Document of The World Bank Report No: ICR00002854 IMPLEMENTATION COMPLETION AND RESULTS REPORT (IBRD-7560-DO) ON A LOAN IN THE AMOUNT OF US$ 42 MILLION TO DOMINICAN REPUBLIC FOR A ELECTRICITY DISTRIBUTION REHABILITATION PROJECT February 28, 2014 Sustainable Development Department Caribbean Country Management Unit Latin America and the Caribbean Region CURRENCY EQUIVALENTS (Exchange Rate Effective: February 9, 2014) Currency Unit = Dominican Republic Peso (DR$) DR$$1.00 = US$0.02311 US$1.00 = DR$42.0920 FISCAL YEAR January 1 - December 31 ABBREVIATIONS AND ACRONYMS CDEEE Corporacion Dominicana de Empresas Electricas y Estatales CNE Comision Nacional de Energia CRI Cash Recovery Index EDEs Empresas Distribuidoras de Electricidad EdeEste Empresa Distribuidora de Electricidad del Este S.A. EdeNorte Empresa Distribuidora de Electricidad del Norte S.A. EdeSur Empresa Distribuidora de Electricidad del Sur S.A. GDP Gross Domestic Product IPP Independent Power Producer LAC Latin America and Caribbean OC Organismo Coordinador del Sistema Electrico Interconectado PIU Project Implementation Unit PROTECOM Oficina de Proteccion al Consumidor de Energia Electrica SIE Superintendencia de Electricidad TA Technical Assistance Vice President: Hasan A. Tuluy Country Director: Sophie Sirtaine Sector Director: Ede J. Ijjasz-Vasquez Sector Manager: Malcolm Cosgrove-Davies Sector Leader: Maria Angelica Sotomayor Task Team Leader: David Reinstein ICR Team Leader: David Reinstein DOMINICAN REPUBLIC ELECTRICITY DISTRIBUTION REHABILITATION PROJECT TABLE OF 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 1. Program Context, Development Objectives and Design......................... 9 2. Key Factors Affecting Implementation and Outcomes ............................. 14 3. Assessment of Outcomes ....................................... ...... 22 4. Assessment of Risk to Development Outcome .......................... 28 5. Assessment of Bank and Borrower Performance ...................... 30 6. Lessons Learned .................................................... 33 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners . ......... 35 Annex 1. Project Cost and Financing ................................... ..... 37 Annex 2. Ouputs by Component .......................................... 38 Annex 3. Economic and Financial Analysis..................................... 40 Annex 4. Bank Lending and Implementation Support/Supervision Processes ........... 41 Annex 5. Beneficiary Survey Results ............................................ 43 Annex 6. Stackeholders Worshop Report and Results................................ 44 Annex 7. Summary of Borrower's ICR and/or Comments on Draft ICR ........... ..... 45 Annex 8. Comments of Cofinanciers and Other Partners/Stakeholders ............... 47 Annex 9. List of Supporting Documents ..................................... 48 1 A. Basic Information Country: Dominican Republic Project Name:Electricity Distribution Rehabilitation Project L/C/TF Number(s): Project ID: P089866 7560-DO 7560-DO ICR Date: 15/02/2014 ICR Type: Core ICR Lending Instrument: IBRD Borrower: GOVERNMENT OF DR Original Total DOP 1,804,950,210.00 Disbursed Amount: DOP 1,781,586,502.58 Commitment: Environmental Category: B Borrower: CDEEE, Dominican Republic Implementing Agency: EdeNorte, EdeSur, EdeEste (Dominican Republic) Co-financiers and Other External Partners: Borrower and other financiers (not specified) IB. Key Dates Process Date Process Original Date Rvsd/Ata Date(s) Concept Review: 03/07/07 Effectiveness: 18/06/09 Appraisal: 04/02/08 Restructuring(s): Approval: 05/20/08 Mid-term Review: Closing: 12/31/2012 09/30/2013 C.1 Performance Rating by ICR Outcomes: Moderately Satisfactory Risk to Development Outcome: Moderate Bank Performance: Moderately Satisfactory Borrower Performance Moderately Satisfactory 1C.2 Detailed Ratings of Bank and Borrower Performance (by ICR) Bank ~~Ratings Broe aig *5liyt nr. saifS mmeS ** - Quality at Entry: Moderately Unsatisfactory Government: Moderately Unsatisfactory Quality of Supervision: Satisfactory Implementing Agency / Satisfactory Agencies: erall Bank Overall BorrowerR Pe rformance: Moderately Satisfactory Pefrac:Moderately Satisfactory Performance iany) PotntalPrble PojctatYes IQuality at Entry (QEA): None any time (Yes/No): Problem Project at any time NoQuality of Supervision Nn (Yes/No): (QSA): DO rating before Satisfactory 2 D. Sector and Theme Codes Closing/Inativnastatus Sector Code (as % of total Bank financing) Transmission and distribution of electricity 100% 100% T heme Code (PrimarylSecondary) Infrastructure services for private sector development 100% 100% E. Bank Staff Positions At ICR At Approval Vice President: Hasan A. Tuluy lPamela Cox Country Director: Sophie Sirtaine Yvonne M. Tsikata Sector Manager: Malcolm Cosgrove-Davies Philippe Benoit TTL: David Reinstein Lucio Monari ICR Team Leader: David Reinstein ICR Primary Author: Gabriela Elizondo-Azuela F. Results Framework Analysis Project Development Objective (from Project Appraisal Document - PAD) The original project development objectives were: (a) increase the Cash Recovery Index (CRI) of the three EDEs, and (b) improve the quality of electricity service. The PDO was not modified during the life of the loan. 3 PDO Indicators Original Target Actual Value Formally Values (fromAcivda -W . civda Indicator Baseline Value RvsdTre Indicator Increase percentage of people satisfied with electricity ec by distribution Quantitativeeo ualitative)l No baseline provided at entry. TBD Values (ISR 3 March Vales srviThe Gallup survey to Values (ISR 3 ~~~~March Vaue (IR30 fTh0alu)sreyt 2 2009): March 2, 2009): assess progress in EdeNorte: 62% No target values EdeNorte: 75% the rehabilitated (Quantitative or EdeSur: 62% provided at entry. EdeSur: 75% circuits will be EdeEste: 75% launched 6 months after the completion of rehabilitation ISR 9 (July 2012) ISR 9 (JulyQ2 of 2014). . 2012) provided confirmed baseline an end target of value at 62%75 75% Date achieved 3/2/2009 12/31/2010 12/31/2010 08/23/2013 The progress on this indicator has been measured at the level of rehabilitated circuits. A Gallup survey on consumer satisfaction at the level of circuits intervened will be Comments launched in Q2 of 2014, a few months after the full completion of rehabilitation works (inl. % to ensure satisfaction is not affected by the disruptions associated with construction. achievement) Consumer satisfaction at the level of circuits (as reported by EDEs for some of the intervened circuits) has remarkably improved as follows: i) between 200 and 633 % in Edeste, ii) 414 and 455 % in EdeNorte, and iii) between 327 and 436% in EdeSur. Indicator 2 Average service availability index (ASAl) in each EDE as a whole (and per circuit with Project investments) No formal revisions Original Values: EdeNorte: 74.2% Value EdeNorte: 73.35% No target values Target values EdeSur: 85.0% offered at entry. (introduced in (Quantitative or EdeSur: 74.9% ISR No.4 of EdeEste: 91.5% Qualitative) EdeEste: 75% November, 2009): EdeNorte: 75% EdeSur: 75% EdeEste: 75% Date achieved 4/18/2008 11/14/2009 11/14/2009 30/09/2013 Comments Baseline values were ratified before the signing of EPC procurement contracts and the start of rehabilitation works (between August, 2010 and Q2 2011). acieemt On average the three EDEs achieved 111 percent of the target. EdeNorte reached 99 achievement) percent, EdeSur 113 percent, EdeEste 122 percent. Indicator 3 Audited increase of Cash Recovery Index (CRI) in each EDE as a whole Value EdeNorte: 71% EdeNorte: 62% (Quantitative or Original Baseline: EdeSur: 75% EdeSur: 67% 4 Qualitative) EdeNorte: 56% EdeEste: 75% EdeEste: 59% EdeSur: 58% EdeEste: 62% Adjusted Baseline: (Based on data adjusted in December 2010 after full revision and update of EDEs data bases) EdeNorte: 63% EdeSur: 53% EdeEste: 62% Date achieved 4/18/2008 12/31/2010 30/09/2013 On average the three EDEs reached only 85 percent of the target. EdeNorte attained 87 percent of the target, Edesur 89 percent, and EdeEste 79 percent. Compared to the baseline, EdeSur registered an increase of 26 percent. The CRI in EdeEste decreased in 5 percent and in EdeNorte it decreased in about 2 percent. Thus, the CRI Comments declined in 2 out of 3 EDEs from the adjusted baseline. (incl. % To some extent, the decrease in CRI in both EdeEste and EdeNorte can be attributed to the achievement) fact that the PRA areas (National Program for the Reduction of Blackouts, created to service the urban poor with a subsidized tariff) had to be redefined when the Program was phased out in 2011. Also, the data used to estimate the baseline in 2010 was less reliable than the data used in 2014 to estimate the progress, as the EDEs had gone through a process of improved data management (cleaning of existing data bases) and reporting. Intermediate Outcome Indicators Original Target Formally ActalVuAcivdt Values (from Indicator Baseline Value Reie agtCompletion or Target approval Values Years documents) Indicator 1 Rehabilitation of selected circuits No formal revisions. Adjustments Vlue EdeNorte: 0 :EdeNorte: 37 introduced in ISR::EdeNorte: 4 quantitative or EdeSur: 0 :EdeSur: 5 of May 2012. EdSr4 Qualitative) EdeEste: 0 :EdeEste: 18 :EdeEste: 5 EdeNorte: 5 EdeSur: 12 EdeEste: 18 Date achieved 4/18/2008 123/0005/31/2010 3/921 Originally the Project was expected to reach 60 circuits (20 circuits with Bank support). Comments The overall target was adjusted to 35 (ISR of May, 30 2012) as follows: 5 in Edenorte, (incl. % 12 in EdeSur and 18 in EdleEste. achievement) With the Bank support (42 million), ultimately 13 circuits were rehabilitated (the remaining circuits are being rehabilitated with resources from the IADB ($42 million), 5 OPEC Fund ($38 million) and GoDR ($20 million)). Total project cost was $152 million. In terms of the number of circuits rehabilitated, the IBRD Project reached 65 percent of the "original" target. However, the total number of customers reached was higher in the 13 circuits rehabilitated than expected with the original target. The original 20 circuits were expected to reach 81,489 customers; the 13 circuits rehabilitated reached 101,197 customers. Thus, although the "original" target was not reached, the number of beneficiaries was ultimately 24.2 percent higher than expected The rehabilitation works of remaining circuits (those being financed by IADB, OPEC Fund) are expected to be completed at the end 2014. Indicator 2 Increase of CRI by circuit intervened EdeNorte LVEG-1 03: 97% LVEG-1 05: 93% WAS-101: 101% EdeNorte: 9%-56% EdeNorte: 64-75% CVTU-101: 94% 37circuits identified 37circuits identified COTC-102:94% by code in PAD by code in PAD No formal EdeSur: 25%-65% EdeSur: 34-84% revision took EdeSur Value place during CUAR102: 102.8% Vuauitaive or 5 circuits identified by 5 circuits identified Plac C AH 010 2.8% quantitative or code in PAD by code in PAD Project COHE103: N.A. Qualitative) implementation to KDIE102: N.A. EdeEste: 28%-76% EdeEste: 84% adjust baselines BAYO102: 96.8% 18 circuits identified 18circuits identified and targets. EdeEste: 84% by code in PAD by code in PAD LM38-CO5: 97% LM38-CO1: 93% LM38-CO2: 101% CAPO-CO2: 94% Date achieved 4/18/2008 12/31/2010 12/31/2010 30/09/2013 The circuits selected for intervention at appraisal were not necessarily the same (in classification and/or scope) as the circuits intervened during project implementation due to changes in population density, discovery of unregistered customers, demand side needs and other technical considerations (number of polygons and borderlines changed, some of the rehabilitations supported by the Project only covered polygons as opposed to whole circuits). Comments No formal revisions took place during Project implementation to adjust original (incl. % baselines and targets of this indicator; the value ranges provided in the original results achievement) framework were maintained and assumed to be relevant. All the circuits intervened achieved a CRI above the maximum original target established for the circuits selected at appraisal. On average, the CRI of intervened circuits in EdeNorte, EdeSur and Edeste reached a value of 28, 19 and 14 percent above the original targets, respectively. Indicator 3 Increase average availability index ASAI by circuit intervened EdeNorte: 50-87% EdeNorte: 37 circuits identified EdeNorte: 87.5%12:9.8 by code in PAD EdNre875SAC0:.8 In all 37 circuits No formal LVEG103:97.36 EdeSur:revision took LVEG15:87.83 Value EdeSur: 87.5% place during WAS101: 99.80 quantitative or 5 circuits identified by In all 5 circuits Project Qualitative) code in PAD implementation to EdeSur: EdeEste: 87.5% revise baselines CUAR102:100.0% EdeEircutstdentified In all 18 circuits and targets. COHE103:97.0% 18 circuits identified KDIE12:98.0% BAYO102: 68.0% 6 EdeEste: LM38-CO5: 74% LM38-C01: 99% LM38-CO2: 74% CAPO-CO2: 64% Date achieved /18/2008 12/31/2010 12/31/2010 30/09/2013 The circuits selected for intervention at appraisal were not necessarily the same as the circuits intervened during project implementation due to changes in population density, discovery of unregistered customers, demand side needs and other technical considerations (number of polygons and borderlines changed). Comments No revisions took place during Project implementation to adjust original baselines and (incl. % targets provided at entry on this indicator; the value ranges provided in the original achievement) results framework were maintained and assumed to be relevant. On average, EdeSur reached an ASAI of 90.75% in intervened circuits (above original target), EdeNorte reached 96.32% which is also well above original target, but Edeste reached 78% which is below original target. Indicator 4 Control of operating costs of the EDEs TOCi < TOCi.1xk TOCSep2013s 3,503 MCOP EdeEste: 2,809 MCOP < 3,528 MCOP Date achieved /18/2008 12/31/2010 30/09/2013 Baseline values were estimated before the signature of EPC procurement contracts and Comments the start of rehabilitation works. (incl. % The Results Framework provides a condition based on a formula for assessing achievement) progress. Data available includes the comparison between total operating costs (TOC) of September 2012 and September 2013 (TOCSep2013 3,503 MCOP * EdeEste: 2,809 MCOP < 3,528 MCOP 3.3 Efficiency The benefits selected for the economic analysis at appraisal included: i) reduction in system technical losses, ii) reduction in theft (through both illegal and unmetered connections, and by not paying bills) and, iii) increased hours of power availability. At appraisal the ERR was estimated in 73 percent with a NPV of US$428 Million (discounted at 10 percent and based on a 10 year stream of benefits). At closing, the NPV is estimated in US$110.95 Million with a corresponding ERR of 63 percent when considered the same benefits (discounted at 10 percent and based on a 10 year stream of benefits). It is important to note that despite the fact that scope of the intervention lowered in terms of the number of circuits, the actual hardware (equipment, materials) was ultimately aligned to the original scope. Annex 3 describes in more detail the assumptions and results of the economic analysis. 25 3.4 Justification of Overall Outcome / Rating The overall outcome rating proposed is moderately satisfactory. The original objectives of the Project today reflect a high overall relevance considering that the design reflected a proper diagnosis of a development priority that continues to be critical. The development objectives were partially achieved mainly due to i) the natural transformation of distribution networks in the three different EDEs that occurs with time (in terms of changes in population density and energy demand) and the reclassification of circuits and circuits' areas which required a redefinition of the scope of the Project (in terms of the number of circuits intervened), ii) the assumption at entry that the Project would have a significant -or at least noticeable- impact on indicators at the level of EDEs (and not only specifically on the circuits rehabilitated). These aspects can be regarded as moderate shortcomings considering that the CRI at the level of circuits rehabilitated surpassed the targets established and the population (or beneficiaries) reached was higher than originally expected. Also, the increase of consumer satisfaction at the level of circuits was remarkable, proving that the interventions were extremely successful. The ASAI targets were met at the level of EDEs, but partially met in the circuits rehabilitated. This is in part explained by the fact that in some cases the Project only supported the rehabilitation of a few polygons within circuits, and not the whole circuits. On the other hand, the EDEs did not have in place the number of meters and reporting system necessary to perform reliable measurements in the areas of intervened circuits. The data for indicators in consumer satisfaction at the level of EDEs is still pending as the survey has to be launched within reasonable period of time after the completion of rehabilitation works and community outreach efforts to avoid the impact of the perception during and soon after the construction works (6 months to a year). 3.5 Overarching Themes, Other Outcomes and Impacts i) Poverty Impacts, Gender Aspects, and Social Development The community outreach component was extremely effective not only in enhancing customer understanding of the service (cost of service, reliability, demand side management) and directly contributing to increase billing collection in the areas intervened by the Project, but also in delivering other co-benefits or unintended effects. First, the rehabilitation of circuits brought public lighting to the areas intervened which had an impact on citizen safety and on the reduction of crime. Second, the areas illuminated created an interest for additional improvements such as the organization of street sport and cultural activities and the setting 26 of green areas or parks. Third, small business and households experienced less accidents associated with the use of electrical installations (less fires and explosions). Fourth, the energy bill of the customers in the intervened circuits was substantially reduced due the increase in the number of hours of service (24 hours), and the elimination in the use of liquefied petroleum gas (LPG) and additional electricity to charge the inverter-battery systems at the household and commercial levels. The impact of this Component on Community outreach was outstanding. Proposal for this project was prepared with focus on this Component and won a prize in Shared Prosperity Days). Most importantly, the rehabilitation of circuits focused primarily on areas with predominance of lower income communities, thereby enhancing their overall electricity access and quality of service. The Project has improved the standard living conditions of poor urban communities as well as the capacity of small businesses to improve their income and efficiency through extended access to the electricity service and improved reliability. ii) Institutional Change / Strengthening Although the Bank intervention ultimately only supported the rehabilitation of a small percent of the total number of circuits, the model established -which also promoted better management at the level of EDEs and a more dynamic and informed consumer participation- was effective and replicated in the rehabilitation of all other circuits not directly supported by the Bank. Indeed, the Project brought to the three EDEs the first challenge in massive rehabilitation and has left a strong managerial and technical system capable of following up with new stages in the rehabilitation of circuits. Ultimately, the improvements in governance at the level of CDEEE and EDEs for managing technical and commercial losses can be directly attributed to the Project. The Project contributed to the organizational and managerial strengthening of EDEs (including the creation of new units to improve data management and reporting and manage the community outreach activities), to increased consumer awareness regarding the service and to the creation of a strong system of social workers and community leaders. One of the issues however that have to be improved in the short term is the formalization of the employment status of social and community workers, who do not have a permanent staff status or earn benefits. iii) Other Unintended Outcomes and Impacts None iv) Summary of Findings of Beneficiary Survey and/or Stakeholder Workshops No beneficiary survey or stakeholder workshop has been launched so far. 27 SECTION 4. ASSESSMENT OF RISK TO DEVELOPMENT OUTCOME 4.1 Risk to Development Outcome Rating can be considerated Moderate The sustainability of development outcomes in the long term depends on several factors, including the following: * Appropriate operation and maintenance of distribution lines to maintain technical losses at a minimum level * The continuous implementation of the social management and commercial strategies to maintain a high levels of billing collection (for example, through the activities conducted by the Comites de Seguimiento y Enlance, COSEs) * Continuous strengthening of the M&E framework * Increased capacity in the execution of implementation, disbursement and procurement plans and the implementation of turn-key projects to continue with rehabilitations and extensions of the network. The project has successfully introduced these elements and has acquired a high level of capacity and experience in the implementation of technical, social, procurement and fiduciary activities. The organizational structure created to manage the Project has continuously evolved and is considered strong (the structure includes today new units attending the commercial, social and environmental aspects of rehabilitation and service provision; all these activities are well aligned and integrated). The model introduced under the Project has already been replicated in the rehabilitation of circuits not directly supported by the Project. Indeed, the Project brought to the three EDEs the first challenge in massive rehabilitation and has left a strong managerial and technical system capable of following up with new stages in the rehabilitation of circuits. The operation is not facing any major uncertainties over its expected remaining useful life. The risks that the long-term flow of benefits in the circuits rehabilitated (lower commercial and technical losses and maintained reliability and quality of service) will not be realized is low. The flow of benefits however could be affected by increased constraints on the financial sustainability front (which could in turn affect the flow of resources to maintenance and community outreach). Indeed, the risk that GODR breaks the social compacts and fails to supply 24-hour electricity even with high repayment due to cash flow problems is plausible. However, the Government and CDEEE are taking steps in the right direction. As described before, the current National Development Strategy (END or Vision 2030) -approved by Congress in March 2012- explicitly prioritizes the development of a reliable, efficient and environmentally sustainable energy sector under its Pillar No. 3 (Productive Development) with two specific objectives: i) ensure a reliable supply of electricity at competitive prices as well as financial and environmental sustainability, and ii) ensure a reliable, diversified supply of fuels in a competitive and environmentally sustainable manner. In 28 the medium term, the END specifically focuses on reducing electricity sector losses and cutting generation costs by adding publicly owned coal generation capacity and negotiating new capacity additions with existing private generators, among other. The risk that the development outcome will not be maintained is therefore considered moderate. 29 SECTION 5. ASSESSMENT OF BANK AND BORROWER PERFORMANCE 5.1 Bank Performance The rating of the overall Bank performance can be considered moderately satisfactory. i) Quality at Entry In general terms, the operation was well designed and appraised. Indeed, the team included the key components necessary to increase the Cash Recovery Index (CRI) and improve the quality of service in the areas where circuits (feeders) were rehabilitated. The Project's design as well as the preparation and appraisal were also aligned to the Bank's fiduciary role. However, the PDOs were not appropriate and consistent with the scope of the intervention. This means that even with the rehabilitation of 60 circuits (as offered at entry), it was unlikely that the Project could have had an impact on the overall CRI at the level of distribution companies. In addition, and as explained before, the CRI at the level of EDEs was affected by factors which were not under the control of the proposed intervention (e.g.; absorption of PRA areas, changes in the number of registered customers, total number of customers billed, etc.). It was also unlikely that an intervention only affecting a small percent of the total number of feeders could have had an impact on the overall consumer satisfaction at the level of EDEs. During preparation and appraisal the team missed to assess the possibility that the scope of the Project -in terms of the number of circuits- could change with time for the reasons explained before and that the Project's impact could be limited only to the circuits rehabilitated under the operation (and not necessarily affect the overall distribution network and EDEs). The risk assessment and Results Framework proposed at entry did not reflect these considerations. However, it is important to note that the Project has indeed been extremely successful in the areas where the circuits were rehabilitated. Also, the Project has had an impact on the overall management and implementation of circuit rehabilitation at the level of EDEs, and the concept and practices introduced -especially the innovative community outreach component- have been further replicated which is expected to have a more generalized impact over time. In particular, the Project has contributed to prepare the companies for the next stage, which includes automation and the introduction of smart metering. The Bank's performance in ensuring quality at entry can be considered moderately unsatisfactory. ii) Quality of Bank Supervision As described before, during implementation the World Bank team took several actions to address the emerging challenges: i) strong and closed supervision focused on the continuous revision of implementation, disbursement and procurement plans, ii) continuous revision and strengthening of 30 operational manual (for example, to update organizational structures and to include standardized formats for performance reporting), iii) strong support to the counterpart during bidding processes, including revisions of the evaluation of "pre-qualified" firms as well as recommendations to ensure competition in the process, iv) continuous review of results framework to adjust scope, define baseline values and confirm target values of some of the proposed indicators due to circuit network transformations over time; among other. The team made a special effort to continuously adjust to contextual changes: i) managerial and organizational restructuring of EDEs, ii) managerial changes in the executing agency, iii) continuous changes in personnel responsible for the operation at the level of EDEs. In addition, considering that baseline and target values in some of the indicators were not in place or trusted at entry, almost all of the M&E had to be defined in the first year of Project supervision. In particular, the experience with similar Bank operations over time has demonstrated the difficulty in defining quantitative baselines in the performance indicators of underperforming distribution companies. It is now widely recognized that it is critical to set targets at the outset and much less important -and feasible- to define baselines from the start, especially in cases were the quality of data is poor (as it was the case in the DR). During the supervision, the team supported all efforts necessary to measure baselines (despite the challenges of the context where frequent changes were registered including the redefinition of circuit categories and geography), and of course achieve targets. The changes in the geography of circuits described before were difficult to predict during supervision, as these modifications were introduced gradually in a relatively short period of time between 2011 and 2013. During supervision, information on the changes in the geography of circuits was not clear until after the signing of EPC contracts and beginning of rehabilitation works. The rehabilitation of circuits also took place very gradually during the period 2011-2013. In reality, the final number of rehabilitated circuits has been a "moving target". In addition, during Project implementation there were 3 different Government administrations which led to a stop-and-go situation as well as multiple delays. This meant that most of the rehabilitation works took place in the last 18 months of Project implementation. During supervision the team considered a restructuring of PDO and project outcome indicators (to reflect increase in CRI only at the level of rehabilitated circuits), but this action would have entailed additional and long delays (due to the requirement of Congress approval) at a time when disbursements were actually starting to flow. Indeed, it is remarkable that the Project was able to disburse most of the resources in the last 18 months of implementation. The Bank's performance in ensuring quality at supervision can be considered moderately satisfactory. 5.2 Borrower Performance The rating of the overall Borrower performance can be considered moderately satisfactory. 31 i) Government Performance During loan preparation, the Government and the implementing agencies demonstrated strong commitment and ownership by actively participating in the design of the project, including technical background information and discussion on project formulation. Nevertheless, the project had a slow start, mainly due to delays in Congress approval of the lending operation as well as in achieving effectiveness. Borrower commitment and ownership varied over the course of implementation, reflecting changes in administration. The project slowed during a period of political transition. After considering the project's relevance, the new administration showed commitment to the project's PDO and it became at the end more supportive. The new administration proved to be an advocate for improved corporate governance of state-owned utilities and for increased sector efficiency. ii) Performance of Implementing Agencies The implementing agency (CDEEE) demonstrated strong commitment in the fulfilling of PDOs and provided adequate internal staff and resources to ensure implementation success despite the difficult enabling environment and changes outside of its direct control. CDEEE complied with all Bank covenants, produced regular progress reports and discharged its fiduciary duties in a highly satisfactory manner. CDEEE played an important interagency coordination and project management role. The major shortcoming of CDEEE during project implementation was the slow pace of procurement processing which in part was due to external factors and the time needed to finalize the detailed engineering of the circuits to be rehabilitated. Despite the numerous challenges and shortcomings, the PlU at CDEEE managed to conclude the project in a satisfactory manner and achieve the targeted values in most of the outcome and output indicators. The overall Borrower performance rating is deemed moderately satisfactory. 32 SECTION 6. LESSONS LEARNED The lessons learned in the design and implementation of the Project are described below. Procurement and Contract Design i) The PIU at CDEEE learned that is necessary to reduce the scope of bidding processes (or to set a cap on the maximum number of areas that a single company can win) to increase the number of participating contractors and reduce the risk of no delivery; without this provision very large companies tend to win the bids increasing the Project's vulnerability to delays and low quality of service. ii) The experience of the PIU is that the contract design needs to be adjusted to include penalizations against delayed delivery and/or low quality of service during the service period as opposed to the very end, when nothing can be done to steer the planned works back on track. For example, in the "supply and installation" contract, payments are delivered in 80 percent against disembarked material and 20 percent against installation. The PIU has suggested to make adjustments so that the highest proportion of payments are delivered against actual installations. Project Design i) One important lesson learned is that the basic and design engineering have to be conducted when the resources are available to pay for the rehabilitation works, considering that long delays in the Project (including in reaching effectiveness and in the launching of bidding process) render the engineering work outdated as the system changes with demand growth and other modifications. In the case of the Project, the engineering conducted in 2008 was not adequate when the actual works started in 2011. Design of Results Framework i) The choice of CRI as an indicator that reflects progress in the financial sustainability of distribution companies was not necessarily appropriate in the context of the power sector in the DR.'7 Indeed, EDEs financial sustainability is affected by a number of other variables including the debt service associated with the loans issued to purchase fuel oil and direct cash transfers from the Government to cover this debt. Thus, the CRI -as it is defined in the Project- maybe positive despite the lack of financial sustainability at the level of EDEs. 1 The CRI formula is: (energy billed by distribution companies/energy purchased by distribution companies) multiplied by (electricity bills paid by consumers)/(total electricity billing issued to consumers) 33 In theory, the CRI should reflect the extent to which power purchased is billed and bills are paid, and provide a good measure of improvements in technical and commercial losses. However, when companies do not bill 100 percent of their customers or when cash inputs other than bills paid are included (for example, reconnection fees or fines) as has happened with the three EDEs, the result is distorted. As an intermediate outcome the Results Framework proposed the monitoring of the evolution of total operating costs (TOC). 8 Together with the CRI this indicator contributes to better reflect improvements in the managerial performance and financial sustainability of EDEs, albeit only partially. Monitoring and Evaluation i) The main lessons learned in M&E is the need to impose third audit access of EDEs' primary data as condition for concessionary support to ensure the quality of M&E results. Today, external auditors have to rely on data supplied by EDEs, but there is concern regarding the quality and reliability of this data. Circuit Rehabilitation Works and Consumer Satisfaction i) The PlU has learned that it is important to include the installation of meters and other accessories (commercial infrastructure) in the works associated with the rehabilitation of circuits to avoid delays in service provision and the regularization of billing. These delays also affect consumer's perception of the quality of service and impact the baseline information in surveys launched soon after the meters are installed. Community Outreach i) The design of the social component, which included a strong Social Management Plan that introduced a seven step process as a strategy for community outreach was extremely successful in both bringing the community on board and substantially improving the revenue collection of the utilities as well as quality of service to consumers. The social activities also derived in multiple benefits and co-benefits, as described before. ii) The PlU has learned that it is critical to formalize the employment status of social and community workers in order to maintain stability in the community outreach activities. Today, most social and community employees work under temporal employment status and do not earn benefits. 18 The Project called for TOCs increasing less than the rate of supply expansion and inflation. 34 SECTION 7. COMMENTS ON ISSUES RAISED BY BORROWER / IMPLEMENTING AGENCIES / PARTNERS Positive Aspects The CDEEE (PIU) and EDEs believe that the DR Distribution Rehabilitation Project marked the beginning of a new vision for the Electricity Sector authorities to confront serious energy losses which have affected for many years the electricity distribution sector. The design and conceptualization of the project were appropriate, especially with regards to the model and methodological process necessary to reach consumers and build a constructive and sustainable relationship between the supply and demand. The project had a high degree of success in improving performance indicators at the level of rehabilitated circuits. At this level, the project successfully implemented the actions necessary to gradually achieve financial recovery of EDEs and improve the quality of service to users. Considering that for many years the power distribution sector of the country has lacked consistent investment and managerial actions focused on measurable results, the Project brought improved practices in Project execution as well as improved supervision/monitoring mechanisms and streamlined practices across the three EDEs. Through implementation of component 2, the Community Outreach and Social Management Strategy, which focused on building a bridge between service users and EDEs, has created an unprecedented milestone in the history of the Dominican electricity sector, which has been highly valued by other sectors of the country. The Project strategy was accompanied by an extensive process of communication and training to service users on both user rights and EDEs's responsibilities, which resulted in a high level of public awareness and understanding of the electricity sector. When processes are performed on technically well designed platforms, as was the strategy used in Community Outreach and Social Management Component introduced for the implementation of the Project, the results are qualitatively superior. The impact of the model and actions introduced have gone beyond the scope of the Project, since the EDEs have replicated the experience learned in the rehabilitation of other circuits, with its own resources and financing from other multilateral institutions. Negative Aspects The lack of clear baseline values for the measurement of results as well as the uncertainty in the detailed engineering and technical specifications, made the execution of the Project challenging and impacted efficiency of the processes. 35 General Recommendations Improve the accuracy of baseline information in future projects and the definition of specific indicators for sub-projects to be impacted in order to meet the proposed goals and targets. Consistency is essential in investment support to this type of projects; our observation through the years is that the lack of continuity in investments creates even greater problems and distortions in the electricity sector. 36 Annex 1. Project Costs and Financing (a) Project Cost by Component (in US$ Million) Appraisal Estimate Actual/Latest Estimate Newrscomponents (U$ilo)(Smlin Percentage of Appraisal (US$million) (US$million) 1. Rehabilitation of Distribution 37.0 38.38 103.73% Networks 2. Outreach to Consumer 3.0 1.88 62.67% Communities 3. Technical Assistance 2.0 1.09 54.50% Total Project Costs 42.0 41.35 98.45% Front-end fee --- 0.105 Total Financing Required 42.0 41.45 98.70% (b) Financing Appraisal Revised Actual/Latest Source of Funds Estimate Amount Estimate Percentage of (USD (USD Appraisal millions) millions) Borrower 10.50 --.-- 60.28 574.10% International Bank for Reconstruction and 42.00 42.00 98.70-% Development Other Financiers 99.60 --.-- 70 70.28% Total 152.10 --.-- 172.28 113.27% 37 Annex 2. Outputs by Component The table below presents the project's outputs by the closing date Objective/Output Indicator Baseline Target * Actual (% accomplished) Increase the Cash Increase the CRI in each EdeNorte: 56% EdeNorte: 71% Actual: Recovery Index CRI of EDE as a Whole EdeSur: 58% EdeSur: 75% EdeNorte: 62% the three distribution EdeEste: 62% EdeEste: 75% EdeSur: 67% companies (EDEs) EdeEste: 59% % Accomplished: EdeNorte 88% Edesur 90% EdeEste 79% Average EDEs: 85% Improve the quality of ASAI: Average Service EdeNorte: 56% EdeNorte: 75% Actual: electricity service Availability Index EdeSur: 58% EdeSur: 75% EdeNorte: 74.2% (Indice de EdeEste: 62% EdeEste: 75% EdeSur: 85.0% Disponibilidad EdeEste: 91.5% Promedio del Servicio), per circuit with project % Accomplished: investment and in each EdeNorte 99% EDE as a whole Edesur 113% EdeEste 122% Average EDEs: 111% Increase percentage of EdeNorte: 62% EdeNorte: 75% EdeNorte: TBD people satisfied with EdeSur: 62% EdeSur: 75% EdeSur: TBD electricity service by EdeEste: 62% EdeEste: 75% EdeEste: TBD :distribution companies Intermediate Outcome Indicators Rehabilitation of EdeNorte: 0 EdeNorte: 37 Actual: selected circuits EdeSur: 0 EdeSur: EdeNorte: 4 EdeEste: 0 EdeEste: 18 EdeSur: 4 EdeEste: 1 % Accomplished: EdeNorte: 10.8% EdeSur: 80.0% EdeEste: 27.3% Increase of CR by EdeNorte: 9%- EdeNorte: 84% EdeNorte circuit intervened 56% 37circuits LVEG-103: 97% 37circuits identified by code [VEG-0S: 93% identified by code in PAD VVAS-101: 101% in PAD COTU-102: 94% EdeSur: 84% SALC-102TENA: 94% EdeSur: 25%-65% S circuits E circuits identified by code EdeSur 38 Objective/Output Indicator Baseline Target * Actual (% accomplished) identified by code in PAD CUAR102: 102.8% in PAD COHE103: N.A. KDIE102: N.A. EdeEste: 28%- EdeEste: 84% BAY0102: 96.8% 76% 18circuits 18circuits identified by code EdeEste: 84% identified by code in PAD LM38-CO5: 97% in PAD LM38-C01: 93% LM38-CO2: 101% CAPO-CO2: 94% % Accomplished: On Average: EdeNorte: 128% EdeSur: 119% EdeEste: 114% EdeNorte: 50- EdeNorte: 87.5% EdeNorte: 87% 37circuits LVEG-103: -- 37circuits identified by code LVEG-105: identified by code in PAD VVAS-101: in PAD COTU-102: EdeSur: 87.5% SALC-102TENA: -- EdeSur: 62% 5 circuits 5 circuits identified by code EdeSur: identified by code in PAD CUAR102: 100.0% Increase averageinPDCH13970 availability index ASAl nPDCH13 70 avalablit inex SAIEdeEste: 87.5% KD1E102: 98.0% EdeEste: 50%- 18circuits BAY0102: 68.0% 75% identified by code 38circuits in PAD EdeEste: identified by code LM38-Cn5:t74% in PAD LM38-CO: 99% LM38-C02: 74% CAPO-C02: 64% EcIeNorte: --TOCi < TOCj-,xk EdeSu r: -- TOCep.... 3,503 M COP EdeEste: 2,809 MCOP < 3,528 COUP 39 Annex 3. Economic and Financial Analysis Cost Benefit Analysis The economic analysis at closing considers a total cost of US$ 42 Million. The benefits are the following: i) reduction in commercial and technical losses, ii) increase in the availability of electricity service to consumers (hours per day), which displaces the use of alternative sources of electricity (e.g.; batteries, other fuels, etc.). The results are summarized in the table below: Reduction in Technical Losses Increase in Power (GWh) Availability (Gwh) ERR NPV (US$ million) EdeNorte 13.7 17.0 59.7% $34.45 EdeSur 41.4 17.1 84.6% $56.42 EdeEste 27.6 9.8 41.5% $20.08 Total 82.8 43.8 62.8% $110.95 Sensitivity Analysis NPV (US$ Benefits From ERR million) Reduction in Technical Losses 29.0% $33.7 Reduction in Technical Losses and Increase In Power Availability 62.8% $110.9 Key Assumptions: > Discount rate: 10 percent > Energy Price: Purchase Price (applied to EDEs) > Increase in energy consumption: At appraisal it was assumed that the overall energy consumption in the rehabilitated circuits would increase in 10 percent due to the increase in the number of hours available to consumers. The analysis at closing is considering the same conservative assumption (although in some circuits the increase in consumption has reached 25 percent). > The costs of alternative uses of energy (batteries, other fuels) are taken as twice the cost of electricity to consumers. This is the same assumption considered at appraisal. 40 Annex 4. Bank Lending and Implementation Support/Supervision Processes (a) Task Team members Names Title Unit Responsibility/ Specialty Lucio Monari Task Team Leader LCSEG [David Reinstein Sr. Energy Specialist LCSEG Maritza A. Rodriguez Financial Mgmt Specialist LCSFM Jayme Porto-Carreiro Energy Specialist Consultant Stephen Ettinger Economist Consultant Vladimir Jadrijevic Power Engineer Consultant Catherine Abreu Procurement Analyst LCSPT [Pedro Antmann Power Engineer Consultant [Elena Correa Sr. Social Dev. Specialist SDV Alejandro Deeb Environmental Specialist Consultant [Fabiola Altimari Counsel LEGLA Mariana Montiel Sr. Counsel LEGLA Miguel-Santiago da Silva Finance Officer LOAFC Oliveira Alma Domenech Program Assistant LCSEG Robert Taylor Peer Reviewer EASTE Malcolm Cosgrove-Davies Peer Reviewer AFTEG iSarah Keener Senior Social Development Specialist LCSSO Alonso Zarzar Casis Sr Social Scientist LCSSO Zoila Catherine Abreu Rojas Procurement Specialist LCSPT Fabiola Altimari Montiel Senior Counsel LEGLE David Reinstein Senior Oil and Gas Specialist, TTL SEGOM Pedro Antmann Senior Energy SEGEN Specialist Maritza A. Rodriguez De Sr Financial Management Specialist LCSFM Pichardo Janina Andrea Franco Energy Specialist LCSEG Salazar Sergio Augusto Gonzalez Senior Energy Specialist LCSEG Coltrinari Frederic Verdol Power Engineer LCSEG [Davide Zucchini E T Consultant LCSPS Andrea Maria Castro Astudillo Program Assistant LCSEG Sandra Solano Couce Consultant LCSEG Juan Carlos Cardenas Valero E T Consultant LCSEG Vladimir T. Jadrijevic Consultant LCSEG Mary Louise Gifford E T Consultant LCSEG 41 Elena Correa Consultant SDV Alma Domenech Senior Executive Assistant SEG James Victor Pannett Operations Officer LCSEG [Laura Wendell Berman Consultant LCSEG Suzanne Casolaro Consultant LCSHH Jayme Porto-Carreiro Energy Specialist Consultant Alejandro Deeb Environmental Specialist ECSAR Stephen Ettinger Economist Consultant Gabriela Elizondo Azuela Senior Energy Specialist SEGEN (b) Staff Time and Cost It is pull up by the system 42 Annex 5. Beneficiary Survey Results Not Available. 43 Annex 6. Stakeholder Workshop Report and Results Not available. 44 Annex 7. Summary of Borrower's ICR and/or Comments on Draft ICR Positive Aspects The CDEEE (PIU) and EDEs believe that the DR Distribution Rehabilitation Project marked the beginning of a new vision for the Electricity Sector authorities to confront serious energy losses which have affected for many years the electricity distribution sector. The design and conceptualization of the project were appropriate, especially with regards to the model and methodological process necessary to reach consumers and build a constructive and sustainable relationship between the supply and demand. The project had a high degree of success in improving performance indicators at the level of rehabilitated circuits. At this level, the project successfully implemented the actions necessary to gradually achieve financial recovery of EDEs and improve the quality of service to users. Considering that for many years the power distribution sector of the country has lacked consistent investment and managerial actions focused on measurable results, the Project brought improved practices in Project execution as well as improved supervision/monitoring mechanisms and streamlined practices across the three EDEs . Through implementation of component 2, the Community Outreach and Social Management Strategy, which focused on building a bridge between service users and EDEs, has created an unprecedented milestone in the history of the Dominican electricity sector, which has been highly valued by other sectors of the country. The Project strategy was accompanied by an extensive process of communication and training to service users on both user rights and EDEs's responsibilities, which resulted in a high level of public awareness and understanding of the electricity sector. When processes are performed on technically well designed platforms, as was the strategy used in Community Outreach and Social Management Component introduced for the implementation of the Project, the results are qualitatively superior. The impact of the model and actions introduced have gone beyond the scope of the Project, since the EDEs have replicated the experience learned in the rehabilitation of other circuits, with its own resources and financing from other multilateral institutions Negative Aspects The lack of clear baseline values for the measurement of results as well as the uncertainty in the detailed engineering and technical specifications, made the execution of the Project challenging and impacted efficiency of the processes. General Recommendations Improve the accuracy of baseline information in future projects and the definition of specific indicators for sub - projects to be impacted in order to meet the proposed goals and targets. 45 Consistency is essential in investment support to this type of projects; our observation through the years is that the lack of continuity in investments creates even greater problems and distortions in the electricity sector. 46 Annex 8. Comments of Cofinanciers and Other Partners/Stakeholders Not available. 47 Annex 9. List of Supporting Documents ABC Consulting Group, Quantum and CNEEE. Informes Auditoria T6cnica del Proyecto de Rehabilitaci6n de Redes de Distribuci6n de Electricidad CDEEE. Informes de Avance Proyecto de Rehabilitaci6n de Redes de Distribuci6n de Electricidad (Financiamiento BM/BID/OFID) World Bank. Aide Memoire for the Project's Preparation and Supervision Missions World Bank. Project Implementation Status Reports (ISRs) World Bank, 2008. Project Appraisal Document on a Proposed Loan in the Amount of US$42 Million to the Dominican Republic for an Electricity Distribution Rehabilitation Project. IMF; 2013. Growth, Employment and Social Cohesion in the Dominican Republic, ILO Background Paper World Bank; 2013. Patronage or reform? Political Economy of Policy Performance in the Dominican Republic: Institutional and Governance Review, Washington DC 48 7T W 71°W 7 W DOMINICAN ATLANTIC OCEAN REPUBLIC 20°N 0 SELECTED CITIES AND TOWNS e PROVINCE CAPITALS Moneo Cristi Luperon ® NATIONAL CAPITAL älls PÖqQ MO JE-bl" RIVERS villevo E~fSPAUlATRvES Sk "s J1pUERTO PATA MAIN ROADS RAJLROADS PROVINCE BOUNDARIES T". 4 SÁNCHEZ<- - INTERNATIONAL BOUNDARIES "AMAn FrNnciAo ThNagua H AI TI ven A AN A SDAJN T TA GOmse SS5iI50 Onf O ul bjdru P9(N 7 L AiV E G A 5. F9° N >MONTE LAO sankar aM~Son om Crilr O noM {283 m)- Yi,o Mn oygoe H ECra o l PlatPE A MaAorvvEl Seibeo corrorsoNnbao IsUAlan ded o c aI ~ taa a . VAl Vj 18NAeera BARANA 18°NNT LAÅ Son HERMANAS MIRABALU D Ca9i beanSe SSÄNCHEZ RAM REZ R °"* 6. MONSENOR NCUE L 7, SAN JOSÉ DE OCOA isla 8. S AN CRIS TOBA L 0 20 4 kioorws Bea a 9. S ANTO DOMiNGO 0-, 02 UG0 AJIkleACI DOMIN -00 e ~ri 49. AO20 7TPVJ WFnt 71°å70W 6° 49lc