanding was reached that the Controller General of Finance, who has independent status and whose audits are acceptable to the Bank in scope and content, would do the audit. Audited financial statements and reports, of such scope and in such detail as the Bank may reasonably request, would be submitted to the Bank within six months after the end of each fiscal year. The auditors' report would include a statement on the adequacy of the accounting system and internal controls and state whether or not Bank funds have been used for their intended purposes. The report would also confirm that SOE correctly reflect detailed records kept by CCDA. Assurances that the above procedures would be followed were obtained at negotiations. 6.17 Reporting. CCDA would submit semi-annual reports on the physical and financial progress of the Project within two months after the end of each semester, including the performance indicators given in Annex 10, and other indicators that may be needed during Project supervision. These reports would include project interim and annual financial statements for the year under review and accumulated transactions of fiscal years from Project commencement. These arrangements were agreed upon at negotiations.  - 23 - VEL PROJECTJTUSTIFICATION AND RISKS A. Production, Yields and Technological Change 7.01 Based on the farmers and UCP's expected to adopt the Project promoted technology (para. 7.07), the following overall changes in cropped areas, animal units, yields and production may be expected in the Project area by full development in Year 15: At Full Development Existing Situation (Year 15) Incremental Area Yields Production Area Yields Production Production (ha) (t/ha) (t) (ha) (t/ha) (t) (t) crops Ourum wheat 130.000 1.0 130.000 122.400 1.35 165.200 35.200 Bread wheat 15.700 1.35 2t.200 17.300 1.7 29.400 8.200 Barley 48.500 1.1 53.400 37.200 1.35 50.200 -3.200 Grain legumes 11.700 0.6 7.000 15.700 1.0 15.700 8.700 Olives 23.900 1.3 31.100 23.900 1.3 31.100 - Hay 22.600 3.8 85.900 25.200 4.0 100.800 14.900 Silage /a - - 18.400 1S.0 276.000 276.000 Medicago /a - - 10.100 - - - Fallow 103.100 - - 85.300 - - Rangeland 44.500 - - 44.500 - -- Total 400,000 400.000 Total animal 116.350.000 FU 211.550.000 FU 95.200,000 FU Feed all sources 1k Animal Yields Production Animal Yields Production% Units Rer AU 1t) Units per AU (t) (tL Livestock Products Sheep meat 82,000 20 kg 1.600 145.000 33.3 kg 4.800 3.200 Sheep wool 2.4 kg 200 2.7 kg 400 200 Dairy cattle milk 22,000 260 1 6.000 38.000 1.330 51.000 45.000 Dairy cattle meat 110 kg 2.000 - 190 7.000 5.000 Lamb meat - - - 33.000/i 1S SOO 500 Steer meat - - - 29.000/t 130 3.800 3.800 /a Although some silage and medicago are produced in limited quantities at present, they are not separated in available statistics. /k Includes feed from all sources on the farm including hay, silage. medicago. crop straw. weeds, fallow and rangeland. /s Individual animals purchased for fattening.  - 24 - 7.02 As the above table indicates, production increases would stem from changes in cropped areas, animal units and yields on private farms and UCPs. The yield increases would be obtained from the incremental on-farm investments, increased input use and better farming practices which are expected to result from Project investment in extension, research, input supply, livestock services, rural roads and training. On-farm investments would include establishment of medicago, purchase of tractors and implements, purchase of dairy cattle, rams and ewes, and construction of cowsheds, sheepfolds and sheep fattening corrals. Incremental input use would include increased use of seeds of high-yielding varieties, fertilizers, pesticides and herbicides, labor, machinery and veterinary services. Better farming practices would consist primarily of rotating cereals with nitrogen-fixing leguminous crops and integrating crop and livestock activities. These production increases would generate incremental gross foreign exchange savings of about US$'00 million per year at full development in year 15. After deducting the foreign exchange c6mponent of operating and replacement costs, incremental net foreign exchange savings would be about US$35 million a year. B. Markets and Prices 7.03 Since most incremental production from the Project would consist of goods for which part of the country's needs are presently met from imports (milk, wheat, feedgrains, mutton/lamb), additional output of these goods would find a ready market as import substitutes (a higher quality product in the case of milk), thereby helping the balance of payments. A sizeable export market for mutton/lamb could also be further developed in neighboring Arab countries. Remaining incremental production, consisting of goods for which there is strong internal demand (grain legumes and wool), would be sold on the domestic market. The farmgate prices used in the financial analysis for the major Project outputs are as follows: (D per ton) Durum Wheat 112 Bread Wheat 102 Sheep Meat 1,380 Dairy Cattle Milk 200 Dairy Cattle Meat 850 These prices are broadly in line with current world prices and provide adequate incentives. At negotiations, Government agreed to take all necessary measures to ensure adequate incentives to agro-industry for milk and meat marketing, and fertilizer distribution at least cost to the Government budget, in order to promote increased agricultural production and the wider availability of fertilizer. In particular, the Government indicated that it would continue to progressively adopt more realistic consumer prices for milk and beef, and maintain adequate distribution margins for fertilizer. C. Farmers' Incentives 7.04 To assess the incentives for private farmers to adopt the new technology to be extended under the Project, six farm models have been prepared. Three represent the situation of small (14 ha), medium (30 ha) and  -25- large (106 ha) farmers in the lower rainfall part of the Project area; three others represent that of the same size farmers in the higher rainfall zone. An additional model representing the situation of UCPs has also been developed. In all cases, the incentives to participate in the Project are substantial, as the following table shows: Net Income Net at Full Income Deve- Increase Increase Financial at lopment in Net in Net Rate of (No.) Present (Year 15) Income Income Return - (constant 1984 D/year) - (%) Farmers Lower Rainfall Zone Small (14 ka) 640 610 1,500 890 +146 over 50 Medium (30 ha) 515 1,410 3,060 1,650 +117 over 50 Large (106 ha) 272 3,340 11,760 8,420 +252 34 1,427 Higher Rainfall Area Small (14 ha) 1,624 860 2,790 1,930 +224 over 50 Medium (30 ha) 980 1,490 5,810 4,320 +290 over 50 Large (106 ha) 538 4,340 17,670 13,330 +307 39 5,142 Sub-total 4,569 DCPs 24 48,000 117,220 69,220 +144 37 D. Cost Recovery 7.05 Apart from the normal cost recovery applied to items bought on credit which account for most Project expenditure, two specific Project activities would be subject to cost recovery: milk collection and animal buying. Following the system already established in the adjacent Northwest Rural Development Project (Loan 1997-TUN), the operating costs of milk collection under the Project would be recovered through a price margin (D 0.010 per liter of milk collected plus D 0.0075 in taxes). To attempt to recover part or all of the investment costs would be impractical since it would imply two cost recovery systems in the same area. For animal buying, being a new activity organized primarily for the benefit of the larger farmers, the recovery charge would cover operating and maintenance costs from the start of operations and, in addition, investment costs within five years. At negotiations, assurances were obtained that from the commencement of operations: (a) the price margin for milk collection would be at least sufficient to cover the operation and  - 26 - maintenance costs of milk collection, and (b) the price margin for the use of the animal buying centers would be sufficient to recover operating and maintenance costs and investment costs as well no later than five years thereafter. E. Economic Analysis 7.06 Using the latest conversion factors for Tunisia,1' updated as necessary for major changes in financial/economic price relationships, the economic rate of return (ERR) of the Project is estimated at 302, with a net present value (NPV) of D 25.0 million, discounted at 122. In updating the conversion factors, border prices for traded items were derived on an import or export parity basis from World Bank projected prices. The economic life of the Project was assumed to be 25 years, based on the useful life of the Project buildings. All incremental costs, including the investment and recurrent costs of the Government agencies involved in the Project, as well as those of the private farmers, UCPs, OC, COCEBLE and CCGC, were included. Benefits were measured on the incremental production of crop and livestock products generated by the Project. Demand for non-family seasonal agricultural labor equivalent to about 3,900 manyears per year would also be created by full development (year 15). In the "without project" situation, production was assumed to grow at 3% per year to reflect general improvements in productivity which would have taken place anyway. The on-farm investments necessary to sustain such an increase were also included. 7.07 Since on-farm development accounts for the bulk of the Project costs (52Z of base costs) and benefits, the approach taken to the economic analysis was to first aggregate the economic farm models into subproject areas (lower rainfall, higher rainfall, and UCP), and then aggregate them together with the incremental economic Project costs to estimate the overall Project ERR and NPV. For the purpose of this aggregation, different rates of farmer entry into the Project were assumed depending on whether the farm is in the lower or higher rainfall subproject area, as follows: Assumed Farmer Entry into the Project Project Year 1 2 3 4 5 6 7 Total -(- } - Lower Rainfall Zone 1 3 0 6 10 7 5 32 Higher Rainfall Zone 1 5 1 8 10 10 5 40 Based on the readiness of their development plans, UCPs were assumed to enter the Project at the rate of 6, 6, 1, 5, and 6, resulting in a total of 24. The probability of a pause in adoption caused by the occurence of drought has been reflected in the relatively low adoption rates assumed for private farmers and UCPs in years 3 and 4. No separate component ERRs or NPVs were calculated (e.g. on extension, roads, etc.) since each component contributes jointly to overall benefits, and has no separate benefits of its own. 1/ Gordon Hughes, "Shadow Prices and Economic Policy in Tunisia", May 1980.  - 27 - F. Risk and Sensitivity Analysis 7.08 The major risk facing the Project would be a slower than expected adoption by farmers and UCPs of the new technology to be extended. By selecting a target group already in the market economy which stands to gain significantly from adoption (para. 7.04), by ensuring the availability of adequate inputs on credit, and by projecting realistic adoption rates, steps have been taken to minimize this risk, but a lower than expected or a slower than anticipated adoption rate would lower the ERR and NPV. An unexpected fall in yields or output prices or increase in costs would also have a negative effect. 7.09 To test the sensitivity of the Project to the risk of slower adoption, the number of adopting farmers and UCPs was cut in half. This gave a 151 ERR. The crossover value of different variables (i.e., the value of the variable at which the NPV of the economic benefits discounted at 12% becomes zero) was then calculated to test the sensitivity of the Project to the other risks. The conclusions that may be drawn from this analysis are that the Project would still be justified even if: (a) benefits decreased by 37%; and (b) costs increased by 58%. In terms of yield increases, average "iwith project" yields (para. 7.01) for the major outputs would only have to rise to the following levels to ensure a positive NPV: durum wheat 1.25 t/ha (instead of 1.35 t/ha); milk 935 1/AU (instead of 1,330 1IAU); and mutton 28.4 kg/AU (instead of 33.3 kg/AU). 7.10 The Project thus has a strong economic justification which is resistant to adverse movements in key variables. VIII. RECOMMENDATIONS 8.01 At negotiations, assurances were obtained that: (a) Government would provide all necessary credit finance [para. 5.04; LA Section 3.01 (a)]. By November 30, 1985, the Government would conclude a subsidiary loan agreement with BNT, satisfactory to the Bank, covering the onlending of credit funds to 3NT [para. 5.04; LA Section 3.02 (b)]. Terms, conditions, and eligibility criteria satisfactory to the Bank would be defined in the action plan [para. 6.13; LA Section 3.02 (c)]; (b) The Project annual budget and financial plans, approved by the Project Coordinating Committee and by the Ministries of Planning and Finance, would be provided to the Bank for information no later than August 31 of each year [para. 5.13; LA Section 3.01 (b)]; (c) A revolving fund with an initial deposit of US$300,000 would be set up as described above [paras. 5.12-5.15; LA Section 2.02 (b) and Schedule 2:;  - 28 - (d) The Project would'be managed by a PIU operating through the CCDA/CRDA structure of the Ministry of Agriculture, with the Central Commissioner for Agricultural Development as Project Manager who would also chair the Coordinating Committee to be created by October 31, 1985. The Regional Commissioners for Agricultural Development would also be appointed subproject managers in their respective provinces by October 31, 1985 (paras. 6.02 and 6.03; LA Section 3.04); (e) Extension staff would be deployed according to an acceptable plan. Progress in meeting the plan would be reviewed annually by the Government and the Bank [para. 6.03 and Annex; LA Section 3.03 and supplemental letter No. 2]; (f) For the milk collection center at El Kef, at the latest 5 years after the start of operations, OEP would sign and maintain a leasing agreement satisfactory to the Bank with a private party or a service cooperative for the management of the center [para. 6.07 (c); LA Section 3.06]; (g) Before final approval, all rural road segments scheduled for upgrading would meet the same criteria as those applied under the ongoing Fifth Highway Project, including an economic rate of return of at least 10X. Government would maintain the entire network of all weather roads in the Project area (para. 6.09; LA Sections 3.07 and 3.08 and supplemental letter No. 4); (h) By June 30, 1985, the Government would adopt an action plan on national credit issues satisfactory to the Bank, including measures concerning interest rates and the improvement of loan repayments; thereafter, the Government would implement the action plan according to the timetable contained therein [para. 6.13; LA Section 3.02 (a)]; *i) The Agricultural Economist to be appointed in the PIU would be responsible for monitoring Project implementation according to the indicators given in Ainex 10 [para. 6.14; LA Section 3.13 (c) and supplemental letter No. 3]; (j) CCDA would submit semi-annual progress reports on the physical and financial progress of the Project within two months after the end of each semester [para. 6.17; LA Section 3.13 (c)]; (k) Government would take all necessary measures to ensure adequate incentives for milk and meat marketing and fertilizer distribution at least cost to the Government budget (para. 7.03 and LA Section 3.09); (1) From the commencement of operations, the price margin for milk collection would be at least sufficient to cover operation and maintenance costs, and the price margin for the use of the store (lean) animal purchasing centers would be sufficient to recover operating and maintenance costs from the start of their operations and, in addition, investment costs no later than five years thereafter (para. 7.05; LA Section 3.10).  - 29 - 8.02 As conditions of effectiveness of the proposed loan, Government would: (a) establish the PIU headed by the Deputy Project Manager, and fill the positions therein of Chief Accountant and Agricultural Economist, and the positions of Deputy and Accountant in each of the CRDAs [para. 6.10; LA Section 6.01 (a) and (b)W; and (b) complete the design of a Project accounting system satisfactory to the Bank [para. 6.15; LA Section 6.01(c)]. 8.03 As a condition of disbursement for the credit funds in the Bank loan, the Government would adopt an action plan on national credit issues, and the Government and BNT would conclude a subsidiary loan agreement, both satisfactory to the Bank (para. 6.13; Schedule 1, para. 4 (b)]. 8.04 The above assurances having been obtained, the proposed Project is suitable for a Bank loan of US$15.0 million to the Government of Tunisia for a term of 17 years including 4 years grace. The Project is expected to be completed by June 30, 1990. 2441E  -30 - ANNEX Page 1 STAFF APPRAISAL REPORT TUNIIA NORTHWEST AGRICULTURAL PRODUCTION PROJECT PLAN OF ACTION FOR THE ORGANIZATION OF THE PROJECT EXTENSION SERVICES A. Structure and Organization 1. The effectiveness and efficiency of extension services in Tunisia are hampered by lack of organization and discipline, ambiguous lines of responsibility, excessive amounts of office work; and at field level, by inadequately trained personnel and lack of direction, support and incentives. This situation leads to unmotivated, poorly trained field level staff dispersed over large areas, lacking close supervision, and precise work programs and responsible for many tasks which often have no direct relation with extension work. Under the Project, the extension service in the Project area would be reorganized along the lines of the T&V system in order to remedy current defects and improve efficiency. Structure 2. CRDA Level: The four CRDAs in the Project area will set up an extension unit dealing exclhsively with extension activities headed by an agricultural engineer and some support staff who would be the link between the CRDA level and field staff. This unit's primary responsibility would be to (a) closely supervise extension staff; (b) coordinate with other ministry departments, research and support agencies at the provincial level; and (c) provide specialized technical support to field staff through a team of subject-matter specialists (para. 12). The CRDA extension director should spend four days per week in the field in order to monitor the work of extension agents. Support services such as input distribution, marketing, credit, mechanization would be managed separately from the extensionL service by specific agencies. Appointment of the heads of extension units would be condition of effectiveness of the proposed loan. 3. Extension Centers (CTVs): There would be on average two extension agents attached to each CTV so that office/storage space would be shared. The number of extension agents will be defined according to the number of farms to be visited, their distribution in the area supervised by each agent, types of farming and cropping patterns. In order to facilitate the extension agents' task and increase the number of farmers in direct contact with them, farmers would be organized in groups of about 20/30. Half a day would be spent every fortnight with each group. One extension agent would would be expected to visit and be responsible for an average of about 250 farmers. This ratio would be lower in very intensive production areas, and higher in more extensive areas. Project extension staff would be deployed according to the following plan.  -31 - Page 2 Deployment of Extension Staff Total Project Year Number Category Required 1 2 3 4 CRDA Deputies for Extension 4 " - - - - Extension Agents at CTV Level 66 15 17 15 4 /a To be recruited prior to loan effectiveness. /b Of which 15 already in place in 15 existing CTVs. Progress in meeting the above staffing levels would be reviewed annually by the Government and the Bank. 4. Farm: The average group of 20/30 farmers, who should all be neighbors, some natural cultural group or applying the same farming system, vould meet once every two weeks with the extension agent. The meeting should be held on each farm in turn; its program would vary according to the needs of the growing season and the type of farming but would, in principle, include: - explanation of recommended cultural practices; - practical demonstrations; and - delivery of papers required for administrative purposes (credit, inputs, etc.) It is recommended that demonstrations be carried out on only part of the farm to faciritate comparisons between traditional and improved cultural practices. The meeting must be followed by visits to some of the other group members' farms. Organization 5. In order to improve efficiency and accountability at all levels from the CRDA down to the farm level, extension staff should: (a) comply with precise visit programs which would enable supervisors to locate and closely monitor their staff at any time and ensure at field level that farmers would get extension support at regularly scheduled times;  MSEX 32 - AN Page 3 (b) be exempt, as much as possible, from all non-extension activities; and (c) be kept well informed of their tasks and the way to carry them out successfully through thorough discussion at monthly meetings, which would involve research staff. Visit Program 6. The strict respect of the visit program (4 days a week) is particularly important at the farmers' level because this new approach would generate a system in which farmers could rely on getting timely and appropriate technical support and administrative help (filling forms for inputs, credit, etc.). It is therefore important that the fortnight visits (para. 3) be carried out the same day at .he same hour during the whole growing season. Farmers would then be sure to have an opportunity to meet the extension agents to discuss their problems. In addition, in case of emergency, they would be able to locate the extension agent during a meeting with another neighboring group, thus reducing the time they have to spend going to the extension center with no assurance of finding the right person. The four days a week visit schedule should be established by the e-:tension agent prior to the growing season and af.er approval by the regional extension chief, it should remain unchanged throughout the season. During the off-season period that schedule would be relaxed. 7. The frequency, content and need for all existing reports should be reviewed and reduced. Office work should be reduced as much as possible and reserved for essential report writing and form filling. It would be carried out over about 1-1/2 days a week. Each extension agent would maintain a daily diary of activities. 8. To be efficient extension staff need transportation at both the CRDA and CTV level, because farmers in the Project area are spread over very large areas. Experience in Tunisia, even when Projects have supplied additional vehicles for individual agents, shows that they are irregularly available, costly to operate often with paid drivers, and not available for private use. Following successful experience in other countries, it is proposed to overcome these problems by having agents purchase and operate their own vehicles. The Project would purchase the vehicles on behalf of the extension agents who would repay an interest free loan of 100Z of the purchase price of the vehicle using in part a monthly payment for the kilometers travelled on official business. 9. Transfer of non-extension activities: A total transfer of non-extension activities to other staff may not be feasible in the immediate future because it could result in increasing official staff and therefore recurrent costs to an unbearable level for Government, but this should be aimed at as soon as feasible since extension agents should, in principle, devote full time to extension activities.  ANNEX 33- Page 4 10. Monthly Meetings: Extension agents would meet regularly during the last week of the month, or more often if required. This frequency would be determined by the type of crops and the timing of the growing season. Meetings would be held at the CRDA, under the chairmanship of the CRDA extension chief and with the participation of research staff required by the problems to be discussed in order to: (a) review field activities of the last month or fortnight; (b) review problems which arose and how they were solved; (c) introduce the work program for the comirng month and discuss it; and (d) give short refresher courses taught by subject-matter specialists prior to the key periods of the growing season. These meetings could eventually be extended over two days if the subjects to be discussed require it. B. Technical SuRport and Coordination with Research 11. Since a well-organized extension service would be meaningless if it cannot propose technically, economically and socially reliable messages relevant to farmers' needs, great attention would be paid to the technical support that research can provide to extension. Extension staff must be kept informed of the latest improvements in cultural practices and should be able to rely on highly qualified technicians to solve critical issues which can occur during the growing season. 12. For that purpose extension agents have to be backed up at the CRDA level by a group of subject-matter specialists whose qualifications would meet Project needs. This group would include experts drawn from existing staff in the CRDAs as far at possible or seconded from the research organization to the Project from at least the four following specialities: training and extension, livestock, pest control and agronomy. They will devote full time to their speciality and would carry out the following activities: - coordinate with research; - prepare booklets on recommended cropping practices for the main crops; - select and prepare extension messages in cooperation with farmers and research; - prepare and carry out refresher courses; - provide emergency technical support to cope with casual problems; - train staff; and - carry out adaptive trials.  ANNEX 34 Page 5 Coordination with research 13. The CRDA will be responsible for organizing formal links between extension and research at the CRDA level or at Project level so that farming requirements can be discussed and research can look for appropriate solutions and get farmers' feedback. To complete and improve that cooperation, action would be taken at two sensitive levels: subject-matter specialists and field trials: (a) In addition to regular monthly meetings frequent contacts between subject-matter specialists and research staff would keep each other informed of the specific problems faced by both parties and of their requirements (i.e., demand for defining the most cost effective - not necessarily the most productive - quantity of inputs to be -used, etc.). (b) At field level also both staff have to be aware of the constraints and requirements of each agency to improve their understanding; therefore trial and demonstration plots at the farmer's level should be established jointly, extension staff would carry out cultural practices and research staff technical treatments and observations. Recomimended Cropping' Practices, Prep aration of Extension Messages and Refresher Courses 14. It will be the responsibility of subject-matter specialists to prepare small booklets which would summarize recommended practices for each Project enterprise. This literature, illustrated by clear and attractive drawings, would be used by extension agents and distributed to literate farmers. Subject-matter specialists, assisted by extension and research staff, would also build in "diagnostic surveys" to study farmers system constraints and in the light of these findings they would prepare messages for the extension agents prior to key periods of the growing season, provide, when necessary, appropriate materials to promote these recommendations and prepare extension agents to carry out those tasks during monthly meetings (para. 10). Emergency Support 15. In addition to information and training activities, subject-matter specialists must be permanently and directly available to the extension agents when they have to cope with critical problems. Since these problems often need immediate action, all foraal and administrative circuits have to be bypassed, but it would be the extension agent's obligation to inform his direct supervisors.  ANNEX ~ 35 - Page 6 C. Training 16. The major part of extension agents training must involve farmwork in order to train staff to demonstrate efficiently in the field the methods they are promoting. Following an initial testing program for aptitude and the basic five-week training course currently practised in Tunisia, initiation would as far as possible be by working with an experienced extension agent. Thereafter three types of training would be envisaged (a) refresher courses over one or two days (para. 14) to Ikeep field agents informed of the latest cropping developments and prepare them for the on-coming tasks; (b) off-season general training carried out over two weeks sessions to improve their general knowledge and their efficiency; and (c) specialized training for agents appointed to handle special farming situations (UCPs for example). Consideration should be given'to establishing orientation training during the initiation period. 17. The two weeks training sessions are aimed at developing general knowledge of the extension agents and extending it to skills which are not generally taught in agricultural schools such as extension methodology, communication, promotion of farmers' associations, cooperatives, etc. This training would be carried out either on Project premises or in the agricultural training center at Sidi Thabet. They would be organized by the extension directors at the provincial, office or Project level, and matters taught would be chosen according to the development targets of each project. They would be supplemented within the framework of the work program and adapted to any changes or requirements of the work program. Teaching would be carried out either by agriculture staff or by teachers from agricultural schoois or university. These courses should be compulsory for all extension staff and would be carried out in winter or other such period when field work is at it lowest level. D. Other Extension Activities 18. The extension service tasks described above are not a substitute for some existing activities which can be usefully integrated as a complement to the work program with the exception c, demonstration plots managed by Ministry officials which, in general, have been found useless. These activities usually include: - films, radio and TV messages which will be intensified according to the farmers' development; - information days for farmers; - performance awards to farmers; and - production contests, plowing contests, etc.  ANNEK -36- Page 7 E. Monitoring and Evaluation 19. Evaluation of the impact of extension on agricultural production and farm income will be one of the main objectives of the monitoring and evaluation unit. That unit should provide Project or regional extension management with a constant flow of information aimed at monitoring Project activities and eventually at modifying the extension approach through adapting extension activities and messages to the situation. The monitoring and evaluation unit should therefore focus on determining: - the acceptability of technological practices to farmers; - the rate of adoption of improved cultural practices and operational progress; - the eventual bottlenecks in improvement and their origin; - the recommendations for adapting the extension approach to farmers requirements; and - the effectiveness of extension services and the impact of extension messages on farmers' farming systems and income. The monitoring and evaluation unit would also provide assistance in developing the overall extension strategy in the Project area. 244lEfp3l  Table 1 -37- STAFF APPRAISAL REPORT TUNISIA NORTHWEST AGRICULTURAL PRODUCTION PROJECT Proiect Components by Year Totals Including Totals Including Contingencies Contingencies (D Million) mUSS Nillion) 1986 1987 198 1989 Total 1986 1987 1988 1999 Total EXTENSION 0.6 0.8 0.9 0.7 2.9 0.8 1.0 1.2 1.0 4.0 RESiCH 0.3 0.3 0.1 0.1 0.6 0.3 0.3 0.1 0.1 0.9 LIVESTOCK SERVICES 0.5 0.6 0.6 0.1 1.8 0.7 0.8 0.8 0.1 2.5 FARM NACHINERY TESTING 0.1 0.1 0.2 0.0 0.4 0.1 0.1 0.3 0.0 0.6 RURAL ROADS 1.1 1.5 1.0 1.0 4.6 1.5 2.0 1.3 1.4 6.3 PROJECT IDFLEIENTATIoN UNIT 0.1 0.0 0.0 0.0 0.2 0.1 0.0 0.0 0.0 0.3 TRAINING, TECH.ASST.9 STUDIES 0.5 0.6 0.3 0.1 1.5 0.7 0.8 0.4 0.1 2.1 ON-FARR IMlESTHENTS 1.0 2.7 2.7 4.8 11.2 1.4 3.7 3.7 6.6 15.4 SEASONAL INPUTS 0.1 0.5 1.0 1.7 3.3 0.1 0.6 1.4 2.4 4.6 INPUT RESALE CENTERS 0.2 0.3 0.4 0.2 1.0 0.2 0.4 0.6 0.3 1.4 Total PROJECT COSTS 4.4 7.3 7.2 8.8 27.7 6.0 10.0 9.8 12.1 37,9 Februari 13., 1985 15:17  TUNISIA NORTHWEST AGRICULTURAL PRODUCTION PROJECT SUHarv Account by Project Cosponent (D '000) Physical FARM PROJECT TRAINING, INPUT Contingencies LIVESTOCK MACHINERY IMPLEMENTATION ECHSASST,A ON-FARM SEASONAL REGALE- EXTENSION RESEARCH SERVICES TESTING RURAL ROADS UNIT STUDIES INVESTMENTS INPUTS CENTERS Total X A1 ount