Tunisia - Rural Roads Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 7182 PROJECT PERFORMANCE AUDIT REPORT TUNISIA: THIRD HIGHWAY PROJECT (LOAN 1601-TUN) April 11, 1988 Operations Evaluation Department 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 authorizatic . TK WOLO BNK IOR OFFICIAL USE OILY THE WORLD SANK Washington. D.C. 20433 U.S.A. Okae of iecato-Gawal Operatims IvAstian April 11, 1988 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Performance Audit Report on: Tuniaa Third Highway Project (Loan 1601-TUN) Attached for information is a copy of a report entitled "Project Performance Audit Report on Tunisia: Third Highway Project (Loan 1601-TUN)" prepared by the Operations Evaluation Department. Attachment This document has a restricted distribution and may be used by recipients only in the perfornae of their oficial duties. Its contents may not otherwise be disclosed without World Bank authorization. FOR OFFICIAL USE ONLY PROJECT PERFORMANCE AUDIT REPORT TUNISIA: THIRD HIGHWAY PROJECT (LOAN 1601-TUN) TABLE OF CONTENTS Page No. Preface. .. . ... .. .-. . .... . .. .. .. .. . ... . . . . . .. ... *. . . . ........ I Basic Data Sheet.................................. 1i Evaluation Sury..............**... ....... ................. v PROJECT PERFORMANCE AUDIT MEMORANDUM I. INTRODUCTION................. ...... ................ 1 II. THE THIRD HIGHWAY PROJECT ............ .... .t. ....... .... 2 III. RURAL ROADS AND RURAL DEVELOPMENT IN TUNISIA............ 5 IV. CONCLUSIONS.................... *...............*...... 7 Notes*.#... . . .....o . ..... ....... . . .o......o..........o 8 Borrower's Comments...... . ....... . ................ ............ 13 PROJECT COMPLETION REPORT: THIRD HIGHWAY PROJECT I. Introduction and Background.................oo..oo..ooo 19 II. Project Preparation and Appraisal ...................... 20 III. Project Implementation... o ............ o ... . .......... 22 IV. Project Cost and Disbursements ......... ............. 35 V. Institutional and Operational Development.......o...... 37 VI. Economic Reaevaluation.. ............ t ....... .t. ...... . 38 VII. Role of the Bank .........o........................... 41 VIII. Conclusions ....................... 42 Tables 1. Execution of Part A: Road Improvements.................. 43 2. Execution of Part B: Road Maintenance Equipment......... 44 3. Execution of Part C: Construction of Extension Centers.. 46 4. Execution of Part Cl Extension Centers Equipping....... 47 5. Status of Part D2 on 12-31-82: Medium-Long Term Credit.. 48 6. Economic Rates of Return . .............. ...... .t..... 49 7. Comparison of Traffic in the with- and without- Project Situations .............................. 50 Annex Disbursement Schedule ... .....................*** *** 51 IBRD 19678 (PCR) - TUNISIAs Third Highway Project 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 authoriation. PROJECT PERFORMANCE AUDIT REPORT TUNISIA: THIRD HIGHWAY PROJECT (LOAN 1601-TUN) PREFACE This is the performance audit of the Third Highway Project for which Loan 1601-TUN in the amount of US$32.0 million was approved in July 1978. The Project Performance Audit Report (PPAR) consists of a Project Performance Audit Memorandum (PPAM) prepared by the Operations Evaluation Department (OED), and a Project Completion Report (PCR) prepared by the Borrower's Executing Agencies (Tunisian Highways Department, and Agricultural Production Department) and by the Bank's Europe, Middle East and North Africa Regional Office. The PPAM is based on a review of the Staff Appraisal (SAR) and the President's Report (PR), of the legal documents and the transcripts of the Executive Directors' meetings which considered the project, related documents in the Bank's files, and information supplied by the Borrower. After studying the file and discussing the project with Bank staff in Washington, OED conducted a field mission to Tunisia in June 1987 to secure the views of Government officials involved in the inception and implementation of the Third and Fifth Projects and in the preparation of the Seventh. The mission travelled on many roads improved under the Third Project in different parts of the country and met with local farmers and laborers. It also met extension workers and staff members of provincial administrations and can testify to the enthusiasm and dedication displayed at all levels of the Tunisian services involved in rural development. The hospitality, cooperation and assistance extended to the mission is gratefully acknowledged. The PCR ably summarizes the experience gained from the design of the project, the results achieved, and the lessons which helped in the preparation of subsequent Bank-assisted operations related to rural road development. The principal focus of the PPAM is on Government and Bank efforts to devise planning procedures for rural roads in particular and for rural development in general. A copy of the PPAR was sent to the Borrower and the comments received are appended to the PPAM. PROJECT PERFORMANCE AUDIT BASIC DATA SHEET TUNISIAs THIRD HIGHWAY PROJECT (LOAN 1601-TUN) Key Prolect Data Borrower Government of Tunisia Executing Agencies Directorate of Public Works Ministry of Agriculture NT (National Bank f Tunisia) Fiscal Year of Borrower January 1 - December 31 Appraisal Actuall Estimate Reestimated (US$ million) Total Project Cost 69.2 11 53.1 11 Loan Amount I 32.0 30.0 Cancelled 2.00 Undisbursed 0.15 Disbursed 29.85 Repaid IBRD 9.81 Repaid Third Party 2.37 Due IBRD 16.65 Due Third Party 1.02 Project Completion Date December 1983 December 1986 Economic Rate of Return 11-272 N/A 11 If short- and medium-term credit to farmers (representing amounts not financed under this project) is included, the total would be increased to US$93.6 million equivalent. 12 The scope of the project was changed, part scaled up and part eliminated. Due to the appreciation of the US dollar, there was both an underrun in terms of dollars and an overrun in terms of dinars. 13 Actual figures from the September 30, 1987 Statement of Loans. Proiect Chronoloxy Project Appraisal October 1977 Negotiations May 1978 Board Approval June 1978 Loan Agreement Signed .July 1978 Loan Ettactive April 1979 Final Disbursement September 1987 Loan Closed March 1986 PCR Received by OED March 1987 Cumulative Disbursements (US$ million) FY79 FY80 FY81 FY82 FY83 FY84 FY85 FY86 F87 Bstiated 0.3 3.8 14.7 20.0 25.9 32.0 Actual 0.5 2.0 6.2 10.8 12.5 21.2 27.0 ... 2 of estimated 12.2 13.5 31.5 41.6 38.9 Country Exchanve Rates (Tunisian Diner (TD)IUS$) Appraisal year averages 0.435 Intervening years averages 0.705 Completion year average: 0.850 Follow-on Projects Project lame: Fourth Highway Project Loan Numbers 184*,-TUN Amount: US$36.5 million Loan Agreement Date: May 1980 Project Name: Second Rural Roads Project (Fifth Klghways) Loan Number: 2108-TUN Amount: US$35.2 million Loan Agreement Date: may 14, 1982 Project Name: Highway Maintenance and Rehabilitation Project (Appraised March 1986) iv Total Bank Staff Time (Staffweeks) FY76 FY77 FY78 FY79 FY80 FY81 FY82 FY83 FY84 FY85 FY86 TOTAL Preappraisal 27.1 7.3 23.2 57.7 Appraisal 99.0 99.0 Negotiation 17.3 17.3 Supervision 0.1 40.9 22.3 16.8 13.3 11.4 11.1 14.4 4.5 134.7 Other 1.7 0.1 0.1 9.5 0.1 11.7 TOTAL 320.3 Mission Data Mssion Number Man Report Date of Staff Weeks Date Identification November 1975 5 14 February 1976 Preparation June 1976 3 3 June 1976 Preparation May 1977 2 2 June 1977 Preappraisal August 1977 3 7 September 1977 Appraisal October 1977 4 18 June 1978 Subtotal 44 Superv. I July 1978 3 9 August 1978 Superv. II September 1978 3 9 September 1978 Superv. III November 1978 2 2.5 December 1978 Superv. IV March 1979 4 4 May 1979 Superv. V June 1979 2 0.5 July 1979 Superv. VI October 1979 2 0.5 November 1979 Superv. VII February 1980 1 1 February 1980 Superv. VIII April 1980 1 2 Nay 1980 Superv. IX July 1980 1 1 July 1980 Superv. I July 1980 5 5 August 1980 Superv. XI September 1980 4 3 October 1980 Superv. XII February 1981 2 1 March 1981 Superv. XIII may 1981 2 2 May 1981 Superv. XIV October 1981 4 2.5 December 1981 Superv. XV April 1982 3 4.5 May 1982 Superv. XVI September 1982 4 4 October 1982 Superv. XVII February 1983 1 0.5 March 1983 Superv. XVIII May 1983 3 1.5 July 1983 Superv. XIX March 1984 3 1.5 May 1984 Suptirv. XX August 1984 2 1 September 1984 Superv. XXI October 1984 2 2 December 1984 Superv. XXII February 1985 3 1.5 March 1985 Superv. XXIII September 1985 2 1 December 1985 Superv. XXIV July 1986 2 0.5 August 1986 Subtotal 61.0 Total 105.0 PROJECT PERFORMANCE AUDIT REPORT TUNISIA: THIRD HIGHWAY PROJECT (LOAN 16013-TUN) EVALUATION SUMMARY Introduction In addition to mining and tourism, agriculture is an important sector in Tunisia. Regional development and better income distribution are among the Government's basic objectives whos" purarit the Bank has been assisting since the early 1960s. Road Improvements, and carefully selected "parallel" investments in a road's "sone of influence*, are indispensable for agricultural growth, and the Thirl Highway Project tested a variety of interventions, to help Government plan and Implement a nation-wide rural development program (PPAM, paras 1 and 10). Objectives The project established a pilot scheme to raise agricultural production and standards of living in rural areas by Improving about 1,000 km of roads and by financing "parallel investments", better extension services, purchases of rural roads maintenance equipment, and technical assistance to the Ministries of Public Works and Agriculture. With the exception of *parallel investments", all other components achieved the targeted objectives (PPAM, para 4 and PCR, para 2.04). Implementation Experience (PPAM, paras 5-6) Project effectiveness was delayed by about five months (PCR, para 3.01). Road components were launched in early 1979. Agricultural components began in late 1979. Counterpart fund shortages delayed project completion from December 31, 198! to December 31, 1986 (PCR, para 3.02). Road improvements (PCR, paras 3.04-3.15) were done on a total of 1,089 km and had a time overrun of 2.5 years. Cost overruns were modest: about 101. Procurement of road maintenance equipment (PCR, paras 3.16-3.20) was delayed, mainly because of uncertainties about budgetary allocations, and registered major cost overruns. Measures to strengthen agricultural extension (PCR, pares 3.21-3.35) were, on the whole, satisfactorily implemented. Credit for complementary investments had disappointing results (PCR, paras 3.36-3.42). Feasibility study for a follow-on project (PCR, paras 3.43-3.46) was satisfactorily completed. Results The project was satisfactorily completed (PPAH, para 9). Total actual cost amounted to US$53.1 million, as against US$69.20 million estimated at appraisal. Comparisons between estimated and actual figures do not yield meaningful conclusions because of the strong appreciation of the US dollar (about 621 over the project period) which led to a 231 decrease of the actual dollar costs and to a 242 increase %f TD costs (PCR, paras 4.01-4.03). Planning and monitoring have been improved at both the national and sectoral levels and led to better preparation of the follow-on Fifth Highway Project vi (PCR, paras 5.01-5.03). Progress has been made in confronting the road maintenance problem and the proposed Highway Rehabilitation and Maintenance Project will continue to pursue reaseonable solutions (PCR, para 5.04). Lack of data precludes the re-estimation of the rate of economic return according to the methodology used at appraisal (PPAM, para 8). Sustainability Institutional benefits are fully sustainable. Roads have been upgraded and are in good repair; improvements in agricultural practices through extension services continue to take place; and Government has developed planning and monitoring procedures to support rural development (PPAMX, para 9). Findings and Lessons The interdependence assumed ac appraisal between road improvements and "parallel investments" could not be quantified after project completion. It might be preferable to separate "road" from "agricultural" components even though this may it hard to justify rural road improvements on "economic" grounds". Rural road improvements will bc better justified if the social benefits, which constitute the principal gains procured by this type of investment, can be quantified and introduced in the economic calculus. This may be a legitimate area for future research (PPAM, para 14). Planning and monitoring procedures developed in the course of the project deserve to be disaggregated so that they can be used at the Governorate level (PPAM, para 15). Furthermore, experience available in Tunisia for rural development planning and monitoring might be,shared with other countries (PPAM, parA 19). The proposed Seventh Highway Project, under preparation by Government, might consider making provision for these two suggestions. The Government and the Bank made serious efforts, through the Third and Fifth Highway Projects, to determine how benefits from an improved transport infrastructure can be internalized within the agricultural and rural development sectors. Considerable experience is now available and it deserves to be properly disseminated. The audit feels that upon the completion of the PCR for the Fifth Highway Project, OED should carry out a study of the overall impact of transport improvements upon the rural environment of Tunisia (PPAM, para 17). PROJECT PERFORMANCE AUDIT MEMORANDUM /4 TUNISIA: THIRD HIGHWAY PROJECT /1 (LOAN 1601-TUN) I. INTRODUCTION 1. Independent since 1956, Tunisia (164.0 thousand km2, 6.4 million inhabitants and US$1310 per capita GNP in 1980) has an increasingly diversified economy, with agriculture playing a significant part and generating about one third of t-tal employment. Agricultural output is varied and can be increased in volume and quality. Rural development is not easy because much of the country is arid or semi-arid, less than five percent of the arable land is irrigated, and areas relying on rain-fed agriculture show wide fluctuations in rainfall from year to year. Mineral resources are modest. The most important raw materials are phosphates, oil and natural gas, the latter two approaching depletion. There is considerable tturism potential. Petroleum, phosphate mining and processing, and tourism provided about 70 percent of total foreign exchange earnings in the early 1980s. 2. Since the early 1960s, Bank lending 11 aimed to support Government efforts in four areas: employment generation; regional development and better distribution of income; export promotion; and development of basic infrastructure. Lending for transport, power, and tourism infrastructure accounted for 33% of the total volume of Bank assistance. Total Bank lending for the transport sector through FY1984 amounted to US$238.5 Million and, apart from roads, covered all modes except air.13 3. Road transport is the dominant mode for both freight and passengers and its share of total demand has been increasing. Starting in 1970, the Bank assisted in strengthening the road sector through a series of projects. The T1ird Highway Project, the object of this audit, was prepared through the Bank's involvement in two previous operations /A and was followed by the Fourth /5 and Fifth Highway Projects./I A Sixth Project was approved in December 1987 and in mid-1987 the Government is preparing a Seventh Project. The Third, Fifth and Seventh Projects are focussed on the rural road network. 1 N Notes are at the end of the text. 2 II. THE THIRD HIGHWAY PROJECT 4. The project succeeded in establishing a pilot scheme to raise agricultural production and standards of living in rural areas by improving about 1,000 km of roads and by financing better extension services, purchases of rural roads maintenance equipment, and technical assistance to the Ministries of Public Works and Agriculture (PCR, para 2.04). Main benefits from the project vTere cheaper transport of agricultural inputs and crops, and increases of farm income and employment. Ptoiect Implementation and Results 5. Project effectiveness was delayed by about five months (to November 1978) due to delays in the official ratification of the Loan Agreement (PCR, para 3.01). Road components were launched in early 1979. Agricultural components began in late 1979. Problems related to scarcity of funds for both the Ministry of Public Works and the Ministry of Agriculture contributed to the delay in project completion from December 31, 1983 to December 31, 1986 (PCR, para 3.02), thereby necessitating three amendments to the Loan Agreelent (PCR, para 3.03). For each of the five major components, implementation experience can be summarized as follows Part A: Civil Enaineerina Works Associated with Road Improvement (PCR, paras 3.04-3.15). Work carried out during 1979-1983. Total road length: planned 1,068 km/ actual 1,089 km. Total time overrun: 2.5 years. Most of the delays were due to financial or labor difficulties encountered by the contractors. Cost overruns were modest: about 10Z Part B: Procurement of Road Maintenance Equipment (PCR, paras 3.16- 3.20). In two phases. Phase I: equipment originally intended for the needs of 8 maintenance brigades. Main contracts awarded in 1980, roughly within the time foreseen at appraisal. Phase II: equipment identified during the implementation of the Fourth Highway Project and the preparation of the Fifth. Most contracts were awarded towards mid-1984 because of procurement complexities and uncertainties about budgetary allocations. This contributed to the extension of the closing date until December 1986. Total cost showed serious overruns: from TD 2.3 million at appraisal to TD 10.3 million at completion. Part Ct Measures to Strengthen Agricultural Extension (PCR, paras 3.21-3.35). The component consisted of construction of regional extension centers; purchase of equipment for the extension services; recruitment of extension agents specia'izing in communications with the rural environment; technical assistaice with the introduction of new crop-growing methods; and training for extension agents. In May 1984 an Amendment was agreed, to add a cold storage depot for seed potatoes. Final cost is TD 1.3 million (42% for extension centers; 522 for equipment; 6% for technical assistance) as against TD 1.2 3 million estimated at appraisal. The potato depot cost TD 974,700 and registered a 2.6% overrun above the TD 950,000 contract price. Part D: Credit for Complementary Investments (PCR, paras 3.36-3.42). Three components related to granting of credits and subsidies through the Banque Nationale de Tunisie (BNT) to fund complementary investments. Co2Monent D1 (TD 3.1 million) was to fund purchases of inputs by farmers (pesticides, fertilizers and feed concentrates). Very little funding provided (partly because farmers secured credit from other sources at more advantageous terms) and the precise amount is not available since BNT accounts are held by province. Component D2 (TD 7.5 million) to fund complementary investments by farmers (purchase of agricultural machinery, dairy cows and sheds, as well as small scale irrigation and well construction). Only TD 0.34 million lent out mainly because other sources of financing were available. Component D3 (TD 4.0 million) for agro-industrial investments. Results were even more disappointing than under D2. Part E: Identification and Feasibility Study for a Second Rural Roads Project (PCR, paras 3.43-3.46). This included economic, technical and financial studies for about 2,500 km of rural rcads and for complementary agricultural investments in their zones of influence. Final cost: TD 632,600 vs. 530,000 estimated at appraisal. The main study (feasibility of 2,600 km of rural roads) began in May 1979 and was completed in December 1980. Othez studies were engineering studies for about 640 km of rural roads improved under the Third and Fifth Highway Projects. 6. The project improved planning at both the national and sectoral levels and this contributed to better preparation of the follow-on Fifth Highway Project (PCR, paras 5.01-5.03). Progress has been made in confronting the road maintenance problem which, in part because of budgetary difficulties, may become serious as more and more rural roads are improved. The proposed Highway Rehabilitation and Maintenance Project will continue to pursue reasonable solutions (PCR, para 5.04). Project Scope, Prolect Cost and Disbursements 7. The project scope was modified in two respectsa first, additional maintenance equipment was procured and put into use for the follow-on project (PCR, para 3.16). Second, the facilities and equipment program to be handled by BNT was largely repltced by a potato cold storage depot (PCR, paras 4.36- 3.41) whose construction was handled by the Ministry of Agriculture and whose management was contracted out to the Groupement Interprofessionel des Legumes (GIL), a vegetable producers and trade association. Total actual cost amounted to US$53.1 million, as against US$69.20 adilion estimated at appraisal. Comparisons between estimated and actual figures do not yield meaningful conclusions because of the strong appreciation of the US dollar (about 62% over the project period) which led to a 23% decrease of the actual dollar costs and to a 242 increase of TD costs (PCR, paras 4.01-4.03). Reduced agricultural investments led to a July 1986 reduction of the loan amount from 4 US$32 to US$30 million. The loan was closed on December 31, 1986; all but US$145,000 was disbursed (PCR, paras 4.04-4.05). Economic Reevaluation 8. The project was a pilot exercise to assist Government in developing a country-wide rural development program. At appraisal, emphasis was placed on multi-sectoral, multi-component project preparation and implementation (SAR, para 5.04). Road Improvements and agricultural investments were taken as an integrated package, because, on their own, road improvements could not show satisfactory rates of return. Since "parallel" investments financed under the project were considerably fewer than anticipated at appraisal, economic re- evaluation would have to be done on the basis of benefits generated by road improvements only. Lack of suitably detailed data precludes the duplication, for comparison purposes, of the analytical framework used at appraisal and, consequently, no re-estimated rate of return can be given. Conclusions 9. Data considerations preclude the formulation of conclusions based on a rigorous "before" and *after" comparison, and this dictated that, at least, a qualitative judgment be made of the project's achievements. Accordingly, the OED mission traveled on about twenty of the roads improved under the project and spent much time talking to local farmers. Their responses support the view expressed in the PCR (paras 6.06-6.08) that although it is difficult to attribute specific percentages of the increased economic activity to road improvements, better roads did contribute to rural development. The audit9s conclusion is that, for numerous, and important, reasons, the project was a justified operation. It was a pilot effort to test multi-sector interventions and lessons have been applied in subsequent projects. Roads have been upgraded and are in good repair. Improvements in agricultural practices through project- financed agricultural extension services continue to take place. National and provincial agencies active in rural development have successfully collaborated on the implementation of the Fifth Project and the preparation of the proposed Seventh Project. Better planning and monitoring, made possible in part by the project, is a valuable institutional resource for the Tunisian authorities which, as shown in the next section of the PPAM, made important contributions to a more precise definition of relationships between rural roads and rural development. 5 III. RURAL ROADS AND RURAL DEVELOPMENT IN TUNISIA 17 10. In the mid-1970s, Government and the Bank agreed to proceed in stages towards the formulation of a nation-wide rural development program. The Third Highway Project would be the pilot operation to identify rural roads (and their "zones of influence"); to indicate the type of road improvements and of "parallel" investments required; to channel funds through competent agencies; and to assist with works execution and with monitoring of the developmental Impact. A major problem was the absence of a methodology for the prioritization of individual road improvements. Acceptable formulas had long bec. available for major highway improvements but not for rural roads which seldom carry more than a few hundred vehicles. This led to a vicious circle: rural areas had productive potential but required roads to exploit it; rural roads could not generate enough benefits to justify the necessary improvements; therefore, rural roads could not be justified in the customary fashion, funds could not be made available, and the absence of road improvements inhibited rural development. 11. Among the attempts to resolve this impasse, one solution, developed by World Bank staff in 1976, /1 suggested that transport cost savings might be translated into increased agricultural production and income. In 1979, the further suggestion was made that road improvement and agricultural development ought to be integrated because the absence of complementary agricultural programs often led to non-achievement of forecasted traffic (and hence of the forecasted economic rates of return) on road improvement packages. 12. Tunisian officials and Bank staff collaborated to translate these concepts into an operational program 1 which would focus on individual rural roads and their zones of influence; would develop an investment package for each zone of influence; and would integrate the individual packages into regional and national development plans. 110 Principles were simple but data processing and implementation of the program was not. For instance, it turned out that if a zone of influence had more than three different agricultural outputs and more than three inputs, analytical operations became intricate and time-consuming. To resolve these difficulties, a computer model was developed: the Rural Roads Computer Program (RRCP). 13. The Third Highway Project was the pilot application of the RRCP. The program was used by the Tunisian authorities to appraise 1073 km of rural roads (and their corresponding zones of influence) but was found to be complicated. 11 The need to cut down on data collection and processing, plus the need to do appraisal work for the Fifth Project, led the Tunisian authorities and their consultants to develop the Modele Agricole d'Optimisation des Niveaux d'Amenaiement (MACON). This was applied on 2700 km, out of which 1200 km were selected for inclusion under the Fifth Project and continuous refinement of this planning tool over the past five years has made it possible for the Government to adopt a new approach for selecting and improving rural roads. /12 In the summer of 1987, the Tunisian authorities are finalizing an Impact Study on the rural roads program supported by the Third and Fifth Projects /13 and, in the audit's opinion, Tunisia is close to having a methodological framework suitable both for appraisal and for ex-post evaluation of rural roads: the definitive version of MACON /14 represents a 6 mechanism that, at the appraisal stage, can reasonably forecast the socio- economic effects of various interventions (including road Improvements) and, after a project has been operational for some years, it can help pinpoiat which interventions had the greatest significance. 14. The Impact Study will discuss in detail the results of the Third and Fifth Highway Projects and its conclusions will be elaborated upon in the forthcoming PCR of the Fifth Project. Meanwhile, a preliminary review of findings related to work done under the Third Project can be sunarized as follows: (a) Road improvements and "parallel" investments produce two types of effects each: a principal, direct effect which manifests itself fairly rapidly; and a secondary, indirect effect which appears with some delay and only if favorable conditions are present. (b) The direct effect of road improvements consists of higher mobility of the population in the zones of influence and of improvements in the quality of life. This effect is demonstrated by the increase of traffic which is largely explained by the improved condition of the road. /15 (c) The direct effect of the "parallel" investments consists of a diversification of crops planted and of an increase in yields. This effect is demonstrated by increases in the volume and value of production which is largely explained by "parallel" investments.116 (d) Appraisal employed complicated models where road improvements were accompanied by "parallel" interventions in the agricultural sector but in a manner which made it difficult to distinguish benefits clearly attributable to one or the other type of investments. In addition, the assumed benefits were somewhat inflated because of optimistic hypotheses imbedded in the model design. At the implementation phase, "parallel" investments hypothesized in the feasibility studies proved to be unrealistic since less than 5% of the planned interventions were started. Upon project completion, economic benefits directly attributable to road improvements turned out to be considerably lower than the appraisal estimates. (e) Given the above, it seems preferable to separate the "road" from the "agricultural" components, and to attribute traffic increases to road improvements and production increases to agricultural investments. (M) Separating the impact of the two interventions will make it hard to justify road improvements on "economic" grounds because of low traffic volumes. However, combining other investments with road Improvements so as to increase the economic rate of return of the latter has not been vindicated by results obtained in Tunisia. 117 7 (g) Rural road improvements will be better justified if one succeeds in quantifying, and in introducing in the economic calculus, the social benefits which constitute the principal, and extremely important, gains procured by this type of investment. This may be a legitimate area for future research. IV. CONCLUSIONS 15. With rural development as the end objective, Government and the Bank worked together since 1976 to devise a rational formula for road selection and evaluation. Collaboration was good in the past and ought to be equally successful in the proposed Seventh Highway Project whose preparatory phase might consider two topics. First, that the planning and monitoring methodology embodied in the MAGON program needs to be disaggregated so as to make elements of the program usable at the Governorate level. This will require funds for the procurement of some personal computers, for program modifications on suitable software, and for some training. Once disaggregation has taken place, it will be easier for regional authorities to make investment decisions on technical and economic grounds thanks to the information which would be placed at the disposal of the Governor of the Region, of the Regional Director of Agriculture, and of the Regional Director of the Ministry of Equipment. The advantages for regional development and local government strengthening cannot be overemphasized. 16. Second, that the rural development planning and monitoring experience available in Tunisia deserves to be disseminatad. Ten years of systematic work have yielded lessons which other countries should not spend ten years of their own to arrive at. The combination of road improvement with "parallel" agricultural investments generates problems which most developing countries have to confront. In the preparation of the proposed Seventh Highway Project, some thought could be given to organizing seminars and workshops where officials from developing countries could be invited to discuss rural development, and rural road planning, issues with their Tunisian counterparts. 17. The Third and Fifth Highway Projects confronted seriously the question of how benefits from an Improved transport infrastructure can be internalized within the agricultural and rural development sectors. The Government and the Bank have amassed considerable experience which deserves to be properly disseminated. The audit feels that upon the completion of the PCR for the Fifth Highway Project, OED should carry out a study of the overall impact of transport Improvements upon the rural environment of Tunisia. 8 NOTES /1 The following documents have been consulted in the preparation of this Project Performance Audit Memorandumt Memorandum on the Economic Position and Prospects of Tunisli (September 1976); Tunisia: Economic Position and Prospects of Tunisia: Review of the Fifth Development Plan, 1977-1981 (May 1977); Tunisia: First Highway Project (Loan 746-TUN) (PPAR, December 1979); Tunisia: Secon4d Higkway Ppolect (Loan 1188-TI) (PCR, May 19085); Tunisia: Third Highway Project (President's Report and Staff Appraisal Report, June 1978); Tunisia: Fifth Highway (Rutal Roads) Project (President's Report and Staff Appraisal Report, February 1982); Republic of Tunisia: Transport Sector Memorandum (July 198). /2 Between 1962 and 1082, the Bank committed to Tunisia 48 loans (for a total of US8988.7 million) and 11 IDA credits (US870.1 million). /1 The Firet Railway Project (Loan US/Credit 1-TUN of 1964) focussed on track renewal and purchase of equipment. The First (Loan 880-TUN of 1964), Second (Loan 578-TUN of 1969) and Third Ports Project (Loan 1797-TUN) were for the rehabilitation and improvement of port facilities. The Gas Pipeline Project (Loan 724-TUN) was approved in 1971. /4 When the First Highway Project was appraised in November 1979, Tunisia's highway network consisted of some 16,6M ko, with about half paved. Roads had originally been built to relatively low standards since traffic volumes were light and since traffic growth in the 1969s had been moderate, no significant betterment works had been required. In the late 198s traffic growth picked up and dictated the construction of new roads and major Improvements of the existing facilities to accommodate agricultural and tourist demand. Also, there was an urgent need for periodic road maintenance which had been neglected because of shortages of maintenance funds. The project was completed in September 1977, with a delay of almost two years. In spite of much higher costs, the reestmated economic return was higher than that estimated at appraisal (281 vs. 16), due to higher than anticipated traffic growth. Despite of a reduction of the project scope by 25%, there was an expenditure overrun of 651 which led to a doubling of the Government0* contribution. The technical assistance for Institutional improvements has been successful where it did not touch upon Government policy. The recommended restructuring of the road maintenance organization was carried out and maintenance operations became more efficient. The project contributed to the improved capability of the Highway Department to carry out construction supervision and local consultants gained experience in feasibility studies and design work. Efforts to create a more effective administrative machinery for intermodel coordination and to reduce excessive regulation were not fully successful. The study on road user taxation was carried out but implementation of the recommendations was slow. The project provided further illustration of the difficulties which are frequently experienced in implementing aspects which are perceived by a population segment as having adverse effects. In this case, the difficulties related to right-of-way acquisition, road transport regulation, and road user taxation. 9 The Second Highway Project was appralsed In April 1971 and consisted of two main elements: (a) Highway improvements totalling 225 km and (b) Consulting services to assist in transportation planning by updating the 1968 transport survey; and to foster rural development through the preparation of a roads Improvement program. The project was completed with a five-year delay (1984 instead of 1979), due primarily to financial allocations which gave preferential treatment to the completion of the expressway and Bizerte bridge components. The project scope was changed, and there was underrun in terms of dollars and overrun in terms of Tunisian diners, both due to the appreciation of the US dollar. Project objectives were achieved: 264 km of highways were rehabilitated (instead of the 225 km foreseen at appraisal). Traffic increases were substantially higher than anticipated and this led to a reestimated economic return of 885. Substantial progress in Institutional and operational development was achieved: the studies carried out under the project contributed to a better understanding of the need for transportation planning and coordination. The rural roads program was prepared according to schedule. Most of the problems encountered during implementation were related to shortage of counterpart funds and to acquisition of right-of-way. The Fourth Highway Project (Loan 1841-TUN of May 1980 for USS88.5 Million) was to help the Government In improving highway planning, maintenance, rehabilitation and safety. Within this project, the Government established a planning unit in the Ministry of Transport and Communications to provide a close investment analysis and continuous monitoring of developments In the transport sector. /6 The Fifth Highway Project of 1982 Is a follow-up to the Third Highway (Rural Roads) Project of 1978 and is described as follows in the PR of February 1982: It continues the same general approach developed for that project, combining road Improvements and complementary agricultural Investments in integrated programs for specific rural roads and the agricultural areas they serve. Modifications have been made based on lessons learned during Implementation of tNe ongoing Third Highway Project. It was identified in July 1977 and has been prepared by the Highway Department and the Ministry of Agriculture with the assistance of consultants and Bank missions. The Fifth Highway Project has the following objectives: (a) improve selected roads to a minimum all weather* standard; (b) Increase agricultural production within the zones of influence of the project roads; (c) help improve standards of living In rural areas by providing better access to social and administrative services; (d) Improve the capability of the road maintenance organization so that project roads continue to provide the desired level of service; (e) enhance the benefits of the project roads by Improving connecting feeder roads in the zones of influence; and (f) strengthen the institutional framework by continuing to assist project preparation, implementation and monitoring, together with some applied research. Project components included: (a) a three-year program (1988- 8S) of rural road improvements; (b) a program of complementary Investments, extending over the same period for agricultural extension and over five years for credit services (1988-87); (c) a program of improved rural road maintenance; (d) spot improvements to feeder roads; (e) equipment for four vehicle inspection stations in provincial centers and one In Tunis; and (f) technical assistance for project preparation, supervision and monitoring; agricultural extension training; research on materials, construction techniques and maintenance techniques for unpaved roads; and a study of the socio- economic, agricultural and transport impact of rural roads. 1 /7 Thisection of the PPAM has made use of the following documents: The Economic Analysis of Rural Roads (World Bank Staff Working Paper No. 241, 1976); Identification and Appraisal of Rural Roads Projects (World Bank Staff Working Paper No. 862, 1979); Projet do Pstes Agricoles (Identification et Factibilites) MACON (Republique Tunislenne, Ministere, do I'Equipment, 1908); Approche Economiue Relative a I'Amenagement des Routes Rurales: Le Modele Macon (A. Chaar, Republique Tunisienn*, Ministere do I'Equipement et do I'llobitat, 1984); and Etude d'Imoact du Programme do Plstes Agricoles: Rapport do Synthese (Draft, STUDI-8CEOM for Le Minister do I'Equippement et do l'Habitat, Republique Tunisienne, May 1987). /8 cf. The Economic Analysie of Rural Road Proects (Staff Working Paper No. 241, August 1976, pp. 2-18). /g The 1979 Bank Staff Working Paper No.862 (Identification and Appraisal or Rural Roads Pro]ects) is the product of this collaboration and provided the theoretical framework upon which the Third Highway Project was based. /10 For each individual road, the Investment Package might consist of rural road Improvements; rural road esint*nance equipment; on-farm investments by farmers in the road's area of Influence; provision of working capital to farmers; and extension services. An analysis of the potential agricultural development In each &one of influence would determine precisely which components of the package would be provided for each zone. Zones of Influence were determined by rural road network surrounding the road under consideration; the distances between farms and local markets; the nature of the terrain; and the means of transport used. Benefits from agricultural investments would accrue primarily to farmers, and benefits from road Improvements te farmers and/or to transporters. The rate of economic return and the not oreseai value of a given investment package would be computed by comparing total benefit (including transport savings accruing to non-agricultural traffic) with total costs of road construction and maintenance and complementary agricultural investments. /c of. A. Cheamari, in his Approche Economique Relative a l'Amenagement des Routes Rureles (Minister, do I'Equipement et do I'abitat, November 1984), compares the RRCP and MAON and notes that for results of comparable rel*ability, the RRCP required 9 instructions (vs. 169 for MAGON) and 418 date cards (vs. 26 for MAU0N). He also notes (p.22, OHypothese de 8 passages par plate) that total cost (including data processing, programming cosat and cost of computer time) amounted to 18 Diners per road for the RRCP and to 1 Diner for Megon. /12 As with the RRCP, the MACON model rolles upon an analysis of variations of producers' surplus within the area of influence following the Improvement of a given road. Improved roads are regarded to be desirable because on the one hand they support efforts for the amelioration of the quality of life and, on the other, through transport cost reductions, they encourage farmers to improve their production methods and increase their output. Benefits from road Improvements are broadly grouped under three categories: the surplus of agricultural producers (after taking Into account the Oconflecations of part of the benefits by transporters and off-farm consumers); the 11 surplus of transporters and consumers associated with the marketing of agricultural products; and the surplus of non-farm related road users. Costs include road construction plus complementary agricultural investments. The "with projec*t' hypothesis assumes the simultaneous Implementation of these two types of investments. Economic calculation is based on the following stages: estimates of exportable surplus by zone of influence (taking into account production and internal consumption by zone In the 'with2 and *without project* hypotheses); estimates of traffic volumes corresponding to average load factors (with additions to these volumes of non-agricultural traffic, i.e., passengers and freight); evaluation of the 'benefits' attributable to road improvements, I.e., benefits for producers and transporters; and, finally, calculation of rate of return and estimation of best year for opening the road to traffic. Data for the above operations are derived from a variety of special studies and surveys. Road Studies are intended to compare appropriate levels of road improvement on the basis of geological, hydraulic and geotechnical Investigations and on the basis of more or less detailed feasibility of the construction and maintenance costs associated with different alternatives. Agricultural Studies encompass four major tasks: delimitation of the zone of influence of each particular road; collection of agricultural data related to the zones of Influence; targeting of agricultural output in the long and medium term; and Identification of high priority agricultural investments in the zones of Influence. Rod Traffic Studies are intended to estimate the level of actual traffic and to make forecasts after road improvements have been made. Origin-deotination surveys are used to estimate diverted traffic. Transport Studies aim to determine actual conditions on the rural roads by investigating vehicle operation costs, by making comparisons between transport costs and officially approved tariffs and rates, and by calculating the costs of Interruption of road service due to poor quality of the existing roads. Economic Studies compare costs and benefits in the customary fashion. / The study has focussed on a sample of 24 roads (with a combined total of 855.7 ka) distributed throughout the arable surface of the country and selected to be representative of all rural road improvements (1,689 for the Third and 2,76 for the Fifth, or about 8,686 km In total) and of all ecological zones in Tunisia. The study began by describing the elements In Its methodological approach (progression from Individual roads in the sample to road network; definition of the Owlth and 6without projectO situationel. It then discussed the parameters and variables used in the analysis (Motorized traffic; Total agricultural production; Quality of life Indicators; Road improvement level; Investments in Irrigation; Planting of Fruit trees; Modernization of growing methods; Population variation In the zones of influence). It continues by describing the two main methods It has used for date elaboration (Calculation of averages and Correlation analysis). Results are presented in averaged and matrix formats and the Interpretation of results is given under two main headings: Impact of the Road Component (Effect on traffic; Effect on production; Effect on population and quality of life) and Impact of gComplementarys actions (Increase of irrigated surface; Increase of cultivated surface; Improvement of technical level). The ex post evaluation of road Improvements is thon carried out by calculating separately the 'social* returns and the economic returns of each road in the sample. /14 Pages 2.8-2.18 of the Prolet do Pstes Agricoles (Identification *t Factibilites)/ MACON (Dossier de l'Utilisatour) (BCEOM- SCET International for the Ministere do I'Equipement, Republique Tunisionne, October 1988) contain the detailed description of the programmatic logic and Annex 2 contains the forms for data inputting. 12 /3L cf. Etud dImpact du Proorla de Pltes A2ricole-- Rapport de Synthese (May 1987) pp. 07-71. /IC cf. Etude d'Impct du Progroame d* Plstes Aaroles-- Rapoert de Synthese (May 1987) pp. 71-77. / of. Etude d'Imact du Progra&~e de Pistes Agr1coles- Rapport de Synthese (May 1987) pp. 87-91. - 13 - 8690349 December 17, 1987 French Tunisia 175/10 DBB:mec Republic of Tunisia Ministry of Equipment and Housing Directorate General of Bridges and Highways Directorate of Studies Tunis, November 28, 1987 Mr. Graham Donaldson Chief, Agriculture, Infrastructure and Human Resources Division, OED World Bank Washington, D.C. Subject: Third Highways Project (Loan 1601 TUN) Project performance audit report We are in receipt of the first draft of the project performance audit report for the Third Highways Project covered by Loan 1601 TUN. This report gives rise to the following comments relating to the recommendations it contains (page 2 of summary of performance audit). While recognizing the difficulties involved in quantifying linkages, we would suggest that the "highways" and "agriculture" components not be treated separately, and that research be directed along these lines. Is/ A. Chamari Head of Engineering, Director of Research - 14 - Republic of Tunisia Ministry of Agricultural Production and Food Ref. No. 192/DGPV/G The Director General of Plant Production December 4, 1987 Mr. Graham Donaldson, Chief, Agriculture, Infrastructure and Human Resources Division, OED World Bank Washington, D.C. Subject: Comments on project performance audit report Please find attached comments relating to the PPAR for the third highways project. The feeder roads pilot project made it possible to implement a new approach to rural development that combines the establishment of a road infrastructure and the implementation of parallel operations aimed at intensifying agricultural production, and to this extent the project has been positive. Thus, although it proved impossible via an evaluation of the project after completion to quantify the extent of the linkages between road improvements and parallel investments, there can be no question but that such a linkage does exist and that the absence of parallel investment reduces the benefits from road improvements and other rural infrastructure projects. This can be seen from the conclusion to the impact study, according to which the economic benefits resulting from road improvements were less than the awounts estimated, which could be in part explained by the fact that less than 5% of the parallel investments were actually made. From this perspective, it would be helpful if the report were to mention the reasons for the disappointing results as regards the implementaticd of parallel investments (part D of the project). - 15 - There are in fact a number of problems inhibiting investment, caused mainly by the farmers' inability to pay. It would therefore be helpful if this report were to analyze the constraints on farmers' liquidity and, in this connection, to study the effect of the size of farms, the parceling of landholdings, the average age of farmers and their educational level; it would also be beneficial if these factors were included in the IBRD approach to agricultural development projects. Similarly, the strict requirements of banking institutions for awarding credits and IBRD's policy of eliminating subsidies are poorly adapted to the situation of Tunisian farmers, who still need substantial support. Hoping that these few remarks find in you a sympathetic audience and that you will include them in your comments concerning the project, I remain, Yours sincerely, /s/ Malek Bensalah Director of Production - 17 - TUNISIA PROJECT COMPLETION REPORT THIRD HIGHWAY PROJECT (LOAN 1601-TUN) (RURAL ROADS PROJECT) March 11, 1987 Regional Projects Department Europe, Middle East and North Africa Region - 19 - I. INTRODUCTION 1.01 Tunisia's road network consists of approximately 9,000 km of classified paved roads, 8,500 km of classified unpaved roads, and 30,000 km of unclassified roads. The road network carriss virtually 90% of Tuniria's passenger traffic and 851 of its freight traffic, which increased at an average annual rate of 15% during the 1970s. Remaining traffic is carried by the rail network, which consists of 1,900 km of Lrack, standard gauge in the north of the country and metric in other regions. 1.02 At the outset of independence in 1956, Tunisia's colonial heritage included a road system which, although fairly well-developed, was of passable geometric design standards, since intended only for light and infrequent traffic. Until 1970, no major system improvements were made; attention was restricted on the whole to maintenance and repair of existing paved roads, while a few rural roads were regraded on a biannual basis. 1.03 It was not until after 1970 that any sizeable projects for the improvement of both inter-city and rural roads were undertaken, with World Bank assistance. The First Highway Project (Loan 746-TUN), begun in 1971 and completed in 1978, mainly involved the rehabilitation of 50 engineering structures and nearly 1,000 km of paved road sections, while a second (Loan 1188-TUN), begun in 1976 and completed in 1984, was designed to improve and modernize 264 km of paved roads. The Third Highway Project (Loan 1601-TUN), begun in 1978 and completed in 1986, and the subject of this Completion Report, was the first to focus on rural roads to the exclusion of bitumen-surfaced roads. It called for improvements to about 1,000 km of rural roads, measures to strengthen agricultural extension services, and related agricultural investments. The Fourth Highway Project (Loan 1841-TUN), begun in 1980, is part of a network-wide rehabilitation maintenance program developed in the course of a road maintenance study carried out as part of Loan 52-TUN. The Fifth Highway Project (Loan 2108-TUN), and the second to concentrate on rural roads, was launched in 1980 and is still in the process of execution. Similar in content to the Third Project, it will affect approximately 1,200 km of rural roads. 1.04 A sixth project, for the improvement of the maintenance management system and the modernization of certain primary- and secondary-network roads, has been appraised. Finally, a "seventh" project, and the third to focus exclusively on the rural network, calls for improvements to 1,500 km of roads in this category in all parts of the national territory. It is being prepared by the Government. 1.05 Paralleling these large-scale projects there have been many smaller-scale regional and local ventures to improve or create rural roads, launched against the backdrop of the country's vast rural - and later integrated rural - development movement. In concrete terms, between 1973 and 1985, some 6,150 km of rural roads were improved or built under various Rural Development Programs, at a total cost of TD 56 million. - 20 - 1.06 It is clear, then, that the Government of Tunisia, well aware of the fundamental importance of major and minor roads in rural development and regional equilibrium, has over the last 15 years made a very considerable investment in such infrastructure, an effort supported by crucial financial assistance from the World Bank. II. PROJECT PREPARATION AND APPRAISAL A. Preparation 2.01 This Third Highway Project was prepared on the basis of feasibility studies carried out by a Franco-Tunisian consulting group in connection with the Second Highway (Fifth Hiabway) Project and intended for what was referred to as a Pilot Project, since it was to be followed by a full-scale rural roads project. However, due to tae innovative approach to rural development through emphasis on rural rovd improvements, the discussions on the project concept and scope took several months, especially within the Bank. The identification mission took place in November 1975 but it was not until May 1976 that the following understanding was reached within the Bank and conveyed to the Government: an integrated approach is warranted because rural road improvements and rural development are highly itterdependent, and although the needed complementary agricultural investments are small compared to rural road investments, successful implementation of the multi-sectoral Pilot Project would require resolution of important policy issues, particularly concerning agricultural credit and land tenure. Three preparation and preappraisal missions were sent in the field during June 1976 and August 1977 to launch and supervise the feasibility studies and to refine the methodology for appraisal. 2.02 The tentative list of project components included: improvements to approximately 1,070 km of roads in eight governorates; measures to upgrade road maintenance administration; measures to strengthen agricultural extension services; the provision of credit to finance complementary investments within the zones of influence of the project roads; and preparation of a Second Rural Roads Project (or Fifth I-ghway Project). B. Appraisal 2.03 This project was appraised in October 1977, the basis for the process being (i) a detailed analysis of about 400 km of rural roads and complementary agricultural investments for which the feasibility studies had been completed; (ii) agreement on the methodology to be used for sub-projects approval and (iii) determination of the Government's capabilities to undertake the preparation of the remaining 600 km of rural roads and complementary agricultural investments. 2.04 The agreed project components were: - 21 - (a) Development of approximately 1,070 km of rural roads (rehabilitation, improvement, or construction) in eight governorates: - Le Kef 106.8 km - Siliana 65.7 km - Naboul 93.7 km - Kairouan (Mahdia) 193.7 km - Kasserine 170.5 km - Sidi Bouzid 267.0 km - Gafsa 65.7 km - Kbili 105.6 km TOTAL 1,068.7 km (b) Procurement of road maintenance equipment, spare parts, and workshop machine tools. (c) Strengthening of agricultural extension services in the zones of influence of the project roads, through: (i) construction of regional extension centers; (ii) procurement of equipment; and (iii) training services, technical assistance, and retraining of extension agents. (4) Provision by Banque Nationale de Tunisie (BNT) of three types of credit: Di: short-term, for purchases of inputs by farmers in the zones of.influence; Da: medium- and long-term, to fund farmers' investment projects in the zones of influence; and Ds: medium- and long-term, to fund agro-industrial investments by state and parastatal agencies (STIL, Office des C6r6ales, SONM) and service cooperatives. (e) Identification and feasibility studies of approximately 2,500 km of rural roads - in other words, preparation of the Fifth Highway Project. 2.05 The project cost estimate and loan allocation were as follows: Loan Cost Estimate Foreign Allocation Item TD USU Cost Component US$ Rural Road Improvement 16.4 37.7 40% 15.1 Road Maintenance Equipment 1.9 4.4 801 3.5 Extension Services 1.0 2.4 55% 1.3 Short-term Credit 2.4 5.5 281-' - Investment Credit 5.7 13.1 40% (3.3)& Agric. Facilities & Equip. 3.0 6.8 531 4.8 Technical Assistance 0.5 1.1 75% 0.8 Contingencies 9.8 22.6 441 5.5 ' Total 40.7 93.6 44% 32.0 1/ Financed by Banque Nationale de Tunisie (BNT) from its own resources. 2/ Financed under Second Agricultural Credit Project. 3/ Including 15% for quantities and 20% for price escalation. - 22 - 2.06 Negotiations took place in Washington from May 12 through May 19, 1978. During negotiations the Government agreed to the following: (i) procedures established for the preparation and Bank approval of sub-projects; (ii) formal establishment of a Coordinating Comittee to supervise the project; and (iii) signature of an agreement between the Government and BNT. Points (ii) and (iii) were made conditions of effectiveness. 2.07 On June 27, 1978, the Board of Executive Directors of the Bank approved a loan of US$32 million at an interest rate of 7.5%, repayable over a period of 17 years with four years of grace. The Loan Agreement was signed on July 24, 1978, came into effect on April 30, 1979 and was to be completed on June 30, 1984. 2.08 The Highways Department, Ministry of Public Works, was declared the executing agency for project civil engineering works (Part A), procurement of maintenance equipment (Part B), and conduct of highway studies (Part E). The Agricultural Production Department, Ministry of Agriculture, was named the executing agency for actions designed to promote agricultural extension (Part C). Finally, BNT was to be responsible for implementation of Part D, which called for short-, medium-, and long-term credits to be made available to individual farmers and certain organizations. III. PROJECT IMPLEMENTATION A. Effectiveness. Start-up and Amendments 3.01 The loan became effective on April 30, 1979, about five months after the November 22, 1978 deadline in the loan documents. This deadline had been prorogated twice, due to the delay of the official ratification of the Loan Agreement. 3.02 The project had a good start, with the first tranche of rural road improvements launched in early 1979 as well as preparation of the subsequent tranches . The agricultural components began only in late 1979, but did accelerate as far as extension services were concerned. Problems related to scarcity of funds accorded to the Ministries of Public Works and Agriculture began to appear in the mid-course of the project and contributed, with other component-specific factors, to the overall delay in project completion. 3.03 Three amendments were made to the Loan Agreement. In May 1984, the Government requested and the Bank agreed to a reallocation of loan funds to cover the purchase of complementary rural roads maintenance equipment (para. 3.16) and to allow construction and equipping of a cold store for seed potatoes directly by the Government rather than under the credit program (para. 3.42). At the same occasion the closing date was extended to June 30, 1985. A second amendment was signed in June 1985 which provided for an additional 9 months (to March 31, 1986) which was finally extended to December 31, 1986 (paras. 3.08 and 3.17). - 23 - B. Execution of Part A: Civil Engineering Works Associated with Road Improvement 3.04 The rehabilitation and improvement of rlads targeted through the project took place, for the most part, during the period 1979-1983, and, again for the most part, went ahead normally. The implementation process was organized by governorate and by lot, each lot consisting of several roads, and each governorate generally requiring two lots or, on occasion, one or three. The major items of information describing the works themselves and related financial particulars are brought together in the accompanying Table 1, by governorate and by lot. The last line of the table gives totals or averages for all the project roads as a group. 3.05 Generally speaking, the work required consisted of: spot improvements and corrections of alignment; corrections of longitudinal section; improvement of lateral and transverse drainage; laying of sub-bases and bases; building of engineering structures. The information given in the table facilitates comparison of initial plans with actual work, from three standpoints: (1) choice of sub-projects, (2) implementation schedule, and (3) cost. (1) Total Road Length and Changes in Choice of Roads 3.06 As the last line of Table 1 indicates, the total length of roads scheduled for improvement through the project was stated in the Loan Agreement as 1,068.7 km, in the eight governorates of Le Kef, Siliana, Kairouan, Nabeul, Kasserine, Sidi Bouzid, Gafsa, and Kbili. However, the actual total improved in those eight regions was only 873.6 km, or 195.1 km (18.3%) less than anticipated. This appreciable difference between anticipated and actual figures is explained by the fact that a number of roads were rejected on grounds of rate of return, computed only after signature of the Loan Agreement and clearly inadequate. The rural roads withdrawn from the project on these grounds are grouped in Table I as national lots 5X, 6X, and 7X. In all, 196.2 km of roads were affected, practically the same figure as the difference of 195.1 km referred to above (the missing 1.1 km reflecting uncompensated minor adjustments between total length of road originally scheduled for improvement and actually improved). 3.07 Since the Loan Agreement referred explicitly to about 1,000 km of rural roads to be improved, the shortfall of nearly 200 km needed to be made up in one way or another. The solution, adopted with Bank consent, consisted in replacing the unprofitable roads with certain roads studied in connection with the Fifth Highway Project (Part E of this project); located in the governorates of Zaghouan, B6jA, and Jendouba, their total length was 215.9 km. Although at the time these replacement roads were chosen their internal rates of return were unknown, the studies not being completed at that point, the available information guaranteed figures high enough to justify the advance choice. This group of roads was handled under what later came to be known as the "complementary program." 3.08 In the end, a total of 1,089.5 km of rural roads was improved under the Third Highway Project, as against the 1,068.7 km initially anticipated, so that finally there was an increase of 20.8 km or 1.9%. - 24 - (2) Implementation Schedules and Delays 3.09 According to the original schedule, the Third Highway Project was to be completed by December 31, 1983, with the loan closing date set for June 30, 1984. The latter has finally been extended to December 31, 1986 and by that date the provisional acceptance of certain lots had still not taken place. This would mean a total project time overrun of at least two and a half years, or 42Z of the six years originally projected as the execution period. 3.10 The delay with execution of Part A which can be attributed to contractors is detailed lot by lot in Table 1, where comparisons are made between contractual and actual implementation periods, the latter taken as the time elapsed between the date of the service order for commencement of works and the date of provisional acceptance. It will be noted that total delay varied between one governorate and another as well as from lot to lot. Some lots were actually ready prior to deadline, as in Sidi Bouzid and in Kairouan. 3.11 An overall picture of the delay ascribable to contractors may be obtained from the last line of Table 1, which gives the average weighted estimated and actual execution periods, the weighting factor being the total length of roads in each lot. The average delay is seen to be two months on an average estimated execution period of about 14 months - in other words, 15%, an acceptable figure and one which reveals that globally execution of Part A of the project was actually not very far at all behind schedule. The few cases of serious delay can be traced to some contractors' financial problems or labor difficulties which sometimes brought work on the particular lot to a complete standstill for many months. It is these few serious cases that are the cause of the overall time overrun of 421 mentioned in para. 3.09. (3) Imlementation Costs and Cost Overruns 3.12 Two kinds of cost overruns can be distinguished in Part A: (i) the difference between the actual cost and the cost estimated at appraisal; and (ii) the difference between the actual cost of each lot and its cost as agreed in the contract entered into with the contractor. On a lot by lot basis, Table I gives the following figures: the cost of the works as estimated originally; any additional amounts approved on change orders, their total giving the size of the cost overrun; and the final, actual cost of the works (initial estimates plus change orders). The final column shows any overrun in percentage terms. 3.13 Overruns varied considerably from one lot to another, being anywhere from 01 on lot 3A, to 5% on lot 12B and 71 on lot 11A, to 76% on lot 63, 88% on lot 4C, and 91% on lot 3B. Their major causes were price escalation and additions to the works scheduled, the latter including paving of certain roads not contemplated initially. Such paving was allowed for in the description of the project, insofar as its economic justification was provided: this was the case for certain roads in Lots 13, 3B, 4B, 4C and 63, where traffic increase was substantial. - 25 - 3.14 Total overrun is shown by the following aggregate figures, given on the last line of Table 1: - total cost of all contracts with successful bidders: TD 17,974,000 - total amount of change orders: TD 6,255,900 - total actual cost of works executed: TD 24,229,900 There was therefore a total project cost overrun of 35%, a figure close to the appraisal estimate of likely physical contingencies and price escalation (34%). At that stage, in 1978, Part A cost was estimated at TD 16.4 million, without allowances for physical contingencies and price escalation, and TD 22 million, including 34% allowance. Comparison of these two figures to total actual base cost (TD 18 million) and total cost of project works (TD 24.2 million), respectively, makes it clear that overruns were no more than moderate: TD 1.6 million, or 10%, in the first instance, and TD 2.2 million, or 10% again, in the second. 3.15 To conclude, it may be stated that the assessment of Part A civil engineering costs made at appraisal was accurate, and that the implementation process was normal from the financial viewpoint. C. Implementation of Part B: Procurement of Road Maintenance Equipment 3.16 The road maintenance equipment was procured by the Highways Department in two phases, details of which can be seen in Table 2. The first phase dealt with the equipment originally intended (at appraisal) for the needs of 8 maintenance brigades. The second phase dealt with additional needs that were identified during the implementation of the Fourth Highway Project and the preparation of the Fifth Highway Project. At negotiations for the latter (January 1982), it had been agreed that funds available under the Third Highway Project would be used to procure this additional equipment. A provision to that purpose was included in the May 1984 amendment to the Loan Agreement. Contract Deadlines 3.17 The main contracts of the first phase were awarded in 1980, roughly within the time-frame foreseen at appraisal. The picture is quite different with the second phase. Most contracts were awarded only toward mid-1984, or more than two years after the agreement reached during the negotiations of the Fifth Highway Project. Such a time lapse is ascribable to the relatively complex nature of the equipment required and the large number of suppliers involved, two factors always likely to make procurement proceedings lengthier than usual (para. 3.49), but also, as far as the second tranche is concerned, to the uncertainties over the budgetary allocation which had held up any action for several periods of time. 3.18 In addition, the arrangements to procure certain equipment had to be conducted all over again from the beginning, creating a further delay of at least two years. The contracts on the 26 four-wheel drive vehicles, the - 26 - 70 commercial vehicles, the 12 tracked bulldozers, and the five tire-mounted bulldozers could not be for various internal reasons (suspension of the authorization to purchase light vehicles (para. 7.02), lengthy processing by the Main Tender Board), awarded until 1986, and this equipment is currently either yet to be delivered or has just been received. This is the main reason why the loan closing date was postponed until December 31, 1986. Quantities and Costs 3.19 There were no significant changes made in the numbers of units of equipment identified as necessary either at appraisal for the first phase or during the preparation of the Fifth Highway Project for the second phase. Once amended, the procurement program was fully completed, except in the case of lot 6, where only 63.6% ef the equipment grouped in it was supplied, as is apparent from Table 2. For the second tranche as a whole, only 1.8% of all equipment ordered was not obtained, either because of an appreciable increase in the base price, or lack of adequate budget allocation, or non-availability of the model required. 3.20 The total cost of equipment procured was TD 10,336,250 million, not counting the 10% increase which the recent (September 1986) devaluation of the dinar will mean in the cost of the last tranche of equipment yet to be delivered. The allocation made in the Loan Agreement to Part B of the project was TD 2.3 million, including provision for physical contingencies and price escalation. The cost overrun here is considerable, but on the other side it has.helped to contain the total cost of the Fifth Highway Project within reasonable limits. D. Implementation of Part C: Measures to Strengthen Agricultural Extension 3.21 Under the terms of the Loan Agreement, the program to strengthen agricultural extension services in the zones of influence of the project roads called for various types of action: (i) construction of regional extension centers; (ii) purchase of equipment for the extension services; (iii) recruitment of extension agents specializing in communications in the rural environment; (iv) technical assistance with the introduction of new crop-growing methods; and (v) organization of short-term seminars and internships abroad to provide refresher training for extension agents. 3.22 The Loan Agreement allocated TD 1.2 million (including contingency allowance) to this program. Shortly after the project started up, the Government decided as a national policy objective for the Fifth Plan to increase the extension services to farmers by placing an extension office in each imadat (smallest administrative entity). For the project this would have resulted in a doubling of the TD 1.2 million allowed for in the Agreement. However, the initial allocation to Part C was not eventually revised upwards due to financial budgetary constraints that appeared in the course of the project. 3.23 The new Part C as defined in the May 1984 Amendment had two components: (i) Part Ct, more or less the same as the original Part C; - 27 - and (ii) Part Ca, which covered the construction and equipping of a cold storage depot for seed potatoes at Soliman. On the financial side, Part C cost as estimated in the 1984 Amendment was as follows: (i) Part C, civil works and equipment: US$3.17 million, of which IBRD financed US$1.3 million (41%); (ii) Part C, training and technical assistance: US$0.13 million, of which IBRD financed US$0.1 million (751); and (iii) Part Ca civil works and equipment: US$1.45 million, of which IBRD financed US$0.8 million (55%). 3.24 The necessary time-phasing of the construction of extension centers with that related roads was ensured through a three-phase implementation sch"dule: the first phase dealt with those investments in rural roads and extension centers for which studies were completed at the project beginning; the second phase dealt with investments identified during appraisal; the third phase was the complementary program (para. 3.07). 3.25 In describing the achievements of this project, it is convenient to do so by type of action: (1) Construction of Extension Centers 3.26 Construction and equipping of the extension centers provided for in the project went ahead in three phases. Table 3 provides details, organized by phase and by governorate, of the number of existing centers and those either leased or built under this Third Highway Project; estimated and actual construction costs are also given. (a) The first phase, launched in 1980, involved rural roads in the governorates of Kairouan-Mahdia, Nabeul, Siliana, and Le Kef. The total number of extension centers involved was 43, 11 of them already in existence, 21 newly leased, and 11 actually built. The extension centers (known in French as cellules techniques de vulgarisation, or CTVs) built through the project were generally located on public land or sites which the particular Regional Agricultural Development Commissions (CRDAs) had taken over with the consent of the regional authorities and made available at no cost. Construction costs for.a single extension center were estimated at TD 13,000 at the time and therefore at TD 143,000 for the 11 that were to be built. Actual construction went ahead uneventfully and was completed on schedule. The final cost figure was TD 138,317, or 3.3% less than estimated. (b) The second phase, launched in 1981, involved roads in the governorates of Gafsa, K6bili, Sidi Bouzid, and Kasserine. The centers in these districts numbered 30 in all, of which 10 were already in existence, 8 newly leased, and 12 actually built. The estimated cost of each center built had now risen to TD 15,000. Completion of this second group of centers was slow, particularly in the governorates of the Sidi Bousid and KMbili. In Sidi Bouzid, for instance, the CRDA began the - 28 - construction of five centers on force account. However, scant allocations to Rural Works Department and limited payment appropriations led to considerable delays and meant the program was not completed until 1985. In K4bili, the contracting firm's bankruptcy and the ensuing cancellation of its contract forced CRDA to proceed with building of the extension centers on force account. The total actual cost of the centers built as part of the second phase was TD 169,280, or 6% less than the estimated figure. (c) The so-called complementary program involved roads in the governorates of Zaghouan, B6ji and Jendouba which had not been included in the original pilot project but which replaced other roads rejected on grounds of inadequate rates of return. The number of extension centers targeted in these three governorates totaled 20, with five of them already in existence, four newly leased, and 11 actually built. Estimated construction costs per center were TD 25,000 in Zaghouan and Bkjh, and TD 22,300 in Jendouba. Although this work was scheduled for 1982 and 1983, delays - ascribable mainly to the inadequacy of payment appropriations - meant the centers were not completed until 1986. Total actual cost was TD 226,522, or 15.2% less than estimated. - (d) Total extension center construction costs, over the three phases. amounted to TD 534,119. The slow implementation pace resulttd in an average unit price of TD 15,710 per CTV, a 20% increase over the initial TD 13,000 unit cost. (2) Purchase of Equipment 3.27 Equipment purchased with a view to strengthening the agricultural extension system can be grouped in three major categories: vehicles, to allow agents the mobility essential to their work; demonstration equipment, needed to allow proper dissemination of knowledge of modern cropping patterns; and office equipment for the extension centers themselves, since agents do a large part of their work there. 3.28 The project equipment purchased is detailed in Table 4 by category and by phase (the phases being those used above in discussing the construction of extension centers). Under the complementary program, considerably less equipment was purchased than was needed. This is explained by a funding shortfall, as the budget allocations had been largely spent on constructing the extension centers, the cost of each center having virtually doubled over the space of a few years from TD 13,000 in 1980 to TD 25,000 in 1985. 3.29 Equipment purchases totaled TD 651,747, of which TD 386,408 (59.3%) was spent on vehicles, TD 225,976 (34.7%) on demonstration equipment, TD 22,174 (3.4%) on office equipment and TD 17,189 (2.6%) on operating expenses. - 29 - (3) Technical Assistance 3.30 The technical assistance program involved three types of training action: (a) Mission by a communications expert: This international consultant, Mr. Mustapha Lamine, was available for a total of 15 months, beLveen June 4, 1981 and November 17, 1982 (including breaks). His services consisted of: (i) selecting of 83 trainees from among 150 candidates nominated by the Agricultural Production Department; (ii) organizing and giving five series of five training courses each for the benefit of the 83 extension agents chosen; (iii) following those training courses and providing assistance for trainees as they were beginning to apply the new techniques learned; and (iv) assisting the Agricultural Production Department in choosing audio visual equipment for the extension centers. Following psychological aptitude tests and a series of briefings, 83 individuals (74 extension agents and nine coordinators) were chosen to receive the following training: - 4 five-day courses; - 1 three-day practice session; - 1 other practice session and individual follow-up assistance. The results of this program were regarded as satisfactory, since only five participants were finally disqualified because they were unable to master the communication techniques and methods taught. (b) Short internships in France: Three 20-day programs of this type were organized for the benefit of 31 Agricultural Production Department professional staff: the first (November 14-December 4, 1982), for 11 higher-level graduates; the second for 10 higher-level graduates; and the third for 10 intermediate-level graduates. The Institut National de Promotion Sup6rieure Agricole, Dijon, France, organized all three programs, which were given at the Centre de Formation Professionnelle pour Adultes in Montargis, Loiret Department, Central France. The goal was to give participants some training in integrated rural development. (c) Seminars: A series of five awareness and evaluation seminars, lasting two or three days each, were organized in conjunction with CNEA for the benefit of extension center managers. They were held in 1983 and 1984 in the following - 30 - order: two seminars for the centers targeted during the first phase, two for the second phase and one for the complementary program. 3.31 Training costs totaled TD 77,362, of which TD 37,740 (48.81) was spent on the programs organized by the communications expert, TD 31,672 (40.91) on the internships in France, and TD 7,950 (10.3%) on the seminars. At appraisal, the sum allocated to technical assistance under Part C was TD 57,000 without allowances and TD 72,500 with allowances for contingencies and price escalation. This latter figure, only 7% less than the actual final cost, is an indication that the initial appraisal of these needs was correct. (4) Final Cost of Part C (later C) 3.32 The final cost incurred under Part C, which became Part C, following the Amendment of May 23, 1984, and which covered the strengthening of agricultural extension services, totaled TD 1,263,228, with 42.3% going to the building of extension centers, 51.6% to equipment for the centers, and 6.1% for technical assistance. 3.33 At appraisal, Part C costs were estimated at TD 1.057 million without allowances and at TD 1.24 million with allowances for contingencies and price escalation. The difference between actual Part C (later C) costs and the appraisal estimate was only 21. Due to financial constraints appearing in the course of the project, a determined effort was made to abide by the original estimates, despita recommendations to strengthen agricultural extension services that would have involved spending about the double (see para. 3.22 above) (5) Part Ca 3.34 Once it was decided not to go ahead at all with implementation of Part D3 of the complementary investment program (see discussion on Part D below), the Agricultural Production Department, with the consent of the World Bank, decided to assume direct responsibility for one major agro-industrial investment, namely a cold storage depot for seed potatoes at Soliman. With a capacity 1,500 metric tons, this facility is designed to meet the demand associated with cultivation of early varieties. Total demand is estimated at 4,500 tons, with two thirds of it being met from a 3,000-ton depot built earlier as part of the cooperation arrangements between Tunisia and Canada. 3.35 International tenders were invited for construction of the second depot, a contract priced at TD 950,000 being awarded to a group of Canadian firms.Despite the considerable difficulties encountered in obtaining the necessary equipment import permits, and the resulting month-long standstill, the building was completed on schedule. Cost overruns amounted to TD 24,700 only, or 2.6%, giving a final cost of TD 974,700. The cold store is operated under a rental agreement by GIL, a vegetable producers and trade association. - 31 - E. Implementation of Part D: Credits for Complementary Investments 3.36 As described in the Loan Agreement, Part D of the project, concerned with the granting of credits and subsidies through Banque Nationale de Tunisie (BNT) to fund complementary investments, had three components, as follows: (1) Component D, 3.37 Here, a sum of TD 3.1 million was allocated to fund purchases of inputs by farmers (such as fertilizers, pesticides and feed concentrates) in the zones of influence of project roads during the project implementation period. The financing arrangement was to be as follows: (i) 19.4% of cost put up by sub-borrowers; and (ii) 80.6% of cost financed with short-term BNT credits utilizing BNT own resources (total of TD 2.5 million to be made available by BNT). In actual fact, BNT provided only 'very little funding of this type, but the precise amount has not been available since the BNT accounts were held by province and not by zone of influence. Also it appears that farmers were provided with credit from other sources at more advantageous terms than that of BNT (for instance from the Cereals Office). (2) Component Da 3.38 Here, a sum of TD 7.5 million was allocated to help fund the designated kinds of complementary investments by farmers (such as purchase of agricultural machinery, dairy cows and sheds as well as small scale irrigation and well construction) in the zones of influence of project roads. Financing was to come from the following quarters: (i) 21.5%, or TD 1.6 million, from sub-borrowers; (ii) 20.3%, or TD 1.5 million, from the Special Fund for Agricultural Development (FOSDA); and (iii) 58.22, or TD 4.4 million, from medium- and long-term credits funded as follows: 67%, or TD 2.9 million, by BNT; and 33%, or TD 1.5 million, by IBRD (to be financed under the Second Agricultural Credit Project). 3.39 Table 5 shows the performance picture as of December 31, 1982: (a) of the total of TD 7.5 million aliocated for funding complementary investments, farmer loan applications would have absorbed no more than TD 2.1 million, or 27.72 of the original allocation. In other words, the potential beneficiaries of the program manifested only slight interest. This can be traced to: (i) deficie-..-ies of extension staff in promoting and preparing agri.cultural credit applications (para. 3.40); and (ii) the large number of alternative sources of financing. Thus much of shallow well development in Central Tunisia was financed through remittances sent by relatives abroad. Elsewhere, small-scale irrigation was partly financed by several bilateral credit schemes and various sources of financing were available for fruit tree development. Wb Of the total of TD 2.1 million actually sought in loans by farmers, BNlT approved only 36.42, or TD 0.76 million. In - 32 - explaining this high rejection rate, the BNT representative assigned to the project indicated that most loan applications were not feesible, either technically or financially. However, the fact that BNT was to bear nearly 40% of the risk involved probably weighed heavily in the decisions to reject. (c) The loan applications actually approved were by no means all carried through to completion; only TD 0.34 million of the TD 0.76 million in applications accepted by BNT was disbursed, giving an implementation rate of 44.6%. Many causes may lie behind the shortcomings of this program, but the most frequent were the sums that borrowers themselves had to put up and the strict conditions attached to disbursement of proceeds, which was tied closely to physical progress; hence the attractiveness of alternative sources of financing. Of the funding actually used, 90.31 was put into agricultural machinery (two combine harvesters, 64 tractors), 5.32 into wells and hydraulic equipment, and 4.42 into cactus and medicago plantations. (d) In short, considering the size of the initial allocation to this project component, actual achievements seem paltry: TD 0.34 million lent out of an available total of TD 7.5 million, giving a 4.5% performance rate only. 3.40 These results show the highly sensitive character of the approach taken to the subject of complementary investment as an accompaniment to rural roads projects. To decide beforehand on what types of complementary investment should be encouraged and what volume of funding they should absorb, and to rely for their implementation on the "cooperation" of a multitude of farmers with a wide variety of motives and attitudes, has been shown up in practice for what it in fact was, an oversimplified view of reality. On the other hand the poor performance of this component under this project has to be pvt in perspective: alternative sources of financing were available and were actually utilized. Also BNT's involvement has to be clarified and improved: under the approach now advocated under ASAL I and the Fourth Credit Project, a predominant role will be assigned to BNT staff in credit promotion and assistance to farmers in preparing loan applications. (3) Component D, 3.41 Here, a total of TD 4 million was to be allocated for agro-industrial investments by: the Tunisian Dairy Industry Company (STIL), in milk collection facilities; the Cereal Board (OC), in grain collection facilities; the National Wechanised Farming Company (SONAM), in maintenance workshops/leasing centers for farm machinery; and service cooperatives, in the purchase of a range of agricultural equipment. 3.42 ' Financing was to come from the following quarters: (1) 29.71, or TD 1.2 million, from sub-borrowers; (ii) 17.62, or TD 0.7 million, from BNT in medium- and long-term credits; and (iii) 52.71, or TD 2.1 million, from IBRD in medium- and long-term credits. As far as actual achievements go, component Ds of the program was even more of a disappointment than Da, - 33 - since (L) STIL did not require credit because as a result of its access to low-cost milk powder imports, it was not profitable to invest in a milk collection infrastructure; (ii) investment Cemane for farm machinery by SONAM was met at more advantageous terms by suppliers' credit, and thus for it has not been feasible to organize service cooperatives for farm machinery rental and repair; and (iii) finally no funds were released at all by BNT, which assumed 100% of the risk involved. Under the circumstances, the Agricultural Production Department decided to take on direct responsibility for construction of a cold storage depot for seed potatoes at Soliman (paras. 3.34 - 3.35). F. Implementation of Part E: Identification and Feasibility Study for a Second Rural Roads Proiect 3.43 As defined in the Loan Agreement, Part E of the project, to be carried out with the assistance of some 130 man-months of consultants' services, consisted of a series of economic, technical, and financial studies on improving approximately 2,500 km of selected supplementary rural roads and on complementary agricultural investments in their zones of influence. 3.44 The main study (feasibility of 2,600 km of rural roads), begun on May 14, 1979 and completed by December 1980, was carried out by an ad hoc grouping of both French and Tunisian consulting firms. Initially, the goal of the study was to identify 2,600 km of roads by using a simplified version of the Rural Road Model (RRM) developed during the course of the pilot project. Subsequently, however, in order to save one whole phase, it was decided to: (M) conduct feasibility studies, using the RRM, on 1,000 km of rural roads which seemed on the face of things to offer prospects of reasonable rates of return; and (ii) examine the remaining 1,600 km of roads, plus an additional 100 km in the governorate of Sidi Bousid, so 1,700 km in all, for identification purposes, using a "simple" model yet to be developed. 3.45 In practice, the RRM proved difficult to use for two reasons. Firstly, it is best designed for the study of a single road serving a large zone of influence; when the number of roads is large, it becomes too unwieldy and so was virtually unusable in the present study, whith was looking at 220 roads. Secondly, it was made available behind schedule, transcribed with difficulty, and proved nearly impossible to apply in Tunis given the personnel and equipment required (80% of the available capacity of CNI) and the considerable problems created by the input process (800 cards). For these reasons, the consultants resorted to the MAGON program, which provided a good compromise between the RRM and the "simplified" model specified in the initial agreement (para. 6.01). 3.46 Two other studies were carried out under the project: (i) engineering studies for 227 km of the "complementary program" (contract amounting to TD 125,400) and (ii) engineering studies for 410 km of the first tranche of the Second Rural Roads Project (contract amounting to TD 225,400). Together with te main feasibility study amounting to TD 281,800, the actual cost of Part E was TD 632,600, an increase of 18% over the appraisal estimate of TD 530,000. - 34 - Reporting 3.47 Progress reports were to cover both road improvements and agricultural activities in an integrated document. Unfortunately this was scarcely the case. Reporting on the road improvement side was good, while on the agricultural side it took longer to set up a satisfactory monitoring system (the first report on the agricultural components was issued in April 1981), and although an effort was then made to report in a satisfactory fashion timewise and qualitywise, this effort was not sustained throughout the project. This is especially true with BNT, which failed to produce separated accounts per zone of influence (para. 3.37). Procurement 3.48 No problem was encountered for the procurement of road civil works. A general prequalification was carried out in 1978 and then updated from time to time. The grouping of lots in the bid packages followed the rules agreed upon in the loan documents, but due to the size and the nature of the works, few foreign firms were interested and Tunisian firms won all the contracts for rural roads improvement. 3.49 Road maintenance equipment proved to be a much more complex matter. The Bank sent during the two first years of the project three missions of mechanical engineers specialized in procurement to assist the Government in (i) preparing adequate specifications and (ii) establishing a suitable multicriteria method.of bid evaluation. This helped the Highways Department to cope with the multiplicity of bids received from local and foreign suppliers, but the internal award process until final approval by the Main Tender Board remained lengthy. As a result, several rebiddings had to take place because the bid validity period had elapsed and suppliers did not maintain their offers. 3.50 For the agricultural components two procurement methods were used. Local competitive bidding was appropriate for the procurement of extension center facilities and small equipment; but even so it happened in several instances that no local firm was interested in the works associated with construction of an extension office in a remote area. Therefore the Government had to resort to force account works, and this was agreed by the Bank in December 1983. International competitive bidding was used for larger pieces of equipment and, of course, for the building and equipping of the Soliman cold store (awarded to a Canadian firm). 3.51 The consulting services were procured through short lists prepared by the Government. Except for the training consultancy to the Ministry of Agriculture (awarded to an individual expert), all other consulting services were awarded to Franco-Tunisian joint ventures. Performance of Participants Consultants 3.52 The feasibility and engineering studies for rural roads and associated agricultural investments were satisfactorily completed by the - 35 - three different Franco-Tunisian joint ventures. Special mention has to be made about the consultant for the training of extension agents: he performed extremely well in terms of the response of the trainees and the enthusiasm of the regional commissioners. Contractors 3.53 The main problems associated with contractors on the project are related to: (i) financial problems and labor difficulties for two contractors entailing complete work standstill for many months, and (ii) an almost general unawareness, at least at the beginning of the contract, of the requirements for earth roads (use of oversized materials, lack of adequate equipment, insufficient drainage). The situation improved after a time (para. 5.05), but this contributed to a large part of the project time overrun. Government 3.54 The Coordinating Committee consisting of representatives from Ministries of Plan, Agriculture, Public Works, Interior and from BNT was very efficient in putting together the various components in one integrated project. The performance of the Highway and Agricultural Production Departments in sub-projects preparation was good, but as far as construction and operation supervision are concerned, they sometimes lacked the necessary trained staff and equipment. It was unfortunately only in late 1985, toward the end of the project, that the Government released the authorization to purchase most of the vehicles needed for the control of construction and the monitoring of the centers operations. Performance of BNT was mixed and most of the issues associated with agricultural credit have had to be dealt with on a broader national level. IV. PROJECT COSTS AND DISBURSEMENTS 4.01 Actual costs and their relationships to appraisal estimates were discussed for each component separately in Chapter III. A summary for the project as a whole is given in the table herebelow. The dollar equivalency has been calculated using an exchange rate of US$1 = TD 0.435 for the appraisal estimates (value as of May 1978), and US$1 - TD 0.705 for the actual costs. In fact the last figure masks a broad spectrum of exchange rates, from 0.404 for the first withdrawal in May 1980 to 0.860 in January 1987 with intermediary high point of 0.890 in March 1985-' and low point of 0.710 in March 1986. 1/ This led to the June 4, 1985 Loan Amendment raising the disbursement percentage for category I (civil works) from 40 to 55, so that the latter better matches the actual foreign exchange component of the category. - 36 - Actual and Appraisal Estimates of Project Costs (in millions) Actual Appraisal % Actual/Appraisal TD US$ TD US$ TD US$ Rural Road Improv. 24.23 34.38 21.97 50.50 1.10 0.68 Road Maint. Equipment 10.34 14.67 2.35 5.40 4.40 2.72 Extension Services 1.19 1.69 1.22 2.80 0.98 0.60 Agricultural Invest. 0.97 1.38 3.96 9.10 0.25 0.15 Technical Assistance 0.71 1.01 0.61 1.40 1.16 0.72 Total 37.44 53.13 30.10 69.20 1.24 0.77 4.02 The table above shows the two important changes in project scope discussed in para. 3.16 for road maintenance equipment (due to the inclusion of additional needs identified under the follow-up project), and in paras. 3.36 to 3.41 for agricultural investments (only one sub-project handled by the Agricultural Production Department instead of the facilities and equipment program to be handled by BNT). The short and medium-term credits to farmers, not financed by this project, are not considered in the table above (they are dealt with in the Second and Third Agricultural Credit Projects). 4..03 The strong appreciation of the US Dollar (by an average 62% over the project period) led to a general decrease of the actual dollar costs in comparison to the appraisal estimates in that currency. Thus, although some changes in the project scope provoked an overrun of 24% in terms of dinar, this resulted in an underrun of 23% when expressed in dollar. Disbursements 4.04 The actual and forecast disbursements are shown in the key project data table and in the annexed graph. The latter shows also what would have been the actual disbursements in case of a constant exchange rate equal to the appraisal rate. The loan was clo.sed on December 31, 1986 with the loan amount reduced in July 1986 from US$32 to 30 million to finally take account of the reduced scope in agricultural investments. 4.05 As of January 1, 1987, US$ 28.64 million have been disbursed and the remaining part of the loan, US$1.36 million, is expected to be disbursed by June 30, 1987. Assuming a constant exchange rate, the initial loan amount (US$32 million) would have been insufficient to cover the total consequent disbursements (US$46.34 million). 4.06 A further analysis of the disbursement rate shows that: (i) the rate foreseen at appraisal was US$5.8 million per year; (ii) the actual rate was about US$3.6 million per year or 62% of the appraisal rate, integrating both delay in implementation schedule and mostly exchange rate variation; and (iii) the currency constant value disbursement rate is - 37 - US$5.8 million per year, same as the appraisal one. In other words, in dinar terms expenditures and disbursements have followed the pace called for at appraisal. V. INSTITUTIONAL AND OPERATIONAL DEVELOPMENT 5.01 Substantial progress in institutional as well as operational development was achieved through the establishment of the Coordinating Committee. It was officially set up by a presidential decree of November 1978 but had already informally worked during the preparation of this project. The President and Vice-President of the Committee were respectively the Minister of Public Works and the Director of the Highway Department. Members were representatives from the Ministries of Agriculture, Public Works, Plan, Interior, and of the BNT. Meetings were scheduled three to four times per year. The Committee was instrumental in solving interface problems between the constitutive bodies, and in bringing the common issues to the adequate level of political decision (for instance the issues related to agricultural credit or rural roads maintenance). Putting these issues into a broader national perspective was beneficial not only to the project but also to the rural development sector. 5.02 The project helped to identify, and further address in follow-up projects, two fundamental policy issues associated with: (i) the need for a national extension services policy and related monitoring system, and (ii) the agricultural credit policy with special emphasis on interest rates and organization of BNT regarding risk arrangements on sub-loans and decentralization. 5.03 The sector approach retained for this project meant that the preparation of about 601 of the rural roads packages (road improvements plus complementary agricultural investments) rested with the Government. Through its Highways Department (as the main executing agency) and Agricultural Production Department (which had to become familiar with Bank procedures) and assisted by consultants, the Government was quite successful in this respect. The experience gained is contributing to a smoother implementation of the Second Rural Roads (Fifth Highway) Project. 5.04 Regarding road maintenance,'the results are mixed. When the project started, maintenance of unclassified tracks was ill-defined, and the project helped to introduce the concept of progressive insertion of the improved roads into the network maintainable by the Highways Department. Unfortunately the budgetary allocations remained behind the increasing needs, and in spite of the commendable action of the Coordinating Committee which succeeded in having a specific rural roads maintenance line in the government budget (Article 60 bis), actual allocations were rare and meager. This issue had to be tackled at a broader level, questioning the appropriate sectoral balance between investments and maintenance, and the proposed Highway Rehabilitation and Maintenance Project is an answer. A' regards the equipment procured under the project, its use in terms of physical maintenance outputs was not monitored under this project but rather under the Fourth Highway Project: only in the latter was the provision of equipment directly tied to well-defined road maintenance programs and physical targets. - 38 - 5.05 Finally on the operational side two results were achieved. Firstly the extension agents were successfully trained to better organize and develop their activities. Secondly the Tustisian technicians, both from the Government and from the enterprises, became more aware of the particulars of each road construction. However the project identified a need for additional assistance in this respect, and a study on unpaved roads is being carried out under the Second Rural Roads Project. VI. Economic Appraisal and Reevaluation A. Appraisal 6.01 Three different models were used in order to assess the value to the national community of improving the various roads targeted by the project: (a) In the first instance, the Rural Road Model (RRM), although especially comprehensive, was very difficult to put into operation by reason of the considerable volume of data utilized (several thousand items for each road). It was particularly sensitive to the changes and developments in local agricultural production techniques introduced by this type of project. It was utilized for the roads located in the governorates of Le Kef, Siliana, Kairouan and Nabeul. (b) The second model, simpler and easier to operate and dependent on a much more limited stock of data, also generated estimates of the major indicators of rate of return on investments, whether in road improvement or in the farming environment itself. This model was used for roads in the governorates of Kasserine, Sidi Bousid, Gafsa and K6bili. (c) The third model, the MAGON (Mod6le Agronomique d'Optimisation des Niveaux d'am6nagement) prograu, was a compromise between the other two. Systematic use was made of it for the roads to be targeted by the Second Rural Roads Project and in particular for roads in the governorates of Zaghouan, BUjA, and Jendouba, which replaced those initially selected for the pilot project but then rejected as not offering adequate rates of return. 6.02 Table 6 provides details of the results of computations of cost-effectiveness. It will be noted that internal rate of return varied from one road and one governorate to another. This illustrates the wide range of rural conditions in Tunisia, and within one particular area, the variations in the agricultural potential of each zone of influence. 6.03 As to the roads grouped in the complementary program (in the govarnorates of Zaghouan, Bji and Jendouba), internal rates of return were calculated for a range of scenarios, but particularly the following: (i) under scenario 1, road improvements plus agricultural measures during the - 39 - first five years; and (ii) under scenario 2, road improvements only. As can be seen in Table 6, the internal rates of return projected were generally high to very high under Scenario 1 and poor to medium under Scenario 2. The logical conclusion was that the project, if implemented without the complementary investments, would not be sufficiently cost-effective. B. Re-evaluation 6.04 The project in its final form corresponded more to Scenario 2 than Scenario 1, since Part D was implemented to only a very limited degree. Has the return obtained on the investment therefore been less than adequate? 6.05 Unfortunately, it is all but impossible to reconstruct the initial economic analysis using the current data, recapture of which would be too complex, problem-prone, and slow; most items of information required for the calculations would be lacking. One can conclude only that ex-post analysis of the original economic justification for the project is out of the question. 6.06 However, in the interest of obtaining an accurate assessment of the repercussions and real benefits of Tunisia's rural roads projects, the Government, in conjunction with the World Bank, has launched an impact study, which is being carried out under the Second Rural Roads (Fifth Highway) Project. This study is limited to a representative example of 24 roads, of which half vere studied as part of the pilot project and half as part of the follow-up project. Detailed traffic, farming, and socio-economic surveys have been conducted in the zones of influence of the roads in the sample, the goal being to characterize the different facets of the present situation and to pinpoint the changes brought about by the project. 6.07 The impact study results available so far already point to certain conclusions. For instance: (a) comparison of the "with" and "without" project situations as shown in Table 7, reveals that traffic on improved roads has seen a threefold increase on average, wAile the proportion of four-wheel motorized vehicles has gone up from 42% to 66%; (b) gross output in the zones of influence of improved roads, when valued at 1985 prices, is seen to have increased by 78% on average when the "with" and "without" project situations are compared. 6.08 The substantial changes in two such significant indicators, which have virtually doubled, demonstrate that the impact of rural roads projects is high. It can of course be claimed that this impact is not ascribable only to the project studied, but that it is also the result of a general developmental trend in the rural world and of other development projects that, either directly or indirectly, have affected the zones of influence of the roads in question. - 40 - 6.09 Although this general statement is obviously prudent, it is nonetheless true that most of the impact observed flows from the rural roads project, and more precisely from its road component, since the physical improvement of roads has proven in the majority of cases to be a sine qua non condition of all development in their zones of influence. Other promotional actions, whether incorporated into the rural roads project itself or other projects, are no more than a complement and are incapable alone of generating the impact noted unless associated with road improvements. Clear verification of this assertion can be obtained in areas where irrigation facilities are being created, a type of project often accompanied by the development of all-weather roads. Inversely, when the type of production operation carried on in the zone of influence of a given road does not require major journeys during the bad-weather season (cereal crop, for instance), the link between road accessibility and development of the zone is a weak one, and improvements to the road are not reflected in any significant increase in either traffic volume or gross output. This can be seen in the case of road 1202a in the governorate of BWji, which serves as a classic instance of the "without project" situation and confirms the hypothesis that either stagnation or a very low level of development are all that can be expected in the absence of a project or its failure to produce the expected results. 6.10 In short, the preliminary results of the impact study justify the conclusion that the project has proved cost-effective, as manifested in appreciable increases in: (i) gross product, and therefore in the value added created by the project, and (ii) general traffic volume, and therefore in savings in transportation costs. This answers to the question of para. 6.04, with the additional comment that complementary investments other than those programmed as part of the pilot project and financed by one of the numerous existing agricultural development funds (e.g., FOSDA, PAM, PDR, etc.) have certainly been undertaken in the zones of influence of the project roads, prompted, at least in part, by the upgrading of these roads. 6.11 In order to get some quantitative assessment of the economic value of the project a simplified analysis can be done by means of a comparison of current traffic on the project roads with that estimated at appraisal. The table herebelow shows the actual and estimated traffic for a few roads that had been analysed in detailed both at appraisal and in the on-going impact study. Actual and Forecast Traffic Total 4-Wheel Annual Rate Economic Motorized Traffic (vpd) of Increase () Rate of 1977L' 1986A' Actual.' Forecast' Return-' Road 103 41 114 12.0 8.4 21.3 Road 203 3 54 37.9 9.2 19.1 Road 401 45 124 11.9 9.7 18.1 Road 406 34 171 19.7 8.9 16.4 Sources: 1/ Projet Pilot des Pistes Agricoles, Setec-Sotuetec-Sotinfor (1977) 2/ Impact Study, Preliminary Traffic Report, Studi-BCEOM (1986) - 41 - Thus it appears that traffic growth has generally been higher than forecast at appraisal. In view of the prominence of the traffic parameter in rural development as discussed in the precedipg paragraphs, one could expect conservatively that in order of magnitude the actual economic return of the various schemes are not lower than the one forecast at appraisal. This would confirm the economic worth of the project. 6.12 It is expected that a more thorough re-evaluation of rural road projects will be undertaken when preparing the Second Rural Roads (Fifth Highway) Project completion report, in principle in 1988, with the impact study conclusions fully available. VII. ROLE OF THE BANK 7.01 During project preparation two approaches to this then-innovative project were envisaged by the Bank: (i) limit its participation in the project to investments for rural roads and covenant the necessary complementary agricultural investments, or (ii) include the required complementary agricultural investments for financing under the project, which would better assure and facilitate the sequencing and timing of rural roads and complementary investments. The Bank chose the second approach, in line with OED recommendations made in 1975 to foster 'multi-sectoral package programs". This entailed departing from such constraints as (a) insistence on traditional rate of return analysis and standards, and (b) sector focus of each Bank Division. On the first point, an interesting rural roads analysis computer model was specially developed and further improved for use in project preparation and execution. On the second point, once it was decided to proceed with the project, coordination and collaboration between all parties concerned within the Bank was excellent, as witnessed by the fact that nearly 50 of supervision missions were made jointly by the Regional Agriculture and Transportation Divisions. 7.02 The Bank's strong representations in the matter of local funds for the project helped to obtain an improvement of the disbursement status as from mid-1984 with the appreciable result described in para. 4.06 (iii). Also the Bank's representatioa letter of May 1985 on the subject of adequate control of operations by the executing agencies led to a release of the light vehicle acquisition programs which had remained blocked for two years. 7.03 More generally the Bank participation in preparation (44 staff-weeks), supervision (:. staff-weeks), review and discussion of the feasibility studies, helped co clarify many issues related to the methodology of sub-project preparation and the implementation of the extension, credit and road improvement components. - 42 - VIII. CONCLUSIONS 8.01 Overall the project achieved well its objectives. 1,090 km of rural roads were improved, and 34 extension centers were built and about 100 were equipped, at an average cost well in line with appraisal estimates. Appreciable although not quantifiable increases took place in gross product (and therefore in the value added by the product) and in general traffic volume (and therefore in savings in transportation costs). Substantial progress in institutional and operational development was achieved through the project studies, on-the-job and overseas training, and the establishment of a Coordinating Committee to handle rural roads projects. 8.02 One important problem disturbed the project execution: the provision of agricultural credit was implemented to only a very limited degree. The project design was right in calling for complementary agricultural investments, but the credit issue could just not be tackled at the project level. This issue was addressed as one important sector issue, under several agricultural follow-up projects, culminating in the proposed Fourth Agricultural Credit Project. 8.03 Ihe principal lessons learned through this project are: (i) there should be a clear understanding from all parties concerned on the procedures to be followed for preparation and approval of sub-projects; in the same time some degree of flexibility has to be built in the project at the implementation stage (for instance use of force account for extension services, or pavement of a road in certain circumstaves); (ii) a key factor to success is the ability of the Government to plan, prepare and evaluate the integrated packages and to coordinate the activities of all the executing agencies involved; (iii) this project level, although not the most appropriate one to tackle fundamental pilicy issues such as agricultural credit, did provide valuable experience in the development of the extension system and the introduction of farmers credit to many of the areas served; and (iv) thorough evaluation of the complex effects of a rural roads project can only be done through an impact study. (77860) l一―--! I州縱織｝，饞！縱！州，：矓矓 &”·“,&&&&,&&&&&，·“1萋 _｝·！〕！.!&!!!l}l!& 44 TABLE 2 TUNISIA PROJECT COMPLETION REPORT THIRD HIGHWAY PROJECT (LOAN 1601-TUM) Execution of Part B% Road Maintenance Equipment Phase I Contract Quan- Amount Contract Description tit-T (7D) Date DeliveEZ Graders Champion 20 663,600 01/10/80 1981 Front end loaders Michigan 8 192,000 01/10/80 1981 Rubber tired bulldozers Clark 8 382,400 01/10/80 1981 Dump trucks Ebro 8 95,920 01/10/80 1981 Liaison vans Renault 36 84,600 01/10180 1981 Compactors Sakai a 212,680 01/10/80 1981 Fork lift TCM 1 13,073 12/02/81 1982 Hydraulic presses 2 5,781 12/02/81 1982 Valve grinders 3 46,506 12/02/81 1982 Toole - 20,729 12/02/81 1982 Injector calibrators 5 28,207 12/02/81 1982 Parts STAP - 51,110 09/11/81 1982 Parts Dalmas - 40,890 09/11/81 1982 Passenger cars Dacia 3 i8,006 01/10/80 1981 4-4& drive vehicles ARO a 41,976 12/02/81 1982 TOTAL FIRST PHASE 1,897,480 Phase 2 First Trancbe Bulldozers Hanomag 5 355,600 1/07/83 1983 Trucks 10/12T Berliet 10 249,000 1/07/83 1983 Wheel-loader Michigan 1 40,230 1/07/83 1983 TOTAL FIRST TRANCHE 645,130 (77940) -45- TABLE 2 (Cont'd) Phase 2 - Second Tranche Contract Quan- Amount Contract Description tity (TD) Date Delivery Trucks 10/12 T Berliet 40 1,058,911 7/31/84 1984 Trucks 6/7 T Berliet 8 152,669 6/21/84 1984 Tractors Berliet 5 186,069 7/14/84 1984 Low-bed trailers 25 T King 5 82.500 6/21/84 1984 4-4heel drive cars Mitsubishi 26 371.644 relet in 86 underway Light vehicles, Renault 70 361.466 relet in 86 1986 TOTAL LOT 1 2,213,259 Graders Komatsu 30 1,199,850 7/20/84 1985 Scarifiers Koatsau 30 116,460 7/20/84 1985 Wheel loaders Michigan 28 784,089 2/04/84 1984 Wheel loaders Fiat Allis 10 416,820 relet in 86 underway Track bulldozers Fiat Allis 12 804,000 relet In 86 underway Tire bulldozers 5 547,075 relet in 86 underway TOTAL LOT 2 3,868,294 Tankers Lebrero 6 180,000 9/17/84 1985 Tar-sprayers 9000L Scania/Lebrero 8 400,800 4/27/84 1984 Tar-sprayers 450L Sindag 13 52,325 7/20/84 1984 TOTAL LOT 3 633,125 Stone-crushers Nakayma 2 274.000 6/22/84 1984 TOTAL LOT 4 274,000 Compactors 6/15 T Sakai 6 159,858 6/08/84 1984 Compactors Richier 4 152,800 2/21/85 1985 Rollers 6/ST Perroni 9 238,365 5/29/84 1984 TOTAL LOT 5 551,023 Mobile servicing units - Intobal 2 30,679 1985 1985 Mobile lube units - Istobal 20 223,258 1985 1985 COMPLETED PART OF LOT 6 253,937 63.6% of Lot 6 TOTAL SECOND TRANCHE 7,793,640 TOTAL PHASE 2 8,438,770 (77860) - 46 - TABLE 3 TUNISIA PROJECT COMPLETION REPORT THIRD HIGHWAY PROJECT (LOAN 1601-TUN) Status of Part Cl: Construction of Extension Centers (CTV) CTV Built Under the Project Estimated Actual Gouvernorate Existing CTV Leased NR Cost (TD) Cost (TD) Kairouan 3 8 3 39,000 39,000 Mahdia 1 2 1 13,000 13,000 Nabeul 3 4 2 26,000 23,492 Siliana 3 3 2 26,000 23,826 Le Kef 1 4 3 39.000 39.000 Total 1st Phase 11 21 11 143,000 138,317 Gafsa 1 1 1 15,000 12,800 Kebili 2 - 3 45,000 44,480 Sidi Bousid 4 7 5 75,000 73,000 Kasserine 3 - 3 45,000 39,000 Total 2nd Phase 10 8 12 180,000 169,280 Zaghouan 3 1 4 100,000 64,388 Beja - - 4 100,000 100,000 Jendouba 2 3 3 67,000 62.134 Total Complementary Program 5 4 11 267,000 226,522 TOTAL GENERAL 26 33 34 590,000 534,119 7788D-in -47- TABLE 4 TUNISIA PROJECT COMPLETION PROJECT THIRD HIGHWAY PROJECT (LOAN 1601-TUN) Execution of Part Cl: Extension Centers Equioins Quan- Total Cost Equipment Description tit_ in TD Vehicles Small carc 64,855 Minibus 5 59,260 light motorcycles 9 2,385 Pick-up 12 53,688 TOTAL 1ST PHASE 180,188 Small cars 19 50,482 Medium cars 12 66,634 Pick-up 3 33,000 Minibus 2 23,704 TOTAL 2ND PHASE 173.820 Small cars 2 6,400 Minibus 2 26,000 TOTAL COMPLEMENTARY PROGRAM 32.400 TOTAL VEHICLES 386,408 Demonstration Equipment Mini-tractors 1st & 2nd phases 39 203,677 Complementary program 3 17,014 Audio-visual equipment 12 2,640 Other toqls - 2,645 TOTAL DEMONSTRATION EQUIPMENT 225,976 Office Equipment Furniture 46 10,540 Calculators 11 1,113 Typing machines 14 5,831 Other office equipment - 4,690 TOTAL OFFICE EQUIPMENT 22,174 General Miscellaneous expenditures 17,189 TOTAL GENERAL 651,747 77860 - 48 - TAGLE 5 PROECT e .ETWen REPOMT THiRD HIGHMAY PROJECT <LOan igo1-TUN% £tatus af Part p3 en 12-21-19n2 t1dige and Löna Term Credttes ta Farrme In the 2aeet Influeneel Reeived Appraved Olsbursed. x 1 Aunt Aunt Appnoved/ Amsunt Disbursed/ LTan ATa lReattent 5 TD .i.gE ..Ree.ived ggg.IRee.tved Cabine harvesters 14 173,438 4 50.868 29.3 2 22.570 44.4 Tractors and tools 294 1.457.414 128 554.777 38.1 64 281.652 50.8 Nydraulte equipmt 81 46.718 58 30.933 66.2 16 9.577 31.0 Wells 16 37.438 6 14.755 39.4 4 8.308 56.3 Cusheds 29 95.167 9 14,747 15.5 0 0 0 Sheepsheds 8 11.086 6 6.741 60.8 0 0 0 Purchaseo cowa 87 33.742 76 29.968 88.8 0 0 0 Purchase of shep 4 1.925 1 735 38.2 0 0 0 Cactus Plantattons 228 26,754 198 22.716 84.9 120 8,344 36.7 Medicago Plantations 121 18.287 117 17.472 96.5 58 6.310 36.1 Frult Trm Plantattons _jg 172 .j 11.74 .-m MM _a 11-4a 0 TOTAL 1.014 2.074.577 614 758.452 36.4 264 336.761 44.6 - - - p -- g e - = ~8S1 - 49- TABLE 6 TUNISIA PROJECT COMPLETION REPORT THIRD HIGHWAY PROJECT (LOAN 1601-TUN) Economic Rates of Return Derived From Feasibility Studies Gouvernorate Road Nbr. ERR Gouvernorate Road Nbr. ERR Le Kef 101 18.4 Sidi Bousid .602 19.6 103 21.3 606 9.3 104 23.1 607 14.6 105 10.3 651 19.6 106 12.9 652 16.5 109 10.5 658 14.0 110 14.4 659 17.0 107 - 108 - Siliana 201 24.7 Gafsa 703 10.9 202 28.1 706 11.5 203 19.1 Kairouan 401 18.1 Kebili 801 37.0 402 28.9 802 19.7 404 11.0 803 12.8 405 14.5 804 14.1 406 16.4 805 - 407 12.1 806 44.4 408 24.8 SCE.1 FCE.2 Nabeul 301 62.0 Zaghouan 1101 14 0 302 74.0 1102 19 13 303 44.0 1103 16 8 304 46.5 1104 18 10 305 23.4 1105 17 0 306 41.5 1106 29 1 307 68.4 1107 17 0 308 31.4 108 Kasserine 501 12.1 Beja 1201 20 7 504 13.0 1202a 28 9 506 12.7 1203 26 22 507 13.6 1204a 24 17 1204b 31 17 Jendouba 1301a 2k 11 1301b 24 11 1302 24 6 1303 18 0 1304a 19 13 1358 30 15 7760 PaECT COMPLETION REPORT THI MIIY PROJECT - LOAU 1601-TIl Camarlson of Daily Traffic in the Situations with and without Prolect Total Motor Vehicles Total Traffic SttuuVais Abicles Except Tractors Tractors Twm-eals AnImAls.o (MR (R) ( ) R) (%) ((R) () Road with' 447 182 40.7 9 2.0 203 45.4 53 11.9 339 301 without2 244 15 6.1 5 2.0 56 23.0 168 68.9 9 ratio w1w.o. 1.8 12.1 1.8 3.6 0.3 3.2 Road with 378 213 56.3 21 5.6 95 25.1 49 13.0 34 305 without 113 54 31.2 13 7.5 11 6.4 95 54.9 164 ratio w/w.o. 2.2 3.9 1.6 8.6 0.5 3.4 Road with 115 43 37.4 11 9.6 36 31.3 25 21.7 7e 203 without 106 a 0.0 3 2.8 a 0.0 103 97.2 2s ratio w/w.o. 1.1 3.7 0.2 2.5 Road with 161 107 66.5 7 4.3 40 24.8 7 4.3 I39 103 without 46 41 89.1 0 0 2 4.3 3 6.5 42 ratio w/w.o. 3.5 2.6 20 2.3 3.3 Road with 859 485 66.5 109 12.7 95 11.1 95 12.7 669 501 without 443 152 34.3 46 10.4 68 35.3 177 26.6 1eN ratio w/w.o. 1.9 3 2.4 1.4 6.5 3.4 1 Road with 729 534 73.3 60 8.2 104 34.3 31 4.3 644 0 507 without 199 102 51.3 l 5.0 56 28.1 31 15.6 143 ratio w/w.o. 3.7 5.2 6 1.9 1.0 4.5 Road with 264 153 58.0 t8 6.8 68 25. 25 9.5 204 406 without 73 32 43.8 2 2.7 16 21.9 23 31.5 45 ratio w/w.o. 3.6 4.8 9.0 4.3 1.1 4.5 Road with 160 112 70.0 12 7.5 27 16.9 9 5.6 162 401 without 62 45 72.6 0 0.0 9 14.5 a 12.9 56 ratio w/W.o. 2.6 2.5 3.0 1.1 2.9 Road with 782 612 78.3 11 1.4 128 16.4 31 4.0 692 651 without 226 17S 77.4 3 1.3 33 14.6 Is 6.6 262 ratio w/w.e. 3.5 3.5 3.7 3.9 2.1 3.4 Road with 113 S 48.7 9 8.0 26 23.0 23 20.4 as 706 without 162 20 12.3 22 1S.6 30 18.5 90 55.6 02 ratio w/w.e. 0.7 2.8 0.4 0.9 0.3 1.0 Road with 1068 592 55.4 19 1.8 447 41.9 10 0.9 791 801 without 392 159 40.6 0 0.0 147 37.5 86 21.9 250 ratio wlw.o. 2.7 3.7 3.0 0.1 3.16 r* I/ Current traffic (1986 survey). Source: Impact Study. Preliminary Traffic Report, STUI-CEIN. 2/ 1977 Survey (without project there would have been no development in the zone of influence, hence the traffic stagnation). Source: Projet Pilote des Pistes Agricoles. SETEC-SOTUETEC-SOTINFOR. October 1977. 1/ Total of pack animals and animal-drawn carts. 1/ Based on the following coefficients: 1 for cars -d pick-ups; 2 for trucks and buses: 0.25 for tw-wheels and pack an'sals. 0.50 for animal -drawn carts and tractors; 0.75 .-a,tors with trailer. (7786D) 47.4 TUNISIA PROJECT COMPLETION REPORT THIRD HIGHWAY PROJECT - LOAN 1601-TUN Disbursement Schedule 32 30 30 -o28.6 25 20 Appraisal Eitimate z/ Actual 15. Constant Exchange Rate 00 1979 1980 1981 9I2 1983 1984 1985 1986 1987 .~...) FISCAL YEA ：以