Tunisia - Northwest Rural Development - Phase I - Project

DoesiM 4 J.1.4- L%Itb - Thc World: For OMCAL u ReponNeN. 9934 PROJECT COMPLETION REPORT TUNISIA NORTHWEST RURAL DEVELOPMENT PROJECT (LOAN 1997-TUN) OCTOBER 4, 1991 Agricultural Operations Division Country Department I Europe, Middle East and North Africa Regional Office This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  Currency Equivalents YEAR US$ TUNISIAN DINAR (D) 1981 1.0 0.4938 1982 1.0 0.5907 1983 1.0 0.6788 1984 1.0 0.7788 1985 1.0 0.8345 1986 1.0 0.7940 196. 1.0 0.8287 1988 1.0 0.8578 1989 1.0 0.9371 1990 (November) 1.0 0.8000 Weight and Measures Metric System Government Fiscal Year January 1 to December 31 Glossary or Abbreviations ASAL - Agricultural Sector Adjustment Loan BNA - Banque nationale agricole CNEA - National Centre for Agricultural Studies CRDA - Commissariat r6gional au developpement agricole CRGR - Centre de recherche du g6nie rural ERR - Economic Rate of Return FAO/CP - FAO - World Bank Cooperative Program GTZ - Deutsche Gesellschaft FQr Technische Zusammenarbeit ME System - Monitoring and Evaluation System NRDP - Northwest Rural Development Project ODESYPANO - Office de d6veloppement sylvo-pastoral du Nord-Ouest PCR - Project Completion Report PMU - Project Management Unit SAR - Staff Appraisal Report WFP - World Food Program  JO OFnCL USZ ONLY TFt WOtLO BANK Washington. 0 C 204 3 U S A Oftee of OlfctKe-Gwfai October 4, 1991 MEMORANDUM TO THE-EXECUTIVE ECTORS AND THE PRESIDENT SUBJECT: Project Completion Report - TunLsia Northwest Rural Develonment Prolect (Ln. 1997-TUN) Attached, for your information, is a copy of a report entitled "Project CompletLon Report on Tunisia - Northwest Rural Development Project (Ln. 1997-TUN)," prepared by the Europe, Middle East and North Africa Regional Office with Part II of the report contributed by the Borrower. No audit of this project has been made by the Operations Evaluation Department at this time. Attachment ; This document ha rastricted diuubuton and may be used by recipients only in the performace of their oficial duties lu contents may not otherwise be disclosd without World Usk authouiutbon.  FOR OFFICIAL USE ONLY PROJECT COMPLETION REPORT TUNXSIA NORTHWEST RURAL DEVELOPMENT PROJFCT (Loan 1997-TUN) TABLE OF CONTENTS Pa,4e No. PREFACE . . . . EVALUATION SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . iii PART I - PROJECT REVIEW FROM THE BANK'S PERSPECTIVE . . . . . . . . . . 1 1. Project Identity . . . . . . . . . . . . . . . . . . . . . . . . . 1 2. Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 3. Project Objectives and Description . . . . . . . . . . . . . . . . 2 4. Project Design and Organization . . . . . . . . . . . . . . . . 3 5. Project Implementation . . . . . . . . . . . . . . . . . . . . . . 5 6. Project Results . . . . . . . . . . . . . . . . . . . 7 7. Project Sustainability . . . . . . . . . . , . . . . . . . 13 8. Bank Performance . . . . . . . . . . . . . . . . . . . . . . . . 13 9. Borrower performance . . . . . . . . . . . . . . . . . . . . . . 14 10. Project Documentation, Data, Accounts and Procurement . . . . . 15 PART II - PROJECT REVIEW FROM BORROWER'S PERSPECTIVE . . . . . . . . 16 Memorandum Concerning the Evaluation of Ph..se I of the Northwest Rural Development Project . . . . . . . . . . . . . . . . . . . . . . 16 1. Basic Infrastructure . . . . . . . . . . . . . . . . . . . . . . 16 2. Reforestation and Water and Soil Conservation Actions . . . . . . 17 3. Livestock Production and Pasture Improvement . . . . . . . . . . 17 4. Agricultural Development and Support Actions . . . . . . . . . . 17 5. Participation of the Population . . . . . . . . . . . . . . . . 18 6. Monitoring and Evaluation . . . . . . . . . . . . . . . . . . . 18 PART III - STATISTICAL INFORMATION . . . . . . . . . . . . . . . . 19 1. Related Bank Loans . . . . . . . . . . . . . . . . . . . . . . 19 2. Project Timetable . . . . . . . . . . . . . . . . . . . . . . . 20 3. Credit Disbursements . . . . . . . . . . . . . . . . . . . . . . 21 4. Project Implementation . . . . . . . . . . . . . . . . . . . . . 23 5. Project Costs and Financing . . . . . . . . . . . . . . . . . . 24 6. Project Results . . . . . . . . . . . . . . . . . . . . . . . . 26 7. Status of Covenants . . . . . . . . . . . . . . . . . . . . . . 29 8. Use of Bank Resources . . . . . . . . . . . . . . . . . . . . . 31 9. Economic Analysis . . . . . . . . . . . . . . . . . . . . . . . 32 This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  PROJECT COMPLETION REPORT TUNISIA NORTHWES- RURAL DEVELOPMENT PROJECT (Loan 1997-TUN_) PREFACE This is the Project Completion Report (PCR) for the Northwest Rural Development Project in Tunisia for which the L,an 1997-TUN in the amount of US$24.0 million was approved on May 15, 1981. The loan was closed on December 31, 1989, two years behind schedule. A sum of US$6.5 million was cancelled in 1987 and a further US$1.1 million was cancelled at closing. The PCR was prepared by FAO/CP (Preface, Evaluation Summary, Parts I and III), and the Borrower (Part II). A Completion mission by FAO/CP visited Tunisia during November 19-24, 1990. The report is based, inter alia, on findings ot .his mission, the Staff Appraisal Report, the Loan Agreement, supervisio- reports and a project completion report prepared by the project authority with the National Centre of Agricultural Studies in Tunisia.  - iii - PROJECT COMPLETION REPORT TUNISIA NORTHWEST RURAL DEVELOPMENT PROJECT (Loan 1997-TUN) EVALUATION SUMMARY Oowectives 1. The primary objective ol the project was to tackle the problems of the backward and depressed Northwest of Tunisia where the poorest sections of the rural population live. The project was the first phase of a 15 year program that aimed to increase self sufficiency, employment generation, the nutritional standing of the poor, and increased energy production, by: environmental actions through the introduction of new approaches to soil conservation and forestry regeneration, the introduction of a new model of rural development, expanded agricultural and forestry production, and improved health and nutrition. Specifically, the SAR noted that the proiect aimed to: (a) increase the income and improve the health and educational levels of the target population; (b) protect natural resources through reforestation and reduction of soil erosion on crops and forage lands; (c) create employment; and (d) develop local energy sources through the production of fuelwood (para. 3.01). Implementation ExRerience and Results 2. The project mainly achieved its targets in construction of infrastructure (building, roads, schools, and water points). This has increased social welfare (improved quantity and qualit:y of water, provided schooling opportunities that were previously not available, etc.) and increased access to backward areas of the country thereby allowing marketing activities, delivery of inputs, and permitting improved delivery of health and administrative services. The achievement of these targets is important since it will now allow the successful development of input and output services during the remaining second phase of this 15-year development program. For soil conservation measures (reforestation, conversion of highly erodible lands from crops to perennial forages, establishment of permanent pastures or improving natural ranges and on-farm erosion control works), the results were less satisfactory when measured against appraisal targets, achieving less than 50 percent of projections (para. 4.02 d,e,f). Project nctivities to increase agricultural production (wheat, edible legumes, milk and meat) were the least effective (para. 4.02 a,b,c).  - iv - 3. The primary reasons for shortfalls in implementation rate were organizational deficiencies, staff shortages, land tenure and use patt,rns, resistance of small-scale, subsistence fatmers to accept perceived risks in changing production systems and failure to promote development of effective producer groups until toward the end of the project period (para. 5.01). More importantly, it appears that the SAR underestimated the time taken to develop the new model of rural development that relied heavily on the participatioll of local populations- -an activity that the Bank is struggling with in many projects today. Following a mid-term review, Government accepted the need for testing the participative/integrated approach which has proven to be successful following its impleme.itation on a pilot basis. The project has also established good relations between the 'Office" and the rural populations. Taking this invo account, and the fact that the objectives of the project should have been more modest from a production point of view (and rather focussed on institutional, organizational and infrastructural aspects that needed to be established in the first five years of the 15-year development period), the project has been relatively successful. Sustairnabilitv 4. The construction of roads and schools will have a long-term sustainability in improvement of input distribution, product marketing sustainability and increased welfare of the rural communities. Health should benefit from improved water supplies and road construction should lead to a multiplier effect as farmeLs in these backward areas become less subsistence- oriented. The sustainability of other actions financed under the first phase project will depend upon continuation of the 15-year program provided that (a) organizational problems can be overcome; (b) lessons learnt during the first phase can be fully taken on board; and (c) the participative/integrated approach can be replicated over the entire project area. Given that the SAR envisaged a 15-year development period, it was not anticipated that the project be "fully- sustainable" at the end of the first phase. Lessons have been learnt and, as noted, there is now a sound intellectual and institutional basis for future development in which production benefits should accelerate. A second phase, drawing on experience from the existing project, is under preparation. Findings and Lessons Learned 5. Sustainable development of areas with a concentration of small-scale producers can only be achieved through beneficiary participation in project development and implementation and promotion of effective producer groups. Projects which expect a participative approach and producer group promotion require long preparation and start up and participation of social science specialists. Projects of rural development should be part of a wider, longer term program approach. On-farm soil erosion investments require a relatively large public sector participation. Where there is a history of failures in public support to development activities, new efforts must include public sector actions of clear and immediate interest and benefits to the target population. Investment and incremental production credit components aimed to assist disadvantaged small-scale producers without land titles and with small individual credit demand must be specifically designed to meet these particular needs (para. 8.02).  PROJECT COMPLETIgNREPORT TUNISIA NORTHWEST RURAL DEVELOPMENT PROJECT (Loan 1997-TUN) PART I: PROJECT REVIEW FROM THE BANK'S PERSPECTIVE 1. Proiect _dentity Project name : Northwest Rural Development Project Loan No. l1997-TUN RVP Unit : EMENA Region Country Tunisia Sector : Agriculture Sub-sector . Rural Development 2. Background 2 01 The NRDP Project was the tenth Bank-group assisted agricultural project in Tunisia. It was identified as an effort to reduce the extreme po\arty in Northwestern Tunisia. At that time, 17 percent of Tunisia's population lived out or below the absolute level of poverty. This was to be achieved through, inter alia, laying the base for increasing crop and livestock production and lowering the rate of soil erosion in the area. The project idea, which was based on two successful pilot operatio. , was presented to the Bank by the Government in 1978. The project was prepared by a public Tunisian consulting firm, the National Center for Agricultural Studies (CNEA), with important assistance from the Ministry of Agriculture Forestry Department and the FAO/World Bank Cooperative Programme (JrAO/CP). The project was appraised by the Bank in September/October 1980, approved by the Board in May 1981, signed in July 1981 and became effective in March 1982. It closed at the end of December 1989, two years behind schedule after two extensions. 2.02 At appraisal the project was expected to cover a five-year time slice of a 15-year development program with a first phase cost of US$61.5 million of which the Bank would provide US$24 million, producers US$4.3 million, German Technical Assistance US$1.7 million and the Government US$31.5 million equivalent. At closure, the actual project costs reached US$46.7 million and the Bank loan amounted to US$16.4 million after cancellation of US$7.6 million (see Part III, Table 5). The program is being continued with interim financing provided by the Bank financed ASAL II loan. The Government, with assistance from FAO/CP, is preparing a second phase project.  - 2 - 2.03 Policy and Sector Context: The project was part of an on-going Bank support of Tunisia's efforts to increase efficient agricultural production although the special feature of this project was its objective to address problems in one of the most backward regions of the country through an integrated approach. Bank projects in the sector have assisted in the development of irrigation, rural roads, agricultural credit, fisheries, forestry, agricultural dnd rural development. In addition, two agricultural sector adjustment loans have supported the Government's efforts to make necessary changes in macro- and micro-economic policies in the sector (Part III, Table 1). 2.04 These Project ar.d Sector Loans have sought to increase agricultural income, reduce the exodus of the rural population and protect the environment through rational soil use thus reducing erosion. The need for strengthening the agricultural sector is demonstrated by the decreasing relative importance of the sector within the overall economy and the increasing dependence on imported food products. Between 1960 and 1988, agriculture's contribution to GDP has decreased from 24 to 13 percent, employment in the sector had dropped from 54 to 30 percent and the proportion of agricultural to all exports from 6C to 8 percent. While these decreases in the significance of the agricultural sector in the economy were largely due to a rapid growth of other sectors, the rate of decline indicates the need for revitalization of the primary production sector. This need was further exemplified by the ratio of agricultural exports to imports which had dropped from 75 in 1975-76 to 42 percent in 1980-81. 3. Project Objectives and Description 3.01 The primary objective of the project was to tackle the problems of the depressed Northwest of Tunisia where the poorest sections of the rural population lived. The project was the first phase of a 15-year program that aimed to increase self sufficiency, employment generation, the nutritional standing of the poor, and increased energy production, by: environmental actions through the introduction of new approaches to soil conservation and forestry production, the introduction of a new model of rural development, expanded agricultural and forestry production, and improved health and nutrition. Specifically, the SAR noted that the project aimed to: (a) increase the income and improve the health and education of the target population; (b) protect natural resources through reforestation and reduced soil erosion on crop and forage lands; (c) create employment opportunities; and (d) develop local energy sources through the production of fuelwood. In retrospect, it appears that the SAR underestimated the time taken to develop the new model of rural development that relied heavily on the participation of local populations--an activity that the Bank is struggling with in many projects today. In this vein, it underestimated the time taken to establish the right organizational structure to implement this form of participative management and overestimated the production benefits that were likely within the first five-years of a 15-year program (para. 4.07).  -3- 3.02 The project had a total of 12 components which fell under the ge eral headings of organization and management, crop and animal production, in.rastructure, human resource development and natural resource conservation. The specific components are discussed in para. 6.01 and 6.02 and detailed in Part III, Table 5. 4. Project Design and Organizat ion 4.01 The general problems of low income, inadequate health and education and land tenure of the rural population in Northwestern Tunisia and the specific need to reduce the sedlimentation rate in the new Sidi Salem reservoir, constructed to supply potable water to the increasingly urbanized eastern coast, provided a clear conceptual base for the project. These problems were interlinked and efforts to solve anyone or a series of them would assist in the resolution of the others, while individual or non- integrated activities would not attain program objecti-ves. Government and Bank staff involved in project appraisal understood and were in agreement on the concept. 4.02 There were three innovative approaches proposed in the project. The first dealt with the presentation of a 15-year program which recognized that the regional development objectives could not be met during a normal Bank project cycle. The second innovation was related to the micro-zone approach whereby the total area would be divided into manageable geographic areas where an integrated approach could be taken to research, extension, marketing, credit and watershed protection. The project proposed that the most critical micro-zones, from the standpoint of the reservoir protection, be given priority in the t :netable of project implementation. The third innovation was the idea of incorporating the rural population into the development process at the initial stage of micro-zone planning ancd to continually strengthen producer groups with the goal of allowing group extension, demonstration, credit activities and especially soil conservation actions. Producers',groups would eventually take control of certain public services such as water point maintenance, input supply, marketing and operation of milk collection centers. 4.03 The scope and scale of the project as presented in the appraisal reporL was reasonable based on the proposed organization, staffing and budgeting of the project. Here again the idea of a 15-year development program tended to reduce the temptation to attempt too much in one project. In light of the lack of previous developmental activity in the region and the insufficiency of social, technical and production data, the project was surprisingly well prepared. There were weaknesses in the evaluation of the ability to overcome traditional cultural attitudes and the specific approach and resources needed to bring about beneficiary ownership of project goals and activities. This was related to the lack of real life experience of both Bank and Tunisian staff in the application of human resource mobilization proposals particularly the issue of the participation of local populations. The appraisal mission was not unaware of difficulties as indicated in Annex 11, page 9, para. 20 of the Implementation Volume, i.e., "Given the fact that for  - 4 - centuries the population has considered iorage as a gift from heaven, the concepts introduced would represent a radical sociological and economical change which can only be viable in the context of the entire project concept where pasture and forage produtction are offered as part of a package of mutual rights and obligations between the project and the population in which the population benefits only through its cooperation and participation." 4.04 Partially a!; a result of this lack of experience, the Project Implementation Unit, the "Office", did not fully understand the responsibility which they were accepting. It is reasonable to assume that if the higher level staff had fully understood the inplication of the producer group proposals, many of then would not have agreed to the concept due to their own traditional attitudes and the negative experiences of the cooperative movement in Tunisia in the 1960's. 4 05 Ir, retrospect, the project organizational proposals appear to be valid. Deficiencies in this asFect during implementation were due to internal difficulties which limited the creration of the organizational structure and staffing proposed 'n the project. There were a few project design features which, in hindsight, limited effective project implementation but it is not obvious that all of these could have jeen foreseen at the time of preparation and appraisal. 4.06 The questionable design features were. (a) Proposals made for soil conservation often did not adapt to the landholding and usf; patterns; (b) Permanent pasture plantings were based on experiences in the Sedjnane Project which were to be extrapolated to other project areas. Ecological differences prevented the adaptation of the same forage species and there was inadequate adaptive research undertaken to develop new production systems. The proposals for enriched or natural regeneration of native range were not based on previous experience; (c) As menLioned in para. 4.03, there wa& no experience in Tunisia in the incorporation of beneficiaries in projecc goal ownership and implementation, nor did the project propose the incorporation of sociology specialists to seek solutions to the problems which could have been foreseen. Failuro to anticipate these problems had an impact on many project components but especially watershed wide soil conservation works, management of communal pasture and ranges, reforestation and transfer of public sector activities to producer groups (para. 5.01 c); (d) Implementation of certain project components depend greatly on contracting research project to INRAT or other agencies. The appraisal mission failed to adequately consider the legal difficulties in contracting to these non-autonomous agencies and their deficiencies in organization, funding and staffing;  (e) The proposal to drastically reduce cereal production on land with greater slope did not adequately appreciate the cultural and production systems bottlenecks in achieving such a change which is likely to occur only with producer groups to resolve co0m1MU'al pasture and range use, availability of appropriate, proven technological packages and an effective extension program; (f) The appraisal mission was over-optimistic in forecasting the eventual staffing level of the recently created "Office"; and (g) The appraisal missicn failed to anticipate the disequilibria in social and technical aspects which would be created by the difference in two sources o, technical assistance for the Project, i.e., GTZ for Mogods-Kroumerie and loan financing for the remaining two regions. 4.07 As noted also, in retrospect the appraisal underestimated the time taken to achieve the production benefits that could be expected from this, the first phase of a 15-year program. The objectives of the project should have been more modest and focussed more on institutional, organizational and infrastructural aspects that needed to be established in the five-year project period. Produclion benefits should have been less stressed; rather, it was important that the project lay the base for production and income generating benefits that would arise over the remaining 10 years of the 15 year development per4od. As stated in paras 5.01 c and 8.01, this was noted during the course of prject implementation. 5. Project Implementation 5.01 Due to the Government delay in approving legal documents and fully authorizing the PMU (Office du Developpement sylvopastoral du Nord-Ouest, ODESYPANO), the project only became effective in March 1982, ten months after Board approval (Part III, Table 2). In spite of the delay in effectiveness, various project activities were initiated and financed for 1981 since financing retroactive to Loan signing was authorized. ODESYPANO made rapid progress in construction of project infrastructure and initial staff recruitment. However, the implementation rate gradually declined due to weak organization and management and a national freeze on staff recruitment. Purchases of additional vehicles and equipment were delayed or prohibited by the Ministry of Finance. These problems lead to a reduction in total project costs and devaluation in the Dinar reduced US$ requirements of the project. A total of US$7.6 million of the loan were cancelled. The final total projec: cost was Dinars 36 million including Dinar 1.2 million from the World Food Programme to provide food for Distribution to project beneficiaries in partial payment for labor in infrastructure construction and loss of production while changing production systems (Part III, Table 5). As the project proceeded, there were a number of variances between planned and actual implementation. The most important of these were:  (a) The expansion of project activities into new micro-zones proceeded much more rapidly than anticipated and beyond the implementation capacity personnel available. This resulted in a lack of adequate planning and incorporation of beneficiaries in the planning and implementation process as well as individual, specific activities rather than an integrated approach to entire watershed or at least douar level protection within micro-zones. The integration of actions at individual farm level was also incomplete. This rapid expansion was largely due to political pressure applied by the five separate Governorates, each of which wanted to have a maximum of project activities within their geographic area. The lack of maturity of the new "Office" as an administrative body and inadequate support from the Ministry of Agriculture probably contributed to this phenomenon; (b) The organization and function of the "Office" suffered from three major problems. First, while the Government appears to have remained--and, indeed remains--highly committed to the objectives of the project (paras. 7.02 and 9.01), there was several changes of managing directors which complicated understanding and continuity of the Project. Secondly, management established an ex ante control system which resulted in too much centralization of authority and too high a percentage of administrative staff. Thirdly, a freeze on recruitment applied by the Ministry of Agriculture in 1984 resulted in a deficiency of technical staff to adequately implement planning, training and agricultural extension activities, including producer group promotion as well as delaying soil conservation and reforestation; (c) Partially as a result of "b", producer groups, which were the basis for success of the crop and livestock production components, installation of soil conservation structures, road maintenance and water point construction and maintenance, never became functional. Nevertheless, a mid-term review and other supervision missions, noted that the group approach, recommended by the SAR--and fundamental for the success of the project--was not being implemented. In 1986/87, the Bank succeeded in persuading Government to try the group approach on a pilot basis. With assistance from GTZ, this approach was tested in the Sedjenane area with some success--lessons will be incorporated into a possible second phase operation; (d) Project authorities decided to improve potable water supply for the human and animal population through simple actions in existing water sources rather than drilling wells and equipping these with pumps (para. 6.02 e). This change allowed a greater positive impact throughout the project area at a much lower cost. Present indications are that the more sophisticated water sources recommended in the appraisal may be needed in certain micro-zones during a second phase due to increased water requirements; and  - 7 - (e) Government requested and the Bank approved Loan financing of 45 percent of road construction costs. 6. Proiect Results 6.01 The project was very effective in implementing the social and physical infrastructure components. This infrastructure will contribute significantly to further development of the region within the guidelirnes of the 15-year development program. Other components had varying levels of results due to difficulties mentioned in earlier sections of this report and others to be discussed in the component analysis. While the project generally did not meet SAR production targets, in retrospect, it is felt that these targets were overly optimistic and that the project should have focussed more on laying the basis for the fu_.vce (para. 4.07). In this respect, the project was quite successful. Firstly the project's infrastructural achievements were significant. Secondly, the project has established good relations between the "Office" and the rural. populations. And thirdly, the new participative approach is promising in the project area with GTZ assiscance. A second phase project has, therefore, a sound intellectual and institutional basis from which to move forward. Details of project results are set out in Table 4 in Part III of the report. 6.02 Specific component results and impact are as follows: (a) Soil Conservation and Pasture Improvement: As in all components there is a severe deficiency in quantified or quantifiable data. There is a notable difference among data provided to supervision missions, those collected by CNEA in the process of preparing a completion and impact report for the Office (1990) and those provided to the completion mission. Where available, these data are given in the tables of Part III; however, the data received by the completion mission have been used in the impact analysis since project staff have made a concerted effort to update information as a result of discussions with the second phase preparation mission in June 1990. Soil Conservation measures were more readily accepted on larger private or communal holdings where installations of structures would enhance or at least not interfere with present farming systems. Thus construction of rock terraces on contour lines which resulted in destoning arable land or planting of olive or almond trees on rangeland was readily adopted. Construztion of earth or vegetated terraces which required a change in soil tillage methods were mainly rejected, especially by small-scale producers with land holdings running from top to bottom of slopes with widths of 25 to 100 meters. In such cases, contour farming will only be applicable where adjoining producers agree to cultivate their land jointly. Rock terraces on contour were constructed on about 3,800 ha of land with an estimated  -8- reduction in erosion from 10-12 to 5.5-7 tons/ha/yr. Where arable lands were dest,ned, cereal production increased by about 600 kg per hectare. Some 2,800 ha of land were affected by vegetated terraces but these have all disappeared. The project anticipated treatment of 41,500 ha. Treatment of ravines. Ravines were treated through construction of stone walls or more temporary structures with wire or bales of straw. All of these actions reduced further erosion but have a long-term effect only if the sides and floor can be planted with trees, bushes or cactus. The maintenance of this vegetation is only possible with the cooperation of local farmers in controlling grazing by their livestock and this was only partially achieved. Some 5,500 ha of ravines were treated far in excess of the 1,700 ha projected at appraisal; however, there is no current evaluation of ravine control sustainability. Reforestation. This included actions on private, communal and state lands and involved planting of pines, eucalyptus, olives, almonds, acacia, atriplex, cactus and fruit trees. This activity has not been well received on private lands due to a lack of producer interest in woodlots, to small holdings divided in many parcels with farmers feeling that all land is needed for other production and perhaps lack of agricultural extension contacts. A total of some 4,200 ha have been planted, well below the projection of 22,000 ha. Permanent Rasture or improved native range. The project proposed that some 28,000 ha of permanent pasture and improved range land would be developed, some of this to be converted from cereal production. The office invested in the establishment of 7,000 ha of pasture and range improvement, however, data provided to the completion mission indicated that only 1050 ha of permanent prairie remain of which 390 ha will be converted to sylvopastoral parcels and 115 ha to cereal production or fruit tree plantations. Additionally, only 240 ha of improved native range and 540 ha of sylvopastoral areas (planting with atriplex, acacia and/or cactus or olive and almond trees) remain. This component suffered severely from the lack of prodticer group promotion, deficiencies in agricultural extension and the lack of appropriate, tested technological packages. Small-scale farmers, the majority with less than 5 ha divided into three to ten parcels were not able to dedicate land to permanent pastures and without producer group collaboration, it was not feasible to establish natural range management practices on communal or state forest land.  -9- (b) Agricultural Development Research. The research component never got off the ground due to legal difficulties in "Office" contracting to research institutions and the weak programning, staffing and financing of these institutions. This had a major negative impact on the availability of technological packages to recommend to producers. Agricultural Extension, The rapid expansion of the geographical area covered by the project and the restrictions on staff recruitment reduced extension activities to a minimum. The technical staff were thinly spread at the intervention centers, agencies, regional offices and central headquarters and devoted less than 20 percent of their time to ad hoc extension activities. Only in 1987 did the Government approve the recruitment of six additional technicians to work full time in extension in six micro-zones. (GTZ directly employed an additional six agents to work in the Mogods-Kroumerie region). Even these agents had difficulty in maintaining a planned program due to assignment of their vehicles for other uses by higher authorities. In 1987 consultants were contracted to assist in identifying producer groups and train extension agents in group promotion. This work is being continued in Mogods- Kroumerie, but there has been no follow up in the Kef and Beja regions. Crop production intensity. The appraisal proposed that crop productivity on the better quality less erodable soils would be greatly increased due to adaptation of improved technology and that cereal production would be decreased or stopped on highly erodable slopen. Neither of these objectives were achieved due to lack of crop production subject matter specialists, weak agricultural extension and the reluctance of subsistence farmers to accept the risk of changing traditional production systems. Farm input and eguiment suRply. This sub-component functioned well. The intervention centers stocked and distributed WFP rations and fertilizers and pesticides (the latter sold or provided on BNA credit to producers) in a timely manner. This availability of inputs produced an average of 300-400 kg/ha increase in cereal production on fertilized lands. The intervention centers also provided a rental service of equipment for land preparation. This was a useful service on flatter land but may have contributed to erosion on more sloping areas. The Office with CRGR made some efforts to introduce improved cultivation equipment for animal traction but this did not receive adequate emphasis or follow-up.  - 10 - Agricultural credit. Short-term production credit was in demand due to extension activities and the ready availability of inputs at the local intervention centres. The processing of this type of credit was handled entirely by the Office for the first two seasons with a revolving fund. Later an agreement was signed with BNA to handle the credit activities, but credit approval became complicated because BNA was not given responsibility on credit approvals. The intervention centre chief prepared the dossiers for producers in his area and passed them to BNA for creditworthiness review, and processing. BNA had no incentives in analyzing small loans requested with about 50 percent of these being below DT180. In reality, the processing of such small loans at subsidized interest rates is not economical for BNA. It would have been better to group farmers into legal entities such as lending groups. Loan recovery was unsatisfactory because of farmers resistance to commercial short-term credit, and BNA's inability to recover loans it did not grant itself. Except for cisterns to transport water and purchase of livestock, there was little demand for medium or long-term credit. This was slightly influenced by the lack of land titles to provide as guarantee to BNA but was primarily due to the resistance of project beneficiaries to commit to long-term indebtedness. (c) Livestock Development. This component did not achieve projected levels of production primarily due to less than anticipated increases in forage availability, weak technical assistance and lack of promotion of producers groups. Medium-term credit for livestock purchases. A total of 8,000 ewes and 250 heifers were financed for project beneficiaries. Given the lack of improvement in forage resources, this may have been too many. Sheep purchased were generally no better than local flocks so that the goal of genetic improvement was not achieved. Heifers were high grade or pure bred and were not well adapted to the management and nutritional levels on beneficiary farms. Although there was no monitoring of production results from these credit purchased animals, there are indications that many of the sheep were sold to meet cash needs of the producers. Genetic improvement. An average of 50-55 bulls (101 in 1989) were made available in 14 breeding centers throughout the project area. It is estimated that some 30,000 cross- bred animals were produced during the life of the project. Females would have produced some 800 litres per year of additional milk while males would have weighed about 150 kg  - 11 - more at 18 months than local breeds. A total of 22 stallions and 38 jacks were also available to produce horses and mules for animals for field work. | - Milk marketing. In order to stimulate milk production and increase producer incomes, the Offices constructed 21 milk collection centres of which 13 were equipped with cooling tanks. Of these, only eight received adequate amounts of milk to operate in 1989. Centers were built before a proper evaluation was made of forage and livestock production potential. A total of 15,520 tons of milk was collected during project life, of this about 5,500 tons were from small-scale beneficiaries and the rest from larger private and state farms. Milk collections from larger-scale producers allowed more economical operation of the centers. Only centers which could cover all amortizacion and operating costs were kept in operation. - Small animal production. Although not anticipated at appraisal, the project promoted bee and rabbit production as an additional source of income for a few beneficiaries. A total of 112 modern bee hives and 119 female rabbits were distributed to producers financed by medium-term credit. (d) Roads. The project achieved about 60 percent of the target for road construction (660 km vs 1,070 kin) aimed at increasing access to these backwards areas thereby allowing marketing activities, delivery of inputs, and improving social welfare delivery of health facilities etc). The shortfall in road construction was mainly due to delay in purchasing road equipment, higher than anticipated costs per km, long hauls required for base materials, shortage of staff and difficulty in maintaining and replacing equipment. (e) Water Supply. With a reduction in the complexity of water points to be constructed, the project was able to far surpass appraisal estimates. A total of 255 water sources were established or improved (28 projected at appraisal). This was a major project achievement given that water supplies in these areas were inadequate in quality and quantity for the healthy maintenance of the population and for the development of livestock. The impact on time spent hauling water, human and animal health, hygiene of the home and total number of beneficiaries was very positive--a strong social impact. (f) Schools. Twenty four classrooms were constructed i.e. 66 percent of the SAR target. Many communities in the Program area had no access to schools. The construction of classrooms have both reduced the average distance travelled by students and the number of students per classroom. The Ministry of Education has assumed responsibility for operations and some 2,000 students have been  - 12 - direct beneficiaries. (g) Technical Assistange and Training. Two different sources of TA operated in the project. In the region of Mogods-Kroumerie, German assistance (GTZ) was provided continually with 4-6 long- term expatriates; the performance of these consultants was highly satisfactory. In the remaining two regions, the Office contracted TA from various sources although most was coordinated by FAO. Financing was provided by the Bank Loan, UNDP and bilateral assistance. There were two long-term contracts (2 years), but most of the TA was provided by short-term consultants. Generally, the quality of these consultants was satisfactory. However, short-term consultants were unable, within the time of their contracts, to come to grips with problems that were essentially long-term in nature. Successive project directors objected to long-term TA and were not efficient in deciding upon the TA needed or the contracting of short-term consultants. There was a disparity created between the two systems to the point that operationally it was as if there were two separate projects. Staff in the Oued Zarga and Oued Tessa regions felt a great disadvantage in TA, training and availability of vehicles and equipment. The training component was not well implemented by the Office due to the difficulty of relieving essential staff to undertake training and the lack of enthusiasm of the various directors for this activity. (h) Monitoring and Evaluation. Project implementation data were collected by the Technical and Infrastructure Departments beginning in 1982 and provided reliable information while the project area and activities were limited. However, as the responsibilities of this staff increased and a greater area was incorporated, there was a need for a specialized M/E unit. The appraisal estimated that such a unit would be functional by December 1982, but this only occurred in 1987. Even then the Unit focussed on collecting management information data with improperly selected indicators, other departments continued to duplicate information, processing was slow and there was no feedback of reports to the regions. All of these deficiencies created resistance of field technicians to the filling out of multiple information requests from which they saw no results. A program for collecting and analyzing project impact and sustainability is still not in place. In large part, delays in establishing an effective M/E system has been due to a lack of project management commitment to this activity. (i) Financial and Economic Rate of Return. As mentioned earlier, project impact data were not available. It was not possible for the completion mission to obtain data allowing for a without and with project comparison of farm models. The calculation of financial rate of return was therefore not possible. The mission working with project technical staff was able to make reasonable  - 13 - estimates of production incieases in each of the three regions. The estimated actual ERR is slightly lower than anticipated at appraisal (6 percent at seven years compared with 7 percent at five years or 16 percent at fifteen years). The reasons for lower than expected production impact are discussed above as well as the need to consolidate production related components in a follow-up project. 7. Project Sustainabilitv 7.01 At appraisal, it was anticipated that project implementation, within the initial micro-zones would integrate all actions and components so that personnel and equipment for infrastructure development could be moved to new micro-zones, producers groups would take over some public sector activities and the project would only be required to maintain credit and extension activities. With the rapid geographical expansion and a deficiency in projected staff numbers, the integrated approach to planning, producer participation and component implementation was not adequately followed. As a result, physical infrastructure goals were largely met while those for agricultural production and natural resource conservation fell short of expectations. The physical and social infrastructure in place will contribute to the increased welfare of the rural communities. Health should improve and road construction will lead to a multiplier effect as farmers become less subsistence oriented. 7.02 The sustainability of the actions under the first phase project will depend upon continuation of the 15-year program providtd that: (a) organizational problems can be overcome; (b) lessons learnt during the first phase can be fully taken on board; and (c) the participative approach can be replicated over the entire project area. Given that the SAR envisaged a 15- year development period, it was not anticipated that the project be "fully- sustainable" at the end of the first phase. Lessons have been learnt and, as noted, there is now a sound basis for future development in which production benefits should accelerate. 8. Bank Performance 8.01 As indicated in section 4, Bank performance in preparation and appraisal was satisfactory. In general, Bank supervision missions and a mid- term review identified implementation problems but for various reasons were often not able to bring about Government or Office actions to remedy these. There was discontinuity in supervision staffing during most of the project and generally only one or ti, staff participated in supervision. In the light of the multiplicity of components, it would have been desirable to have had more intensive supervision and for various specialists to participate in missions in order to identify and make recommendations to resolve technical and administrative problems.  - 14 - 8.02 The lessons learned from project experience include: a) programs or projects to be implemented in areas with a high concentration of small- scale farmers must promote strong poducer groups if Government services and credit are to be provided efficiently and in order for producers to become collectively responsible for mechanization, input supply and product marketing; b) projects which depend upon a participative approach and producer group promotion require a long start up and involvement of sociology and group dynamics specialists with real life experience in the particular culture; c) the program approach was appropriate in order to allow for slow start up but appraisal estimates of projected production accomplishments in the five-year project were optimistic and underestimated the difficulties in organization set up and staffing; d) the problem associated with a, b and c above are further complicated when small-scale producers are disadvantaged by topography, soil quality, education and poverty; e) soil erosion works and investments must include a relatively high public sector participation since financial results on individual farms are often not immediately demonstrated; f) because of past fragmented and largely unsuccessful government assistance programs in the region, the project needed to rapidly implement actions of clear interest and benefit to the population in order to obtain their participation. In this case, the construction of roads and schools and improvement of water sources caused the potential beneficiaries to be more receptive to changes in attitudes and production systems. Unfortunately, the failure to adequately implement agricultural extension and watershed protection activities reduced the advantage of these new attitudes; g) credit systems should be adapted to the particular needs of small-scale farms requiring very small production credit loans and without land titles 'zo guarantee medium and long-term investment loanis; and h) the early implementation of an effective M/E system and provision of adequate supervision (from the standpoint of time and staff specialities) are both crucial for effective project implementation. 9. Borrower Performance. 9.01 The Government's greatest strength on this project was its commitment to help develop this backward region of the country by first reducing soil erosion generally and slowing down the sedimentation rate of the Sidi Salem reservoir specifically and, second, complying with provisions for programmed and approved budgetary support. The main weaknesses were in: a) failure to maintain continuity of qualified managers; b) failure to monitor and correct deficiencies in organization; c) inadequate provision of technical staff (the freeze on new recruitment was understandable since Government was required to reduce expenditure, however, redeployment of staff would have been possible with adequate commitment to project needs; d) failure to understand the necessity for and to promote producer groups; and e) inadequate field level monitoring and evaluation of project progress from the Ministry level. The Borrower performed well during preparation and appraisal. The lessons to be learned by the Borrower from project experience are: a) the necessity for strong producer groups where the majority of farmers are small-scale and the need to provide adequate numbers and specialties of staff to promote such groups; b) the implementation of actions of national interest (in this case  - 15 - reduction of soil erosion) must be based on actions acceptable to the affected population because of their immediate, medium-and long-term effect on producer income; c) specific public sector activities which have an immediate impact on the populations's living standards (roads, schools, water points) are helpful in convincing the target group of the seriousness of Government's efforts and in obtaining an open mind to recommended clbanges in production systems and in attitudes toward resource conservation and use; d) continuity of an effective organization and high quality project management is essential for efficient utilization of investment and operating funds; and e) differential sources and levels of funding and/or technical assistance should not be allowed in order to maintain harmony, staff morale and standard implementation procedures throughout the project area. 10. Project Docuinentation. Data. Accounts and Procurement 10.01 The staff appraisal report was used by the Office as a guide during project implemer.tation and rroved to be useful. Unfortunately, the Bank did not adequately clarify that it was only a guide and that changes in activities and procedures required by experience were expected. Project management often stuck too close to the letter rather than the spirit of the document. For unexplained reasons, the Implementation Volume annexes were never translated nor forwarded to the Borrower. These could have been of great value during implementation. 10.02. As indicated earlier in the report, the M/E unit was late in being established so that adequate reliable data on physical implementation and project impact were not available. This greatly complicated the evaluation of project impact and preparation of the PCR. On the other hand, disbursement data, financial records, project audits and procurement were fully satisfactory.  - 16 - PROJECT COMPLETION REPORT TUNISIA tIORTHWEST RURAL DEVELOPMENT PROJECT (Loan 1997-TUN) PART II: PROJECT REVIEW FROM BORROWER'S PERSPECTIVE (Translated from the French Version) MEMORANDUM CONCERNING THE EVALUATION OF PHASE I OF THE NORTHWEST RURAL DEVELOPMENT PROJECT' The first phase of the Northwest Rural Development Project involved actions in the following component areas: - Basic infrastructure measures (construction, farm access roads, water supply points); - Water and soil conservation and reforestation measures; - Agricultural producLion measures (livestock production, crop production and pasture improvement); - Monitoring and suppcrt measures (extension services, credit, verification of land tenure, research, training, technical assistance and monitoring and evaluation. During this first phase, the Office conc.ntrated its efforts on the complementarity of and linkages among actions in the various micro-zones. The results achieved' varied from one component to !nother, depending on the special characteristi.cs of the subzones (land tenure considerations, resources used, etc). 1. Basic Infrastructure Project performance with respect to farm access roads was satisfactory (711 km completed out of the 765 km foreseen in the SAR). Problems relating to the terrain and the shortage of equipment for the work units are still the main obstacles to program implementation. * A detailed analytical report prepared by the Office is available in the Project File. .1/ Cumulative achievements as at December 31, 1990.  - 17 - Water supply point deveiopment (267 compared to the 15 planned) and school construction (16 schools + 22 claserooms compared to the 10 planned) exceeded project targets, thanks to the positive impact of these actions on the beneficiaries, who increased their support and contributions to the costs involved. 2. Reforestation and Water and Soil Conservation Actions If the planting done in gullies is included, the SAR estimate of 5,472 ha was far exceeded (4,370 ha plus 2,251 ha of gullies planted with acacia cyanophylla). This action primarily involved public land, similar measures on privately-owned land being pending. Water and soil conservation actions were highly diversified because of the special characteristics of the regions and the uncertainty that often surrounded land tenure. Performance in this area also far exceeded SAR estimates. 3. Livestock Production and Pasture Improvement The objective of this component was partially met in terms of stock improvement, with the number of bulls increasing from 12 in 1982 to 120 in 1990 and the number of inseminations totaling 65,432 (approximately 40,000 of which involved cross-breeding). Credit distribution focused on the acquisition of 249 pregnant heifers, 9,327 ewes, 109 rams and 207 modern beehives, this being ac-hieved despite the unavailability of good breeding stock on the market. In the area of pasture improvement, plans to extrapolate from the Sedjenene pilot project experience (permanent pastures) encountered numerous practical problems (edaphic and climatic conditions and land tenure considerations). A wide variety of actions were therefore carried out (inclusion of forage crops in the cropping pattern, improvement of natural rangeland and planting of trees and pastures). Performance figures include: 2,845 ha of permanent pastures, 2,236 ha of improved rangeland and 660 ha of silvo-pastoral development. However, the reduction in WFP support, the elimination of the subsidy for th3 purchase of forage crop seed in 1986-87 and the prolonged drought have all seriously affected development efforts. This resulted in certain pastures being taken back by their claimants and a sigrificant reduction in the areas earmarked for this component. 4. Agricultural Development and Support Actions The establishment of "intervention" centers within each micro-zone made services accessible to project beneficiaries and ensured that the project's agricultural extension role could be performed. The introduction of extension, adapted and structured in accordance with the specific characteristics of the project actions being undertaken in the individual micro-zone was not accomplished as planned, however, and only six micro-zones were able to benefit from intensive extension on an experimental basis from 1987 onward. The fact that agricultural credit is not adapted to specific needs is still one of the main constraints to crop diversification and intensification.  - 18 - The agricultural research component did not match development efforts in the various components. It focused primarily on adapting animal traction equipment to mountain farming, in conjunction with the Centre de Recherche du G6nie-Rural, and on the impact of CES operations (Travaux) in collaboration with the Instltut de Recherche Forestier, the Direction des Sols (Department of Soils) and ORSTOM. 5. ParticiDation of the Population The participation of the population has occurred on an informal basis and has been both direct (work participation days, construction materials, etc.) and indirect (leasing of pastures, plots of land, etc.). The value of this participation over the last five years is estimated at D 2 million. 6. Monitoring and Evaluation The Office was rather late in establishing a monitoring and evaluation system after the start-up of the Northwest Project. The incorporation of the monitoring system within the administrative, accounting and financial organization of the Office has therefore not been studied. The monitoring of program actions in physical terms was handled quite well but certain aspects of the project's impact that are more difficult to investigate will be covered by a more extensive effort during Phase II.  -19 - PROJECT COMPLETION-RUPORT TUNISIA NORTHWEST RURAL DEVELOPMENT PROJECT (Loan 1997-TUN) PART III: STATISTICAL INFORMATION 1. Related Bank Loans Loan Number Yoar of Prpoe Loan Amount Status apprval (US$ miion) - TUmN 1983 Central Tuia 13.7 Under irgation Projec implementation zs0 - TUN 1985 Northwest Ariculturai 83 Under Production Project implementauion 2573 - TUN 1985 Irrigtion Managemen 22o Under and Improvement implementation Project 2605 - TlJN 1985 Gabes Irrnbaton Project 2L7 Under implementation 2865 - TUN 1987 Fourth Agiculnural 30.0 Under Credit Project implementation 2S70 - TUN 1987 Forey Development 20.0 Under Project implementation 2962 - TUN 1988 Sector Adjument Loan 150.0 Under implementation 3078 - TUN 1989 Second Sector 84.0 Under Adjustment Loan implementadon 347-TUN 1990 Research and Ee =o 17.0 Not yet effective Project  - 20 - 2. Project Timetable Item Planned Date Rei4sed Date Actual Date Reconnissance VI 1978 Idend cation X 1978 Preparation X 1978 VII 1979 Pre-appraisal VI 1980 Appraisal V 1980 VI 1980 Dx 1980 Nigocations m 1981 IV 1981 Board Approval V 1981 V 1981 Loan Signature V 1981 VII 1981 Loan Effectiveness vn 1981 -m 1982 Loan Clsg rx 1987 -xn 1989 Loan Completion 1m19 -88 VI 1990  - 21 - 3. Credit Disbursements A Cumulative Estimated and Actual Disbursements MDA Fbc c Qmzw Euii S AcuaW co 192 xI 1981 . m 11982 . Vl1198K2 500 500 0 IX 198 550 1SW0 0 XII 1982 lt250 2 100 437 437 3 21 m 1983 1 0so 3 L0 2101 2 538 IS 81 VI1 9 1 05 4 200 521 3 059 19 73 1984 K 1993 1040 5 240 766 3 S52 23 74 XII 1983 1 040 6 280 3 &52 23 61 m 1984 1 040 7 320 957 4 782 29 65 VI 1984 1040 8 360 67 S 409 33 65 18S K 1984 1160 9 S20 230 5 639 34 59 XI 1984 1160 10 680 609 6 248 38 59 m 1985 1160 11840 388 6 636 40 56 VI 1985 1160 13 000 877 7 S3 46 58 1986 K 1980 1 240 14 20 457 797C 48 56 xC1I19S5 1240 15480 179 8 149 S0 53 m 1986 1 240 16 720 1 042 9 191 56 55 VI 1986 1 240 17 960 800 9 991 61 S6 197 1986 1 510 19 470 6 9 991 61 51 XI[ 1986 1 510 20 980 582 10 579 64 50 m 1987 1510 22 490 518 11 097 67 49 V' 1987 1 510 24 000 91 11 188 68 47 1988 K 1987 - 1114 12 302 75 51 XE 1987 - 232 12 534 76 52 m1188 * * 750 13 284 81 55 VT 1988 - 289 13 573 83 57 1988 K 1988 - . 693 14 266 87 59 M 119B8 1 186 15 452 94 64 1P1989 - 154S2 94 64 V1 1989 - - 15 452 94 64 1990 DX19S9 . . 15452 94 64 XM 1989 534 1S 986 97 67 l1990 - . - LS986 97 67 VI 1990 - 460 16 446 100 69  Credit disbunememts B Disbursements by category * ___ , .. . .__________________________________ Category FY FY FY FY FY FY FY FY TOTAL SAR Actual as 1982 & 1983 1984 1985 1986 1987 1988 1989 10 EstAmst Estoate Iri' j1~~~~4 ~~~'S uz TOTAL EstImat~~~~Esl%mof ............... __US$' 0 .......... % Construction of 319 522 282 345 197 1 036 196 651 3 548 3 100 114 buildings Vehicles, equipment 2 460 699 328 313 372 183 21 174 4 550 7 300 62 and livestock Soil conservation, 233 907 659 1 187 495 512 229 482 4 704 4 700 100 forest and pastures M. and L. term credit - - 110 77 - 269 91 122 669 1 800 37 Incremental ST credit - - 293 195 - 292 288 479 1 547 400 387 Consultants' services 26 163 351 203 113 94 60 85 1 095 2 4cX1 46 Studies 21 59 81 158 20 - - - 339 1 000 34 Agricultural research - - - - - - I - I 500 0 Unallocated - - - - - - - - - 2 800 - Sub-total 3 059 2 350 2 104 2 478 1 197 2 386 886 1 993 16 453 24 000 69 "Fund - 7 TOTAL 16 446  - 23 - 4. Project Implementation Indicators Unit Appraisal Actual or PCR I I Estimate I/ Estimate 2/ _ Buildings Number 25 40 Milk Collection Centres Number 14 21 Water Points Number 23 278 Classrooms Number 36 24 Roads kms 1065 657 Animal Reproduction Number 23 14 Centres Bulls Acquired Number 211 232 Maximum Bulls in Service Number 50 101 Reforestation ha 9 468 4 324 Pastures ha 28 140 7 000 .3/ Erosion Control ha 10 470 10 115 4/ Ravine Correction ha 810 2 015 4/ Intensified Cropping ha 11 175 2 000 Microzones Affected Number 45 42 Area Covered by Project ba 90 000 123 000 Stamfg Central Directors and Staff Number 71 77 Regional Directors and Number 45 51 :/ Planning Subject-Matter Specialists Number 40 8 Extension Agents Number 59 19 Credit Agents Number 22 20 Veterinary Aides Number 23 0 Milk Collection Agents Number 9 8 Monitoring and Evaluation Number 7 4 1] Based on seven years of projected 15-year programme. 2/ Based on incomplete data. .3/ Reported to supervision missions. Office data to completion mission indicates 1 330 ha actually 3dsL I4/ Actual susainability not known. / No planners at present except in Mogods-Kroumene.  _ 24 _ S. Project Costs and Financing A Project Costs Item SAR Esdmate 1 Aca Costs Actwa Costs as % of SAR Estimates I___________ Local I Foreig I Total LoalI Forcign Total _ Infrastrucmms 2 580 2 039 4 619 5 363 2 357 7 720 167 roads, schools, wvater supplies, itorvention centres Soil conervadon: 6 278 1 252 7 530 2 420 3 675 6 095 81 pastures forestry and woodlots Equipment 3 677 6 258 9 935 1 362 3 493 4 855 49 Construction: Ofrice 198 156 354 839 606 1 445 408 buildinp Extension ;/ 51 202 253 655 70 725 287 Rescarch 289 199 498 3 14 17 3 studias 10 398 498 48 272 320 64 Technicul asitance 1 953 1 953 21 3 518 3 539 178 On-farm investment 1 064 1 067 2 131 502 551 1 053 49 / Revolving fund for 124 495 619 991 1 333 2 324 375 ST acedi Operadon and 11 978 3 482 15 460 7849 7 849 253 maintenance cosM 4/ TYTOTAL 26 349 17 501 43 850 20 053 15889 35 942 82 J/ Based on first seven years of projected 15-year prograumme. 2/ Actual costs include purchasing costs of bulls and rams their maintenance costs and costs of deaionstra