Document of The World Bank Report No: ICR0000965 IMPLEMENTATION COMPLETION AND RESULTS REPORT (IBRD-70770) ON A LOAN IN THE AMOUNT OF US$ 35.69 MILLION TO THE REPUBLIC OF COLOMBIA FOR A WATER SECTOR REFORM ASSISTANCE PROJECT April 28, 2011 Sustainable Development Department Colombia and Mexico Country Unit Latin America and Caribbean Regional Office CURRENCY EQUIVALENTS (Exchange Rate Effective April 19, 2011) Currency Unit = Colombian pesos 1.00 = US$ 0.00054 US$ 1.00 = Col$ 1,848 FISCAL YEAR January 1-December 31 ABBREVIATIONS AND ACRONYMS CAS Country Assistance Strategy CDF Country Development Framework CPS Country Partnership Strategy CONPES Comisión Nacional de Política Economica y Social (National Comission for Economic and Social Policy) CRA Comisión Reguladora de Agua (Water Regulating Committee) DAPS Dirección de Agua Potable y Saneamiento (Water and Sanitation Directorate) DNP Departamento Nacional de Planeación (National Planning Department) FINDETER Financiera de Desarrollo Territorial (a second tier financial intermediary for regional and local infrastructure investment established in 1990) FM Financial Management FNG Fondo Nacional de Garantias (National Guarantees Fund) IBRD International Bank for Reconstruction and Development INSFOPAL Municipal Development Institute NPV Net Present Value PAD Project Appraisal Document PDA Planes Departmentales de Agua (Departmental Water Plans) PDO Project Development Objectives PIU Project Implementation Unit PME Programa de Modernización Empresarial (Corporate Modernization Program) POI Programa de Obras e Inversion (Program of Works and Investments) PSP Private Sector Participation RPA Regional Procurement Adviser SEA Sectoral Environmental Assessment SEE Strategic Environmental Evaluation SIAS Sistema de Informacion Agua y Saneamiento (Sector Information System) SSPD Superintendencia de Servicios Publicos Domiciliarios (Superintendency of Domestic Public Services) WSS Water Supply and Sanitation ICR Implementation Completion Report IRR Internal Rate of Return ISR Implementation Status Report M&E Monitoring and Evaluation MAVDT Ministerio de Ambiente, Vivienda, y Desarrollo Territorial (Ministry of Environment, Housing, and Regional Development) Vice President: Pamela Cox Country Director: Gloria Grandolini Sector Manager: Guang Zhe Chen Project Team Leader: Greg Browder ICR Team Leader: Greg Browder/Eric Dickson COLOMBIA Water Sector Reform Assistance Project CONTENTS Data Sheet A. Basic Information B. Key Dates C. Ratings Summary D. Sector and Theme Codes E. Bank Staff F. Results Framework Analysis G. Ratings of Project Performance in ISRs H. Restructuring I. Disbursement Graph 1. Project Context, Development Objectives and Design ................................................... 1 2. Key Factors Affecting Implementation and Outcomes .................................................. 3 3. Assessment of Outcomes ................................................................................................ 8 4. Assessment of Risk to Development Outcome ............................................................. 15 5. Assessment of Bank and Borrower Performance ......................................................... 17 6. Lessons Learned............................................................................................................ 18 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners............... 20 Annex 1: Project Costs and Financing .............................................................................. 22 Annex 2: Project Performance Indicators ......................................................................... 27 Annex 3: Economic Analysis: Evaluation of Investment Efficiency ............................... 37 Annex 4: Financial Analysis ............................................................................................. 41 Annex 5: Specific Lessons for the Corporate Modernization Program ............................ 50 Annex 6: Bank Lending and Implementation Support/Supervision Processes ................ 52 Annex 7: Stakeholder Workshop Report and Results....................................................... 54 Annex 8: Summary of Borrower's ICR and/or Comments on Draft ICR ......................... 55 Annex 9: List of Supporting Documents .......................................................................... 56 Maps: IBRD COL38454, COL38455 A. Basic Information WATER SECTOR REFORM Country: Colombia Project Name: ASSISTANCE PROJECT Project ID: P065937 L/C/TF Number(s): IBRD-70770 ICR Date: 04/28/2011 ICR Type: Core ICR REPUBLIC OF Lending Instrument: SIL Borrower: COLOMBIA Original Total USD 40.0M Disbursed Amount: USD 35.7M Commitment: Revised Amount: USD 35.7M Environmental Category: B Implementing Agencies: Ministry of Environment, Housing and Territorial Development Cofinanciers and Other External Partners: B. Key Dates Revised / Actual Process Date Process Original Date Date(s) Concept Review: 11/21/2000 Effectiveness: 06/18/2002 Appraisal: 06/11/2001 Restructuring(s): Approval: 10/25/2001 Mid-term Review: 09/30/2004 01/16/2006 Closing: 06/30/2007 10/31/2010 C. Ratings Summary C.1 Performance Rating by ICR Outcomes: Satisfactory Risk to Development Outcome: Substantial Bank Performance: Satisfactory Borrower Performance: Satisfactory C.2 Detailed Ratings of Bank and Borrower Performance (by ICR) Bank Ratings Borrower Ratings Quality at Entry: Satisfactory Government: Moderately Satisfactory Implementing Quality of Supervision: Satisfactory Satisfactory Agency/Agencies: Overall Bank Overall Borrower Satisfactory Satisfactory Performance: Performance: i C.3 Quality at Entry and Implementation Performance Indicators Implementation QAG Assessments Indicators Rating Performance (if any) Potential Problem Project Quality at Entry Yes None at any time (Yes/No): (QEA): Problem Project at any Quality of No None time (Yes/No): Supervision (QSA): DO rating before Satisfactory Closing/Inactive status: D. Sector and Theme Codes Original Actual Sector Code (as % of total Bank financing) Central government administration 2 2 Sanitation 49 49 Water supply 49 49 Theme Code (as % of total Bank financing) Environmental policies and institutions 13 13 Other financial and private sector development 25 25 Other urban development 24 24 Rural services and infrastructure 13 13 Urban services and housing for the poor 25 25 E. Bank Staff Positions At ICR At Approval Vice President: Pamela Cox David de Ferranti Country Director: Laura Kullenberg Olivier Lafourcade Sector Manager: Guang Zhe Chen Danny M. Leipziger Project Team Leader: Greg J. Browder Menahem Libhaber ICR Team Leader: Greg J. Browder ICR Primary Author: Eric Dickson Greg J. Browder F. Results Framework Analysis Project Development Objectives (from Project Appraisal Document) The development objectives of the project are: (i) to support water sector reform by facilitating the incorporation of the private sector into the management and operation of ii water utilities participating in the project for the purpose of creating an enabling environment for improving the efficiency and sustainability of these utilities, and by providing them with financial support to ensure their viability; (ii) expand the coverage of water and sewerage services in participating municipalities; and (iii) facilitate the access of populations in low income areas of participating municipalities to water and sewerage services. Revised Project Development Objectives (as approved by original approving authority) The PDO was not revised during the project. (a) PDO Indicator(s) Original Target Formally Actual Value Values (from Revised Achieved at Indicator Baseline Value approval Target Completion or documents) Values Target Years Indicator 1 : Number of PSP Contracts Signed Value Medium Cities: 2 Medium Cities: 18 quantitative or None Small Cities: 15 Small Cities: 7 Qualitative) Date achieved 06/30/2001 10/31/2010 10/31/2010 Comments (incl. % Medium Cities target greatly exceeded, while small cities less successful. achievement) Indicator 2 : Coverage Rates for Water Supply Infrastructure Value quantitative or 61% 80% 84% Qualitative) Date achieved 06/30/2001 10/31/2010 10/31/2010 Comments (incl. % Target Achieved achievement) Indicator 3 : Coverage Rates for Sewerage Infrastructure Value quantitative or 40% 65% 50% Qualitative) Date achieved 06/30/2001 10/31/2010 10/31/2010 Comments (incl. % Target Not Achieved. achievement) Indicator 4 : Continuity of Water Supply Service: 24 Hours/7 Days a Wee Value quantitative or 2% 80% 49% Qualitative) Date achieved 06/30/2001 10/31/2010 10/31/2010 Comments Target not achieved, but in many cases residents who do not have continuous (incl. % supply experienced a significant increase in hours per day of service. achievement) iii Indicator 5 : Water Utility Efficiency: Staff per 1000 Connections Value quantitative or 7.6 4 5.6 Qualitative) Date achieved 06/30/2001 10/31/2010 10/31/2010 Comments There were significant decreases in staff but the target value of 4 was not (incl. % achieved. achievement) Indicator 6 : Water Utility Financial Sustainability: Working Ratio Value quantitative or 0.96 0.6 0.84 Qualitative) Date achieved 06/30/2001 10/31/2010 10/31/2010 Comments (incl. % Target value was not reached and financial sustainability remains a concern. achievement) (b) Intermediate Outcome Indicator(s) Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised approval Completion or Target Values documents) Target Years Indicator 1 : There are IO indicators for the data sheet Value (quantitative or Qualitative) Date achieved Comments (incl. % achievement) G. Ratings of Project Performance in ISRs Actual Date ISR No. DO IP Disbursements Archived (USD millions) 1 11/27/2001 Satisfactory Satisfactory 0.00 2 06/03/2002 Satisfactory Satisfactory 0.00 3 11/27/2002 Satisfactory Satisfactory 1.00 4 05/15/2003 Satisfactory Satisfactory 2.16 5 05/29/2003 Satisfactory Satisfactory 2.16 6 11/14/2003 Satisfactory Satisfactory 2.92 7 06/03/2004 Satisfactory Satisfactory 4.94 8 06/30/2004 Satisfactory Satisfactory 4.94 9 12/03/2004 Satisfactory Satisfactory 8.06 10 04/29/2005 Satisfactory Satisfactory 11.89 iv 11 06/05/2006 Highly Satisfactory Satisfactory 21.98 12 12/18/2006 Highly Satisfactory Satisfactory 26.89 13 05/02/2007 Highly Satisfactory Satisfactory 29.45 14 12/04/2007 Highly Satisfactory Satisfactory 32.48 15 06/30/2008 Highly Satisfactory Satisfactory 35.69 16 12/10/2008 Satisfactory Satisfactory 35.69 17 05/09/2009 Satisfactory Satisfactory 35.69 18 12/15/2009 Satisfactory Satisfactory 35.69 19 06/14/2010 Satisfactory Satisfactory 35.69 20 11/14/2010 Satisfactory Satisfactory 35.69 H. Restructuring (if any) Not Applicable I. Disbursement Profile v 1. Project Context, Development Objectives and Design 1.1. Context at Appraisal Providing water and sanitation services to Colombia‘s 1,122 municipalities, most of them small and relatively poor, has always been a challenge. Prior to 1987 water service in Colombia was provided through two different systems: municipal companies in large cities such as Bogotá, Medellin, and Cali, and directly by the National Government through the Municipal Development Institute (INSFOPAL) in small and medium sized cities. In 1987 INSFOPAL was abolished and financing responsibility was shifted to FINDETER, the Regional Development Financing Institution. The water sector then entered a transition stage until 1994, when the National Planning Department (DNP) was assigned the function of planning and technical assistance for the sector. The 1991 Colombian Constitution defined a model of increased decentralization for the provision of public services and promoted a shift away from direct provision of services by municipal government towards one where service is provided by public or private companies. In support of these processes, beginning in the mid-1990s Colombia implemented a set of institutional reforms in the sector that provided a framework for private participation in service provision. The 1994 Law of Public Services (Law 142) mandated that all public service entities convert to legally autonomous stock companies with the possibility of being municipal-owned, private, or mixed. Law 142 also established a national tariff regulator, the Comisión Reguladora de Agua Potable y Saneamiento (CRA). The regulatory framework was complemented by Law 60 of 1993 and Law 715 of 2001 which created the General Revenue-Sharing System (SGP), a tax- sharing system which determined the resource distribution of national government transfers to the municipalities, part of which is allocated to the water sector. In 1998 the Corporate Modernization Program (PME) for the introduction of private sector participation in the water sector was established in the Ministry of Economic Development. The goal of the PME is to provide technical assistance to decentralized entities (municipalities and departments) for partnerships with the private sector. The PME focused on the Caribbean coastal region, where coverage and quality were most lacking, and institutional weaknesses evident. The PME adopted a novel approach whereby the national government provided most of the investment financing while still providing financial incentives for improved performance by the operators. The World Bank-financed Water Sector Reform Assistance Project (―the Project‖) was designed to support the PME and concentrated on the departments of the Caribbean coast. The Project design was partly motivated by the positive experience from two successful private operator contracts, signed for Cartagena in 1995 and for Barranquilla in 1996. The Project was considered a pilot project to introduce specialized operators to small and medium sized cities, and was considered a high risk-high reward project. As shown in this ICR the Project is evaluated as a successful pilot project that had a significant impact on Colombia‘s water sector as well as improving water service in specific subproject cities. 1 1.2. Original Project Development Objectives (PDO) and Key Indicators The development objectives of the Project 1 were to: (i) support water sector reform by facilitating the incorporation of the private sector into the management and operation of water utilities participating in the Project for the purpose of creating an enabling environment for improving the efficiency and sustainability of these utilities, and by providing them with financial support to ensure their viability; (ii) expand the coverage of water and sewerage services in participating municipalities; and (iii) facilitate the access of populations in low income areas of participating municipalities to water and sewerage services. Physical investments were limited to the eight departments of the Caribbean coast. These objectives were expected to be achieved by providing: (i) technical assistance for incorporating private sector participation in the management and operation of the water and sewerage services in the utilities in about 3 medium size cities or regional associations of municipalities, with populations of up to about 300,000 inhabitants, and in the utilities of about 25 small municipalities with populations of up to about 12,000 inhabitants; and (ii) financial support to the utilities which have successfully incorporated the private sector; and which will be directed at benefiting the poor, as well as ensuring the financial viability of these utilities. The table below presents a streamlined version of the indicators presented in Annex 1 of the PAD. Since all of the Project municipalities have over 90% of their population in the lower economic brackets (see Annex 2), project development indicators (ii) expand service and (iii) serve low-income areas have been merged into one set of indicators for the ICR. PDO PDO Indicators 1. Support water sector reform by PSP incorporated in medium size cities and small facilitating the incorporation of the municipalities private sector into the management and operation of water utilities 2. Expand the coverage of water and A. Water supply infrastructure coverage sewerage services and efficiency of B. Sewerage infrastructure coverage operators in participating C. Coverage rates for continuous water supply municipalities D. Service Efficiency: Employees/1,000 connections E. Service Sustainability: Operator Working Ration 1.3 Revised PDO and Key Indicators: The PDOs were not revised. 1.4 Main Beneficiaries According to the Project Appraisal Document (PAD), the primary target group comprised a population of an estimated 700,000 people in low-income communities in Colombia‘s Caribbean 1 The Project Loan Agreement also includes a 4th objective ―(i) to improve environmental protection practices in connection with the provision of water and sewerage services and (ii) to define rural water and sanitation policies, and to develop adequate methods for increasing coverage, both in connection with water and sewerage services rural sector.‖ Since there were no PDO indicators for these objectives, and the amount of Project resources allocated for these activities were limited, the ICR adopts the PAD objectives. 2 coastal departments. These populations are typically assigned to the first three (and lowest income) consumer categories in Colombia‘s six point classification (strata) of consumers.2 The primary target group was not revised during project implementation, however, beneficiaries increased to approximately 1.6 million people largely as a result of an unexpected number of medium-sized municipalities participating in the project. 1.5 Original Components (as approved) Component A (US$61.8 million): Investments in water supply and sanitation works in medium- size cities and in small municipalities. Component B (US$0.8 million): Environmental management capacity strengthening for the sector, with a focus on wastewater management. Component C (US$ 0.9 million): Development of a rural water and sanitation policy. Component D (US$6.5 million): Project management and training. 1.6 Revised Components: The Project components were not formally revised. 1.7 Other significant changes: There were no significant changes in the Project design. 2. Key Factors Affecting Implementation and Outcomes 2.1 Project Preparation, Design and Quality at Entry The key design element of the Project was the requirement to utilize private operators in order to access loan funds. Although the Project design built upon the successful introduction of private operators in Cartagena and Barranquilla, introduction of private operators in small and medium- sized cities was a much more risky undertaking and the overall PAD risk rating was evaluated as ―high‖. The pre-dominant concerns were that: i) municipalities would not want to participate in the program; and ii) there would be little appetite from private operators to work in poor small and medium sized cities in the Caribbean which have historically suffered from poor governance. The Project design benefitted from the insight that excessive risk could not be shifted to the private operator until the deteriorated systems had been improved to the point where the operating revenue streams had stabilized at a sustainable level. Accordingly the Project was prepared with a number of key design aspects such as (i) generous public financing to rehabilitate the run-down systems in the targeted municipalities; (ii) the use of domestic, rather than international, private operators in order to build a national pool of small and medium sized private operators; (iii) generous and continuous technical assistance from the National 2 The classification is derived from the property values of the dwellings in which the families live under the assumption that higher property values are directly linked to higher household incomes, and vice versa. 3 Government project implementation unit to the small-scale private operators; and (iv) the drafting and refinement of transparent bidding documents for two types of schemes to partner with specialized operators. The two standard contract types differed in their focus on financing of investments as described below.  The ‗construction with operation‘ model was designed for small municipalities, generally with a population less than 12,000. The basic concept is that the national government would define the investment program, and the municipality would then contract a ―constructor-operator‖ for both the infrastructure construction in the first 2-3 years, and operation of the system for ten or more years. It was assumed that contractors would be enticed to secure the construction contract in exchange for the less-than lucrative operational obligations. In this model the municipality is the Employer and the ―constructor-operator‖ is the contractor.  The ‗operation with investment‘ model was designed for medium-sized municipalities and followed a more traditional concession model. The national government would define the investment program, and then operators would bid for the concession contract based upon the ―lowest subsidy‖ required to meet the pre-defined investment amounts. In theory, the Operator would be able to generate a positive cash flow from operational efficiencies to finance part of the investment program. In this model the operator serves as the Employer for all construction contracts. The Borrower for the Loan is the Republic of Colombia, and the Project was designed to be managed by the national government, originally through the Ministry of Economic Development. The actual implementation of the Project, however, in terms of contracting would be through the municipalities, in the case of the constructor operator model, and the operators, in the case of the operation with investment model. The Project was prepared in a programmatic and flexible manner with none of the participating municipalities identified during appraisal. The Project was prepared and brought to the Bank Board of Directors in less than one year. This was possible since the Project was prepared by the project implementing unit in the Ministry of Economic Development under the on-going Bank financed Municipal Development Project (3336-CO). The commitment of the Borrower was high as evidenced by: i) amending the on- going Municipal Development Project Loan Agreement to create a new component for water utilities' modernization and reallocating US$5 million for the sector policy reform program; ii) a strong information dissemination effort regarding the government‘s policy, which resulted in generating a demand by mayors for participating in the project; iii) numerous private sector participation processes being initiated during project preparation in both medium size cities and small municipalities; and iv) a financial contribution of nearly half the project cost through national government funding. 4 2.2 Implementation The Project closing date was extended from June 30, 2007 to October 31, 2010 through three extensions. The Project experienced a number of adjustments during implementation which are highlighted below. National Government Reforms-Creation of MAVDT: With the advent of a new presidential administration in 2002, the Ministry of Economic Development was merged with the Ministry of Environment to form the Ministry of Environment, Housing, and Regional Development (MAVDT). The rapid creation of the MAVDT proved initially disruptive to implementation of the Project due to high turnover of staff and frequent management changes. Over the lifetime of the Project four different ministerial counterparts were assigned, each of whom understandably required time to learn the Project‘s design and status of implementation. The appointment of the first Vice-Minister of Water in 2006 helped to stabilize project implementation and staffing. Loan Cancellation: Disbursements under World Bank financed projects are fully integrated within the Colombian central government budget. When projects are implemented slower than forecast, the portion of the annual appropriations for investments under the central government budget that cannot be utilized in a given year is lost and cannot be recuperated through budget appropriations in later years. As a result of the slow Project start-up US$4.31 million of budgeted resources were ―lost‖ and subsequently cancelled from the US$40 million loan, resulting in a final disbursement of US$35.69 million. Shift to Medium-Sized Cities: Only three medium-sized cities were expected to participate in the Project under the ―operation with investment model‖. During implementation, however, this expanded to 18 municipalities participating in the Bank-financed Project as a result of the MAVDT‘s focus shifting to medium-sized cities during the early phases of the Project. The national government provided more financing then originally estimated, increasing from US$30 million to US$45.9 million for the entire PME program. The Project loan resources were therefore spread out in smaller amounts to a greater number of medium-sized cities but this was partially compensated by higher level of counterpart funds. Conversely the number of small municipalities participating under the ―construction with operation‖ model was only seven, under three separate contracts, considerably less than the original forecast of fifteen. Private sector interest and municipal interest in the smaller municipalities was less than originally anticipated, and the MAVDT focus on medium-sized cities diverted attention from the smaller cities. Lingering Sub-Projects: Two sub-projects experienced significant delays which resulted in multiple extensions of the Project closing date. The operator originally selected for Magangué (Aguas Kapital SA ESP) was extremely slow in implementing the investment program, and was replaced by a new operator (Aquaseo SA ESP) in 2010. The La Linea subproject in Bolivar experienced delays in the installation of a critical pipeline due to complications with a parallel highway construction project. 5 2.3 Monitoring and Evaluation (M&E) Design, Implementation and Utilization M&E Design. The PAD was formulated in 2000, and there have been subsequent changes in the Bank‘s guidelines. Overall the ICR considers that the selected monitoring indicators were appropriate to measure the project development objectives. The PDO indicators in Annex 1 of the PAD were reorganized for the ICR to be consistent with the current methodology and more congruent with the stated project objectives. The original objectives call for expanding water and sewerage services in participating municipalities (PDO 2), and also specifically among poor communities (PDO 3). Since the Project only intervened in poor municipalities (see Annex 2) where over 90% of the residents fall into the lowest three strata, only one set of indicators is used in the ICR to measure the improvement of water and sewerage services. 2.4 Safeguard and Fiduciary Compliance Environmental Safeguards (OP4.01): The Project was classified as a category B project. Most of the investments comprised rehabilitation of existing water supply and sewerage systems without any significant negative environmental impact. No wastewater treatment plants were built under the Project. Each of the construction contracts included standard environmental and safety requirements, and the contracts were supervised by independent construction supervisors to ensure compliance with environmental and safety requirements. Procurement: Procurement related activity took place at two levels: Selection of Operators: The Project Loan agreement required the World Bank to provide a no- objection to the selection of the operator. It is important to note that this no-objection is not for a Bank-financed contract, but rather for the selection of an Operator based on agreed-upon World Bank-MADVT procedures and criteria. During the early stages of the Project, the Bank worked with MAVDT to develop model concession bidding documents, which incorporated the following aspects: i) the investment program is defined by consultants contracted by MAVDT; ii) the long-term tariffs are established in the bid documents; and iii) the contract is awarded to the operator which proposes the ―least-cost‖ subsidy necessary for meeting the established investment amount. The selected operator is provided flexibility to adjust the investment program, but not the amount, over the course of time. During the course of the Project, 18 municipalities signed contracts under the ―operation with investment‖ model and received the required Bank no-objection. The operator procurement process and subsequent implementation did not proceed smoothly however. Some of the common complications include: i) no bidders or only one bidder; ii) failed concession contracts; and iii) directly negotiated contracts. The award of the concession contracts needed the approval of both MAVDT and the Bank, which helped to ensure a certain level of contract integrity and transparency. Annex 5 contains a summary of lessons learned in the selection/procurement of specialized operators. 6 Works, Goods, and Service Contracts: For the ―operation with investment‖ model the operators were the implementing agencies acting as the Employer; for the ―construction with operation‖ model the municipalities were the implementing agencies acting as the Employer. In total there were 12 operators and three municipal entities (two serving multiple municipalities) acting as implementing agencies and contracting under Bank procurement guidelines. This decentralization of procurement responsibility was only possible due the technical support and supervision provided by the MAVDT, which had to train the implementing agencies and oversee the procurement process. The Project‘s procurement performance was generally rated as highly satisfactory/satisfactory throughout the life of the Project and there were no cases of mis- procurement. The Project was the subject of an INT investigation in early 2009 associated with the La Linea sub-project. The INT report has been completed and concluded that there was no evidence of malfeasance. Financial Management: The auditors issued unqualified (clean) opinions on the projects financial statements for the audit periods 2006 to 2009. The audit reports contained several internal control and non-compliance findings, issues that as of the project closing date have been corrected. However, the two latest audit reports have been submitted 2 months late, and the contracting of the auditors for the 2010 audit is delayed, despite the Bank's No Objection to the Terms of Reference provided on January 13, 2011. Therefore the final audit report, for the period January 1, 2010 to October 31, 2010, will not be submitted by the due date of April 30, 2011. The MAVDT plans to have the auditors contracted by May 31, 2011 and issue the final report by August 15, 2011. The quarterly interim financial reports (IFRs), which are not audited, were submitted within the contractual date and in general they have been acceptable. The ISR Financial Management (FM) rating up to March 2009 was assessed as Moderately Unsatisfactory (MU), mainly due to the following issues : (a) the Government Official information system (SIIF) was not capturing all the project transactions, thus manual records had to be kept; (b) deficiencies in the internal controls; and (c) delays in the submission of withdrawal applications. These conditions have improved, and as of April 2010, the FM rating was upgraded to Moderately Satisfactory (MS). 2.5 Post-completion Operation/Next Phase During the course of the Project a number of legislative reforms and executive decrees were adopted to improve the enabling framework and sustain the reforms. By 2005 the MAVDT and the Bank realized that in order to expand private sector participation and strengthen service in smaller municipalities, it would be necessary for the Department governments to take more of a leadership role. The Department of La Guajira was chosen as a pilot, and the Bank approved the on-going La Guajira Water and Sanitation Infrastructure and Service Management Project in 2007 (P096965). Based mainly on experience from this project and partly on the La Guajira model, the national government created the Departmental Water Program (PDA) in 2007 (CONPES Policy Paper 3463) which channels national government funds for water and sanitation through the department government and requires municipalities to ensure water utilities sustainability stimulating private sector participation.. 7 3. Assessment of Outcomes 3.1 Relevance of Objectives, Design and Implementation The Project was considered a pilot project to test a novel approach to private sector participation in the small and medium-sized cities. The results of the Project were perceived to have an important impact on Colombia‘s overall national water and sanitation policy and investment program. The Project also responded directly to the need to improve water and sanitation services in the Caribbean region. At the start of the project only 60% of the residents had access to the distribution network, which only operated intermittently. At a higher policy level, the Project directly responded to the Colombia Country Assistance Strategy (CAS) of 1997, which identified the poor and deteriorated state of the country‘s infrastructure as a critical constraint to economic growth. The government‘s strategy emphasized private sector participation to expand water and sanitation services and improve wastewater treatment infrastructure. The CAS identified three key sector-related objectives: (i) creation of mechanisms to facilitate private sector participation in the provision of urban public services, (ii) increase in quality and coverage of water supply and sanitation services, and (iii) achieve sustainable economic growth by improving infrastructure services. The project contributed to the achievement of these objectives. The Project also played an instrumental role in strengthening MAVDT‘s Corporate Modernization Program (PME) and helped lay the ground work for the national policy on the Departmental Water Plan program. The current Country Partnership Strategy (CPS), dated March 4, 2008, is derived from the 2006- 2010 National Development Plan which comprises five development pillars, each of which has clear goals and progress indicators. The Project directly supports the pillars of ―Promotion of Equity‖ and the goals of ―Livable Cities‖ and ―Infrastructure for Development‖. Based on the close alignment of the Project Development Objectives with Colombia’s development priorities and the Country Partnership Strategy, the ICR rates the relevance of the Project as “high”. 3.2 Achievement of Project Development Objectives (Efficacy) The first development objective of the Project is to: ―support water sector reform by facilitating the incorporation of the private sector into the management and operation of water utilities participating in the Project for the purpose of creating an enabling environment for improving the efficiency and sustainability of these utilities, and by providing them with financial support to ensure their viability.‖ Annex 1 provides a summary of overall project costs, financing and a list of Bank-financed and non-Bank financed projects by municipalities. Under the overall PME program, 36 municipalities in the Caribbean region now have private operators; twenty five of these municipalities received World Bank financing. The total cost of the PME as of 2010 was $75 8 million, of which the Bank loan financed US$24 million 3 . The table below summarizes the indicators for the incorporation of private sector operators in the Caribbean region. The indicators are only for the Bank-financed subprojects. PDO Indicator 1: Incorporation of PSP in Medium and Small Sized Cities Baseline Target 2010 Actual From PAD Operation-Investment 0 2 11 Contracts with 18 municipalities; (Medium-Cities) Total Population of 1,561,884 Construction-Operation 0 15 3 Contracts with 7 municipalities; (Small Cities) Total Population of 101,285 Operation with Investment - Achieved: The number of PSP operations with investment contracts greatly exceeded the estimate for medium sized cities in the PAD. In addition, the MAVDT expanded the program beyond the Bank- financed subprojects. Construction with Operation - Not Achieved: The number of PSP contracts for smaller cities using the Constructor-Operator model was significantly lower than PAD targets. This was not necessarily a bad outcome as in general the Constructor-Operator model is not functioning particularly well. The table shows that the Project was highly successful for medium-sized cities and greatly exceeded original targets. The experience with smaller cities was less successful however as only three ―construction-operation‖ contracts were signed with seven municipalities. This is partly explained by a conscious and rational decision by MAVDT to focus on medium sized cities as a priority. Since the Project was designed in a flexible and programmatic manner, the shift in focus to medium sized cities is not a shortcoming. Based on the success in introducing private sector participation in medium-sized cities combined with modest achievements for small cities, the ICR rates the achievement of PDO 1 as “substantial”. The second PDO relates to the expansion of water and sewerage services in participating municipalities, and facilitating access for low-income populations. Since all of the participating municipalities have over 90% their populations in the low-income stratum (Strata 1, 2, and 3) only one set of indicators are used with the assumption that all of the service improvements are directed at the low-income groups. Each indicator is analyzed separately below. PDO Indicator 2.A: Coverage rates for water supply infrastructure % Coverage PAD Target % Coverage Standard (Population in All (Efficacy (Average Per Deviation Municipalities) Grade) Municipality) Baseline 2010 2010 2010 Total Project 61% 84% 78% 22% Operation-Investment 61% 85% > 80% 83% 15% (Medium-Cities) (High) Construction-Operation 64% 65% > 90% 66% 33% (Small Cities) (Modest) 3 Refers to Component A1 and A2 only. See Annex 1. 9 There was a significant increase in the water supply infrastructure coverage for medium-sized cities, increasing from 61% to 85% which is above the target value of 80%. The 15% standard deviation of the municipal average, however, reflects the fact that 5 of the 18 municipalities did not meet the target value. There were some successes in medium sized cities, including significant expansions of water supply infrastructure in El Banco (Operagua el Banco), five municipalities in Atlantico (Triple A), four municipalities in Cordoba (Uniaguas), Maicao in La Guajira (Aguas de la Peninsula), San Marcos (Aguas de la Mojana) and Sincelejo (Aguas de la Sabana) in Sucre and Sincelejo. There was no significant increase in water supply infrastructure coverage for the small municipalities with the construction-operation model. The efficacy rating for small cities, however, is evaluated as modest, as there were some improvements in the water supply infrastructure in all three subprojects. For example, in the case of the La Linea sub-project, which covers four municipalities in Bolivar, the primary focus was on improving the main water distribution line connecting the cities rather than expanding the distribution network. PDO Indicator 2.B: Coverage rates for sewerage infrastructure % Coverage Target % Coverage Standard (Population in All From PAD (Average Per Deviation Municipalities) Municipality) Baseline 2010 2010 2010 Total Project 40% 50% 25% 31% Operation-Investment 43% 53% > 65% 34% 33% (Medium-Cities) (Modest) Construction-Operation 5% 6% > 75% 4% 11% (Small Cities) (Negligible) The provision of sewerage services in medium-sized cities (all without treatment) lagged behind water supply, and the target of 65% was not achieved, although there was a significant increase in coverage from 43% to 53%. The only cities which met the target were Sincelejo (81%), Corozal (79%), Sabanagrande (79%), Soledad (71%), and Sahagun (72%). It should be noted that all of these cities started the project with coverage rates above 60% and thus the increase was not dramatic. There is essentially no sewerage service in small cities. The lack of focus on sewerage is understandable given that usually the first priority for operators and customers in a constrained economic environment is water supply. PDO Indicator 2.C: Continuity of Water Supply Service % with 24 Hour Supply Target % with 24 Hour Standard (Population in All From PAD Supply Deviation Municipalities) (Average per Municipality) Baseline 2010 2010 2010 Total Project 2% 49% 32% 42% Operation-Investment > 80% 45% 44% (Medium-Cities) (Substantial) 2% 53% Construction- 0% 0% > 90% 0% 0% Operation (Small City) (Negligible) 10 Provision of water supply infrastructure coverage does not guarantee adequate service, as most cities suffer from the problem of intermittent supply. There was a significant increase in continuity in medium cities, with continuous supply coverage increasing from 2% to 53% for the total Project population. The systems operated by Triple A in Atlántico (5 cities) and Uniaguas in Córdoba (2 cities) achieved 90% continuous supply coverage. In almost all other cities, even if continuous supply was not achieved, the level of service, in terms of hours per day, increased significantly. Thus even though the target was not achieved the efficacy for medium cities is rated as substantial. Water supply service did not improve in small cities, although there are prospects for future improvements due to targeted investments made to strengthen the primary networks. The overall rating for Project efficacy is calculated using a population weighted scoring system as shown in the table below, and the overall rating for efficacy is considered substantial. Indicator Operation with Construction with Score Investment Operation Medium Cities Small Cities Weight: 94% Weight: 6% Water Supply Infrastructure Coverage High Modest 3.88 Sewerage Infrastructure Coverage Modest Negligible 1.94 Continuity of Service Substantial Negligible 2.88 Final Average 2.9 Notes: Total Population = 1,663,169; Medium = 1,561,884 (94%); Small = 101,285 (6%) Scores: Negligible = 1; Modest = 2; Substantial =3; High = 4 3.3 Efficiency Project efficiency is evaluated on two levels: i) the efficiency of the water utility as measured by staff per 1,000 connections; and ii) efficiency of Project investments, as qualitatively measured by a water engineering expert. Both parameters are rather crude measures of efficiency, as an operator can be efficient even with a relatively high staffing ratio, and evaluation of investment efficiency contains a certain level of subjectivity. Nevertheless, these parameters provide useful insights into efficiency of investments and operations. An analysis of global investment efficiency was undertaken for six subprojects (15 cities), which serves as a type of least-cost economic analysis. A traditional economic cost-benefit analysis was not undertaken due to the large number of subprojects, the data intensive nature of the exercise, and the common recognition that providing water and sanitation services provides significant health and other economic benefits. As shown in Section 4, the ICR analysis placed great emphasis on financial analysis to examine sustainability issues. 11 PDO Indicator 2.D: Water Utility Efficiency Staff Per 1,000 Target Staff Per 1,000 Standard Connections PAD Connections Deviation (Population Weighted) (Average per Municipality) Baseline 2010 2010 2010 Total Project 4.9 1.8 7.6 5.6 Operation-Investment <4 4.4 1.7 (Medium-Cities) 7.8 5.6 (Substantial) Construction-Operation <5 5.9 1.6 (Small Cities) 5.1 5.7 (Negligible) For medium sized cities, the population weighted value of staff per 1,000 connections decreased significantly from 7.8 to 5.6, while the average value was 4.4, indicating that at least some operators were close to the target value of 4. The story the construction-operation contracts in small cities is less clear. It appears that staffing may have increased slightly, but this could be due to the construction activities undertaken by the contractor. Investment Efficiency Indicator Investment Efficiency Investment Efficiency Standard (Population Weighted) (Average per Deviation Municipality) Total Project 2.7 1.4 3.5 Operation-Investment 3.6 0.7 (Medium-Cities) 3.7 (High) Construction-Operation 1.0 0.0 (Small Cities) 1.0 (Negligible) Scores: Negligible = 1; Modest = 2; Substantial =3; High = 4 A scoring system was developed for qualitatively assessing investment efficiency in each subproject. In total six subprojects (15 cities) were evaluated and the methodology and results are presented in Annex 3. Operators in medium-sized cities generally made good investment decisions, particularly the two largest operators: Triple A in Atlantico and Aguas de la Sabana in Sincelejo and Corozal. The evaluation indicated that the investment programs in the two construction-operation contracts (La Linea and San Onofre) were not well formulated. It is important to highlight the ―constructor-operator‖ was not primarily accountable for the definition of the investment program which was defined in advance by the MAVDT/municipalities through consultancy contracts. For example, during ICR preparation the operator for the regional sub- project La Linea (Giscol) noted continual challenges in negotiating amendments to the existing contract to adjust the investment program. This highlights a built-in tension in the ―constructor- operator‖ model in which the Employer has to control the scope of construction but the constructor/operator needs flexibility to adjust the investment program according to operational requirements. The overall rating for Project efficiency is calculated using a population weighted scoring system as shown in the table below, and the overall rating for efficacy is considered substantial. 12 Indicator Operation with Investment Construction with Score Medium Cities Operation Weight: 94% Small Cities Weight: 6% Water Utility Efficiency Substantial Negligible 2.9 Investment Efficiency High Negligible 3.9 Final Average 3.4 Notes: Total Population = 1,663,169; Medium = 1,561,884 (94%); Small = 101,285 (6%) Scores: Negligible = 1; Modest = 2; Substantial =3; High = 4 3.4 Justification of Overall Outcome Rating The ICR rates the overall outcome of the Project as “satisfactory” based on a “high” rating for relevance, and “substantial” ratings for efficacy and efficiency. The Project was successful in introducing the private sector operator model to the Caribbean region of Colombia, and improvements in service were demonstrated by three operators in medium sized cities covering 12 municipalities: Triple A, Aguas de la Sabana, and Uniaguas. The construction-operation model has not yet demonstrated its viability, but fortunately the Project scaled back on this model during implementation and focused on the operation with investment model. 3.5 Overarching Themes, Other Outcomes and Impacts (a) Poverty Impacts, Gender Aspects, and Social Development The Project has had a dramatic impact on poverty. Approximately 1.3 million people in the lower economic stratum (1, 2, and 3) received improved water supply service, although the level of improvement depends on the specific municipality. There was also a 23% increase in the number of households connected to the water distribution systems, from 61% to 84%, certainly almost of all them from low-income households. Households not connected to the distribution systems typically purchase water from water tankers, often at high prices. Higher connection rates do not only imply greater convenience and access to safe water, but the cost of water also is significantly reduced. This is particularly true in Colombia where low-income households receive implicit subsidies through lower utility tariffs. (b) Institutional Change/Strengthening Institutional strengthening has occurred at multiple levels. At the national level, the Project helped build the capacity of MAVDT, particularly through the Corporate Modernization Program (PME), in the implementation of the Project. Component D financed a significant share of the specialized consultant team in MAVDT that helped promote, structure, and supervise the introduction of private operators under the Project and throughout the country. Efforts to improve environmental management (Component B) and rural water and sanitation (Component C), however were less successful than expected. 13 Component B: Environmental management capacity strengthening for the sector, with a focus on wastewater management. Wastewater management smaller municipalities in Colombia is a particular challenge with only treatment levels of less than 30%, and treatment plants often not functioning properly. The Project financed the development of a decision support tool for assisting in the planning of smaller municipal wastewater treatment plants. Generic plans, costs, specifications, and manuals were developed for over 13 treatment technologies. The planning and decision support tool allows users to work through different options, pick the most appropriate treatment technology, and then provide guidance on the planning and design. Although potentially useful, the ICR team understands that the planning tool has not been formally disseminated and put into practice. Component C: Development of a rural water and sanitation policy. To address the challenges water services in rural areas, the following activities were financed through the Project: i) the preparation, publication and dissemination of the document "Public Policy Guidelines for Drinking Water and Sanitation for Rural Colombia‖, ii) a Rural Health Inventory in 18 departments of the country; and iii) a series of capacity building exchanges and workshops at the community, local government and international levels. While policy directions were identified, the MAVDT‘s rural water program became inactive and a national rural water policy was not formally defined. During the course of the ICR evaluation the Bank team was informed the MAVDT is currently formulating a rural water policy to address this recognized gap in sector policy that is directly linked to the PDAs and which will be structured based on the outputs from this Project component. The Project also helped to build a base of private operators who gained experience in dealing with smaller municipalities. Under the Project there were 14 private sector concession contracts which have helped to widen and strengthen the pool of competent operators in the sector. The general trend is for operators to gain experience in a larger municipality and then spread out into neighboring smaller municipalities, bring their experience and expertise to bear in the small markets that usually would be unattended. This was certainly the case for Triple A in Atlántico, which expanded outwards from the city of Barranquilla. The Project also helped to at least partially break down the short-term view of water service delivery at the municipal level where mayors are elected on four year cycles. Previously there was a tendency of mayors to rely on inefficient direct municipal management which afforded the opportunity for political interference. Such short-term horizons have proved highly detrimental to the coverage and quality of water supply and sewerage operations that are capital-intensive and require professional management. Through the stability afforded by long-term operator contracts the previous ―short-termism‖ has been replaced by steady improvement of the capacity and asset base of the private operators which has then resulted in higher coverage and quality of services. 14 (c) Other Unintended Outcomes and Impacts (positive or negative) 3.6 Summary of Findings of Beneficiary Survey and/or Stakeholder Workshops An ICR review workshop was conducted in Bogota on April 11, 2011 with representatives of the MAVDT. The Bank team presented the following components: i) evaluation process methodology, ii) identified strengths and weaknesses of project design, iii) principal results including technical and financial analyses, iv) overall scoring of the Project and stakeholders, and v) recommendations and lessons learned. See Annex 7 for a description of discussion. MADVT endorsed the overall results of the ICR including the overall evaluation of the Project as satisfactory. MADVT also endorsed the satisfactory performance rating of the World Bank and highlighted that the World Bank and MADVT worked in partnership in Project design and implementation. Given the overall success of the Project, the PME, and significant effort expended by MADVT in Project design and supervision, MADVT representatives felt that a satisfactory rating was warranted. 4. Assessment of Risk to Development Outcome Policy Risk: The national government promotion and support of the private sector option in Colombia has been instrumental in achieving the Project results. The Colombian government has formerly adopted this policy through the CONPES Document 3385 of 2005, which established that allocation of National Government resources would be linked to processes of business modernization. In addition, the promotion and consolidation of improved operational management is one of the main objectives of the PDA, as adopted in 3463 CONPES documents 2007, Act 1151 of 2007, and Act 1176 of 2007. The risk that in the future the national government will reverse its strong policy support operators is therefore considered low. Contractual Risk: This risk refers to breaches in contract, either by the municipality or the operator, which would result in a cessation of the contract. Most of the operator contracts have experienced multiple amendments and the arrival of new municipal administrations often results in a reexamination of the contractual arrangements. There have already been a number of sub- projects in which the original operator was unable to comply with contract stipulations and has since been replaced with new operators that have demonstrated their capacity to sustain operations. For example, in Magangué the original operator (Aguas Kapital) was not performing well and was replaced with MAVDT facilitation by another (Aquaseo). The contractual risk is therefore considered to be substantial. Financial Risk: This refers to the risk that the water business in a specific municipality may not be profitable over the long-run and thus not sustainable. The financial risk rating is considered substantial. A comprehensive financial analysis of five operators covering 14 municipalities was undertaken and the results are presented in Annex 3. A fundamental indicator is the working ratio, which is also a PDO indicator, is presented below: 15 PDO Indicator 2.E: Water Utility Financial Sustainability Working Ratio Target Working Ratio Standard (Population Weighted) PAD (Average per Municipality) Deviation Baseline 2010 2010 2010 Total Project 0.96 0.84 1.1 0.4 Operation-Investment 0.96 0.87 < 0.6 1.2 0.4 (Medium-Cities) Construction-Operation 0.92 0.77 < 0.6 0.8 0 (Small Cities) For the operation with investment model, three operators were analyzed. The weighted average working ratio decreased slightly to 0.87, which is still significantly above the target of 0.6. In 2010, three largest cities all had 2010 working ratios under 1.0: Soledad, Sincelejo, and Maicao. The average working ratio however is 1.2, which indicates that most operators are struggling financially—particularly the smaller ones. For example, in the five Triple A municipalities in Atlántico that were analyzed, only Soledad had a working ratio under 1.0. In Atlántico, Triple A subsidizes operations in the smaller municipalities for social objectives, primarily because it has sufficient cash flow from its two major operations in Soledad and Barranquilla. Similarly, for Aguas de la Sabana the more lucrative operations in Sincelejo allow it support the smaller neighboring city of Corozal. In general, the operators showed improved operational and commercial performance, as measured by collection rate (85%-96%), micro-metering (> 80%), and falling operating costs. The key driver for financial sustainability appears to be tariff and transfer practices in each city. Tariffs were originally defined in the concession contract, but in most cases the tariff regime has been modified to either follow the national regulatory regime or to annually negotiate tariffs and transfer with the municipal government. In general, operators which have linked their tariff regime to national regulations have achieved, or are close to, full cost recovery levels. In contrast, operators which tariffs linked to the original contract, or negotiate tariffs with the municipality, are generally struggling financially. The most illustrative case of the importance of the tariff regime is the case of operator Aguas de la Sabana. In 2002 when the operator initiated its contract in Sincelejo and Corozal, tariffs in the two municipalities were largely similar. However, Sincelejo has since adjusted tariffs to align with the national regulatory framework while Corozal has not. By 2010, Sincelejo‘s tariffs were twice that of Corozal‘s tariffs. Although both municipalities have decreased their reliance on municipal transfers, the operation in Sincelejo is now profitable, while the operator is struggling financially in Corazal. The overall rating of risk to the Project development objectives is based on the policy risk level (low), the contractual risk level (substantial) and the financial risk level (substantial) and is considered as substantial. 16 5. Assessment of Bank and Borrower Performance 5.1 Bank Performance (a) Bank Performance in Ensuring Quality at Entry The ICR rates Bank performance during preparation as “satisfactory”. The Project design was well aligned with the Government‘s priorities, incorporated the experience from private sector participation both within Colombia and abroad, and helped structure model bidding documents for two types of novel contracts: i) operation with investment; and ii) construction with operation. The Project was designed in a flexible and programmatic manner, which allowed it to evolve over time as necessary. The PAD explicitly identified the main risks and accurately identified the Project risk level as ―high.‘‖ Minor shortcomings during preparation include the somewhat misplaced emphasis on small cities utilizing the ―construction with operation‖ model and the over reliance on model bidding documents which in practice did not prove very effective during implementation—shortcomings that only became evident during project implementation. (b) Quality of Supervision The ICR rates Bank performance during supervision as “satisfactory”. The Project was prepared and supervised by same Bank task team leader until July 2009, until the very closing stages of the Project. The Bank team provided considerable technical support to both the operators and the MAVDT during Project implementation. (c) Justification of Rating for Overall Bank Performance Rating: The ICR rates the overall Bank performance as “satisfactory” given that Bank performance during both preparation and during implementation has been rated as ―satisfactory‖. 5.2 Borrower Performance (a) National Government Performance The Ministry of Economic Development was initially responsible for the project preparation, and the Project was prepared by the project implementing unit for the then on-going Bank financed Municipal Development Project. The Project was approved in October 2001, and the Uribe presidential administration took office at the start of 2002. The new administration merged the Ministry of Economic Development into the newly created MAVDT. In the early years of the Project there was at times insufficient staff and frequent management changes from MAVDT which hindered Project implementation. The formal establishment of a Vice Minister of Water in MAVDT helped to stabilize Project staffing and management, and the Project moved into high gear after 2006. MAVDT provided significant support and supervision for the Project, helping to structure and supervise the concession contracts, entering into agreements with municipal governments for the transfer of Bank funds, and supervising Project implementation, in coordination with the Bank team. 17 The Bank-financed Project was part of the broader MAVDT Corporate Modernization Program (PME) which supported the introduction private operators in an additional 10 municipalities (See Annex 1). Actual counterpart funding was 100% greater than originally forecast, increasing from $15 million to $30 million 4. In addition, the government adopted a series of resolutions and decrees to strengthen private sector participation. Although there were early short comings in the national governments implementation of the Project, these were compensated by the increasing acceleration of implementation and expansion of the Project concept over time. The overall national government performance is therefore rated as “moderately satisfactory.” (b) Implementing Agency or Agencies Performance The Project was primarily implemented by the operators, who were responsible for executing the works, either through construction contracts in the case of the 11 operation with investment contracts or directly in the case of three construction with operation contracts. As expected, the performance of the operators varied considerably with some clear leaders, such as Triple A and Aguas de la Sabana, and obviously ineffective operators such AguasKapital in Magangué. The overall operator performance is judged as satisfactory given the amount of works executed as well the overall improvement in infrastructure quality and service in most subprojects. (c) Justification of Rating for Overall Borrower Performance While the National Government performance was rated as “moderately satisfactory”, the performance of the Implementing Agencies, who were responsible for the bulk of projection execution, was rated as “satisfactory". Therefore the overall borrower performance is rated “satisfactory.” 6. Lessons Learned General Lessons for Colombia’s Water Sector: 1. Specialized (Private) Operators Are the Preferred Approach for Medium-Size Cities: After decades of stagnation in improving water service in medium-sized cities, the Project demonstrated that under the right circumstances a private operator can significantly improve water services. The Colombian experience in the Latin America context where private sector participation is increasingly rare is noteworthy. The key ingredient is national government support, including: an enabling policy context, infusion of financing, and technical assistance coupled with supervision. The Project, however, only represents the beginning of a process in Colombia that has yet to be fully consolidated and has many shortcomings to address. 2. National Government’s Technical Assistance Program Is Critical: The Departmental Water Plans (PDA) represents the second generation national government policy initiative to 4 Refers only to Component A3. See Annex 1. 18 promote and consolidate private operators by channeling resources through and assigning more sector responsibility to the Departments. Nevertheless, there will always be a need for comprehensive and significant national government technical and regulatory support and oversight to departments, municipalities, and operators. This requires an adequate allocation of budget, staffing, and continuous attention by the responsible national government agency. 3. Operator Selection Process Needs to Be Refined: The original Project concept of awarding concession contracts based on the lowest cost subsidy was theoretically sound, but did not work well in practice. Original investments plans were not well formulated, tariff regimes not maintained, and actual subsidy levels varied widely. The quality of the operator is the most important factor, and in practice this has little to do with which bidder offers the lowest subsidy requirement during bidding. An alternative approach would be to select operators based on experience and quality (similar to consultant selection) rather than primarily on financial considerations (such as a civil works contractor). 4. Need to Better Incorporate Technical Dimensions into Performance Goals and Investment Plans: A number of common technical issues were encountered which hindered improvements in service, including: inadequate raw water supply sources, frequent power outages, high levels of unaccounted for water, low levels of micro-metering, and lack of attention to network sectorization. Sophisticated water utilities (such as Triple A) are usually able to address these technical challenges, but most of the smaller operators are overwhelmed by these problems. In some cases, unrealistic performance targets were set without a clear plan to resolve these issues. There is a need to improve technical assistance to the operators and municipalities. 5. Must Guarantee that Pledged Subsidies Are Provided to the Operator: Subsidies are provided to both support investments and to cover below-cost tariffs for low-income strata in Colombia. The Project established clear mechanisms for channeling national investment subsides through independent fiduciaries that ensure funds go directly to investment project (as opposed to being diverted by municipal governments). Operators, however, faced a constant challenge in ensuring that municipal governments provided their agreed-upon subsidies. This issue is being slowly addressed through a number of national government decrees, but direct negotiations between operators and mayors is still the norm and can potentially undermine the financial sustainability of the operator. 6. Utilization of National Tariff Regime for Private Operators Is Preferred. The ICR analysis shows that when tariffs are set according to national regulatory framework administered by the CRA, cost recovery improves, dependence on municipal transfers decreases and financial results are enhanced. In contrast, tariff regimes formulated at the start of the concession contract were often based on inaccurate investment and operational assumptions that resulted in operator under-funding. Likewise, tariffs which are based on direct negotiations with municipal governments are often disadvantageous to operators and based on political considerations rather than actual financial needs. 7. Strategy for Small Municipalities Still Lacking: The construction-operation model for small municipalities (less than 12,000) did not perform well under the Project and the long-term 19 sustainability of the contracts is uncertain. There is a need to develop new approaches for serving this segment of the population. 8. Wastewater and Stormwater Drainage Policies and Programs Need to Be Developed: The expansion of sewerage coverage under the project was minimal and none of the Project municipalities have wastewater treatment. Although an initial focus on water supply is understandable, the problems associated with collection and treatment will need to be addressed in the future, requiring new policies, strategies and investments. None of the municipalities have storm-water drainage systems, resulting in significant damage to roads and economic losses during storms. There is a need to develop clear national government policies and strategies to address these pressing issues. 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners (a) Borrower/implementing agencies: MADVT representatives objected to the national government performance rating of moderately satisfactory, noting that the delay in the final audit report was a relatively minor issue. (b) Cofinanciers N/A (c) Other partners and stakeholders : N/A (e.g. NGOs/private sector/civil society) 20 ANNEXES 21 Annex 1: Project Costs and Financing (a) Project Cost by Component (in USD Million equivalent) Actual/Latest Appraisal Estimate Percentage of Components Estimate (USD millions) Appraisal (USD millions) Investments in subprojects in A. medium size cities and small 61.80 75.55 122% municipalities Loan investments in W&S A1. works in medium size cities in 15.0 24.06 160% the Caribbean region Loan investments in W&S A2. works in small municipalities in 16.8 5.63 34% the Caribbean region Government counterpart funds for investments in medium and A3. 15.0 29.81 199% small municipalities in the rest of the country Private Sector and Law 60 A4. investments in medium size 12.5 12.94 104% cities Private Sector and Law 60 A5. investments in small 2.5 3.11 124% municipalities Sector environmental B. management capacity 0.80 0.38 48% strengthening Rural water and sanitation C. 0.90 0.21 23% policy development Project management and D. 6.5 5.40 83% training Project Management and D1. 2.5 2.5 100% Training TA for PSP in medium size D2. 1.0 1.0 100% cities and supervision of works TA for PSP in small D3. municipalities and supervision 3.0 1.9 63% of works Total Project Cost 70.00 81.54 116% Front-end fee PPF 0 0 -- Front-end fee IBRD 0.4 0.4 100% Total Financing Costs 70.40 81.94 116% 22 (b) Financing Appraisal Actual/Latest Type of Percentage of Source of Funds Estimate Estimate Cofinancing Appraisal (USD millions) (USD millions) Borrower (Government of Colombia) 30.00 45.86 153% International Bank for Reconstruction 40.00 35.68 89% and Development Total Financing Required 70.00 81.54 116% 23 Sub-Project Costs and Financing Sources SUB-PROJECT A1 A2 A3 A4 A5 TOTAL •World Bank Financing ($IBRD) ($IBRD) ($COUNTERPART) ($PRIVATE) ($PRIVATE) ($USD) Ο Counterpart Financing • El Banco 2,324,602 1,743,594 4,068,196 • Soplaviento 2,499,951 2,499,951 Santa Rosa Villanueva San Estanislao De Kostka • Magangue 1,159,031 1,159,031 • Calamar – Mahates 1,639,991 1,639,991 • Arjona y Turbaco 4,207,674 4,594,917 1,158,485 9,964,076 • Baranoa yPolonuevo 1,955,684 833,488 993,585 3,782,757 • Cereté 5,850,637 4,786,398 2,678,321 13,315,356 Sahagún Ciénaga De Oro San Carlos • Maicao 3,532,642 517,986 160,668 4,211,297 • San Juan de Nepomuceno 425,747 461,814 742,017 1,629,578 • San Marcos 1,160,194 171,262 1,070,230 2,401,686 • San Onofre 1,494,974 922,691 206,180 2,623,844 • Santo Tomas 311,550 437,736 42,908 792,195 Sabanagrande • Sincelejo 1,956,096 5,777,087 2,459,710 10,192,892 Corozal • Soledad 883,804 610,277 1,606,596 3,100,677 Ο Agua de Dios 436,500 2,174,740 2,611,239 Tocaima Ο Carmen de Bolivar 1,013,279 1,013,279 Ο Cumaral 611,039 466,642 1,077,681 25 Ο Nataga 206,108 70,405 276,513 Ο Ponedera 449,100 119,593 568,693 Ο Pto Carreño 187,676 71,137 258,813 Ο Riohacha 1,882,042 1,882,042 Ο San Andres 4,972,523 282,189 5,254,712 Ο San Jacinto 383,116 383,116 Ο Tunja 574,111 574,111 TOTAL 24,063,296 5,634,916 29,809,751 12,938,302 3,108,696 75,554,961 26 Annex 2: Project Performance Indicators The following table presents only those private sector contracts signed under World Bank financed sub-projects. Supervision of these contracts was undertaken by a task team comprised of the World Bank and Vice-Ministry of Water. The PDO targets were at least 1 medium city with PSP by mid-term and at least 2 by closing date, and at least 5 small municipalities with PSP by mid-term and at least 15 by closing date. PDO Indicator 1: PSP incorporated in medium size cities and small municipalities Operator Municipality Contract Contract Contract Initiation Duration Model Operagua el Banco El Banco 2003 16 years Operation with Investment Giscol Soplaviento Constructor – Santa Rosa Operator Villanueva 2008 18 years San Estanislao de Kostka Calamar 2009 15 years Constructor – Mahates Operator Aquaseo Magangué 2007 18 years Operation with Investment Acualco Arjona 2004 20 years Operation Turbaco with Investment Triple A Baranoa 2005 20 years Operation Polonuevo with Investment Sabanagrande 2005 20 years Operation Santo Tomas with Investment Soledad 2002 20 years Operation with Investment Uniaguas Cereté Operation Sahagún with Ciénaga De Oro 2004 20 years Investment San Carlos Aguas de la Peninsula Maicao 2001 16 years Operation with Investment Aguas de la Costa San Juan De Nepomuceno 2002 10 years Operation with Investment Aguas de la Mojana San Marcos 2002 15 years Operation with Investment 27 Insergrup San Onofre 2006 15 years Constructor – Operator Aguas de la Sabana Sincelejo 2003 20 years Operation Corozal with Investment 28 PDO 2: Expand the coverage of water and sewerage services and efficiency of operators in participating municipalities For the purposes of project evaluation and preparation of the ICR, PDO 2 (Expand the coverage of water and sewerage services in participating municipalities) and PDO 3 (Facilitate the access of population in low incomes areas of participating municipalities to water and sewerage services) defined in PAD were merged. The information provided below indicates the proportion of population in each municipality officially categorized as ‗poor‘, with an average of 94% across all participating municipalities. Data is based on official statistics provided by DANE (Departamento Nacional de Estadística) and operators. Department Municipality Number of People in % Strata 1,2,3 Service Area in Service Area Magdalena El Banco 34,397 90% Bolívar Soplaviento 8,095 100% Santa Rosa 13,299 100% Villanueva 16,947 100% San Estanislao de Kostka 11,399 100% Magangué 83,504 93% Calamar 18,369 94% Mahates 13,396 99% Arjona 51,937 95% Turbaco 62,069 92% San Juan de Nepomuceno 25,135 98% Atlántico Baranoa 46,040 98% Polonuevo 12,017 98% Santo Tomás 23,718 97% Sabana Grande 27,657 97% Soledad 541,133 98% Córdoba Cereté 68,789 92% Sahagún 46,162 92% Ciénaga de Oro 26,410 98% San Carlos 8,064 98% La Guajira Maicao 141,917 92% Sucre San Marcos 31,517 97% San Onofre 19,780 95% Sincelejo 292,290 87% Corozal 43,128 96% 29 PDO Indicators 2.A and 2.B: Evolution of Coverage for Water Supply and Sewerage Infrastructure5 Operator Municipality Water Supply Water Sewerage Sewerage Baseline Supply 2010 Baseline 2010 Operagua el Banco El Banco 39.0% 94.5% 26.0% 36.7% Giscol Soplaviento 81.0% 86.2% 0.0% 0.0% Santa Rosa 0.0% 0.0% 0.0% 0.0% Villanueva 41.6% 44.0% 0.0% 0.0% San Estanislao 93.0% 92.1% 0.0% 0.0% de Kostka Calamar 77.6% 81.6% 0.0% 0.0% Mahates 70.5% 72.4% 0.0% 0.0% Aquaseo Magangué 52.5% 63.9% 17.1% 17.5% Acualco Arjona 60.7% 72.9% 0.0% 0.0% Turbaco 25.2% 63.5% 0.0% 0.0% Triple A Baranoa 64.5% 96.0% 0.0% 0.0% Polonuevo 78.0% 91.9% 0.0% 0.0% Sabanagrande 87.1% 92.8% 77.2% 78.7% Santo Tomas 50.8% 88.1% 42.4% 60.7% Soledad 66.0% 96.2% 60.2% 71.1% Uniaguas Cereté 69.2% 90.6% 36.4% 41.6% Sahagún 74.8% 84.0% 64.5% 71.6% Ciénaga De 68.6% 88.5% 30.3% 37.2% Oro San Carlos 0.0% 92.5% 0.0% 0.0% Aguas de la Maicao 41.4% 67.0% 34.3% 30.0% Peninsula Aguas de la Costa San Juan De 39.1% 43.5% 0.0% 0.0% Nepomuceno Aguas de la San Marcos 68.8% 99.2% 0.0% 0.0% Mojana Insergrup San Onofre 84.3% 82.9% 25.5% 28.7% Aguas de la Sabana Sincelejo 73.4% 83.3% 66.5% 81.9% Corozal 73.7% 82.6% 61.6% 77.9% 5 The baseline indicators of each subproject correspond to the start of the concession contract. See PDO Indicator 1 for contract starting year and duration. 30 PDO Indicator 2.C: Evolution of Coverage of Continuous Water Supply Operator Municipality Continuous Intermittent Supply7 Supply6 Baseline Final Baseline Final Operagua El Banco 0.0% 0.0% No data 8 hours per days/7days el Banco a week Giscol Soplaviento 0.0% 0.0% No data 8 hours per days/7days a week Santa Rosa 0.0% 0.0% No data 3 water tanker deliveries per week Villanueva 0.0% 0.0% No data 12 hours per day/7days a week San Estanislao 0.0% 0.0% No data 8 hours per day/7days de Kostka a week Calamar 0.0% 0.0% 16 hours per day 14 hours per day/7days (urban center) /7days a week (urban center) a week Mahates 0.0% 0.0% 4 Sectors: 12 hours per day/7days 8 hours day /2 days a week (urban center) per week Aquaseo Magangué 0.0% 0.0% 5-6 hours per 3 Sectors: 11.5 hours day/7days a week per day/7days a week Acualco Arjona 0.0% 0.0% 2 days per week 22 hours per day, /7days a week Turbaco 0.0% 0.0% 2 days per month 2 Sectors: 24 hours per day, /7days a week 3 Sectors: 24 hours per day, 1 day per week Triple A Baranoa 0.0% 96.0% 19.4 hours per 24 hours per day/7days day/7days a week a week Polonuevo 0.0% 91.9% 12 hours per 24 hours per day/7days day/7days a week a week Sabanagrande 87.1% 92.8% 12 hours per 24 hours per day/7days day/7days a week a week Santo Tomas 0.0% 88.1% 12 hours per 24 hours per day/7days day/7days a week a week Soledad 0.0% 96.2% 12 hours per 24 hours per day/7days day/7days a week a week 6 Defined as 24 hour per day/7 days per week 7 Source: Operator information provided to MAVDT 31 Uniaguas Cereté 0.0% 90.6% 6 hours per day/7days 24 hours per day/7days a week a week Sahagún 0.0% 84.0% 1.5 hours per 24 hours per day/7days day/7days a week a week Ciénaga De Oro 0.0% 0.0% 2 hours per day/7days 6 hours per day/7days a week a week San Carlos 0.0% 92.5% 2 hours per day/7days 24 hours per day/7days a week a week Aguas de la Maicao 0.0% 29.0% No Data 24 hours per day/7days Peninsula a week Aguas de la San Juan De 0.0% 0.0% No Data 7 Sectors: 1 day per Costa Nepomuceno week Aguas de la San Marcos 0.0% 0.0% 2.5 hours per day/ less 6.5 hours per day, Mojana than 7 days per week /7days a week Insergrup San Onofre 0.0% 0.0% 2 Sectors: 2 Sectors: 14 hours per 12 hours per day/1 day, /7days a week day per week Aguas de la Sincelejo 0.0% 14.0% 14 Sectors: None with 1 Sector: 24 hours per Sabana 24 hour service day (Pilot Sector; 17% of population) Corozal 0.0% 29.7% 5 Sectors: 6-7 hours 5 Sectors: ~15 hours per day/7days a week per day 1 Sector: 24 hours per day 32 PDO Indicator 2.D and 2.E: Efficiency and Sustainability of Service Operator Municipality Staff per 1,000 Working Ratio Connections Baseline Final Baseline Final Operagua el Banco El Banco 3.0 2.2 - - Giscol Soplaviento 6.7 7.2 0.92 0.77 Santa Rosa Villanueva San Estanislao de Kostka Calamar 3.4 4.7 - - Mahates - - Aquaseo Magangué 4.7 4.2 - - Acualco Arjona No Data 3.5 1.45 0.87 Turbaco 0.87 0.94 Triple A Baranoa 9.7 5.5 1.55 1.37 Polonuevo 2.49 1.54 Sabanagrande 11.8 6.9 1.89 1.40 Santo Tomas 2.20 1.78 Soledad 13.6 7.8 0.77 0.80 Uniaguas Cereté 4.1 4.0 - - Sahagún - - Ciénaga De Oro - - San Carlos - - Aguas de la Peninsula Maicao No Data No Data 0.69 0.97 Aguas de la Costa San Juan De 6.8 4.9 - - Nepomuceno Aguas de la Mojana San Marcos 1.8 2.7 - - Insergrup San Onofre 1.8 4.4 Aguas de la Sabana Sincelejo 4.6 4.7 0.86 0.63 Corozal 1.06 1.47 33 Water Quality Monitoring8 The issue of water quality was captured through a broader indicator related to disinfected supply which was interpreted to be intended to capture data of water potability in participating municipalities. Given that there was no indicator specified in the PAD that related directly to water quality, it was not included as part of the evaluation of the PDOs. Monitoring of this indicator was initiated, however, in 2007 by the National Institute of Health through the Risk Index of Water Quality for Human Consumption (IRCA) which defines the physical, chemical and bacteriological quality of water through monthly samples. The most recent data collected (December 2010) through the IRCA for each of the participating municipalities is presented below. Operator Municipality Water Quality Level of Water (December 2010) Quality Risk Operagua el Banco El Banco 8.9% Low Giscol Soplaviento 46.7% High Santa Rosa 16.6% Medium Villanueva 32.1% Medium San Estanislao de Kostka 33.5% Medium Calamar 47.8% High Mahates 28.5% Medium Aquaseo Magangué 15.1% Medium Acualco Arjona 10.0% Low Turbaco 11.0% Low Triple A Baranoa 2.2% Low Polonuevo 2.8% Low Sabanagrande 0.0% Low Santo Tomas 1.4% Low Soledad 0.8% Low Uniaguas Cereté 0.6% Low Sahagún 0.0% Low Ciénaga De Oro 0.0% Low San Carlos 0.0% Low 8 The IRCA defines water quality risk as the following: (i) 5.1%-14% (Low), (ii) 14.1%-35% (Medium), (iii) 35.1%- 80% (High), (iv) 80.1%-100% (Unsanitary). 34 Aguas de la Peninsula Maicao 19.7% Medium Aguas de la Costa San Juan De 38.2% High Nepomuceno Aguas de la Mojana San Marcos 2.8% Low Insergrup San Onofre 18.0% Medium Aguas de la Sabana Sincelejo 1.0% Low Corozal 5.6% Low 35 Annex 3: Economic Analysis: Evaluation of Investment Efficiency9 This Annex presents the engineering economic analysis of selected municipal water and sewerage systems comprised of a qualitative assessment of the efficiency of investments in regards their effect of increasing service. To evaluate the efficiency of investments a four point scale was applied as defined below: Investment Efficiency Criteria Negligible Refers to those investments that have not generated a real impact on the conditions of service to users and have not positively influenced performance indicators. Example: Extending network coverage with limited production capacity Moderate Refers to those investments that have generated a limited impact on the conditions of service, but have been defined through inappropriate prioritization criteria. Example: Construction of storage tanks as part of an operation characterized by intermittent supply. Substantial Refers to those investments that have generated a modest impact on the conditions of service, even if they were defined with appropriate prioritization criteria. Example: Expansion of treatment capacity in a system whose source is subject to intermittent electricity. High Refers to those investments that have generated a positive impact on the conditions of service, and are the outcome of proper planning and prioritization. Example: Installation of back-up power in a system reliant on pumping stations. 9 The engineering economic analysis presented in this Annex is based on the professional findings of a recognized expert in Colombian water supply systems with extensive experience and familiarity of water systems along the country‘s Caribbean coast. Juan Camilo Gil works for ACUAGEST S.A.S. 37 Evaluation of Investment Efficiency Operator Municipality Evaluation of Investment Efficiency Ranking Giscol Soplaviento Weaknesses were found in the planning of the works defined in the Negligible structuring of the project, particularly when taking into account uncertainties of water availability and costs of selling water in bulk Santa Rosa from the Cartagena system to meet demand in Santa Rosa. There was limited perception of the need to strengthen operational management in Villanueva addition to investments in system capacity. In specific no investments had been made to ensure continuity of energy needed by the system San Estanislao de resulting in a high degree of intermittency. Kostka Acualco Arjona Investments to date have undoubtedly strengthened the system and Substantial reduced vulnerabilities in some of its major components. However important investment requirements have lagged especially in relation to Turbaco storage and power supply which would significantly enhance the continuity of service. Triple A Baranoa The investment approach followed by the operator of strengthening High production infrastructure to ensure a continuous supply to the population produced a visible impact on service conditions, and Polonuevo appropriately put the system‘s virtually unlimited source and installed capacity to use. Sabanagrande The prioritization of works reflects the perception of the operator on the High need to optimize the principal components of the system in such a way that would generate rapid impact on the continuity of service. In the Santo Tomas absence of challenges relating to its source (Rio Magdalena) the operator has focused on improving the capacity of each of the major system components particularly intake, conduction, treatment and distribution which has generated tacit acceptance of installing household meters. Soledad Appropriate prioritization of works defined in the short term action plan High was aimed at improving the lines of the primary distribution network from Barranquilla, optimization of distribution networks and 38 implementation a macrometering system. This prioritization allowed significant improvements in service continuity, minimum pressures and water quality. The combination of these outcomes has resulted in service provision 24 hours 7 days a week. Aguas de la Maicao The level of investments in the municipal water system have been Modest Peninsula considerable, however the net effect on service conditions has been limited particularly in reference to continuous supply. The operator has been focused on investing in works relating to system optimization but with limited clarity on the associated requirements of operational management. There has also been a prioritization of investments in sanitation rather than in water supply, despite awareness of the existing limitations of service. Insergrup San Onofre In response to energy intermittency, the operator has not focused on Negligible energy alternatives that would facilitate the operation and continued improvement of continuity. Investments in water supply have focused on strengthening production components (rehabilitation of wells, pumping stations, and the expansion of semi-buried storage tanks) as well as optimizing the distribution network. This approach has strengthened the capacity which was necessary to improve conditions of service. However it is worth noting that in the prioritization of investments, major works were left out such as optimizing or replacing the conduction line from the tank to the distribution network, the primary distribution network configuration and the sectorization of networks with macrometering, which would have positively impacted conditions of service, particularly the continuity of supply. Aguas de la Sincelejo Based on system dependency of groundwater at depths of between 600- High Sabana 900 meters, investments have focused on improving installed 39 Corozal production capacity through drilling of new wells, pumping stations, with corresponding transmission lines, construction of new chlorination systems, storage tanks and distribution lines. However investments aimed at improving the operational control of the system, particularly in regards to the sectorization of networks, macrometering and pressure control are still incipient. 40 Annex 4: Financial Analysis The PAD stated that ―the proposed reform program will be successful only if it is structured to attract potential investors and lenders. In order to ensure success, the proposed PSP structures must allow sufficiently attractive rate of return to potential investors and provide adequate financial comfort to lenders‖. The ICR therefore assessed efficiency of the Project based on a financial analysis of 14 operators, representing 67% of all World Bank financed subprojects. To examine how successful the project was a financial analysis of the service as well as of the operators was carried out. This analysis was based on: the utilities‘ audited historical and pro-forma financial statements from the year the contract started to 2009, the non-audited financial statement of 2010, and the projected financial conditions to the end of the contract. The following indicators were measured: (i) estimated net present value and internal rate of return from operation during the contract lifetime; (ii) estimated net present value of shareholders‘ cash flow and internal rate of return during the contract lifetime; (iii) cost recovery through tariffs and fund transfers from Municipalities; and (iv) financial performance of the operation. The evaluation is complemented with a sensitivity analysis, which shows room the operators have for additional improvements and returns. The financial results from the operation were analyzed based on historical results of the operation complemented with the expected results for the remaining time of the contract. The revenue stream of the operation was estimated for each municipality when information was available; while the revenue stream of shareholders was estimated for each operator, which in a number of cases served more than one municipality. Results The results of the analysis indicate that all operators have realized improvements since the initiation of their contracts, yet a number are still showing losses. Improvements are most evident in the working ratio, which was selected in the PAD as one of the performance indicators, and measures the relation between operating costs (without depreciation) and operating revenues. Most operators have decreased their working ratio, however, only one (Giscol) reached the expected target. Upon further analysis arising from known problems in service provision in this specific case, the principal driver for having reached the working ratio target is the contractual model (constructor -operator) which is designed to allow the operator to generate revenues not only from the operation, but more importantly from the construction of civil works. Working Ratio PAD Years of P&L Summary First year 2009 or 2010 operation Target Complies? operation Atlántico Triple A Atlantico Baranoa 1.55 1.37 No 6 < 0.6 Polonuevo 2.49 1.54 No 41 Sabanagrande 1.89 1.40 No Santo Tomas 2.20 1.78 No Subtotal 1.92 1.50 No Soledad (Triple A Barranquilla) 0.77 0.80 >6 < 0.6 No Bolivar GISCOL (LA LINEA ) 0.92 0.77 3 < 0.9 Yes ACUALCO: Arjona 1.45 0.87 5 < 0.7 No Turbaco 0.87 0.94 5 < 0.7 No Arjona&Turbaco 1.19 0.91 5 < 0.7 No Sucre INSERGRUP (San Onofre) 0.94 0.90 4 < 0.8 No ADESA: Sincelejo 0.86 0.63 7 < 0.6 No Corozal 1.06 1.47 7 < 0.6 No Sincelejo&Corozal 0.89 0.72 7 < 0.6 No La Guajira Maicao 0.69 0.97 9 <0.6 No In assessing the aggregated financial performance of the operators with contracts in multiple municipalities it is of relevance to note that all but one show positive results, with net margins ranging between 5% and peaking as high as 22%. Yet when disaggregated at the municipal level only four of the fourteen operators chosen for the sample show positive net income, with losses as high as 91% of revenues. Importantly, the reason for these differences is that operations in some municipalities are profitable and therefore compensate the losses of neighboring municipalities served by the same operator. The variability of results can largely be explained by differences in tariffs, transfers from municipalities, operating costs and commercial performance. Operating margin Net margin P&L Summary First year First year 2,009-2010 2,009-2010 operation operation Atlántico Triple A Atlantico Baranoa -55% -37% -56% -38% Polonuevo -149% -54% -151% -56% Sabanagrande -89% -40% -90% -40% Santo Tomas -120% -78% -121% -79% Subtotal -92% -50% -92% -50% Soledad (Triple A Barranquilla) 23% 20% 5% 5% Bolivar GISCOL (LA LINEA) 8% 23% 1% 22% ACUALCO: Arjona -45% 13% -59% 5% 42 Turbaco 13% 6% 5% -3% ACUALCO (Arjona&Turbaco -19% 9% -30% 1% Sucre INSERGRUP (San Onofre) 6% 10% 0% 0% ADESA: Sincelejo 14% 37% -3% 19% Corozal -6% -47% -26% -91% ADESA (Sincelejo&Corozal) 11% 28% -6% 6% La Guajira Maicao 31% 3% 2% 2% Tariffs and transfers were initially defined in the respective operation contracts, however with appropriate amendments some municipalities and operators have since either agreed to adjust tariffs to align with the national regulatory framework, or to annually negotiate tariffs and corresponding transfers for subsidies provided to low income customers. Depending on these agreements the sharing of both sources of revenues (tariffs and transfers), and the resulting cost recovery vary widely among operators. In general those that have adjusted tariffs to the regulatory framework have achieved or are close to achieving full cost recovery, have decreased dependence on transfers from the municipality, and have improved their financial results. Meanwhile, those with tariffs and transfers negotiated on a yearly basis or with tariffs that remain linked to the operation contract, have neither reached full cost recovery, nor achieved positive returns. 2010 Tariffs according Share on revenues of to Regulatory Operation Cost transfers (from the Framework? yields positive Recovery Nation or the returns? municipalities) Atlántico Triple A Atlantico Baranoa No 59% No 19% Polonuevo No 55% No 15% Sabanagrande No 61% No 14% Santo Tomas No 46% No 19% Subtotal 56% No 18% Soledad (Triple A Barranquilla) Yes 91% Yes 19% Bolivar GISCOL (LA LINEA) No 41% Yes 68% ACUALCO: Arjona Yes 85% Yes 27% Turbaco No 69% No 35% ACUALCO Arjona&Turbaco 77% 31% Sucre INSERGRUP (San Onofre) No 50% Yes 53% ADESA: 43 Sincelejo Yes 120% Yes 10% Corozal No 54% No 5% ADESA (Sincelejo&Corozal) 105% 9% La Guajira Yes if under the Maicao ceiling 65% Yes 37% These differences have resulted in important variations in tariffs among some neighboring municipalities which share same operator, same water production system, and same socioeconomic conditions. When ADESA initiated its contract in Sincelejo and Corozal in 2002, for example, tariffs in the two municipalities were largely similar. However, Sincelejo has since adjusted tariffs to align with the national regulatory framework while Corozal did not and by 2010, Sincelejo‘s tariffs were twice that of Corozal‘s. Although both municipalities have decreased their reliance on municipal transfers, the operation in Sincelejo is now profitable, while Corozal has increased its operational losses and also decreased cost recovery. Share of Funds transferred by Cost Recovery through tariffs Municipalities on Revenues P&L Summary First year of First year of 2009 2,009 operation operation Atlántico Triple A Atlantico Baranoa 56% 59% 13% 19% Polonuevo 36% 55% 11% 15% Sabanagrande 39% 61% 25% 14% Santo Tomas 36% 46% 21% 19% Subtotal 42% 56% 16% 18% Soledad (Triple A Barranquilla) 89% 91% 23% 19% Bolivar GISCOL (La Linea) 29% 41% 74% 68% ACUALCO Arjona 19% 85% 36% 27% Turbaco 20% 69% 59% 35% Arjona&Turbaco 19% 77% 47% 31% Sucre INSERGRUP (San Onofre) 13% 50% 88% 53% ADESA: Sincelejo 73% 120% 20% 10% Corozal 60% 54% 20% 5% Sincelejo&Corozal 70% 105% 20% 9% La Guajira Maicao 46% 65% 55% 37% 44 Operational and commercial performance has generally improved through improved revenue collection rate, micro metering, and operating costs. In a number of municipalities for example, the revenue collection rate is between 85% - 96% and household metering has increased from to levels higher than 80% in most municipalities. Importantly, notable improvements in service continuity has generated tacit acceptance of installing household meters. While operation costs per cubic meter produced have decreased in most cases analyzed, there is still room for improvements particularly through reductions in unaccounted for water and gains in efficient energy usage. Revenue Collection Rate Micro-metering Commercial Indicators First year of operation 2010 2010 Atlántico Triple A Atlantico Baranoa 29% 95% 93% Polonuevo 23% 93% 96% Sabanagrande 47% 81% 87% Santo Tomas 38% 94% 96% Subtotal 39% 90% 90% Soledad (Triple A Barranquilla) n.a 95% 85% Bolivar GISCOL (La Linea) 18% 26% n.a ACUALCO Arjona* 47% 111% 61% Turbaco 32% 77% 34% Arjona&Turbaco 40% 95% 56% Sucre INSERGRUP (San Onofre) 14% 48% 15% ADESA: Sincelejo 62% 91% 79% Corozal 62% 91% 86% Sincelejo&Corozal 62% 91% 80% La Guajira Maicao 0% 16% 39% *In Arjona the 2010 revenue collection rate is higher than 100% because it includes accounts payable in arrears. The capital structure of the operators is also revealing in regards to how each company is financing its overall operations and provides insight into its comparative level of risk exposure. Using the balance sheets of the operators results show that both Triple A and Giscol present sound debt-equity ratios, with 40% or more of their assets financed with equity. However in Giscol this result could be explained by the Government transfers received in anticipation of the works to be implemented, and therefore the situation is not that sounds as it appears; no further analysis could be made due to lack of information. 45 The other utilities though show deteriorated debt-equity ratios, with high indebtedness levels up 95% in the case of Insergroup (San Onofre) and 87% in Acualco (Turbaco and Arjona). Utility Capital structure Has the utility a high First year 2,009 indebtedness level in 2009? Triple A Barranquilla Short term liability 21% 32% Long term liability 40% 26% No Equity 38% 42% Triple A Atlantico Short term liability 71% 49% Long term liability 0% 0% No Equity 29% 51% GISCOL (La Linea) Short term liability 16% 26% Long term liability 66% 34% No Equity 17% 40% ACUALCO (Turbaco and Arjona) Short term liability 23% 12% Long term liability 30% 75% Yes Equity 47% 13% INSERGRUP (San Onofre) Short term liability 30% 91% Long term liability 63% 4% Yes Equity 7% 5% ADESA (Sincelejo and Corozal) Short term liability 13% 40% Long term liability 0% 32% Close to high Equity 87% 28% AGUAS DE LA PENINSULA (Maicao) Liability 39% 48% Equity 61% 52% No Financial returns were evaluated for the provision of service (by municipality) and the operator as a whole, the net cash flow was estimated using actual values from the year of contract initiation to 2009, and projected values until the end of the contract. For the projection it was assumed that all indicators shown in 2009 remained constant, that is, service coverage, operating costs, tariffs, municipalities‘ transfers, and unaccounted for water. Tariffs and transfers from municipalities were projected based on 2010 values adjusted only by inflation. The net cash flow of the each municipal service was estimated without financing. The net cash flow for the operator was estimated using existing practices for dividend distribution. Cash flows were transformed to 2009 prices to show results in real terms without inflation. 46 The results show that the operation in Soledad (Triple A), La Linea (Giscol), and Sincelejo (ADESA) will yield positive returns assuming current conditions prevail. However, as previously noted the returns obtained by Giscol are based on payments received for construction of works rather than the operation through improved service provision. In Soledad and Sincelejo, although improvements could be achieved and therefore improved financial returns obtained, existing returns have come based on service improvements and efficiency gains. Base Case Scenario Utility Operator NPV cash flow (2009 COP NPV cash flow million) IRR (2009 COP million) IRR Atlantico Triple A Atlantico (35,399) n.a Baranoa (5,416) n.a Polonuevo (2,147) n.a Sabanagrande (6,830) n.a Santo Tomas (8,190) n.a Subtotal (22,582) n.a Soledad (Triple A Barranquilla) 16,381 10,511(1) Bolivar GISCOL (La Linea) 522 308 72% ACUALCO: Arjona (3,830) n.a Turbaco (4,549) n.a ACUALCO (Arjona and Turbaco) (8,379) n.a (10,074) n.a Sucre INSEGRUP (San Onofre) (685) n.a (571) n.a ADESA: Sincelejo 13,167 27% Corozal (17,552) n.a ADESA (Sincelejo & Corozal (4,385) 4% (5,590) 6% La Guajira AGUAS DE LA PENINSULA (Maicao) (2,860) n.a (3,044) n.a (1) In the case of Soledad, the returns for the operator correspond to the present value of the annual payment shareholders are getting from the operation, which corresponds to 4.5% of revenues . This amount is registered as an administrative cost of the operation. Capital and Dividends were not estimated due to lack of information. A sensitivity analysis was conducted to measure the impact on returns. The chosen variables for this analysis were: tariffs, municipalities‘ transfers, and operating cost. The analysis was done for those operators and utilities whose returns were lower than 10% in the base case scenario. The results were varied indicating that some municipalities remain very sensitive to small changes in the selected variables, while in others the required changes are significant as is the case of Corozal. 47 Required changes of one of the following variables to get 10% return RETURNS FOR THE SHAREHOLDER Increase in Increase Decrease on Municipalities’ Tariffs Operating cost transfer Atlántico Triple A Atlantico 173% 1000% 94% Soledad (Triple A Barranquilla) Bolivar ACUALCO (Arjona & Turbaco) 75% 262% 38% Sucre INSEGRUP (San Onofre) 31% 36% ADESA (Sincelejo & Corozal) 23% 790% 60% La Guajira AGUAS DE LA PENINSULA (Maicao) 38% 154% 25% In the case of ADESA (Sincelejo and Corozal), the sensitivity analysis was carried out applying the changes only in Corozal, given that Sincelejo is already showing high returns. Required changes on one of the following variables to get 10% return RETURNS FOR THE UTILITIES Increase in Increase Transfer by Tariffs municipalities Atlántico Triple A Atlantico Baranoa 131% 573% Polonuevo 208% 1157% Sabanagrande 129% 798% Santo Tomas 251% 1122% Subtotal 153% 831% If variables change at same time 100% 100% Bolivar ACUALCO (Arjona and Turbaco) 93% 321% Arjona 84% 302% Turbaco 97% 270% Sucre INSERGRUP (San Onofre) 31% 35% ADESA Sincelejo Corozal 251% 4876% If variables change at same time 190% 100% ADESA (Sincelejo & Corozal) 65% 1141% La Guajira AGUAS DE LA PENINSULA (Maicao) 46% 185% In the case of Sincelejo and Corozal, the sensitivity analysis was carried out applying the changes only in Corozal, given that Sincelejo is already showing high returns. 48 Lessons learned from the financial analysis A. The contracts should provide incentives to the operators for improving the service and financing as much as possible the operation and the required investment. In the financial analysis of the sample of the utilities some flaws were noticed, which can be addressed in the new contracts. B. Current contracts can allow operators to obtain profits even when they provide poor service. Such is the case of construction-operation contracts, which gives more incentives for doing the construction than the operation. Same with the concession contracts such as the one in Maicao, which provides more incentive in supervising the works than doing the operation. The new contracts should focus on operation with investment, paying more attention on incentives for efficiency improvement. C. The financial analysis of the sample of utilities showed that when tariffs were set according to National Regulatory Framework, cost recovery improves, dependence on municipal transfers decreases, and financial results improves. When tariff reflects the cost of providing the service, and this is accompanied by commercial efficiency, the customers get the right signal for water efficient usage. If this is the case, the transfers from the municipalities would be used not only to cover operating costs just to compensate for subsidies of poor income customers, but also to partially cover investment. New contracts should promote the design of tariffs according to the regulatory framework and established targets that give incentive for efficiency improvements. D. Some of the municipalities have been subsidizing others; such is the case of Sincelejo with Corazol. This can put in risk the whole operation. The new contracts should allow the financial balance on the operation of each of the municipalities included in the contract. Tariffs and amount of transfers should be measured in detail to avoid deficit in one municipality to be covered by another municipality. 49 Annex 5: Specific Lessons for the Corporate Modernization Program10 The following points highlight some of the lessons learned in the development and selection of specialized operators under the PME program. Programming of the Resources Contributed by the Nation: Structuring of the early concession contracts underestimated the time needed for processing national government contributions, and as a result some operators received resources much later than expected, thereby reducing service improvements expected of the new operators. Write-Off of Pre-Existing Liabilities: In some processes, the liabilities of the old public companies, prior to the arrival of the private operators, were not covered and the responsibility was delegated to the municipalities. This generated the risk of labor unrest as workers made legal claims against the municipality. In one case, the start date was delayed for almost two years because the electric power company did not provide service until the municipality paid the arrears. Bid Preparation Time: In most processes, short periods were assigned for bid preparation considering the agreements have durations of ten or more years. Periods of less than two months for the preparation of proposals were identified in at least nine processes. This tendency is likely to provoke uncertainty, as it forces bidders to use high safety factors in their estimates, and discourages participation of investors who are more risk averse or who have business models that demand a good knowledge of the infrastructure. Uncertainty of Investment plans at Pre-Feasibility Level: Many municipalities that initiated the concession process had large investment backlogs, lacked basic information on the extent and quality of the infrastructure, and did not have Master Plans. However, given the urgency of presenting a solution to limited water services, and the short time frame with which city mayors were granted to initiate the process, concession agreements were structured based on conceptual investment plans and poorly defined Master Plans. Definition of Scope and Quality of the Concession Contract Supervision (Interventor): The supervision role should be based on meeting the established commitments but in most cases the contract did not include a definition of the monitoring tasks, nor of the indicators to be monitored. In addition municipalities often appointed contract supervision firms that did not have the correct profile, and they evidenced more interest in intervening in the operator‘s administrative decisions than in the achievement of contract targets. Rapid Implementation of the Immediate Investment Action Plan: The initial PME model allowed for a selected operator to only have a conceptual investment plan and that Adopted from Anders, A., Sislen, D., and Marin, P. (eds) (forthcoming); Charting a New Course: Structural Reforms in Colombia’s 10 Water Supply and Sanitation Sector. World Bank. 50 studies at the design level were initiated once the operation started. The operator was then responsible for submitting the project for approval and beginning the necessary proceedings to sign the financial support agreement with the Nation, and to later implement the bidding process for the procurement of works. This proved to be a lengthy cycle which resulted in a negative effect on community perceptions of the process as a result of limited improvements in the quality of service. Recent processes have included schemes allowing for the essential works to begin immediately following the arrival of the operator. Continuity of Municipal Political Support: Often concession processes that were awarded during a specific mayor‘s term were subject to the opposition of other party candidates during the next mayoral election campaigns. A newly elected opposition mayor often presented obstacles to the operator by delaying municipal contributions and using the service as a political tool. The lesson learned is that the process to establish an arrangement with a new operator must involve not only the municipal administration but also the community in general and the other political groups in the region. 51 Annex 6: Bank Lending and Implementation Support/Supervision Processes (a) Task Team members Responsibility/ Names Title Unit Specialty Lending Lead Water and Sanitation Menahem Libhaber LCSUW Task Team Leader Specialist Private Sector Fernando Troyano Consultant Development Specialist Paula Pini Social and Community Specialist LCSUW Social assessment Environmental Juan David Quintero Sr. Environmental Specialist LCSUW assessment Economic Maria Angelica Sotomayor Sr. Economist LCSUW assessment Luz Maria Gonzalez Consultant Financial analysis Supervision/ICR Lead Water and Sanitation Task Team Greg Browder LCSUW Specialist Leader/ICR author Local project Carlos A. Uribe Water and Sanitation Engineer LCSUW officer Eric Dickson Urban Economist LCSUW ICR author ICR engineering Juan Camilo Gil Consultant economic analysis Procurement Jose Martinez Senior Procurement Specialist LCSPT supervision ICR financial Luz Maria Gonzalez Consultant analysis Luz Zeron Financial Specialist LCSFM FM supervision Carmen Yee-Batista Water and Sanitation Specialist LCSUW Team member 52 (b) Staff Time and Cost Staff Time and Cost (Bank Budget Only) USD Thousands Stage of Project Cycle No. of staff weeks (including travel and consultant costs) Lending FY00 6.27 20.2 FY01 24.03 56.5 FY02 14.11 19.1 Total: 44.41 96.0 Supervision/ICR FY03 11.98 26.8 FY04 12.96 42.3 FY05 10.65 21.3 FY06 13.24 23.3 FY07 6.76 9.0 FY08 4.58 36.8 FY09 5.95 41.3 FY10 12.82 12.1 FY11 13.74 16.8 Total: 92.68 230.2 53 Annex 7: Stakeholder Workshop Report and Results An ICR review workshop was conducted in Bogota on April 11, 2011, between the ICR team and representatives of the MAVDT. The presentation comprised the following key components: i) evaluation process methodology, ii) identified strengths and weaknesses of project design, iii) principal results including technical and financial analyses, iv) overall scoring of the Project and stakeholders, and v) recommendations and lessons learned. The workshop was well attended and outputs equally well received from MAVDT participants. There was general consensus with the Project scoring based on its overall relevance, compliance with PDO indicators, efficiency of investments, and level of financial risk. Following the workshop, the MAVDT submitted a formal request that the scoring of their performance be increased from Moderately Satisfactory to Satisfactory in light of i) a significant increase in counterpart funding throughout the life of the project ($30M to $45M), ii) the expansion of the project using counterpart funds to include municipalities outside of the Colombian Caribbean coast, iii) the direct link that the project has to the existing national policy as defined in the Planes Departmentales de Agua, and iv) the close and continuous contact that the Ministry had with the operators involved in the project. Discussion focused on several lessons learned, the most salient of which included:  Recognition that the key variables of project design include water tariffs, available financial resources for investment, and the defined service delivery targets for an operator. Given that the relative weight/importance of these variables changes across different municipalities, the quality and experience of the selected operator was noted to be of critical importance.  Debt liabilities assumed by newly installed operators from their public predecessors was noted to be a particularly challenging problem in establishing a steady pace of project implementation and ensuring the sustainability of the new operators.  The role of the supervisor/interventor for the operator (contracted by the Municipal Government) in supervising the performance of the operator was highlighted as a critical but often neglected issue. It is important to define clearly the supervisor‘s role and provide sufficient budget for the task. 54 Annex 8: Summary of Borrower's ICR and/or Comments on Draft ICR 1. The Ministry of Environment, Housing and Territorial Development believes that significant benefits were obtained through the implementation of the Business Modernization Program (WEP), supported by the World Bank through IBRD loan 7077- CO, in improving services water and sewage in terms of coverage, continuity, quality, sustainability, water culture and reasonable costs in those participating municipalities, with major advances in most cases. 2. In addition to the improvements achieved in the indicators of service delivery in participating municipalities, the loan was instrumental in the political dynamics of linking private-sector involvement in the administration and operation of water and sewage systems in the country, and promoting institutionalization of regional schemes as a tool for municipalities where a lack of sustainable institutional policies and a profound weakness in planning for the use of funds from different sources were evident. 3. The implementation of the Project is considered to be the essential input for the design of the current national government policy of the Departmental Water Plans (PDA), which includes within its main objectives the acceleration and expansion of the institutional arrangements for the provision of public services of drinking water and basic sanitation as well as furthering the use of regional schemes to capitalize on potential economies of scale. 4. The focus of the Project on medium and small cities has strengthened the market for specialized operators in the country, which increases the viability of processes occurring in the context of the PDA. Thus there are currently around 160 municipalities with specialized private in the country that are serviced by at least 30 business groups. 5. Regarding the results and lessons presented and discussed at the workshop held on April 11, 2011, it is considered that the analysis and conclusions presented by the Bank's evaluation team provided an integrated overview of the aspects of the implementation experience of the loan, without neglecting those who are not adequately represented by quantitative indicators of the project. However, with respect to the overall performance rating of the Ministry presented at the workshop, the MAVDT reiterates the request that was respectfully communicated of April 13, 2011, so that it is reconsidered. 55 Annex 9: List of Supporting Documents A. Bank Staff Assessments  World Bank (2001), Project Appraisal Document on a Proposed Loan in the Amount of US $40 million to the Republic of Colombia for a Water Sector Reform Assistance Project.  World Bank (to 2010), Implementation Status and Results (ISR) Reports  Anders, A., Sislen, D., and Marin, P. (eds) (forthcoming); Charting a New Course: Structural Reforms in Colombia’s Water Supply and Sanitation Sector. World Bank/PPIAF. B. Other  Equity Investment S.A. Banca de Inversion (2008), Ensayos de Supervisión: Informe Final  Gonzalez, Luz Maria (2011), Financial Analysis (7077 ICR)  Gil, Juan Camilo (2011), Technical Analysis (7077 ICR)  Ministerio deAmbiente, Vivienda y Desarrollo Territorial (2011), Lecciones Aprendidas BIRF 7077 (ICR) 56 IBRD 38454 75°W 70°W 0 80 160 240 320 Kilometers Puerto Bolívar 0 40 80 120 160 200 Miles RA Ríohacha JI A U Santa Marta G To C aribbean Barranquilla LA Maracaibo AT L Á N T I C O COLOMBIA Se a A N Cartegena Valledupar LE A D 10°N Maganqué CESAR Lago de G A Maracaibo M Sincelejo PA N El Baneo AM SU SU Monteria A CR CR AR Acandí NORTE DE R.B . DE R. B. E E LÍV Ocaña SANTANDER Turbo A OB To VE NEZU ELA VENEZUELA BO RD Mérida CÓ Cúcuta uca To Ca Guasdualito Bucaramanga Yarumal a al en Arauca gd Socorro ANTIOQUIA M a ARAUCA CÁ Atr ER AND a to Casanare SANT YA Medellin Puerto BO P A C IFIC Carreño Quibdo Chiquinquirá Puerto OCEA N S CASANARE CHOCÓ LDA Tunja Nueva CA Manizales CUNDINA- Yopal 5°N RISARALDÁ Santa Rita 5°N Cartago Pereira MARCA ta VICHADA Atacavi QUINDIO Armenia BOGOTÁ Me San Pedro Gaviotas Ibaque Girardot Villavincencio Chaviva a da Vich MA Buenaventura Buga VALLE DEL LI DISTRITO Puerto TO Palmira CAPITAL Inirída CAUCA Cali M E TA San Juan de Arama Neiva Guaviare GUAINÍA CAUCA HULA San José Mapiripana Tabaquén Guapí Popayan del Guaviare Brujas San Vicente Garzon a del Caguán Calamar Patí Tumaco Neg NARIÑO Florencia G U AV I A R E ro San Rafael Miraflores Mitu Pasto Cag Mocoa Vau pés n uá Yavarate To Ipiales PUTUM CAQUETÁ Ibarra Puerto AY VA U P É S Asis O Macujer Puerto Huitoto 0°N Puerto 0°N AD EC U A D O R Leguízamo Puerto Pizarro Lérida Ca qu e tá Puerto COLOMBIA Pt Santander Locas de RAZI B RA Z I L ZI um Cahuinari DEPARTMENTS WITH WORLD BANK ay o La Pedrera FINANCED SUB-PROJECTS El Encanto AMAZONAS DEPARTMENTS WITH WORLD BANK FINANCED SUB-PROJECTS MAIN CITIES AND TOWNS DEPARTMENT CAPITALS PER U PERU NATIONAL CAPITAL This map was produced by the MAIN ROADS Map Design Unit of The World Bank. The boundaries, colors, RAILROADS Leticia denominations and any other information shown on this map do not imply, on the part of The DEPARTMENT BOUNDARIES World Bank Group, any judgment on the legal status of any INTERNATIONAL BOUNDARIES 75°W 70°W territory, or any endorsement or acceptance of such boundaries. MARCH 2011 IBRD 38455 74°W 73°W 72°W 71°W COLOMBIA 14°N 14°N MUNICIPALITIES WITH WORLD BANK FINANCED SUB-PROJECTS 0 50 100 150 MUNICIPALITIES WITH WORLD BANK KILOMETERS FINANCED SUB-PROJECTS Area of Map DEPARTMENTS WITH WORLD BANK COLOMBIA FINANCED SUB-PROJECTS MAIN CITIES AND TOWNS 13°N 13°N DEPARTMENT CAPITALS MAIN ROADS DEPARTMENT BOUNDARIES INTERNATIONAL BOUNDARIES 76°W 75°W This map was produced by the Caribbean Sea 12°N Map Design Unit of The World 12°N Bank. The boundaries, colors, denominations and any other information shown on this map do not imply, on the part of The World Bank Group, any judgment on the legal status of any territory, or any endorsement or Riohacha acceptance of such boundaries. Maicao Gulf of Venezuela Santa Marta LA GUAJIRA 11°N Baranquilla ATLÁNTICA Baranoa Soledad Sabanagrande Polonuevo Santo Tomás San Estanislao de Kostka Santa Rosa Cartagena MAGDALENA Valledupar Villanueva Soplaviento Turbaco Calamar Arjona Mahates R. 10°N ar Ces San Juan de Nepomuceno Maracaibo San Onofre Lake Mag da le na CESAR Sincelejo Corozal R. mb o R. tu Magangué a at C 9°N SUCRE El Banco Cereté Sahagún Ciénaga de Oro Zulia R. Montería San Carlos San Marcos . Sinú R BOLÍVAR NORTE DE CÓRDOBA VENEZUELA SANTANDER . Cúcuta ge R San Jor R. hí Nec . Sog am aR oso uc R. Ca Arauca R. Bucaramanga ANTIOQUIA S A N TA N D E R BOYACÁ ARAUCA Op ó R. te R. n Cr a v o Nor MARCH 2011