Document of The World Bank

Report No: 27171

IMPLEMENTATION COMPLETION REPORT(CPL-37820 SCL-3782A SCPM-3782S CPL-37830)

ON A

LOAN

IN THE AMOUNT OF US$29 MILLION

TO THE

REPUBLIC OF TUNISIAAND

A LOANIN THE AMOUNT OF US$29 MILLION

TOSOCIETE NATIONALE D'EXPLOITATION ET DE DISTRIBUTION D'EAU

WITH THE GUARANTEE OF THE REPUBLIC OF TUNISIA

FOR A

WATER SUPPLY AND SEWERAGE PROJECT

December 22, 2003

Finance, Private Sector and Infrastructure GroupMiddle East and North Africa Region

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

CURRENCY EQUIVALENTS

(Exchange Rate Effective June 30, 2003)

Currency Unit = Tunisian Dinar (TD) TD 1 = US$ 0.7846US$ 1 = TD 1.2745

FISCAL YEARJanuary 1 - December 31

ABBREVIATIONS AND ACRONYMS

CAS Country Assistance StrategyGIS Geographic Information SystemICB International Competitive BiddingICR Implementation Completion ReportMENA Middle-East and North AfricaNCB National Competitive BiddingNDF Nordic Development FundNIB Nordic Investment BankONAS Office National d'AssainissementPSP Private Sector ParticipationSAR Staff Appraisal ReportSONEDE Société Nationale d'Exploitation et de Distribution des EauxUFW Unaccounted For WaterWTP Water Treatment PlantWWTP Wastewater Treatment Plant

Vice President: Christiaan J. PoortmanCountry Director: Theodore O. Ahlers

Sector Manager/Director: Emmanuel Forestier Task Team Leader/Task Manager: Ines Fraile

TUNISIAWATER SUPPLY AND SEWERAGE PROJECT

CONTENTS

Page No.1. Project Data 12. Principal Performance Ratings 13. Assessment of Development Objective and Design, and of Quality at Entry 24. Achievement of Objective and Outputs 65. Major Factors Affecting Implementation and Outcome 126. Sustainability 137. Bank and Borrower Performance 148. Lessons Learned 169. Partner Comments 1710. Additional Information 21Annex 1. Key Performance Indicators/Log Frame Matrix 22Annex 2. Project Costs and Financing 26Annex 3. Economic Costs and Benefits 30Annex 4. Bank Inputs 37Annex 5. Ratings for Achievement of Objectives/Outputs of Components 39Annex 6. Ratings of Bank and Borrower Performance 40Annex 7. List of Supporting Documents 41Annex 8. Borrower's Contribution to ICR 42

Project ID: P005680 Project Name: TN-WATER SUPPLY AND SEWERAGE

Team Leader: Ines Fraile TL Unit: MNSIFICR Type: Core ICR Report Date: December 22, 2003

1. Project DataName: TN-WATER SUPPLY AND SEWERAGE L/C/TF Number: CPL-37820; SCL-3782A;

SCPM-3782S; CPL-37830Country/Department: TUNISIA Region: Middle East and North

Africa Region

Sector/subsector: Water Supply (75%); Sewerage (25%)Theme: Pollution management and environmental health (P); Rural services

and infrastructure (P); Access to urban services for the poor (P); Water resource management (S); Public expenditure, financial management and procurement (S)

KEY DATES Original Revised/ActualPCD: 02/15/1993 Effective: 01/18/1995

Appraisal: 12/05/1993 MTR: 11/08/1998Approval: 07/28/1994 Closing: 06/30/2002 06/30/2003

Borrower/Implementing Agency: GOVERNMENT/SONEDE/SONEDE/ONASOther Partners:

STAFF Current At AppraisalVice President: Christiaan J. Poortman Caio Koch-WeserCountry Director: Theodore O. Ahlers Daniel RitchieSector Manager/Director: Emmanuel Forestier A. Amir Al-KhafajiTeam Leader at ICR: Ines Fraile Mohammed BenouahiICR Primary Author: Mohammed Dalil Essakali; Ines

Fraile

2. Principal Performance Ratings

(HS=Highly Satisfactory, S=Satisfactory, U=Unsatisfactory, HL=Highly Likely, L=Likely, UN=Unlikely, HUN=Highly Unlikely, HU=Highly Unsatisfactory, H=High, SU=Substantial, M=Modest, N=Negligible)

Outcome: S

Sustainability: L

Institutional Development Impact: M

Bank Performance: S

Borrower Performance: S

QAG (if available) ICRQuality at Entry: S

Project at Risk at Any Time: No

3. Assessment of Development Objective and Design, and of Quality at Entry

3.1 Original Objective:

1. This project was prepared in a context where the water sector had a high priority in the development program of the Government of Tunisia owing to the scarcity of water resources in the country.  Expansion and improvement of water supply and sewerage services have been the responsibility of two autonomous national public utilities, the Société Nationale d'Exploitation et de Distribution des Eaux (SONEDE) in charge of production and distribution of potable water, and the Office National d'Assanissement (ONAS) responsible for sewerage collection and wastewater treatment.

2. The project development objectives were to assist the Government in:

(a) implementing a suitable demand management program and institutional building measures so as to increase the operational efficiency of SONEDE and ONAS and strengthen their technical and financial capabilities;(b) providing safe, adequate and easily accessible water supply and sanitation services in urban and rural areas;(c) promoting the reuse of treated sewage for irrigation and industrial use; and(d) encouraging SONEDE and ONAS to contract some of their operations to the private sector.

3. The project development objectives were clear, well defined, and mostly realistic.  The staff appraisal report included a set of relevant performance indicators to monitor the financial, investment and operational performance of SONEDE and ONAS, which facilitated project supervision and reporting.  Retrospectively, SONEDE and ONAS, the implementing agencies, proved to be capable of achieving these ambitious goals and implementing such a complex operation, with one important exception, i.e. ONAS' financial performance.  The objective and indicators related to ONAS' financial performance proved to be hard to achieve as ONAS' finances have deteriorated instead of strengthened over the project period.  However, even this objective, which failed to materialize, remains a desirable goal to this date, and although the financial targets were ambitious, they were not too much out of line with the past performance of ONAS.

4. Moreover, the objectives were fully consistent with the Government priorities in the water sector, as well as with the Bank's Water Policy Paper, the MENA Water Strategy and the Country Assistance Strategy for Tunisia presented to the Board on May 11, 1993.  The project and its objectives helped implement the Government long-term strategy to satisfy the country's potable water needs until the year 2010 and avoid a severe water shortage.  These objectives also fitted well with the 1993 Bank's assistance strategy for Tunisia of supporting the Government's efforts to strengthen the management and conservation of its water resources.  The objectives adequately responded to the context and status of development of the water sector in Tunisia as they rightly recognized that despite commendable achievements in the level of water services in urban areas, more remained to be done in rural areas, in mobilizing water resources, implementing adequate cost recovery schemes and improving the efficiency of, and private participation in, SONEDE and ONAS' operations.  The objective of increasing water supply services in rural areas also followed previous efforts by SONEDE under the sixth and seventh Bank-supported water supply projects.  The objectives remain highly relevant today and are consistent with the Government priorities and strategy regarding basic infrastructure and sustainable development as stated in its Tenth Development Plan covering the 2002-2006 period.  The objectives are also in line with the Bank’s 2000-2002 Country Assistance Strategy to Tunisia, the 2003 Bank Water Supply and Sanitation Business Strategy, the 2002 Bank Water Resources Strategy, the 2003 Bank Infrastructure Action Plan, and the Millennium

- 2 -

Development Goals.

3.2 Revised Objective:

5. The project objectives were not revised.

3.3 Original Components:

6. The project had four main components:

(a)  rehabilitation and expansion of the water supply systems in Greater Tunis;(b)  provision of piped water systems to approximately 320 small villages (about 200,000 people) in rural Tunisia;(c)  expansion of sewerage collection systems and construction of sewage treatment facilities in the towns of Tozeur, Tataouine and Jemmal-Zermeddine; and(d)  institutional development with a view to enhancing water resources management, and the financial viability and efficiency of both SONEDE and ONAS. This component included staff training, studies, technical assistance, purchase of equipment, and a campaign to raise public awareness on better use of water and sanitation services.

7. Rehabilitation and expansion of the water supply systems in Greater Tunis (implemented by SONEDE, estimated cost of US$ 33.4 million excluding taxes and customs duties and including physical and price contingencies).  This component included (i) increasing the capacity of the Ghdir-El-Goula water treatment plant (WTP) by 2 m

3

/s; (ii) restructuring of the Tunis water supply mains of Kasseb and Oued-Ellil; (iii) replacement of the Ras-Tabia/Bir-Kassaa pipeline; (iv) rehabilitation of transport and distribution pipelines; (v) construction of reservoirs totaling an additional storage capacity of 25,000 m

3

; and (vi) acquisition and installation of control system equipment.  This component is directly related to development objective (b), as it contributed to satisfying new demand as well as ensuring adequate water supply.  This component also contributed to reducing unaccounted for water, and hence helped achieve development objective (a).

8. Provision of piped water systems to approximately 320 small villages (about 200,000 people) in rural Tunisia (implemented by SONEDE, estimated cost of US $32 million excluding taxes and customs duties and including physical and price contingencies).  This component aimed at supplying piped water to some 320 small villages with a total population of about 200,000 inhabitants.  The project evaluation estimated that most systems would be gravity systems and that the average cost per village would be around US$100,000.  This component directly contributed to achieving development objective (b) and it was a continuation of past Bank-supported water supply projects in Tunisia.

9. Expansion of sewerage collection systems and construction of sewage treatment facilities in the towns of Tozeur, Tataouine and Jemmal-Zermeddine and institutional development for ONAS (implemented by ONAS, estimated cost of US$ 24.6 million excluding taxes and customs duties and including physical and price contingencies).  This component was co-financed with the Nordic Development Fund (NDF) and the Nordic Investment Bank (NIB) and included: (i) construction of waste water treatment plant (WWTP) and extension of sewerage network in the towns of Tozeur (20,000 inhabitants), Tataouine (26,000 inhabitants, financed by NDF/NIB) and Jammel-Zermeddine (29,000 inhabitants, financed by NDF/NIB); (ii) carrying out studies including preparation studies for the Fourth urban sewerage project, rural sewerage strategy and the geographic information system; and (iii) institutional strengthening, technical assistance, training and laboratory equipment.  This component

- 3 -

contributed to achieving all development objectives.

10. Institutional development: Demand management, studies and technical assistance to SONEDE (implemented by SONEDE, estimated cost of US$ 5.6 million excluding taxes and customs duties and including physical and price contingencies).  This component included the acquisition of remote communication and remote measuring equipment for demand management (US$2.7 million) and studies including: (i) updating water supply master plans for Greater Tunis and for southern Tunisia; and (ii) examining different options related to the private sector participation in potable water.  This component helped achieve development objectives (a) and (d).

11. The components were clearly defined and appropriately designed to meet the project objectives.  The project design took into account best practice approaches for water and sewerage projects and the specificity of the Tunisian context.  This included linking the institutional capacity strengthening to results optimization on the ground.  Components definition and design also took advantage of lessons learned from past projects with SONEDE and ONAS, namely the demand management through tariffs, the unaccounted for water program, securing water supply for Greater Tunis, and providing potable water to rural areas all benefited from past Bank experience in Tunisia.  Implementing agencies had demonstrated in previous Bank operations that they were capable of successfully executing these types of components.  Yet the studies and activities included in the institutional components may have not been sufficient to fully achieve the institutional strengthening objectives, specifically regarding the improvement of the financial capacity of the executing agencies.

12. Furthermore, the targets and monitoring indicators included in the Staff Appraisal Report were well suited to conduct project evaluations and relate some project outputs to development objectives.  In particular, the indicators monitored very well the overall technical and institutional capacity of SONEDE and ONAS.  However, impact indicators for the specific contribution of each component to the stated development objectives were not as detailed and clearly defined at project appraisal as the institution-wide indicators, and SONEDE and ONAS also receive financial and technical support from other bilateral and multilateral financial institutions.  Still, key indicators for most components/development objectives were included in the Project Supervision Reports early in project implementation.

3.4 Revised Components:

13. The physical components were slightly revised during project implementation.  The project was extended for 12 months to fully achieve the physical objectives of the Greater Tunis water supply component, and carry out some additional works with available funds under the sewerage and rural water supply components.

14. Rehabilitation and expansion of the water supply systems in Greater Tunis.  This component was revised to take into account the findings of the updated Greater Tunis water supply master plan.  The financing of the capacity extension of the Ghdir-El-Goula water treatment plant was finally separately provided by the Kuwaiti Development Fund, and the Bank loan financed the stabilization of the site only.  Furthermore, following a detailed technical study, the restructuring of the water supply mains of Oued-Ellil and Kasseb was abandoned and replaced with a more efficient solution which included a new treatment plant in Medjez-El-Bab (0.4 m

3

/s).  Finally, the acquisition of water meters was added to the project in order to further assist SONEDE in reducing unaccounted for water and improving water demand management.  This component finally included: (i) site stabilization of the Ghdir-El-Goula water treatment plant and construction of the Medjez-El-Bab water treatment plant; (ii) replacement of the Ras-Tabia/Bir-Kassaa pipeline; (iii) rehabilitation of transport and distribution pipelines; (iv) construction

- 4 -

of reservoirs totaling an additional storage capacity of 25,000 m3

; and (v) acquisition and installation of control system equipment.  The institutional development component for SONEDE, including demand management, studies and technical assistance was also merged with this component.

15. Expansion of sewerage collection systems and construction of sewage treatment facilities in the towns of Tozeur, Tataouine and Jemmal-Zermeddine and institutional development for ONAS.  The following activities were added to this component during project implementation to fully utilize the funds available under the loan: (a) strengthening of sewerage network in the town of Tozeur (adding two more phases to the original two); (b) executing two rural sewerage pilot projects (studies and works) in two rural centers: Chorfech and Chouigui.  Works for this latter center were later transferred to the Fourth sewerage project; (c) training for ONAS staff; (d) acquisition of maintenance equipment; and (e) carrying out studies and construction works for two compact sewage treatment plants in the towns of Boumerdes and Fernana.  In addition, for the part financed by NDF/NIB, two actions were added: (a) carrying out studies and construction works for one compact sewage treatment plant in the town of Mornag; and (b) construction of sewerage transfer system from the town of Ghomrassen to the sewage treatment plant of Tataouine.  On the other hand, the study for the information system master plan was dropped from this project and it may be included in a subsequent sewerage project.

3.5 Quality at Entry:

16. The quality at entry of the project was overall satisfactory.  This rating is based on (i) the strong consistency of the project objectives with the Government and the Bank's water sector development priorities; (ii) good assessment of implementing agencies' capacity at appraisal; (iii) high quality of project design; and (iv) adequate project preparedness at appraisal.

17. Strong consistency of project objectives with the Government and the Bank's water sector development priorities.  As stated in section 3.1, the project objectives reflected the main water sector development priorities of the Government.  Furthermore, one of the five key objectives of the 1993 Bank's assistance strategy for Tunisia was to support the Government's efforts to strengthen the management and conservation of its water resources.  The project was the result of a continuing dialogue in the water sector between the Bank, the Government of Tunisia and the two implementing agencies SONEDE and ONAS.

18. Good assessment of implementing agencies' capacity at appraisal.  Although the project suffered from some delays and benefited from a 12-month extension of the implementation period, it also surpassed its initial objectives in some respects.  Project implementation was overall smooth and successful.  Both SONEDE and ONAS demonstrated their ability to successfully complete this complex project, and showed strong ownership throughout all stages of the project cycle.  These are well run and managed public utilities, which are considered two of the best utilities in the Region.

19. High quality of project design.  Although the project was not assessed for quality at entry, a Quality Assurance Group (QAG) review of supervision concluded that "the project design was sound, though there has been some disruption on Bank Project because of in-and-out of Kuwait Fund taking on some elements of project."  Project components were comprehensive and well designed to achieve the development objectives.  Initial assumptions were realistic.  The staff appraisal report documented the project and its background particularly well.  Performance indicators on both the technical and the financial aspects of the two utilities were more comprehensive and pertinent than in most other water and sewerage projects, and they facilitated monitoring of project outputs and outcomes.  The risks were also well assessed as the staff appraisal report clearly indicated that the main risk would be the Government not living up to its obligation to allow appropriate and timely tariff adjustments.  To mitigate this risk,

- 5 -

agreement was reached with the Government to maintain the practice of annual increases, and tariff increase covenants were included in the loan and project agreements.  Unfortunately, this risk materialized, especially with regard to ONAS and to a lesser extent with regard to SONEDE.

20. Adequate project preparedness at appraisal.  Satisfactory execution and timely completion of important works and activities such as the rehabilitation part of Greater Tunis water supply component as well as the sewerage and rural water supply components demonstrate adequate project preparedness at appraisal.  However, the project suffered from some delays, namely the expansion part of the Greater Tunis water supply component started substantially late, mostly because updating the Greater Tunis water supply master plan took more time than originally foreseen.

4. Achievement of Objective and Outputs

4.1 Outcome/achievement of objective:

21. Overall, the project achieved its development objectives.  Physical outputs have in some cases exceeded initial targets.  Bank missions’ reports as well as implementation monitoring reports document how project objectives have been achieved.  The project played a critical role in improving SONEDE's performance.  SONEDE greatly benefited from the project and from the Bank's assistance as improving Unaccounted For Water (UFW) ratio and improving rural population access to potable water are directly linked to the project.

22. Development Objective (a): implementing a suitable demand management program and institutional building measures so as to increase the operational efficiency of SONEDE and ONAS and strengthen their technical and financial capabilities.  This objective was partially achieved.  Making SONEDE and ONAS more autonomous and self-financing turned out to be hard to achieve.  SONEDE managed to strengthen both its operations and finances (see performance indicators in annex 1).  Despite having an average tariff below the appraisal target, SONEDE has been able to get sufficient tariff increases (an average tariff increase of 9.2 percent in 1996, 5 percent in 1999, 7 percent in 2001, and 5.7 percent in 2003) and generate adequate revenues from its operations.  SONEDE has also been successful in recovering arrears (TD 25 million) from public administration entities, chemical industries, and municipalities.  On the other hand, SONEDE's overall supply and distribution UFW has reached 20 percent, thus SONEDE's actual performance on this not only remains one of the best in the region but has also exceeded appraisal expectations.  Achieving this UFW level strongly contributed to improve SONEDE’s financial situation besides helping preserve water resources.  The reduction of the number of employees per 1,000 connections from 6.8 to 4.1 during the project implementation period also shows an improvement in SONEDE’s efficiency which surpasses initial targets.  Moreover, SONEDE has implemented a new cost accounting system and is planning to implement a geographic information system and a project management system, which will further strengthen its management capacity.  On another front, SONEDE is carrying out a tariff study in order to determine an optimal tariff system that would be conducive to better cover operation and maintenance costs, increase its self-financing capacity to invest, and protect poor consumers. Sustaining such good performance in the future will however require periodic tariff adjustments as well as additional efforts to improve the company’s efficiency.  SONEDE is indeed engaged in ambitious investment programs which would result in increasing its marginal costs while decreasing its marginal revenues: water supply in dispersed rural areas and water production from non-conventional sources such as desalination.

23. ONAS' operational performance has also exceeded initial targets in terms of sewerage coverage levels and percentage of wastewater treated at the national level.  ONAS has also improved its efficiency,

- 6 -

although to a lesser extent than SONEDE, as it is shown by the reduction of the number of employees per 1,000 connections which has decreased from 6.4 to 5.5 during the project implementation period, in line with initial targets.  ONAS' financial situation is however deteriorating.  This is mainly due to a tariff freeze for almost five years (July 1998 to March 2003) combined with a rapid expansion in sewerage activities and an involvement in non-core businesses such as storm drainage and rural sanitation services.  This unsustainable trend needs to be reversed.  It has not negatively affected ONAS' operations as yet because the Government has decided to almost double the consumption subsidy in 2003.  A tariff increase of 5.7 percent granted to ONAS in March 2003 is a positive step, but further measures need to follow.

24. Development Objective (b): provision of piped water systems to approximately 320 small villages (about 200,000 people) in rural Tunisia.  This objective was achieved.  SONEDE has successfully expanded its networks to new urban and peri-urban areas in Greater Tunis.  The project was critical in enabling SONEDE to respond to increased water demand in the Greater Tunis area.  The number of connections in the area increased by 40 percent in 8 years from 343,286 in 1994 to 482,995 in 2002; hence some 139,709 new connections were added.  Satisfying this new demand would not have been possible without the increased production capacity and the reduced unaccounted for water achieved thanks to the project.

25. The rural water supply component was highly satisfactory given its social impacts, physical outputs and the successful ownership of this component by SONEDE.  During project implementation SONEDE has taken adequate actions to ensure that the initial coverage and connection targets are achieved.  The depreciation of the Tunisia Dinar in relation to the Dollar enabled SONEDE to extend its initial program since civil works are locally procured.  This partially explains the need for a one-year extension of the project implementation period.  At project closing, some 459 groups of rural centers, compared to 320 planned at appraisal or 44 percent more, were connected to SONEDE's national water supply network.  Works for two other localities were being completed.  In terms of population benefiting from this component and connected to the network, SONEDE estimates that, as of December 31, 2002, 178,950 people were connected to the network through 30,652 household connections, and 8,869 people were covered by 44 collective standpipe systems.  Extensions to water networks in these 459 rural centers financed by SONEDE and the Government further added 5,273 household connections benefiting 30,663 people and 3 collective standpipe systems covering 258 people.  The total population connected is thus 209,613 through 35,925 household connections in addition to 9,127 covered by 47 collective standpipe systems.  Latest estimates as of June 30, 2003 are in annex 1 and show a total population connected at about 235,485 and household connections at 40,343.  These figures surpass the appraisal target of 200,000.  On the other hand, household connections rate to SONEDE network also increased from 27.1 percent in 1995 to 35.1 percent in 2002.  During this same period, new water connections in rural areas carried out by SONEDE under the project are estimated to have contributed by some 38 percent to this increase.  The project also partly contributed to increasing SONEDE’s water supply coverage in rural areas.  This coverage went up from 33.4 percent in 1995 to 42.4 percent in 2002.  It should be mentioned that the total water supply coverage in Tunisia in 2002 was estimated at 83.3, the Génie Rural covering some 40.9 percent of rural population. 

26. At project closing date, ONAS' operations were covering 144 municipalities for a total population of 5.4 million inhabitants, and the volume of treated water is 150 million m

3

against a target of 149 million m

3

at appraisal.  The project allowed the improvement of living conditions of some 150,000 inhabitants compared to an appraisal target of 130,000.  A sewerage treatment capacity of 26,200 m

3

/day has been added thanks to the three wastewater treatment plants and the three compact stations.  ONAS also launched and completed a pilot project on rural sanitation in the village of Chorfech that has been successful. 

- 7 -

Further studies on the institutional aspect of rural sewerage are being undertaken by ONAS in order to determine the most efficient management and institutional settings for rural sanitation.

27. Development Objective (c): promoting the reuse of treated sewage for irrigation and industrial use.  This objective was achieved.  Treated water coming from the new Tozeur wastewater treatment plant, built under the project, will be reused to water a 65-hectare golf course in Ras-El-Ain tourist district.  Works for the transfer system at a cost of some TD 1.5 million are being financed by the Ministry of Tourism and ONAS. The Ministry of Agriculture and ONAS have also undertaken other initiatives to reuse treated wastewater with separate financing such as the pilot project of groundwater recharge, which is part of the Greater Tunis Sewerage and Water Reuse Project. The Staff appraisal report did not have specific targets to measure the outcomes of this component. Treated wastewater reuse rate in Tunisia is currently estimated at 24 percent, half of which is used to water golf courses.

28. Development Objective (d): encouraging SONEDE and ONAS to contract some of their operations to the private sector.  This objective was achieved.  SONEDE is subcontracting its network expansion works to the private sector.  In 2002, about 94 percent of these activities were subcontracted to private companies.  Its connections installation activities have also been subcontracted to private companies but at a lower rate, 63 percent in 2002.  SONEDE has carried out a sub-contracting study whose recommendations for medium and long-term implementation are being discussed internally.  The terms of reference of this study were as ambitious as its recommendations were timid and conservative.  The recommendations of the study for the short and medium term covered certainly a wide range of activities, but in reality these recommendations were a mere conclusion that SONEDE should focus on its core business, that is production and supply of potable water, and abandon in-house construction and civil works as well as the security services of its installations.

29. More important steps have been taken by ONAS in order to have private sector participation in some sewerage activities such as operation and maintenance of part of its network and WWTPs.  A pilot project was initiated in 1997 by contracting out operations of 185 kilometers of sewerage network in El-Menzeh district in Greater Tunis.  This experience showed that delegating operation and maintenance of the network to the private sector could considerably help reduce costs while providing the same or improved level of service.  ONAS has also contracted out operations of four other sewerage networks in 2001 and 2002.  According to the objectives of the Tenth Development plan (2002-2006), ONAS is planning to involve the private sector in the management of some 1,800 Km of its sewerage collection network and 26 of its wastewater treatment plants.  Furthermore, ONAS' private sector participation (PSP) action plan will focus on carrying out different studies on sector strategy, market analysis, social aspects, regulatory framework and PSP participation in rural sanitation.

30. As for actions directly associated with the project, ONAS has contracted out operations of Tataouine WWTP, built under the project, and 95 km of the sewerage network to the private sector.  The contract starting date was March 1st, 2003.  At project closing date, ONAS was in the process of bid evaluation of proposals by private firms to operate Tozeur and Nafta WWTPs.  ONAS was also finalizing procurement documents for contracting out of operations of three other WWTPs: Jammel-Zermeddine, Monastir-Frina and Sayada-Lamta.

31. Beyond the project development objective, it should be noted that the legal and institutional framework for PSP remains underdeveloped in Tunisia, which hinders the development of more advanced forms of PSP other than sub-contracting.  Operation contracts duration is limited to a maximum of five years for instance.  Institutional and legal reforms remain necessary, and the Government is currently conducting an assessment of obstacles to PSP and outlining different options for reform.  The time frame

- 8 -

for the current Government effort contrasts however with the rapid change of conditions in the water and sewerage sector.

4.2 Outputs by components:

32. Rehabilitation and expansion of the water supply systems in Greater Tunis. This component output is rated satisfactory. Status of implementation of this component at project closing was as follows: (i) water treatment plant works were completed; (ii) Ras-Tabia/Bir-Kassa water pipeline was completed and has been operational since 1998; (iii) rehabilitation and improvement works of mains were completed; (iv) reservoir works were completed except for one reservoir where final tests were underway; and (v) control and regulation equipment component was carried out.  The project increased water treatment capacity by 207,360 m

3

/day and water storage capacity by 25,000 m3

.  Overall, some 80 kilometers of large-section water supply pipelines were laid.  The rehabilitation of water mains using ductile iron pipes and experimenting new materials such as fiberglass helped improve the durability of the network and reduce water losses as well as the number of breakages.  Old water meters were replaced by more accurate and reliable ones, and metering and leak detection equipment were purchased.  Under the project, SONEDE also introduced new technologies such as remote control and remote management of its installations including reservoirs and WTPs, and the regulation system of Ghdir-El-Goula WTP was completely modernized.  Sensitization campaigns were carried out by SONEDE and included television ads, booklets, site visits as well as the production of a water user guide.

33. Provision of piped water systems to approximately 320 small villages (about 200,000 people) in rural Tunisia. This component output is rated satisfactory.  SONEDE has provided piped water to some 459 rural centers, against a target of 320 at appraisal, and the total population connected thanks to these investments is more than the appraisal target of 200,000 (see paragraph 25).  At project closing, some 40,343 household connections and 47 collective standpipe systems had been completed.  In total, SONEDE installed more than 1,687 kilometers of pipes of various types and diameters under this component.

34. Sewerage.  This component output is rated satisfactory.  All WWTPs and network works were completed.  Tozeur WWTP has been operating since July 2000.  Treated wastewater from this plant will be used to water a 65-hectare golf course.  Works for the sewerage network in Tozeur were completed.  Some 86 kilometers of new sewers were laid, and 4,750 new connections to the sewerage network were added.

35. All sewerage and wastewater treatment studies were completed.  The geographic information system was installed and is being tested.  Training sessions on GIS aimed at ONAS staff were conducted.  This system will be extended to include 18 more cities.  The institutional strengthening component, which includes training and human resources development, was completed.  Training sessions as well as study tours for ONAS staff were conducted and included various themes such as rural sanitation, new sewerage technologies and private sector participation in the sewerage sector.  The cost accounting system was put in place with a one-year delay.  Studies for the preparation of the fourth sewerage project were completed.  Operations equipment was purchased under the project.  The enterprise information system strategy sub-component was transferred to the Greater Tunis Sewerage and Water Re-use Project.

36. A rural sanitation sector strategy study carried out under the project identified a pilot project to experiment an adapted sanitation system.  ONAS chose the rural center of Chorfech for the first phase of the pilot project and the rural center of Chouigui for the second phase, which is not financed under this project.  Works for Chorfech were completed and the system is operational.  In order to evaluate this pilot project, ONAS is planning to carry out a technical, institutional and operational study for the management of rural sanitation systems.  This study would be undertaken under the rural infrastructure project co-financed by the Bank.

- 9 -

4.3 Net Present Value/Economic rate of return:Ex-post evaluation of the Greater Tunis Water Supply Component

37. The economic analysis of the Greater Tunis Water Supply Component shows an ex-post economic rate of return (ERR) of 18.5 percent, which is higher than the 16.6 percent estimated at appraisal. The following two factors explain this higher value of the ERR: (i) the net present value of the project cost over the period 1995-2030 is slightly lower than estimated at appraisal; and (ii) while the real water demand was lower than projected in 1995 and 1996, it was on average 7 percent higher during the rest of the implementation period. The ex-ante economic analysis assumed that the bulk of the investments would occur in the 1996-1998 period. In reality, investments were spread over a longer period (1996-2003). Accordingly, the ex-ante evaluation estimated a net present value of project costs of TD 39.7 million at a discount rate of 12 percent in the period 1995-2030, whereas the ex-post net present value is TD 37.8 million. Water demand in 1995 and 1996 was respectively 0.5 and 2.9 percent lower than the ex-ante projections. Thus, the saturation of the system which had been predicted to occur in 1996 in the ex-ante evaluation was avoided because of the improvement in UFW. Real water demand in 1997 and 1998 was slightly higher than projected in the ex-ante evaluation, and 7 to 11 percent higher afterwards, which leads to an average 8 percent higher over the 2003-2030 projection period. Since the project benefits reflect the additional demand that the “without project” situation cannot satisfy, a higher demand led to higher benefits and a higher ERR.

Water supply consumption in Greater Tunis (million m3)1995 1996 1997 1998 1999 2000 2001 2002

Ex-ante projections 76.0 77.5 79.1 80.7 82.3 84.0 85.7 87.4Real consumption 75.6 75.3 80.2 82.9 88.0 91.2 95.2 94.9Difference -0.5% -2.9% 1.4% 2.7% 6.9% 8.6% 11.1% 8.5%

Ex-post evaluation of the Rural Water Supply Component

38. The ex-post assessment of 29 rural sub-projects implemented under the Project shows higher net social benefits than those initially estimated during the ex-ante appraisal phase.  Therefore, these projects meet the net social gain criterion ex-post, as they met it ex-ante.  Differences are due to generally larger population covered and lower investment cost than initially projected.  They are also due to a significantly higher cost of acquiring water in the situation without the project than in the ex-ante assessment.  Cutting the time spent on water provision by one half shifts five out the 29 projects from socially profitable with positive net social gains to socially unprofitable with negative social gains.  The methodology is sound and provides a good guide for selection of rural projects, but cost components, particularly those of water supply in the situation without project, need to be defined and assessed more reliably.

Discrepancy between Projected and Actual Values Average

ex-anteAverage ex-post

Discrepancy (%)

Proportion of ex post higher than ex-ante

Population (inhabitants) 1,577 2,164 + 37% 80%Consumption per household 91 88 - 3.3% 50%Cost under the project

Operation cost (TD/m3) 0.508 0.508 0% -

Investment cost (TD/m3) 0.531 0.531 0% 40%

Total investment (TD) 317,600 240,000 -25% 17%Investment per household 1,480 1,213 -20% 24%

Cost without the project (TD/m3) 4.260 6.060 +42% 96%

Net social gain (TD/m3) 1.830 3.760 +100% 93%

- 10 -

A more detailed economic and social evaluation is in annex 3.

4.4 Financial rate of return:

39. N/A

4.5 Institutional development impact:

40. The project institutional development impact was modest.  The project had an important impact on strengthening operational and technical capacities of ONAS and SONEDE, but it could not prevent the deterioration of ONAS's finances.  Among the positive institutional development impacts, the project made a critical contribution to the effective use and preservation of the country scarce water resources.  SONEDE's performance and service delivery were significantly improved.  Investments in water production, distribution and demand management proved to be beneficial both in reducing UFW from 28 percent in 1993 to around 20 percent since 1999 and ensuring continuous water supply in Greater Tunis.  SONEDE also expanded its service to rural areas where its coverage went from 33.4 percent in 1995 to 42.4 percent in 2002, and its household connection rate from 27.1 percent in 1995 to 35.1 percent in 2002.  The project also enhanced ONAS's technical performance and increased its coverage as well as wastewater treated.  Some 26,200 m

3

/day wastewater treatment capacity was added and sewerage networks were enhanced in six towns.

41. Whereas both SONEDE and ONAS benefited from the project through implementation of improved financial management and accounting systems, their financial capacity did not improve.  SONEDE had a good financial situation prior to the project and managed to keep it owing to the water tariff increases granted during project implementation.  ONAS's financial situation has been deteriorating during the past few years, mainly because of the freeze of sewerage tariffs for almost five years.

42. A healthy financial capacity is however essential in order to preserve and improve the operational and technical capacity of these two entities in the medium and long run.  ONAS and SONEDE are autonomous but they rely upon the Government for tariff increases as well as for approval of their 5-year development program.  The current trend of ONAS's deteriorating financial situation, which is thus becoming more and more dependent on increasing state subsidies, points out to a lack of corporate autonomy and a weakness of the regulatory framework of the current model.  The project did not have an adequate vehicle to assist the Government in analyzing and coming up with solutions to such difficulties in an institutional framework otherwise with many solid features, which has served Tunisia well for more than 30 years and enabled good sector performance, as shown by several performance indicators included in this report.  The Tunisian Government had a solid track record in terms of tariff adjustments in the past, and tariffs had been adjusted periodically.  Correcting the current financial situation without negative effects on the operational and technical performance of ONAS is thus still possible, even likely.  Including a component to assist the Government in strengthening the overall institutional framework of the sector could have probably helped increase the institutional development impact of the project.

43. Regarding PSP, the project objective was achieved, but this objective was modest given current standards.  SONEDE still carries out in-house activities that are normally sub-contracted by most sector utilities all over the world in developing and developed countries.  This practice hinders the development of local small and medium size private enterprises specialized in the water supply sector.  ONAS seems more open and willing to have increased PSP in its operations, but it is constrained by the Tunisian legislation.

- 11 -

5. Major Factors Affecting Implementation and Outcome

5.1 Factors outside the control of government or implementing agency:

44. The influence of factors outside the control of Government or implementing agencies has been mixed.  Bank performance affected project achievements positively.  However, at least three Bank project teams led this project, which caused some delays, mainly in procurement, at every transition period.  The second and major outside factor was the devaluation of the Tunisian Dinar which went from almost parity at appraisal to US$ 0.75 during most of the project implementation period.  The appreciation of the dollar indirectly generated more resources for ONAS and SONEDE, which allowed them to adjust some components and add new ones.  The one-year extension was partly justified by the fact that funds still available at that time could be used for very efficient actions such as demand management and water metering upgrading.

5.2 Factors generally subject to government control:

45. Lengthy procurement procedures in Tunisia contributed to the delays incurred in launching the expansion works of the Greater Tunis water supply component.  In addition to extending project preparation and implementation periods, which was taken into account at project preparation, these procedures become an additional constraint when the investment plans change to incorporate improvements, as it was the case in this project, or to respond to unforeseen circumstances.  In addition to the Bank’s no objection, implementing agencies had to report to internal, regional, or high procurement commissions, depending on the size and type of the contract.  The prequalification of consulting firms is a mandatory step in Tunisian procurement law, and it sometimes doubles the time needed to hire consultants.  The procurement issue could have been mitigated if the "Commission nationale des marchés" were aware of Bank procedures and had showed a proactive attitude in the context of the project.

46. Water and sewerage tariff adjustments need to be approved by the Government.  The appraisal report clearly identified as a risk the Government not living to its obligation to allow the appropriate tariffs.  Unfortunately, this risk did materialize, and the sewerage tariff freeze for almost five years damaged ONAS' financial situation.

47. By law, operations contracts with the private sector are limited to five years maximum.  This restriction is a clear obstacle to the development of private sector participation in the water and sewerage sectors.

48. Although of a lesser importance than the above-mentioned issues, combining different components to be implemented by two separate entities (SONEDE and ONAS) under a single loan (No. 3782) and a shared special account brought about numerous practical complications.  It obscured the proper accounting of disbursements and allocation of Government subsidies between the two implementing agencies.  The complications associated with this were compounded by factors beyond the control of the Government such as Bank staff errors in the labeling of forms 384, and exchange rate variations due to the Government decision to change the denominating currency half way during implementation.  This problem was particularly cumbersome towards the end of project execution when the implementing agencies wanted to accurately ascertain whether they had sufficient funds under the loan.  It also consumed considerable scarce Bank supervision time on an issue that had no developmental impact.

- 12 -

5.3 Factors generally subject to implementing agency control:

49. The strong technical and institutional capacity of both ONAS and SONEDE explains in part the success of the project.

50. Some technical studies were not ready at the time the project became effective, and their finalization was delayed.  The Greater Tunis Water Supply Master Plan, which was critical for some of the expansion works of the Greater Tunis water supply component, was delayed by more than a year not only because of procurement procedures, but also because SONEDE's technical studies department, which carries out almost all technical studies inhouse, was very reluctant to have an independent consultant carry out this major study.

51. The procurement schedule could have been better managed.  Both SONEDE and ONAS internal procurement procedures are cumbersome.  In addition to the lengthy national procurement procedures, the internal procurement cycle including documents preparation and approval is far from efficient.  Coordination among different department needs to be improved.

5.4 Costs and financing:

52. The total project cost, including physical and price contingencies, was estimated at appraisal at some US$ 110.9 million taxes and customs duties included and US$ 95.8 million without taxes and customs duties.  Bank financing was set at US$ 58 million.  The actual cost of the project is estimated at US$ 104 million excluding taxes and customs duties, of which some US$ 99 million correspond to activities completed by project closing date.  Disbursements from the project proceeds reached US$ 55.71 million.  The undisbursed loan amounts were US$ 2 million from Loan 3783 and Euro 0.5 million from Loan 3782.  Details of these figures appear in Annex 2.

6. Sustainability

6.1 Rationale for sustainability rating:

53. The sustainability rating given to the project is likely.  While the Greater Tunis project component is a part of SONEDE's everyday business, the outputs of the rural water supply component are less obvious to operate and manage.  This is all the more true than rural water supply is a relatively new activity to SONEDE.  However, the project focused on relatively dense rural centers and similar to peri-urban zones, and hence SONEDE does have the expertise to manage this type of service and ensure the sustainability of the rural water supply component.  The rural water supply strategy study undertaken under the project already made recommendations on strategic options, technologies as well as the institutional setting needed for providing services to this type of areas.

54. The sustainability of project outputs implemented by ONAS may be questioned given its deteriorating financial situation.  However, ONAS' past track record, recent actions taken by ONAS to reverse this trend, and the clear Government commitment to support ONAS in particular and the sewerage sector in general reduce the risk of not achieving the sustainability of these outputs.  The Bank dialogue with Tunisia in the sewerage sector in the context of Greater Tunis Sewerage and Water Re-use Project will also contribute to ensuring that ONAS' financial situation improves.

6.2 Transition arrangement to regular operations:

55. Most project outputs are already integrated in the everyday business of ONAS and SONEDE,

- 13 -

which reduces the need for special transition arrangements to regular operations.  The rural water supply component is however part of an ongoing SONEDE long-term investment program and might have a negative impact on the cost and revenue structures of SONEDE.

56. ONAS has already opted for private sector participation in some of its operations carried out under the project.  The adoption of this institutional setting by ONAS could ensure efficient operations of the project outputs.  Redeployment of staff might however be necessary in order to fully extract the benefits and avoid extra-costs.

57. Both ONAS and SONEDE have prepared a note on transition arrangement to regular operations, and it is included in the Borrower's contribution to ICR in annex 8.

7. Bank and Borrower Performance

Bank7.1 Lending:

58. The Bank performance during project preparation is rated satisfactory.  At the time of preparation and appraisal, the project was fully consistent with Bank policies and the Country Assistance Strategy and clearly responded to a national priority.  Bank dialogue with Tunisia in the water supply and sewerage sector has been continuously going on, and project preparation took advantage of the Seventh water supply project which was ongoing at the time to design this project.  The economic evaluation of the project went beyond evaluating the economic rate of return of the project and developed an economic and social model to evaluate and prioritize SONEDE’s investments in rural areas.  The economic evaluation also estimated the long-term marginal cost of the water supply service and was a catalyst for the water tariff study.

7.2 Supervision:

59. The overall Bank performance in supervision is rated satisfactory. A Quality of Supervision Assessment was carried out by QAG and rated project supervision highly satisfactory.  The Bank staff showed flexibility and understanding in responding to evolving circumstances.  Bank teams were well staffed with a good skill mix.  Management and staff maintained strong continuity in the dialogue.  The mid-term review of the project was carried out in November 1998 and revealed no major deficiencies that warranted change in the development objectives or amendment of existing loan agreements.  Supervision of the environmental impacts mitigation plan was probably not as systematic as a category-A project would have required.  However, there is no indication that this had had negative effects on the project environmental performance.  On the procurement side, more forceful application of Bank procedures could have speed up implementation.  Changes in Bank teams were smooth, although some procurement delays occurred during transition periods.

60. The only area where the Bank could and should have been more proactive was regarding the lack of sewerage tariff increases for about 5 years, and the deteriorating financial situation of ONAS that ensued.  Raising the dialogue to Ministerial and Government level on this issue should have been done earlier in project implementation.  While the weight of the dialogue on this issue was transferred to the ongoing Greater Tunis Sewerage Project, which provided a more suitable basis since it was being solely implemented by ONAS, strong action was taken altogether too late.  As a minimum, and although already too late, the Bank could have established the approval of the long overdue sewerage tariff increase as a pre-condition for granting the extension requested by the Government.  Retrospectively, granting an extension without using its leverage on the tariff issue seems surprising.

- 14 -

7.3 Overall Bank performance:

61. The overall Bank performance is rated satisfactory.  Comprehensive teams were fielded in lending and supervision, with many members maintained throughout the project.  Supervision missions included the required expertise and staff who continuously monitored project activities.  One of the success stories of project was the use of highly skilled local consultants during preparation and supervision.  Also, the Bank was flexible in adjusting project components as conditions changed.  Finally, the Bank assisted the Borrower to minimize project delays and use the additional local currency generated by the appreciation of the dollar.

Borrower7.4 Preparation:

62. The Borrower performance in lending is assessed as satisfactory.  The Government as well as the two utilities worked in close collaboration with the Bank to prepare the project.  The project size and scope had changed many times during preparation, until the final design was agreed on.  This process may explain the fact that right after appraisal, the water treatment plant was taken out of the components financed by the loan to be financed by the Kuwaiti Fund.

7.5 Government implementation performance:

63. The Government implementation performance is assessed unsatisfactory due to the five-year freeze in sewerage tariffs, heavy procurement procedures and restrictions on PSP development.  Tariff increases for water supply granted by the Government to SONEDE were not in accordance with the loan agreement either, although the impact on SONEDE's financial situation was not as serious as in the case of ONAS.

7.6 Implementing Agency:

64. The implementing agencies (ONAS and SONEDE) performance is rated satisfactory.  Project complexity as reflected in the existence of two implementing agencies and the multiplication of components was very well managed by the two entities.  The rural water supply component was complex to implement as project area was scattered all over the country, but SONEDE's action was a success.  SONEDE was also very efficient in adapting to the changes in the Greater Tunis component and using available funds to increase the importance of the demand management component. On the other hand, project co-financing by NDF/NIB was very well managed by ONAS, and no problem was reported in relation to this issue.  ONAS also very well adapted to the changes in the sewerage component, as it was very efficient in completing the studies for the Fourth Sewerage Project.  Both utilities had problems however in managing delays incurred at the start of the project.

7.7 Overall Borrower performance:

65. The overall performance of the Borrower is rated satisfactory.  SONEDE and ONAS were qualified to implement the project.  In addition to their difficulties to manage the delays in project implementation, the Government stand on the tariff adjustment issue was a clear factor in not fully achieving development objective (a).

- 15 -

8. Lessons Learned

66. A key factor of the project success in improving overall sector performance and in achieving such ambitious development objectives resides in ONAS and SONEDE's strong institutional capacity.  The absence of isolated project implementation units exclusively created for the purpose of the project allowed a large impact on these entities' staff and procedures from the institutional strengthening instrinsically associated to project dialogue.  This positive aspect required however extra effort from Bank teams, who had to interact with several departments and units within each entity.  Other lessons can be learned from this project implementation experience, namely the following:

Sector policy dialogue

The deterioration of ONAS' financial situation shows that the Tunisian model in the water supply land sewerage sector is under strain and that further and faster reforms may be required to preserve the achievements of the past and face the challenges of the future. 

The Bank was not forceful enough during the early years of sewerage tariff freeze in raising lawareness with the Government and ONAS on the predictable negative impacts of the absence of sewerage tariff adjustments over a long period.  This issue should have been raised earlier to Ministerial or Government level.  Furthermore, the Bank should have established the approval of the long overdue sewerage tariff increase as a pre-condition for consideration of the project closing date extension.  This leniency was possibly due to the Bank's long and successful dialogue with the Government and the concerned utilities.

However, it should be noted that ONAS and SONEDE's past good performance, as reflected in lmany of today's performance indicators, also made it hard for the Bank to justify the need or to promote more ambitious or comprehensive sector reforms.  Tunisia's gradualist approach to its macroeconomic and development policy agenda has generally maintained stability and brought about steady quality of life improvements.  Sector policy dialogue and reform proposals need accordingly to be well adapted to this context, and be very respectful of local circumstances. 

Adding a policy reform component at the ministerial level to the original project could have helped lin accelerating sector analysis and in introducing adequate reforms.  This issue is getting more and more important given the current status of the sector, and it should be taken into account in future projects.  One area to explore could be how to improve coordination of water and sewerage policies.

Institutional capacity

SONEDE is still reluctant to hire consultants and contract out operations more broadly.  The lcompletion report of the Seventh Water Supply Project, dated August 1994, already mentioned that the long and successful dialogue with SONEDE had made the Bank staff too flexible and lenient towards this reluctance.  These two actions could however help SONEDE make considerable cost savings as it could strengthen its technical capacity.

Lessons should be drawn from the rural sanitation pilot project undertaken by ONAS.  ONAS lshould establish, in close cooperation with the Government, a clear institutional and financial framework to ensure the sustainability of any rural sanitation investments.

The rural water supply component used an economic model to prioritize and make investment l

- 16 -

decisions on the basis of preset objective criteria.  SONEDE is now using the same model for other rural water supply projects.  This practical tool is a step further toward rationalizing investment decision-making, especially in rural areas.  Still, as with any other model, the outputs of this model are as good as its inputs, and SONEDE needs to work harder in ensuring that the data used is up-to-date and accurate, and that the assumptions made are valid and solid.

Project design and project preparedness at appraisal

Finalizing technical studies before project effectiveness is critical in order to avoid delays in works land project implementation.  Completing such studies can take considerable time even in the case of strong institutions.  Carrying out feasibility studies of future operations during an ongoing operation shortens these delays, as it was the case of the Greater Tunis Sewerage and Water Re-use whose feasibility studies were financed under this project.

Combining two different components to be implemented by two separate entities (SONEDE and lONAS) under a single loan (No. 3782) and a shared special account brought about numerous practical complications and difficulties in proper accounting of disbursements, and allocation of loan proceeds among the two implementing agencies, and it should be avoided in the future.

9. Partner Comments

(a) Borrower/implementing agency:

- 17 -

- 18 -

- 19 -

- 20 -

(b) Cofinanciers:

(c) Other partners (NGOs/private sector):

10. Additional Information

None.

- 21 -

Annex 1. Key Performance Indicators/Log Frame Matrix

Key Performance Indicators

Outcome / Impact Indicator Baseline Mid-term End of

ProjectOutcome/Impact Indicator Target Actual Target Actual

DO (a): SONEDE (water supply) Debt Service Ratio (DSR) 1.40 1.84 1.98 2.70 4.61DO (a): SONEDE (water supply) Internal Cash Generation (CG) in % 36.5 51.3 40.6 66.5 61.7DO (a): SONEDE (water supply) Average Tariff in DT/m3 0.327 0.459 0.391 0.526 0.419DO (a): SONEDE (water supply) UFW in % 27.5 26.1 21.2 25.2 20.2DO (a): SONEDE (water supply) Staff per 1000 connections 6.8 5.3 5.2 4.7 4.1DO (a): ONAS (sewerage) Debt Service Ratio (DSR) 1.00 1.7 2.55 2.1 1.9DO (a): ONAS (sewerage) Internal Cash Generation (CG) in % 13.5 18.9 31.0 20.6 15.0DO (a): ONAS (sewerage) Average Tariff in DT/m3 0.187 0.415 0.384 0.553 0.417DO (a): ONAS (sewerage) Staff per 1000 connections 6.42 6.0 5.8 5.6 5.5DO (b): SONEDE (water supply) Population Served in 000’s 6,097 6,988 7,168 7,313 7,760DO (b): ONAS (sewerage) Population Served in 000’s 3,780 4,900 4,900 5,200 5,420DO (d): SONEDE (water supply) Subcontracting work to PSP - Network extension in % 66.0 84.0 89.0 90.0 94.0DO (d): SONEDE (water supply) Subcontracting work to PSP - Number of Connections in % 10.0 75.0 42.0 85.0 63.0Targets for ONAS include changes made in the context of Greater Tunis Sewerage and Water Re-use Project.

ONAS Performance IndicatorsINDICATORS 1992 1993 1994 1995 1996 1997

Connections (000) 499 520 578 616 650 690New connections (000) 43 21 58 38 34 40New connections due to the project (cumulative) 0 0 0 500 1500 2500Population covered (000) 3,780 3,935 4,101 4,330 4,560 4,701New population covered (000) 180 155 166 229 230 141New connections due to the project (cumulative) 0 0 0 10 30 50Employees Permanent 2,599 2701 2832 2949 2990 3,128 Temporaries 600 723 811 923 925 940 Total 3,199 3,424 3,643 3,872 3,915 4,068Number of towns covered by ONAS 80 81 91 100 111 117Connection rate ONAS 75.0% 76.0% 76.6% 76.7% 77.0% 78.0%SONEDE Water volume distributed (m3 million) 151 153 163 164 167 177ONAS Wastewater collected (m3 million) 119 125 130 135 138 141Average tariff (0.001 TD per m3) 186 218 248 268 298 344Operating ratio 91.1% 94.0% 95.0% 97.0% 93.0% 89.0%Self-financing ratio (incl. "recette compensatoire") 69.7% -7% 21% 29% 67% 23%Self-financing ratio (without "recette compensatoire") 13.5% -27% -4% 6% 45% 4%Debt service ratio (without "recette compensatoire") 0.90 1.48 1.61 2.46 2.91 2.87Debt service ratio (incl. "recette compensatoire") 1.00 1.97 2.89 4.80 4.58 5.33Working capital 80.5% 105% 104% 119% 103% 105%Debt/net assets 20.5% 28.2% 28.2% 21.6% 27.9% 27.3%Investments (TD million) 64 44.7 44.2 48.9 52.4 54.2

- 22 -

ONAS Performance Indicators (continued)INDICATORS 1998 1999 2000 2001 2002Connections (000) 725 765 805 994 1,060New connections (000) 35 40 40 189 66New connections due to the project (cumulative) 3500 4000 4500 5500 6180Population covered (000) 4,900 5,095 5,245 5,330 5,414New population covered (000) 199 195 150 85 84New connections due to the project (cumulative) 70 90 110 130 150Employees                       Permanent 3,246 3,421 3,714 4,063 4,236                                         Temporaries 940 1,133 1,313 1,487 1,148                                         Total 4,186 4,554 5,027 5,550 5,384Number of towns covered by ONAS 125 133 142 144 144Connection rate ONAS 79.0% 80.0% 81.0% 82.0% 83.0%SONEDE Water volume distributed (m3 million) 181 192 200 210 217ONAS Wastewater collected (m3 million) 161 163 181 187 194Average tariff (0.001 TD per m3) 348 392 380 394 395Operating ratio 91.0% 93.0% 97.0% 102.0% 110.0%Self-financing ratio (incl. "recette compensatoire") 31.0% 33% 23% 21% 15%Self-financing ratio (without "recette compensatoire") 15.0% 17% 8% 6% 4%Debt service ratio (without "recette compensatoire") 2.61 2.33 2.31 1.40 1.09Debt service ratio (incl. "recette compensatoire") 4.36 5.20 3.61 2.45 1.90Working capital 105.0% 104% 103% 102% 100%Debt/net assets 30.5% 32.3% 33.4% 37.5% 40.3%Investments (TD million) 67.7 83.5 90.2 96.1 102.5

SONEDE Performance IndicatorsINDICATORS 1993 1994 1995 1996 1997

Connections (000) 1,118 1,187 1,250 1,305 1,361Population covered (000) 6,365 6,560 6,718 6,876 7,030Employees 7,356 7,744 7,507 7,301 7,395

Permanent 5,195 5,360 5,420 5,405 5,738Temporaries 2,161 2,384 2,087 1,896 1,657

Employees / 1000 connections 6.6 6.5 6.0 5.6 5.4Water produced (m3 million) 305.8 322.3 310.5 307.8 319.1Water sold (m3 million) 223.2 237.5 230.4 229.3 245.4Unacounted For Water UFW (%) 27.0% 26.3% 25.8% 25.5% 23.1%Network extension (km) 1,445 1,440 1,247 1,222 1,066

In house % 20.0% 15.0% 12.0% 19.0% 17.0%Sub-contracting % 80.0% 85.0% 88.0% 81.0% 83.0%

New connections 72,039 66,913 65,839 57,689 58,477In house % 100% 90% 80% 70% 60%

Sub-contracting % 0% 10% 20% 30% 40%Average cost (TD per m3) 0.379 0.370 0.382 0.460 0.434Average tariff per m3 (without fixed connection fee) (TD) 0.381 0.392 0.324 0.381 0.392Average tariff (incl. fixed connection fee) (TD) 0.379 0.446 0.459Average tariff / Average cost (incl. fixed connection fee) 0.98 1.02 0.99 0.97 1.06Average tariff / Average cost (without fixed connection fee) 1.00 1.06 0.85 0.83 0.90Gross operating ratio (after depreciation & provision) 93.7% 93.1% 96.5% 100.5% 92.1%Net operating ratio (before depreciation & provision) 59.7% 62.1% 65.6% 68.2% 67.5%Self-financing ratio (without subscribers' contribution) 33.8% 4.9% 38.3% 22.1% 6.0%Self-financing ratio (including subscribers' contribution) 56.1% 25.8% 57.9% 44.4% 23.3%Debt service ratio (without subscribers' contribution) 1.27 1.22 1.54 1.37 1.26Debt service ratio (incl. subscribers' contribution) 1.54 1.45 1.84 1.69 1.49Current Asset / Current Liabilities 1.1 1.1 1.1 1.2 1.1Debt / Net Asset 35.4% 33.9% 31.8% 30.0% 32.9%Investments (TD million) 42 40 50 42 40

- 23 -

SONEDE Performance Indicators (continued)INDICATORS 1998 1999 2000 2001 2002

Connections (000) 1,419 1,477 1,548 1,623 1,706Population covered (000) 7,201 7,338 7,480 7,630 7,760Employees 7,350 7,244 7,119 7,110 7,038

Permanent 5,804 5,786 5,773 5,825 5,865Temporaries 1,546 1,458 1,346 1,285 1,173

Employees / 1000 connections 5.2 4.9 4.6 4.4 4.1Water produced (m3 million) 327.2 338.6 350.3 373.4 373.4Water sold (m3 million) 255.8 271.6 285.1 301.0 298.1Unacounted For Water UFW (%) 21.8% 19.8% 18.6% 19.4% 20.1%Network extension (km) 1,098 1,328 1,443 1,359 1,320

In house % 11.0% 9.0% 7.0% 8.0% 6.0%Sub-contracting % 89.0% 91.0% 93.0% 92.0% 94.0%

New connections 59,373 61,706 67,801 74,667 85,000In house % 57% 52% 49% 42% 37%

Sub-contracting % 43% 48% 51% 58% 63%Average cost (TD per m3) 0.462 0.462 0.471 0.477 0.505Average tariff per m3 (without fixed connection fee) (TD) 0.391 0.408 0.406 0.430 0.419Average tariff (incl. fixed connection fee) (TD) 0.459 0.477 0.474 0.500 0.494Average tariff / Average cost (incl. fixed connection fee) 0.99 1.03 1.01 1.05 0.98Average tariff / Average cost (without fixed connection fee) 0.85 0.88 0.86 0.90 0.83Gross operating ratio (after depreciation & provision) 98.5% 94.6% 96.3% 93.1% 98.4%Net operating ratio (before depreciation & provision) 71.7% 67.6% 69.6% 70.8% 74.6%Self-financing ratio (without subscribers' contribution) 22.9% 61.5% 50.8% 63.3% 41.5%Self-financing ratio (including subscribers' contribution) 39.1% 79.7% 64.4% 87.3% 61.7%Debt service ratio (without subscribers' contribution) 1.37 2.01 3.20 4.06 3.37Debt service ratio (incl. subscribers' contribution) 1.71 2.51 3.77 5.44 4.61Current Asset / Current Liabilities 1.1 1.1 1.0 1.1 1.4Debt / Net Asset 31.0% 37.2% 33.4% 34.8% 39.9%Investments (TD million) 50 70 70 65 75

Rural Water Supply IndicatorsNew connections in rural areas due to the project

As of December 31, 2002

As of June 30, 2003 (Latest Estimates)

New household connections because of the project 30,652 34,403New household connections because of SONEDE's extensions 5,273 5,940Total household connections 35,925 40,343New population connected because of the project 178,950 200,917New population connected because of SONEDE's extensions 30,663 34,568Total population connected 209,613 235,485New standpipe systems because of the project 44 44New standpipe systems  because of SONEDE's extensions 3 3Total standpipe systems 47 47New population covered by standpipes because of the project 8,869 8,869New population covered by standpipes because of SONEDE's extensions 258 258Total population covered by standpipes 9,127 9,127

Annual distribution of new connections in rural areasYear 1995 1996 1997 1998 1999 2000 2001 2002 2003 TotalNew Connections 5,074 5,230 4,081 6,519 4,542 4,243 3,756 2,480 4,418 40,343

Output indicatorsPolyethylene  Asbestos-cement Ductile Iron Total

Pipes (km) 795 871 21 1,687

- 24 -

Greater Tunis Water Supply Component IndicatorsGreater Tunis Total ConnectionsYear 1994 1995 1996 1997 1998 1999 2000 2001 2002 TotalNew Connections 17,119 13,516 16,545 14,378 16,518 19,341 19,528 22,764 139,709Total Connections 343,286 360,405 373,921 390,466 404,844 421,362 440,703 460,231 482,995 -% Increase 5.0% 3.8% 4.4% 3.7% 4.1% 4.6% 4.4% 4.9%

Output indicatorsDiameter 160 200 300 400 400 500 500 600 600 800 1000 1200 TotalType PEHD PEHD DI DI CO DI CO DI CO DI DI DIPipes (km) 0.100 0.600 10.986 3.000 0.300 26.945 2.200 5.093 16.400 0.800 1.800 8.500 76.724DI: Ductile Iron. CO: Concrete

Sewerage Component IndicatorsAppraisal Actual

New population covered 130,000 150,000WWTP 3 3WWTP total capacity (m3 per day) - 18,750Compact WWTP - 3Compact WWTP total capacity (m3 per day) - 7,450New sewerage pipes (km) 82 86New connections 4,547 4,750

- 25 -

Annex 2. Project Costs and Financing

Project Cost by Component (in US$ million equivalent)(*)

Component Appraisal Estimate US$

million

Actual/Latest Estimate

US$ million

Percentage of Appraisal

Water Supply in Greater Tunis (Loan 3783) 1) Water Treatment Plant 7.80 8.60 110%    - Ghdir El Golla (WTP) 5.45 7.12 131%    - Ghdir El Golla (Site stabilization) 2.34 0.27 12%    - Medjez El Bab (WTP) 0.00 1.21 2) Pipeline Restructuring 2.72 0.00 0%3) Rehabilitation of Ras Tabia Pipeline 9.34 7.16 77%4) Master Distribution Pipe 8.32 14.92 179%5) Reservoirs 4.02 3.02 75%6) Regulating Equipment 1.21 0.48 39%7) Institutional Strengthening and Demand Management (including water meters acquisition)

2.68 9.20 343%

8) Studies 2.92 1.37 47%Subtotal 39.01 44.75 115%Rural Water Supply (Loan 3782 SONEDE) Subtotal 32.05 32.02 100%Sewerage (Loan 3782 ONAS) Construction of Tozeur Wastewater Treatment Plant 3.48 3.06 88%Construction of Tataouine Wastewater Treatment Plant 4.23 2.63 62%Construction of Jammel-Zeramdine Wastewater Treatment Plant 5.35 3.74 70%Tozeur Sewerage Network 1.50 1.78 119%Tataouine Sewerage Network 2.50 1.88 75%Jammel-Zeramdine Sewerage Network 1.75 2.97 170%Construction of Mornag Wastewater Treatment Plant & Ghomrassen Sewerage Network

2.37

Construction of Boumerdes and Fernana Wastewater Treatment Plant

1.15

Studies for 4th Sewerage Project 4.64 4.05 87%Studies Rural Sanitation (appraisal did not include pilot project) 0.24 0.72 294%Geographic Information System 0.37 0.08 21%Technical Assistance 0.24 0.01 2%Training 0.12 0.09 73%Laboratory Equipment 0.24 0.24 101%Operating Equipment 0.22 Others 0.54 Subtotal 24.66 25.51 103%Total Project Cost 95.72 102.28 107%

(*) Costs do not include taxes and customs duties.

- 26 -

Project Cost by Procurement Arrangements (Appraisal Estimate) (US$ million equivalent)Procurement Method

(1)

Expenditure Category ICB NCB Other(2) NBF Total

CostTotal Project1. Works 6.45 27.78 0.00 8.93 43.16

(3.42) (13.89) (0.00) (0.00) (17.31)2. Goods 26.90 9.78 1.50 4.90 43.08

(22.73) (7.13) (1.35) (0.00) (31.21)3. Services 0.00 0.00 9.48 0.00 9.48

(0.00) (0.00) (9.48) (0.00) (9.48)4. Others 0.00 0.00 0.00 0.00 0.00

(0.00) (0.00) (0.00) (0.00) (0.00)Total 33.35 37.56 10.98 13.83 95.72

(26.15) (21.02) (10.83) (0.00) (58.00)

- 27 -

Project Cost by Procurement Arrangements (Actual/Latest Estimate) (US$ million equivalent)Procurement Method

(1)

Expenditure Category ICB NCB Other(2) NBF Total

CostWater Supply in Greater Tunis (Loan 3783)1. Works 0.73 9.92 0.07 7.12 17.83

(0.44) (5.11) (0.04) (0.00) (5.59)2. Goods 25.12 0.21 0.11 0.00 25.44

(19.83) (0.10) (0.08) (0.00) (20.01)3. Services 1.48 0.00 0.00 0.00 1.48

(1.42) (0.00) (0.00) (0.00) (1.42)4. Others 0.00 0.00 0.00 0.00 0.00

(0.00) (0.00) (0.00) (0.00) (0.00)Subtotal 27.33 10.13 0.18 7.12 44.75

(21.69) (5.21) (0.11) (0.00) (27.02)Rural Water Supply (Loan 3782 SONEDE)1. Work 0.00 13.95 1.32 0.00 15.27

(7.97) (3.41) (0.65) (0.00) (12.03)2. Goods 10.93 4.88 0.94 0.00 16.75

(0.00) (7.01) (0.66) (0.00) (7.68)3. Services 0.00 0.00 0.00 0.00 0.00

(0.00) (0.00) (0.00) (0.00) (0.00)4. Others 0.00 0.00 0.00 0.00 0.00

(0.00) (0.00) (0.00) (0.00) (0.00)Subtotal 10.93 18.82 2.26 0.00 32.02

(7.97) (10.43) (1.31) (0.00) (19.71)Sewerage (Loan 3782 ONAS)1. Works 1.95 1.87 0.00 8.67 12.49

(1.15) (1.13) (0.00) (0.00) (2.28)2. Goods 1.84 0.80 0.08 4.90 7.62

(1.29) (0.57) (0.00) (0.00) (1.86)3. Services 0.39 0.00 4.46 0.01 4.86

(0.39) (0.00) (4.45) (0.00) (4.84)4. Others 0.00 0.00 0.22 0.32 0.54

(0.00) (0.00) (0.00) (0.00) (0.00vSubtotal 4.18 2.67 4.75 13.90 25.51

(2.83) (1.70) (4.45) (0.00) (8.98)Total Project1. Workd 2.68 25.73 1.39 15.79 45.60

(9.56) (9.65) (0.69) (0.00) (19.90)2. Goods 37.90 5.89 1.12 4.90 49.81

(21.11) (7.69) (0.74) (0.00) (29.54)3. Services 1.87 0.00 4.46 0.01 6.34

(1.81) (0.00) (4.45) (0.00) (6.26)4. Others 0.00 0.00 0.22 0.32 0.54

(0.00) (0.00) (0.00) (0.00) (0.00)Subtotal 42.45 31.62 7.19 21.02 102.29

(32.49) (17.34) (5.87) (0.00) (55.71)

- 28 -

(1) Figures in parenthesis are the amounts to be financed by the Bank Loan. All costs include contingencies.(2) Includes civil works and goods to be procured through national shopping, consulting services, services of contracted staff of the project management office, training, technical assistance services, and incremental operating costs related to (i) managing the project, and (ii) re-lending project funds to local government units.

Project Financing by Component (US$ million equivalent)Appraisal Estimate Actual Estimate

Bank CoF. SONEDE ONAS Govt. Total Bank CoF. SONEDE ONAS Govt. TotalWater Supply in Greater Tunis (Loan 3783)Works 7.74 0.00 7.74 0.00 0.00 15.47 5.59 7.12 5.12 0.00 0.00 17.83Goods 15.25 0.00 2.69 0.00 0.00 17.94 20.01 0.00 5.44 0.00 0.00 25.44Services 5.52 0.00 0.09 0.00 0.00 5.60 1.42 0.00 0.06 0.00 0.00 1.48Others 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00Subtotal 28.50 0.00 10.51 0.00 0.00 39.01 27.02 7.12 10.62 0.00 0.00 44.75Rural Water Supply (Loan 3782 SONEDE)Works 7.75 0.00 0.00 0.00 7.75 15.49 12.03 0.00 0.00 0.00 3.24 15.27Goods 12.75 0.00 0.00 0.00 3.81 16.56 7.68 0.00 0.00 0.00 9.07 16.75Services 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00Others 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00Subtotal 20.50 0.00 0.00 0.00 11.55 32.05 19.71 0.00 0.00 0.00 12.31 32.02Sewerage (Loan 3782 ONAS)Works 1.63 6.25 4.31 0.00 12.19 2.28 8.67 0.00 1.54 0.00 12.49Goods 1.76 4.41 0.69 0.00 6.85 1.86 4.90 0.00 0.86 0.00 7.62Services 5.61 0.00 0.00 0.00 0.00 5.61 4.84 0.01 0.00 0.01 0.00 4.86Others 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.32 0.00 0.22 0.00 0.54Subtotal 9.00 10.66 0.00 5.00 0.00 24.66 8.98 13.90 0.00 2.63 0.00 25.51Total ProjectWorks 17.11 6.25 7.74 4.31 7.75 43.16 19.90 15.79 5.12 1.54 3.24 45.60Goods 29.76 4.41 2.69 0.69 3.81 41.35 29.54 4.90 5.44 0.86 9.07 49.81Services 11.13 0.00 0.09 0.00 0.00 11.21 6.26 0.01 0.06 0.01 0.00 6.34Others 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.32 0.00 0.22 0.00 0.54Total 58.00 10.66 10.51 5.00 11.55 95.72 55.71 21.02 10.62 2.63 12.31 102.29

- 29 -

Annex 3. Economic Costs and Benefits

Ex-post evaluation of Greater Tunis Water Supply component

The same methodology used in the ex-ante economic evaluation was used to evaluate the economic rate of return (ERR) at completion.  Real values for consumption, UFW and investment costs were used for the period 1995-2002.

The Urban Water Supply component comprised extension of water treatment capacity, reservoirs, pipes and equipments.  The component was aimed at expanding potable water production and distribution capacity and reducing the rate of unaccounted for water to meet a growing demand.  Without the project, the existing capacity was likely to become inadequate to meet all the demand (in 1995-1996 according to ex-ante evaluation). Some kind of rationing, which may take different forms would have been unavoidable (tariff increases, decrease in pressure, spot cutoffs, or even suspending service for some consumers or service expansion in areas like El-Menzeh and Ariana). With the project, new capacity was installed and UFW was reduced, which made it possible to meet the anticipated demand, and avoid rationing to consumers, particularly the urban poor.

The benefit of the project is the value for the consumer of the additional demand (volume of water) to be met by the project and which would have remained unmet without it.  This value is estimated quarterly taking into account three main assumptions:

(a) The benefit stream from the expansion of treatment capacity sub-component (i.e. increased water production) represents the benefits of the project up to the level of the demand.  This benefit already includes the contribution from the required accompanying measures to meet this level of demand such as increasing the capacity of the reservoirs and strengthening the distribution network.(b) The benefits accruing from the project are quantified in terms of the volume of additional water effectively used by consumers.  This quantification is based on the expressed demand that will be met by the project and would not be without it.  Demand was estimated in the ex-ante evaluation by projecting through 2030 the trend observed over the period 1981-92 (quarterly data), which gave an average annual growth rate of 2% over this period.  The ex-post evaluation used the real consumption values for the 1995-2002 period, and projected the 2002 values using a 2% annual increase.  The projection is done for each quarter so as to allow for seasonal fluctuations. The elasticity of demand used in the ex-post evaluation is the same as the one used at appraisal.  The latter was calculated on the basis of an econometric model using the quarterly statistics for the period 1981/Q1 through 1992/Q4, and is set at 0.68, a relatively high figure, which is for all water uses combined.  (c) The value of the project contribution is estimated on the basis of the consumers' willingness to pay.  This concept refers to the price that consumers are willing to pay for one cubic meter of water in different situations, with and without the project.  The price Po, which represents the price consumers are willing to pay for the last cubic meter of water without the project, is calculated as follows, using the elasticity of demand in Greater Tunis:

Po = Ps[1-(1/e)*(Qo-Qs)/Qs]

Where e is the price elasticity of demand, Ps is the average tariff per m3 paid by the consumers for the current capacity.  Qo is the quarterly production capacity of the system without the project, and Qs is the quarterly demand that can be met with the project.

- 30 -

The total cost of the Greater Tunis project has been estimated in constant 1994 prices, including physical contingencies in the ex-ante evaluation, but excluding taxes and custom duties. The project cost was adjusted to take into account: (i) real border prices by using an exchange rate factor to convert official exchange rate into real values (this factor is equal to 1.2 in the base scenario, which means the official exchange rate TD/US$ is 20% higher than the border exchange rate); and (ii) shadow wage for unskilled labor in Tunisia by using a wage factor (this factor is equal to 0.8 in the base scenario, which means the shadow wages are 80% of the official wages).  Operating costs in economic values were estimated at TD 0.179 per m3.  The same value was used in the ex-post evaluation.

With these assumptions, the ERR for the Greater Tunis project has been estimated at project completion at 18.5 percent compared to 16.6 percent at project appraisal.  This higher value of the ERR could be explained by two factors: (i) the net present value of the project cost over the period 1995-2030 is slightly lower than what was estimated at appraisal; and (ii) real water demand was lower than projected in 1995 and 1996, while it was on average 7 percent higher during the rest of the project implementation period.  The ex-ante assumed that the bulk part of the investment would occur in the 1996-1998 period.  In fact, the real investment was spread over a longer period (1996-2003).  In other words, the net present value of project costs in the period 1995-2030 at a discount rate of 12 percent were estimated in the ex-ante evaluation at TD 39.7 million, whereas in the ex-post evaluation they were evaluated at 37.8.  The water demand in 1995 and 1996 was 0.5 and 3 percent respectively than the ex-ante projections.  Thus, the saturation of the system that was predicted in the ex-ante assumption to occur in 1996 was avoided thanks to the improvement in UFW.  Real water demand in 1997 and 1998 was slightly higher than that of the ex-ante evaluation, and 7 to 11 percent higher afterwards, which leads to an average 8 percent higher over the projection period 2003-2030.  Since the project benefits reflect the additional demand that the “without project” situation cannot satisfy, a higher demand led to higher benefits and a higher ERR. 

Sensitivity analysis was conducted to capture the impact of the analysis assumptions.  The following tables summarize the final results.  An Excel datasheet is kept in project files.

Water supply consumption in Greater Tunis (million m3) 1995 1996 1997 1998 1999 2000 2001 2002Ex-ante projections 76.0 77.5 79.1 80.7 82.3 84.0 85.7 87.4Real consumption 75.6 75.3 80.2 82.9 88.0 91.2 95.2 94.9Difference -0.5% -2.9% 1.4% 2.7% 6.9% 8.6% 11.1% 8.5%

Project economic investment costs (TD million) 1995 1996 1997 1998 1999 2000 2001 2002 2003Ex-ante estimates 1.6 11.9 11.9 10.1 6.4 0.0 0.0 0.0 0.0Ex-post real costs 0.0 5.3 2.0 2.7 8.9 8.0 6.1 8.2 7.9

Sensitivity analysis

Operating costsIncrease in economic operating

costs per m3 estimate (%)Economic rate of Return (%)

Ex-post0 18.510 18.020 17.430 16.940 16.350 15.7

- 31 -

Shadow Exchange RateExchange rate factor Economic rate of Return (%)

Ex-postEconomic rate of Return (%)

Ex-ante1.0 18.0 17.81.1 17.4 17.21.2 16.9 16.61.3 16.3 16.1

Shadow WageWage rate factor Economic rate of Return (%)

Ex-postEconomic rate of Return (%)

Ex-ante0.6 19.2 17.20.7 18.8 16.90.8 18.5 16.60.9 18.2 16.31.0 17.9 16.1

Water Demand ElasticityElasticity Economic rate of Return (%)

Ex-postEconomic rate of Return (%)

Ex-ante-0.42 21.7 19.8-0.50 20.4 18.5-0.68 18.5 16.6-0.70 18.3 16.5-0.80 17.6 15.8-0.90 17.1 15.3-1.00 16.6 14.9

In a pessimistic scenario where the elasticity is taken equal to -1.0, exchange rate factor equal to 1.3 and shadow wage factor equal to 1.0, the ERR is estimated at 15.6% compared to 14.0% at appraisal.  In this scenario, if the operating costs per m3 were increased by 50%, the ERR falls to 12.7% (no value was estimated at appraisal).

Ex-post evaluation of the Rural Water Supply component

Introduction

The Water Supply and Sewerage Project included rural water supply sub-projects that should be selected on the basis of a criterion of social profitability different from the criterion which had been used before the start of this Project. Prior to the latter, the criterion used by SONEDE was the amount of investment per capita. A regression line was estimated for the relationship between per-capita investment and population served. As expected, the slope of this line was found to be negative, implying that the larger the population, the smaller is per capita investment. The criterion, which also allowed for some margin above the estimated line, consisted of accepting a project if its per-capita investment fell within the margin above the regression line.

Because this criterion was thought to be crude and not well grounded in the framework of economic project appraisal, an alternative criterion was designed under the Project, that of the net social gain. The new

- 32 -

approach has been applied to the assessment of all rural sub-projects that SONEDE envisaged under the Project. A standardized evaluation spreadsheet was prepared for each sub-project using small software developed at SONEDE with Bank assistance. For each project, assessment was based on estimates of the population to be covered, as well as of investment and operation cost. Since the numbers of rural sub-projects and of the population targeted under the Project have already been reached, it is important to assess how the methodology has been used and whether actual benefits and costs enable the projects already implemented to pass the selection criterion ex post. This ex post evaluation is based on a sample of 29 rural centers out of the 459 centers that have been financed under the Project. The centers are located in different regions of the country so as to have the widest representation of the country possible. Methodology The evaluation model designed for this purpose relies on the assessment of economic benefits and costs of the project, considering the state of meeting water needs in the two situations of with and without the project. Benefits of the project for households are:

avoidance of the cost of water supply under the initial conditions without the project, lwhich includes both current and investment expenses;larger volume of water consumed at lower prices. The increase in consumption and the ldifference in cost depend on whether the household will be individually connected or will use a public standpost.

These benefits are supposed to grow as population and per capita income rise. Households also incur a fixed cost when connected, which is the connection fee. Additional benefits due to higher volumes of water and lower purchase price are captured by a consumer surplus that grows in value with population and income per capita. For the household, the net benefit is equal to the difference between this surplus and the differential in fixed cost in the with and without project situation. Rural projects are subsidized by the state, as revenues are not expected to cover total investment and operating cost. The cost of the subsidy, properly adjusted to reflect the social cost of public funds, should therefore be netted out from the value of household benefit to get the net social benefit of the project. The final decision criterion is to implement the project if the net social benefit (in absolute terms or per cubic meter) is positive (formula below). Application of the approach by SONEDE Small software was designed and put in place at SONEDE in order to facilitate the application of this methodology. The software distinguishes between parameters, the values of which should depend on local conditions and estimated investment cost for the rural area to be served, and computational steps leading to the estimate of net social gain. Main parameters involved are the social discount rate, the rate of growth of water consumption, the opportunity cost of time spent on water supply in the situation without the project, the minimum rural wage, social weights for the welfare of rural households and the opportunity cost of public funds. Before implementation of the methodology SONEDE staff were trained to use the software. 

Overall, the software has been correctly used, and the net social gain has been properly assessed. Values were set for the parameters of the model in line with what the model requires. However, while the values of

- 33 -

these parameters are meant to vary according to local or at least regional conditions for each sub-project, the values used in SONEDE’s evaluation hardly vary from one sub-project to another. The same values were indeed used for time, growth of consumption, the operation cost of water under the project, as well as for the social weights associated with the increase in household welfare. Before proceeding with ex-post project appraisal and the comparison between the latter and ex-ante appraisal, the following points can be made concerning the use that has been made of the approach:

The net social gain is very sensitive to the value of time spent on water provision in the lsituation without project. It can vary widely between high positive values and negative values.Including a cost other than the value of time of acquiring water, the price, which is assumed to lbe TD 0.189, TD 0.300 and TD 0.666 depending on the project, is not justifiable when the household is assumed to travel a long distance to acquire water. Including such additional cost overestimates the cost of water in the without project situation and therefore increases its economic return.  The investment cost in the initial situation without the project is very low (around TD 0.003 to lTD 0.004).The social gain is large due to the saving in time spent on getting water from distant sources or lalternatively to buy water from water vendors. In the case of projects supplying water to public outposts without house connections, perhaps time saved is lower than assumed in the computations. Nevertheless, even in this case it remains the main component of the project benefits.Consumption of water was initially set according to norms: 73 m

3

in the North, 67 m3

in the l

Center and 120 m3

in the South, not on the basis of observed consumer behavior under similar projects.The fact that operation cost is the same for all areas means that it has been determined at the lcentral level. In the future, SONEDE regional districts should be more involved in providing information for the project. Operation cost without the project: it is not clear how the figures were chosen. In some cases l

they are too high, as high as TD 12 per m3

. The question may be raised concerning the time involved in bringing water back home, which is often high in SONEDE estimates, knowing that households may have water tanks enabling them to stock water for a few days.

Ex post evaluation

The table below gives a summary of the evaluation results and a comparison with ex ante evaluation. This table brings out the following remarks:

The population covered by the project turned out to be larger than initially estimated at the ltime of the appraisal in 23 rural areas out of a sample of 29 areas. If the sample is representative, that means that the population is higher than initially estimated in 80% of the cases. The projects served on the average 37% more people than initially estimated.

Consumption: per capita consumption was initially set at levels equal to norms set for three l

regions: 73 m3

per year for a household of five living in the north, 67 m3

for the center and 120 m

3

for the south. No reference was made to average actual consumption recorded for similar connected rural areas. The average volume consumed turned out to be lower than expected in about 50% of the cases and higher than expected in the other half. The shortfall with respect to expectations may be due, in principle, to delays in connections, but this cannot account for the

- 34 -

gap in this case since more people have had access to SONEDE water than initially expected. On average, actual consumption is very close to the projected one, estimated at about 88 m

3

/year/household against a volume of 91 m3

in initial project appraisal. Finally, total actual consumption turned to be about 38 percent higher than in the initial estimates, the larger population effectively covered accounting for the discrepancy.

Operation cost under the project: this cost includes materials, labor cost, financial cost, etc., lwith labor being the dominant item. This cost is assumed to be the same for all rural projects, set at TD 0.508 per m

3

. The standardization of this cost implies that it was set at the central level with very little input, if any, from the regional districts. Since SONEDE has not developed cost accounting yet, particularly at the district level, it is difficult to assess the bias resulting from assuming the same operation cost everywhere.

Investment cost: based on the sample of 29 rural projects the actual investment cost is lsignificantly lower than assumed in appraisal, with the exception of one single project for which actual cost is higher than initially projected. The actual average cost of the project averaged TD 240,000 against an expected cost of TD 317,000, thus resulting in an average discrepancy between the two amounts equal to about 25 percent of initial cost. Thus a larger rural population has been covered at a lower cost than expected.  The investment cost per m

3

is about the same in the initial estimates as after the implementation of the project, equal to about TD 0.531, which is almost the same as the operation cost of the project. Generally in water projects, investment cost is higher than the latter, but because some of these projects supply water through public outposts and not house connections, capital cost tends to be lower, in relative terms, than for urban water supply projects.  Thus, the total cost per m

3

, capital and operating cost, is estimated at about TD 1.040 for the period 1996-2000.

Cost of water supply without the project: the investment component of this cost includes water ltanks and eventually transportation animals. The operation cost includes the value of household’s time spent on bringing water to the house as well as expenses on the purchase of water. While investment cost is generally low, current cost is far higher than under the project, averaging over TD 6 against TD 0.5 under the project and exceeding TD 12 per m

3

in some cases. Actual cost also exceeds the initially estimated cost by a margin of 42 percent, which results in a larger net social gain since the cost of water supply without the project becomes a benefit when the project is implemented.

Since this cost has a high incidence on the value of net social gain, it is important to assess it in the most reliable way. A sensitivity analysis is carried out by reducing by one half the cost of time assumed in SONEDE initial project appraisal.  The sensitivity analysis also removes the simplifying assumption of an infinite lifetime of the project and replaces it with a twenty-year lifetime.

Outcome of the net social gain criterion of project selection: As initially assessed by SONEDE, the 29 rural projects fulfill the criterion of positive net social gain, with the latter varying between 0.350 and TD 3.7 per m

3

and averaging TD 1.830. Calculated net social gain is negative (TD –0.104) for Balta’s project while SONEDE’s record shows a positive value of TD 0.350.  The ex-post evaluation yields much larger net gains, averaging TD 3.760 and exceeding in some cases TD 7. These significant differences are due to the much higher cost of water supply without the project in SONEDE’s ex-post evaluation.

The rationale behind the changes introduced in the cost of meeting water needs in the absence of the project is not explained and it is not clear whether it is justifiable or not. We simulate the effect of a cut by one half

- 35 -

in the cost of time spent on water provision. The result is a drop in the average value of net social gain to TD 0.940 per m

3

, and an even larger drop from positive to negative values for five rural areas, which would have resulted in the rejection of the corresponding projects.

Conclusion

The ex-post assessment of 29 rural sub-projects that have been implemented under the Project shows higher net social benefits than those initially estimated during the ex-ante appraisal phase. Therefore, these projects meet the net social gain criterion ex-post, as they met it ex-ante. Differences are due to generally larger population covered and lower investment cost than initially projected. They are also due to a significantly higher cost of acquiring water in the situation without the project than in the ex-ante assessment. Cutting the time spent on water provision by one half shifts five out the 29 projects from socially profitable with positive net social gains to socially unprofitable with negative social gains. The methodology is sound and provides a good guide for selection of rural projects, but cost components, particularly those of water supply in the situation without project, need to be defined and assessed more reliably. 

Discrepancy between Projected and Actual Values Average

ex-anteAverage ex-post

Discrepancy (%)

Proportion of ex post higher than ex-ante

Population (inhabitants) 1,577 2,164 + 37% 80%Consumption per household 91 88 - 3.3% 50%Cost under the project

Operation cost (TD/cum) 0.508 0.508 0% -Investment cost (TD/cum) 0.531 0.531 0% 40%Total investment (TD) 317,600 240,000 -25% 17%Investment per household 1,480 1,213 -20% 24%

Cost without the project (TD/cum) 4.260 6.060 +42% 96%Net social gain (TD) 1.830 3.760 +100% 93%

Note:

where: [ ] SGnGnnnNSG BA βα −++= )()/(1 2121

n1 and n

2  are number of households respectively house-connected and users of public outposts, NSG is net

social gain (a measure of net change in welfare induced by the project), GA is the net change in

house-connected household (consumer surplus minus the change in fixed cost after connection), GB is the

net change in welfare for the user of public outpost, S is government subsidy. The coefficients α and β are the social weights associated respectively with the welfare of the project beneficiaries and public funds. 

The net change in house-connected household welfare is: 2/))(()( 10101 QQPPaIIG OA +−+−−=

where GA is net welfare gain, I

1 household investment under the project, I

0 household investment in the

situation without project, P0 is the cost of one cum of water without project and P

1 is SONEDE price of

water. Q0 and Q

1 are respectively the quantities of water consumed by the household in the without and with

project situations. The coefficient (a) is the difference between the real interest rate and the rate of growth.

- 36 -

Annex 4. Bank Inputs

(a) Missions:Stage of Project Cycle Performance Rating No. of Persons and Specialty

(e.g. 2 Economists, 1 FMS, etc.)Month/Year Count Specialty

ImplementationProgress

DevelopmentObjective

Identification/Preparation2/11/1993 3 FINANCIAL ANALYST/TASK

TEAM LEADER (1); SANITARY ENGINEER (CONS) (1); ENV. SPECIALIST (CONS) (1)

5/11/1993 3 FIN. ANALYST/TASK TEAM LEADER (1); ENGINEER (1); ECONOMIST (1)

10/5/1993 4 FIN. ANALYST/TTL (1); ECONOMIST (1); ENGINEER (1); DIVISION CHIEF (1);

Appraisal/Negotiation12/24/1993 6 FIN. ANALYST/ TTL (1);

ECONOMIST (1); ENGINEER (1); ECONOMIST (CONS) (1); ENGINEERS (CONS) (2)

Supervision10/28/1994

4/06/1995

1

2

FIN. ANALYST/ TTL (1)

FIN. ANALYST/TTL (1); ENGINEER (1)

HS HS

04/13/1996 2 CONSULTANT (2) S HS03/25/1997 2 FIN. ANALYST/TTL (1);

SANITARY ENGINEER (1)HS HS

10/23/1997 2 FIN. ANALYST/TTL (1); SANITARY ENGINEER (1)

HS HS

05/21/1998 3 FIN. ANALYST/TTL (1); SANITARY ENGINEER (1); ECONOMIST (1)

HS HS

06/11/1999 4 FIN. ANALYST/TTL (1); ECONOMIST/FIN. ANALYST (1); ENGINEER (2)

HS HS

11/12/1999 1 SR. SANITARY ENGINEER (1) HS HS02/18/2000 4 TASK TEAM LEADER (1); SR.

SANITARY ENGINEER (1); ECONOMIST/FIN ANALIST (1); PRIN. FIN. ANALYST (1)

HS HS

06/29/2000 3 ECON/FIN. ANALYST (1); SR. SAN. ENGINEER (1); CONS. (1)

S HS

10/20/2000 3 MISSION LEADER (1); FINANCIAL ANALYST (1); CONSULTANT (1)

S HS

- 37 -

03/09/2001 3 MISSION LEADER (1); FINANCIAL ANALYST (1); CONSULTANT (1)

S HS

10/27/2001 2 SR. SANITARY ENGINEER/ TTL (1); SR. ECONOMIST (1)

S HS

05/02/2002 4 SR. INFRASTR. SPEC./TTL (1); SR. SAN. ENGINEER (1); ECON./FIN. ANALYST (1); SAN. ENGINEER (CONS.) (1)

S HS

1/22/2003 5 SR. INFRASTR. SPEC./TTL (1); LEAD FINANCIAL ANALYST (1); SAN. ENGINEER (CONS.) (1); ENV. SPECIALIST (1);ENGINEER (1)

S S

ICR7/15/2003 4 SR. INFRASTR. SPEC./

TTL (1); SAN. ENGINEER (CONS.) (1); ENV. SPECIALIST (CONS.) (1); ENGINEER (1)

S S

(b) Staff:

Stage of Project Cycle Actual/Latest EstimateNo. Staff weeks US$ ('000)

Identification/PreparationAppraisal/Negotiation N/A 454Supervision N/A 539ICR N/A 25Total 1,018

- 38 -

Annex 5. Ratings for Achievement of Objectives/Outputs of Components(H=High, SU=Substantial, M=Modest, N=Negligible, NA=Not Applicable)

RatingMacro policies H SU M N NASector Policies H SU M N NAPhysical H SU M N NAFinancial H SU M N NAInstitutional Development H SU M N NAEnvironmental H SU M N NA

SocialPoverty Reduction H SU M N NAGender H SU M N NAOther (Please specify) H SU M N NA

Private sector development H SU M N NAPublic sector management H SU M N NAOther (Please specify) H SU M N NA

- 39 -

Annex 6. Ratings of Bank and Borrower Performance

(HS=Highly Satisfactory, S=Satisfactory, U=Unsatisfactory, HU=Highly Unsatisfactory)

6.1 Bank performance Rating

Lending HS S U HUSupervision HS S U HUOverall HS S U HU

6.2 Borrower performance Rating

Preparation HS S U HUGovernment implementation performance HS S U HUImplementation agency performance HS S U HUOverall HS S U HU

- 40 -

Annex 7. List of Supporting Documents

1. Staff Appraisal Report, Republic of Tunisia, Water Supply and Sewerage Project, World Bank Report. No. 12939-TUN, July 6, 1994 2. Loan Agreement with Republic of Tunisia, September 21, 1994 3. Guarantee Agreement with Republic of Tunisia, September 21, 1994 4. Project Agreement with ONAS, September 21, 1994 5. Missions aide-memoires, PSRs and BTOs:

- Pre-Identification mission, January 24-February 11, 1993- Identification mission April 19-May 11, 1993- Pre-appraisal mission, September 9-October 5, 1993- Appraisal mission, December 5-24, 1993- Supervision mission, October 16-28, 1994- Supervision mission, March 26-April 6, 1995- Supervision mission, March 25-April 13, 1996- Supervision mission, March 6-25, 1997- Supervision mission, October 6-21, 1997- Supervision mission, May 7-21, 1998

- Mid-Term review mission, November 9-23, 1998- Supervision mission, May 24-June 11, 1999- Supervision mission, June 14-29, 2000- Supervision mission, February 21-March 9, 2001- Supervision mission, October 16-28, 2001- Supervision mission, April 15-May 1, 2002- Supervision mission, January 6-22, 2003- ICR mission, June 29-July 15, 2003

6. QAG Report 1998 7. Rapport d'achèvement du projet. République tunisienne, SONEDE et ONAS. Septembre 2003. 8. Contrats Programmes SONEDE-ETAT 1997-2001 and 2002-2006. 9. SONEDE Rapports Statistiques: 2001 and 2002. 10. ONAS Project implementation monitoring reports: December 2002 and April 2003. 11. SONEDE Project implementation monitoring reports: December 2002 and June 2003. 12. Etude économique sur l'eau potable en Tunisie. SONEDE, Lahouel and al. December 1994. 13. Rapport d'Achèvement du Projet Eau Potable et Assainissement. République Tunisienne: Ministère du Développement et de la Coopération Internationale; Ministère de l'Agriculture, de l'Environnement et des Ressources Hydrauliques; SONEDE; ONAS. Septembre 2003.

- 41 -

Additional Annex 8. Borrower's Contribution to ICR

1.Préambule

1. L’objet du présent rapport est de passer en revue les différentes étapes d’évaluation, de mise en œuvre, d’exécution et d’achèvement du projet "Eau potable et Assainissement".

2. A cet effet, deux prêts de la Banque de 29 millions de dollars chacun ont été accordés respectivement au Gouvernement Tunisien (prêt 3782 avec une composante alimentation en eau potable en milieu rural pour 20 millions de dollars et une composante assainissement pour 9 millions de dollars) et à la SONEDE (prêt 3783 pour le Grand Tunis).

3. Pour le projet de l’ONAS, deux agences d’aide multilatérale ont participé au financement, il  s’agit de la Banque Nordique d’Investissement (NIB) et du Fonds Nordique de Développement (NDF) pour un montant total de 7 millions de dollars. (annexes 1 et 2 )

4. Pour le secteur de l’eau potable, la collaboration de la BIRD est devenue une tradition qui date de la création de la SONEDE en 1968 et totalise une dizaine de prêts ayant contribué à  des taux de desserte élevés pour la Tunisie atteignant en 2002, 100% en milieu urbain et 83% en milieu rural.

5. Pour aider le Gouvernement Tunisien à atteindre les objectifs d’assainissement, la BIRD a consenti également en faveur de ce secteur 6 prêts pour la réalisation de plusieurs projets depuis la création de l’ONAS en 1974. (annexe 3).

6. Approuvé par le conseil d’administration de la Banque le 28 juillet 1994, mis en vigueur le 18 janvier 1995, le prêt a été clôturé le 30 juin 2003 après un an de prorogation.            

2. Objectifs

2.1 Objectifs d’origine

7. Les objectifs du projet tels qu’ils ont été définis dans l’accord de prêt étaient de :

a. Mettre en œuvre un programme approprié de gestion de la demande et de renforcement institutionnel pour accroître l’efficacité de l’exploitation de la SONEDE et de l’ONAS et renforcer leurs capacités techniques et financières et ce  par la formation du personnel, les études, l’assistance technique, l’acquisition de matériel de mesure, une campagne de sensibilisation du public en vue d’une rationalisation dans l’usage de l’eau et des services d’assainissement.

b. Fournir aux zones urbaines et rurales des services d’alimentation en eau et d’assainissement qualitativement et quantitativement satisfaisants et facilement accessibles par l’expansion et la réhabilitation d’équipements existants, en particulier pour les ménages à faible revenu , ce qui se traduit par : la remise en état et l’extension des réseaux d’approvisionnement en eau pour le Grand Tunis (cette composante étant financée par le prêt 3783-TUN), le raccordement de 320 localités rurales regroupant 200.000 habitants aux réseaux SONEDE  et l’amélioration des conditions de vie en matière d’assainissement de plus de 130.000 habitants répartis sur plusieurs villes (financé par le prêt 3782-TUN).

c. Promouvoir la réutilisation des eaux épurées dans l’industrie et l’agriculture.

- 42 -

d. Encourager la SONEDE et l’ONAS à sous traiter certaines de leurs activités au secteur privé. 

8. Ces objectifs reflétaient les principales priorités de la Tunisie en ce qui concerne le développement du secteur. Ils étaient cohérents avec la stratégie du Gouvernement pour faire face aux besoins du pays jusqu’à 2010 et d’éviter les risques de  pénuries d’eau.

 2.2 Objectifs révisés

9. Les objectifs du projet tels qu’ils ont été définis dans l’accord de prêt n’ont pas été révisés. Toutefois, certaines composantes ont été revues sans pour autant compromettre la réalisation des objectifs.

2.3 Degré de réalisation des objectifs

10. Dans l’ensemble, le projet a atteint ses objectifs :

a. Au niveau du Grand Tunis, bien que le prêt soit clôturé le 30/06/03, toutes les composantes ont été engagées avant cette date. Au 30/06/03 l’ensemble des composantes a été réalisé avec un pourcentage d’exécution moyen de 90%, le reste est en cours d’exécution par les fonds propres de la SONEDE.

b. Pour les zones rurales, 461 localités ont été approvisionnées en eau potable contre 320 initialement prévues, faisant bénéficier directement 195.000 habitants par des branchements individuels et ayant permis de desservir une population additionnelle de 25.000 habitants par bornes fontaines et ce en recourant à des sources de financement autres que celles de la Banque. Ainsi la population totale desservie a atteint 220.000 habitants.

c. Pour le secteur de l’assainissement, le projet a permis le raccordement au réseaux de plus de 150.000 habitants contre 130.000 prévus initialement.

d. En matière de réutilisation des eaux épurées , celles de la station de Tozeur  seront réutilisées pour l’ irrigation d’un terrain de golf situé dans la zone touristique de « RAS EL AIN » d’une superficie de 65 hectares et dont l’aménagement est en cours. Les travaux de transfert des eaux épurées vers les terrains de golf d’un coût d’environ 1,5 millions de dinars seront à la charge du Ministère de Tourisme, du Commerce et de l’Artisanat (1 MDT) et de l’ONAS (0,5 MDT).

e. Sur le plan institutionnel plusieurs actions ont été menées par la SONEDE et l’ONAS dont notamment :

i. La mise en place, des deux modules finances et  approvisionnements & gestion des stocks du progiciel standard intégré JDEdwards OneWorld. Outre le partage et l’accès rapide à l’information par tous les utilisateurs, cet outil a rendu la justification des comptes plus aisée, l’information comptable et financière de meilleure qualité, l’analyse des coûts par centre de coûts ou par activité plus développée et la gestion des approvisionnements et  les stocks plus maîtrisés.

ii. la composante études SONEDE comprenait des études sur (i) le Plan Directeur d’alimentation en eau potable du Grand Tunis, qui est achevé sauf la 5ème

mission ; (ii) le Plan Directeur du Sud Tunisien, arrêté à sa 2ème

mission, (iii) l’étude sur la sous-traitance de certaines tâches de la SONEDE, aussi achevée. La SONEDE a décidé de surseoir au lancement de la 2ème partie de l’étude du développement de la participation du secteur privé et se consacrer à la mise en application  des  recommandations de l’étude sur la sous-traitance. En outre la

- 43 -

SONEDE a confié au privé la réalisation des extensions des réseaux (94%), l’exécution des branchements (63%) , l’activité gardiennage et certaines études de plans directeurs. La SOENEDE envisage également de réaliser une station de dessalement d’eau de mer éventuellement selon la procédure BOT.

iii. Concernant la sous-traitance des activités d’assainissement, l’ONAS a délégué l’exploitation de la station d’épuration de Tataouine, réalisée dans le cadre du projet, au privé avec 95 km de réseau. Le contrat de délégation d’une durée de 5 ans est entré en vigueur le 01/03/03. l’office est également en face de dépouillement des offres pour la délégation de l’exploitation des stations d’épuration de Tozeur et Nafta au privé. Enfin le dossier d’appel d’offres pour la délégation de l’exploitation  au privé de trois autres stations d’épuration (à savoir Jammel – Zarmdine, Monastir – Frina et Saiada – Lamta) est bien avancé. Le programme de l’ONAS prévoit également d’impliquer le secteur privé dans l’exploitation de 1800 km et 26 stations d’épuration.

iv. Deux autres études sont en cours de réalisation :1. une étude organisationnelle portant sur la restructuration des activités régionales. Cette étude permettra notamment de décentraliser certaines activités du siège vers les régions. 2. une étude sur la restructuration des tarifs de l’eau qui s’achèvera avant la fin de l’année 2003.

v. Des actions  de formation parmi les composantes du prêt ont été prévues ; cependant, la SONEDE en recourt à d’autres sources de financement pour réaliser son programme de formation.

3. Composantes du projet

3.1 Composantes d’origine

11. A) Alimentation en eau des zones urbaines : Cette composante concerne :

a. l’extension de la capacité  de traitement de la station de Ghdir el Goula, par la construction d’une nouvelle unité de traitement et la réalisation des travaux de confortement du site.b. la réhabilitation de certaines sections de conduites, le remplacement de 8.5 km de la conduite · 1200 Ras Tabia–Bir Kassâa hors service à l’époque,  le raccordement de collectivités par l’installation de 29 km de conduites .c. la Construction de réservoirs totalisant une capacité de 25.000 m

3 situés dans le

Grand Tunis.d. l’installation de diverses conduites nécessaires pour relier ces réservoirs et améliorer la distribution de l’eau dans le Grand Tunis (amélioration du comptage, des conduites d’amenée).

12. B) Alimentation en eau des zones rurales : Exécution de sous projets afin d’approvisionner 200.000 habitants répartis sur 320 localités.

13. C) Composante Assainissement a. la construction de 3  stations d’épuration à Tozeur, Tataouine et Jammel-Zeramdine.b. la pose d’environ 82 km de conduites et le raccordement de 4.547 logements au réseau public d’assainissement.

- 44 -

14. D) Renforcement institutionnel : Cette composante a pour objectif de permettre à la SONEDE et à l’ONAS d’améliorer la gestion et ce par :

a. l’organisation d’une campagne de sensibilisation des abonnés de la SONEDE pour une meilleure utilisation de l’eau : spots télévisés, diffusion  des brochures et  production d’un guide de l’usager.

b. Des actions de formation du personnel de la SONEDE.

c. L’acquisition par la SONEDE du matériel nécessaire à la détection des fuites d’eau et des appareils de mesure, de télégestion, et de télécommande.

d. l’acquisition de matériel spécialisé pour les études et de matériel de laboratoire nécessaire pour contrôler et suivre le traitement des eaux usées domestiques et industrielles.

e. la fourniture de service d’assistance technique nécessaire à travers :i. la mise à jour des plans directeurs d’eau potable du Grand Tunis et du Sud Tunisien.ii. l’élaboration des études d’exécution du quatrième projet d’assainissement urbain,iii. l’élaboration d’une étude permettant d’identifier des techniques d’évacuation et de traitement des eaux usées adaptées aux zones rurales.iv. la mise en place d’une base de données cartographique sur les réseaux d’assainissement (GIS).v. l’examen des options concernant la participation du secteur privé dans les services d’alimentation en eau potable.

3.2  Composantes révisées

15. Il est à rappeler que le projet a été évalué sur la base d’études sommaires réalisées en 1994. Au cours de l’élaboration des études d’exécution et durant la réalisation du projet, certains ajustements (des ajouts et des transferts) ont été effectués en concertation avec la Banque.

3.2.1  Alimentation en eau potable de zones urbaines

16. Les révisions et les principaux ajustements ont touché les composantes suivantes :

a. La station de traitement de Ghdir El Goula a été construite moyennant un financement du Fond Koweïtien de Développement Economique Arabe (FKDEA). Une partie du montant réservé a cette rubrique a été utilisée pour financer la construction d’une nouvelle station de traitement de 400 l/s. Cette station a permis d’avoir une ressource supplémentaire pour le Grand Tunis et d’exploiter d’une façon rationnelle les capacités des conduites de transfert de Oued Ellil et Kasseb. L’autre partie du montant a été utilisée dans les autres rubriques.

b. La réhabilitation de conduites d’amenée d’eau a été supprimée ; le montant y afférent a été employé dans la rubrique « conduites maîtresses » et dans les autres composantes essentiellement celles relatives au renforcement institutionnel et équipement de comptage.

3.2.2  Composantes eau potable rurale

17. Le nombre de groupements retenus est de 461 contre 320 initialement prévus. Soit 141 groupements en plus , ce qui représente une augmentation de 44%. Ces groupements totalisent

- 45 -

une population de 195.000 habitants, les extensions du réseau de ces localités ont permis par d’autres fonds, de desservir une population additionnelle de près de 25.000 habitants, ramenant ainsi la population totale desservie à 220.000 habitants.

18. Sur les 461 groupements, 459 ont été alimentés. Pour les deux autres projets restants, les travaux sont achevés à 99% pour le premier (projet EL Hoshassia situé au gouvernorat de Siliana). Pour l’autre projet celui d’El Herri situé au gouvernorat de Béjà, le prêt  BIRD n’a financé que la composante fourniture de conduites. Les travaux de pose seront réalisés sur d’autres sources de financement.

19. le nombre d’abonnés généré par le projet au terme de l’année  2002 est de 35.925 dont 47 abonnés GIC qui correspondent à la population additionnelle de  25000 habitants.

3.2.3  Composante assainissement Les ajouts ont concerné :

20. Pour la partie financée par la BIRD :a. le renforcement des réseaux d’assainissement dans la ville de Tozeur (Assainissement des cités de Ezdihar et El Mattar de population à faible revenu).b. la réalisation de deux projets pilotes d’assainissement rural (Chorfech et Chouigui).c. la réalisation d’un cycle de formation au profit de 40  cadres de l’ONAS.d. l’acquisition de matériel d’exploitation et d’entretien (en deux lots).e. l’étude et la réalisation de deux stations d’épuration type « compact » dans les villes de Boumerdès et Fernana.

21. Pour la partie financée par la NIB/NDF :a. la réalisation d’une station d’épuration type « compact » dans la ville de Mornag.b. la réalisation d’un système de transfert des eaux usées de la ville de Ghomrassen vers la STEP de Tataouine.

22. Les transferts ont concerné deux composantes du prêt 3782 vers le prêt BIRD N°4175-TUN. Il s’agit de l’élaboration du Schéma Directeur de l’Informatique, des travaux d’assainissement du centre rural de Chouigui et de la réalisation d’un système de transfert des eaux usées de la ville de  Boumerdès.

4.  Résultats engendrés par le  prêt

23. Le prêt a permis d’agir positivement sur plusieurs volets conformément aux objectifs assignés aux contrats programmes et au prêt :

4.1 Mesures de renforcement institutionnel

24. Sur le plan des mesures de renforcement institutionnel, il a été procédé à :

a. La réalisation d’une campagne de sensibilisation des abonnés à de bonnes méthodes d’utilisation et de conservation de l’eau. Cette campagne comprend des spots télévisées, des brochures, des visites de sensibilisation sur le terrain et la production d’un guide de l’usager.b. l’acquisition de matériel nécessaire à la détection des fuites pour la SONEDE,c. l’acquisition de matériel de comptage, de micro portable, d’auto relève pour la SONEDE. d. l’élaboration en 1999 d’une étude sectorielle sur la stratégie en assainissement rural. Cette étude a permis d’établir un programme défini sur plusieurs étapes avec en particulier un projet pilote touchant 30 localités à réaliser au cours du X

ème plan.

- 46 -

25. Dans le cadre de la stratégie nationale,la SONEDE a entrepris des actions d’économie d’eau qui ont permis de porter le taux de rendement du réseau d’eau potable de 73 % en 1994 à 80 % en 2002. 4.2 Résultats engendrés pour les zones rurales : " eau potable "

26. Eau potable rurale :a. le prêt a permis de mieux rationaliser la desserte de l’eau, de minimiser les taux de fuite d’eau dans les réseaux, d’acquérir des moyens de comptage fiables et de desservir une population totale additionnelle de 220 000 habitants à faible revenu.b. il a permis le raccordement, jusqu’à fin  2002, de 35925 nouveaux abonnés et 47 abonnés GIC.c. depuis la mise en vigueur du prêt, l’évolution annuelle du nombre additionnel des nouveaux abonnés en milieu rural est satisfaisante avec une moyenne de 4483 abonnés par an (voir annexe 5).d. le prêt a contribué à l’augmentation significative du taux de desserte en milieu rural atteignant 83 % entre le Génie Rural et la SONEDE contre un taux de 60,6% avant la mise en vigueur du prêt en 1994.e. Afin de s’assurer de l’éligibilité effective des projets, la SONEDE a procédé à l’actualisation des fiches d’éligibilité de 30 projets pris comme échantillon par l’introduction des données réelles ayant servi à l’élaboration des dites fiches, à savoir la population réelle, le coût du projet et la consommation moyenne enregistrée. Les résultats ont montré que le gain social par m

3 d’eau consommé demeure toujours positif

(annexes 6).f. le linéaire total des conduites commandées et posées pour le réseau eau potable rurale de différents matériaux et tous diamètres confondus est de 1685 km (795 km en polyéthylène, 871 km en amiante ciment et 21 km en fonte) contre 1585 km initialement prévus ; soit un dépassement de 6% (annexes 7.1 et 7.2).g. l’investissement pour la composante eau potable rurale par localité est de 72 000 $US contre 100 000 $US prévu initialement. Cette baisse s’explique en partie par l’appréciation du dollar et de l’euro, à la chute du prix du pétrole qui a une incidence directe sur le prix de la fourniture des conduites en polyéthylène représentant 50% environ du linéaire posé.

4.3 Résultats engendrés pour l’assainissement

27. Assainissement :

a. la pose de 86 km de réseaux d’assainissement.b. la réalisation de 3 grandes stations d’épuration de capacité totale 18750 m

3/j dans

les villes deTozeur, Tataouine et Jemmal Zeramdine.c. la réalisation de 3 stations d’épuration type compact dans les localités de Boumerdès, Fernana et Mornag de capacité totale 7450 m

3/j.

d. la réalisation de 4750 nouveaux raccordements aux réseaux d’assainissement

28. La  réalisation des stations d’épuration a été précédé par des études d’impact sur l’environnement approuvées par l’ANPE.

4.4 Résultats engendrés pour les  zones urbaines

29. le prêt a  permis l’introduction dans le réseau :a. de conduites en fonte ductile améliorant ainsi la durabilité du réseau, et l’amélioration du rendement par la diminution des pertes et casses.

- 47 -

b. de conduites en fibres de verre nouvellement testées à la SONEDE dans les ouvrages hydrauliques, elles permettront d’éviter les problèmes de corrosion constatée avec les conduites en acier généralement utilisées.

30. Outre l’introduction des conduites en fonte et en fibre de verre citées précédemment, le prêt a permis le remplacement d’une grande partie de compteurs de faible précision par d’autres de classe supérieure et de meilleure précision. Dans le même but, et afin de mieux gérer son réseau, la SONEDE a introduit pour le réseau du Grand Tunis  de nouvelles techniques telles que la télémesure et la télégestion sur plusieurs sites de réservoirs. Ceci  a permis d’avoir des informations continues, fiables et instantanées avec un meilleur archivage des données pour une exploitation ultérieure. Le système de régulation au sein du pôle principal de production de Ghdir El Goula qui assure plus de 80% des besoins en eau du Grand Tunis dont le réseau a  été intégralement modernisé.

31. La  mise en service de ces composantes avec l’introduction de conduites de gros diamètres en fonte ductile, et l’augmentation des réserves de 25 000 m

3 ont permis de maintenir une

continuité dans le service malgré les fluctuations importantes dans la consommation entre les saisons et les journées de pointe, ce qui a permis d’étendre le réseau, d’augmenter le nombre d’abonnés et par conséquent d’augmenter la consommation d’eau.

32. le prêt a contribué à l’augmentation du nombre d’abonnés qui est passé de 343.286 en 1994 à 482.995 abonnés en 2002 soit un nombre additionnel de 139.709 abonnés et une évolution globale  de 40,7 %  engendrant un nombre additionnel d’abonnés annuel moyen de 17.463 ; soit  une évolution de 5 % ( annexe 8). Cette augmentation importante du nombre d’abonnés s’explique par le fait que les infrastructures réalisées dans le cadre du prêt ont permis de subvenir aux demandes sans cesse croissantes des différents organismes chargés du secteur de l’habitat tels que l’AFH, AFI, SNIT …

33. Sur un total de 96,4 km de conduites achetées tous diamètres confondus dans le cadre du projet,  79,4 km ont été posées au 30 juin 2003 ; soit un pourcentage de 82%.  Le reste est en cours de réalisation. Le montant y afférent sera supporté par la SONEDE. (annexe 9).

34. le volume consommé est passé de  89 millions de m3 en 1994 à 103 millions de m

3 en 2002

; soit une évolution annuelle de 2%. Cette consommation projetée en 2030 serait à 206 millions de m3 avec une croissance  annuelle de 2,5% (  annexe 10).

5. Décaissements

35. A  la date de clôture du prêt (30 juin 2003), les retraits sur le prêt 3782 ont atteint :

a. pour la composante assainissement les décaissements  ont été évalués à 1,25  M$US pour la partie du prêt libellée en Dollar; soit 100% et 6,18 M€ pour la tranche libellée en Euros; soit 89% du montant total de cette tranche (6,94 M€). Au 31 octobre 2003 le reste du prêt sera totalement décaissé. (annexe 11).

b. pour la composante eau potable rurale, les retraits ont atteint 18,301 millions de dollars soit un pourcentage de décaissement de 92%. Au  31 Octobre 2003, la part réservée à la composante rurale fixée à 20 M$ millions de dollars sera totalement consommée. (annexe 12).

36. Au 30 juin 2003  les  décaissements du prêt 3783 pour le Grand Tunis ont atteint 24,331 M$ millions de dollars soit  84% du montant total . Au 31 octobre 2003 ils  passeraient à 26,515 M$ millions de dollars soit 92% du montant total du prêt fixé à 29 millions de dollars ( annexe 12 ).

- 48 -

6. Mise en œuvre du projet

L’exécution du projet a fait ressortir les enseignements suivants

37. l’ouverture d’un compte spécial unique pour  la SONEDE et l’ONAS n’a pas facilité la gestion de ce compte. A l’avenir il est recommandé de créer des comptes spéciaux séparés pour chaque agence d’exécution en cas d’emprunt conjoint.

38. les missions de la Banque ont permis d’assurer un suivi précis des investissements, des décaissements et de l’avancement physique du projet. Toutefois compte tenu de la fréquence de ces missions qui nécessitent la mobilisation de tous les intervenants sur le prêt avant l’arrivée de la mission et en cours de la mission, Il serait  judicieux de créer une unité de gestion du projet « UGP » au sein de chaque agence d’exécution.

39. l’appréciation du dollar et de l’Euro au cours de la période du prêt a engendré de nouvelles ressources imprévues en dinars, ce qui amené les agences d’exécution à engager de nouvelles composantes afin de résorber le prêt.

40. Au moment de l’évaluation du prêt, la non disponibilité des études détaillées est parfois la cause de retards dans la réalisation des composantes pendant la période des prêts.

41. les procédures d’obtention des différentes autorisations auprès des autorités compétentes pour la pose des conduites ont été à l’origine des retards pour la réalisation de certaines composantes.

42. les recommandations récurrentes de la Banque dans ses « Aides Mémoires » ont amené le Gouvernement tunisien à :

a.  Apurer les arriérés de l’Administration vis-à-vis de la SONEDE et l’ONAS.b. Procéder régulièrement à des ajustements tarifaires de l’eau et des redevances d’assainissement.

Difficultés rencontrées au niveau des procédures

43. les procédures de passation des marchés tant au niveau de la banque qu’au niveau local sont lourdes et lentes, causant ainsi des retards dans l’exécution de certaines composantes. De même la multiplicité des procédures des différents bailleurs de fonds qui financent les projets de l’ONAS et de la SONEDE complique la gestion des marchés. Il serait donc souhaitable d’assurer une formation pour tous les intervenants et ce pour mieux connaître les différences entre les procédures tunisiennes et celles  des bailleurs de fonds.

44. le cumul des procédures tunisiennes avec celles de la Banque, quand elles ne sont pas contradictoires, rallongent les délais de réalisation . Il serait souhaitable dans le futur de réviser à la hausse les seuils des montants des appels d’offres pour la « Non Objection à priori » de la Banque.

45. afin de faciliter la gestion du prêt et de diminuer les délais de procédures, il serait aussi souhaitable de créer un bureau local de la Banque qui disposera des prérogatives nécessaires.

7. Performances des agences d’exécution

7.1 Performances de la SONEDE

46. Grâce à la performance du personnel de la SONEDE tant au niveau des études, du suivi   des travaux qu’au niveau de l’exploitation, les composantes ont été réalisées d’une façon

- 49 -

satisfaisante, ce qui s ‘est traduit par des taux de réalisation de 90 % pour le Grand Tunis et 100 % pour le milieu rural.

47. l’étude du plan directeur du Grand Tunis, financée dans le cadre du prêt 3783, a accusé un retard important qui s’est répercuté en partie sur les délais d’exécution de certaines composantes du projet. Malgré ce retard, la SONEDE a pu exploiter les différentes ébauches  de l’étude jugées pertinentes pour mieux atteindre les objectifs qui lui ont été assignés.

48. l’exécution du projet a contribué positivement à l’amélioration des indicateurs de performance tant financiers que techniques (annexe 14). 7.2 Performances de l’ONAS

49. l’ONAS a porté toute son assistance technique et ses moyens matériels et humains durant toute la période d’exécution du projet. De plus, une unité de coordination et de gestion du projet a été crée au sein de l’ONAS. Cette unité qui est composée d’un coordinateur et d’un chef de projet a contribué au bon suivi du projet.

50. le projet a contribué en partie à l’amélioration de certains indicateurs de performance. Ainsi, il a permis le raccordement de 6.200 nouveaux abonnés aux réseaux d’assainissement.

8. Performances des autorités Tunisiennes

51. Le Gouvernement a soutenu toutes les propositions de la SONEDE et de l’ONAS concernant le déroulement du projet et a approuvé toutes les recommandations de la Banque durant l’évaluation du projet pendant son exécution et ce compte tenu de la conjoncture du pays .

9. Performances de la BIRD

52. La Banque a effectué régulièrement des missions de supervision qui ont porté sur l’ensemble des projets qu’elle a financées.

53. Elle a également  soutenu certaines propositions de la SONEDE et de l’ONAS auprès du Gouvernement pour respecter en particulier les clauses financières de l’accord du prêt.

54. Elle  s’est montrée compréhensive à l’adaptation de la consistance du projet aux réalités constatées en cours d’exécution dans le but de mieux suivre l’évolution de l’urbanisation.

55. Les recommandations de la Banque ont aidé les deux agences d’exécution à instaurer un système de comptabilité analytique, une action d’apurement des arriérés ainsi qu’un ajustement tarifaire afin d’améliorer leur situation financière.

10 . Performances des bureaux d’études et des entreprises

56. Certaines entreprises ont enregistré des retards voir même des résiliations de marché par manque d’organisation et pour des difficultés financières .

57. Le groupement de bureaux d’étude pose certains problèmes de gestion. Une meilleur définition des tâches de chaque membre du groupement devra être précisée au niveau du contrat. Une formation pour les ingénieurs d’étude pour le suivi des études réalisées par les consultants est souhaitable.

11 . Conclusion et recommandations

- 50 -

58. Les différents objectifs visés par le projet ont été réalisés voir même dépassés. Les  réseaux et ouvrages ont amélioré l’état sanitaire dans les villes bénéficiaires du projet et ont eu un impact direct sur la protection de l’environnement et l’amélioration des conditions de vie des habitants.

59. De même, la SONEDE et l’ONAS ont amélioré leur capacité technique, institutionnelle et financière.

60. Le délai réaliste pour l’exécution d’un projet  d’assainissement ne peut pas être inférieur à 8 ou 9 ans. Ce délai doit servir de base aux prochaines évaluations de projets d’importance similaire.

61. Le fait d’avoir un seul prêt géré par deux agences d’exécution (SONEDE et ONAS) avec un seul compte spécial , a montré ses limites. Cette situation, combinée avec la conversion du prêt du Dollar et en Euros au cours de l’exécution,  a crée des problèmes au niveau du suivi des décaissements du prêt. Ce constat doit servir pour les prochains projets.

62. Ce projet a servi pour la préparation du projet d’assainissement du Grand Tunis (pour les études d’exécution). Ces deux projets étant financés par le même bailleur de fonds. Ainsi, les études étaient prêtes avant le démarrage du nouveau prêt. Ce fait pourra servir d’exemple pour les autres projets.

63. La partie financée par la NIB/NDF aurait due être séparée en un projet indépendant. Cette partie n’a bénéficié  ni d’une supervision particulière de la part des nordiques (NIB et NDF) ni de la part de la Banque mondiale.

64. Le présent projet a permis d’introduire de nouvelles technologies d’épuration à savoir les stations d’épuration type compact , de tester de nouvelles pratiques de réalisation des stations d’épuration en l’occurrence le clé en main et de fournir un outil de base pour le secteur de l’assainissement rural à savoir l’étude sectorielle de la stratégie en assainissement rural.

65. Il est recommandé d’inclure dans le prochain prêt une composante formation des cadres techniques, administratifs et financiers dans les technologies de pointe et le suivi des études sous traitées.

66. Afin d’éviter les différences d’interprétation entre les agences d’exécution et les bureaux d’étude il est recommandé de recourir à une assistance externe pour l’élaboration des termes de références.

67. Il serait judicieux pour les prochains projets d’envisager la création d’une unité de gestion et d’un bureau local de la banque.

- 51 -

ANNEXESAnnexe 1: Détails du Prêt 3782 "Eau Potable et Assainissement" - ONASAnnexe 2: Détails du Prêt 3782 Assainissement - ONASAnnexe 3: Liste des projets ONAS financés par la BIRDAnnexe 4: Liste des composantes révisées (3783 Grand Tunis)Annexe 5: Prêt BIRD 3782-TUN: Nombre d'abonnésAnnexe 6: Prêt BIRD 3782-TUN: Evaluation économique de 30 projetsAnnexe 7: 1-Linéaire des conduites initialement prévu. 2-Linéaire des conduites posées.Annexe 8: Evolution du nombre d'abonnés dans le Grand TunisAnnexe 9: Linéaire de conduites posées dans le Grans TunisAnnexe 10: Volume consommé dans le Grand TunisAnnexe 11: Tableau des décaissements du Projet 3782 ONASAnnexe 12: Prêt BIRD 3782 SONEDE: Situation et prévisions de décaissementsAnnexe 13: Prêt BIRD 3783 SONEDE: Allocation initiale et actualiséeAnnexe 14: Indicateurs de performance de la SONEDEAnnexe 15: Tableau des réalisations par composante 3783Annexe 16: Photographies d'ouvrages ONASAnnexe 17: Evolution des indicateurs de performance de l'ONAS entre 1989 et 2002Annexe 18: Coût initial du projet 3782 ONASAnnexe 19: Coût actualisé du projet 3782 ONASAnnexe 20: Planning de réalisation du projet 3782 ONASAnnexe 21: Note sur les arrangements pour la poursuite des opérations et l'exploiatation des investissements réalisés dans le cadre du Projet 3782 ONASAnnexe 22: Prêt 3782-TUN: Retraits sur le compte du prêtAnnexe 23: Prêt 3783-TUN: Retraits sur le compte spécial 4024064840 et suivi des demandes de reconstitutionAnnexe 24: Prêt 3782-TUN: Situation des retraits et balance du Prêt en dollars US (Eau potable)Annexe 25: Prêt 3783-TUN: Situation récapitulative des demandes de décaissementAnnexe 26: Prêt 3783-TUN: Situation des retraits et balance du Prêt en dollars USAnnexe 27: Prêt 3783-TUN: Situation des retraits par nature de demandesAnnexe 28: Suivi des aspects environnementaux (ONAS): Suivi de l'exploitation des stations de Jemmel-Zeremdine, Tozeur, Tataouine, Fernana et Boumerdes

Annexes 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 22, 23, 24, 25, 26 and 27 are in project files with original Borrower's contribution.

- 52 -

Annexe 6Composante alimentation en eau potable des zones rurales

Evaluation économique de 30 projets

FICHES INITIALES NOUVELLES FICHES ACTUALISEESProjets /Gpts Populatio

nCoût du projet

(103 DT)

Cons. (m3/foyer

/an)

Invest. / foyer (DT)

Gain social par m3

(DT)

Population

Coût du projet

(103 DT)

Cons. (m3/foyer

/an)

Invest. / foyer (DT)

Gain social par m3

(DT)EL BASSATINE 5115 500.000 73 488.759 1.759 14666 289.000 86 118.249 1.438C. KABOUTI 3106 451.000 73 1919.149 3.820 2236 206.000 110 460.850 3.412BEJOU 2501 393.000 73 811.983 3.345 3078 325.000 76 546.218 3.304TRIFAYA 210 38.000 73 1085.714 4.270 288 89.000 63 1854.167 4.456KERMIA (KELBIA) 700 140.000 73 1102.362 4.805 586 123.000 166 1149.533 3.992KERKOUAN 1200 237.000 73 1247.368 3.170 1506 116.000 74 423.358 3.156AIN SALLEM 1723 495.000 73 1666.667 2.797 1854 362.000 72 1113.846 2.808BALTA 3200 1459.000 73 2335.938 0.352 1921 1157.799 51 3015.102 0.050C. SOUK ESSEBT 7844 1015.000 73 1079.787 0.322 9778 447.812 107 274.731 0.280MAHJOUBA 640 300.000 73 2343.75 3.717 626 318.000 35 2544.000 4.309CITE ESSAADA 270 70.000 73 1396.296 3.063 360 253.285 114 3780.373 3.046OULED HABBASSI 126 30.000 67 1500.000 4.890 232 28.388 83 727.897 4.683C. EL GHOUALIF 4766 771.000 37 829.032 2.907 5380 614.181 46 570.800 3.471NEFFOUSSINE 162 134.000 67 4962.963 8.885 192 111.447 85 3482.719 8.535ACHBA CHIEB 1513 446.000 67 1689.394 4.688 2964 396.293 78 802.213 4.584C. ENNABCH 2347 601.000 67 1375.286 3.482 2628 476.656 59 997.188 3.737HDAIDIA 370 53.000 67 981.481 1.757 558 80.419 83 864.720 1.694C. MSAADIA NORD

540 78.400 67 871.111 3.527 558 46.439 63 499.344 3.733

C. BEDARNA 1380 442.000 120 1921.739 7.214 2430 223.757 105 552.486 6.408C. O.FERJANI 605 29.198 120 224.600 5.254 966 103.951 103 590.631 4.721C. OGLA 900 370.000 160 1466.667 2.175 1782 276.233 199 930.077 1.762DGHOUMES 1381 145.000 120 630.435 5.226 1455 161.425 119 664.300 4.500EL ASSAIDA 1095 216.000 120 981.818 3.007 1374 157.950 125 574.364 2.632OULED HAMDI 300 140.000 120 2333.333 8.123 78 52.837 94 3032.313 7.619ETTATAOUER 1760 110.000 120 314.286 3.675 2034 154.640 92 379.951 3.428CITE BOUZNIF 385 121.000 120 1728.571 4.095 558 115.463 39 1143.198 5.203GMANA 2 300 128.000 120 2327.273 3.572 246 110.339 46 2451.978 4.001KSAR O. SOLTAN 1987 97.000 120 244.332 4.187 2242 74.072 68 165.339 4.292HACHEN DOUIRET

240 35.000 120 729.167 4.589 307 40.497 92 525.935 4.280

OUELGET GMANDA

385 97.000 120 1385.714 4.166 450 96.713 90 1179.427 3.918

- 53 -

Annexe 21

Ministère de l’Agriculture, de l’Environnement et des Ressources HydrauliquesOffice National d'Assainissement (ONAS)

Septembre 2003 PROJET « EAU POTABLE & ASSAINISSEMENT »

(Prêt BIRD N°3782-TUN)RAPPORT D’ACHEVEMENT

Note sur les arrangements pour la poursuite des operations et l’exploitation des investissements

réalisés dans le cadre du projet

La présente note présentera les différents arrangements engagés par l’ONAS en vu de poursuivre les opérations et l’exploitation des investissements réalisés dans le cadre du projet « Eau potable et assainissement ».

A signaler que la plupart des activités du projet ont été intégrées dans les opérations régulières de l’ONAS ce qui a minimisé le besoin d’arrangements particuliers pour la transition.

En total, le présent projet a permis la pose de 86 km de réseaux d’assainissement et la réalisation de 6 stations d’épuration dont 2 de type « compact ». Ces ouvrages sont répartis, de point de vu exploitation, comme suit :

- 2 STEP’s à savoir Fernana et Boumerdès sont en cours d’exploitation par les services de l’ONAS ;- 1 STEP à savoir Tataouine dont l’exploitation vient d’être délégué au privé (à partir       de 01/03/203)- 2 STEP’s à savoir Tozeur et Jammel-Zeramdine ou leur exploitations sont en cours d’être déléguées au privé ;- 1 STEP à savoir Mornag est en cours de construction et dont la mise en exploitation par les services de l’ONAS est prévue début 2004.

Les travaux de construction des stations de Tozeur, Fernana et Boumerdès ont été financés par la BIRD, alors que celle des stations de Tataouine, Jammel-Zeramdine et Mornag ont été financés par la NIB/NDF.

* La STEP de Tozeur, mise en exploitation depuis 13/07/2000, a été exploitée en premier lieu par les services de l’ONAS du département régional de Gafsa- Kebelli - Tozeur (Direction Exploitation). Quatre ouvriers et un chef de station ont été recrutés pour assurer l’exploitation de cette station.Dans le cadre du programme de participation du secteur privé dans le secteur de l’assainissement, cette station sera déléguée au privé. Ainsi, les offres pour la délégation de cette station (avec celle de Nafta et 95 km de réseau) sont en phase de dépouillement.Les principaux caractéristiques techniques de cette station sont les suivantes :

Capacité hydraulique :  6654 m3/jl

Capacité actuelle de traitement : 2573 m3/jl

Rendement épuratoire : 95%lCoût d’exploitation de la station : 164.460,452 DTlEquivalent habitant : 55484l

Le réseau d’assainissement réalisé dans le cadre du projet pour cette ville (38 km) est actuellement entretenu (curage) par les services du district de Tozeur du département régional de Gafsa- Kebelli - Tozeur (Direction Exploitation). Les abonnées paient une redevance d’assainissement variable suivant le volume d’eau consommée.

- 54 -

* La STEP de Tataouine, mise en exploitation depuis 11/07/1999, a été exploitée en premier lieu par les services de l’ONAS du département régional de Gabès-Medenine-Tataouine (Direction Exploitation). Neuf (09) ouvriers ont été recrutés pour assurer l’exploitation de cette station.Dans le cadre du programme de participation du secteur privé dans le secteur de l’assainissement, cette station a été déléguée au privé. Ainsi, l’exploitation de cette station est assurée actuellement par le privé. Le contrat de délégation est entré en vigueur le 1

er mars 2003.

Les principaux caractéristiques techniques de cette station sont les suivantes :Capacité hydraulique :  5430 m

3/jl

Capacité actuelle de traitement : 1413 m3/jl

Rendement épuratoire : 89%lCoût d’exploitation de la station en 2002 :103.170 DTlNombre d’abonnées : 38554l

Le réseau d’assainissement réalisé dans le cadre du projet pour cette ville (33 km) est actuellement entretenu (curage) par les services du district de Tataouine relevant du départment régional de Gabès-Medenine-Tataouine (Direction Exploitation). Les abonnées paient une redevance d’assainissement variable suivant le volume d’eau consommée.

* La STEP de Jammel-Zeramdine, mise en exploitation depuis janvier 2000, a été exploité en premier partie par les services de l’ONAS du département régional de Monastir- Mahdia (Direction Exploitation). Quatre ouvriers ont été recrutés pour assurer l’exploitation de cette station.Dans le cadre du programme de participation du secteur privé dans le secteur de l’assainissement, un dossier d’appel d’offres pour la délégation au privé de l’exploitation                  de 3 STEP’s y compris cette station est en cours de préparation. Les principaux caractéristiques techniques de cette station sont les suivantes :

Capacité hydraulique : 6700 m3/jl

Capacité actuelle de traitement : 3549 m3/jl

Rendement épuratoire : 96%lCoût d’exploitation de la station/an : 136.465 DTlNombre d’abonnés : 11420l

Le réseau d’assainissement réalisé dans le cadre du projet pour cette ville (16 km) sont actuellement entretenu (curage) par les services du centre de Jammel-Zeramdine du département régional de Monastir-Mahdia (Direction Exploitation). Les abonnées paient une redevance d’assainissement variable suivant le volume d’eau consommé. * La STEP de Boumerdès, mise en exploitation depuis mars 2003, est exploitée actuellement par les services de l’ONAS du département régional de Monastir- Mahdia (Direction Exploitation). 01 ouvrier a été recruté pour assurer l’exploitation de cette station. Les principaux caractéristiques techniques de cette station de type « compact » sont les suivantes :

Capacité hydraulique :  700 m3/jl

Charge polluante en DBO5 : 350 kg/jl

Capacité actuelle de traitement : 250 m3/jl

Rendement épuratoire : 96%lNombre d’abonnés : 3180l

Cette station avec celle de Fernana constitue une première expérience de l’ONAS en terme de procédé de traitement de type compact. * Pour la STEP de Fernana, mise en exploitation depuis mars 2003, est exploitée actuellement par les services de l’ONAS du département régional de Jendouba- El kef (Direction Exploitation). 02 ouvrier ont été recrutés pour assurer l’exploitation de cette station.

- 55 -

Les principaux caractéristiques techniques de cette station de type « compact » sont les suivantes :Capacité hydraulique : 270 m

3/jl

Charge polluante en DBO5 : 162 kg/jl

Capacité actuelle de traitement : 180 m3/jl

Rendement épuratoire : 95%lCoût d’exploitation de la STEP :l

Energie : 1000 DToPersonnel : 800 DToAnalyses : 300 DTo

Abonnés à la station : 90% (1880 habitants)l Remarque Pour la STEP de Mornag : les travaux de construction de cette station sont en cours (Génie civil : 100%  et équipements en cours de livraison). La mise en exploitation de cette station est prévue pour début 2004, alors que son exploitation sera assurée par le département régional de Ben Arous (Direction Exploitation). Les principaux données de dimensionnement de cette station sont les suivantes :

o Population : 34700 habitants ;o Equivalent habitant : 42500 Eq. Habitants ;o Débit moyen journalier : 3200 m

3/j ;

o Charge en DBO5 : 1700 kg DBO

5/j soit 500 mg/l ;

o Azote total : 288 kg/j soit 90 mg/l ;o Azote ammoniacal : 192 kg N-NH

4/j soit 60 mg/l ;

o Phosphore total : 80 kg/j soit 25 mg/l ;o Phosphate : 54 kg/j soit 17 mg/l.

Il ressort de ce qui précède que les ouvrages réalisés dans le cadre du projet ont été intégrés dans les services de l’ONAS compétents dans les régions correspondantes.L’importance du réseau réalisé dans le cadre du présent projet n’est pas de nature à perturber la marche des services concernés et ne nécessite pas de renforcement des moyens de ces services. Pour les stations d’épuration, certains ont été ou vont être confiées au secteur privé pour assurer leur exploitation. Pour le reste, l’ONAS a recruté le personnel nécessaire pour assurer un fonctionnement adéquat. Pour le suivi de fonctionnement de la station d’épuration, des protocoles d’analyses et de mesure (pH, débit, DBO

5, DCO, etc.) sont réalisés journalièrement et consignés dans un rapport d’exploitation. Pour

le réseau, l’activité d’exploitation est suivi par des rapports journaliers sur le mètre linéaire curé dans chaque centre d’exploitation de l’ONAS.

- 56 -

Société Nationale d'Exploitation et de Distribution des Eaux (SONEDE)Projet Eau Potable et Assainissement

Prêt BIRD 3782-TUN

Note sur les arrangements des opérations et l’exploitation des investissements

après la clôture du prêt

La SONEDE, ayant la responsabilité d’assurer la continuité de la distribution d’eau sur tout le territoire, a mis en œuvre une organisation souple et efficace en matière d’exploitation et de maintenance permettant une utilisation optimale des réseaux et des équipements.Pour assurer cette fonction, la SONEDE a mis en place des structures locales et régionales qui oeuvrent ensemble, et d’une façon cohérente au maintien en bon état des réseaux et des équipements.

ORGANISATION DE LA GESTION TECHNIQUE DES RESAUX

Concernant la gestion des réseaux et des ouvrages hydrauliques, la SONEDE ne distingue pas entre urbain et rural. Elle applique les mêmes règles et les mêmes procédures pour les deux milieux.

1. Sur le plans organisationnel, la SONEDE dispose de quatre directions régionales (Nord - Grand-Tunis – Centre et Sud) et de trente huit Districts qui couvrent l’ensemble du territoire, et qui assurent les deux fonctions commerciale et technique.Un District est doté de plusieurs unités de travaux qui opèrent dans des secteurs bien délimités et qui sont dimensionnés en moyens humains et matériels selon des critères précis (nombre d’abonnés, rayon d’action, importance de l’infrastructure hydraulique etc.).La SONEDE dispose de 180 unités de travaux pour 1.750.000 abonnés, soit environ 10.000 abonnés/unité.Ces unités sont composés généralement d’un Chef de travaux (Ingénieur Adjoint), d’un certain nombre de surveillants de réseaux et d’équipes d’interventions.

2. Gestion technique des réseauxLe District à travers les Services Techniques (Etudes et Exploitation) et les unités de travaux assure l’exploitation courante des réseaux, l’entretien curatif (la gestion des interventions) et l’entretien préventif par la réalisation des programmes annuels préalablement établis, discutés et validés.

Programme d’entretien préventifLe programme d’entretien préventif comprend : Les travaux de réhabilitation de conduites et de branchements.Annuellement, la SONEDE programme environ 150 Kms de conduites représentant 0,5 % de la longueur totale du réseau. L’objectif consiste à atteindre 1%.Elle remplace les branchements en plomb (300.000) par du polyéthylène  à concurrence de 25.000/an. L’objectif à remplacer les branchements restants à raison de 50 000/an.

Recherche de fuites - Sous-traitance :  1.200 km/an - En régie :  1.000 km/an      (Soit 6% de la longueur totale des réseaux)      Les zones qui font l’objet d’une recherche de fuites sont identifiées et dégagées       suite à un travail de sectorisation.Comptage et régulationTous  les  systèmes  hydrauliques  ( quelque  soit  leur  importance )  sont  régulés  et   comptés. La surveillance est assurée à distance pour la majorité des systèmes  par desinstallations de télémesure. L’objectif tracé consiste à généraliser la télémesure et de doter les pôles

- 57 -

importants par des installations de télégestion.

Amélioration des centres rurauxLe programme d’investissement prévoit annuellement un budget pour l’amélioration des centres ruraux. Ces améliorations peuvent toucher les ressources, les adductions, les réseaux et éventuellement les réseaux e distribution.

Réalisation des programmes - Les travaux d’entretien curatif sont réalisés à 40 % par sous-traitance, et le reste en régie. - Les travaux d’entretien préventif sont réalisés à 80 % par sous-traitance, et le reste en régie. - Le programme d’amélioration des centres ruraux est réalisé en totalité par sous-traitance.L’objectif de la SONEDE est de développer la sous-traitance en vue de couvrir toutes les activités de la Société même celles nécessitant une technicité élevée.

3. Maintenance préventive systématiqueDans le but de réduire les pannes et d’optimiser les coûts d’exploitation, la SONEDE à travers ses structures régionales de maintenance élabore annuellement un programme de maintenance préventive selon des critères prédéfinis.Ces programmes fixent clairement les travaux à faire, le matériel à acquérir, la date de la dernière intervention, l’opérateur, etc. Ils permettent de suivre les équipements et de réduire les pannes mécaniques.

CONCLUSION

Suivant la politique instaurée par la SONEDE en matière d’exploitation, de gestion et de maintenance des différents réseaux et équipements, après la clôture du prêt 3782, les systèmes ruraux seront gérés de la même manière que les systèmes urbains situés à proximité avec les moyens humains et matériels disponibles qui seront renforcés en cas de nécessité.

- 58 -

Annexe 28

Ministère de l’Agriculture, de l’Environnement et des Ressources HydrauliquesOffice National d'Assainissement (ONAS)

Septembre 2003 PROJET « EAU POTABLE & ASSAINISSEMENT »

(Prêt BIRD N°3782-TUN)RAPPORT D’ACHEVEMENT

Note Sur le Suivi des aspects environnmentaux des Stations d’epuration réalisees dans le cadre du projet

Suivi  de l’Exploitation des Stations d’épuration

 de  Jemmel-Zaremdine ,Tozeur ,Tataouine, Fernena   et Boumerdas

I / données générales : I-1 ) Année de mise en service des STEP :

STEP Jemmel-Zaremdine Tozeur Tataouine Fernana BoumerdesAnnée de mise en service 2000 2000 1999 2003 2003

I-2 ) Procédé de traitement :

STEP Jemmel-Zaremdine Tozeur Tataouine Fernana BoumerdesProcédé Aération

prolongéeAération

prolongéeAération

prolongéeCompacte  Aération

prolongée

Compacte  Aération

prolongée

I-3 ) Paramètres de dimensionnement :

STEP Unité Jemmel-Zaremdine Tozeur Tataouine Fernana BoumerdesDébit m

3/j 6700 6654 5430 270 700Charge

biologiqueKg DBO5/j 3127 2845 2225 162 350

I-4 )  Provenance des eaux usées :

STEP Jemmel-Zaremdine Tozeur Tataouine Fernana BoumerdesDomestique 91 % 85 % 90 % 100 % 100 %Touristique 0 % 3 % 1 % - -Industriel 9 % 12 % 9 % - -

II Suivi de la qualité des eaux usées : - Le suivi de la qualité des eaux brutes et épurées s’effectue dans  les laboratoires :

· Des Stations d’épuration·  Régionaux ·  Du Centre International de Technologie de l’Environnement.

- Les paramètres  et les fréquences d’analyses sont établis selon un programme bien déterminé . - Les résultats d’analyses sont réexaminés par les services centraux (DATC)  en proposant aux exploitants des solutions pour une meilleure gestion de ces stations . III)  Quantité et qualité actuelles des eaux(valeurs  moyennes jusqu'au Juillet 2003) :

- 59 -

Paramètres Unité JemmelZaremdine

Tozeur Tataouine

Débit m3

/j 3549 2573 1413 Entrée – sortie Entrée – sortie Entrée – sortie

Charge en  DBO5 Kg/j 1640 64 1536 75 406 45

RendementEpuratoire

% 96 95 89

Entrée – sortie Entrée – sortie Entrée – sortieMES mg/l 536 21 751 30 692 16

IV ) Gestion actuelle des boues - Les boues produites des Stations d’épuration posent actuellement des problèmes de gestion (difficultés d’évacuation ),pour cela , l’ONAS envisage de réaliser une étude de gestion qui permettre de présenter les solutions techniquement faisable et économiquement rentable  (valorisation , la mise en décharge , l’enfouissement in situ et toutes autres formes  d’escamotage . - Une fois les boues sont séchées et vue la restriction de leur vente et l’absence des normes pour leur valorisation,  les  boues sont  stockées au niveau des stations d’épuration à part les boues de la station Tataouine  qui est envoyée à la décharge publique. - Les volumes des boues produites par les trois stations sont indiqués dans le tableau suivant :

STEP Jemmel-Zaremdine Tozeur TataouineVolume de  boue humide en m

3

/an         (2002 ) 8395 9284 2849Volume de boue humide m3/ mois (jusqu’au  juillet

2003 )777 895 1573

V ) Gestion des nuisances : odeurs et insectes

TRAITEMENTSTEP Odeurs Insectes

Jemmel – Zaremdine Chaux Insecticides *Tozeur Sulfate de fer,

Chaux, désodorisant RC 600Insecticides *

Tataouine

Chaux Insecticides *

* Les insecticides sont utilisés selon les recommandations du ministère de la santé publique. VI ) Destination des eaux épurées : - Le rejet  actuel des eaux épurées de ces trois stations d’épuration est  indiqué dans le tableau suivant :

STEP Jemmel-Zaremdine Tozeur TataouineMilieu de rejet Oued EL MALEH Chott EL JERID Oued Tataouine

- Le programme de réutilisation des eaux épurées durant le X

ème plan est mentionné ci dessous :

Réutilisation STEP Jemmel-Zaremdine Tozeur Tataouine

Type de périmètre Périmètre irrigué Terrain de Golf

Parcelle expérimentale Situation projetée ( X

ème

Plan ) Superficie     (ha) 180 150 4

VII) Renforcement des capacités du personnel exploitant des STEP Durant l’année 2003 l’ONAS a organisé  en collaboration avec des centres spécialisés (Exemple :

- 60 -

CITET) des activités de formations multidisciplinaires pour le personnel exploitant (cadres et agents) des STEP’s avec plusieurs thèmes : Epuration et gestion des stations d’épuration, Exploitation des biofiltres et traitement des odeurs dans les STEPs, la sécurité dans les laboratoires et les STEPs, désinfection des eaux épurées, Traitement des effluents liquides et industriels, les nouvelles techniques d’épuration, conception et dimensionnement des STEPS  des petites collectivités rurales , conception et dimensionnement des STEPS urbaines à boue activée, Techniques d’analyses des métaux lourds par absorption atomique (analyses chromatographiques, observations microscopiques)  et prélèvement et échantillonnage dans le cadre du contrôle de la qualité des eaux. 

- 61 -

- 62 -