Immoved Water I House Connection Sanitation I Urban I 54% I 24 % I 51 % I Rural 54 % 3% 16 % Total 54 % 10 % 28 %

Document of The World Bank

FOR OFFICIAL USE ONLY Public Disclosure Authorized Report No: 38080

PROJECT APPRAISAL DOCUMENT

ON A Public Disclosure Authorized PROPOSED GRANT

IN THE AMOUNT OF SDR 3.4 MILLION (US$5 MILLION EQUIVALENT)

TO THE

REPUBLIC OF HAITI

FOR A

Public Disclosure Authorized RURAL WATER AND SANITATION PROJECT

December 2 1,2006

This document has a restricted distribution and may be used by recipients only in the

Public Disclosure Authorized performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. CURRENCY EQUIVALENTS (Exchange Rate Effective September 30,2006)

Currency Unit = Gourdes (HTG) HTG38.1700 = US$1 $1.48020 = SDR 1

FISCAL YEAR

January 1 - December 31

ABBREVIATIONS AND ACRONYMS

CAEP ComitC d’ Approvisionnement en Eau Potable (Drinking Water Supply Committee) CAEPA ComitC d’ Approvisionnement en Eau Potable et Assainissement (Drinking Water Supply and Sanitation Committee) CAMEP Centrale Autonome MCtropolitaine d’Eau Potable (Autonomous Metropolitan Water Company) CDD Community-Driven Development CSA Country Social Analysis DC Direct Contracting DHS Demographic and Health Survey EA Environmental Assessment EGRO Economic Governance Reform Operation EGTAG Economic Governance and Technical Assistance Grant EMP Environmental Management Plan FAES Fonds d’Assistance Economique et Social (Haitian Social Fund) GOH Government of Haiti ICB International Competitive Bidding ICF Interim Cooperation Framework IDB Inter-American Development Bank ISN Interim Strategy Note JMP Joint Monitoring Programme for water supply and sanitation of UNICEF and UNDP LAC Latin America and Caribbean Region LICUS Low-Income Countries Under Stress MDG Millennium Development Goal FOR OFFICIAL USE ONLY

MTPTC Ministkre des Travaux Publics, Transports et Communications (Ministry of Public Works, Transport and Communications) NCB National Competitive Bidding NGO Non-Governmental Organization O&M Operation and Maintenance OBA Output-Based Aid PMU Project Management Unit PAHO Pan-American Health Organization PADF Pan-American Development Fund PFM Public Financial Management POCHEP Postes Communaux d’ Hygikne et d’Eau Potable (Communal Hygiene and Drinking Water Posts) PRODEP Projet de DCveloppement Participatif (Participatory Devlopment Project, a CDD project) PRSP Poverty Reduction Strategy Paper RWSS Rural Water Supply and Sanitation SBD Standard Bidding Document SNEP Service National d’Eau Potable (National Drinking Water Company) TSS Transitional Support Strategy URSEP Unit6 de RCforme du Secteur de 1’Eau Potable (Water Sector Reform Unit) WHO World Health Organization

Vice President: Pamela Cox Country Managermirector: Caroline D. Anstey Sector Director: Makhtar Diop Sector Manager: John Henry Stein Task Team Leader: Manuel Schiffler

This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not be otherwise disclosed without World Bank authorization. REPUBLIC OF HAITI Rural Water and Sanitation Project

CONTENTS

Page A . STRATEGIC CONTEXT AND RATIONALE...... 1 1. Country and sector issues ...... 1 2 . Rationale for Bank Involvement ...... 3 3 . Higher level objectives to which the project contributes ...... 5 B. PROJECT DESCRIPTION...... 6 1. Assistance Instrument ...... 6 2 . Project development objective and key indicators ...... 6 3 . Project components.,, ...... 7 4 . Lessons learned and reflected in the project design ...... 11 5 . Alternatives considered and reasons for rejection ...... 11 C. IMPLEMENTATION ...... 13 1. Partnership arrangements...... 13 2 . Institutional and implementation arrangements...... 13 3 . Monitoring and evaluation of outcomeshesults ...... 16 4 . Sustainability ...... 17 5 . Critical risks and possible controversial aspects ...... 18 6 . Grant conditions and covenants ...... 20 D. APPRAISAL SUMMARY ...... 20 1. Economic and financial analyses ...... 20 2 . Technical ...... 21 3 . Fiduciary ...... 23 4 . Social ...... 24 5 . Environment ...... 25 6 . Safeguard policies ...... 26 7 . Policy Exceptions and Readiness ...... 26 Annex 1: Country and Sector or Program Background ...... 27 Annex 2: Major Related Projects Financed by the Bank and/or other Agencies ...... 35 Annex 3: Results Framework and Monitoring...... 38 Annex 4: Detailed Project Description ...... 42 Annex 5: Project Costs ($ million) ...... 46 Annex 6: Implementation Arrangements ...... 47 Annex 7: Financial Management ...... 51 Annex 8: Procurement Arrangements...... 58 Annex 9: Economic and Financial Analysis ...... 63 Annex 10: Safeguard Policy Issues ...... 75 Annex 11: Social Analysis ...... 77 Annex 12: Project Preparation and Supervision ...... 82 Annex 13: Documents in the Project File ...... 83 Annex 14: Statement of Loans and Credits...... 84 Annex 15: Country at a Glance ...... 85 Annex 16: Map ...... 87

HAITI

RURAL WATER AND SANITATION PROJECT

PROJECT APPRAISAL DOCUMENT

LATIN AMERICA AND CARIBBEAN

LCSFW

Date: December 20,2006 Team Leader: Manuel Schiffler Country Director: Caroline D. Anstey Sectors: Water supply (90%);Sanitation (10%) Sector Managermirector: John Henry Stein Themes: Infrastructure services for private sector development (P) Project ID: PO89839 Environmental screening category: Partial Assessment Lending Instrument: Specific Investment Loan

[ ] Loan [ 3 Credit [XI Grant [ ] Guarantee [ ] Other:

For Loans/Credits/Others: Total Bank financing (US$m.): 5.00

INTERNATIONAL DEVELOPMENT 5.00 0.00 5 .OO ASSOCIATION Total: 5.25 0.00 5.25

Borrower: Ministry of Economy and Finance Haiti

Responsible Agency: Service National de 1'Eau Potable (SNEP) Port-au-Prince Haiti Tel: 509-2465300 or 2464131

Estimated disbursements (Bank J?YAJS$m) ;Y 7 8 9 10 11 12 bual 0.50 1.oo 2.00 0.50 0.50 0.50 hmulative 0.50 1.50 3.50 4.00 4.50 5.00 Project implementation period: Start March 15,2007 End: December 31, 201 1 Expected effectiveness date: 3/15/2007 Expected closing date: 12/3 1/2011 Does the project depart from the CAS in content or other significant respects? [ ]Yes [XINO Ref. PAD A.3 Does the project require any exceptions from Bank policies? Ref. PAD 0.7 [ ]Yes [XINO Have these been approved by Bank management? [ ]Yes [ IN0 Is approval for any policy exception sought from the Board? [ ]Yes [XINO Does the project include any critical risks rated “substantial” or “high”? [XIYes [ ]No Ref. PAD C.5 Does the project meet the Regional criteria for readiness for implementation? [XIYes [ ] No Ref. PAD D. 7 Project development objective Ref. PAD B.2, Technical Annex 3 Development Objective: Increase access of water supply and sanitation services in participating rural communities. Specific Objectives: 1. Increasing the sustained and effective use of safe drinking water in participating communities; 2. Improve use of sanitation and hygiene practices in participating communities; 3. Strengthen the capacity of SNEP, local water committees, and professional operators in cooperation with local government.

Project description Ref. PAD B.3.a, Technical Annex 4 Component 1: Capacity building for rural water supply and sanitation (RWSS), monitoring and evaluation and project management. Component 2: Water supply, basic sanitation and hygiene promotion. See PAD for further details.

Which safeguard policies are triggered, if any? Ref. PAD 0.6, Technical Annex 10 See ISDS. The safeguard policies on environmental assessment, cultural property and natural habitat are triggered.

Significant, non-standard conditions, if any, for: Ref. PAD C.7 Board presentation: No.

Loadcredit effectiveness: Subsidiary agreement satisfactory to IDA between MTPTC and SNEP signed. Approval by SNEP and MTPTC of an operational manual satisfactory to IDA.

Covenants applicable to project implementation: No. A. STRATEGIC CONTEXT AND RATIONALE

1. Country and sector issues

1.1. Country Background

Haiti is the poorest country in the Western Hemisphere. Average per capita income is comparable to the poorest countries of Sub-Saharan Africa, and Haiti ranks 154* (out of 177 countries) in terms of the Human Development Index. About 78 percent of its population of 8.6 million lives below the poverty line, with 54 percent in extreme poverty. The country’s health and social indicators are on the same level as those of the world’s poorest countries: Haiti’s child mortality rate of 118 deaths per 1,000 is far higher than the Latin American regional average of 33, and closer to sub-Saharan Africa’s average of 171.’

The rural areas exhibit the highest rate of poverty in the country. Of the roughly two-thirds of the population who live in rural areas, 86 percent live on less than US$2 a day and 69 percent live on less than $1 a days2Rural education and economic opportunities are extremely limited, and basic social services are severely lacking. Basic rural infrastructure (e.g., water and irrigation, feeder roads, electricity, and sanitation) is virtually absent or severely depleted, reinforcing isolation and exclusion. These conditions contribute to rural-urban migration, especially into the capital, Port-au-Prince.

Rural water coverage in Haiti is the lowest in the hemisphere. It was estimated at 54 percent in 2000, compared to an average of 73 percent for rural areas in LAC.3 The rural population relied on the following water sources: 4 percent house connection, 32 percent standpipes, 6 percent protected wells, 10 percent protected springs, 2 percent rainwater collection, totaling 54 percent having access to safe sources. 46 percent fetched water from unsafe sources, including 3 1 percent from unprotected springs, 6 percent from unprotected wells and 9 percent from rivers or canals. This means that an estimated 2 million Haitians in rural areas rely on unprotected water sources, jeopardizing their health, as evidenced by very poor health indicators. In addition, significant time is spent fetching water, which affects in particular women, children, and the poor.

Access to sanitation is also the poorest in the region. Haiti has the lowest degree of access to improved rural sanitation at 16 percent in 2000, compared to an average of 49 percent for rural areas in LAC. The Demographic and Health Survey (DHS) of 2000 shows a breakdown of only 12 percent having access to a private improved latrine and 4 percent using more basic traditional latrines of a type considered adequate. Another 4 percent use basic traditional latrines that are not considered adequate, 14 percent have access to shared improved latrines, and 10 percent can

WDI indicators, 2003 * World Bank (2006), Haiti Country Economic Memorandum. Data on water and sanitation coverage from different sources vary widely. The data in this report are based on the most recent survey on coverage undertaken in Haiti, which is the Ministry of Health’s Demographic and Health Survey (DHS) carried out in 2000. This is also quoted by the global Joint Monitoring Program (JMP) for Water Supply and Sanitation of WHOAJNICEF, from which regional comparator data are quoted. A new DHS was carried out in 2005 and preliminary results have been published, but the final results including water and sanitation data are expected to be published only in February 2007.

1 access shared traditional latrines, all of which offer only partial protection from the spread of fecal contamination. Fifty-six percent of the population resort to open defecation. Over a quarter of children under five years of age had experienced an episode of diarrhea in the two weeks preceding the survey. Diarrhea is particularly common between the ages of 6 to 24 months, exceeding 40 percent in the two weeks before the survey. These statistics highlight the importance of improving sanitation and hygiene in the domestic environment, where such children spend the most time and are most vulnerable to infection.

1.2 Sector Background

Institutional capacity for rural water supply is low, and nonexistent for rural sanitation. Rural water supply in Haiti is formally under the responsibility of the Service National de 1’Eau Potable (SNEP), the implementing agency for the proposed project. According to the 1977 law creating SNEP (Loi Organique du SNEP) it is an “autonomous state entity with civil and legal personality enjoying all the rights and prerogatives derived from this status”. SNEP is responsible for water supply outside the metropolitan area. SNEP has only limited capacity in rural areas and until recently focused its limited resources on urban water supply in secondary towns. The Ministry of Health also has a rural water unit Postes Communaux d’ HygiBne et d’Eau Potable (POCHEP), but this unit is small and inactive for lack of funds. There is no institutional home for sanitation.

Without an institutional presence in rural areas, rural water supply and sanitation investment has been unplanned and piecemeal. Most investments in rural areas are carried out by non-governmental organizations (NGOs) and the Social Fund (Fonds d ’Assistance Econornique et Sociale-FAES). Between 2000 and 2004, over 40 different NGOs and donors contributed an average annual total of approximately $3 million to the rural water sectors4The Social Fund invested $1.1 million in nine water and sanitation projects in 2004-2005. There is little knowledge about levels of coverage and needs, and thus it is difficult to prioritize investments, which are carried out in a piecemeal manner and do not necessarily reach the neediest.

Local communities do not maintain infrastructure well. Many rural water systems are managed by water committees (Comite‘ d ’Approvisionnement en Eau Potable-CAEP), consisting of unpaid volunteers elected by the community, as well as in some cases by a plumber hired by the committee. The performance of water committees varies widely, ranging from some that regularly collect funds, deposit them in a Bank account and perform routine maintenance to those that are practically defunct. Most water committees seem to not perform their functions adequately and collect insufficient funds for operation and routine maintenance. The Ministry of Public Works and SNEP now refer to these committees officially as water and sanitation committees (Comite‘ d ’Approvisionnement en Eau Potable et Assainissement-CAEPA) to reflect the broader role the committees are expected to play in the future.

The government of RenC PrCval elected in early 2006 intends to reform the water and sanitation sector through a framework law (hi-Cadre). Since 1996, a draft water and sanitation framework law has been discussed in Haiti, but has never been adopted. The new

PAHONHO Haiti Water and Sanitation Monitoring System.

2 government has taken the initiative to substantially rewrite the framework law and intends to submit it to Parliament. The framework law aims at the separation of policyhegulatory functions from operating functions, decentralization and the introduction of public-private partnerships. It aims at creating a water and sanitation directorate within the Ministry of Public Works (MTPTC) and regional water and sanitation companies. Furthermore, POCHEP will be dissolved and integrated into the new structures. Ultimately, the law envisages the gradual transfer of the responsibility for water and sanitation to those municipalities that have proven their capacity to undertake these responsibilities. Municipalities would in turn have the option to delegate service provision to the private sector, municipal water companies, or water committees. The law would also--for the first time in Haiti--define legal responsibility for sanitation within the public sector.

The substance of the law is still under discussion. The government does not expect the new structures to be put in place earlier than in three years. After the envisaged adoption of the Loi-Cadre, another Loi Organique for the regional water and sanitation companies would have to be passed and then their internal statutes would have to be elaborated and adopted before any new structures could be put in place.

2. Rationale for Bank Involvement

The project was identified and prepared according to the proposed Interim Strategy Note (ISN) for Haiti. The ISN aims at delivering hope to the population by helping the Government deliver quick wins in the provision of basic services and job creation; and restoring credibility in Haitian institutions by deepening reforms that promote long-term good governance and institutional development. The strategy, which is a continuation of the 2005 Transitional Support Strategy (TSS), will focus on selective interventions in areas where the Government has requested support from the Bank, and where the latter has a comparative advantage and complements other donor support. The ISN will be presented to the Bank’s Board of Directors concurrently with this project.

The project supports the ISNin terms of helping the Government deliver quick wins in the provision of basic services and restoring credibility in Haitian institutions. By using the public agency in charge of the sector to deliver services, unlike in the case of most other projects in the sector, the project promotes the restoration of credibility in Haitian public institutions. This is also in line with the recommendation of the Haiti Country Economic Memorandum to improve the capacity within state utilities.

The emphasis on strengthened public institutions is reflected in the proposed project’s focus on capacity building within the public institutions in the sector. The project would work with the staff of SNEP, which is the national entity responsible for rural water supply. SNEP has demonstrated capacity, and their performance under an earlier LICUS grant (see below) has been positive. It is expected that SNEP will remain in place for at least the next three years and that within this period the bulk of the project activities can be executed. Once the new structure of the sector foreseen in the Loi-Cadre will be in place, responsibility for executing the remainder of the project (mainly post-construction support) is expected to be largely transferred to the regional water and sanitation company in the project’s intervention area. Some functions are expected to remain at the national level with the water and sanitation directorate in MTPTC

3 or an associated structure for rural areas. Support to the national-level institutions will be provided primarily under the parallel IDB operation (see below).

A LICUS grant has provided the Bank with useful hands-on experience in working with SNEP as the implementing agency. The project involves two components: (i)preparation of a national strategy for rural water supply and sanitation, and (ii)rehabilitation of water systems in three communities in the Southern Department. Progress on both components is satisfactory and has led to an increase in SNEP’s capacity and motivation. A national workshop was held in November 2005 to discuss the national strategy. Under the second component, new management and technological approaches are being tested to increase system sustainability.

The national strategy for rural water supply and sanitation foresees a strengthening of MTPTC’s policy functions of the water and sanitation sector and SNEP’s capacity to intervene in rural areas. It also aims at the establishment of a closer partnership among SNEP, NGOs and local government. The strategy involves two parts: a largely completed operational strategy that will determine the content of SNEP’s cooperation agreements (protocoles d’accord) with NGOs in terms of technical specifications and community contributions; and a global strategy that will include an investment program to be funded by various donors and a better definition of the roles of various actors in the sector. The operational strategy is expected to be finalized first, at the request of SNEP, and the global strategy will be finalized at a later stage once the framework law has been passed.

The project will use a comprehensive methodology known under its French term as inginien’e sociale, for which there is no appropriate English tran~lation.~Zngknierie sociale is defined as a set of support activities that will:

Ensure that project activities respond to user demand, and that users are consulted in every important phase of the project; Create all necessary relationships among stakeholders for joint activities, and to formalize these relationships through contracts if necessary;

0 Ensure that conditions are created that will allow services to be provided on a sustainable basis; and * Strengthen the capacity of community-level stakeholders who will be involved in all phases of the project.6

The Rural Water and Sanitation Project thus aims to marry the benefits of a community-focused, demand-driven approach with the benefits of having a competent sector agency involved in

Ingknierie sociale is not to be confused with the English term “social engineering” which has a completely different meaning. Community development, community mobilization, community outreach and community participation all describe only specific facets of this comprehensive approach. The concept of ingknierie sociale has been developed in Haiti in the mid-1990s with the highly successful project to provide water to slum areas of Port- au-Prince using local water committees in partnership with the utility CAMEP, using a service provider (NGO) to facilitate the relationship. SNEP, StratCgie Eau Potable et Assainissement en Milieu Rural, Directive OpCrationnelles, June 2006, p. 18

4 terms of procurement, financial management and technical know-how in the design and supervision of good-quality works.

The project will build on the beneficial social dynamics created under the Community Driven Development (CDD) project through its strengthening of community-based organizations for those communities where the projects will overlap. The training and capacity building for community-based organization under the CDD project would support the implementation of RWSS. SNEP will have to coordinate with the PL-480 management office under the Ministry of Planning and External Cooperation, the executing agency of the CDD project, and its service provider in the Southern Department (PADF) in order to ensure that the two complementary approaches are clearly presented to communities,

While many aspects of the project are similar to the CDD approach, there is one key difference: Communities are not expected to manage funds or to undertake procurement. Most works contracts are expected to be well above the threshold of $20,000 currently used under the on-going Haiti CDD project ($38 million). Having communities manage such large sums has not been tested so far and would entail risks.

This project is closely coordinated with the IDB, which is the focal point for the sector according to the Interim Cooperative Framework (ICF.) The project is being executed in parallel with a $15 million IDB-supported rural water and sanitation pr~gram.~The IDB will intervene in four departments (Grande Anse, Nippes, 1’Ouest and 1’ Artibonite), while the proposed project will intervene in one (South) and in small parts of Nippes that are best reached from Les Cayes, the capital of the Southern Department. Both banks will use a similar selection mechanism for participating localities and the same operating manual. The Bank’s financial contribution, though modest, has enabled it to play a valuable role in the design and preparation of the program. Elements of the program that have resulted from Bank involvement include the use of SNEP as an implementing agency, professional operators and incorporation of a hygiene and sanitation component.

3. Higher level objectives to which the project contributes

The project is expected to improve governance and capacity within the public sector in Haiti. The project will, for the first time in Haiti, carry out a systematic assessment of needs and coverage levels for water supply and sanitation in one region, linked to a transparent system of selecting participating communities on the basis of objective criteria. These criteria will include estimates of needs, costs, and community commitment. Furthermore, the project’s focus on capacity building within the implementing agency concerning financial management and procurement contributes to the overall strengthening of public sector institutions and economic governance at a higher level.

This project will contribute to the higher level objective of improved health in Haiti. Intestinal infection (largely diarrhea) is cited by the Pan-American Health Organization (PAH0)8

~ ’ The IDB uses the term “program” for its overall intervention and the term “project” for each intervention in a !articular locality. Infant mortality in Haiti, www.paho.ordEnglishDPED/Infant mortalitv.mt, downloaded 4/27/2006.

5 as the leading cause of death in children under five, and the second leading cause of infant mortality (exceeded only by perinatal infections occurring in the first month of life). According to the most authoritative available estimates, 87 percent of the burden of diarrheal disease in LAC is attributable to unsafe water and sanitation.’ The evidence from global epidemiological experience is clear-cut, however, and indicates that water, sanitation, and hygiene interventions can reduce diarrheal disease by as much as fifty percent-far more than simple water supply interventions alone.

The project will contribute to social cohesion through helping communities make decisions collectively. Water is a good entry point for a community development project as it is used by all and serves a variety of needs. It provides a natural common issue around which the community can unite. The LICUS grant has given the Bank valuable experience in how to effectively mobilize communities around water management.

B. PROJECT DESCRIPTION

1. Assistance Instrument

The lending instrument is a Specific Investment Grant.

2. Project development objective and key indicators

Development Objective: Increase access to and use of water supply and sanitation services in participating rural communities.

Specific Objectives:

Increase the sustained and effective use of safe drinking water in participating communities; Improve use of effective sanitation and hygiene practices in participating communities; and Strengthen the capacity of the implementing agency, local water committees, and professional operators in cooperation with local government.

Given the low capacity in the Haitian public sector as well as previous implementation experience with Bank projects in Haiti, (i)project design has been kept very simple, (ii) substantial capacity building has been included, and (iii)objectives have been kept modest. At the same time, coordination with other stakeholders (government agencies, donors, and NGOs) is being actively sought, complementing the existing partnership with the IDB.

The Global Burden of Disease and Risk Factors, Disease Control Priorities Project, The World Bank, 2006.

6 Key indicators will be: number of beneficiaries gaining access to and using sustainable” water systems’ in participatin localities, and number of beneficiaries gaining access to and using basic sanitation systems r3 in participating localities.

3. Project components

The project has two components, the first one aimed at strengthening the capacity of SNEP and the second one aimed at directly providing services to communities. The project design has been kept very simple, while differing in important respects from how interventions in the sector have been carried out so far:

The project will build, for the first time in rural water supply and sanitation in Haiti, a close partnership between a public entity (SNEP) as the implementing agency and a service provider (an NGO or a firm) working under the direction and leadership of the public entity. This approach is being promoted both at the national and departmental level (see implementation arrangements for details). The project will, for the first time in rural water supply in Haiti, select participating communities based on objective and transparent criteria in a systematic, department-wide process. Building on the experience of the LICUS grant, the project will involve professional operators in the operation and maintenance of water systems and introduce volumetric tariffs, in order to promote greater sustainability. Concerning sanitation, the project will not build entire individual latrines, but it will test a more cost-effective approach by providing only slabs and working closely with communities to motivate them to build superstructures and to dig holes (sanitation promotion).

Component 1: Capacity Building and Project Management

The activities under this component aim at managing the project and strengthening the capacity of SNEP. The component includes four main activities.

First, it will strengthen SNEP’s regional office in Les Cayes through the recruitment of two consultants who are expected to become staff after project closure, as well as through the provision of cars, office space and coverage of operating expenses. Second, it will finance services for capacity building in project management, documentation, reporting, information technology, and environmental management. This loThis condition is expected to be fulfilled if beneficiaries are using systems and (i)tariffs are set at a level that guarantees operation and routine maintenance of the system; (ii)at least 80% of revenues due are collected; (iii)the water committees meet on a regular basis; and (iv) professional operators fulfill their contractual obligations. I’ “Water systems” in terms of this project means systems that are “improved” according to the Joint Monitoring Programme (JMP) definition of “improved” water supply. (e.g. Improvement in the access to piped water (for those households without access to piped water) or protection of a previously unprotected source.) “basic sanitation systems” in terms of this project means systems that satisfy the JMP definition of “basic” sanitation contributing towards the MDG Target. (e.g., both simple and Ventilated Improved Pit latrines, twin pit pour flush latrines, septic tanks etc) which protect the user and the community from fecal contamination,

7 activity is foreseen primarily during the first two years of project implementation, with the aim of making the assistance unnecessary during the remainder of the project. This activity will also include the financial audits of the project. Third, the Bank would finance services for surveys to monitor the achievement of performance indicators one year after the construction of water and sanitation systems. Fourth, it will assist SNEP in finalizing and disseminating the national rural water supply and sanitation strategy initiated under the LICUS grant mentioned above.

The component complements $3 million for capacity building to be provided by the IDB under its parallel financing. The IDB financing covers two areas: First, it aims at strengthening SNEP's functions in rural areas in close relation with the investment component, with an emphasis on community development and procurement. Second, it will strengthen SNEP' s overall capacity throughout its mandate, including rural and urban areas. It foresees the hiring of an advisor to the Director General for the first two years of the project and the strengthening of the technical and financial departments of SNEP. It also foresees an assessment of SNEP's human resources in view of a rationalization of its personnel and salary structure. In addition, the IDB will finance vehicles, equipment and recurrent costs necessary for the project.

Component 2: Water supply, basic sanitation and hygiene promotion

This component will enable participating communities to access and manage water supply and basic sanitation services. This includes the identification of participating communities based on objective criteria; ingbnierie s~ciale'~using a participatory approach allowing communities a series of choi~es;'~engineering design; system construction and supervision; and support during the early years of system management. The project will promote the innovative management approach piloted under the LICUS grant involving professional operators, increased cost recovery, metering, and the use of water kiosks instead of standpipes.

Investments under this component are expected to consist mainly of piped water systems. The piped systems will feed water kiosks and house connections. Water kiosks consist of a small concrete structures with faucets on the outside and taps on the inside, thus allowing kiosk operators to sell water by the bucket. Both water kiosks and house connections would be metered. Most piped systems are expected to be gravity-fed by springs. If pumping should be necessary and cost-effective, the project would give a strong preference to renewable energy. Given widespread theft of photovoltaic cells in Haiti, wind energy would be the preferred source of energy, relying on hurricane-proof new technologies as they are about to be tested in Haiti under a pilot project supported by the World Bank's Development Marketplace. Other water investments include hand pumps and rainwater harvesting systems.

Concerning sanitation, the project would support the construction of household latrines as well as blocs sanitaires next to schools and health centers. Sanitation investments will only be

l3See explanation further above on p. 4. l4 The proposed key choices are: choice between different water service levels (house connection or water kiosk) and sanitation service levels (pit latrine and Ventilated Improved Pit latrine) implying different costs to users; choice in the tariff structure (pure volumetric or combined flat rate and volumetric) and level (covering only operating costs and basic maintenance, or including a reserve for major maintenance).

8 supported in localities where water systems will be built, in order to be able to carry out hygiene promotion (see further below) in a credible manner and to maximize health impacts. A bloc sunitaire is a concrete structure with toilets and basic water supply fixtures in two compartments, one for men/boys and one for womedgirls.

Sanitation and hygiene promotion will be carried out in the same communities in which water services will be improved. This allows for the creation of synergies among water supply, sanitation and improved hygiene interventions. SNEP and Haitian NGOs have experience in the construction of blocs sanituires for health centers and schools, and the project will support the continuation of this work, with a focus on their operation and management as well as upon their physical construction.

The approach to sanitation will stress subsidies to promote the construction of individual household facilities. While other sanitation initiatives in Haiti have involved subsidies in the order of $100-$500 per family, this project will explore the viability of an approach in which only the basic slab and seat are subsidized, accompanied by promotion activities. Keeping the subsidy as low as possible while focusing on promotion has two advantages: (1) reducing the per-unit subsidy enables the project to provide access to sanitation services to a greater number of people; and (2) a low-cost subsidy of appropriate technology may sow the seeds for a small- scale industry that designs basic sanitation at a price families can afford. The sanitation component will thus focus heavily on demand stimulation at the household level.

While sanitation interventions will be coordinated with hygiene promotion activities, the differences between the two must be clearly understood. Hygiene promotion will thus be based on early studies to (a) determine high risk behaviors related to hygiene, sanitation and water supply; (b) identify specific motivators for change, including concepts of cleanliness, modernity, education and attractiveness; (c) identification of key channels of communication most effective in reaching mothers and children; and (d) development of appropriate messages to promote improved hygiene behavior. The hygiene promotion campaigns can draw substantially from recent experience in hand washing campaigns in LAC and elsewhere. SNEP will have the institutional responsibility for hygiene promotion under this project. However, since SNEP has no experience in hygiene promotion, and given the weakness of other potential stakeholders such as the Public Hygiene Directorate (Direction de Z’Hygidne Publique--DHP) in the Ministry of Health, the service provider (see under implementing arrangements) will also undertake hygiene promotion. Solid waste management is not considered a priority issue in rural areas and is thus not included under this project.

The financial policy under the program includes the following key elements:

9 Up-front financial contribution: Communities are required to come up with an up-front financial contribution before technical studies will be initiated. The purpose of the contribution is to be an indicator of willingness to pay. It should cover, at the minimum, three months of operation and routine maintenance costs of the system. These funds would not be mingled with project funds, but would rather be used by the community for small initial purchases (e.g. for chlorine, a motorcycle etc.) or be kept in a reserve for repairs. In addition, households that receive house connection would pay connection fees. Pass-through conditions for investment funds: Following common practice in Haiti and in many other countries, project funds for investments will be passed through to the community as a grant, given limited ability to pay.

Participating localities will be selected based on objective, transparent criteria. A tentative long list of about 40 localities has been established that would be reduced to a short list of about 12 localities based on an objective eligibility and priority criteria. Eligibility criteria include the number of inhabitants (less than 10,000); availability of a nearby, sufficiently productive, good quality source of water; protection of the source or indication of willingness and likelihood that the source will be protected; estimated per capita costs of less than $150; acceptance to pay volumetric tariffs covering operation and routine maintenance costs; acceptance of the professional operator model; a formal letter requesting to be included in the project; and payment of an up-front financial contribution into a fund managed by the water committee. If the number of eligible localities is greater than about 12, participating localities will be selected based on priority criteria. These include greatest needs (as measured by average walking distance to currently used sources of water) and lowest estimated unit costs (as measured by estimates of the required works and the number of houses in the locality assessed based on satellite pictures).

Total project costs are $5.25 million, including $0.92 million for the first component and $4.33 million for the second component. The costs will be financed through the grant ($5 million) and counterpart funds ($0.25 million).

Table 1: Estimates of Project Costs, by Components ($ million)

Components Grant Counterpart Total Cost 1. Capacity Building, 0.67 0.25 0.92 and Project Management 2. Water Supply, Basic 4.33 0.00 4.33 Sanitation and Hygiene Promotion TOTAL Project Costs 5 -00 0.25 5.25

The Government of Haiti has requested and the Bank has approved a Project Preparation Facility (PPF) advance of $0.7 million to prepare the project, of which $0.22 million will be used for component 1 and $0.48 million for Component 2. These costs are included in the total grant amount shown above.

10 4. Lessons learned and reflected in the project design

The following lessons have been incorporated in the project:

Design simple and well-defined operations with modest objectives (Operations Evaluation Department Country Evaluation). The reality and the complexity of the Haitian environment have demonstrated the importance of having a clear project design, to facilitate preparation and implementation. .) The proposed project keeps its objectives simple and well targeted. Build on the experience of a pilot project. .) The project builds on a LICUS grant pilot project currently underway in three communities in the Southern Department. Support transparent and timely implementation of the project with adequate procurement expertise. Previous projects have shown that the lack of clarity in procurement procedures can seriously delay project implementation and eventually undermine its success. .) The implementing agency has gained procurement experience during the LICUS grant execution. In addition, it has received training from IDB’s procurement staff based in Haiti. Further training is foreseen. No objection will be required for every single contract, which in itself entails a learning process that was begun under the LICUS grant. Strong involvement and ownership of the implementing agency in order to strengthen its capacity. Many projects in Haiti heavily relied and rely on foreign consultants and PMUs, which deteriorates capacity and motivation in the implementing agency itself. .) The project will not use a PMU structure. Instead, it will rely on staff of the implementing agency to make all key decisions, with intensive capacity building and supported by consultants on an as-need basis. The perspective is that any long-term consultants will be integrated into the implementing agency’s staff in the future. Need to further improve coordination among donors. The 2004 Interim Cooperative Framework (ICF) and the 2006 Extended ICF highlighted the importance of donor coordination. + The project has been prepared jointly with the IDB. A national workshop for the preparation of the national rural water supply and sanitation strategy has been held with the participation of donors, and at least one more workshop is planned under the project. The water and sanitation sector table created through the ICF is another forum for donor coordination.

5. Alternatives considered and reasons for rejection

Focus on urban water supply and sanitation: An urban water supply and sanitation project in the Port-au-Prince Metropolitan area was considered, but was rejected for the time being, as a rural focus would build on the progress made under the LICUS grant pilot project. However, in the future, IDA may consider scaling up the existing operation to include a pilot for urban areas.

Using a different implementing agency: POCHEP, a water and sanitation unit under the Ministry of Health, was considered as a possible implementing agency, but its capacity appears to be weaker than the capacity of SNEP. Furthermore, Haitian law attributes the responsibility for rural water supply and sanitation to SNEP under the Ministry of Public Works. SNEP also is successfully implementing the LICUS grant pilot project, and the IDB has decided to work with

11 SNEP on its rural water supply and sanitation project. A reform unit within the Ministry of Public Works, URSEP, has gained good procurement capacity through the implementation of the water reform project. However, it is not considered a permanent entity and working through it would not contribute to strengthening SNEP as the agency legally responsible for the sector.

Working with the Ministry of Health on hygiene promotion: The Ministry of Health would seem like a natural institutional home for the hygiene promotion activities under the project. Unfortunately the Public Health Directorate in the Ministry of Health is extremely weak and has no presence on the ground. It seemed overambitious to try to strengthen that directorate, which would also mean that the project would be executed by several implementing agencies. As a result, this option was rejected and the hygiene promotion activities will be carried out by specialized NGOs working for SNEP.

Greater focus on Output-Based Aid: A Bank-administered study financed by the Global Partnership on Output-Based Aid (GPOBA) assessed the potential of attracting private small- scale operators to invest in the water sector in Haiti, similar to a successful experience in Paraguay. These operators would be paid by the Bank once works are completed. The study’s results show that the Haitian private sector and NGOs face major obstacles to invest in the water sector, including lack of pre-financing and an unfavorable legal framework. In addition, public sector agencies have very limited capacity to administer an OBA project. Another reason for the rejection of the OBA option at this stage are the mixed signals that would be sent if the Bank supports one type of approach (OBA) under certain projects and different approaches (mainly CDD) under other projects.

A more substantial role for local government: Local government in Haiti is very weak, in particular in rural areas, where municipal administrations often do not exist. Previous water supply pilot projects in small towns have highlighted the occasional difficulty of working with local government in Haiti (see Annex 1 for details). The government deems that municipal administrations do not yet have sufficient capacity to plan, carry out, operate and maintain infrastructure. The revised draft water and sanitation framework law, therefore, foresees the creation of regional water and sanitation companies. The regional companies would have the ability to delegate services to municipalities if they have proven to have sufficient capacity. In rural areas, community-level water committees and professional operators would continue to operate infrastructure independently of local government. The project’s approach concerning local government is flexible: it will attempt to include them where advisable, keeping them at the very minimum informed of all project activities. Closer cooperation will be sought on a case-by- case basis.

Greater emphasis on sector reform: The policy dialogue on sector reform is led by the IDB, which is the focal point for the sector as part of the ICF. Given this agreement, the option of focusing on sector reform as part of the current project has been discarded. However, the Bank remains in close contact with the government and the IDB regarding sector reform initiatives, and the project has given the Bank an opportunity to stay engaged on the issue.

Establishing a project steering committee: The establishment of a steering committee comprising representatives of various Ministries has been considered. However, the idea was

12 rejected for various reasons. First, steering committees have proven to be largely ineffective, both in Haiti and in other countries, because they do not take actual decisions, there are few incentives for representatives to participate, and they often tend not to meet at all in reality. Second, the IDB has opted not to use a steering committee and the Bank wants to streamline procedures under the joint program. Nevertheless, the project would disseminate results and seek cooperation on a case-by-case basis in the absence of a steering committee.

Establishing a department-level consultative committee: In order to strengthen the perception of objectivity in the choice of participating localities, the Bank considered the creation of a department-level consultative committee. It would have consisted of elected officials and civil society representatives, and it would have vetted the selection process undertaken by SNEP. However, concerns were expressed concerning unmanageable delays due to potential disputes between members of the committee, ultimately jeopardizing the delivery of results on the ground.

Using Community Participation in Procurement and Management of Funds: A small share of the works considered under the project is small-scale in nature. This includes the construction of individual latrines, and of rooftop rainwater harvesting systems for residential homes, schools and health posts in higher-altitude areas where rainfall is plentiful. For such small-scale works, community participation in procurement and management of funds, as practiced under the Haiti CDD project, was considered. However, given the small total amount of these works (less than $250,000), the fiduciary risk and administrative complexity of using this arrangement with an implementing agency that has no prior experience with them, this alterative was rejected.

C. IMPLEMENTATION

1. Partnership arrangements

The proposed project constitutes part of a rural water and sanitation program for which the IDB contributes a $15 million loan. The IDB loan has been approved by its Board in September 2006 and is currently awaiting ratification by the Haitian Parliament, most likely in parallel with the World Bank project. The IDB’s rural project will be implemented by SNEP and is expected to use the same implementation mechanisms as the Bank project. The IDB has secured project preparation funds under arrangements that are coordinated with the Bank. Both projects will seek to implement the operational part of the national rural water and sanitation strategy. IDB is the focal point for the water and sanitation sector as part of the ICF. Its overall intervention in the sector is described in more detailed in the last paragraph of section A.2.

2. Institutional and implementation arrangements

The implementing agency for the project will be the Service National de 1’Eau Potable (SNEP). According to the 1977 law creating SNEP (Loi Organique du SNEP), it is an “autonomous state entity with civil and legal personality enjoying all the rights and prerogatives derived from this status”. According to the law, SNEP is under the supervision of a Board (Conseil d’Administration) presided by the Ministry of Public Works, and its day-to-day activities are managed by a Director General. In reality, the Board has not met in recent history,

13 thus leaving SNEP directly under the tutelage of the Ministry of Public Works. SNEP is responsible for water supply in all areas outside the metropolitan area of the capital.

The institutional arrangements of this project are largely based upon the implementation arrangements under the ongoing LICUS grant for rural water supply. The Ministry of Public Works (MTPTC), the government ministry responsible for water supply and sanitation in Haiti, would sign a subsidiary agreement with SNEP, as under the LICUS grant. While the implementation arrangements for the LICUS grant have proven effective, the proposed program supported by the Bank and IDB will require a much larger absorption capacity. In order to ensure the ability of SNEP to implement the activities under the program, the following measures have been initiated or are envisaged:

0 Recruitment of an experienced service provider. 0 Individual consultants hired by SNEP with the expectation that they would become SNEP staff when the project ends. 0 Capacity building within SNEP for project management and procurement.

The service provider (an NGO or a firm) will assist SNEP in key activities throughout the sub- project cycle, including the selection of participating communities, feasibility studies and technical designs, supervision of construction, sanitation promotion, hygiene promotion and post-construction support. In carrying out these activities the service provider will not substitute for SNEP, but rather complement SNEP’s activities. The service provider will have a specific mandate to strengthen SNEP’s capacity.

The institutional characteristics of the proposed program and the LICUS grant are compared in the table below:

Table 2: Comparison of proposed program and predecessor pilot project

Proposed Program Predecessor Pilot Project (WB and IDB) (LICUS grant) Ingknierie sociale Carried out by an experienced service Carried out in-house by SNEP with support provider, in partnership with SNEP from two international volunteers from 1 Engineers Without Borders-Canada (EWB) Procurement and I SNEP, supported by consultants and I SNEP with intensive Bank supervision and project management significant training with intensive I capacity building program and Bank I support by EWB I supervision; continued support by EWB 1 Technical design- I Carried out by the same service provider I Engineering firm I I (consortium) that will undertake-the I I ingknierie sociale Construction Contractors Contractors Construction Engineering firm or service provider Individual consultant Supervision Geographical About 50 localities in 5 departments 3 localities in 2 departments Cover age Duration 5 years without the preparation phase 2 years incl. preparation phase Amount $20 million $0.85 million

14 SNEP will have the fiduciary responsibility for the project (financial management and procurement). The project’s financial management will be carried out by SNEP using the same arrangements as under the LICUS grant. For the implementation of the grant, SNEP staff have already set up a separate accounting system, developed an operational manual and familiarized themselves with Bank’s financial management and disbursement procedures. Additional staff will be hired by SNEP to accommodate the higher workload due to the project. SNEP has selected one of its current employees to become a full-time procurement staff. He and other SNEP staff have been trained by the procurement specialist of the IDB office in Haiti and by the National Procurement Commission, which receives support from the Bank and IDB.

In order to assist in the process of ingknierie sociale, and to carry out engineering designs, SNEP is in the process of selecting a service provider (NGO, consulting form or a consortium of both) for the Southern Department on a competitive basis. The service provider will have as one of its tasks to strengthen the capacity in the local office of SNEP. It will be selected partly based on the service provider’s methodology and track record in working with and strengthening of public institutions.

Local water committees (CAEP) would have important responsibilities during the feasibility and construction stage. The CAEP is a community-based organization composed of five volunteers elected by the community. Given the weakness of most existing water committees, substantial assistance would be provided to existing and new water committees in order not to overload them during these crucial phases. While water committees have an important role, communicating with them is no substitute for communication with the entire community. SNEP and its service provider would therefore organize community consultations on key aspects of the project. These would include community contributions, design and location of infrastructure, tariff levels, as well as explaining the concept of and rationale for professional operators and water meters. During the construction stage, the CAEP will organize the community contribution, check and approve the work and assist SNEP and its service provider in resolving any differences between the community and the contractor.

In the operating phase, the role of the water committees will change compared to the role they have played in the sector so far. As mentioned earlier, most water committees do not have the capacity to adequately operate and maintain water systems. Under the proposed project, water committees would, therefore, delegate operation and maintenance to remunerated professional operators. Professional operators are expected to operate systems (chlorination, cleaning of reservoirs, operation of kiosks, meter reading, billing, collection and management of funds generated by tariff revenues), undertake routine maintenance and repair, and undertake minor upgrading and expansion using revenues from tariffs and connections fees. Operators will be selected by communities based primarily on the level of trust put in them under simple contracts drawn up by the community. These contracts will specify tariff levels and simple service quality standards. Contracts are expected to be for one year and will be renewable, with the objective of establishing a long-term relationship. Professional operators will be individuals from the respective communities, who will in turn hire a small number of employees to assist them. They will collect and own revenues, but they will pay a small fee to the water committee to be used for miscellaneous administrative expenses. Water committees will supervise professional operators by controlling their contractual performance on behalf of the community.

15 3. Monitoring and evaluation of outcomes/results

“Apre kase mayi, n ’a konte boujon.” “After harvesting the corn, we’ll count the sprouts.” (Haitian proverb cautioning not to count on results prematurely.)

Monitoring and evaluation of outcomes/results will use indicators and mechanisms defined in detail in Annex 3. In particular, indicators will include (i)the number of beneficiaries gaining access to sustainable water systems; and (ii)the number of beneficiaries gaining access to basic sanitation s ys tems.

The monitoring and evaluation system clearly distinguishes between intermediate and final outcomes. For example, in water supply, not only will data on beneficiaries in the service area of new or rehabilitated systems based be collected through reports by the implementing agency (intermediate outcome), but more meaningful outcome indicators will be collected at two levels through an independently conducted survey one year after construction. At a first level, the survey will assess if beneficiaries actually use the system. Reasons why some beneficiaries in the service area of a water system do not have access or do not use it may include: (i)the yield of the source is insufficient, (ii)diversion of water at the head of the system (vandalism); (iii)the number of actual connections exceeds the number determined in the technical design based on water availability; (iv) users are reverting to unsafe water sources because of lack of affordability or taste. At a second level, the survey will assess if the institutional and financial prerequisites are met for the system to remain sustainable after the project ends. Systems are considered to be sustainable if beneficiaries use systems and: (i)tariffs are set at a level that guarantees operation and routine maintenance of the system; (ii)at least 80% of revenues due are collected; (iii)the water committees meet on a regular basis; and (iv) professional operators fulfill their contractual obligations.

Health outcomes are not listed in the project results framework because the methodology of health outcome evaluation of water, sanitation, and hygiene interventions for an individual project such as this is complex, uncertain, and expensive.

A baseline survey will be undertaken by the service provider assisting SNEP. A second survey will be undertaken about one year after the hygiene and sanitation promotion campaign that will be executed in parallel with the construction of water facilities. In order to avoid a conflict of interest, this second survey will not be carried out by the service provider, but by an independent entity.

Project Contribution to the IDA-14 Results Measurement Agenda: Water and sanitation projects funded by IDA-14 are required to consider national sector monitoring data and capacity; where appropriate, the project should identify and respond to opportunities to strengthen national monitoring capacity.

16 National access statistics As noted on page 1, a variety of sources were consulted to assess the current national picture on access to water and sanitation. Relevant national coverage statistics as estimated by the World Health Organization’s (WHO) and UNICEF’s Joint Monitoring Program (JMP), the global body tasked by the UN for tracking progress towards the water and sanitation targets of the Millennium Development Goals (MDGs), are reported below:

Table 3: Water Supply and Sanitation Coverage

Water Surmlv Sanitation Access to Access to 1 Access to improved I immoved water I House Connection sanitation I Urban I 54% I 24 % I 51 % I Rural 54 % 3% 16 % Total 54 % 10 % 28 %

It should be noted that these estimates are based on data collected for the Demographic and Health Surveys (DHS) of 2000; a DHS has been carried out in 2005 and preliminary results have been published, but final results including water and sanitation coverage data are expected to be published only in February 2007.

Opportunities to strengthen national monitoring capacity. The IDB loan will fund the establishment of a basic national database on rural water supply systems. Collection and analysis of data for this system will be among the first steps to increasing SNEP’s understanding and control of its assets, and will serve as a basis for monitoring the sector and its performance. In addition, application of this project’s Results Framework (Annex 3) will not only determine access within the communities included in the project, but will also provide valuable experience to SNEP and others in the rural water sector in modem methods of assessing access to water and sanitation.

4. Sustainability

N Kabrit anpil m8t mouri grangou nan sol8y. >> “The goat of many owners dies hungry in the sun.” (Proverb used by some Haitians to explain why community-managed infrastructure has failed to be maintained in Haiti.)

A key problem in rural water supply systems in Haiti is the lack of sustainability, due to low-quality construction and lack of maintenance. Many water systems fall apart less than 10 years after construction, leading communities to petition donors to rehabilitate them. The low quality of construction is due to poor technical designs, poor quality of execution of works, inadequate supervision of works (often done by the communities themselves, who lack technical expertise). The proposed project will address this problem by using experienced engineering firms (working in a consortium with NGOs or firms specialized in ingknierie sociale) to produce technical designs. Furthermore, experienced contractors will be hired to organize the execution of works under the supervision of engineers. The whole process will be closely monitored by the

17 specialized staff of the implementing agency. This will reduce the risk of poor design and execution of works which has been so detrimental to the sustainability of infrastructure in Haiti.

Lack of preventive maintenance and repair, in turn, is due to poor management and insufficiency of funds. About two-thirds of investments in the sector are dedicated to rehabilitation rather than the construction of new infrastructure. This is widely attributed to the lack of payment for water by the majority of users (those receiving water from standposts), the low tariffs charged for house connections, and to the lack of a specific person that can be held accountable for the functioning of the water system. Poor maintenance leads to a vicious circle of declining service quality, reduced willingness to pay and further reduction of funds available for repair. Thus, community management of infrastructure is widely perceived as having failed at being sustainable in Haiti. In order to increase the likelihood of maintenance and repair, the project promotes a new model that foresees the hiring of professional ouerators by the water committees on behalf of the community. Professional operators would be contractually accountable for the functioning of the system and will be remunerated. Concerning the physical and financial aspects of systems, instead of providing water for free at standposts, water would be sold in buckets at metered water kiosks. House connections would also be metered and households would be billed based on their consumption.

Vandalism and misuse also compromise sustainability. Occasionally farmers in Haiti close to a source that supplies a village damage pipes to divert water for irrigation purposes. This risk would be mitigated through intensive community participation in setting rules for water use that are being perceived as fair by the community, avoiding the need for heavy-handed enforcement measures. If such enforcement should be necessary, professional operators and water committees would defer the issue to local government, in particular to the CASEC (Conseil d’Administration de la Section Communale) of the respective communal section to which the locality belongs.

The risk of lacking environmental sustainability exacerbates the problem, as the dry season yield of many springs has declined significantly because of deforestation and possibly because of climate change. One of the eligibility criteria under the project is a dry-season yield sufficient to supply the expected future population served by a piped system. In addition, communities will be encouraged to protect and, if need be, to reforest the catchment area of springs. Funding for reforestation could be provided from a basin management project supported by the IDB in part of the Southern Department (watershed of Les Cayes and Cavaillon).

5. Critical risks and possible controversial aspects

Risk Risk Minimization Measure(s) I Rating Insecurity and civil unrest Insecurity largely affects Port-au-Prince, where MTPTC and SNEP central office are located. Project’s proposed area of operations in the S Southern Department does not present a major security risk.

Limited human and Substantial effort at capacity building, including M institutional capacity in recruitment of junior staff, use of senior

18 ~~ SNEP consultants, incentive payments to staff and substantial training targeted at key deficiencies. Lack of support at the Discussion of new models as part of preparation community level for new of the national water and sanitation strategy and water management models flexibility in the application of the principles; testing of the approach under the LICUS grant; M survey of the private sector under the study on the applicability of Output-Based Aid in the water sector. Insufficient donor Intensive dialogue with key donors, going beyond coordination the mechanisms from the sector table under the ICF. Parallel financing by IDB. Dialogue on N national strategy for rural water supply and sanitation. Inability to achieve cost Investment funds will be passed on as grants (as recovery levels necessary for opposed to loans) to communities in order to keep sustainability payments affordable; demand-responsive S approach based on willingness to pay; modest and realistic formulation of cost recovery targets. Transition from SNEP to Requiring an assessment of the fiduciary capacity new structures to be created of the new structures before transferring function as part of planned reforms from SNEP; Maintaining trust and open M (the new structures are communication channels with senior staff in expected to be created in MTPTC; close consultation with IDB as the focal about three years) point for the sector. Limited financial Strengthening of financial management capacity management capacity for the of SNEP through training and technical M project in SNEP assistance; regular audits; focus on financial management in supervision Limited procurement Strengthening of procurement capacity already capacity for the project in under way, but needs to be scaled up; hiring of a H SNEP full-time procurement expert to cover both the IDB and World Bank projects. Lack of physical Use of LICUS grant experience to improve sustainability of rural water sustainable systems; emphasis on good quality supply systems in construction and community ownership; participating localities assessment of current and expected future land use in catchment areas during project preparation; M assignment of priority to localities that have put in place measures to protect catchments and elimination of localities with catchments that are at risk.

19 Lack of capacity in Intensive technical assistance in program design; systematic, behavior-focused quality control and supervision of activities; M hygiene promotion training for NGOs and SNEP staff.

OVERALL PROJECT M RISK RATING

H - High S - Substantial M - Moderate N - Negligible or Low

6. Grant conditions and covenants

The project does not have any board or disbursement conditions.

Effectiveness Conditions:

Subsidiary agreement satisfactory to IDA between MTPTC and SNEP signed. Approval by SNEP and MTPTC of an operational manual satisfactory to IDA.

Legal covenants The following standard financial management conditions should be included in the legal agreement: Maintenance of an adequate financial management system throughout the life of the project. Submission of half-yearly interim financial reports as part of the project progress report no later than 30 days after the end of each semester. Appointment of a project auditor acceptable to the Bank selected according terms of reference acceptable to the Bank. Submission of the annual audit reports of the project’s financial statements and SNEP financial statements no later than four months after the end of each fiscal year of the borrower.

D. APPRAISAL SUMMARY

1. Economic and financial analyses

Economic benefits from the water supply systems are estimated based on a comparison of net benefits in a “with project” scenario and a “without project” scenario for newly constructed piped rural water supply projects. The “without project” scenario consists of a situation where a small amount of water fetched from unprotected and often distant sources is consumed, while in the “with project’’ scenario a larger amount of water is used at or near the house. In order to estimate the benefits from improved water supply, a demand curve has been constructed with data on water consumption and distance to source, which is used as a proxy for water price. The data come from a field investigation in 84 localities in the four participating

20 departments. The benefits are estimated based on the increase in consumer surplus, divided into benefits from (a) reduced cost to provide the quantity of water currently consumed, and (b) benefits from increased water consumption.

The economic internal rate of return for the selected sample is 28 percent. Sensitivity analysis is expected to show that the rate of return is robust against key risks such as an increase in investment costs, construction delays, increase in Operation and Maintenance (O&M) costs and a decrease in benefits. Additional health benefits are expected, but have not been included in the analysis since the benefits listed above already largely justify the project interventions by themselves.

The financial analysis of piped water projects is based on two scenarios: (a) traditional system with stand posts and un-metered house connections; and (b) a system with water kiosks and metered house connections, as recommended under the project. The analysis confirms that the traditional system is not financially sustainable. The proposed intervention is financially sustainable in the sense that, at a minimum, the costs of operation and routine maintenance are expected to be recovered. The analysis is based on stated willingness to pay by users in 84 localities in the four departments proposed for this project, as well as the general experience that collection efficiency is high if community mobilization and service quality are high, two aspects that are particularly emphasized in the proposed project. (See Annex 9 for more details.)

2. Technical

The project will meet water supply needs through a combination of a limited number of technical options, based on the specific conditions in each community. These include: (1) spring-fed gravity piped networks, (2) individual or collective boreholes (using hand pumps, solar or wind power) or protected hand-dug wells, and (3) individual rainwater catchment systems. Pros and cons of each or these options are assessed below:

Spring-fed gravity piped networks offer a number of important advantages. This is a technology with which the client is familiar, and offers high quality water with a minimum of complexity. Capital costs can be high where sources are distant from populations, but operating costs and maintenance requirements are minimal, as neither treatment nor pumping is required (unless localities are located higher than springs). Where high quality springs are close enough to the population to be served, this is technically a near-ideal solution, but the other technologies become more advantageous where springs are distant from population centers, and/or where populations are much dispersed.

% Individual or collective boreholes have been constructed, as well as hand-dug wells, in both local practice and previous projects. Boreholes can generally be located close to the populations to be served, minimizing distribution costs. In addition, they can penetrate the water-bearing layers (aquifers) more effectively than hand-dug wells, thus making more water available. Groundwater is usually of high quality, eliminating the need for treatment, although this will depend to some degree upon location, the depth of aquifer, the quality of construction, and the sanitary protection of the environs of the well. Boreholes, however, have at least 3 significant drawbacks: (1) their construction costs can be high depending upon the soil conditions, the

21 number of holes drilled in a given region, the drilling technology chosen, and the depth required; (2) they require pumping to lift water to the surface, implying higher operation and maintenance costs; and (3) whenever the pumping mechanism fails, the tributary population will have no access to water.

Protected hand-dug wells do not require such complex technology for construction as boreholes do, and eliminate the problem that arises when mechanical or hand pumps fail, since the population can still obtain water through the use of buckets. However, hand-dug wells are limited in depth, cannot penetrate very far below the water table, and cannot draw from deeper aquifers, which often yield greater flow at better quality. They are most likely to be attractive options in the flood plains adjacent to rivers and streams, and in these situations will contribute more to water quality improvement than to reduced travel time.

Whenever pumping should be necessary and cost-effective, the project would give a strong preference to renewable energy. Given widespread theft of photovoltaic cells in Haiti, wind energy would be the preferred source of energy, relying on hurricane-proof new technologies as they are about to be tested in Haiti under a pilot project supported by the World Bank’s Development Marketplace. Pumping could be required both for boreholes and, potentially, for springs if the locality to be served is located higher than the spring.

Rainwater catchment may be an attractive option to consider where (1) rainfall is substantial and dry periods relatively short, (2) groundwater is very deep, (3) source springs for gravity-fed systems would be distant, and/or (4) populations are much dispersed. Operation and maintenance of individual rainwater catchment systems are minor, and can legitimately be assigned to the household. Maintenance of collective rainwater catchment systems is technically simple, but institutionally more complex. Generally speaking, rainwater catchment is most appropriate as a critical supplement to water supply rather than as a complete source for all domestic water needs. It can, however, offer higher quality water with minimal treatment requirements at the household level. Experience in Haiti has shown this technology to be most commonly applied in the “mornes” or highlands, in those particular locations where springs and other groundwater sources lie too far below the elevation of the community to be served.

The project is not considering surface water collection and treatment, on the grounds that the treatment required will be technically complex, relatively expensive, and beyond the capacity of the local management structures foreseen. Solutions requiring fossil fuels are not being considered because of the high operating (and institutional) costs of their management. Solar or wind power may be considered as sources of energy for pumping; while the operating costs of solar and wind power are minimal, the capital costs are significant, and the risks of solar panel thefts or damage from hurricanes are real.

Sanitation needs will be met through a combination of community and individual interventions. Blocs sanituires will be constructed at schools and health centers and onsite options will be promoted for households. The individual options consist largely of latrines, with the possibility of some septic tanks where individual water supply house connections are made.

22 Onsite sanitation options to be promoted will be variants on the standard low-cost sanitation technologies widely used and proven elsewhere. These include pit latrines, ventilated improved pit latrines, and pour flush toilets as options where water use is low, and some form of septic tank where water use is high and soil conditions permit. (Alternatively, pit latrines with separate soak ways or disposal for sullage may be considered where the infiltration capacity of the soil is very limited, or where there is demand for reuse of sullage.) Sewerage is not being considered a sanitation option in this project.

3. Fiduciary

3.1 Financial Management The financial management capacity assessment of the project implementing entity (SNEP) has been assessed. The overall conclusion is that SNEP has adequate financial management capacity. Although Haiti public financial management is still affected by significant weaknesses despite major improvements during the last three years, and is confronted with a high level of perceived corruption, the impact of these risks on the project is likely to be limited given the defined implementation arrangements. These arrangements do not rely on the central government’s system. Instead they build on a continuing entity that has established an adequate financial management system, procedures and staffing, and successfully implemented a LICUS Trust Fund project. In addition, specific financial management procedures developed solely for the project with adequate level of controls will be followed. The project itself does not present any significant risk and its overall financial management risk is rated moderate. The detailed assessment is presented in Annex 7 of the PAD and in the FM assessment report.

3.2 Procurement

The key issues and risks concerning procurement implementation for the project arise from the many weaknesses in the Haitian public procurement system. These weaknesses were identified by a Country Procurement Assessment Report prepared by Bank in 1999 and confirmed by the Government and multiple donors in the context of the Interim Cooperation Framework (ICF) prepared in 2004, the Extended ICF of 2006 and the Interim Poverty Reduction Strategy Paper of September 2006. The proposed Economic Governance Reform Operation (EGRO 11) project and ongoing Economic Governance Technical Assistance grants (EGTAG 1 and 2) provide support to the Government for the implementation of procurement reforms.

In addition to the reforms supported by the EGRO I1and the EGTAG grants, procurement risks will be mitigated in the case of the proposed project through the use of close and frequent supervision, including full prior review of all contracts financed by the grant. Further, as a result of the Government’s commitment to procurement reform and the application of the Bank’s Guidelines in the other Bank-financed operations, government agencies such as PL-480, the Ministry of Finance, and the Direction de la Protection Civile, have become familiar with the basic principles of open and fair procurement, as SNEP did through its implementation of the LICUS grant for Rural Water Supply. This familiarity, combined with the ongoing training and

23 broad dissemination of information on new procurement procedures to all purchasing agencies, is expected to mitigate risks.

Nonetheless, the procurement risk is still considered high. Once SNEP’s procurement staff have been fully trained in the Bank’s requirements and project implementation is underway, their capacity will be further assessed during the course of the Bank’s intensive supervision of the project.

4. Social

The proposed project, by virtue of its geographic focus on rural areas, targets some of the poorest and most underserved of Haiti’s citizens. Rural areas exhibit the highest rates of poverty in the country, and rural infrastructure is severely lacking.

Specifically, the project will have the greatest impact on women and children, who represent 49.5 percent and 45 percent of the rural population, respectively.” Women and children are generally responsible for fetching water, a task that can be very time consuming. Furthermore, lack of access to safe water and sanitation has the greatest health impact on children, as they can be more susceptible to water borne illnesses. In a small study done in three villages in the Southern Department as part of the LICUS project, 38 percent of the 7000 families surveyed had had at least one member with diarrhea in the week preceding the survey. On the national level, Haiti’s average infant mortality rate of 76 deaths per 1,000 live births is far higher than the regional average of 27.16

Women and children are critical to the spread of good practices in hygiene and sanitation. Hygiene messages delivered through local schools in the proposed project will rely on children to share lessons learned with their families. Mothers who adopt these practices can have a strong impact on their families’ hygiene and sanitation habits, as they are generally in charge of cooking and cleaning. The experience of the Community Driven Development (CDD) pilot project in Haiti showed that women were involved in the community organizations and project development councils. Special efforts will be made to encourage involvement of women in the water committees short of mandating their participation.

Participatory Approach in the Haitian Context

Haiti has a tradition of deference to individual leaders, but collective efforts on a smaller scale do function. A recent World Bank study17 determined that membership in an organization reduces the probability of being poor. Examples include collective work groups (konbits); microcredit initiatives managed by a small group; farming “by half,” meaning that the owner of a garden rents half of his terrain to a farmer who farms the entire property and gives half the harvest to the owner; the quarry owner in the town of Pemerle in the Southern Department (where the LICUS grant has intervened) who formed a committee of five people to collect a tax l5EnquQte Modalit&, Morbidite et Utilisation des Services, Haiti 2000: httu://www.measuredhs.coduubd~df/FR121/02chauitre02.udf (Children under 15 years of age.) l6UNDP Human Development Report (2005). ” World Bank (2005). Haiti Agriculture and Rural Development: Diagnostic and Proposals for Agriculture and Rural Development Policies and Strategies.

24 from the community; and water committees that manage small-scale water systems (hand pumps) as a group.

The proposed project will take advantage of the experience with village leadership dynamics and participatory techniques gained by the existing LICUS pilot water supply project (see above under section 2, Rationale for the project). Among the many lessons of the project are: i)Village structure in Haiti is hierarchical, with many community members deferring to acknowledged local leaders to make decisions on their behalf; ii)interventions following the inginierie sociale approach need to be more intensive and require more skills and resources than those of SNEP’s single community animator in the region.

The social sustainability of the project relies in large part on the service provider that would provide the functions of ingknierie sociale under the project. The proposed project requires a strong emphasis on skilled community animators to avoid marginalizing community members and sacrificing sustainability. However, the challenge of this should be recognized in a country with limited resources/capacity in this area, and a cultural tradition of decision making by a small group of leaders.

Communication is key to an effective participatory approach. Because of the limited capacity of SNEP in this field it will be assisted by a competent service provider, as mentioned above, who will take the lead in communicating with participating communities. Community animators face a communications challenge in a number of regards: logistics of arranging meetings, keeping records of what was saiaagreed upon, eliciting the opinions of the population without leading or raising expectations, and working against a traditional top-down decision making approach. The LICUS project underscored the importance of keeping written records of meetings with communities, as this is not in the local tradition and can prove useful when accounts of what was agreed upon differ. Project staff must also make sure that communities receive technical designs in time before launching bids for works at a formal meeting where they are asked for comments, and that the result is ratified by a vote that is documented in written minutes, even if that takes more time. Trade-offs are an effective way of presenting the choices to be made by the communities, such as the pros and cons of more kiosks vs. more house connections. It also must be made clear that budgets cannot be amended once the works contract is launched. As a lesson from earlier experience, the respective responsibilities of communities, SNEP and its service provider will be included in a grant agreement to be signed between SNEP and the local community represented by its water committee at the feasibility stage of the project.

5. Environment

An environmental framework has been completed, reviewed by the Bank and posted on SNEP’s as well as the Bank’s website. The framework was reviewed by the Bank and found to be satisfactory. It analyzes the project’s main environmental impacts, proposes measures to mitigate them, and its Environmental Management Plan outlines steps to be taken for the environmental assessment of sub-projects. The environmental assessment also satisfactorily covers issues related to natural habitat and cultural property (see Annex 9 for details).

25 6. Safeguard policies

Safeguard Policies Triggered by the Project Yes No Environmental Assessment (OP/BP 4.01) [XI [I Natural Habitats (OP/BP 4.04) [XI [I Pest Management (OP 4.09) [I [XI Cultural Property (OPN 11.03, being revised as OP 4.11) [XI [I Involuntary Resettlement (OP/BP 4.12) [I [XI

7. Policy Exceptions and Readiness

No policy exceptions are being sought.

~~ ~~~ * By supporting the proposed project, the Bank does not intend to prejudice the final determination of the parties' claims on the disputed areas

26 Annex 1: Country and Sector or Program Background HAITI: Rural Water and Sanitation Project

A. Country Background

Haiti is the poorest country in the Western Hemisphere and one of the most disadvantaged in the world. The vast majority of its population of roughly 8.3 million people lives in extreme poverty. This is especially true of the nearly two-thirds of the population who live in rural areas, 86 percent of whom live on less than $2 a day and 69 percent of whom live on less than $1 a day.” The country’s health and social indicators are on the same level as those of the world’s poorest countries: Haiti’s child mortality rate of 118 deaths per 1,000 is far higher than the Latin American regional average of 33, and closer to sub-Saharan Africa’s average of 171.”

Poverty strikes hardest in rural areas, where education and economic opportunities are extremely limited, and basic social services and infrastructure are severely lacking. Basic rural infrastructure (e.g., water and irrigation, feeder roads, electricity, and sanitation) is virtually absent or severely depleted, reinforcing isolation and exclusion. These conditions contribute to rural-urban migration, especially into the capital, Port-au-Prince.

Haiti is crippled by pervasive and long-standing social and political turmoil. For the past twenty years, the country has struggled to emerge from a cycle of political instability and internal conflicts that has devastated its economy, further weakened state institutions, multiplied poor governance practices, augmented levels of poverty and, at times, prompted the withdrawal of external assistance to the government. The most recent political crisis occurred in February 2004, when President Aristide resigned under pressure and left the country. The political vacuum was filled by a Transition Government under President Boniface Alexandre and Prime Minister GCrard Latortue. The Transition Government has fulfilled its mandate to hold presidential and legislative elections, leading to the election of Rent PrCval in February 2006.

The Transition Government attempted to increase political and social stability and jumpstart the economy through the Interim Cooperative Framework, which focused on “restoring security, reestablishing public services, and providing economic opportunities for those most affected by the crisis.” The strategy of the transition government was articulated through four focus areas: 1) strengthening political governance and promoting national dialogue; 2) strengthening economic governance and contributing to institutional development; 3) promoting economic recovery; 4) improving access to basic services.*’

The PrCval government has focused on a program of social appeasement (Programme d’Apaisement Sociale), with broad geographic coverage including scaled-up activities in the slum areas of Port-au-Prince. At a donors conference in July 2006, the government outlined its plans for social, economic and institutional development for the first 15 months in office. Haiti obtained some $750 million in donor pledges to cover the extension of the Interim Cooperation Framework (ICF) until September 2007.

l8World Bank (2006), Haiti Country Economic Memoradum. WDI indicators, 2003 Interim Cooperative Framework Sumary Report, Government of Haiti, July 2004.

27 B. Water and Sanitation Coverage

Although data quality is poor, it is generally accepted that water and sanitation coverage in Haiti are the lowest in the Western Hemisphere. Estimated coverage is especially low in rural areas: the 2000 DHS estimates that 54 percent of rural population has access to “improved water,” which includes a household connection, public standpipe, borehole, protected dug well, protected spring and rainwater collection. The breakdown for Haiti’s rural population is as follows: 4 percent house connection, 32 percent standpipes, 6 percent protected wells, 10 percent protected springs, 2 percent rainwater collection, 3 1 percent from unprotected springs, 6 percent from unprotected wells and 9 percent from rivers or canals.

Sanitation coverage is significantly lower than water coverage: The DHS estimate for rural access to improved sanitation is 16 percent, which includes connection to a septic system, pour- flush latrine, simple pit latrine or a ventilated improved pit latrine. The information base for coverage figures is very weak and any coverage figures from Haiti, therefore, have to be taken as rough estimates. Those who do not have access to improved sanitation usually have to resort to open defecation. Those who are not served by an improved water source often have to walk long distances to fetch insufficient quantities of water of dubious quality. In urban areas, they pay high prices to water tankers and for bottled water.

Sewer systems and wastewater treatment are nonexistent in Haiti. Storm water drainage, where it exists, consists of open drains that are often clogged by solid waste. This exacerbates already precarious sanitary conditions, especially in slum areas that are home to about half of Haiti’s urban population. Septic tanks in urban areas contribute to the pollution of aquifers that serve as sources for public water supply. In some cases, sewage is directly discharged into open watercourses.

Service Quality

The above coverage rates do not give an indication of service quality, which is generally quite poor. In rural areas, systems have often fallen into disrepair. They either do not provide any water service at all or provide service only to those close to the source, with those at the end of the system (“tail-enders”) remaining without water. In almost all urban areas water supply is intermittent.

Institutions

The main institutions in the Haitian water sector are community-based local water committees and two state-owned enterprises, each created by its own law: CAMEP (Centrale Autonome Me‘tropolitaine d’Euu Potable), responsible for the Port-au-Prince metropolitan area, and SNEP (Service National d’Eau Potable), responsible for secondary cities and rural areas. There is no institutional responsibility for sanitation in Haiti, since the mandates of CAMEP and SNEP currently do not include sanitation. Both entities theoretically are under the authority of Boards including representatives of several Ministries. Since these boards have not met for more

28 than a decade, both entities are de facto under the sole control of the Minister of Public Works, Transport and Communications (MTPTC).

Sector agencies have lost qualified and trained staff, often to NGOs and donor agencies, because of their low pay levels. Nevertheless, there are competent and motivated managers and staff in the public Haitian water sector. Unfortunately, they lack the political guidance and support as well as the financial means to be motivated and productive.

Other public entities active in the sector are FAES (Social Fund) and POCHEP, a small unit under the Ministry of Health that used to execute rural water supply and sanitation projects. Municipalities have almost no role in the sector, although they are expected to play a larger role in the future under a proposed new water law. NGOs perform a wide variety of functions in the sector and often attract the most qualified and motivated staff due to their higher salary levels. They are particularly active in rural areas, but also in small towns and slums.

The functions and performance of each of the institutions enumerated above is described in more detail below:

Water committees: Managing systems in rural areas, small towns and slums

There are hundreds of water committees, called CAEPs (Comissions d’Aprovisionnement en Eau Potable) or simply Comitks d’Eau, in charge of water systems in rural areas and small towns. Their degree of formalization and effectiveness varies considerably. The best water committees meet regularly, closely interact with the community, regularly collect revenues, hire a plumber who performs routine repairs, have a bank account and are registered and approved by SNEP. However, many water committees fall short of these expectations. There is no national or regional registry of water committees or water systems and, unlike in other countries, there are no associations of water committees at the municipal, departmental or national level. The Ministry of Public Works now refers to these committees officially as water and sanitation committees (Comitk d’Approvisionnement en Eau Potable et Assainissement-CAEPA) to reflect the broader role the committees are expected to play in the future.

While most water committees manage village water systems, fifteen of them manage small town systems, including Ouanaminthe, a few small towns in the Southern Department, Fort LibertC in the Northwest, and all the small towns in the Northern Department. In Port-au-Prince, numerous water committees operate water kiosks and manage funds generated from water sales for community projects. They work in close partnership with CAMEP, from which they buy their water.

SNEP: Managing systems in secondary towns and supporting water committees in rural areas

SNEP is organized in a national office, 24 operating centers (centres d’exploitation, also called BDS - Bureaux Dkconcentrks du Sewice) in secondary cities and small towns. SNEP’s primary activity is the operation and maintenance of water systems in small towns. All its revenues are generated from that activity, which involves the great majority of its staff.

29 SNEP is headed by a Director General. Three directors report to the Director General, all based in the national office in Port-au-Prince:

(i) the director of planning; (ii) the technical director; and (iii) the administrative and financial director.

In addition, the recently created rural unit (cellule rurule) reports directly to the Director General. In terms of planning and technical backstopping the rural unit so far has functioned autonomously, due to the weakness of SNEP’s planning and technical departments. The rural unit also carries out its own procurement under the LICUS grant, since SNEP has not undertaken procurement activities for many years in the absence of donor funding and thus has no procurement capacity outside the rural unit. The rural unit has no accounting system of its own. It uses the services of SNEP’s administrative and financial department for that purpose. The Director General intends to strengthen the planning and technical departments and aims at eventually mainstreaming some of the functions carried out by the rural unit into the regular structure of SNEP.

Historically, SNEP has not been able to be very active in rural areas, except for a few scattered projects, such as the CARE/USAID/SNEP project in the Southern Department in the early 1990s. During the execution of the LICUS grant, SNEP gained valuable experience in rural water supply, including working closely with local communities.

Since early 2005, SNEP has implemented a World Bank-financed LICUS grant of $855,000 for community-driven rural water supply and sanitation services. The rural unit of SNEP was created to implement this project. The two components of the grant were (i)the elaboration of a national rural water supply and sanitation strategy; and (ii)implementation of pilot projects in community driven rural water supply and sanitation in the Southern Department. The LICUS grant served as a pilot project for the proposed project. Currently, a draft strategy has been produced and construction of water systems in three rural communities is underway. SNEP’s rural unit has worked extensively with the communities through a participatory approach, ensuring that the principle of sustainability is central to the beneficiaries’ understanding of the project.

SNEP annual tariff revenues are about $550,000 a year, but have been much lower at times when bill collection efficiency dropped. Revenues per each of the 27,300 connections average $2 per month. Tariffs are determined roughly based on ability to pay, and vary between $1 in the central plateau and $7.30 in Kenscoff. The last tariff increase of about 35 percent was in January 2004. There were some protests, such as in Mirebalais, where SNEP had attempted to raise the tariff to $2.45 and ultimately was forced to lower it to $1.95.

SNEP receives a direct subsidy of almost $200,000 from the Ministry of Finance to cover the salaries of some employees of its central office. Another portion of the costs of the central office is paid for by a 25 percent share of revenues from the regional offices. That share is envisaged to be reduced as part of the effort to provide the operating centers with more autonomy and means. The central office, therefore, has to mobilize other sources of funding and make better use of its existing staff.

30 The central office of SNEP is expected to be strengthened with the support of the parallel IDB project in the following ways:

(i) a technical advisor to the Director General; (ii) a procurement advisor; (iii) a community development specialist in the national office (already recruited); and (iv) a database operator to help in establishing a sector information system.

In addition, in each of the four departments where the IDB will support interventions one engineer and one community development specialist would be recruited. In the same vein, the WB would finance the recruitment of an engineer and a community development specialist in the south.

In addition, the IDB provides financing for training in areas such as planning, project management, financial management, community development and various technical skills.

CAMEP: Serving the Metropolitan Area of Port-au-Prince

CAMEP is the utility in charge of serving the Metropolitan area of Port-au-Prince, consisting of six municipalities: Delmas, Pttionville, PAP, Tabarre, Carrefour and Citt Soleil. In theory, CAMEP has a board consisting of several Ministers. The six municipalities are not represented on the board. In reality, the Board does not meet and the Minister of Public Works chooses the director general and provides the policy directions.

There were reportedly 54,000 regular house connections (of which 35,000 were active) in CAMEP’s service area in 2004.*’ One connection is estimated to directly serve 6 people, plus another 15 people through improvised connections to neighbors, for a total of 21 people. Water is supplied between 2 and 10 hours per day in Port-au-Prince at least 3 days per week, depending on the neighborhood. Some neighborhoods in the city of Petionville (in the hills above Port-au- Prince) receive water just once a week. The rotations schedule is fixed, announced in offices and communicated verbally by CAMEP staff. Unaccounted for Water (UFW) is estimated at 45-65 percent, including a large share of stolen water.

Twenty public fountains on water mains provide free water, because people refuse to pay for water. If the water were not given for free, according to CAMEP management, the mains would be damaged. CAMEP would like to convert these fountains to water kiosks.

Metering is not widespread. CAMEP has only 3,000 metered connections, or about 8 percent of its active connections. CAMEP meters all the water committees in the informal settlements and its industrial customers. Some major consumers, such as luxury hotels, have disconnected from the public network and receive all their water via tankers.

~ ~ ~ ’*This number varies according to sources. Another source quotes 60,000 connections, out of which 43,000 are active. Reasons for inactivity could be voluntary disconnection by large consumers or forced disconnection because of non-payment.

31 Tariffs are flat rates due to the absence of metering for most customers. Tariffs have been regularly increased at substantial rates to keep pace with inflation, or even exceeding inflation. Price increases have been implemented by CAMEP after simple notification of Board members on a no-objection basis. Tariff increases are not implemented in service areas with poor service, such as in the isolated system of Petionville that is supplied by a source with low yield in summer. Tariffs are not transparent. There are six different categories of connections, each with different tariffs corresponding to a different consumption level, totaling 30 tariffs.

Annual revenues of CAMEP are about $3 million, about four times more than SNEP. The revenues barely cover operating costs, leaving insufficient resources for maintenance. Water cutoffs are frequent to enforce payments, partly because payments cannot be enforced through the legal system. Users have to pay to be reconnected. Staff in the billing collections department is rewarded for enforcing payments, as they receive bonuses if cost recovery increases.

CAMEP has about 750 employees, or 14 per 1,000 water connections-more than twice the efficient.leve1. CAMEP pays higher salaries than does SNEP. CAMEP has a unit charged with service provision for slums, called the UCQD (Unite‘ Centrale des Quartiers Dkfavorisks), which works closely with water committees that manage a total of 148 water kiosks in slums supplied with bulk water by CAMEP.

Ministry of Public Works, Transport and Communications (MTPTC)

The MTPTC does not have a water and sanitation directorate, although its creation is foreseen. Together with the fact that the boards of the enterprises are inactive, this led to a situation of ad- hoc interventions by the Minister in the absence of any long-term policy or performance targets. MTPTC now envisages creating a water and sanitation directorate as part of the Loi-Cadre. In the interim period, an advisor to the Minister is fulfilling this role. The unit would be in charge, among others, of developing a sector policy and of supervising the public enterprises in the sector.

The water reform unit (URSEP) in MTPTC is the project management unit of an IDB-funded water reform project. One of URSEP’s main functions is the implementation of water and sanitation interventions in small towns with systems operated by SNEP. It also promotes greater autonomy for the local branches of SNEP. URSEP may be integrated into the new water and sanitation directorate of MTPTC as part of the on-going reform process.

Ministry of Public Health and POCHEP

The Ministry of Public Health still houses POCHEP (Postes Communautaires d’HygiBne et d’Eau Potable), the Project Management Unit of an IDB rural water and sanitation project closed in 1998. Despite its name, the IDB’s POCHEP project covered water supply but not sanitation. POCHEP has one staff member and a database with contacts for contractual staff, in the hope of attracting donor funds. In the past, POCHEP worked with water committees all over the country, limiting it to gravity systems in order to reduce maintenance problems and recurrent costs. Now, POCHEP helps communities on occasion with spare parts from their stocks and executes small projects with government funding. Without major new works, however, its personnel are unmotivated and feel marginalized.

32 The Social Fund: more involved in funding than in a dialogue on sector policy

The Social Fund (FAES) is under the authority of the Ministry of Finance. Its board includes several representatives of ministries and professional associations, but no service NGOs. It receives funds from IDB, the International Fund for Agricultural Development and the UN Capital Fund. About 12 percent of its expenditures are for rural water supply and sanitation, supporting 103 projects under the IDB-funded local development project. It has an inoperative cooperation agreement with SNEP. So far, FAES has not worked extensively with NGOs or local governments, but has expressed a desire to involve mayors in its work. Its 60 staff members are located in the capital and in recently created regional offices. They are paid higher salaries than their public sector counterparts. Their projects are executed through contractors.

Although it executes a large number of water projects, FAES does not have a strong sector expertise in water or, for that matter, in sanitation or hygiene promotion. This sometimes leads to poor quality projects and low project sustainability. In at least one case, FAES has built relatively expensive water kiosks without asking users to pay.

Local government: marginalized, but expected to play a more important role

Municipalities have so far practically no role in water supply and sanitation. There has been an attempt to involve the municipalities indirectly in water and sanitation through the formation of advisory committees in small towns that are served by SNEP. These advisory committees, called COCEPAs (Comitks Communaux d’Eau Potable et de I’Assainissement), are made up of representatives of users, SNEP and municipalities and are supposed to play an advisory role and perform community outreach. In reality, very few COCEPAs exist.

There are various initiatives underway to give municipalities a greater role in managing their water systems. The IDB is helping some regional offices (Saint Marc and Port-de-Paix; afterwards possibly Les Cayes and Jacmel) to set up their own administrative, financial and accounting systems; undertake water and sanitation master plans and tariff studies; and introduce metering and perform community outreach.

In 2000, a pilot project was undertaken by the NGO GRET in partnership with two municipalities to attempt to give them more control over management of their water systems, in accordance with a proposed new law that ultimately was blocked. The results were mixed. The project, which was conducted in two small towns, Aquin and Mirebalais, included the installation of meters and foresaw volumetric pricing. Also included were surveys to understand user practices and water points, mapped in a GIs; a communication campaign; and a choice of different technologies at different costs. Although meters were installed, volumetric tariffs were never applied. There is some confusion about the reasons, but it seems that they were neither social nor technical, but purely political. The volumetric billing was accepted by the population and the meters functioned well, but following municipal elections the new local authorities refused to issue volumetric bills and decided to continue to bill on a lump-sum basis, despite the efforts undertaken previously.

33 Non-governmental organizations: performing a wide range of functions

NGOs in the water sector vary widely in size and scope of activities. Some are large and work in many areas throughout the country. Others are small and active only in one region. Some are branches of international NGOs, while others are local NGOs, with many shades in between. Some NGOs work in partnership with state institutions, while others work on their own. Many cover activities in various sectors and work only to a limited extent in water and sanitation, while others have water specialists among their staff. It seems that only few NGOs have developed specific expertise in sanitation and hygiene education.

Sector Reform Initiatives

The newly elected government intends to reform the sector through a water and sanitation framework law. The framework law (Zoi cadre) aims at creating autonomous regional water and sanitation companies replacing SNEP and the Port-au-Prince water utility, CAMEP. It also envisages the creation of a water and sanitation directorate in the Ministry of Public Works that will propose policies for the sector. Furthermore, a rural water and sanitation unit in the Ministry of Health, called POCHEP, will be dissolved and integrated into the new structures. Ultimately, the law envisages the gradual transfer of the responsibility for water and sanitation to those municipalities that have proven their capacity to undertake these responsibilities. Municipalities would in turn have the option to delegate service provision to the private sector, municipal water companies, or water committees. The law would also-for the first time in Haiti-define legal responsibility for sanitation within the public sector.

The proposed framework law constitutes a significant step forward compared to the existing piecemeal legal framework, which does not define clear responsibilities for policy and regulatory functions, does not allow for private sector participation, and does not define responsibilities for sanitation. The proposed structure of regional companies strikes a pragmatic balance between the necessity and political will to decentralize on the one hand, and the risk of fragmenting the sector through over zealous municipalization in the context of very weak capacity on the other. The passing of the law and an associated sector strategy is also expected to bring a new momentum to the sector, which was somewhat paralyzed and dispirited given that an earlier draft framework law that has been in existence since 1996 has not been passed in the absence of a functioning Parliament.

34 Annex 2: Major Related Projects Financed by the Bank and/or other Agencies HAITI: Rural Water and Sanitation Project

Bank-Financed Projects

1. Community-Driven Development Project (CDD) ($38 million): The Bank’s Community-Driven Development (CDD) project transfers resources to local community organizations in poor communities by (i) improving access to basic infrastructure and support income-generating activities at an estimated average cost of $20,000 per activity, and (ii) improving governance and building social capital of communities by increasing citizen participation and transparency. The CDD project will ultimately cover almost half the country’s rural communities. Concerning water supply, investments are limited to a small range of technical options (essentially hand-pumps). The water and sanitation project complements the CDD project, because it allows the financing of high-priority investments that could not be included under the CDD project because of their size. The proposed project will overlap geographically with the first phase of the CDD project in the Southern Department. The CDD project is implemented by the PL-480 Management Office under the Ministry of Planning and NGOs.

2. Emergency Recovery and Disaster Management Project (ERDM) ($12 million): The ERDM project aims at rehabilitating areas affected by two tropical storms in 2004, strengthening the country’s capacity to manage natural disasters, and to reduce the vulnerability of communities through risk identification and risk mitigation activities. The rehabilitation works will be partly carried out in areas covered by the proposed project (GonaYves in the Artibonite and Fonds Verettes in the Western Department), thus allowing for support in managing rehabilitated water systems from SNEP’s departmental offices, which will be strengthened under the proposed project. Concerning efforts to reduce disaster vulnerability, a CDD approach will be used in 70 communities, half of them in rural areas and including some in areas covered by the proposed project. Investments considered do not cover water supply and sanitation, but provide important benefits to water systems in terms of watershed protection (reforestation and terracing). Furthermore, the proposed project will strengthen SNEP’s capacity to respond to requests to possible requests by the Civil Protection Directorate (DPC) for support in the rehabilitation of water systems damaged by natural disasters. The project is implemented by the DPC of the Ministry of Interior and PL-480 Management Office.

3. Transport and Territorial Development Project ($16 million): This infrastructure rehabilitation project aims at creating growth poles in selected rural areas. Investments in the areas covered by the TTDP could be picked up under the current project to maximize the growth impact by improving public health and reducing the time burden of water fetching. Synergies could be obtained at least in both areas covered by the TTDP: The Dondon area in the North includes the municipality of Marmelade in the Artibonite department covered by the proposed project; and the Southeastern area includes the municipality of Fonds Verettes in the Western department included under the proposed project. The project is implemented by the PL-480 Management Office.

35 4. Economic Governance Reform Operation (EGRO 11) ($23 million): A proposed second EGRO operation (EGRO 11, of $23 million in two tranches) will support the continuation and deepening of the reforms introduced in the area of economic governance by EGRO I(which was fully disbursed in early 2006) and help improve the policy environment for increased Bank and other donor-supported investment lending in key sectors. Specifically, the EGRO I1 will: (i) promote efficiency, transparency, and accountability in public resource use through public financial management and procurement reforms; (ii)strengthen the public sector’s human resource management and employees’ accountability; (iii)improve efficiency and transparency in public infrastructure management; and (iv) support governance reforms in the education sector to promote accountability and transparency in the use of public education funds. The operation would also support ongoing efforts to help create a culture of public participation in the monitoring of economic governance reforms.

5. Economic Governance Technical Assistance Grant (EGTAG I and 11) ($2 million each): These projects aim to assist the Government of Haiti in strengthening its institutional capacity in the areas of budgetary planning and execution, investment and project planning, internal and external control, procurement, accountability, participatory monitoring and reform constituency building. The projects seek to: (i)strengthen the Government’s capacity to prepare and execute a budget and enhance financial management and procurement policies and procedures; (ii)support the development of a Human Resource Management System and the strengthening of human resources capacity under the direction of the Human Resource Unit within the Prime Minister’s Office; (iii)strengthen the Ministry of Finance’s Anticorruption Unit and its special program for civil society monitoring of economic governance reforms; and (iv) strengthen the Ministry of Economy and Finance’s (MEF) program for communication of economic governance reforms to the public and enhance MEF’s ability to coordinate donor programs in economic governance

6. Electricity Loss Reduction Project (PREPSEL) ($6 million): This project has two main objectives: (1) To lay the foundation for improving the operational and technical performance of Electricit6 d’ Haiti (EDH) through investments in Port-au-Prince; and (2) to reduce and stabilize the fiscal burden caused by the electricity sector on public finances through better governance of the sector and commercial re-orientation of EDH. The project has two components: loss reduction in Port-au-Prince and development of a medium-term strategy for the electricity sector.

Projects Financed by Other Donors

IDB rural water and sanitation project. The IDB has negotiated a $15 million rural water supply and sanitation project in close coordination with the Bank, which was approved by the IDB Board in September 2006. The IDB’s rural project will be implemented by SNEP and uses the same implementation mechanisms as the Bank project.

The IDB also has a water sector reform project and a local development project with the Social Fund. The other donors engaged in the rural water and sanitation sector in Haiti are the European Union, France, Belgium, Spain, Switzerland and a large number of NGOs. Coordination with

36 these donors will be strengthened through the ongoing consultation process as part of the preparation of the national rural water and sanitation strategy.

37 Annex 3: Results Framework and Monitoring HAITI: Rural Water and Sanitation Project

Results Framework

Increase access to and use of water supply and sanitation services in participating rural communities .22 Increasing the sustained and Number of beneficiaries gainin Assessment of efficacy of effective use of safe access to and using sustainable B approach to construction and management of rural water drinking water in water systems24 in participating supply, for broader use in participating localities. Haiti. communities. Improved use of Number of beneficiaries gaining Assessment of efficacy of in access to and using basic sanitation “limited subsidy” approach to sanitation sanitation, for broader use in participating systems2’ in participating Haiti. communities. localities,

22 Communities with 1,000-10,000 inhabitants. 23 Systems are considered to be “sustainable” if beneficiaries use systems and: (i)tariffs are set at a level that guarantees operation and routine maintenance of the system; (ii)at least 80% of revenues due are collected; (iii)the water committees meet on a regular basis; and (iv) professional operators fulfill their contractual obligations. 24 “Water systems” in terms of this project means systems that are “improved” according to the Joint Monitoring Programme definition of “improved” water supply. (e.g. Improvement in the access to piped water (for those households without access to piped water) or protection of a previously unprotected source.) *’“Basic sanitation systems” in terms of this project means systems that satisfy the Joint Monitoring Programme definition of “basic” sanitation contributing towards the MDG Target. (e.g., both simple and Ventilated Improved Pit latrines, twin pit pour flush latrines, and septic tanks) which protect the user and the community from fecal contamination.

38 Use of Intermediate Outcome Monitoring COMPONENT 1: Establishment and use of a sector This indicator will Increased capacity within information management system within reflect not only the SNEP for overseeing and the program (establishment and use of a quality and utility of the managing rural water standardized electronic database of rural monitoring system supply and sanitation water systems and water committees in developed by SNEP (RWSS). participating departments; and under this program, but maintaining an up-to-date database of also its potential utility investment plans by NGOs and FAES). for other investments. COMPONENT 2: See indicators by subcomponent below. Expanded access to water supply, basic sanitation and hygiene promotion in participating communities

Subcomponent 2A: Number of beneficiaries gaining access Assessment, for mid- Construction and/or to improved water supply systems. course correction, of rehabilitation of water project implementation. supply systems in participating communities. Subcomponent 2B : Percentage of households in participating Assessment, for mid- Systems in place in communities paying for water services. course correction, of participating communities sustainability of to effectively manage management models for water supply systems. water services. Subcomponent 2C: Number of households satisfying Assessment of efficacy Hygiene and sanitation preconditions of programme-specified of approach to promotion activities labour and/or cash investment in construction of latrines, underway in participating sanitation for provision of latrine slab. for broader use in Haiti. communities. Subcomponent 2D: Number of schools, health centres with Assessment of efficacy Construction of sanitation access to blocs sanitaires, and with well- of approach to facilities (blocs sunituires) defined plans and sources of funds for construction of blocs in schools and health their maintenance and management. sanitaires, for broader centres of participating use in Haiti. communities.

39 Arrangements for results monitoring

0 Institutional issues: How will monitoring and evaluation (M&E) complement project management? How will participatory M&E be integrated into management and capacity building for the communities involved (if applicable)?

The monitoring and evaluation program will be broadly conceived, and reflect issues of water supply service and sustainability, sanitation access and personal hygiene. Results will be fed back not only to SNEP, but also to the local CAEP. In addition, the CAEP and community will be involved in the participatory elements of overall hygiene assessment in the baseline and follow up surveys.

Data collection: If the project is drawing on data collected by Government statistical offices or line agencies, which statistics would be used and what is the reliability of this information? Where information is to be derived specifically for measurement of project results and outcomes, what are the associated costs and institutional responsibilities?

“Access” data will be based on representative household surveys conducted during feasibility studies (baseline) and on follow-up surveys conducted either by the implementing NGOs at baseline or by an independent consultant at project close, and are thus not determined by government data. The overall institutional management responsibility lies with SNEP, but the implementation will be subcontracted. In addition, participatory hygiene assessment is proposed at baseline and close as a way to determine the appropriate methods and channels for hygiene promotion. The cost of associated surveys will be financed by the project.

Capacity: Where there is limited capacity in the country to derive the necessary information, how will local capacity be supplemented through the project, and what will be the costs of doing this?

The experience of the Bank and other institutions (including that of the WHOLJNICEF Joint Monitoring Program for Water Supply and Sanitation) in the development and administration of household surveys for water and sanitation will be shared with the implementing consultant/NGOs (service provider) responsible for the household surveys. In addition, participatory methods of hygiene evaluation are already documented and available in French on the web, and are specifically mentioned in the Terms of Reference for the service provider. Bank staff will review the survey plans and methods proposed by the consultants.

40 8 P b B .-8

v)5 c

E s

Q tN

0 0

.n Annex 4: Detailed Project Description HAITI: Rural Water and Sanitation Project

Component 1: Capacity Building and Project Management

The activities under this component aim at managing the project and strengthening the capacity of SNEP. The component includes four main activities.

0 First, it will strengthen SNEP’s regional office in Les Cayes through the recruitment of two consultants who are expected to become staff after project closure, as well as through the provision of cars, office space and coverage of operating expenses. 0 Second, it will finance services for capacity building in project management, documentation, reporting, information technology, and environmental management. This activity is foreseen primarily during the first two years of project implementation, with the aim of making the assistance unnecessary during the remainder of the project. This activity will also include the financial audits of the project. 0 Third, the Bank would finance services for surveys to monitor the achievement of performance indicators at least one year after the construction of water and sanitation systems. Fourth, it will assist SNEP in finalizing and disseminating the national rural water supply and sanitation strategy initiated under the LICUS grant mentioned above.

The component complements $3 million for capacity building to be provided by the IDB under its parallel financing. The IDB financing covers two areas: First, it aims at strengthening SNEP’s functions in rural areas in close relation with the investment component, with an emphasis on community development and procurement. Second, it will strengthen SNEP’ s overall capacity throughout its mandate, including rural and urban areas. It foresees the hiring of an advisor to the Director General for the first two years of the project and the strengthening of the technical and financial departments of SNEP. It also foresees an assessment of SNEP’s human resources in view of a rationalization of its personnel and salary structure. In addition, the IDB will finance vehicles, equipment and recurrent costs necessary for the project.

Component 2: Water supply, basic sanitation and hygiene promotion

This component will enable participating communities to access and manage water supply and basic sanitation services. This includes the identification of participating communities based on objective criteria; ing4nierie ~ociaZe*~using a participatory approach allowing communities a series of choices;30engineering design; system construction and supervision; and support during the early years of system management. The project will promote the innovative

See explanation further above. 30 The proposed key choices are: choice between different water service levels (house connection or water kiosk) and sanitation service levels (pit latrine and Ventilated Improved Pit latrine) implying different costs to users; choice in the tariff structure (pure volumetric or combined flat rate and volumetric) and level (covering only operating costs and basic maintenance, or including a reserve for major maintenance).

42 management approach piloted under the LICUS grant involving professional operators, increased cost recovery, metering, and the use of water kiosks instead of standpipes.

Investments under this component are expected to consist mainly of piped water systems. The piped systems will feed water kiosks and house connections. Water kiosks consist of a small concrete structures with faucets on the outside and taps on the inside, thus allowing kiosk operators to sell water by the bucket. Both water kiosks and house connections would be metered. Most piped systems are expected to be gravity-fed by springs. If pumping should be necessary and cost-effective, the project would give a strong preference to renewable energy. Given widespread theft of photovoltaic cells in Haiti, wind energy would be the preferred source of energy, relying on hurricane-proof new technologies as they are about to be tested in Haiti under a pilot project supported by the World Bank’s Development Marketplace. Other water investments include hand pumps and rainwater harvesting systems.

Concerning sanitation, the project would support the construction of household latrines as well as blocs sanitaires next to schools and health centers. Sanitation investments will only be supported in localities where water systems will be built, in order to be able to carry out hygiene promotion (see further below) in a credible manner and to maximize health impacts. A bloc sunituire is a concrete structure with toilets and basic water supply fixtures in two compartments, one for menhoys and one for womedgirls.

Sanitation and hygiene promotion will be carried out in the same communities in which water services will be improved. This allows for the creation of synergies among water supply, sanitation and improved hygiene interventions. SNEP and Haitian NGOs have experience in the construction of blocs sunituires for health centers and schools, and the project will support the continuation of this work, with a focus on their operation and management as well as upon their physical construction.

43 institutional responsibility for hygiene promotion under this project. However, since SNEP has no experience in hygiene promotion, and given the weakness of other potential stakeholders such as the Direction de Z’Hygi2ne PubZique (DHP) in the Ministry of Health, the service provider (see under implementing arrangements) will also undertake hygiene promotion. Solid waste management is not considered a priority issue in rural areas and is thus not included under this project.

The financial policy under the program includes the following key elements:

Tariffs: Communities will charge tariffs that recover at a minimum full operation and routine maintenance costs. All consumption from piped systems is expected to be metered and tariffs will be volumetric. There will be tariffs for water sold in buckets at water kiosks, as well as tariffs for house connections, both residential and for public buildings. House connections will have a fixed portion including a minimal consumption and a volumetric portion for consumption exceeding the minimum. Tariffs for the use of hand pumps will be flat rates per household and month. Up-front financial contribution: Communities are required to come up with an up-front financial contribution before technical studies will be initiated. The purpose of the contribution is to be an indicator of willingness to pay. It should cover, at the minimum, three months of operation and routine maintenance costs of the system. These funds would not be mingled with project funds, but would rather be used by the community for small initial purchases (e.g., for chlorine and a motorcycle) or be kept in a reserve for repairs. In addition, households that receive a house connection would pay connection fees. Pass-through conditions for investment funds: Following common practice in Haiti and in many other countries, project funds for investments will be passed through to the community as a grant, given limited ability to pay and given that similar practice is exercised in urban areas.

Participating localities will be selected based on objective, transparent criteria. A tentative long list of about 40 localities has been established that would be reduced to a short list of about 12 localities based on an objective eligibility and priority criteria. Eligibility criteria include the number of inhabitants (less than 10,000); availability of a nearby sufficiently productive, good quality source of water; protection of the source or indication of willingness and likelihood that the source will be protected; estimated per capita costs of less than $150; acceptance to pay volumetric tariffs covering operation and routine maintenance costs; acceptance of the professional operator model; a formal letter requesting to be included in the project; and payment of an up-front financial contribution into a fund managed by the water committee. If the number of eligible localities is greater than about 12, participating localities will be selected based on priority criteria. These include greatest needs (as measured by average walking distance to currently used sources of water) and lowest estimated unit costs (as measured by estimates of the required works and the number of houses in the locality assessed based on satellite pictures). Data on estimated costs and benefits will be collected by the service provider on behalf of SNEP.

44 Total project costs are $5.25 million, including $0.92 million for the first component and $4.33 million for the second component. The costs will be financed through the grant ($5 million) and counterpart funds ($0.25 million).

Estimates of Project Costs, by Components ($ million)

Components Grant Counterpart Total Cost Capacity Building, and 0.67 0.25 0.92 Project Management (1) Water Supply, Basic 4.33 0.00 4.33 Sanitation and Hygiene Promotion (2) TOTAL Proiect Costs 5 .OO 0.25 5.25

45 v) In cu 0 €9 co 0 8 8 s 2 €9 2 2

v) 2

v) v) r- 0 co 0 8 8 2 2 2 2

c Q a, '0 E a, C 5 0 3 .-c ea 53 .- a, C 5 a C U .- s .I-t -h C Q a 8 3 C v) 0 .-.I- t 5 Q; 3 0 - v) 2 C 0 B 2 .-.I- W a B 3 z c - v) c w 9 0 a, W P) C .-8 .-8 .I- d .- 5 C 0 0 v) w a, C C C - .-C a, 0 a, 5 0 .-.I- w C C cb w .I-a .- .- .- C .-.I- .- .I- .I-; C C a 2 5 s z 8 5 0 v) s 7 Y m t t l- l- l- l- x 04 Annex 6: Implementation Arrangements HAITI: Rural Water and Sanitation Project

MTPTC (Ministkre des Travaux Publics, Transports et Communication)

SNEP (Service National d’Eau Potable)

Service Provider (NGO or consulting firm)

Professional Operators

47 General The institutional arrangements of this project, summarized in the above table, are largely based upon the implementation arrangements under an ongoing LICUS grant for rural water supply and sanitation. The Ministkre des Travaux Publics, Transports et Communications (MTPTC), the government ministry responsible for water supply and sanitation in Haiti, would sign a subsidiary agreement with SNEP, as under the LICUS grant. While the implementation arrangements for the LICUS grant have proven effective, the proposed program supported by the Bank and IDB will require an absorption capacity that is more than ten times larger ($4 milliodyear compared to $0.425 milliodyear). In order to ensure the ability of SNEP to implement the activities under the program, the following measures have been initiated or are envisaged: Recruitment of an experienced service provider (see below). Individual consultants hired by SNEP with the expectation that they would become SNEP staff when the project ends. Capacity building within SNEP for project management and procurement, including supervision of the service provider.

The institutional characteristics of the proposed projectlprogram and the LICUS grant are compared in the table below:

Table 1: Comparison of proposed program and predecessor pilot project

Proposed Program Predecessor Pilot Project WB and IDB) (LICUS grant) Ingknierie Carried out by an experienced Carried out in-house by SNEP with sociale service provider, in partnership support from two international with SNEP volunteers (EWB-Canada) Procurement and SNEP, supported by consultants SNEP with intensive Bank project and significant training with supervision and support by EWB management intensive capacity building program and Bank supervision Technical design Carried out by the same service Engineering firm provider or consortium that will undertake the ingknierie sociale Construction I Contractors Contractors Construction Engineering firm or service Individual consultant Supervision provider Geographical About 50 localities in 5 3 localities in 2 departments Coverage devartments Duration 5 years without the preparation 2 years incl. preparation phase 1 vhase Amount I $20 million $0.85 million

SNEP is the executing agency for the proposed project. SNEP will have the fiduciary responsibility for the project (financial management and procurement). SNEP has selected one of its current employees to become a full-time procurement staff. He and other SNEP staff have

48 been trained by the procurement specialist of the IDB office in Haiti and by the National Procurement Commission, which receives support from the Bank and IDB.

In order to assist in the process of ingknierie sociale, and to carry out engineering designs, SNEP is in the process of selecting a service provider (NGO, consulting firm or a consortium of both) for the Southern Department on a competitive basis. The service provider will have as one of its tasks to strengthen the capacity in the local office of SNEP and they will be selected partly based on their methodology and track record in working with and strengthening of public institutions.

Local water committees (CAEP) would have important responsibilities during the feasibility and construction stage. The CAEP is a community-based organization composed of five volunteers elected by the community. Given the weakness of most existing water committees, substantial assistance would be provided to existing and new water committees in order not to overload them during these crucial phases. While water committees have an important role, communicating with them is no substitute for communication with the entire community. SNEP and its service provider (NGO or consulting firm) would, therefore, organize community consultations on key aspects of the project. These would include community contributions, design and location of infrastructure, tariff levels, as well as explaining the concept of and rationale for professional operators and water meters. During the construction stage, the CAEP will organize the community contribution, check and approve the work, and assist SNEP and its service provider in resolving any differences between the community and the contractor.

In the operating phase, the role of the water committees will change compared to the role they have played in the sector so far. As mentioned earlier, most water committees do not have the capacity to adequately operate and maintain water systems. Under the proposed project, water committees would delegate operation and maintenance to remunerated professional operators. Professional operators are expected to operate systems (chlorination, cleaning of reservoirs, operation of kiosks, meter reading, billing, and collection and management of funds generated by tariff revenues), undertake routine maintenance, and upgrade and expand systems on a limited basis using revenues from tariffs and connections fees. Operators will be selected by communities based primarily on the level of trust put in them under simple contracts drawn up by the community. These contracts will specify tariff levels and simple service quality standards. Contracts are expected to be for one year and will be renewable, with the objective of establishing a long-term relationship. Professional operators will be individuals from the respective communities, who will in turn hire a small number of employees to assist them. They will collect and own revenues, but they will pay a small fee to the community to be used for miscellaneous administrative expenses of the water committee. Water committees will supervise professional operators by controlling their contractual performance on behalf of the community.

In order to attract and retain qualified staff and build capacity in SNEP, two approaches will be used in parallel: recruitment of consultants and salary top-ups. Both approaches have been tested in SNEP, and both approaches entail their own benefits and risks. Recruitment of consultants, some of whom are recruited from outside and some of whom are SNEP employees who have resigned from their positions, entails a lower financial management risk. However, it

49 increases the risk that staff may not return to SNEP after their contracts end, and it may create tensions between employees and consultants. Salary top-ups are eligible for Bank financing under OP/BP 6.00. Salary top-ups are routinely used by donors in Haiti and the Ministry of Economy and Finance (MEF) supports their use.

The Bank’s approach to government salary top-ups and the use of consultants--as reflected in this project--is to ensure transparency, fairness, sustainability, compatibility with local markets, and avoidance of conflicts of interest. In the case of the proposed project, the Bank and the IDB share a common approach, stemming from the commitment to work through a government agency as opposed to creating a self-standing PMU. Top-ups are proposed by the Director General based on recommendations by the project coordinator. Their level is in many cases 50 percent of the base salary and does not exceed 100 percent of the base salary of each employee. These levels, together with per diems, have shown to be sufficient for staff to significantly increase their motivation. Likewise, fees for consultants will not exceed the double of comparative salaries and will often remain below that level. These arrangements are meant to increase the probability that consultants will accept to be recruited as employees after the project ends. Top-ups are reviewed by each donor ex-ante for every eligible staff member in order to reduce the risk of abuse and to ensure that they will be linked to actual performance. Ex-post reviews will also be undertaken on a six-month basis to ensure the fairness of the mechanism. Up to now the mechanism has not been perceived as unfair or inconsistent, mainly because of the way the Director General has administered it. Top-ups have not creamed off top staff from other activities within the institution, because of the relatively small fraction of eligible staff. There is also no evidence so far that top-ups or consultant contracts have weakened other public sector entities, since the only outside consultant recruited so far has not worked for the government before. Top-ups and consultant contracts are expected to continue to contribute to an atmosphere of elevated standards and performance that would further strengthen the implementing agency, as has been the case during the LICUS grant.

Counterpart funding is expected to create a precedent for annual budget allocations to the public agency in charge of rural water supply and sanitation, thus fostering institutional sustainability. Certain recurrent costs, including individual consultants, per diems and operating costs are expected to be funded through the government’s counterpart contribution beginning in the Haitian fiscal year 2007-2008. The total of these costs is estimated to be $250,000 over the life of the project. The counterpart financing remains to be confirmed through the annual budget law voted by Parliament. If counterpart funding should not be sufficient to cover these costs, the Bank would have to finance these costs from the contingency allocation.

The IDB has not requested counterpart funding and will finance all costs, including consultants, per diems and top-ups, through its credit. It covers these cost items for staff and consultants working in SNEP’s national office that would work on the entire program, no matter if financed by the World Bank or the IDB. This would include the project coordinator, procurement staff, a specialist in inginierie sociale and a sanitary engineer. It will also include these cost items for local staff in the four departments to be supported by the IDB. However, it would not support these costs for local staff in the Southern Department, making it necessary to fund them through government counterpart contributions or, if necessary, through the grant.

50 Annex 7: Financial Management According to Bank’s policy, a financial management capacity assessment (FM assessment) was performed for the Haiti Rural Water and Sanitation Project. The objective of the assessment is to determine whether the project’s implementing entity has acceptable financial management arrangements including the accounting system, reporting, auditing, and internal controls. The specific activities and circumstances of the project were also factored in to ensure that the entity’s financial management system is also commensurate with the needs and the unique nature of the project. The report’s findings are based on the conclusions of the capacity assessment for LICUS Trust Fund-financed Haiti Community-managed Rural Water and Sanitation project (given that the same implementing entity is involved), the results of the supervision of this project, as well as consideration of larger scope of the current project. Overall conclusions The overall conclusion of the FM assessment is that the project implementing entity, which is the Service National d’Eau Potable (SNEP), has adequate financial management capacity. Although there are significant country risks because of major weaknesses in the public financial management system and the high level of perceived corruption in Haiti, the impact of these risks on the project is likely to be limited given the defined implementation arrangements. These arrangements do not rely on the central government’s system. Instead, they build on a continuing entity that has established adequate financial management system, procedures and staffing, and successfully implemented a LICUS Trust Fund project. In addition, specific financial management procedures developed solely for the project with adequate level of controls will be followed. The project itself does not present any significant risk and its overall financial management risk is rated moderate. Risk Assessment and Mitigation The following matrix summarizes the financial management risk assessment for the project.

1 AssessmentRisk I H S M N Comments Inherent Risk Country X See text below on country risk. Entity X Project X Overall Inherent Risk X Control Risk 1. Implementing Entity X Adequate FM capacity and successful previous experience implementing WB -funded I operation.

51 - 2. Funds Flow X No major issue except for the proposed opening of a 30-day advance account at the provincial level. But amount of the advance is small (maximum $2,000) and adequate reporting and control procedures will be developed in the project operations manual. 3. Staffing Staff with adequate qualification and experience commensurate with the scope of the project already in place. 4. Accounting Policies Implementing entity already has a and Procedures manual specifying accounting policies and procedures. 5. Internal Audit No specific internal audit function but adequate controls provided for in the operations manual. 6. External Audit Audit will be required to be done by an independent private auditor acceDtable to the Bank. 7. Reporting and X Arrangements will be clearly Monitoring defined in the operations manual - to be approved by the Bank. 8. Information Systems X Arrangements to be clearly defined in the operations manual to be - approved by the Bank. Overall Control Risk X H - High S - Substant lerate N - Negligible or Low Country Risks The information collected through a number of recent studies (the 2006 Country Economic Memorandum and the 2006 Gestion de D4penses Publiques: Evaluation et Plan d’Action) and as part of the Interim Cooperation Framework exercise and the Economic Governance Reform Operations supported by the Bank provide a reasonable basis to assess the overall performance of the country’s public financial management (PFM). Prior to 2004, the government had at times operated without approved budgets or with budgets approved late into the fiscal year. Procedures for budget formulation and execution were weak and a significant share of public resources was channeled through multiple comptes courants3’ held by individual ministries and used non-transparently. In addition, internal controls were impaired by the lack of a well-structured accounting system or external audits of government budgets. Recent government measures to address these weaknesses include: (i)passage of a new

31 The comptes courunts (Ministries’ accounts) are meant to be used for unforeseen or non-budgeted needs such as for assistance to those affected by a natural disaster or unexpected travel by policy makers.

52 Organic Budget Law3* and adoption of a new budget classification and chart of accounts; (ii) timely approval of the budgets and regular public dissemination of key budget allocations and execution information; (iii)a drastic reduction of discretionary spending through ministerial comptes courants; and (iv) strengthening the external audit function with a decree on the organization and functioning of the supreme audit institution (Cour Supe'rieure des Comptes et du Contentieux Administratif-CSCCA). The CSCCA is in the process of catching up on government accounts audits which, together with the resumption of Parliament oversight functions, will strengthen external controls. These measures contribute to improving the overall performance of Haiti PFM that is, however, still affected by significant weaknesses in the budget process, as well as internal and external control. Also, Haiti has traditionally ranked very low in the Transparency International index indicating a high level of corruption perception. Without undermining the significance of the issues above, they will not affect directly the current project. The implementation arrangements of the project including procurement and financial management will not rely on the country's procedures. Specific procedures complying with Bank's policy will be implemented by a unit outside the central government ministries and with adequate staff and controls. Consequently, although the country fiduciary risk is high, its impact on the project is moderate. Project and Entity Risk Given the nature of the project activities and institutional set up, no major inherent fiduciary risk is identified. The implementing entity has adequate financial management capacity and the activities of the project are not complex. For this reason, both project and entity risks are rated moderate. Project activities The summary below provides a highlight of the key features affecting the scope and coverage of the financial management aspects of the project. Implementing entity The project will be implemented by the Service National d'Eau Potable (SNEP) through its Rural Water Unit and the Administration and Finance Department (AFD), respectively for the technical and financial management aspects. The structure of the AFD is as follows:

Admin. and Financial Department I Deputy Director (Admin) Deputy Director (Finance) I I Personnel Accounting -Sales

32 In addition to establishing the outlines of a new budgetary process, the Law mandates the creation of a new accounting system, creates the position of internal ex-ante controllers and a new internal auditing office.

53 The positions of Director Administration and Finance and Deputy Director Finance are currently vacant. The Deputy Director Administration is acting for these other two positions. The accounting unit consists of 5 staff, including the Chief Accountant, three accountants and an account clerk. Except for the chief accountant, the other accounting staff have not completed yet their university degree in accounting. For the purpose of the project, the Deputy Director Administration and one accountant will be responsible for the financial management. The team will be strengthened with the recruitment of a Finance Assistant financed under the IDB-funded project. This level of staffing is deemed adequate to handle SNEP’s as well as the WB and the IDB-funded projects. Based on its experience in implementing a LICUS TF grant and given the largely similar implementation arrangements between this grant and the current project, SNEP staffs are familiar and have a good understanding of World Bank financial management requirements. Budgeting The budget of the project will come mainly from the WB funding ($5.0 million). The Government is requested to provide counterpart funds estimated at $0.25 million over the life of the project (5 years). There will be no financing arrangements between the Bank and the government. The counterpart funds will be used to finance discrete expenditures. This will help alleviate the risk of unavailability of funds from the government on the project’s overall implementation. Accounting system SNEP maintains accounts for its own operations using a manual and accrual based accounting system. It has also developed an operations manual describing key financial management procedures. For the implementation of the LICUS Trust Fund-financed Haiti Community- managed Rural Water and Sanitation project, SNEP has also installed off-the-shelves commercial accounting software. The software has been used to produce the interim financial reports for the LICUS TF. It will also be used for the current project. It has all relevant features and is adequate for the project. SNEP will have to customize the account classification to be able to generate the interim financial reports according to the project activities. Internal Control and Internal Auditing There is no formal internal audit function at SNEP. However, adequate controls are defined in the operations manual. Funds Flow and Disbursement Arrangements The World Bank funds will be disbursed into a designated account denominated in US dollars opened at the borrower’s central bank (Banque de la RCpublique de Haiti). A local currency account may also be opened to finance local expenditures. The designated account will be opened and managed directly by SNEP. The authorized allocation is expected to be $300,000. Eligible expenditures paid from the Designated Account will be reported on the basis of statements of expenditures. In addition, the project can use direct payments. Given the implementation of the activities in the Southern Department and the need to finance expenditures at the local level, a 30-day advance account will be opened and managed by the Administrator of Les Cayes regional office. Since this account will be used mainly to finance minor operating expenses, the amount of the advance will be limited to $2,000. Detailed

54 Expenditure Category Amount in $ Works 2,875,000 Goods 495.000 I Consultant Services I 1.230.200 I I Operating cost I 226,000 Unallocated 423,800 Total Project Cost 5,250,000

Reporting Annual entity financial statements are regularly prepared albeit with significant delay. The review of the latest available financial statements shows that they are overall adequately prepared. SNEP will be required to provide the Bank with interim financial report, in form and substance satisfactory to the Bank. The format of the interim reports will be included in the operations manual. The interim financial reports will be integrated to the project progress reports and follow the same six-month periodicity. External Auditing SNEP financial statements are normally subject to audit by the supreme audit institution, the Cour Supe‘rieure des Comptes et du Contentieux Administratif (CSCCA), but this has not been done for several years now as CSCCA does not have the capacity to carry out its mandate properly. For this reason, the project audit will be done by a private sector audit firm acceptable to the Bank in accordance with International Standards on Auditing issued by the International Federation of Accountants. In addition to the project’s audit, SNEP will be required to submit its entity’s annual audited financial statements. The list of audit reports required is presented below.

55 Audit report Schedule Audit Report Periodicity Due Date Project specific financial Annual 4 months after the statements end of each SNEP’s fiscal vear. Special Opinions 4 months after the . Statement of end of each SNEP’s Expenditures fiscal year. . Designated Account Entity financial statements 4 months after the end of each SNEP’s fiscal year.

Action Responsibility Completion Date Audit SNEP No later than Appoint the auditors for the project six months after acceptable to the Bank according to TORS project also acceptable to the Bank. effectiveness. Accounting System SNEP As part of the Adapt the accounts classification to reflect selection the structure of the project activities. process of the NGOs. Counterpart funds MEF At the time of Commit adequate level of counterpart funds annual budget in the budget. discussions.

Legal covenants The following standard financial management has been included in the legal agreement: . Maintenance of an adequate financial management system throughout the life of the project. Submission of half-yearly interim financial reports as part of the project progress report no later than 30 days after the end of each semester. . Appointment of a project auditor acceptable to the Bank selected according terms of reference acceptable to the Bank. . Submission of the annual audit reports of the project’s financial statements and SNEP financial statements no later than 4 months after the end of each fiscal year of the borrower.

56 Supervision plan Supervision of the project financial management will be based on: H Desk review of the half-yearly interim financial reports and the annual audit report. On-site supervision: given the relative simplicity of the project and the assessment of the risk, at least one supervision mission should be done per year.

57 Annex 8: Procurement Arrangements HAITI: Rural Water and Sanitation Project

A. General

Procurement for the proposed project will be carried out in accordance with the World Bank's "Guidelines: Procurement under IBRD Loans and IDA Credits" dated May 2004; and "Guidelines: Selection and Employment of Consultants by World Bank Borrowers" dated May 2004, and the provisions stipulated in the Legal Agreement. The various items under different expenditure categories are described below. For each contract to be financed by the Loadcredit, the different procurement methods or consultant selection methods, the need for pre- qualification, estimated costs, prior review requirements, and time frame are agreed between the Recipient and the Bank in the Procurement Plan. The Procurement Plan will be updated at least annually or as required to reflect the actual project implementation needs and improvements in institutional capacity.

Procurement of Works: Works procured under this project will include: rehabilitation and construction of small rural water systems, drilling and installation of wells, installation of rain water harvesting systems and construction of public washrooms in schools and clinics. It is expected that the works for rehabilitation and construction of water supply systems will be packaged to create two contracts of roughly $1 million to $1.5 million each, to be procured using international competitive bidding (ICB) procedures. For these and the other works to be financed by the project, procurement will be carried out using the Bank's standard bidding documents (SBD) for international competitive bidding (ICB) and national SBD agreed with the Bank for national competitive bidding (NCB) and shopping procedures.

Procurement of Goods: Goods procured under this project will include: water meters (estimated value $120,000), latrine bases (estimated value $125,000) and small quantities of office equipment, computers, furniture, and vehicles. Procurement of these goods will be carried out using the Bank's SBD for all ICB and national SBD agreed with the Bank for other procurement methods.

Procurement of non-consulting services: As needed, small amounts of non-consulting services required for project execution, such as those needed for training events and project information campaigns, will be acquired in accordance with the Bank's Procurement Guidelines, as appropriate. This procurement will also be carried out using National SBD based on the Bank's SBD and agreed with the Bank.

Selection of Consultants: The project is expected to finance contracts with consulting firms for, inter alia: social outreach programs and mobilization of local community participation; engineering design and supervision of the rehabilitation / construction of rural water supply systems. In the case of the major contract for community outreach and mobilization services, competition will be open to NGOs active in the targeted regions, as well as to firms experienced in managing community-based development programs. In order to neutralize the advantages of any subsidized, low-cost NGOs, it is expected that the service provider for the community outreach and mobilization contract will be selected using the Quality-based selection method.

58 Short lists of consultants for services estimated to cost less than $100,000 equivalent per contract may be composed entirely of national consultants, in accordance with the provisions of paragraph 2.7 of the Consultant Guidelines.

Where teams of consultants are not required, individual consultants will be hired to provide specialized advisory services and services to support project implementation and monitoring, as well as technical assistance to SNEP and other agencies critical to project implementation. Universities, government research institutions, training institutions, NGOs and other specialized non-commercial organizations may be contracted to provide technical assistance and carry out research in their areas of specialization. Such service providers will be selected, to the extent possible, through competitive processes based on the Bank’s Consultant Guidelines.

Operating Costs: Sundry items, utilities, logistics costs for training, and other incremental recurrent costs may be procured using SNEP’ s administrative procedures which have been reviewed and found acceptable to the Bank. The procurement procedures and SBDs to be used for each procurement method, as well as model contracts for works and goods procured, are included in the project Operational Manual.

B. Assessment of the agency’s capacity to implement procurement

Procurement activities will be carried out by SNEP, which has assigned responsibility for project implementation to experienced staff who will perform the roles of project coordinator (head of the rural unit), procurement specialist and financial management specialist for the project. The SNEP staff member selected to become a full-time procurement specialist will be supported by the head of the rural unit, who was closely involved in procurement under the LICUS grant and worked directly with volunteers from Engineers Without Borders (EWB), the Canadian organization that helped to strengthen SNEP’s project management and procurement capacity during 2005-2006. As was the case under the LICUS grant, the Ministihe des Travaux Publics, Transports et Communications (MTPTC), which is responsible for water supply and sanitation in Haiti, will be involved in project procurement only at the final stages of the contracting process, based on SNEP’s recommendations for award.

The Bank’s assessment of SNEP’s capacity to implement procurement actions for the project has been based largely on SNEP’s satisfactory performance to date in implementing the LICUS grant, as well as in developing the project operational manual and procurement plan for the current project. An additional review of SNEP’ s organization and staffing for procurement conducted during project preparation confirmed that the minimum resources are in place to ensure that procurement is executed in accordance with Bank Guidelines. As part of the agreed action plan for strengthening SNEP’s capacity further, all SNEP staff involved in procurement will be trained by a local consultant to be hired under the project during the start-up period, with backstopping by the field-based IDB procurement staff and Bank procurement staff as needed. This assessment of SNEP’s capacity to implement procurement will be updated once the hands- on training is completed and project implementation is underway. If additional strengthening of SNEP’s organization and staffing is required at that time, an action plan will be prepared to address any remaining shortcomings.

59 The key issues and risks concerning procurement implementation for the project arise from the many weaknesses in the Haitian public procurement system. These weaknesses were identified by a CPAR prepared by the Bank in 1999 and confirmed by the Government and multiple donors in the context of the Interim Cooperation Framework (ICF) adopted in 2004. The Bank’s Economic Governance Reform Operation (EGRO 11) and Economic Governance Technical Assistance grant (EGTAG Iand 11) are providing support to the Government to address these problems through their procurement components and the Government has already taken several actions to improve procurement institutions and practices. These include the adoption of a Procurement Decree which creates the Commission Nationale des Marchb Publics (CNMP) with a mandate, inter alia, to improve procurement practices and develop standard bidding documents for use in all public procurement. The Decree also formally establishes that competitive bidding practices are the norm and direct contracting the exception for public procurement. In addition, the Decree’s coverage of government contracting practices has been expanded by including new provisions governing the procurement of intellectual (consulting) services.

In addition to the reforms supported by EGRO I1and the EGTAG grants, procurement risks will be mitigated in the case of the proposed project through the use of close and frequent supervision, including full prior review of all contracts financed by the grant. Further, as a result of the Government’s commitment to reform and the application of the Bank’s Guidelines in various recent B ank-financed operations, some government agencies such as PL-480, the Ministry of Finance and the Direction de la Protection Civile are becoming increasingly familiar with the basic principles of open and fair procurement, as SNEP did through its implementation of the LICUS grant for Rural Water Supply. This familiarity, combined with the planned broad dissemination of information on new procurement procedures to all purchasing agencies, is expected to mitigate some risks. Nonetheless, the overall project risk for procurement remains high.

C. Procurement Plan

The Recipient, at appraisal, developed a procurement plan for project implementation which provides the basis for the procurement methods. This plan was agreed between the Recipient and the Project Team on November 22, 2006 and is available at SNEP headquarters in Port-au- Prince. It will also be available in the project database and in the Bank’s external website. The Procurement Plan will be updated in agreement with the Project Team annually or as required to reflect the actual project implementation needs and improvements in institutional capacity.

D. Frequency of Procurement Supervision

Supervision will be carried out primarily through the prior review by the Bank of all procurement actions by SNEP. In addition, day-to-day procurement supervision will be supplemented by supervision missions at least twice a year.

E. Details of the Procurement Arrangements Involving International Competition

1. Goods, Works, and Non Consulting Services (a) List of contract packages to be procured following ICB and direct contracting (DC):

60 Ref Contract Estimated Procur P-Q Domestic Review Expected Comme . (Description) Cost (USD) ement Preference by Bank Bid- nts No. Metho (yesho) (Prior / Opening d Post) Date E2 Satellite $ 3,500 DC No No Prior Jan. 2007 photos E5 Water meters $ 120,000 ICB No No Prior Feb. 2009 E6 Latrine bases $ 125,000 ICB No No Prior Feb. 2009 T-1 Rehabilitation/ $1,260,000 ICB No No Prior Mar. 2008 Construction of rural water supply systems (SAEPS) - East T-2 Rehabilitation/ $ 900,000 ICB No No Prior Sept. 2008 Construction of rural water supply systems (SAEPS) - , West

(b) All ICB contracts and all direct contracting will be subject to prior review by the Bank. (Please refer to table below for prior review arrangements for other contracts.)

2. Consulting Services

List of consulting assignments with short-list of international firms. 1 Ref. No. Description of Estimated Selection Review Expected Comments Assignment cost Method by Bank Proposals (Prior/ Submission 1I Post) 1 Date I 1 Service $605,600 QBS or I Prior I Feb.07 Provider for Community outreach and mobilization

(a) All selection processes for consultancy services will be subject to prior review by the Bank.

(b) Short lists composed entirely of national consultants: Short lists of consultants for services estimated to cost less than $100,000 equivalent per contract may be composed entirely of national consultants in accordance with the provisions of paragraph 2.7 of the Consultant Guidelines.

61 F. Thresholds for Procurement Methods and Prior Review

Recommended thresholds for the use of the procurement methods specified in the grant agreement are identified in the table below. These thresholds, as well as the requirement for Bank prior review of all contracts, are common to all World Bank projects in Haiti. In the event that it is justified by a future assessment of the procurement capacity of the implementing agency (SNEP), thresholds for prior review may be introduced. The negotiated version of the plan provides for Bank prior review of 100 percent of contracts financed by the Grant.

Expenditure Contract Value Procurement Contracts Subject to Category (Threshold) Method Prior Review US $ thousands 1. Works > 1,000 ICB All 100- 1,000 NCB All loo ICB All 25-100 NCB All e25 Shopping /All Regardless of value Direct Contracting /All ~~~~______3. Consulting Services -3.A Firms Regardless of value QCB S ,QBS ,FBS ,LCS All

Note: ICB = International Competitive Bidding NCB = National Competitive Bidding QCBS = Quality- and Cost-Based Selection QBS = Quality-Based Selection FBS = Fixed Budget Selection LCS = Least-Cost Selection CQS = Selection Based on Consultants' Qualifications

62 Annex 9: Economic and Financial Analysis HAITI: Rural Water and Sanitation Project

Economic Analysis

I.Methodology

The present analysis is based on a household survey undertaken in 2005 in 84 randomly selected localities within 25 municipalities, in 4 departments covered by the program. Municipalities have been selected because they are shown to have lower aggregate levels of access to potable water according to the 2004 Poverty Map of Haiti. Within municipalities localities were selected randomly, including both villages with and without potable water systems. In each village, one or two households were interviewed about their daily water consumption and the distance to source. The number of households is very limited, so results have to be interpreted cautiously, in particular since estimates of the distance to source given by households in Haiti tend to be unreliable and distance to the source was not verified by the survey team.

Based on an estimate of the opportunity cost of time, the daily “price” of each bucket of 18.9 liter of water (called bokit in creole) and the daily “cost” of fetching water in terms of the opportunity cost of time have been estimated. The results are shown below.

Table 1. Average Household Costs and Consumution of Water in Rural Areas Daily Water Price of Water Distance to Source Daily Cost of Water Consumption Consumed (minutes) (gourdedfamil y) (bokitdfamily) (gourdedbokit) ARTIBONITE 27 15.65 3.57 4.62 GRAND ANSE 33 19.80 3.68 5.93 LOUEST 18 9.78 3.79 3.05 TOTALAVERAGE I 26 15.27 3.65 4.57

There are usually no other sources of water supply to the rural population (bottled water or distribution of water by truck are rare in rural areas of Haiti). The distribution of household by time spent fetching water is skewed to the left (see Figure 1 below), reflecting the fact that although most of the population spends less than 30 minutes fetching water, there is a long tail of households that do spend over half an hour and even over one hour.

It should be noted that the Southern Department and Nippes are not included in the data, because at the time when the data were collected for the economic analysis these departments were not included in the program. Conditions in the Southern Department, where the World Bank would support activities, are similar to those in the Grand’Anse, except that rainfall tends to be lower, which may entail even greater distances to sources. In any case, eligible communities will be prioritized based on lowest per capita costs and longest walking distance to source, which would weed out localities with potentially low rates of return.

63 Figure 1.

Distribution of HHs by Time Spent Fetching Water

30 al E 25 E c7? 20 aJ r, 15 .-C g lo I 55* 0 e1 0 10x20 20x30 30x40 40x50 50x60 60x70 >70 Minutes Spent Fetching Water

In terms of the consumption of water, this varies between 2 to 8 bokits per day. Figure 2 below shows the distribution of households per consumption of water. Most of the households consume 4 bokits daily or less (the equivalent of 75 liters or less). The correlation between the distance traveled to fetch water and the price (or cost) of water per bokit is very high (coefficient of 0.91). Furthermore, as Figure 3 below shows, almost 90 percent of the people carry the water by hand or on their heads. Only 10 percent use animals and 1 percent have private connections.

Figure 2.

Distribution of HHs by Consumption of Water

30 -0 g 25 1 s' 20 C 15 -I* '0

0 2 3 4 6 8 Consumption (bokites I day Ifamlly)

64 Figure 3.

MEANS OF CARRYING WATER

ByTube ByW By Animal 1% 7 4%

On Head 8 5%

111. Estimating a Rural Potable Water Demand Curve for Haiti

Given the water demand curve represented by equation (1) we calculated the price of water (P) based on the opportunity cost of fetching water. The opportunity cost of time was estimated as the daily legal agricultural wage (75 gourdes). Given that some approaches to valuing time have used less than 100 percent and others more than 100 percent3’, this value will be modified in a sensitivity analysis to assess how robust the results are in case the opportunity cost of time is lower given the prevalence of seasonal unemployment and underemployment in rural areas (see section VII).

P=PQ” (1)

where P is price of water and Q is quantity of water.

The price of water is not only a function of the time spent, but how many peopc per household went to fetch water, if they were adults or children, what volume was carried and how many trips were taken. It is estimated that it is often an adult (mother) with one or two children (sons or daughters) that are responsible for fetching water. Therefore, an estimate of 1.5 people (the opportunity cost of the child was estimated at 50 percent) was calculated for each trip to collect water. These assumptions are modified in the sensitivity analysis further below to test the robustness of the results.

The linear expression of equation (1) is shown in equation (2). We use equation (2) to regress quantity of water on price to obtain the coefficients p and a that allow us to estimate the demand curve. As can be seen in the scattered plot of Figure 4 below, the data already show a downward sloping demand curve, indicating the decreasing marginal value of water. This means that the first bokits per day per family are extremely valuable as they are necessary to sustain the

33 Whittington, Dale; Mu X. and Roche R. “Calculating the Value of Time Spent Collecting Water: Some Estimates from Uganda, Kenya.” World Development, Vol. 18, No.2. Pp. 269-280.

65 household needs. For each additional increment of water, the marginal value declines as the household puts the resource to less valuable uses.

In P = In p + a In Q (2)

Figure 4.

PRICE OF WATER CONSUMED (gddboquite)

30 00 - ; *

25.00 - ,I i *I * 3 20 00 - '0 0 1500- 0 1000- 5.00 - -. ; 000 -

The results show significance to the 95 percent confidence interval, obtaining results of p = 5.28 and a = -0.44. The estimated water demand curve takes the shape shown in Figure 5 below. To give an idea of the curve's shape, we calculated an increase in price of 1 Gds. at the base price of 2 and 5 Gds. would result in a decrease in water consumption of 62 percent and 34 percent respectively. This shows that as price of water increases, the response in terms of quantity of water consumed diminishes. We have then calculated the price elasticity (e,) of demand for a 25 percent increase in the price of water to be 1.6. This is highly elastic given the often inelastic (<1) price elasticities found in other studies.

66 Figure 5

WATER DEMAND CURVE

8 7 6 g5 t;4 E3 2 1 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5 7 7.5 8 8.5 9 9.5 10 BOKITES

In order to find out if there were substantial differences in the water demand curves of the different provinces (dkparternents), separate water demand curves were estimated. Figure 6 below superposes the various demand curves. The analysis shows that there are differences, but that these are not substantial, especially beyond the level of consumption of 2 bokits per family per day. Furthermore, due to the reduce sample size some of the coefficients of the water demand curves, especially in Grand Anse, loose statistical significance.

67 Figure 6

WATER DEMAND CURVES (by reglon)

-P (TOTAL) - P (ARTIBONITE) P (GRAND ANSE)

0.5 i 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5 7 7.5 a 8.5 9 9.5 io BOKITES

IV. Cost-Benefit Analysis of New Rural Potable Water Systems

In order to provide insight on the per capita investment levels for new rural potable water systems that would yield the minimum economic rates of returns, a cost-benefit analysis was undertaken based on the water demand curves estimated in part I11 above. The benefits from installing a rural potable water system can be divided between the benefits that arise from the incremental demand and those that arise from the non-incremental demand for water. The non- incremental benefits come mainly from cost savings on obtaining the current water consumption volume, while the incremental benefits result from the additional volumes consumed by the price drop.

On average, rural households consume approximately 3 bokits per day. It is estimated that the investment program under preparation by the SNEP will reduce the time for fetching water to almost zero, bringing the price down from POto PI, and will increase the daily volume available to the households at the water fountain from QO(3 bokits) to at least Q1 (10 bokits). Such increase in volume consumption and reduction in price will produce consumer surplus gains (areas A and B in Figure 7 below) by moving along the water demand curve from point “e” to point “f”.

68 Figure 7. Consumer Surplus Gains from Moving Along the Water Demand Curve

Price

b Qo QI Quantity

Area A is the consumer surplus gain produced by cost savings from the time spent fetching water (benefits from non-incremental demand). Such savings are only 19 percent of the total benefits to households. The larger portion of benefits (81 percent) come from the incremental demand for water, Area B. Table 2 below show the different consumer surplus gains by region and by type of benefit. Grand Anse shows the greater overall benefits, mainly due to the larger average distances traveled to fetch water (see Table 1). The survey also recorded the household willingness to pay for water, even though no details were provided to the interviewees about the future system to be constructed. This willingness to pay was an average of 0.74 gdslbokit. This translates into 7.4 gds for 10 bokits consumed under the new system. This is slightly above the aggregate benefits from cost savings, which is a conservative figure given the fact that the interviewees did not have information about the benefits from additional volumes provided by the investment.

Table 2. I Benefits (Consumer Surplus) - Gourdeshouseholdldas I Benefits from Cost Benefits from Additional Total Benefits Savings (Non- Consumption (Incremental (consumer surplus) incremental Demand) Demand) ~TIBONITE 1.66 I 19.46 I 21.12 I GRAND ANSE 6.52 19.55 26.07 L’OUEST 6.54 10.22 16.75 AGGREGATE 4.02 16.82 20.84

It is estimated, conservatively, that the investments in rural potable water systems would produce benefits for at least 10 years. The cost benefit analysis used a discount rate of 12 percent and an estimated operation and maintenance (O&M) costs of 5 percent of the total investment cost. The objective was to obtain the maximum level of investment per capita that would at least yield an internal rate of return (IRR) of 12 percent. Obtaining such threshold investment level would then allow the SNEP to identify those projects that are sure to have “automatic” economic

69 justification by looking at the level of investment per capita proposed to establish the water system. The results show that on an aggregate level for new rural potable water systems that have investment levels of less than $170 per capita, the IRR will be of 12 percent or higher. Such thresholds vary per region as shown in Table 3 below. Again, Grand Anse has higher investment threshold levels due to the relative longer distances for households to fetch water.

ARTIBONITE 170 GRAND ANSE 210 LOUEST 135 AGGREGATE 170

V. Other Considerations: Health Benefits and the Value of Water

Besides the factors considered above, health benefits and the opportunity costs of water itself also influence the economic analysis. However, compared to the costs and benefits assessed above, these factors are both difficult to estimate and are expected to be relatively modest for the reasons outlined below.

Health Impacts

Health benefits are expected from hygiene promotion and sanitation under the project. Based on international experience, these health benefits are expected to ,be higher than health benefits derived from improved water supply. However, because of the difficulties in measuring health benefits enumerated above, the costs of these activities and its benefits are not included in the present economic analysis.

Adverse health impacts could result if the increased volume of wastewater (grey water) is not managed properly, and if the sanitary conditions around water fountains are unacceptable. These factors have been carefully considered during project design. Since the proposed project by SNEP includes investment in sanitary blocks, facilities, and latrines, it is likely that the project health benefits will be even higher.

The Value of Water

In some rural localities water is scarce, especially during the dry season, and agriculture and human consumption compete for the resource. However, the quantities used for rural water supply are low compared to water use in agriculture and the value of water use in agriculture is low compared to the value of drinking water. For example, in the Artibonite valley the value of water in agriculture has been estimated at 15 gourdeslm3 or 0.06 gourdeslbokit. This is only 6 percent of what users are willing to pay for drinking water (1 gourdelbokit) at water kiosks.

70 VI. Rates of Return and Investment Parameters for New Rural Potable Water Systems

Given the thresholds levels calculated in part IV above it is recommended to maintain the aggregate threshold level of $170 per capita for the whole program (without dividing by d2partement) as the maximum level under which no economic analysis is needed. This recommendation is based on the fact that: (i) differences between regions may not represent the situation of specific villages within those regions; and (ii) calculated benefits may be underestimated since there are various benefits that have not been quantified (as mentioned in Part V above). Therefore, all proposed new rural potable water systems that show an investment cost of less than $170 per capita would be accepted without requiring additional economic feasibility analysis. For proposed systems with higher than $170 per capita investment costs, a specific economic analysis should be undertaken, especially if the population to be served is spending above 30 minutes to fetch water.

On the other hand, systems that present more than $480 per capita investment costs are unlikely to show economic net benefits from pure opportunity cost of time estimations. Unless there are tangible and quantifiable health benefits from the new potable water system or unless the potential beneficiary population face substantially higher opportunity costs from fetching water, then it is recommended that proposals showing costs above $480 per capita do not be considered. This “ceiling” threshold was calculated by extending the durability of the system from 10 years to 20 years, reducing O&M costs from 5 percent to 4 percent and assuming that incremental water consumption benefits would extend to up to 26 bokits per family per day (the equivalent of 100 liters). Figure 10 below shows the scale for analyzing the economic feasibility of projects presented to the rural potable water program of SNEP.

Figure 10. Scale for Analyzing the Economic Feasibility of Rural Potable Water Investments

Less 6 170 per $480 per More expensive capita capita expensive I 4 I b

-L Y >- No analysis required - economic analysis nquued Not accepable - automatically accepted Unless exception to undertake economic analysis

A sample of 40 rural villages without potable water systems were visited and cost estimates were presented for investments in construction of piped potable water systems with water fountains (no private connections). From Figure 11 below, we observe that 73 percent of the projects in the sample of 40 villages fall under the $170/capita threshold (in fact the majority, 55 percent, fall under $100/capita) and would be accepted without requiring economic analysis. We also observe that 20 percent of them would need to be evaluated further for estimating the net benefits from investments, and 8 percent of them would be excluded (unless already mentioned exceptions arise that justify further economic evaluation).

71 Figure 11.

RURAL POTABLE WATER INVESTMENTS BY PER CAPITA COSTS

.-! 15

g 10 -*0 5

0 400 1 OOcz170 17003W 3000480 +480 USUCAPITA

For investments that require further economic analysis, this can be done by: (i) quantifying time spent traveling to fetch water and inputting it in the estimated demand function of Part 111; or (ii) doing a survey of the potential beneficiaries and estimating their own water demand function to calculate the net benefits. Finally, taking the sample of 40 proposed projects and excluding those with per capita investments costs of above $480, Table 5 below shows the aggregate IRR by region. This shows that the aggregate economic IRR of the SNEP’s rural potable water program (as indicated by this sample of potential investments) could be expected to be around 60 percent.

Table 5. IRR on a Sample of Potential Rural Potable Water Projects by Region I Samde of Potential Rural Potable Water Proiects Aggregate

ercent)

VII. Sensitivity Analysis

The economic rates of return shown above for the Grande Anse and for the DCpartement de 1’Ouest are unusually high, mainly due to the relatively low per capita costs. As mentioned earlier, the assumptions concerning the opportunity cost of time and whether adults or children fetch water are debatable and may account for the high rates of return. For these reasons, the following sensitivity analyses were undertaken:

1. Reduce the opportunity cost of time by one third from 75 Gourdes to 50 Gourdes (about $1) per day to reflect underemployment and unemployment in rural areas. 2. Modify the assumption that one adult and one child undertake every trip to fetch water jointly, assuming that either an adult goes alone or two children go together. The

72 opportunity cost per trip would thus be reduced from 1.5 times the opportunity cost of an adult to 1 time, maintaining the assumption that the opportunity cost of a child is 50 percent of that of an adult.

The impacts on the economic rate of return for the overall project are as follows:

Reduction of the opportunity cost of labor to 50 gourdes per day: 28 percent Variation in the assumption of participants in each trip: 28 percent Combined impact of the two changes: 9 percent

The sensitivity analysis shows that the results are very sensitive to assumption on the opportunity cost of labor and the number of participants in each trip to the source. Although the average rate of return for all project analyzed remains high if only one sensitivity analysis is applied, the rate of return of individual projects may well be below the 12 percent threshold rate of return expected of each sub-project.

It is, therefore, suggested to carefully screen sub-projects on the basis of unit costs. It is proposed to (i)either eliminate sub-projects costing more than $100 per capita, or (ii)to subject each such sub-project to an economic analysis to determine if the higher costs are justified by associated benefits .

Financial Analysis

The financial analysis of piped water projects is based on two scenarios: (a) traditional system with stand posts and un-metered house connections; (b) a system with water kiosks and metered house connections, as recommended under the project. The latter scenario is the preferred mode of intervention under the program, since the traditional system has proven not to be very sustainable.

It is assumed that initially 80 percent of households will be served through either stand posts or water kiosks, and 20 percent through house connections. Over the lifetime of the system the share of users served by house connections is expected to increase, based on experience in existing systems. This observed trend is likely to improve the financial sustainability of the systems, since communities tend to set tariffs in such a way that users with house connections pay somewhat higher water bills than users buying at water kiosks.

Under “scenario A” revenues accrue from house connections only, using a fixed fee. Revenues are likely to be lower and somewhat more predictable than under “scenario B”. Current expenditures are for personnel for billing, collection, chlorination and repairs. Other expenditures are for small repairs (meter replacement, valve replacement, cleaning of reservoirs, pipe breaks). The cost of major repairs is not included in this financial analysis, although communities are strongly encouraged to build up reserves to be prepared to pay for major repairs.

Under “scenario B” revenues accrue from two sources: sales at water kiosks, and sales through house connections. Revenues and expenditures are expected to be higher than under the traditional system. Additional expenditures are for the staffing of water kiosks and for meter

73 reading. Quality of service is expected to be significantly better in “scenario B” over the long run, since the likelihood of sustainability is much higher. This will also increase the likelihood that tariffs will actually be paid and thus improve the amount of revenues actually collected.

Under scenario A, which is the predominant model of rural piped water supply throughout Haiti, collected revenues per family with a house connection are in the order of 50 gourdes per month (approx. $1.22), or often even less. Since only one out of five households is assumed to have a house connection, revenues per family are only 10 gourdes, well below estimated O&M costs of 25 gourdes per month and family (including 5 gourdes for disinfection, 10 gourdes for the salary of a plumber and 10 gourdes for maintenance).

Under scenario B, the proposed model, tariff levels and collection efficiency would be higher. Based on the data collected for the economic analysis, families served through water kiosks would buy 120 buckets per month, corresponding to 2.4m3/month or 16 litedcapitdday for a family of five. If they were charged a price of 0.75 gourdes per bucket, their monthly payment for water would be 90 gourdes. For the purposes of the financial analysis it is assumed that families served through house connections would use 1Om3/month at a price of 12 gourdes/m3, corresponding to 66 litedcapitdday and a monthly water bill of 120 gourdes. The average revenue per household would be 96 gourdes, under the assumptions about the share of water connections made above. O&M costs are estimated to be 70 gourdes per connection, including 55 gourdes for salaries, 5 gourdes for disinfection and 10 gourdes for routine maintenance. This would allow a surplus of 26 gourdes per connection, part of which would be passed on to the community (represented by the water committee) as a reserve for major repairs and replacement, and part of which would accrue to the operator for system expansion and profit. It should also be noted that for a locality with 500 households 10 jobs (kiosk attendants, plumbers, meter readers, treasurer and professional operator) are expected to be created, thus adding significantly to local employment.

74 Annex 10: Safeguard Policy Issues HAITI: Rural Water and Sanitation Project

Environmental Assessment

The development objective of the project is to provide water and sanitation services to rural communities on a sustainable basis. More specifically, the project will (i) provide water services in participating communities using innovative approaches in view of laying the foundations for project sustainability; (ii)improve sanitation practices in participating communities; and (iii) strengthen the capacity of SNEP, local government, local water committees, and professional operators.

The project, at the present design stage, includes two main components: 0 Capacity building for rural water supply and sanitation (RWSS), monitoring of the National Strategy for RWSS, impact evaluation and project management ($2 million). Water supply, basic sanitation and hygiene promotion ($8 million).

The project triggers three of the safeguard policies (OPBP 4.01 on Environmental Assessment, OPBP 4.04 on Natural Habitats and OPN 11.03 on Cultural Property). The processing of these three policies has been accomplished through the preparation of an Environmental Assessment report and its adoption by the Government of Haiti (GoH). While there were concerns that the project might involve land acquisition and thus trigger OPBP 4.12 on Involuntary Resettlement, assurance was provided before appraisal that land acquisition would not happen under the project. Thus, OPBP 4.12 on Involuntary Resettlement is not triggered by the project.

The project was categorized as Category B for Environmental Assessment (OPBP 4.01) as a precautionary measure to ensure that no negative environmental impacts would be produced, in particular from the second component. The Environmental Assessment of the project was conducted during project preparation and it produced the following findings and recommendations.

The bulk of project beneficiaries will be served through piped systems feeding water kiosks and standpipes, a common technology throughout concentrated rural areas (with about 500 inhabitants and up) and small towns in Haiti. A smaller part of beneficiaries would be in dispersed rural areas, which would be served by hand pumps or rainwater harvesting systems.

This project will identify participating communities based on objective criteria; community development using a demand-responsive approach allowing communities a series of choices: engineering design; system construction and supervision; and support during the early years of system management. To the extent feasible, the project will promote the innovative management approach piloted under the LICUS grant involving professional operators, metering, and the use of water kiosks instead of standpipes.

The project’s sanitation and hygiene promotion will be carried out in the same communities in which water services will be improved, because these activities can only be effectively carried out if water services are introduced or restored. Sanitation services will focus on the construction

75 of latrines using minimum standards, while using simple techniques in order to avoid concentration of benefits on a few beneficiaries.

The impacts that may be generated by the project without proper mitigating measures have been characterized as follows: Positive impacts: Increased volume of available potable water and improvement of the quality of drinking water. Improved access to water resources. 0 Decreased risk of water-borne diseases.

Negative impacts: Competition on water resources may generate scarcity or worsen resource distribution. Increased risk of nuisance and health issues from increased waste water production. Nuisance during construction and maintenance works. Pollution and accident risks during construction and maintenance works. Tension within the communities before, during and after the completion of the works.

Mitigating measures are included in the Environmental Management Plan (EMP). They basically apply to: The screening phase, with additional studies required to introduce environmental and social criteria. 0 The selection and design phase, with additional baseline and engineering studies. The implementation and supervision phase, with the construction of latrines, the restoration of sewage systems wherever required, additional water filtering and workers protection equipment, spring protection, and some degree of watershed management. The community involvement phase, with facilitation and capacity building activities.

The Environmental Assessment assigns responsibilities for the implementation of the EMP and for its monitoring. The implementing agency's institutional capacity for safeguard policies will be strengthened under the project. In particular, SNEP staff will be advised and trained in how to ensure contractors' compliance with safeguard policies related to the environment, natural habitats (capturing of sources that may be located in nature reserves) and cultural property (chance finds or archeological sites during construction).

76 Annex 11: Social Analysis HAITI: Rural Water and Sanitation Project

Impact on Vulnerable Populations The proposed project, by virtue of its geographic focus on rural villages and small towns, targets some of the poorest and most underserved of Haiti’s citizens. Haiti is more predominantly rural, and its rural population poorer, than the rest of Latin America and most of the developing world. The vast majority of Haiti’s population lives in extreme overty. This is especially true of the nearly two-thirds of the population who live in rural areas‘, 86 percent of whom live on less than $2 a day35 and 69 percent of whom live on less than $1 a day.36 Only 8 percent of rural households have access to water in their home or on their property.37 The World Bank Agricultural and Rural Development study estimates that 41 percent of the rural population spend a significant amount of time (or of their income) to fetch clean water every day. This has implications in terms of the present opportunity cost of time and increased agricultural prod~ctivity.~~

Specifically, the proposed project will have the greatest impact on women and children, who represent 49.5 percent and 45 percent of the rural population, re~pectively.~~Women and children are generally responsible for the often time-consuming task of fetching water. Furthermore, lack of access to safe water and sanitation has the greatest health impact on children, as they can be more susceptible to water borne illnesses. In a small study done in three villages in the south as part of the LICUS project, 38 percent of the 7000 families surveyed had had at least one member with diarrhea in the week preceding the survey. On the national level, Haiti’s average infant mortality rate of 80 deaths per 1,000 live births is far higher than the regional average of 27, and closer to sub-Saharan Africa’s average of 94.40

Women and children are also critical to the spread of good practices in hygiene and sanitation. Hygiene messages delivered through local schools in the proposed project will rely on children to share lessons learned with their families. Mothers who adopt these practices can have a strong impact on their families’ hygiene and sanitation habits, as they are generally in charge of cooking and cleaning. The experience of the Community Driven Development (CDD) pilot project in Haiti showed that women were involved in the community organizations and project development councils. Special efforts will be made to encourage involvement of women in the water committees short of mandating their participation.

34 SEDLAC database: http://www,depeco.econo,unlp.edu.ar/cedlas/sedlac/statistics.htm 35 SEDLAC database. 36 SEDLAC database. 37 SEDLAC database. 38 Agricultural and Rural Development, M 39 Enqutte Mortalitk, Morbidite‘ et Utilisation des Services, Hai‘ti 2000: httu://www. measuredhs.com/uub%df/FR121/02cha~itre02.odf (Children under 15 years of age.) World Population Datasheet 2005, Population Reference Bureau.

77 Rural Institutions and Leadership Rural areas suffer from institutional, economic and physical isolation. The World Bank’s Haiti Country Social Analysis (CSA) describes the institutional situation this way: “In the place of an overly complex formal structure, there is in reality an institutional void that perpetuates the absence of channels of political articulation on the part of the rural population in particular, and prevents the flow of information between local communities and central authorities (and international donors) that could enhance the effectiveness of development policies.” Economic and physical isolation are linked, given that the transaction costs of any productive activity are higher due to distance to markets, poor transport, lack of access to accurate information, and general weakness of infrastructure.

Local or central government plays a very limited role in the lives of villagers in rural Haiti. Historically, the basic administrative unit for both civilian and military administration was the section rurale, governed by a chefde section. While the chefde section post no longer exists officially, there is still a reliance on local leaders as de facto decision makers and authority figures. The smallest administrative unit is the section communale, which forms part of a commune, of which there are several in each of Haiti’s ten dipartements. The 1987 constitution included provisions for decentralization of administrative authority to the sections communales, but little progress has been made in reality. Local governance was designated to the administrative councils of the sections communales, or CASECs, but these councils vary greatly in their operational status. The commune is the only administrative body apart from the central government that is empowered to levy taxes, but few taxes are collected, leaving communes with very limited resources.

At the community level, the absence of formal institutions means that there is little sense of connection to the political and economic center of the country, Port-au-Prince. Land ownershi is largely informal and only four percent of rural households have access to a savings account.R Most use livestock as a form of saving, to sell in a time of great need. Seventy-eight percent of rural households own some form of livestock.

The project aims to increase the capacity of SNEP at the central, departmental and local level. A central rural unit has been established under the LICUS pilot in Les Cayes, and departmental rural units are planned under this program. This structure is designed to contribute to creating a flow of information from localities to SNEP’s rural units, and ultimately to the central bureau in Port-au-Prince. At the national level, the project also includes a corollary proposal to hire a consultant to increase SNEP’s capacity to coordinate the work of NGOs and other donors working in the water sector and implement the national strategy on rural water supply and sanitation.

At the local level, improved communication and increased presence of SNEP should contribute to filling the void between the rural population and government. The project aims to empower communities to articulate their needs, a process that is dependent upon decentralization to the extent that communities have access to SNEP and its representatives and have some sense that their input will be considered. SNEP and its planned experts on community

41 (( Gouvernance rurale et institutions locales en HaTti: Contraintes et opportunitks pour le dkveloppement, )) Nathalie Brisson Lamaute, Gilles Damais, and Willy Egset, 15 Avril 2005, p. 24.

78 outreach will actively monitor project sites. The NGO responsible for community outreach will be required to make a certain number of field visits and provide regular written reports to SNEP.

The attainment of any further administrative goals should not affect the success of the project in the communities it aims to serve. If passed under the new government, a draft water and sanitation framework law (Zoi cadre) would address the lack of decentralization. The law includes provisions to gradually transfer the responsibility for water and sanitation to municipalities, which in turn could delegate services to the private sector or water committees.

The institutional void at the local level also translates into an absence of police or any official body to protect the population and provide security. Although much attention has been paid to the political instability and violence creating Haiti’s current “environment of conflict,” rural areas are much less affected by this. The majority of violence occurs in urban areas, and primarily in the slums of Port-au-Prince. According to the 2001 household survey, most rural dwellers feel safe. “While a significant minority in rural areas also expresses fear of visiting markets and other towns, the figures demonstrate that for three-quarters of the rural population fear is not a major, daily

Community Selection The project aims to identify, through a two-stage process, communities with the need and motivation for improved access to and use of water and sanitation services. The identification and feasibility stages are designed to gauge communities’ needs and demands in order to select those who not only suffer from lack of access to water and sanitation, but also are mobilized to address this need along the parameters of the project. SNEP and project staff responsible for community development will be the primary level of contact with the communities for the project. Community animators of SNEP and the service provider contracted by SNEP to work on community development will work with the inhabitants of each locality to identify needs, strengthen local water committees, and enhance local capacity for the operation and maintenance of water and sanitation systems, with the option to hire a private operator.

The project will, for the first time in the history of the Haitian water sector, select participating rural communities in a systematic way based on objective criteria. It will carry out a systematic assessment of needs and coverage levels for water supply and sanitation in the intervention region, linked to a transparent system of selecting participating communities on the basis of objective criteria. These criteria will include estimates of benefits, costs, and community commitment. In addition, the project will ensure that selection of participating communities is also perceived by key local stakeholders as being objective. For that purpose, a regional advisory group will be set up consisting of civil society representatives and mayors that would review the selection of communities and advise the implementing agency in its final choice.

The project’s emphasis on community mobilization as a prerequisite for participation could result in the selection of better organized communities to participate while those with marginally greater needs are not selected. This risk of not serving communities with the absolute greatest need is difficult to avoid given the lack of data from which to select communities and the value the project places on community participation as a contributor to

42 World Bank (2006). State Failure and Social Resilience in Haiti: A Country Social Analysis, p. 40.

79 sustainability. While many NGOs and donors have built water systems in rural Haiti over the years, communities as a whole have been unable to operate and maintain the systems in a sustainable fashion. Innovative approaches are required to break the harmful cycle of systems falling into disrepair and communities waiting for another system to be built in its place.

A transparent process will help manage expectations. Many communities, through Haiti’s informal communications network (or teledjol), have already heard of an upcoming project and submitted requests to SNEP to participate. Such self-identification, while not the primary basis for selecting communities, can demonstrate the mobilization and motivation of a community. The feasibility stage of the project is designed to more thoroughly assess the communities selected in the initial identification stage. The firm specializing in inginierie sociale will be responsible for communicating the criteria used to select communities to ensure a clear understanding of the process.

The project will rely heavily on community outreach to create realistic expectations for what can be accomplished. The LICUS pilot project has shown that participating communities expect project work to proceed at a much faster pace than is realistically feasible. The rural population’s low access to news media makes it difficult to implement a communications strategy to ensure mass dissemination of clear messages about project timeframes, as well as community selection and SNEP’s role in rural water supply. According to the 2001 household survey, only three to four percent of the rural population had read a newspaper or watched TV in the week preceding the interview. The most important media by far is radio, to which one-third of the respondents had listened in the preceding week.43 The teledjol, while effective in spreading news quickly, cannot be depended upon for accuracy or reliability.

Once the community is selected, animators will work with communities to ensure they have the needed tools to make decisions. For instance, communities should establish rational criteria to choose professional operators. The LICUS experience showed that communities chose a trusted community member to serve as the professional operator, instead of selecting someone with appropriate technical skills. To address this weakness, communities should receive guidance from the community animators in the qualifications needed in a professional operator and the specific tasks that will be required of the operators. Furthermore, training of professional operators should be offered.

Participatory Approach in the Haitian Context Haiti has a tradition of deference to individual leaders, but collective efforts on a smaller scale do function. The World Bank Agriculture and Rural Development study determined that membership in an organization reduces the probability of being poor. Examples include collective work groups (konbits); micro-credit initiatives managed by a small group; farming “by half,” meaning that the owner of a garden rents half of his terrain to a farmer who farms the entire property and gives half the harvest to the owner; the quarry owner in the town of Pemerle in the Southern Department (where the LICUS grant has intervened) who formed a committee of five people to collect a tax from the community; and water committees that manage small-scale water systems (hand pumps) as a group.

43 (( Gouvemance rurale et institutions locales en Hayti: Contraintes et opportunitks pour le dkveloppement, N Nathalie Brisson Lamaute, Gilles Damais, and Willy Egset, 15 Avril 2005, p. 24.

80 The proposed project will take advantage of the experience with village leadership dynamics and participatory techniques gained by the existing LICUS pilot water supply project (see above under section 2, Rationale for the project). Two among the many lessons of the project are: i) village structure in Haiti is hierarchical, with many community members deferring to acknowledged local leaders to make decisions on their behalf; and ii)interventions following the ingknierie sociale approach need to be more intensive and need more skills and resources than those of SNEP’s single community animator in the region. These two factors combined made it difficult for the LICUS pilot project to be truly participatory, in the sense that it is not known if the views of marginalized or poorer village inhabitants were fully taken into account. SNEP’s community animator dealt primarily with local leaders and relied too much on imparting information and not enough on seeking the input of the leaders and others to inform the project’s response. However, given time and resource constraints, a different approach would have faced serious difficulties.

The social sustainability of the project relies in large part on the service provider that would provide the functions of ingbnierie sociale under the project. The proposed project requires a strong emphasis on skilled community animators to avoid marginalizing community members and sacrificing sustainability. However, the challenge of this should be recognized in a country with limited resourceskapacity in this area, and a cultural tradition of decision making by a small group of leaders.

Communication is key to an effective participatory approach. The community animators face a communications challenge in a number of regards: logistics of arranging meetings, keeping records of what was said/agreed upon, eliciting the opinions of the population without leading or raising expectations, and working against a traditional top-down decision making approach. The LICUS project underscored the importance of keeping written records of meetings with communities, as this is not in the local tradition and can prove useful when accounts of what was agreed upon differ. Project staff must also make sure that communities receive technical designs in time before launching bids for works at a formal meeting where they are asked for comments, and that the result is ratified by a vote that is documented in written minutes, even if that takes more time. Trade-offs are an effective way of presenting the choices to be made by the communities, such as the pros and cons of more kiosks vs. more house connections. It also must be made clear that budgets cannot be amended once the works contract is launched. As a lesson from earlier experience, the respective responsibilities of communities, SNEP and its service provider will be included in a grant agreement to be signed between SNEP and the local community represented by its water committee at the feasibility stage of the project.

81 Annex 12: Project Preparation and Supervision HAITI: Rural Water and Sanitation Project

Planned Actual PCN review 12110/2005 12/08/2005 Initial PID to PIC 12/20/2005 1212 1/2005 Initial ISDS to PIC 12/20/2005 1212 1l2005 Appraisal 0911512006 9/14/2006 Negotiations 11/20/2006 11/20/2006 BoardRVP approval 1/25/2007 Planned date of effectiveness 311 512007 Planned date of mid-term review 12/08/2010 Planned closing date 1213 1/20 11

Key institutions responsible for preparation of the project:

The project was prepared by the Service National de 1’Eau Potable (SNEP) #1, Les Residences du Soleil, Rue Georges Sylvain, Delmas 29, Port-au-Prince, Haiti with the assistance of international and local consultants and the technical guidance provided by the Bank task team.

Bank staff and consultants who worked on the project included:

Name Title Unit L Manuel Schiffler Task Team Leader and Senior Economist LCSFW Peter Kolsky Sr.Water Supply and Sanitation Specialist EWDWS Ulrich Myboto Consultant - Operations Analyst LCSER Katherine Shafer Coleman Consultant - Social assessment LCSFW Alexandre Brailowsky Consultant - Policy reforms LCSFW Jean-Roger Mercier Lead Environmental Specialist ESDQC Patricia Macgowan Sr. Procurement Specialist LCSPT Ahmadou Moustapha Ndiaye Lead Financial Management Specialist LCSFM Jennifer Sara Peer Reviewer LCSFP Mukami Kariuki Peer Reviewer EWDWS Patricia Acevedo Program Assistant LCSFW Yao Wottor Procurement Specialist LCSPT Fily Sissoko Senior Financial Management Specialist LCSFM

Bank funds expended to date on project preparation: 1. Bank resources: $229,365 2. Trust funds: 3. Total: $229,365

Estimated Approval and Supervision costs: 1. Remaining costs to approval: $62,178 2. Estimated annual supervision cost: $110,000

82 Annex 13: Documents in the Project File HAITI: Rural Water and Sanitation Project

Cadre De Cooptration Inttrimaire (CCI), Groupe Thtmatique : Eau Potable Et Assainissement. Rapport Final, Mai 2004. (Final Report of ICF Thematic Group on Water and Sanitation.)

“Compte-Rendu d’ Atelier National pour l’tlaboration d’une strattgie eau potable et assainissement en milieu rural, 10 et 11 novembre 2005. ” (Minutes of a November workshop on the National Strategy for Rural Water Supply and Sanitation)

“Projet Don LICUS Rapport d’Etape No. 2,” (LICUS grant stage 2 progress report), June 2005- March 2006.

T. Niyungeko, “Strattgie Nationale et Dtveloppement du Secteur Eau Potable et Assainissement en Milieu Rural. Partie 1 : Diagnostic et Analyse du secteur. ” Dtcembre 2005.

“Water Resources Assessment of Haiti,” US Army Corps of Engineers, August 1999.

Enqutte Sur Les Conditions de Vie en Hui’ti, Institut Haytien de Statistique et d’Informatique, 200 1.

Enqu2te Mortalitk, Morbidite‘ et Utilisation des Services, Haiti 2000. (Haiti DHS survey.)

Government of Haiti, “Interim Cooperative Framework Summary Report”, July 2004.

Nathalie Brisson Lamaute, Gilles Damais and Willy Egset, “Gouvernance rurale et institutions locales en Haiti: Contraintes et opportunitts pour le dtveloppement, ” 15 Avril2005.

PAHONHO Haiti Water and Sanitation Monitoring System for Haiti.

PAHO, “Infant mortality in Haiti”, 2006.

83 Annex 14: Statement of Loans and Credits IDA Portfolio in Haiti As of 1211112006

Difference between expected and actual Onginal Amount m $ Millions disbursements

Project ID FY PUIPOSe IBRD IDA SF GEF Cancel. Undisb. Orig. Frm. Rev’d PO98531 2007 HT Electncity Project 0.00 6.00 0.00 0.00 0.00 6.08 0.00 0.00 PO93640 2006 HTCDDProject 0.00 38.00 0.00 0.00 0.00 33.24 6.23 0.00 PO95371 2006 HT Economic Governance TAG I1 0.00 2.00 0.00 0.00 0.00 2.06 0.00 0.00 PO95523 2006 HT Transport and Terntorial Development 0.00 16.00 0.00 0.00 0.00 16.46 0.00 0.00 PO90159 2005 HT Emergency Recovery & Disaster 0.00 12 00 0.00 0.00 0.00 9.46 4.61 1.54 Management PO93936 2005 HT Governance Technical Assistance Grant 0.00 2.00 0.00 0.00 0.00 1.51 1.46 0.00 Total, 0 00 76.00 0.00 0.00 0.00 68.81

Statement of IFC’s Held and Disbursed Haiti Portfolio As of 1013112006 In Millions of US Dollars

Committed Disbursed IFC IFC FY Approval Company Loan Equity Quasi Partic. Loan Equity Quasi Partic 2006 Digicel Haiti 30.00 0.00 0.00 0.00 30.00 0.00 0.00 0.00 1998 Microcredit 0.00 0.27 0.00 0.00 0.00 0.27 0.00 0.00 Total portfolio: 30.00 0.27 0.00 0.00 30.00 0.27 0.00 0.00

Approvals Pending Commitment FY Approval Company Loan Equity Quasi Partic.

Total pending commitment: 0.00 0.00 0.00 0.00

84 Annex 15: Country at a Glance HAITI: Haiti Rural Water and Sanitation Project

Latin Key Development Indicators Amenca LOW Age dlatrlbutlon, 2005 Haiti 8 Carib. income (2005) Female

Population, mid-year (millions) 8.8 551 2,353 Surface area (thousand sq. km) 28 20,418 29,265 Population growth (%) 1.4 1.3 1.8 Urban poplation (%of total population) 39 78 31

GNI (Atlas method, US$ billions) 3.9 2,210 1,364 GNI per capita (Atlas method, US$) 450 4.008 580 GNI per capita (PPP, international $) 1,840 8.111 2,486

GDP growth (%) 1.8 4.4 7.5 GDP per capita growth (“A) -0.2 3.1 5.6

(moat recent eat/mate, 2000-2M)

Poveriy headcount ratio at $1 a day (PPP, %) 54 9 Under4 morlallty rate (per 1,WO) Poveriy headcount ratio at $2 a day (PPP, %) 78 23 1 Ufe expectancy at birth (years) 52 72 59 Infant mrtallty (per 1,000 live births) 76 27 80 Child malnutrition (“h of children under 5) 17 7 39

Adult literacy, male (“h of ages 15 and older) 91 73 Adult literacy,female (%of ages 15 and older) 90 50 Gross pdmary enrollment, male (“6 of age group) 121 110 Gross pnmary enrollment, female (%of age group) 117 99

Access to an improved water source (%of population) 54 91 75 Access to improved sanitation facilities (% of population) 30 77 38 I OHanl QLatin Amenca &the Caribbean I

Net Aid Flows I980 19gO 2000 2005 a

(US$ millions) Net ODA and official aid 105 168 208 243 IGrowth of GDP and GDP per capita (Oh) Top 3 donors (in 2004): United States 35 50 91 91 10 I Canada 5 10 20 37 I France 7 32 11 25 0 Aid (%of GNI) 7.3 5.9 5.6 6.3 Aid per capita (US$) 19 25 26 29 .IO

Lonp-Term Economic Trends .m 90 95 03 os Consumer prices (annual % change) 17.8 21.3 11.5 14.8 GDP implicit deflator (annual %change) 21.4 14.1 11.2 +GDP -GDPpercapila

Exchange rate (annual average, local per US$) 5.0 5.0 21.5 40.7 Terms of trade index (2000 = 100) 225 132 100 87 1980-90 1990-2000 20W-05 (average annual growrh %) Population, nud-year (mllllons) 5.5 6.9 7.9 8.8 2.3 1.4 1.4 GDP (US$ mlilions) 1,462 2,864 3,716 4,3m -0.2 -1.5 -0.5 (%of GDP) Agriculture 33.0 27.8 25.6 -0.1 -7.4 -1.2 Industry 22.0 16.2 16.0 -1.7 3.2 0.5 Manufacturing 16.0 8.8 8.3 -1.7 -8.4 -2.2 Sewices 43.0 64.1 51.1 0.5 -0.3 -0.4 Household flnal consumption expenditure 81.9 81.4 86.0 93.3 0.9 General gov’t Rnal consumption expenditure 10.1 8.0 7.6 6.3 -4.4 Gross capital formation 16.9 13.0 26.6 23.0 -0.6 8.6 1.5

Exports of goods and sewices 21.6 17.5 12.3 14.0 1.2 -10.3 7.7 Imports of goods and services 30.5 20.0 32.5 41.0 2.3 -5.2 10.6 Gross savings 6.4 21.8 23.5

Note: Figures in Italics are for years other than those specified. 2005 data are preliminary estimates. .. indicates data are not available. a. Aid data are for 2004.

85 Haiti

Balance of Payments and Trade 2000 2005 Governance IndlCStOts, 2000 and 2004 (US$ millions) Total merchandise exports (f.0.b) 331 458.9 Total merchandise imports (f.0.b) 1,086.7 1,308.5 Voice and amuntaMllty Net trade in goods and services -852.1 -1162.7 Polibcal mabillty Workers' reminances and compensation of employees (net) 578 925 ReQUlatOtYquality

Current account balance -260 -270 Rule 01 law as a % of GDP -6.6 -6.3 Control 01 wmpbon

Net foreign assets (Central Bank, US$) millions 172.3 191.2 0 25 YI I5 loa

2w4 Gauntr(8 percenhle rank (C-tW) Central Government Finance np., "al"e4 Unpr DMOI RmDI (% of GDP) Revenue 8.4 9.6 Scum: Kauimann-Kraay-Maszi, Wcdd Bank Tax revenue 7.7 6.3 Expense 7.4 13.8 Technology and Infrastructure 2000 2004 Cash surpius/deliat -2.2 -0.7 Paved roads (%of total) 24.3 Highest marginal tax rate (96) Fixed line and mobile phone Individual subscribers (per 1,000 people) 16 e4 Corporate High technology exports (36 of manufactured exports) 3.5 External Debt and Resource Flows Environment (US$ millions) Total debt outstanding and disbursed 1,169 1,336.0 Agricultural land (% of land area) 58 58 Total debt service 44 51.0 Forest area (% of land area, 2000 and 2005) 4.0 3.8 HlPC and MDRi debt relief (expected; flow) - Nationally protected areas (%of land area) ,. 0.4

Total debt (% of GDP) 31.5 31.0 Freshwater resources per capita (a.meters) .. 1,548 Total debt service (% of exports) 4.0 8.5 Freshwater withdrawal (%of internal resources) .. 7.6

Foreign direcl investment (net inflows) 8.0 9.5 CO2 emissions per capita (mt) 0.18 0.22 PorHoiio equity (net inflows) GDP per unit of energy use (2000 PPP $ per kg of dl equivalent) 7.0 6.3

Energy use per capita (kg of dl equivalent) 257 270

(US$ millions)

IBRD Total debt outstanding and disbursed 0 0 Disbursements 0 0 Prinapal repayments 0 0 interest payments 0 0

IDA Total debt outstanding and disbursed 480 507 Disbursements 8 9 Private Sector Development 2000 2005 Total debt servim 10 50 Time required to start a business (days) - 203 iFC (fiscal year) Cost to start a business (% 01 GNI per capita) - 153.1 Total disbursed and outstanding portfolio 0 0 Time required to register property (days) - 683 of which IFC own acwunt 0 0 Disbursementsfor IFC own account 0 0 Ranked as a major constraint to business Portfolio sales, prepayments and (% of managers surveyed who agreed) repayments for IFC own account 0 0 n.a. n.a. MlGA Gross exposure Stock market capitalization (% 01 GDP) Bank branches (per 100.000 people)

Note: Figures in italics are lor years other than those speafied. 2005 data are preliminary estimates. 9/19/06 .. indicates data are not available. -indicates ObSeNatiCm is not applicable.

Development Ewnomia. Development Data Group (DECDG).

86 MAP SECTION

such boundaries. any endorsement or acceptance of ntelglsau faytrioy or status of any territory, on the legal The World Bank Group, any judgment this map do not imply, on the part of n n te nomto hw on other information shown and any h onais oos denominations The boundaries, colors, Map Design Unit of The World Bank. This map was produced by the 18¡30’ 19¡00’ 19¡30’ d’Hainault RURAL WATER AND SANITATION PROJECT Dame-Marie Les Irois 43’74¡00’ 74¡30’ 74¡30’ Anse U.S.A. Tiburon CUBA Chambellan JAMAICA Abricots La Cahouane

Moron

GRANDE- INTERNATIONAL BOUNDARY DEPARTMENT BOUNDARIES COMMUNE BOUNDARIES NATIONAL CAPITAL DEPARTMENT HEADQUARTERS COMMUNE HEADQUARTERS SELECTED TOWNS RIVERS DIRT ROADS GRAVEL ROADS ASPHALT ROADS AREA OF INTERVENTION SUPPORTED BY THE IDB LICUS LOCALITIES AREA OF INTERVENTION SUPPORTED BY THE WB AREAS OF INTERVENTION: Bass Droit

Caribbean Sea Bonbon

Chardonnieres E Gebeau

HAITI B

Chaude Source A

H Cape Rose

Anglais Grande M

Les A ot -Piment Port- S Lon Marche

V Anse o

oh-Bateau Roche- ANSE ld r

o DOMINICAN teaux C gu e REPUBLIC JEREMIE

Roseaux

ATLANTIC OCEAN Point Gravois 74¡00’ Roseaux Port-Salut

HAITI Beaumont Rico Puerto Camp-Perrin

Charles Carrefour

(US) SUD Chantal Arniquet 01020304050 Corail

Marceline Cayemites

Islands s

Crrf. Joute Crrf. e

Torbeck n t

l du Sud St. Jean e

Pestel s Cavaillon Maniche LES CAYES Baraderes Baraderes KILOMETERS Carrefour

Cavaillon Sudre Vache Island 20¡00’ Ile a Vache

Plaisance NIPPES St. Louis du Sud Morisseau de Nippes Petite Trou LE ST. NICOLAS M Occo Morne 73¡30’ 33’7¡0 23’72¡00’ 72¡30’ 73¡00’ 73¡30’ Bombardopolis

L’Asile Grande Zanglais Aquin d

Aquin O VE GON

Anse-a- ISLAND Veau

Nippes Gon ve Channel ve Gon Vieux Bourg d’Aquin Petite Riviere

de Nippes NORD-OUEST Baie de Henne Jean-Rabel

Price

Pemele Hennes

Pointe a Raquette

Saint Marc Channel Marc Saint

Bay

Point Jean-Rabel

i n

CARIBBEAN SEA NE MIRAGO

s Des Ruisseaux Crrf.

d e

F e

r Anse-Rouge iaone Mirago otMrg ne Pont Mirago

73¡00’ Lake tes-de Fer C PORT-DE-PAIX

M Island C ou ol st

Grande Saline om iq

B u Point Lapiere b e

Bassin-Bleu

a ie

r i

n ve Go Petite

e Tortue

t Tortue Channel Tortue

Terre-Neuve Anse a Galet

Blockauss e OAVES GONA

Bainet

Montrouis s

St. Marc r ove Go Grand

o Chansolme

Gras Morne sR

i v

Desdunes e r

e St. Louis Logne s Ile de la Tortue La Valle

Etienne

ARTIBONITE C

Trouin Quinte o L’Estere

Sond St.

u l

Pont

n Poteaux Les

Carrefour Dufort Bayonnais e -Foleur Anse- Arcahaie

Gr a Momance

E Pilate

St. Antoine n

d s

e t Port-Margot

72¡30’

J r

Cape Jacmel a Verrettes Gressier e c PORT-AU-PRINCE

G Le Borgne JACMEL m Sentard

o Dessalines

s e Port-Au-Prince

s l Cabare

e Carrefour SUD - EST l D li o n n Limb

e Ennery

y Plaisance

ATLANTIC OCEAN Bay e Croix des Missions n de l’Artibonite Pte. Riv. de L’Attalaye La Chapelle

St. Michel

Cayes-Jacmel rc NORD L

i A e m b lle e

R Marmelade Sta. Terrienne

Hasco T I

B Bas Limb

Bon Repos O

N Acul du

I Nord

T Acul Bay

Ville Bonheur E

e St. Raphael

F Dondon

Saut d’Eau o l Marigot n e Delmas t Kenscoff Ville Petion- G

r Milot a Shada Plaine Nord

s Fonds Jean Noel OUEST

s du

G ssade Ma

Boucan Carre

s Crois des Bouquets

e B

u TIEN CAP-HA

Bahan c

C Pignon B

Thomazeau n Quartier Morin Quartier a

o du Nord Grd. Riv.

C u n Trou Du-Nord

Caimen Lake

y o

a Barrage

Trianon a

t r CENTRE r

r Limonade

e t

T B

l Ganthier

Crrf. Mace Crrf. o Belle-Anse o

c HINCHE

Mirebalais t

Peligre Dam Ranquitte o

e Thomonde Ste. Suzanne

e n

Saumatre Lake Saumatre

- n

Grand Gosier a Parisien

72¡00’ e

- C La Victoire

Coracol Fond

h e S Vallieres

arna Perches

v na a Rouge Terrier Crochu Monbin

Casse Cerca Carvajel Bodane l Cornillon

Verrettes Lascahobas NORD Fonds

FORT LIBERT

EST

Peligre Lake M Thiotte

Anse a Pitres a

Roysec r

io Mancenille Bay Delegram n

Canice

- To Barahana Thomassique Ouanaminthe Croix Fer

A Savanette r t

Baptiste i b Cerca-la-Source o Capotille

n Organis Mont it

eldre Bellad e

NOVEMBER 2006

To Santiago

IBRD 35101 DOMINICAN REPUBLIC 18¡30’ DOMINICAN 19¡00’ 19¡30’ 20¡00’