Tunisia - Second Urban Sewerage Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 7917 PROJECT PERFORMANCE AUDIT REPORT TUNISIA SECOND URBAN SEWERAGE PROJECT (LOAN 1675-TUN) JUNE 30, 1989 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 disclost- without World Bank authorization. THE WORLD BANK FOR OFFICIAL USE ONLY Washington DC 20433 USA OIfere ad DeerCe.eral Oe6MM Ialuauaion June 30, 1989 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Performance Audit Report on Tunisia Second Urban Sewerage Project (Loan 1675-TUN) Attached, for information, is a copy of a report entitled "Project Performance Audit Report on Tunisia - Second Urban Sewerage Project (Loan 1675-TUN)" prepared by the Operations Evaluation Department. Attachment This document has a restricted ditribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. FOR OFFICIAL USE ONLY PROJECT PERFORMANCE AUDIT REPORT TUNISIA SECOND URBAN SEWERAGE PROJECT (LOAN 1675-TUN) TABLE OF CONTENTS Page No. Preface ................................................. ........ i Data Sheets ....................................................... ii Evaluation Sunmary .............................................. iv PROJECT PERFORMANCE AUDIT MEMORANDUM I. BACKGROUND . .................... ..... 1 II. THE PROJECT ..............................................3 III. PROJECT IMPLEMENTATION ........................... ...... 6 IV. PROCUREMENT ........8....................................8 V. DISBURSEMENT ........8....................................8 VI. PERFORMANCE OF SUPPLIERS, CONTRACTORS AND CONSULTANi: ......8..................................8 VII. PROJECT COSTS ............................................9 VIII. TECHNICAL ASSISTANCE .......................... ......... 9 IX. OPERATIONS ............................................. 10 X. INSTITUTIONAL PERFORMANCE ................................ 10 Staffing and Training ............................... ..... 11 XI. FINANCIAL PERFORMANCE .......................... ......... 12 XII. BANK PERFORMANCE ................................. ........ 15 XIII. CONCLUSIONS ............................................ 16 Sustainability ............................... ............ 17 Lessons to be Learnt ........................... .......... 17 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 WGrld Bank authorization. TABLE OF CONTENTS (continued) Page No. PROJECT COMPLETION REPORT I. Introduction............................................ 21 II. Project Identification, Preparation and Appraisal....... 22 III. Project Implementation and Cost......................... 25 IV. Financial Performance................................... 29 V. Institutional Performance..................... ........ 32 VI. Bank Performance............... ........................ 33 VII. Project Justification........................ ........ 33 VIII. Conclusions and Lessons Learned......................... 35 ANNEXES MAPS 1. IBRD No. 13845R 2. IBRD No. 13846R 3. IBRD No. 13847R PROJECT PERFORMANCE AUDIT REPORT TUNISIA SECOND URBAN SEWERAGE PROJECT (LOAN 1675-TUN) PREFACE 1. This Project Performance Audit Report (PPAR) presents the results of a performance audit of the Tunisia Second Urban Sewerage Project for which Loan 1675-TUN of US$26.5 million was approved on March 22, 1979. The loan was made to the Government of Tunisia (GOT) for on-lending to Office National d'Assainissement (ONAS), the Government-owned agency responsible for the planning, operation, maintenance, and development of sewerage fa- cilities in Tunisia. The loan agreement was signed on April 13, 1979 and following the completion of the subsidiary loan agreement and other docu- ments, became effective on August 31, 1979. There were changes made in the scope of the original project, however, it was not deemed necessary to amend the Loan Agreement. The project was completed in June 1988 after a delay of four years and at a cost some 50Z above appraisal estimates. Project implementation delays caused a lag in disbursement and the closing date was extended on three occasions for a total of three years to December 31, 1987. Commitment fees on the loan amounted to US$1.1 million. 2. The Project Performance Audit Report consists ot a Project Perfor- mance Audit Memorandum (PPAM), and a Project Completion Report (PCR) pre- pared in June 1988 by staff of the Europe, Middle East and No-th Africa Region Office. The audit involved reviews of the PCR, the Appraisal and President's Report, legal documents, Bank files, and the transcripts of the Executive Director's meeting at which the project was approved. An OED mission to Tunisia in January 1989 had discussions with officials of ONAS in Tunis and Sfax, all of whom had knowledge of the facts. 3. The PCR provides a very good account and assessment of the project experience, and discusses the performance of the Bank and the project execu- ting agency. the PPAR summarizes and provides comments on the major aspects of the project. 4. Following standard OED procedures, copies of the draft PPAR were sent to Government and the Borrower. No comments were, however, received. -it - PROJECT PERFORMANCE AUDIT REPORT TUNISIA THE SECOND URBAN SEWERAGE PROJECT (LOAN 1675-TUN) BASIC DATA SHEET KEY PROJECT DATA Appraisal Actual or Actual as X of Item Expectation Current Estimate 4ppre*al Estimate Total Project Costs (USS militon) 77.6 110.5 15. Loan Amount (US$ million) 26.5 28.5 166 Date Physical Component Completed 8/60/64 6/36/8 177 L Proportion completed by that date (3) 10 52 - Economic Rate of Return 16i -21 CUMULATIVE ESTIMATED AND ACTUAL DISSURSEMENIS FY90 FY01 FY82 FY88 FY64 FY6 FY6 FY87 FY88 Appraisal Estimate (USS million) 6.5 4.4 12.9 29.0 26.5 Actual (USS million) - .7 2.6 9.6 11.4 15.6 19.6 21.9 26.5 Actual as 9 of Appraisal (S) - 16.9 23.3 41.7 48.6 51.6 74.6 01.5 166.6 Date of Final Disbursement: February 26, 1966 PROJECT DATES Original Plan Revisions Actual First Mention in Files 6/19/77 - - Negotiations 12/16-21/78 - - Board Approval 8/22/79 - 8/22/79 Signing (Credit Agreement Date) 4/18/79 - 4/13/79 Effectiveness 7/12/79 8/81/79 8/81/79 Closing Date 12/31/64 11/81/85 12/81/87 12/81/94 21/31/87 / From date of sign;ng. - Mi STAFF INPUTS (staff weeks) FY?6 FY77 FY76 FY79 FYN FYM1 FYI2 FYS FYg4 FY65 FY4 FY67 FYI6 Total Identificatlon/ PreparatIon 0.2 1.1 9.9 0.2 - - - - - - - - - 11.4 Appraisal - - - 22.6 - - - - - - - - - 22.6 Negotiations - - - 14.1 - - - - - - - - - 14.1 Supervision - - - 0.2 5.1 6.8 13.7 4.2 6.4 6.9 6.0 1.4 7.7 68.9 Other - - - 0.4 - - - - 6.6 - - - - .9 Subtotal 6.2 1.1 9.9 37.5 5.1 6.3 13.7 4.2 5.4 6.9 6.6 1.4 7.7 107.6 MISSION DATA Date No. of Specializations (mo./yr.) Persons Represented Preparation 12/77 3 ENG FNA Appraisal 3/78 1 X 6-7/78 2 X X Supervision I 12/79 2 X X Supervicion II 8/86 2 X X Supervision III 12/80 1 X Supervision IV 1/01 1 X Supervision V 11/81 2 X X Supervision VI 6/32 2 X X Supervision VII 2/88 2 X X Supervision VIII 9/63 2 X X Supervision 'X 6/84 1 X Supervision X 12/84 2 X X Supervision XI 6/85 3 X X Supervision XII 6/86 3 X X Supervision XIII 10/87 1 X Supervision XIV 2/88 1 X Completion 6/88 2 X X OTHER PROJECT DATA Borrower: Republic of Tunisia Executing Agencies: Office Naticnal D'Assainlesement kONAS) - iv - PROJECT PERFORMANCE AUDIT REPORT TUNISIA SECOND URBAN SEWERAGE PROJECT (LuAN 1675-TUN) EVALUATION SUMMARY 1. This loan for the Urban Sewerage Project was the second Bank loan to the urban sewerage subsector in Tunisia. Since water resources are scarce, the Government of Tunisia (GOT), is promoting the rational use of water, including the reuse of waste water. In pursuit of the principle of maximizing benefits from the limited available resources (including finan- cial resources) GOT directed the main thrust of its water supply towards the densely populated areas, and the national water authority, Societd Nationale d'Exploitation et de Distribution des Eaux (SONEDE) has followed that directive since 1977. Presently 651 of the total population has ac- cess to safe piped water supply; this includes 97Z of the urban population of which about 80Z is supplied througb private house connections (PPAM, paras. 1.01-1.03). 2. Increased water supply leads to increased volumes of waste water to be cc ed, treated and disposed of. GOT's nain objective for the sewerage ector is to bring sewerage services to a level comparable to that of potable water supply, through the rehabilitation and expansion of existing sewer and drainage systems, and the development of treatment and disposal facilities to safeguard the environment and abate health hazards (PPAM, para. 1.04). 3. The Bank has been active in the water and sewerage subsector since 1966, with the object of assisting GOT to attain its service level objec- tives for both water supply and sewerage services, and to step up public enterprise savings through management efficiency, and tariff policy. The sewerage subsector has received five Loans and one Credit totalling US$112.5 million from the Bank Group (PPAM 1.05-1.06). In 1975, under the First Urban Sewerage Project (Loan 1088-TUN) GOT, with Bank assistance, created Office National de l'Assainissement (ONAS) to be responsible for the planning, operation, maintenance, renewal, and construction of all sewerage works located within the limits of the communes, and within tour- ism and industrial development zones. 4. The OED report on the evaluation of the First Urban Sewerage Project (Loan 1088-TUN) concluded that the project was successful in all major aspects, however, the project confirmed the importance of adequate project preparation. The project was appraised on the basis of feasibility studies, and without an updated urban development plan for Greater Tunis. This resulted in revisions to the project scope during implementation (PPAM, para. 1.07). 5. Tue first mention in the files of a Second Urban Sewerage Project was in a Secuor Brief of May 19, 1977. The Brief proposed a project which would include a second phase of sewerage for the Greater Tunis district, collection and treatment works for a number of towns, 13 be selected on the conclusion of an ongoing sector review; and part of a program to provide more tertiary aewers and house connections in all centers to be served by CNAS. GOT meanwhile was seeking co-financing for the project from the European Investment Bank (EIB) and the European Development Fund (FED). EEC had suggested to GOT that the project proposed by the Bank be divided into two parts: the facilities in Tunis and Sfax being financed by the Bank, and those for seventeen other towns by EEC. The project, including only components for Tunis and Sfax, was appraised by Bank staff in June/July 1978 (PPAM, para. 2.02). 6. The project consisted of a second stage of sewerage eztensions for Greater Tunis, and a sewerage project for Sfax. The total cost of the project was estimated at US$77.5 million of which US$26.5 million would be foreign exchange. The foreign costs of the project would be financed by the Bank; local costs would be financed in part from ONAS' cash flow and in part from GOT. The project was expected to be completed by December 31, 1983. The project was in line with GOT's main objectives for the sewerage subsector of bringing sewerage services to a level comparable with that of potable water supply, and promoting the reuse of treated sewage for indus- trial or agricultural use in order to supplement the country's scarce water resources. The scope and configuration of the project had been planned to provide a package of least-cost investments in those centers, which repre- sented a logical development of facilities already in operation or under construction; and to further develop the managerial and technical capabil- ity of ONAS <PPAM, paras. 2.03-2.06). 7. Two aspects of ONAS' management gave rise to concern during the appraisal: (a) The relative inexpezience of staff having responsibility for proj- ect execution, and the inadequacies of their planning and control procedures; and (b) the large number of managers reporting to the President Director General (PDG) and the resultant risk of a aecision-making bottle- neck. It was agreed that ONAS would retain a firm of consultants to coordinate all planning of new works, to supervise construction, and to introduce and hand over to ONAS a proper system of planning and control, and the foreign cost of the assistance would be financed from the loan. ONAS was also required to submit, for consideration at negotiations, proposals for strengthening top management to allow a greater degree of delegation from the PDG. Negotiations took place December 15-21, 1978 (PPAM, paras. 2.07-2.11). The loan was approved on March 22, 1979, and after a short delay, became effective on August 31, 1979 (PPAM, 2.12-2.13). - vi - 8. The project was expected to be completLd by June 30, 1984 (project implementation period had been extended six months after appraisal). Al- though the delays on individual components varied, the overall project was finally completed by June 30, 1988, with a delay of four years. Reflecting inadequate project preparation, much of the delay was due to changes in the scope of the work in both centers; other delays were also caused by the lengthy import and land expropriation procedures. One of the Sfax collec- tors still presents a potential problem and requires urgent attention if the investment is to be effective and fulfill the use for which it was intended. Reportedly sever connection rates in Tunis and Sfax at the end of 1987, were 91Z and 642 respectively; on the basis of these figures ONAS achieved its target for Tunis and considerably exceeded the target for Sfax (PPAM, paras. 3.01-3.04). 9. Although there were some delays, procurement for the project pte- sented little difficulty. ONAS had no difficul.z in complying with the Bank's guidelines for procurement (PPAM, para. 4.01). 10. Disbursement lagged significantly throughout the implementation period. The closing date was extended three times, between December 31, 1984 to December 31, 1987. Final disbursement was ma.e on February 26, 1988. Commitment fees (US$1.1 million) were heavy (PPAM, para. 5.01). 11. Suppliers and contractors, both local and foreign, generally per- formed well. The consultants were competent and carried out their duties satisfactorily (PPAM, paras. 6.01-6.02). 12. The final costs of the project reflected the lack of adequate preparation of the project. Due to underestimation at appraisal, changes in the scope of the project, implementation delays, and the devaluation of the dinar, the final costs of the overall project in Dinar terms was 50Z higher than the cost forecast at appraisal. The Tunis component exceeded the forecasts by 50Z; costs on the sewer system and sewage treatment facil- ities in that city exceeded the estimates by 49Z and 402 respectively. The sever system works were substant1illy changed after appraisal and con- siderable additional construction and improvements were cirried out on the pumping station, force main, canal and outfall. The final costs of the Sfax component were 1082 over the forecasts. The actual costs of the storm water system exceeded the estimates by 144Z, and the cost of 'work on the combined system and sewers exceeded the appraisal forecast by 47% and 11% respectively. Although there were changes in the scope of the project in Sfax, the principal reasons for the high costs were underestimation and project implementation delays. There was a 25% saving on the cost of con- sultants services (PPAM, para. 7.01). 13. A technical assistance component was included in the project to cover the cost of consultants for strengthening ONAS' planning and con- struction capability and putting in place a project planning and control system. After experiencing difficulties in recruiting a suitable engineer, GOT and ONAS requested the Bank to second a Bank staff engineer to head the - vii - planning unit for the duration of the project. Although supervision mis- sions reported that the new unit was proving useful to ONAS .anagement, the latter considered that the arrangement was not appropriate in the Tunisian scene, and had not achieved the results expected (PPAM, para. 8.01). 14. Since 1975 ONAS has considerably extended its sever system opera- tions and is now operating countryvide through four regional orgrnizations, covering thirteen districts. ONAS has taken over 55 municipal systems since 1975 and has 327,000 connected customers. The widespread nature of ONAS's operations, the size of the facilities, the poor physical state of the systems taken over, and the limited financial resources available, pose a problem for ONAS' maintenance capability. The problem is exacerbated by ONAS' reported need for a systematic preventive maintenance programs. Unfortunately, rather than increase, ONAS has been forced to reduce this part of its operations because of lack of funds. The implications of this situation need to be brought to the attention of GOT (PPAM, paras. 9.01-9.02). 15. The appraisal of the project showed ONAS' management to have had difficulties in implementing its growing investment program for three main reasons: (i) staff responsible for project execution was relatively inexperi- enced, and project planning and control procedures were inad.- quate; 'ii) ONAS' financial accounts were being prepared with increasing delay; and (iii) an excessive number of managers (eleven) were reporting to the PDG. At the request of GOT and ONAS, the Bank seconded a Bank staff engineer to head the planning unit; subsequently in June 1982, his local counterpart took over the unit and was appointed Director of Planning and Studies. The Unit has been working satisfactorily for several years. To correct the accounting problems ONAS appointed a local management consulting firm to upgrade its financial, administration and management reporting systems; these measures, together with staff changes and new recruitment, have a- chieved the desired results, and ONAS' accounting and financial administra- tion has showed progressive improvement and is satisfactory. The improve- ment in the accounting function has also been reflected in the elimination of the delays in the production of the audit reports. Concerning the third problem ONAS has r2vised the organizational structure on three occasions since 1978. At the present time, only six executives and managers report to the PDG (PPAM, paras. 10.01-10.04). 16. ONAS management have tightly controlled the staffing, and as of end 1988, total staff numbered, 1970, which represented about six employees per 1,000 customers which signals a high level of efficiency. ONAS has developed a comprehensive in-house training program; each year about 202 of total staff undergo short term training. In addition, senior staff are sent overseas for training in sanitary engineering and other related - viii - fields. Staff training has also been given by the equipment suppliers s part of their contracts, and also from the consultants engaged on the project (PPAM, paras 10.05-10.06). 17. Despite the large increase in customers, because of the revenue policy adopted when ONAS was created, and the subsequent failure of GOT to permit ONAS to make timely and adequate tariff increases, ONAS continues to be heavily dependent on non-customer related sources of revenue; which are not related to costs and are outside the control of ONAS. When ONAS was created in 1975, 752 of total revenues came from GOT subsidier (502), and muni-cipal taxes (25Z); despite the large increase in sewerage services since then, only 52Z of ONAS's revenue was derived from the customers in 1987; 22? came from GOT subsidies and 25% from munic4pal taxes (PPAM. paras. 11.01-11.02). 18. Under the First and Second Sewerage Projects (Loans 1088 and 1675), the financial covenants required ONAS to provide annually funds from internal sources equivalent to 4Z of its net fixed assets, including work in progress. Under the Third Project (Loan 2255) the covenant was changed to a simpler--and nn1y tenuously related to the preceding--parameter, which required ONAS to meet not less than 13? of its annual investment and fund requirements from internal sources. The new covenant also required ONAS to eliminate Government operating subsidies after 1986. Both these require- ments should be regarded unreason&bly low considering the nature of the investments but were apparently justified on the grounds of the extensive social benefits expected to result from the projects (PPAM, paras. 11.0?-11.04). 19. During the period 1979-1982 ONAS financed 34? of its investments from internal sources (which included GOT operating subsidies equivalent to 19? of investments), although, in 1981, it provided only 2.9?, below that covenanted. ONAS' financing performance steadily deteriorated after 1983 and in 1986 and 1987 it had negative cash flows (despite GOT operating subsidies equivalent to 21? of investments) and so failed to comply with the new (13%) covenant. Needless to say it also failed to comply with the requirement to eliminate Government subsidies. Overall, during the period 1979-1987 ONAS contributed only 10? to investments from internal sources. ONAS has only been permitted to increase its customers' tariffs on four occasions in the nine year period 1979-87. In addition, no adjustment has been made in the rate on which GOT subsidies are calculated and the rate remains the same as in 1977. On the basis of the 1987 accounts, a substan- tial increase in tariffs is necessary. Unless measures are taken to cor- rect ONAS' poor financial situation, ONAS will not only have difficulty in completing its development program but, because of inability to carry out quality maintenance, will find that the works which have been carried jut in recent years will require rehabilitation. There is no evide.ice that serious efforts have been made to improve ONAS' financial situation during the project period and the Bank appears to have condoned the violation of the financial covenants of the Loan Agreement (PPAM, paras. 11.05-11.11). - ix - 20. The past close relationship between the Pank and ONAS was contin- ued during the implementation of the project. The Bank's project supervi- sion effort was normal, and procurement was dealt with expeditiously. The project seriously lacked adequate preparation; this weakness was reflected in the large number of changes made in the project as appraised, the large variations in costs, delays in implementation, high commitment fees, and higher than expected investment. The Bank did not profit from the lesson learnt from the First Urban Sewerage Project. Given the weakness in proj- ect preparation, the Bank had no alternative but to be flexible in agreeing to the changes. The secondment of the Bank staff member did not achieve the expected results. The Bank brought about greater awareness of the need to improve investment planning and control, and in financial management. In 1983, after the revenue covenant was changed, sewage flows, which have implications for customers revenue, were overestimated for 1985-1987 (PPAM, para. 12.01). 21. Although the project was completed with significant delay and, due to changes in its scope, at substantially higher than estimated cost, the final project essentially achieved its physical objectives in both the Tunis and Sfax areas. The project brought considerable benefits, not only to the direct users of the facilities, but also to the population at large, through the health benefits and improvement to the environment. The insti- tutional component provided continuation of the institutional building started under the first project (PPAM, para. 13.01). 22. Provided the installed facilities are satisfactorily maintained, and expanded to meet increased demand, there should be nc difficulty in sustaini.ig the benefits from the pro4ect. The sustainability o the insti- tutional improvements should present no problem provided they are updated and developed as ONAS progresses (PPAM, para. 13.02). 23. The principal lessons to be learnt from this project are: (i) Sewerage projects are usually complex and require particularly careful preparation. The Bank must resist the urge to lend quick- ly and discourage unrealistic programs by its borrowers. (ii) Cost recovery in waste disposal projects is an imperfect art and careful monitoring of the financial performance of the entity, particularly in relation to agreed covenants, is essential so that appropriate remedial actions may be Levised in good time. The Bank should have had such action initiated before ONAS' financial position was deteriorated to its present state (PPAM, para. 13.03). - 1 - PROJECT PERFORMANCE AUDIT MEMORANDUM TUNISIA SECOND URBAN SEWERAGE PROJECT (LOAN 1675-TUN) I. BACKGROUND 1.01 Tunisia is located on the southern shores of the Mediterranean Sea about midway between Gibraltar and Suez. The country is poorly endowed with natural resources; water resources are scarce and at present 60Z of total usable water resources are under exploitation and by the year 2000 more than 90% will be utilized. Urban and agricultural water demand could soon exceed available regional water resources. Because of the scarcity of water in Tunisia, the Government of Tunisia (GOT), in promoting the ration- al use of water, attaches great importance to reuse, particularly for irri- gation. The principle that treated sewage will eventually be used for irrigation has been adopted and, in fact, the first of such schemes is about to be put into operation in an agricultural area near Tunis. 1.02 Administratively, Tunisia is divided into 20 governorates, 136 counties and 162 communes. Half of the estimated total population of about 7.7 million , which is growing at an average annual rate of 2.5% is consid- ered urban, and more than one-third of the urban population is concentrated in Tunis and Sfax, the two largest metropolitan areas. The rural popula- tion is widely dispersed in more than 3,500, often isolated locatiots, with low populations which impede rapid implementation of water supply and 3ani- tation systems. 1.03 In pursuit of the princip13 of maximizing benefits from the lim- ited available resources (including financial resources), GOT directed the main thrust of its water supply policy towards the densely populated areas. Societ4 Nationalq d'Exploitation et de Distribution des Eaux (SONEDE), the nation.l water authority, has followed that directive since 1977 and pres- ently, about 652 of the total population has accers to safe piped water supply; this includes 97% of the urban population of which about 80% is supplied through private house connections. 1.04 Increased water supply leads to increased volumes of waste water to be collected, treated, and disposed of and GOT's main objective for the sewerage subsector is to bring sewerage services to a level comparable to that of potable water supply. This objective is to be achieved by the rehabilitation, upgrading, and expansion of existing sewer systems, the provision of flood control and drainage, and the development of treatment and disposal facilities in order to safeguard the environment and abate health hazards. - 2 - 1.05 In 1974, under the First Urban Sewerage Project (Loan 1088-TUN), GOT created Office Nationale de l'Assainissemet.t (ONAS) under the tutelle of the Ministry of Equipment, to be responsible for the planning, opera- tion, maintenance, renewal, and construction of all sewerage works located within the limits of the communes, and within tourism and industrial devel- opment zones. ONAS started its work in July 1, 1975, and embarked on a program of taking over existing sewer systems from the municipalities who had previously been responsible. The first priority was to rehabilitate the sewer systems and restore the treatment plants to operating condition. To complement the effort of SONEDE, and more precisely define the overall objective set by Government, ONAS has set itself the task of sewering all dwellings receiving water supply in built up areas, within five years of taking over operational responsibility in each area. At the time of Lp- praisal of the project under audit (June/July 1978), ONAS had assumed ra- sponsibility for the systems in thirteen municipalities and in five tourism zones, and while the rate of expansion would depend on ONAS' capacity to implement new and remedial works, and satisfactorily operate and maintain the systems, it was estimated that the takeover of the remaining urban systems would be complete by 1983. However, ONAS had overestimated the task. At the appraisal of the Third Sewerage Project in early 1983, 26 municipality systems had been taken over; and as of January 1989 this num- ber had been increased to fifty-five with a further twenty-five systems to be taken over during the next five years. To implement its expansion, ONAS retained foreign consultants to carry out a general study of the sewerage subsector and establish a program of priority works. Using the results of the study, together with tl-e results of another study for an update of the sewerage master plan for greater Tunis, ONAS prepared an investment program through 1983 which embraced five projects. Three projects, the First Urban Sewerage Project, the Tourism Infrastructure Project, and the project under audit were financed by the Bank; one was financed by Swedish aid, and a secondary cities project was partially financed by EEC. 1.06 The Bank has been active in the Tunisian water an1 sewerage sector since 1966; and heiped to establish both SONEDE and CNAS, and has main- tained a continuous and close dialogue with GOT on sector policies. The Bank's strategy is to assist GOT to attain its service level objectives for the water supply and sewerage sector, and to step up public enterprise savings through improvements in management efficiency and tariff policy. To date the sewerage subsector has received five loans and one credit, totalling US112.5 million, from the Bank Group. The last project, the Third Urban Sewerage Project, for which a loan of US$34 million was made in 1983, is still under implementation. 1.07 The OED report on the evaluation of the First Urban Sewerage Project (Loan 1088-TUN) concluded that the project was successful in all major aspects, however, the project confirmed the importance of adequate project preparation. The project was appraised on the basis of feasibility studies, and without an updated urban development plan for greater Tunis. This resulted in revisions to the projecr scope during implementation. - 3 - II. THE PROJECT 2.01 The first mention in the files of a second urban sewerage project was in a Sector Brief dated May 19, 1977, which had been prepared following a supervision mission in April 1977 for the First Urban Sewerage Project. The Brief proposed a project, for processing in FY79, which would include a second phase of sewerage for the greater Tunis district; collection and treatment works for a number of towns to be selected on the conclusion of an ongoing sector review; and part of a program to provide more tertiary severs and house connections in all centers to be servEd by ONAS. The Minister of Equipment had indicated to the mission that he would be in favor of a project, with a total cost of about DT 16.6 (US$39.1) million, the foreign exchange component being about DT 10.0 (US$23.5) million. The mission proposal was confirmed in letters to GOT and ONAS in September 1987. GOT meanwhile was seeking cofinancing for the project, firstly from the European Investment Bank (EIB), and then the European Development Fund (EDF). A preparation mission visited Tunisia in December 1977, and pro- posed a project of US$70.0 million which might be financed as follows: US$ millions World Bank 28 EIB 28 GOT 14 Total 70 A further preparation mission visited Tunisia in March 1973. The mission was presented with revised financial forecasts, prepared by ONAS' consul- tants, and based on additional revenues from a tariff increase, which was expected to be implemented before the end of 1978, ONAS believed that the investment program could be increased to US$96 million and financing might be as follows: US$ millions World Bank 28 FED (EEC) 28 GOT 40 Total 96 The program would consist of the construction of a second stage of facili- ties in greater Tunis and works in eighteen other cities, including Sfax, however, should the tariff increase not be approved, the program would have to be brought down to US$70 million. 2.02 In May 1978, EEC confirmed its interest in participatin,; in fi- nancing a sewerage project for secondary towns. Subsequently, Bank Pro- grams learnt that ECC had suggested to GOT that the project proposed by the Bank should be split into two parts, with the facilities in Tunis and Sfax being financed by the Bank, and those in the seventeen other towns by EEC. Therefore, a project confined only to facilities for Tunis and Sfax, was appraised by Bank staff between June 16 and July 13. 1978. Discussions took place with EEC in Brussels, on July 14, 1978, and the EEC representa- tive indicated that there was no objection on the part of EEC to the approach the Bank wished to take. However, reportedly, EEC had not formal- ly decided to include such a sewerage project in its lending program at that time, and was not able to indicate that participation would be poss- ible, within the time-frame envisaged by the Bank for a wider project. Accordingly, although Programs, in a memo as late as August 8, 1978, asked Projects for further discussions on the matter, reportedly for reasons of differences in EEC and Bank procedures and the time factor, the Bank de- cided to limit the project to facilities in Tunis and Sfax. 2.03 The project,l/ represented part of the capital works program of ONAS for the remainder of the Tunisian Fifth Plan (1977-81) and the early years of the Sixth Plan, and comprised: (a) A second stage of sewerage for greater Tunis. This component would consolidate the achievements of the First Urban Sewerage Project (Loan 1088-TUN) by installing additional treatment capac- ity to protect the Lake of Tunis from pollution, and by extending service to an increasing proportion of the population; (b) a sewerage project for Sfax, the second largest city of Tunisia. Treatment facilities were already under construction in thic city with Swedish aid funds, and this component of the project would renovate and extend the sewer system, so as to provide the berne- fits of those facilities to the population as quickly and widely as possible; (c) technical assistance to strengthen th-- project planning and control capability of ONAS. 2.04 The estimated total cost of the proj%ct was US$77.5 million in- cluding a foreign exchange component of US$26.5 million which would be financed by the Bank. Local costs would be financed in part from ONAS' own cash flow and the remainder in the form of capital contributions from the Government to ONAS. The project was expected to be completed by December 31, 1983. 2.05 The project was in line with GOT's main objectives for the sewer- age subsector, which were to bring sewerage services to a level comparable to that of potable water suDply, and promote the reuse of treated sewage for industrial or agricultural use in order to supplement the country's scarce water resources. 1/ Full details are given in the PCR, para. 2.04. 2.06 The scope and configuration of the project had been planned with the following considerations in mind: (a) to provide a package of investments in centers which represents a logical development of facilities already in operation or under construction, and to provide needed services at the least overall cost; (b) to further develop the managerial and technical capability of ONAS to discharge its growing responsibility for sector operation. 2.07 Two aspects of the management of ONAS gave rise to concern during the appraisal: (a) the relative inexperience of staff having responsibility for proj- ect execution, and the inadequacies of their planning and control procedures; (b) the large number of managers currently reporting to the President Director General (PDG) and the resultant risk of a decision-making bottleneck. 2.08 To respond to the first of these problems the PDG accepted the recommendation of the mission that ONAS retain a firm of consultants to coordinate all planning of new works (not merely those in the proposed Bank project), to supervise construction, and to prepare and hand over to ONAS a proper system of project planning and control. The foreign cost of these services (US$2.6 million) would make up the technical assistance component of the loan. 2.09 While the technical assistance measures would help overcome tech- nical weaknesses in ONAS' management, they would not substitute for execu- tive control and decision making. A reluctance to make decisions--even on routine matters--was apparent among many of the eleven managers and direc- tors who reported to the PDG, with the result that too many matters com- peted for his attention. The evident danger of the PDG becoming overload- ed, and decisions delayed as a result, was magnified by the rapidly growing scale of ONAS' operations. The mission considered that this situation was prejudicial to the timely implementation of the project and recommended that ONAS be required to submit proposals, for consideration at negotia- tions, as to how its top management was to be strengthened to allow a greater degree of delegation from the PDG. 2.10 During the review stage of the project, the Loan Committee re- quired that the submission of the main findings and recommendations of the auditors' report on ONAS' 1977 accounts should be a condition of negotia- tions. ONAS was also asked to submit before negotiations proposals for the improvement of its administrative and financial management. - 6 - 2.11 Negotiations took place December 15-21, 1978. ONAS, in a letter dated December 13, 1978, informed the Bank of the measures taken since ap- praisal and to be taken in 1979 to strengthen ONAS' management. These measures were deemed satisfactory, Bank management agreed, as a result, to waive the condition of negotiations relating to submission by ONAS of its 1977 Audit Report. 2.12 At negotiations, the Tunisian delegation requested that the terms of the loan (17 years, including four years of grace) be varied and made the same as those for the First Urban Sewerage Project (25 years including four years of grace). This request was found to be unacceptable to the Bank because ONAS' projected financial operations did not justify more lenient terms than the standard terms proposed by the Bank. The Bank nego- tiating team was prepared, however to recommend to management more lenient terms (19 years including four years of grace) provided the Government agreed to reduce, proportionately, terms of one or more loans scheduled for FY79 so as to remain within country limits. In the absence of Government instructions on this point, negotiations were not completed. Subsequently the Planning Minister cabled the Bank to request an increase in the terms of the proposed loan to 19 years, provided those of future FY79 loans weIld not be reduced. This proposal was also unacceptable to the Bank. Agree- ment was eventually reached in Tunis and the proposed Bank terms of 17 years, including four years of grace were accepted by GOT. 2.13 A number of questions on the project were raised by Executive Directors' staff before the Board meeting (see also PCR, para. 2.07). However, these wer0 all answered satisfactorily by Bank staff; no questions were raised at the Board meeting on March 22, 1979 when the loan was approved. The loan and project agreements were sigred on April 13, 1979 and the loan became effective on August 31, 1979 after a short delay. III. PROJECT IMPLEMENTATION 3.01 During appraisal it was considered that the greatest risk to the project would be construction delays and consequent cost escalation. Al- though ONAS was a newly created agency, the implementation experience dur- ing the First Urban Sewerage Project had been disappointing. However, the causes for the delay in that project, some of which had been outside the control of ONAS, were taken into account in formulating the second project. The principal measures to address the controllable causes of implementation delay were: (i) the strengthening of ONAS' top management; and (ii) the provision of technical assistance to coordinate and strengthen ONAS' project planning and control capability. It was considered that by addressing the executive and technical bottle- necks, there would be reasonable grounds to expect an improved implementa- tion performance from ONAS. 3.02 The project was expected to be completed by June 30, 1984. Al- though the delays on individual components varied, the overall project was finally completed by June 30, 1988; with a delay of four years. Much of the delay was due to changes in the scope of the works, both in Tunis and in Sfax. In Tunis, significant project changes were made in sewer works because of the construction works for a new city transport system; changes were also made in the design of the principal treatment plant which caused a delay in awarding the contract; other delays were caused b- import proce- dures, lengthy land expropriation procedures, and the failure of the munic- ipality to complete road and sidewalk surfacing on time. In Sfax, a col- lector was taken out of the project because of estimated high cost and later reinstated; another collector was replaced by an open canal; a third collector was realigned; and, in addition, there were variations in other sewer works. These changes, in both centers, reflected inadequate prepara- tion in the initial project. In fact, these first impressions were later confirmed during the audit, when it was learned that the Sfax component had been prepared from a master plan, without a detailed feasibility study. 3.03 In Sfax, one of the collectors constructed under the project still does not function; this could pose a potential danger, given that the area is subject to flooding. The PCR correctly states in para. 3.03 that in that city there was a modification of the alignment of Collector L to take into account the new embankment of the railway; however, it should be added that the realignment was also made to avoid obs+ruction of the outfall from the huge deposits of phosphate residue, which had been dumped by the state- owned phosphate firm on a site opposite the outfall. When the initial realignment was made, ONAS understood that no further dumping of resid"e would be made. Unfortunately, that has not been the case and the dumping continued until the plant was closed at the end of 1988. Collector L is presently ineffective, since it is blocked by phosphate residue; it will therefore be necessary to dredge a channel to open water to allow storm- water to discharge. ONAS has drawn the attention of the responsible Ministry to this and the matter is under review. Collector L is an impor- tant component in the Sfax flood prevention scheme and its construction was a condition for the Bank making a loan for the Sfax Flood Prevention Project. 3.04 It was reported to the audit mission that, at t-ie end of 1987, the sewer connection rates in Tunis and Sfax were 91Z and 64Z respectively; on the basis of these figures, ONAS had therefore achieved its target for Tunis and considerably exceeded the target for Sfax. 3.05 Finally, regarding the overall delays in priject completion, there is some evidence that the Government has deliberecely slowed down all in- vestment projects in view of country economic difficulties during the project period. - 8 - IV. PROCUREMENT 4.01 The files indicate that although there were delays, procurement for the project presented little difficulty and was satisfactory. There was some delay in Bank clearance of the award for the Choutrana Sewerage Works contract and the Bank called for additional information from ONAS; however, eventually the award was made according to ONAS' evaluation. The PCR correctly states that ONAS had no difficulty in complying with the Banks guidelines for procurement. V. DISBURSEMENT 5.01 Because of procurement and implementation delays loan disbursement lagged significantly throughout the implementation period. At appraisal it was expected that the loan would be fully disbursed by December 31, 1984. However, as of that date disbursement was only 51% complete. The closing date was extended three times, firstly to December 31, 1985, then to December 31, 1986 and finally to December 31, 1987. The loan was fully disbursed, final disbursement being made on February 26, 1988. Because of delayed procurement commitment fees on the loan (US$1.1 million) were rela- tively heavy. VI. PERFORMANCE OF SUPPLIERS, CONTRACTORS AND CONSULTANTS 6.01 Suppliers and contractors, both locAl and foreign, in general, performed well. Reportedly, ONAS had problems with one contractor who was unable to complete his contract, and the uncompleted work was executed by other contractors. 6.02 Prior to loan approval, ONAS had retained Dutch consultants to carry out a review of the sewerage sector and establish a national program of works by order of priority. The Bank had reviewed and cleared the coi,- sultants terms of reference. Subsequently, ONAS appointed four firms of consultants, one firm each, for the Tunis outfall, the Tunis sewer system, the Tunis treatment plant and the Sfax component. The consultants, with the assistance of ONAS' staff, prepared the detailed design and bidding documents, ebzisted ONAS in the evaluation of bids and in the supervision of the construction of the components in Tunis. Implementation in Sfax was essentially supervised by ONAS. The consultants were competent and carried out their duties satisfactorily. VII. PROJECT COSTS 7.01 Due to underestimations at appraisal, changes in the scope of the project, project delays, and the devaluation of the dinar, the final costs of the whole project in dinar terms was 50Z higher than forecast at ap- praisal (paras. 3.14-3.16 and Annex 2 of the PCR provide detailed costs with relevant explanations), and reflect the lack of adequate project prep- aration. In Tunis, final overall costs of the component (TD41 million) were 502 over the appraisal estimate of TD27.4 million. The rationaliza- tion of the sever system in that city, which was estimated to copt TD3.2 million, a t.ually cost TD8.0 million, an increase of 1492; this componeAt was significantly changed after appraisal, and there was a substantial increase in the work as compared with that foreseen. Higher than expected costs also occurred on the pumping stations, force main, canal and outfall which in total exceeded the appraisal forecast by 140?, considerable addi- tional construction and improvements were carried out on these works. The cost of work on the sewer and storm water system (TD15.0 million) exceeded the estimate by 28?. In Sfax, the final overall costs (TD6.6 million) exceeded the appraisal figures (TD3.2 million) by 108%. The actual :osts of the storm water system was 144? above the forecast and the cost of work on the combined system and sewers exceeded the appraisal forecasts by 47? and 11Z respectively. Although thr!re were some changes in the scope of the project in Sfax, the principal reasons for the cost overruns were underes- timation and project delays. There was a 25% saving on the cost of consul- tants services. VIII. TECHNICAL ASSISTANCE 8.01 As a newly formed agency ONAS overcame its initial difficulties in staffing through active recruitment and training and the judicious use of expatriate technical assistance and consultants. However, during the ap- praisal it was perreived that ONAS, in particular, needed to strengthen its Alanning and construction management capability and also put in place a project planning and control syste-. A technical assistance component was therefore included in the project, and the Bank agreed to finance 72 man- months of consultants' time for this purpose, the intention being that such appointee(s) would report directly to the President of ONAS and would as- sist ONAS' divisions with planning and monitoring the programs. After experiencing difficulties in recruiting a suitable engineer, GOT and ONAS requested the Bank to second one of its engineers to head the planning unit for the duration of the project. The Bank agreed to the request, and the engineer was seconded for a three year period as from June 1980. Although supervision missions reported thereafter that the new unit was proving useful to management, the unit lacked authority, and in November 1981, the PDG expressed the opinion that the concept was too expensive and another - 10 - arrangement had to be found. The Bank staff engineer returned to Washing- ton in June 1982. During the audit the efficacity of this type c: techni- cal assistance was discussed. The explanatior given from the Tunisian side was that the arrangement of having a highly paid expatriate engineer in charge of competent local staff. was not appropriate to the Tunisian acene, and for this reason the secondment had not achieved the results expected. 8.02 During the implementation period ONAS delegated staff to work closely with the foreign consuLtants on the various physical components of the project and also with tne local consultants on the financial, admini- strative and management improvement component. The consultants were not only executing the tasks for which they had been appointed, but they were training ONAS staff who had been seconded to work with them, and at the same time giving technical assistance in the form of advice to ONAS manage- ment to assist them in making the correct decisions. IX. OPERATIONS 9.01 ONAS is carrying out countrywide sewer system operations, through four regional organizations; metropolitan Tunis, north, center, and south, in all covering thirteen districts of the country. ONAS has taken over 55 municipal systems, and operates 3840 km of sewer systems, 157 pumping sta- tions, and 26 treatment stations. The total number of connected customers is 327,000. Considering that ONAS only started operations in 1975 this is a creditable performance. Techniral operations are well organized and customers' services are effiiently managed. Customers billing and collec- tions are carried out by SONEDE on UNAS' behalf; the arrangement is satis- factory. 9.02 The widespread nature of ONAS' operatio,s, the extent of the fa- cilities, the poor state of the systems taken over from the municipalities, and the limited financial resources available, however, pose a problem for ONAS' maintenance capability. The problem is exacecbated by ONAS' reported need to increase maintenance through systematic preventive maintenance programs. Unfortunatel rather than increase maintenance ONAS has been forced to reduce this part of its operations because of lack of funds. The implications of this situation have been brought to the attention of GOT has recent Bank missions and ONAS has begun to improve its operations. X. INSTITUTIONAL PERFORMANCE 10.01 The appraisal of Loan 1675-TUN in 1978 showed ONAS' management to have had difficulties in implementing its growing investment program for three main reasons. Firstly, the staff responsible for project execution - 11 -. was relatively inexperienced and ONAS' project planning and control proce- dures were inadequate. [econdly, ONAS' financial administration was unsat- isfactory and financial accounts were being prepared with increasing de- lays. Thirdly, an excessive number (11) of managers were reporting to the PDG, which, despite his own creditable performance, burdened him unduly and hampered decision-making. It was therefore proposed that consultants be recruited to (i) coordinate all planning and design of new works and super- vise the execution of ONAS' program, (ii) develop and install a proper system of project planning, monitoring and control and train ONAS' staff to use it. As regards the financial administration and accounting problem it was recommended that a Director of proven capability be appointed who could assume control of personnel, administration, and finance. It was believed that with this latter change, the pressure on the PDG would also be re- lieved. 10.02 As explained in Chapter VIII, ONAS obtained from the Bank the servicea of a staff engineer to take charge of a coordination and planning unit, newly created for the purposes mentioned in the previous paragraph. The local counterpart to the Bank staff engineer took over when the latter left in June 1982, and was appointed Director of Planning and Studies. This unit has been working satisfactorily for several years. 10.03 To correct the accountiig problems, ONAS appointed a local manage- ment consulting firm in June 1980 to upgrade its financial, administration and management reporting systems. This measure, which has been a success, together with staff changes and new recruitment, have achieved the desired results, and ONAS' ccounting and financial administration has showed pro- gressive improvement and is satisfactory. The improvement in the account- ing function has also been reflected in the elimination of the delays in the completion of audit reports. The quality of audit -eports is satisfac- tory. 10.04 As regards the third problem, since 1978 ONAS has reviewed and revised the structure of its organization on three occasions, 1980, 1987 and more recently in January 1989. The number of executives and managers who reported to the PDG in 1978 was eleven; as a result of the changes in 1980 and 1987 they were reduced to eight and finally in 1989 they have been brought down to six. Under the latest restructuring, the responsibility for personnel, administration, and finance have been given to one depart- mental head. Staffing &nd Training 10.05 ONAS' management have tightly controlled the staffing of the agency. At the end of 1988, total staff numbered 1970, which represented about 6 employees per 1000 customers; at the appraisal of the Third Urban Sewerage Project in late 1982, it was estimated that total staff of ONAS in 1988 would be about 2,525 and the ratio of employees to 1,000 customers would be 7.8. - 12 - 10.06 ONAS has developed a comprehensive in-house training program with the assistance of experts from EEC and Federal Republic of Germany. Each year about 202 of total staff undergo short-term training covering differ- ent aspects of ONAS' operations. In addition, senior staff are sent over- seas for training in sanitary engineering and other related fields. Staff training has also been received from the equipment cnDpliers as part of the contracts, and also from the consultants engaged on the project. In the case of the accounting and financial upgrading, the consultants implement- ing the improvements gave ONAS' staff, relevant training in the new proce- dures. XI. FINANCIAL PERFORMANCE 11.01 The PCR correctly describes ONAS' financial performance. The following paragraphs provide further clarification. 11.02 ONAS' revenues are derived from three main sources: (i) Customers charges. Charges to connected customers by way of a two-part tariff to cover operating and mainte- nance costs for waste water treatment, and part of the costs of storm water drainage and treatment; charges to clients for work done and for sales of by-products; (ii) Contributions from municip-lities taxes. An 8Z contri- bution from the "Funds Communs des Collectivites" to cover part of the costs of storm water drainage and treatment; (iii) Government SubsiLies. Subsidies from GOT which since 1977 have been calculated at the rate of 40 millimes/m3 for the water supplied in areas served by ONAS. The percentages of total revenue derived from the three sources for FY76, ONAS' first year of operations, FY77 and FY87 are given below: FY76 FY77 FY87 (Z) (Z) ( ) Customers Charges 25 17 52 Municipality contributions 25 29 26 GOT Subsidies 50 54 22 100 100 100 The table shows the imbalance in the sources of revenue available to ONAS. Because of the manner in which the revenue policy was formulated when ONAS - 13 - was created; the subsequent failure of GOT to permit ONAS to make timely and adequate tariff increases; and despite a large increase in sewage flows, ONAS continues to be heavily dependent on non-customer related sources of revenue, which are not related directly to costs and which are outside the control of ONAS. 11.03 Under Loan 1088-TUN and Loan 1675-TUN, ONAS was requited to pro- ide from internal sources, funds for investment, which would annually be equivalent to not less than 42 of the net value of fixed assets including work-in-progress. Funds from internal sources wire defined as gross reve- nues (including subsidies), less costs of operation (excluding deprecia- tion), maintenance, administration, debt service, taxes, and increases in working capital other than cash. In 1983, following a review of ONAS' financial policy, the co-enant was changed and a more straightforward re- quirement was adopted under Loan 2255-TUN (the Third Urban Sewerage Projec- t); ONAS was then required to finance, from internal sources, not less than 132 of annual capital investments. It was also required that the subsidies from GOT should cease after 1986. 11.04 During the implementation period of the project (1979-1987), ONA. failed to meet the first covenant in 1981, contributing only 2.6z of fixed assets instead of the 4Z required. It also failed to meet the second cove- nant in 1986 and 1987, when its cash flows were negative. Needliss to say, ONAS was also unable to eliminate the compensatory payments in 1987 or thereafter. 11.05 The undermentioned table, which has been compiled from the figures in the PCP (annexes 3 and 4) shows ONAS' financing performance between 1979 and 1987. A B C FYs 1979-1982 / FYs 1983-1987 /b FYs 1979-1987 Forecast N Actual 3 Forecast X Actual 3 Forecast % Actual 3 TO millions Capital Investaent 72.1 72.6 135.5 101.3 207.6 173.8 Financed From: Internal Sources / 13.8 19 24.1 34 23.9 18 (6.8) (7) 37.7 18 17.3 10 Gov't Conts. Grant. 24.3 34 22.7 31 65.9 41 86.1 64 80.2 39 87.8 60 Long-term Debt 34.0 47 25.7 35 55.7 41 43.0 43 89.7 43 68,7 40 TOTAL 72.1 100 72.5 100 135.5 i00 101.3 lee 207.8 100 173.6 100 Gov't Operating Subsidy / 15.4 14.0 14.3 21.6 29.7 35.6 /a Forecast figures for Loan 1675-TUN. /b Forecast figures for Loan 2255-TUN. /c Includes consumers contributions. / Included in internal sources. - 14 - 11.06 Between 1979 and 1982 (Part A of the table), ONAS financed 34Z of capital investment from internal sources which included operating subsidies from GOT equivalent to 192 of investment. In addition to the operating subsidies, GOT also made contributions equivalent to about 30Z of ONAS' investment. Although ONAS failed to meet the financial covenant in year 1981, over the four-year period ONAS' actual financing performance exceeded the fore,:ast made at appraisal time. 11.07 The deterioration in ONAS' finances was clearly reflected in the Agency's financing performance during the period 1983-1987 (Part B of the table). Despite operating subsidies, which were equivalent to about 21Z of investment, ONAS was unable to make any contribution to investment from internal resources during that period. In fact ONAS had a negative cash flow which had to be covered by borrowing from GOT. 11.08 For the overall nine-year period (Part C of the Table), ONAS, even after taking into account operating subsidies which were equivalent to 191 of investment, was only able to contribute from internal sources, 1OZ to investment, instead of the covenanted 13Z. The remaining 90% of investment was financed by GOT (50%) and borrowings (40%). 11.09 The reasons for ONAS' unsatisfactory financing performance go back to the revenue policy adopted in 1975 when ONAS was created. At that time, GOT and the Bank, considered that there was justification for GOT to subsi- dize about 50Z of ONAS' operational costs, because of the benefits which ONAS' facilities, would bring to the general population, in the form of improved public health, and to GOT from the higher land valies on property adjacent to the cleaned-up Lake of Tunis. Accordingly, these benefits were converted into a charge to GOT, based on the total wate- consumed in areas served by ONAS; the initial charge of 16 millimes/m3 in 75, was increased to 40 millimes/m3 in 1977, a rate which is still operLtive. The initial charges to domestic and industrial customers and tourism agencies were 8, 23, and 15 millimes/m3 respectively. Although customers rates were subse- quently substantially increased in July 1978, the average customers sewage charge was still only 34 millimes/m3 in 1979 or 822 of the rate charged to GOT. It was the intention of GOT in 1975 to progressively reduce the sub- sidy as the number of connections to ONAS' system increased and as tariffs increased. Successive loan agreements have provided for annual reviews of tariffs, and during the appraisal of Loan 2255-TUN in 1983 it was envisaged that all subsidies would cease after 1986. However, for economic and po- litical reasons, GOT has not been able to permit ONAS to increase its cus- tomers charges to such levels as would allow the agency to replace the subsidies, or to provide adequate funds for operational and development requirements. In this respect, ONAS has only been permitted to increase its tariffs on four occasions in the nine year period 1979-1983. In addi- tion, of course, ro adjustment has been made in the rate on which subsidies are calculated, and the rata remains the same as in 1977. A substantial increase in tariffs is necessary. The situation in which ONAS now finds itself is paradoxical. On the one hand, the agency is being pushed to assume wider responsibilities in the subsecor and to take over, rehabili- tate and develop further poorly maintained municipal facilities, while on - is - the other hand, it is unable to have the financial resources to carry out its mandate. Unless measures are taken to correct this situation, ONAS will not only have difficulty in completing its development program, but, because of inability to carry out quality maintenance, will find that the works which have been carried out in recent years will require rehabilita- tion. 11.10 ONAS' financial position has been deteriorating or, at least, stagnating it an inadequate level since the early stages of the project. There is no evidence that serious efforts have been made to improve the situation either by ONAS or the Bank even after the 1983 review of ONAS' financial policies. The nature of sewdrage services makes "conbumer related" charging difficult and there is no universally accepted practice. Indirect changes are common in many parts of the world and ONA5' problems are clearly more related to the level of charges rather than the structure. The Bank appears to have condoned the violation of the loan covenants, at least tacitly. It is a noble target by the Government to aim for compre- hensive sever coverage but only if it is accompanied by appropriate charg- ing policies. Refusal to permit tariff increases in 8 out of 12 years (1976-1987) does not meet this criterion. 11.11 Documentation indicates that only in 1988 did the Pank take up this issue seriously which resulted in a 20Z tariff increase in July 1988 and in the preparation of a cumprehensive sewerage cost recovery study. A satisfactory completion of such a study, the implementation of its recom- mendations, given ONAS' mature and reasonably efficient organization, would ensure the satisfactory continuation of the subsectors' development and the sustainability of its facilities' extensive benefits. XII. BANK PERFORMANCE 12.01 During the preappraisal, appraisal and implementation phases of the project, the past close relationship between the Bank and ONAS, which had been developed under the First Urban Sewerage Project was continued. Considering that the project was implemented over a period of eight years, the Bank's supervision effort of 59 manweeks was normal. The following are general comments on Bank performance during the project experience: (i) The project lacked adequate preparation in both centers, particu- larly so for the Sfax component, which was prepared from a master plan, without a detailed feasibility study. This overall weakness was reflected in the large number of changes which were made in the scope of the project, and the large increase in the investment made as compared with that originally estimated. The lack of preparation was partly responsible for the delays in project im- plementation, and the high commitment fees (US$1.1 million on a - 16 - loan of US$26.5 million), and the higher than forecast investment which had to be financed by GOT. The Bank did not profit from the lesson learnt from the First Urban Sewerage Project (see para. 1.07). (ii) Given the weakness in project preparation, the Bank had no alter- native but to be flexible in agreeing to changes. (iii) Throughout project implementation ONAS' financial situation was deteriorating and, in several years, it was in default of the financial covenants of the Loan Agreement. The Bank did not appear to have pursued this issue with any vigor which cast doubts on the effectiveness of its supervision activities. (iv) The project lacked a component for feasibility studies for further centers to be taken over by ONAS. (v) The Bank brought about the greater awareness for the need to im- provement in investment planning and control, and in financial management, although the secondment of the Bank staff member did not achieve the expected results. (vi) The Bank dealt promptly with procurement matters. (viii) In 1983, when the revenue covenant was changed, sewage flows, which have implications for revenue, were overestimated for 1985-1987. XIII. CONCLUSIONS 13.01 Although the project was completed with significant delays and, due to changes in the scope of the project, at substantially higher than estimated cost, the final project essentially achieved its physical objec- tives. The extensions La the waste water collection, treatment and dispos- al facilities in both the Tunis and Sfax areas brought considerable bene- fits not only to the direct users of the facilities, but also to the popu- lation at large, through health benefits and improvement to the environ- ment. In Tunis, the project also extended and consolidated the important environmental benefits achieved in the Tunis Lake under the First Urban Sewerage Project, while the Sfax component also provided important facili- ties essential to the elimination of storm water flooding to which the city has been prone. Under the institutional component of the project, the improvement of the management, financial, and technical capability of ONAS was a continuation of the institution-building started under the first project. On the other hand, the apparent lack of effort to put ONAS' fi- nances and cost recovery policies on sound footing was a serious omission, by all parties, which makes remedial action in the future all the more difficult politically as well as administratively. - 17 - Sustainability 13.02 Provided the installed facilities are satisfactorily maintained and expanded to meet the increased demand, which will largely depend of effective pricing policies, there should be no difficulty in sustaining the benefits from the project. The sustainability of the institutional im- provements should present no problem provided they are updated and devel- oped as ONAS progresses. Lessons to be Learnt 13.03 The two principal lessons to be learnt from this project are not revelations either to the Bank or to Tunisia: (i) Sewerage projects are, by nature, complex and tend to face unfore- seeable problems. Proceeding, with almost undue haste, with such a large project, before the lessons of the first two projects could be absorbed, and on the basis of prelixinary designs only, was asking for trouble. The subsequent delays and substantial cost increases fully confirm this. The Bank must resist its own urge to lend and discourage unrealistic programs by its borrowers. (ii) Inadequate project preparation included insufficient analysis of the financial aspects of the project. Cost recovery in waste disposal projects is an imperfect art the world over and, given the fast expanding business of ONAS, continuous monitoring of the financial performance should have been a high priority followed by appropriate remedial action programs. As it was, the Bank per- mitted sustained non-compliance with vital financial covenants with the inevitable result that a relatively efficient agency sinks into virtual bankruptcy and with every passing year, recov- ery becomes more and more politically unpalatable. It is yet to be seen whether the most recent efforts can salvage the situation and ensure sustainability of the investments made under the pi:oject. - 19 - REPUBLIC OF TUNISIA SECOND URBAN SEWERAGE PROJECT LOAN 1675-TUN PROJECT COMPLETION REPORT Infrastructure Division Country Department II Europe, Middle East and North Africa Region - 21 - REPUBLIC OF TUNISIA SECOND URBAN SEWERAGE PROJECT (LOAN 1675-TUN) PROJECT COMPLETION REPORT I. INTRODUCTION 1.01 The Bank has been active in Tunisia's water and sewerage sector since 1966, when it helped establish and develop both sector utilities, the Societe Nationale d'Exploitation et de Distribution des Eaux, SONEDE, in 1;68, and the Office National d'Assainissement, ONAS, in 1974. Before 1974, twenty municipalities had some form of partial sewer systems; eight sewage treatment plants were extant but were either overloaded or badly maintained. This situation had resulted from the fragmentation of respon- sibility between municipalities, the low level of investment and lack of experienced staff. Recognizing the need to rapidly upgrade sewerage facilities and improve their operation, the Government passed, on August 3, 1974, a law creating ONAS, to be responsible to the Ministry of Equipment for the planning, operation, maintenance, renewal and construction of all sewerage works. Close cooperation between ONAS and SONEDE was expected and has been achieved. ONAS started its work on JL1y 1, 1975, and embarked on a program of takeover of existing sewer systems from the municipalities. So far it has taken control of Tunis, Sfax and 27 other cities, providing service to a population of 1,9 million inhabitants, which is about 50% of Tunisia's urban population and about 24% of the total population of about 7.9 million inhabitants in 1988. 1.02 The Government's objectives in the sewerage subsector have been to increase wastewater collection, treatment and disposal facilities to elimi- nate the harmful effects of untreated wastewater. Investments in the sub- sector since the late 1960s have been dimensioned so as to catch up with unmet needs and with increasing sales of water. Government's recognition of the need for sewerage facilities is evidenced by the growth of budgetary allocations, which increased gradually from TD 8.9 million for the Third Plan (1969-72) to TD 35 million for the Seventh Plan (1987-91). ONAS' priority has been the rehabilitation of existing systems and the restora- tion of treatment plants to operating condition and then sewering all dwellings receiving water supply in built-up areas within five years of taking over operational responsibility in each area. To implement this expansion of urban sewerage, ONAS retained foreign consultants to carry out a general study of the sewerage subsector and to establish a national program of works in order of priority. The Second Urban Sewerage Project had its origins in this study. It was designed to continue development of the sewer and drainage systems in Tunis and Sfax. 1.03 The Bank financed two sewerage related projects before the Second Urban Sewerage Project through two loans and one credit: the Tourism Infra- structure Project (Loan 858-TUN for US$14.0 million and Credit 329-TUN for US$10.0 million, both of September 28, 1972) included US$16.64 million - 22 - for sewerage facilities in six tourism zones, and the First Urban Sewerage Project (Loan 1088-TUN of February 18, 1975 for US$28.0 million), which supported the establishment of ONAS in 1974 as operating agency, the cleaning up of Lake Tunis and the improvement of the sewerage facilities of the Greater Tunis area. The First Project had a major environmental impact because of the depollution of Lake Tunis and the suppression of obnoxious odors. Although the two projects suffered substantial delays, these have tended to diminish in later years as ONAS has matured and gained experience. The Operations Evaluation Department (OED) has published a Sector Operations Review on the Water Supply and Waste Disposal Program in Tunisia (Report 4146 of October 20, 1982) which states that there have been few examples of the Bank's involvement in the Water Supply and Waste Disposal Sector as successful as the Tunisian experience. 1.04 The Second Urban Sewerage Project, here under review, included a second stage of sewerage expansion works for Tunisia's two largest cities, Greater Tunis and Sfax, and technical assistance to strengthen ONAS' project planning and control capabilities. This project represented thus a continuation of the efforts initiated under the First Urban Sewerage Project. Its cost was estimated at US$77.5 million at the time of appraisal and its impiementation period at about 5 years. II. PROJECT IDENTIFICATION, PREPARATION AND APPRAISAL Origin, Preparation and Appraisal 2.01 Since its creation, ONAS gave priority to the rehabilitation and then to tne expansion of urban sewerage. To implement this expansion, ONAS retained Dutch consulting engineers, to carry out a general study of the sewerage subsector, to establish a national program of works in order of priority and to update the sewerage master plan for Greater Tunis. These studies were partly financed by Loan 1088-TUN in the framework of the First Urban Sewerage Project. On the basis of these studies ONAS planned its investment program, which included the works of the Second Urban Sewerage Project. 2.02 In April 1977, the Government requested the Bank's assistance for financing the Second Urban Sewerage Project. The project was identified in December 1977 and appraised six months later, in June 1978. ONAS, with the assistance of its consultants, prepared the project. At the time of appraisal, only preliminary engineering had been completed. The project included, therefore, the hiring of consultants for final design and construction supervision. Targets and Goals 2.03 The only quantified objective included in the SAR was the service coverage, which was expected to reach 89% of dwellings in Greater Tunis by 1985, compared with 77% in 1978. The actual 1985 value was about 75%, which means that ONAS could not satisfy the higb increase in demannl for - 23 - sewerage services. The SAR mentiond the key indicators that would be used during project execution to evaluate the performance of ONAS but, due to lack of adequate data, only three financial ratios were quantified. Project Description 2.04 The project was carried out in Greater Tunis and Sfax and included an institutional component. The project description is as follows: Greater Tunis. The project consisted of facilities to improve the sewerage and drainage systems. A new treatment plant was to be built at Choutrana, to serve Greater Tunis beyond the capacity of the Cherguia treatment plant. This plant was to discharge treated sewage into the Khelidj canal, and thence, into the Mediterranean Sea; at a later stage, the treated wastewater would be used for irrigation. Thus, no sewage or treated sewage was to be discharged into the Lake of Tunis. The project works in Greater Tunis included: (a) recalibration of the combined system of the City of Tunis (about 3.5 km); (b) extension of the trunk sewer system (total length about 31 km); (c) extensions to the sewer system to serve built-up areas then without service (394 ha); (d) drainage systems for the suburbs; (e) construction of pumping stations and force mains relating thereto; (f) construction of a new treatment plant at Choutrana west of the Sebkhet Erriana; capacity: BOD 20,000 kg/d, 5,000 m3/h; (g) installation of a gravity collector (diameter 1,800 mm, length about 8,500 m) to the treatment plant, a lift station and a treated water canal from the treatment plant to the Khelidj canal (length about 6,900 m); and (h) consulting services for design and supervision of construction. Sfax. The project consisted of facilities to improve the sewerage system and to convey sewage from the combined sewerage system to a new treatment plant financed by SIDA. A pumping station was to be rebuilt. - 24 - The project works in Sfax included: (a) rebuilding of the "5 aofit Street" pumping station; (b) combined system: approximately 2,200 m of trunk collec- tors; (c) drainage system: approximately 2,850 m of collectors; (d) sewer system: extensions to service built-up areas without services (about 15,000 m); and (e) consulting services for derign and supervision of construction. Institutional. The technical assistance component consisted of the strengthening of ONAS' project planning and control capabilities. 2.05 At appraisal, the project was expected to cost US$77.5 million, with a foreign component of US$26.5 million, or 34% of total cost. The cost of the technical assistance component was estimated at US$4 million with a foreign component of US$2.6 million. The project was scheduled to be completed by June 30, 1984. Negotiations and Approval 2.06 Negotiations for a Bank loan of US$26.5 million to finance the foreign exchange cost of the project were held in Washington, D.C, from December 15 to 21, 1978. Final agreement was reached on January 15, 1979. Agreement was quickly reached on the project contents and objectives. Discussions took place on the strengthening of ONAS' top management which the appraisal team considered weak as too many and indecisive managers and directors (11) reported to the President. After ONAS' indications of the measures already taken and to be taken to improve this situation, the Bank required a covenant to ensure that future action would be taken (para. 2.08). The terms of the proposed loan were also discussed, with the Tunisian delegation requesting 2, years including four of grace compared to the 17 and four years suggested by the Bank. The Bank made a counter-offer of 19 and four years on the condition that the Government agreed to reduce proportionately terms of one or more loans scheduled for FY79 so that the Bank would be able to remain within the country limits. The Government settled for the original Bank position of 17 and four years. 2.07 The loan was approved by the Board on March 22, 1979. The Board questioned the adequacy of the measures taken to strengthen ONAS' manage- ment, the exemption of ONAS from the policy of full cost recovery, the adequacy of project monitoring arrangements and the advisability of 50% average increases in sewer charges forecast for 1980. The agreement was signed on April 13, 1979 and became effective on August 31, 1979. There was no special condition of effectiveness. - 25 - Loan Covenants 2.08 The following main covenants were set out in the Project Agreement: 1. ONAS should take not later than December 31, 1979, measures required for the strengthening of its structures concerning the management and control of accountancy and financial services (Section 5.04). 2. ONAS should produce annually funds from internal sources (including C overnment subsidies) equivalent to at least 4% of the net value of its fixed assets including works in progress (Section 4.03). In 1983, the Project Agreement for Loan 2255-TUN changed this covenant for one requiring ONAS to produce annually funds from internal sources (including the Government subsidy) equivalent to at least 13% of its average investment for the current and next following years (Section 4.04). 3. ONAS shall not incur debt without the Bank's concurrence unless its aet revenue for a 12-month period shall not be less than 1.5 times the maximum debt service requirements for any succeeding year on all the debt of ONAS. 2.09 The Loan Agreement does not rontain any special covenant. UI. PROJECT IMPLEMENTATION AND COST Effectiveness and Start up 3.01 There were no special conditions of effectiveness and the Loan was declared effective, with a slight delay, on August 31, 1979, about three and one half months after loan signing. 3.02 Simultaneously with the loan approval by the Board, consulting firms were engaged for final engiieering, preparation of bidding documents and for construction supervision. The project start up was timely. Revisions to the Scope of Works 3.03 The following revisions to the scope of works were undertaken with the concurrence of the Bank: 1. Greater Tunis (see map 13846R IBRD) (i) increase of the recalibration of the combined sewer system of the city of Tunis from about 3.5 km to 5.3 km to ensure appropriate operation of the system after construction of the light rail in Tunis; (ii) increase of the extensions to the sewer systems in residential areas from 394 ha to about 575 ha. to keep pace with the rapid urbanization of Tunis; - 26 - (iii) reduction of the drainage systems from 2180 to 867 ha. caused by the slow implementation of road and sidewalk surfacing programs of the municipalities; (iv) construction of three additional pumping stations in Carthago to convey wastewater from the sewer systems to the treatment plant at CotiLre Nord and cancellation of one pumping station which was subsequently financed under a KFAED loan; and (v) construction of an additional force main of 1100 mm. diameter for the transport of treated wastewater from Cherguia to Choutrana. This force main would originally have been financed under loan 1088-TUN. 2. Sfax (see map 13847R IBRD) 'i) cancellation of one collector W which has been replaced by an open canal of two hundred meter length; (ii) modification of the alignment of canal L taking into account new embankment works of the Railway. The Bank did not consider necessary to amend the Loan Agreement as this was a continuation of the First Urban Sewerage Project. 3.04 The implications of the revisions to the scope of works can be summarized as follows: (i) the increase of the recalibration and of the extensions to the sewer systems as well of the additional pumping stations caused a delay in the execution of works but increased the sewer service coverage; (ii) the additional force main which was laid in the same trench as the gravity collector between Cherguia and Choutrana did not influence project implementation; (iii) there were no important adverse effects related to the reduction of the drainage systems, the cancellation of collector W, and the realignment of canal L in Sfax. Implementation Schedule 3.05 Implementation of the physical project components (para 2.04) should have been completed by mid 1984, while in fact all but one were completed after this date as indicated below: Completion date Appraisal estimate Actual Recalibration mid 83 mid 85 Sewer networks mid 84 mid 88 Treatment plant mid 84 mid 86 Gravity collector mid 83 mid 84 Force main and canal 3.06 An analysis of the reasons for the delays in completing the works shows that: - 27 - (i) the recalibration studies took more time than foreseen because the consultants were obliged to restudy the net- works to take into account the impact of the light rail project. Moreover the light rail project also caused delays in the execution of the sewer works which had to be subdivided to give priority to the works in its alignment. These delays were also increased by coordi- nation problems related to other urbaa infrastructure; (ii) the need to increase the service coverage caused un- expected delays in the sewer works. Four additional areas were thus sewered at a later stage between mid 1985 and mid 1988 to Carthago, UV4, Rades and El Aouina. The initial sewer works were completed by mid 1985 or one year after thi original completion date. Moreover the late supply of concrete sewer pipes, which were produced locally in a new factory, caused some delays in project execution; (iii) procurement procedures for the treatment plant, which were based on separate bidding for equipment and civil works, caused some delays because detailed engineering and bidding for the civil works had to be carried out after contract award for the equipment. These procedu- res, which increased competition, caused a delay of about one year. An additional delay of one year was caused by difficulties in obtaining import licences for specialized construction materials for the civil works such as metal forms, pumps, etc; and (iv) the execution of the works for the gravity collector, the force main, and the canal, which were partly constructed outside residential areas, was delayed by one year because of lenghty expropriation procedures. It was also often difficult due to lack of documentation to identify landowners. Reporting 3.07 The reporting consisted of quarterly reports, which were submitted on a regular basis. Performance of Consultants and Contractors 3.08 The engineering consultants who assisted ONAS in the design, bid- ding procedures and supervision of the project were competent, and the basic design criteria and approaches which they proposed were followed. Execution delays were incurred and changes were made to project scope as stated in para. 3.03. A project coordination unit was established with two years assistance of a Bank staff member. This unit, which included also a Tunisian counterpart who continued its activities after the departure of the Bank staff member, was successful in strengthening ONAS project - 28 - planning and control capabilities. During the preoaration of final designs, some components required more time than anticipated, particularly the recalibration component where coordination was required with the light rail project. 3.09 Most of the civil works contracts were awarded to local contrac- tors. These contractors performed well and generally met expected quality standards. For the more specialized works -- such as the combined systems in Tunis, the laying of large collectors between Cherguia and Choutrana and the equipment of the treatment plant -- contracts were awarded to foreign contractors, who performed well. Only one local contractor was not able to complete his contract and ONAS commissioned the execution of part of his works to other contractors. 3.10 All bidding documents were prepared by four consultants with assistance from ONAS for the general portion of the bidding documents. Preparation was timely and Bank response took place within one month of document receipt. Bid evaluation was performed by ONAS with consultants assistance and was completed within a reasonable period of about six months. 3.11 In general, procurement of pipes and equipment, and of pipelaying and related civil works was done separately. Coordination and supervision of the supply and pipelayinj as well as of the civil works contractors and equipment suppliers was carried out appropriately by consultants and ONAS staff in Greater Tunis and CNAS staff in Sfax. Procurement 3.12 Except for some minor project components, which were awarded on the basis of local advertising, all civil works and equipment required for the project were procured on the basis of international competitive bidding in accordance with the Bank's Guidelines for Procuremen'. ONAS had no dif- ficulty in complying with these guidelines and procurement was accomplished in a satisfactory manner. Disbursements 3.13 Because of the implementation delays noted in para. 3.06, actual disbursements lagged considerably behind the appraisal estimate. Estimates foresaw full disbursement of the loan at the end of the fifth year, when actually about 50% of the disbursement had taken place at the time. In all, it took eight years to disburse the loan instead of five as original.y forecast. Annex 1 provides a comparison of actual and forecast disburse- ments. Project Costs 3.14 Actual costs amounted to TD 49.5 million as compared with appraisal estimates of TD 37 million, adjusted to reflect the project revisions noted in para. 3.03. A detailed cost comparison appears in Annex 2. The cost overrun was met by Government equity contributions. It is estimated that of the total net increase of TD 12.5 million, about TD 1.9 million is the result of higher prices due to the fact that roughly - 29 - 50% of the project was completed after the original completion date of June 1984. 3.15 The devaluation of the Tunisian dinar in reference to the expenses in foreign exchange contributed to about TD 2.5 million increase of the original estimate. The remainder of the cost uverrun is due to project revisions. 3.16 Major ;ost variations occurred in the following components: (a) recalibration, the cost of these works was originally estimated at TD 3.2 million at appraisal. Because of the construction of the light rail and because of underestimated construction costs, actual cost of the works increased to TD 8.0 million or about 2.5 times the original estimate; (b) pumping stations and force main, gravity collector and canal, the cost of these works was originally estimated at TD 3.6 milion at appraisal. Due to the additional construc- tion of the force main and some improvement works, actual cost of tnese works increased to TD 8.7 million or about 2.4 times the original estimate; (c) combined system in Sfax, the cost of these works was origi- nally estimated at TD 1.7 million. Because of underestimated construction costs, actual cost increased to TD 2.-,' million or about 1.5 times the original estimate; and (d) stormwater system in Sfax, the cost of these works was origi- nally estimated at TD 1,1 million. Because of underestimated construction costs, actual cost increased to TD 3.7 milion or about 3.4 times the original estimate. An overview of actual versus cost at appraisal of all project components is given in Annex 2. IV. FINANCIAL PERFORMANCE 4.01 ONAS' appraisal and actual income statements, cash flow state- ments, balance sheets, and financial statistics for the period 1978-87 appear in Annexes 3, 4, 5, and 6. Appraisal figures for the Second Urban Sewerage Project cover up to 1984; for the period 1985-87 the figures for the appraisal of the Third Urban Sewerage Project (Loan 2255-TUN) were used. Operating Results 4.02 The appraisal forecast of the volume of water sold by SONEDE, the water supply company, to ONAS' customers for the period 1979-84 was reasonably accurate (just 5% above actual volume). The appraisal forecast for 1985-87 was substantially higher than the actual (23%) due to delays in ONAS' take-over of new systems and a decrease in the per capita consumption as an increasing proportion of new connections are low income, low - 30 - consumption customers. Over the period 1979-87, revenues were lower than expected at appraisal (12%) and so were operating expenses (4%), but their combined effect on net income, which was 86% lower than estimated, was moderated by lower than expected interest expense (17%). Most of the dif- ference in operating expenses was originated in the personnel expense item, which was 9% lower than anticipated due to the double effect of low salary iicreases allowed by Government as part of its program to control inflation and an improvement in ONAS performance (the number of employees per 10,000 connections improved from 8.9 in 1980 to 5.9 in 1987). Consumer Charges and Operating Subsidy 4.03 ONAS maintains an array of consumer charges. In the period 1979-87, about 38% of these charges was linked to the volume of water consumed (appraisal 53%), 28% to property taxes (appraisal 18%), 4% to the cost of new connections (appraisal 6%), 2% to other sources (appraisal 2%), and 28% to Government operating subsidy (appraisal 21%). The first of these charges stems from the sewerage tariff, the evolution of which appears in Annex 7. Sewerage tariffs, which increased ONAS' revenue per m3 by 218% between 1979 and 1987 (65% in real terms), lagged behind those estimated at appraisal (expected increases of 30L% or 137% in real terms). While the appraisal estimated the average revenue per m3 to be at TD 2.16 by 1987, it was only TD 0.11, or 30% lower than expected. The Lag in the average revenue per m3 was partly offset by the maintenance of '-he Govern- ment subsidy to ONAS' operations, which amounted to TD 0.0516 per m3 in 1987 but which should have been replaced by tariff increases by -nd 1)86 (para 4.04). Financial Covenants 4.04 ONAS met its financial covenant in 1979, 1980 and 1982 otly, i.e. it produced funds from internal sources (including Government subsidits) in excess of the prescribed minimum of 4% of its net fixed assets (including work in progress). In 1983, the Project Agreement for Loan 225-TUN changed the financial covenant for a less stringent one requiring ONAS to obtain a self-financing ratio of 13% computed on its average investment for the current and next following years, and including as a source of revenue the (declining) Government subsidies. ONAS met the new covenant in 1983 -85 but failed to meet it in 1986 and 1987 (negative self-financing tatios). If the old covenant would have been kept, ONAS would not have met it in the entire period 1983-87. Loan 2255-TUN introduced another covenant requiring ONAS to replace the Government operating subsidy by 1986, which UNAS has not complied with. Financing Plan 4.05 The 1979-87 investments, totalling TD 173.8 million (appraisal TD 207.6 million), were financed 10% from internal cash generation (inr uding the Government subsidy) and consumers' contributions (appraisal 14%), 50% from Government contributions and grants from the EEC (apprais,l 50 %) and 40% from borrowings (appraisal 43%). - 31 - 4.06 The almost 20% difference betwen ONAS' actual 1979-87 investment expenditures and those expected at appraisal are explained by a financial constraint that ONAS has faced throughout the period. In fact, ONAS' revenues, Government operating subsidies and Government contributions have been insufficient to cover its programmed investments. Confronted with this lack of resources, and to preserve its financial solvency, ONAS has had to resort to (a) cut back its programmed investments; (b) slow down the pace of its construction activities; (c) reduce expenditures needed to maintain its installations with their consequent deterioration; (d) obtain loans from the Government to service a sizeable proportion of its debt (TD 6.0 million Government loan); and (e) pass on to Government the exchange losses it has experienced on most of its loans. TLt ori6in of this situation can be traced back to 1975, when the first tariff that ONAS was allowed to charge its customers was set at the nominal rate of TD 0.008 per m3. The idea at the time was that customers should be charged a small amount for severage services at the initiation of those services and then progressively to raise charges to the level necessary to fully cover costs while at the same time phasing out the Government subsidy which made up the difference between charges and costs. Although this strategy may have seem appropriate at the time, the ONAS experience shows that once a tariff has been set for the first time, Government will regard any increase designed to bring tariffs to adequate levels as it regards any other price increase, which is politically and economically painful and to be carefully dosed or avoided whenever possible. Tariff increases were not granted in 1976, 1977, 1979-81, 1983, 1985 and 1986, that is in 8 out of 14 years of tariff history. ONAS' dependency on Government funds can be gauged by the amount of transfers, both for operating and investment expenditures, which reached TD 123.4 million in the period 1979-87, or more than 70% of total (reduced) investments, of which TD 35.6 million, went to cover 33% of operating expenses. This excessive dependency has affected ONAS negatively as the Government, faced with its own fiscal difficulties, has consistently reduced the budgetary allocations requested by ONAS. Financial Position 4.07 ONAS' overall financial structure has been sound. Its debt to equity ratio increased gradually from the very low level of 13:87 in 1979 to 33:67 in 1987, which is a satisfactory level. Its current ratio has been low, mainly because of an increase in suppliers' credits and delayed payments to the Government of interest on most of international loans repassed by the Government to ONAS, including IBRD loans. Its receivables have steadily increased from a level of about 2.7 months of sales in 1979 to 6.4 months of sales at the end of 1987, which is still an acceptable level, as the increase of receivables is due to ONAS' long-term financing of new connections provided to low-income consumers. - 32 - V. INSTITUTIONAL PERFORMANCE Organization and Management 5.01 ONAS' organization structure, which was modified in 1987, appears in Annex 8. ONAS continues to be ruled by a Board of Directors which represents various interested ministries and entities. Its chairman is also the General Manager and runs the company. At its inception three departments--Finance and Administration, Engineering and New Works and Operations--reported to the General Manager. In 1980, four departments (Planning and Studies, Operations and Maintenance, Projects and New Works-Tunis, and Engineering and New Works-Interior) and two divisions (Administration and Finance), and currently seven departments (Planning and Studies, Engineering and Works, Operations and Maintenance, Greater Tunis Projects, Administration, Finance, and Organization and Audit) report to the General Manager. Personnel and Training 5.02 Annex 9 shows the evolution of ONAS' staff. Total staff stood at 1869, or 5.9 per 10,000 connections at the end of 1987. This is rather low considering the level of education of ONAS' personnel, its decentralized operations and the need for proper maintenance. For 1987, the appraisal forecast a total staff of 2,360 or 7.9 per 10,000 connections. Billing and Collection 5.03 Billing and collection continues to be done by SONEDE. This function is being performed efficiently. Delays in funds transfer from SONEDE, which also does the collection for ONAS, have now been solved. Accounting 5.04 During the period 1975-77, the preparation of accounts was subject to increasing delays and the backlog of %w.ork had increased to such an extent that it had become impossible to determine the results of one year before the end of the next. The Bank's emphasis on remedial action, followed by various measures including the reorganization of the Finance and Administration Department, the recruitment of additional staff, and the hiring of a consulting firm under the First Project and followed-up under the Second Project yielded results and from 1978 onwards, significant improvements were noted in ONAS' accounting function. Its accounts, inventory control and payroll have been computerized and transactions recorded properly. To monitor costs more effectively, ONAS introduced a more sophisticated computer-based system which provides cost information by region, branch and project. Further improvements, among which stand out the finalization of the inventory of fixed assets and improvements to the management of receivables, are under way and ONAS, on the Bank's request, is preparing an action program on these two areas. - 33 - Audits 5.05 Because of delays in the preparation of accounts, audit reports reached the Bank until 1982 with considerable delays. Such delays, how- ever, have been gradually reduced: the audit for 1986 was completed in April 1987 and the audit for 1987 in April 1988. Audit reports have been comprehensive and valuable to ONAS and the Bank. Their coverage has declined considerably in 1985 and more so in 1986 and 1987 as by now the auditor makes no changes tr ONAS-prepared financial statements. The only continuous observation of the auditors concerns the fact that ONAS has not completed the valuation of all its fixed assets. Thi& is bound to continue to happen each time ONAS aasumes responsibility for the sewerage facilities of a new urban center but the backlog work is close to completion. VI. BANK PERFORMANCE 6.01 The Bank was instrumental in the creation of ONAS, and has continued to assist in its development. In particular, the Bank played a key role in convincing ONAS to gradually improve its organization to cope with its growing responsibilities. 6.02 Bank supervision was relatively continuous between 1979 and 1986 where few changes occurred in project officers during project implementa- tion. Near the completion of the project in 1986 and 1987 the Bank's supervision effort was reduced to only one mission per year. This does not appear to have affected project execution. 6.03 The Bank was flexible in agreeing to proposed modifications in the project scope which enhanced the performance of the sewerage systems. 6.04 However, the Bank should have taken a more aggressive role in requesting Government to raise tariffs to compensate for cost increases, to replace the Government subsidy, to allow ONAS to finance its proposed investment program and the maintenance of its facilities VII. PROJECT JUSTIFICATION Institutional Aspects 7.01 An important impact of the Second Urban Sewerage project has been the improved institutional performance of ONAS with technical assistance from the Bank. Also the development of its accounting was remarkable and as a result ONAS can be counted amongst adequately-run organizations which provide services at the least overall cost. Improvements to the Greater Tunis Sewerage System 7.02 The recalibrazion works in the center of Greater Tunis and the sewer and drainage works in Sfax completed the works of the First Urban Sewerage project. As a result, overflows have virtually stopped in the streets and continuous discharges into the receiving waters are eliminated. - 34 - 7.03 As a result of the Second Urban Sewerage project, the wastewaters of the northern suburbs of Greater Tunis are now treated and the waste- waters of Sfax are conveyed to its treatment plant. The effluent of these plants will be used for irrigation, which is very important in a country with limited water resources. Pollution Control of the Lake of Tunis 7.04 The Lake of Tunis, which is divided into a northern and southern area by a navigation canal, has been for years the source of obnoxious odors. These odors disappeared completely in the northern area through the construction of the pollution control works (conveyors, treatment plants and discharge canals) in the First and Second Urban Sewerage projects. A Government created Company "Societe de Promotion du Lac de Tunis" which is owned by the Tunisian Government and the Saudi Al Baraka Investment and Development Co started subsequently in 1984 to develop the northlake shores by dredging part of the lake thus creating about sixteen hundred ha of arable land. The Company started to sell its first lots for housing construction in May 1988. As a result of the dredging works, fishing in the lake was abandoned in 1986 and 1987 but is expected to restart in 1988. As a result of these activities the im,rovements in and around the northern part of the lake are spectacular and specially the water quality of the lake, which showed originally high BOD5 counts and released obnox- ious odors, improved to the quality of bathing %ater. Internal Rate of Return 7.05 The appraisal team computed a rate of return only on tne pollution control works, including main sewer trunks, treatment plants and treated sewage outfalls. These works amoL.nted to about 40% of the combined cost of the First and Second Urban Sewerage Projects. A rate of return on the remainder of the project was not computed as it is not possible to evaluate its benefits accurately since, among others, (i) demand for sewerage services and sewerage charges cannot be established separately from those for water supply, and (ii) public health data do not permit an evaluation of sewerage benefits independently. The same methodology was followed in this report. 7.06 The rate ot' return for the pollution control infrastructure calculated on the same basis as at appraisal is estimated at negative 2% while the appraisal estimate reached 18%. The difference is mostly due to the omission of land development costs in the project appraisal net benefits estimated at appraisal at TD 130.5 million and at TD 19.2 million at the time of preparation of this PCR, and the timing and the period of time necessary to market developed sites (13 years estimated at appraisal, between 1983 and 1995; 19 years estimated at the time of preparation of this PCR, between 1988 and 2002). The estimated rates of return, however, consider only development carried out or to be carried out by the Societe de Promotion du Lac de Tunis. They do not consider other adjacent tracts of land which have already been developed and where a large number of com- mercial and Government buildings have been built. This land, and other tracts being now developed, could not have been used without the project. Also, other non-measurable benefits have accrued in the provision of a pleasant and safer environment for Tunis residents as well as for tourists, tourism being one of the important economic sectors in Tunisa. Annex 10 provides computational details. - 35 - VIII. CONCLUSIONS AND LESSONS LEARNED Conclusions 8.01 The project was successful in most of its major aspects: (a) the national sewerage authority has continued its development into a financially sound (para. 4.07) and adequately-run organization (para. 7.01); (b) the infrastructure provided performs as expected; (paras. 7.02 to 7.04); and (c) a company, owned 51% by Government, is proceeding with plans to urbanize the priority zone on the shores of the Lake of Tunis (para. 7.04). 8.02 The project was not successful in making ONAS a financially inde- pendent public utility that would permit it to embark on its investment program with adequate financial resources (para. 4.06). Lessons Learned 8.03 This project confirms the importance of adequate project prepara- tion. The project was appraised on the basis of feasibility studies without knowing later urban developments in Greater Tunis and Sfax. This resulted in revisions to the project scope during implementation (para. 3.03). 8.04 Drainage works consisting of primary and secondary collectors should not be carried out where road and sidewalks are not surfaced (para. 3.03). 8.05 It appears that the time required for execution of sewer and drainage works in fully developed urban sites where coordination problems occur with other urban infrastructure might have been more realistically reflected in the project implementation schedule (para. 3.06). 8.06 The experience with ONAS, since its creation, highlights the fact that unless charges for sewerage services are realistically assessed since the beginning, the utility is bound to face a financial constraint that will limit its ability to cope with in:reasing demand for its services. Alternatively, the lesson learned indicates that public utilities should not finance part of its operations with Government subsidies, as it is dif- ficult to replace them by adequate tariffs (para. 4.06). 8.07 The experience with ONAS is also a good example of the maintenance deficiencies found in many countries (paras. 4.06, 5.02). Part of these deficiencies may be attributed to the economic juncture (mostly fiscal and balance of payments deficits). Most importantly they are the consequence of disregard of the long-term effects of inadequate maintenance practices and of failure to highlight maintenance expenditures, including in foreign - 36 - exchange, at the time of investment decision. In this sense, the Bank could improve upon this situation by emphasizing in the SARs and during supervision the need f;r proper maintenance practices and the associated costs. - 37 - ANNEX 1 REPUBLIC OF TUNISIA SECOND URBAN SEWERAGE PROJECT (LOAN 1675-TUN) PROJECT COMPLETION REPORT LOAN DISBURSEMENTS (US$ '000) Bank Fiscal Appraisal Estimate Actual Year Annual Cumulative Annual Cumulative 1980 December 31, 1979 200 200 June 30, 1980 300 500 1981 December 31, 1980 900 1,400 100 100 June 30, 1981 3,000 4,400 570 670 1982 December 31, 1981 3,400 7,500 470 1,140 June 30, 1982 4,500 12,000 1,700 2,840 1983 December 31, 1982 4,400 18,400 3,840 6,680 June 30, 1983 4,200 20,600 1,890 8,570 1984 December 31, 1983 4,100 24,700 1,720 10,290 June 30, 1984 1,800 26,500 1,150 11,440 1985 December 31, 1984 2,170 13,610 June 30, 1985 1,420 15,030 1986 December 31, 1985 2,930 17,960 June 30, 1986 1,670 19,630 1987 December 31, 1986 710 20,340 June 30, 1987 1,550 21,890 1988 December 31, 1987 4,610 26,500 - 38 - REPUBLIC OF TUNISIA SECOND URBAN SEWERAGE PROJECT (LOAN 1675-TUNI PROJECT COMPLETION REPORT PROJECT COSTS (TO '000) Appraisal Increase Increase Estimatesi/ Actual (Decrease) (Decrease) I. IUNI A. Sewer System Recalibration of Combined System 3,213 7,990 4,777 149 B. Sewer and Stormwater System Sewers, Collectors and Urainage 11,728 15.019 3.291 28 C. Sewage Treatment Pumping Station, Force Main, Canal to Plant and Outfall 3,621 8,717 5,096 140 Treatment Plant 8,834 9,361 527 6 TOTAL TUNIS 27,396 41,087 13,691 50 II. SFAX Combined System 1,677 2,462 785 47 Stormwater System 1,069 3.687 2,618 144 Sewers 435 484 49 11 TOTAL SFAX 3,181 6,633 3,452 108 TOTAL TUNIS & SFAX 30,577 47,720 17,143 56 III.CONSULTANTS' SERVICE 2,360 1.780 (580) 25 TOTAL PROJECT COSTS 32,937 49,500 16,563 50 " Includes Physical and Price Contingencies TUNISIA PROJECT COMPLETION REPORI SECOnu URBAN SEWERA6E PROJECT (LOAN 1675-TUN) Actual and Forecast Income Statements of OKAS - 1978-1987 * ln 8llions of4 D1 Fisra, Year 9 7 B 1 9 7 9 1 9 8 o 19 8 1 1 9 8 2 1 9 8 3 1 9 8 4 L 9 9 5 1 9 8 6 l 9 8 7 1979-1983 1979-1987 Fort. Act. Fort. Act. Fort. Act. Fort. Act. Fort. Act. Fort. Act. Forc. Act. Fort. Act, Forc. Act. Forc. Act. Forc. Act. Forc. Act. Revenues: Sewerage charges 1.0 0.8 2.4 1.8 2.7 2.0 4.8 2.2 5.3 4.5 6.5 5.6 7.4 5.9 12.0 8.1 15.3 8.4 19.2 9.8 21.7 16.1 75.6 48.3 ^onnections and other works 0.2 0.1 0.3 0.1 0.5 0.2 6.8 0.5 1.0 0.4 1.2 0.7 1.3 0.5 0.9 9.9 1.0 0.6 1.5 0.6 3.8 1.9 8.5 4.5 Sales o4 by-products 0.0 0.2 0.0 0.1 0.1 0.1 0.1 0.3 0.1 0.4 0.2 (.4 0.2 0.3 0.9 0.4 1.1 0.5 1.3 0.5 0.5 1.3 4.0 3.0 Municipalltv contributions 1.6 2.0 1.7 2.0 1.8 2.4 1.9 2.8 2.0 3.4 2.1 4.2 2.2 4.8 4.4 5.6 4.8 5.3 5.3 5.4 9.5 14.8 26.2 35.9 Governcent subsidy 2.9 2.9 3.1 3.2 3.7 3.4 4.1 3.6 4.5 3.8 5.0 4.0 5.3 4.0 2.5 4.3 1.5 4.6 0.0 4.7 20.4 18.0 29.7 35.6 Tot l revenues 5.7 6.0 7.5 7.2 8.8 8.1 11.7 9.4 12.9 12,5 15.0 14.9 16.4 15.5 20.7 19.3 23.7 19.4 27.3 21.0 55.9 52.1 144.0 127.3 Operating expenses: Payroll 1.9 1.6 2.3 2.0 2.8 2.4 3.4 3.1 4.0 3.7 4.4 4.2 4.8 5.5 6.0 5.9 7.2 6.3 8.6 6.4 16.9 15.4 43.5 39.5 Electricity 0.2 0.2 0.3 0.4 0.3 0.4 0.5 0.6 0.6 0.9 0.6 1.6 0.7 1.2 2.1 1.4 2.5 1.6 2.7 2.1 2.3 3.9 10.3 10.2 gaterials 0.3 0.4 (k.5 0.4 0.8 0.9 0.9 1.0 1.0 i-0 1.2 1.1 1.3 1.7 1.6 1.5 1.8 1.7 2.1 2.1 4.4 4.4 11.2 11.4 Depreciation 1.4 1.4 1.8 1.7 2.2 2.4 2.9 2.6 3.9 3.5 4.3 4.0 5.0 4.7 5.9 5.3 6.6 6.2 7.4 7.4 15.1 14.2 39.9 37.8 Taxes and doubtful debts 0.3 0.5 0.4 0.5 0.4 0.5 0.5 0.8 0.6 1.0 .7 1.0 0.@ 0.9 1.1 1.5 1.3 1.5 i.5 1.5 2.6 3.8 7.3 9.2 Totjl operating expenses 4.1 4.1 5.3 5.0 6.5 6.6 8.2 8.1 10.1 10.' 11.2 11.9 12.6 14.0 16.6 15.6 19.4 17.3 22.3 19.5 41.3 41.7 112.2 108.1 Net operating Income 1.6 1.9 2.2 2.2 2.3 1.5 3.5 1.3 2.8 2.4 3.8 -.0 3.8 1.5 4.1 3.7 4.3 2.1 5.0 1.5 14.6 10.4 31.8 19.2 Interest 0.3 0.2 1.6 0.4 1.6 0.7 1.7 1.2 2.0 2.0 2.2 2.5 2.4 3.4 3.8 3.1 4.3 3.4 4.9 3.6 9.1 6.8 24.5 20.3 Cthers 0.0 (0.4) 0.O 40.1> 0.0 9.1) 0.0 10.11 0.0 (1.0) 1.0 1.4 0.0 k2.0) 0.0 (0.3) 0.0 (0.11 0.0 0.3 1.0 0.1 1.0 (2.0) get income 1.3 2.1 0.6 1.9 0.7 0.9 1.8 0.2 0.8 1.4 0.6 0, 91 1.4 0.1 0.3 0.9 0.0 (1.2) 0.1 (2.4) 4.5 3.5 6.4 0.9 Rate of return on average net fiked a%sets iV' 6.0 12.5 4.6 5.3 3.6 2.6 4.4 1.8 2.9 2.2 3.6 2.9 3.5 1.3 3.1 2.9 2.5 1.5 2.5 1.0 f 1978-84 foarecast figures iere taken from the SAR for the Second Urban Sewerage Project, thrse for 1985-87 from the SAR for the Third Urban Semerage Project. 27-3u1-8 TUNISIA PROJECT CONPLETION REPORT SECOND URBAN SElERAGE PROJECT (LOAN 1675-TUNI Actual and Forecast Flow of Funds of NAS - 1978-197 4 (in millions of OTI FiscalYear 1978 1979 1980 1991 1982 1983 1984 1985 1996 1967 1979-1963 1979-1987 Forc. Act. Forc. Act. Forc. Act. Forc. Act. Forc. Act. Forc. Act. Forc. Act. Forc. Act. Forc. Act. Fort. Act. Forc. Act. Fort. Act. SOURCES Cash flow fro operations: Aet operating income 1.6 1.9 2.2 2.2 2.3 1.5 3.5 1.3 2.8 2.4 3.8 3.0 3.4 1.5 4.1 3.7 4.1 2.1 5.0 1.5 14.6 10.4 31.4 19.2 Depreciation 1.4 1.4 1.8 1.7 2.2 2.4 2.9 2.6 3.9 1.5 4.3 4.0 5.1 4.7 5.8 5.3 6.6 6.2 7.4 7.4 15.1 14.2 40.0 37.8 Decrease (increase) wkg.cap. 1.2 2.6 £0.1) 0.3 0.0 2.4 41.5) 0.7 0.1 1.7 40.8) 0.4 0.2 2.8 (0.9) 0.9 10.5) (3.41 2.5 (12.2) (2.3) 5.5 1.0) (6.4) Cash flow from operations 4.2 5.9 3.9 4.2 4.5 6.3 4.9 4.6 6.8 7.6 7.3 7.4 B.7 9.0 9.0 9.9 10.4 4.9 14.9 13.3) 27.4 30.1 70.4 50.6 Government contribs. and grant 7.2 2.0 7.2 6.2 4.1 7.8 6.6 3.4 6.4 5.3 4.3 8.7 11.6 8.9 12.0 14.4 14.5 13.5 13.5 19.6 28.6 31.4 P0.2 87.8 Customer contributions 0.0 0.0 0.0 0.3 0.0 0.0 0.0 0.3 0.0 0.0 0.0 0.3 0.0 0.4 1.8 0.5 2.3 0.4 3.2 0.5 0.0 0.9 7.3 2.7 Long-term borrowings 4.7 2.4 7.2 3.3 10.8 4.7 9.5 9.6 6.5 8.2 5.0 7.6 12.7 6.3 13.6 5.8 11.9 4.5 12.5 19.8 39.0 33.3 89.7 68.7 Total Sources 16.1 10.3 18.3 13.9 19.4 28.8 21.0 17.9 19.7 21.1 16.6 23.9 33.0 24.7 36.4 30.6 39.1 23.3 44.1 35.6 95.0 95.7 247.6 209.8 USES Debt service 0.3 0.4 2.0 1.0 2.0 1.6 2.1 2.4 2.6 3.2 2.8 4.5 6.3 6.3 6.0 6.2 7.4 7.6 9.3 7.9 11.5 12.6 41.3 40.6 Capital erpenditures 17.1 6.4 16.1 19.9 19.7 16.0 19.3 18.0 17.0 18.5 12.8 19.1 23.6 18.6 30.8 18.2 29.2 13.8 39.1 31.7 84.9 91.5 207.6 173.6 Total uses 17.4 6.6 18.1 20.9 21.7 17.6 21.4 20.4 19.6 21.7 15.6 23.6 29.9 24.8 37.6 24.5 36.6 21.4 48.4 39.6 96.4 104.1 246.9 214.4 Net cash flow (1.31 3.7 0.2 (7.0) (2.31 1.2 (0.4) (2.51 0.1 (0.6) 1.0 0.4 3.1 (0.12 (1.2) 6.1 2.5 2.9 24.31 (4.0) (1.41 (3.51 (1.3 14.6) Deb' service coverc,* (times) 14.0 14.3 2.0 4.1 2.3 4.1 2.3 2.0 2.6 2.4 2.6 1.7 1.4 1.4 1.3 1.6 1.4 0.6 1.6 (0.4 2.4 2.4 1.7 1.2 Annual self-financing -atirl1) 23 88 12 17 13 30 15 14 25 24 35 17 10 17 13 23 18 (17) 23 (34) 0 0 0 0 Total Government funds received (TD millions) 10.1 4.9 10.3 9.4 7.8 11.2 10.7 7.0 10.9 9.1 9.3 12.7 16.9 12.9 14.5 18.7 16.0 18.1 13.5 24.3 49.0 49.4 109.9 123.4 x Total Govnet.funds recvd.as 1 of capital expeneitures 59 80 64 47 40 70 55 39 64 49 73 67 72 70 47 103 55 131 35 77 58 54 53 71 * 1978-83 forecast figures were taken from the SAP for the Second Urban Sewerage Project; those for 1984-87 from the SAP for the Third Urban Sewerage Project. ** The item Capital Contributions includes the figures for Grants received from the EEC. CM4- Mill TUNISIA PROJECT CONPLETtOl REPORT SECOND UIM SEERAGE PROJECT (LOM 1675-TUN, Actual and Forecast Balance Sheets of DMAS - 1978-197 8 (to millions of DTI YereadinqDecesaer3l 197 12979 19 0 1931 2992 1983 1964 1985 1986 198' Fort. Act. Fort. Act, Fort. Act. Forc. Act. Fort. Act. Forc. Act. Fort. Act. Fore, Act. rorc. Act. Fort. Act. ASSETS Fixed assets: At cost 41.5 26.2 54.1 31.8 6'.9 39.9 96.2 64.9 0.o 74.9 128.3 84,Q 130.3 1l0.0 162.1 220.0 191.0 13.3 221.0 169.6 Less: usulative dereciation 5.1 5.4 6.9 '.1 9.1 9.5 12.0 12.1 15.q 15.6 20.2 20.0 25.2 24.9 31.0 30.2 37.6 35.9 45.1 43.5 Net fixed assets in operation 36.4 20.8 47.2 24.' 58.8 30.4 94.2 12.8 92.7 59.3 10.1 64.9 105.1 85.1 131.1 89.8 153.4 102.4 176.0 126.1 Construction in progress 3.9 12.0 7.4 26.3 13.3 34,1 4.3 26.9 8.9 35.4 2.0 44,0 0.0 1'.4 31.0 45.4 30.7 4(.3 38.7 40.9 Total net fixed assets 40.3 32.8 54.6 51.0 72.1 64.5 88.5 79.' 101.6 94.' 110.1 08.9 105.1 122.5 162.1 135.2 184.1 142.7 214.7 167.0 Otht fired assets 0.0 0.4 0.0 0.4 'A 0 .5 0." 0.8 0." 1.4 0.0 1.9 0.0 2.2 3.2 2.4 3.8 2.5 4.9 2.5 Current assets: Stock 0.2 0.3 0.2 0.6 .3 0.0 0.3 0.8 0.4 1." 0.4 1,4 Q.5 1.6 .9 2.5 1.1 1.? 1.2 2.5 Accounts receivable 0.2 0.4 0.4 0.5 0,5 0.7 ".3 1I0 0.9 1.5 1.0 2.3 1.5 2.9 2.5 5.4 3.1 4,8 3.8 5.8 Other receivable 0.6 0.2 0.9 0.8 1.0 1.3 1.5 1.8 2.1 2.1 2.3 4.0 2.5 7.0 3.3 ,.9 4.1 6,3 5.2 1.3 Cash and equivalents 3.9 9.0 4.1 2.1 1.8 !.2 1.4 0.8 1.5 0.2 2.5 0.3 5.0 0.1 4.0 6.1 6. 7.8 2.3 3.8 Total current assets 4.9 0.9 5.6 4.0 3.6 5.8 I.o 4.4 4.9 4.8 6.2 9.0 Q,5 11.6 10. 19.Q 14, 2.6 12.5 22.4 Debt revaluationlt 0.0 0,v 0. 0.0 0.0 0.0 0.) 2.7 0.0 n 1 .0 12.1 0. 19. '2 1i.1 0 15.1 U. 11t8 Total assets 45.2 43.1 60.2 55.4 '5.7 I). a 92.5 87.6 106.5 P,'7. 116,3 1 3 9 114.6 1;4,' 1'.0 168.6 2o2.8 190.o 232.1 20:.7 EQUITY AWE LIABILITIES Capital and reserves: Capital and grants 118 32.1 31.7 39.3 38,1 43.4 45.9 5,.t 44.6 56.4 54.9 60.7 64, i O.' 71,9 90., 87.2 106.9 100.0 123.5 117,1 Retained earnings 4.0 2.2 4.6 4,3 5.3 5.4 7.1 5.0 7.9 5.2 q.5 4.8 10.9 6.1 8.1 6.e lu.1 6.4 12.2 5.6 Total capital and reserves 36.1 33,9 43.9 42.4 48.7 51.3 57.1 54.6 64.2 60.1 '0.2 68.8 71.6 78.2 09.1 93.8 116.9 10.4 135.7 122.7 Long-term debt 6.1 3.8 12.9 6.2 23.3 9.7 32.4 21.6 38.3 34.1 42.7 44.0 42.0 53.3 65.q 49.0 '2.8 52.0 79.9 60.4 Current loabilities: Short-term at.of L-T debt 0.4 0.2 0.4 0.9 o.4 1.2 0.4 :.3 ".6 2.: 1.5 3.0 1.0 2.q 3.1 2.0 4.4 4.2 6.7 6.5 Suppliers 1.2 3.8 1.4 4.6 1.7 6.2 1.6 '.2 2.0 6.1 0.5 6.e 0.0 6.2 7.9 6. '.5 5.2 9.9 6.0 Other current liabilities 1.4 1.4 1.6 1.3 1.6 2.4 1." 2.9 1.3 5.4 1.4 8.5 fI,V 13.7 1.0 16.5 1.2 13.1 1.3 7.1 Total current liabilities 3.0 5.4 3.4 6.8 3,7 9.8 7.0 11.4 3.9 13.' 7.4 18.1 1.0 22.8 i2.0 25.8 13.1 22.5 17.1 19.6 Total equity and liabilities 45.2 43.1 60.2 55.4 75.7 70.8 92.5 8'.6 106.5 107.0 116.3 130.Q 114.6 154.3 176.V 168.6 202.9 180.9 232.1 202.7 Debt as a percentage of capital+reserves+debt 14 1 23 13 32 16 36 20 !7 3v 7 39 3' 41 4", 34 38 33 17 33 8 1978-84 forecast figures were taken from the SAP for the Second Urban Sewerage Project: those for 1995-87 frog the SAP for the Third Urban Sewerage Project. IS ONAS charged the revaluation of debt denominated in foreign currencies directly to losses in 1978 and 1980: valued its debt at the historical echange rates in 1979. Since 1981, it charges a pfrovision for losses witth the changes in the eschanee rates for debt maturing within one year, and to assets those for debt maturing in more than one Year. II8 The item Capital includes the figures for Grants recei-ed from the EEC. - 42 - ~C CREIØom 9%SECT CW.flS M ET mcm u ~M ma<CT 4PO 1473-10) Statittics . 197w-1917 197 1979 1930 191l 1'Ø2 1933 1934 1983 2994 11 i' Forc. Act. Forc. kt. Fort. kt. Firc. Act. Forc. kt. Forc. Act. Forc. ct. Forc. Act. Fc. kt. FOrc. Ac'. Suae f4low (lillin 83) m.00 37.00 å0.00 33.15 63.0Ø 63.64 69.00 67,04 76.00 72.93 1360 79.40 9.06 81.18 9.71 9',31 ve.o 8@.88 !!Q., Q' lisaber øf 11custrs(thousands) 91.0 112.9 100.0 131.3 119,0 147.1 140.0 163.5 163.0 132.1 193.6 203.0 213.6 240.0 2!'.3 2a. .e. 2Q.' '34 Inflatiom infex ract.1972-100100.0 100.0 106.0 107.7 112.4 118.3 119.1 129.1 126.2 146.7 33.8 159.9 141.9 173,2 l3.2 18.1 i 1 I A'eralt »e«ral# charge ITO per 031 0.017 0.014 0.040 0.034 0.043 0.031 0.070 0.033 0.070 0.062 0.077 0.0't ?.062 o.,'3 Q N20 . vg,eraecNarges in 19,8 pricet YD per e3p 0.017 ).014 0.038 0,031 0.038 0.27 0.039 0025 )05 0.042 0.35" 0.043 .03 '.4: 74 0 9 4 Aeraw total r eveu 0 p# a, )0.09 M .305 0.125 1.135 0.14' 0.127 0,1' .'40 0.1'; 0.171 .' 0.0 : 190 19-0 1' A,1e.totil re'efue i, 970 Ivices TD Øer a l .09 0 13 ),Ile 0.12 0.!24 . l .142 2.119 9.134 0.11' 0.134 0 119 3.29 .:',:12 A'erige cish OPerating :Ost ; Te ger e31 c046 "B ,.0 62 .00 . 0.0 " 0 0)ó 0.0- 0."0 0.082 n00 0.nel 0V >94 14 li' A'je.cå5h oerating cost in 1978 prie fI9 r *3 *0.04> .4 .6 1' '#234 0,,4 ' .,04 A.C6 .,.~. .O ' 6 . . . Average opat,gq .: e, 70 per .3! .'0. .. A,g9.opprat:n9 c.ost 2Q' 0,(e IT,D238 Q44 pe r 0 ic 39.0 14 >7 11 lS 4 >0. 4 l 35 4' Rael te e.av#,&QP net +!,ed as5 Se %1 6.8 ".q :A1 3e 2e btt 14 h 2: 33 3 t 3e 28 3' 36 3e 39 3' 4 4 Oeht sgr.Ice coveråge 7tie' t4.0 14.1 _2 4. 2.3 4.1 2.! ,.4 3,4 i.' i IleT Irteral C.sh generation ti allonW Ln 5..4 lQ 3,3 25 4. 2.8 2.6 4.2 4,4 4.3 3.2 2.4 7.2 4 4 . 9 Net internål rish g@eation inc.cust.coniex,h'7.,ue. 'ID m illions' 3.' 2.5 ' .2' .; 3. 1 l.3 -.3 3,01 '0.31 :.6 9.51 ,, .291 -.l • ' .. 2 n'ritefit , 1978 p,lces 'Te alllbonil 17.1 b.2 t. 18.5 l'.3 133L .2 13,0 3L. 32, ., . !> in . 2 C o I b 0,b Annual sti-lfinincing ic.o,t. %uh.,cut.contrib. I1) 23 n0 32 7 33 30 13 14 2N 24 33 ' l' n 27 '4 A490.04elf-1n4ncin inc.loit. su4. ,cuit.contrib. !>ourrnt plus art follouiNg year;u3it 27 42 il 19 13 28 13 14 20 23 23 !' 90 ; 2t . 4 Internal cash generation Incl.sLbsidy at 1 of 4,ge.total net #tied asseti 12.. 19., L L. 31 6.6 2.9 2.o 4." 4 . 4.2 2.> .0 2.2 7 2.t . 2.' Total GOvernmfnt funds receied (TO iin.,H i 1) 10.J 4.0 10. 0,4 7.8 !l.2 10.' '.0 10.9 9.1 0,3 32.' it.Q 12. 14, j@,' le.' 0 275 24. S0'ern~ut fund% rctiled al I af InvstAeit 39 90 b4 47 40 70 33 39 64 49 73 67 '2 7" 4'313 35133 35 3 Colliction perlod (months) 2.0 3.3 1.8 2.7 1.9 3.8 1.' 4.0 1,' 3,7 !.b 4.2 2.i L1 2. r o.e L. : .0 4 Liquidity ratio lcorreut allets/curren! liabilities! 1,9 1.9 1.9 0.7 1.1 0.7 1.5 n.4 1.5 0.4 3.3 0.5 0.J 02.6 I.2 0.9 3.' I . , ., TUNI1A1 PROJECT COMPLETION REPORT SECOND URBAN SEWERAGE PROJECT (LOAN 1675-TUN Evolution of ONAS' Tariffs Qreef_of 3/29/75 Decree of 11/7/78 Decree of 2/26/82 D2cree of 10/2/82 Decree of 11/7184 Decree of 12/27/86 pef Cnsume Fixed Charge Variable Charge F E V V y V m3 per quarter) ITD/0uarter) D/m11) -on polluting: 0-20 - - 0.75 0.75 - 0.75 - 0.75 - 0.85 - 21-40 - - 0.75 - 0.75 0.010 0.75 0.020 0.75 0.025 0.85 0.030 41-70 - 0.008 0.75 0.020 1.00 0.040 1.00 0.045 1.50 0.060 1.60 0.065 greater than 70 - J.008 0.75 0.025 2.00 0.060 2.00 0.065 2.50 0.090 2.90 0.105 Industrial: not connected - 0.015 0.75 0.025 2.00 0.060 2.00 0.065 2.50 0.090 2.90 0.105 weak pollution 0.008 0.75 0.025 2.00 0.060 2.00 0.065 2.50 0.090 2.90 0.105 medium pollution - 0.023 0.75 0.035 2.00 0.075 2.00 0.080 2.50 0.120 2.90 0.140 high pollution - 0.023 0.75 0.050 2.00 0.100 2.00 0.105 2.50 0.150 2.90 0.170 'uurism - 0.0585 0.7S 0.085 2.00 0.140 2.00 0.150 2.50 0.180 2.90 0.200 M TJNISIA SECOND URBAN SEWERAGE PROJECT COMPLUTION REPORT OMce National dAssainIvsement (ONAS) Organization Chart 1988 4cm... I,,.,IFIZI II J '--3 c ---a234 E~SoE - 45 - REPUBLIC OF TUNISIA SECOND URBAN SEWERAGE PROJECT ILOAN 1675-TUN) PROJECT COMPLETION REPORT Evolution of ONAS' Personnel by Categaor 12m 1"IU 12m 12m 128 12m 128 12m Managers 70 71 77 89 100 113 113 115 Professionals 299 318 320 331 340 343 348 427 workers jjU 1.J6I8 1i 12.J 12.4 120 1.241 12Z Total 1.308 1.457 1,552 1,681 1,734 ',776 1,702 1,869 Number f Emploeiees who he.vg -ndergone Training im& 1.981 2u 1im 19IL4 128 198 128Z 91 40 27 138 100 182 392 433 Total 1,403 - 46 - ANNEX 10 Page 1 REPUBLIC OF TUNISIA SECOND URBAN SEWERAGE PROJECT (LOAN 1675-TUN) PROJECT COMPLETION REPORT Pollution Control Internal Rate of Return Economic Analysis 1. The economic analysis covers the period 1977 through 2005. Costs and benefits, expressed in TD of 1978, are detailed on page 3 of this annex. Incremental Costs 2. Costs include the costs net of taxes and duties of actual and future investments and ONAS' incremental operating and maintenance expenses related to the pollution control elements of the project. Incremental Benefits 3. Benefits consist of the appreciation in the value of the land sur- rounding the Lake of Tunis, being developed by SPLT, net of investments necessary to bring it to market, and the increased value of commercial fish production. The value of treated wastewater for irrigation, sludge for fertilizer and the increased value of assets in the city of Tunis boarding the Lake has not been taken into account. 4. Although the entire city of Tunis has benefited from the clean-up of the Lake and of the beaches of the Gulf of Tunis it is not nossible to separate and quantify the effects of the project from other factors to determine the increase in property value which took place in Tunis and would be related to the project without a very detailed and costly study. The benefits relating to increased property values have been calculated for an area of 1,600 ha surrounding the Lake on the basis of a conservative cash flow study prepared by SPLT in 1987 which takes into account (i) investments in land development, (ii) marketing and operating expenses and (iii) revenues from sales of land. 5. Before the project, commercial fishing yielded equal tonnage from the northern and southern parts of the Lake of Tunis, but because of increasing pollution, this tonnage was constantly decreasing. After the project, in 1982, the fish yield in the northern part started to increase while that of the southern part continued decrease. The difference in fish production between the two parts of the Lake is attributed to the project. In 1985 and 1986, there was no fishing but fishing was resumed in 1987 and is expected to reach the original yield in the near future. - 47 - ANNEX 10 Page 2 Rate of Return 6. The rate of return for the pollution control infrastructure is estimated at minus 2%. Sensitivity analysis shows that if the benefits from appreciation in land value increased by 30%, then the project's costs and benefits would break even. A doubling of benefits would raise the ROR to a marginal vpl-ue of 0.1%. REPUBLIC OF TUNISIA SECOND URBAN SEWERAGE PROJECT (LOAN 1675-TUN) PROJECT COMPLETION REPORT Pollution Control Internal Rate of Return Costs (TO million of 1978 Benefits Investments Fish Production Net IAas 1st Proiect 2nd Proiect Oeratina Iotal Land Values With Prciect Without Proiect ML IOlow Cast/Somfits 1976 1977 1.3 1.3 0.4 0.4 (1.3) 1978 0.8 0.8 0.4 0.4 (0.8) 1 v79 1.4 1.4 0.4 0.4 (1.4) 1980 2.1 0.8 2.9 0.4 0.4 (2.9) 1981 2.1 1.5 3.6 0.4 0.3 0.1 0.1 (3.5) 1982 1.6 2.7 4.3 0.4 0.3 0.1 0.1 (4.2) 1983 0.4 1.3 0.3 2.0 0.4 0.2 0.2 0.2 (1.8) 1984 0.1 2.0 0.3 2.4 (8.3) 0.4 0.1 0.3 (8.0) (10.4) 1985 1.6 0.3 1.9 (5.4) - - (5.4) (7.3) 1986 1.0 0.4 1.4 (5.4) - - (5.4) (6.8) 1987 0.2 0.5 0.7 (18.7) 0.4 0.4 (18.3) (19.0) 1988 0.5 0.5 (9.3) 0.5 0.5 (8.8) (9.3) 1989 0.5 0.5 3.2 1.1 1.1 4.3 3.8 1 1990 0.5 0.5 (1.5) 1.2 1.2 (0.3) (0.8) 1991 0.5 0.5 2.3 1.3 1.3 3.6 3.1 co 1992 0.5 0.5 7.5 1.5 1.5 9.0 8.5 1993 0.5 0.5 6.1 1.6 1.6 7.7 7.2 1994 0.5 0.5 5.6 1.8 1.8 7.4 6.9 1995 0.5 0.5 4.8 2.0 2.0 6.8 6.3 1996 0.5 0.5 3.7 2.0 2.0 S.7 5.2 1997 . 0.5 0.5 6.1 2.0 2.0 8.1 7.6 1998 0.5 0.5 6.4 2.0 2.0 8.4 7.9 !999 1.1 0.5 1.6 6.8 2.0 2.0 8.8 7.2 2000 0.5 0.5 7.2 2.0 2.0 9.2 8.7 2001 0.5 0.5 2.0 2.0 2.0 4.0 3.5 2002 5.2 0.5 5.7 6.1 2.0 2.0 8.1 2.4 2003 0.5 0.5 2.0 2.0 2.0 1.5 2004 0.5 0.5 2.0 2.0 2.0 1.5 2005 0.5 0.5 2.0 2.0 2.0 1.5 IRR = (2%) 0 AA 9O <b' z -. Oz -i 00 Ø Z'b 0 () 'be 叩 IBRD 13847R NW TUNISIA T u N l s l Å SECOND URBAN SEWERAGE PROJECT SFAX '4. Combined S~m Sto~a~ N~ Force ~ni PurnIping Stanomi Re~ ~ Sa* N~ork "-up ~B sak sa" 0 2.5 oý 10 AlLES 015 KiLOMETERS JUNE 1988