Document of The World Bank Report No: NCO00003531 NOTE ON CANCELLED OPERATION REPORT (IBRD-81190) ON A LOAN IN THE AMOUNT OF JPY 3,842,600 MILLION (US$50.0 MILLION EQUIVALENT) TO THE DEVELOPMENT BANK OF THE PHILIPPINES FOR A REGIONAL INFRASTRUCTURE FOR GROWTH PROJECT December 10, 2015 Social, Urban, Rural and Resilience Global Practice East Asia and Pacific Region CURRENCY EQUIVALENTS (Exchange Rate Effective June 19, 2015) Currency Unit = Peso US$ 1.00 = PhP 45.07 PhP 1.00 = US$ 0.02 US$ 1.00 = JPY 123 FISCAL YEAR January 1 – December 31 ABBREVIATIONS AND ACRONYMS ADB Asian Development Bank CAS Country Assistance Strategy DBM Department of Budget and Management DBP Development Bank of the Philippines DILG Department of Interior and Local Government DOF Department of Finance FIL Financial Intermediary Loan GFI Government Financial Institution GoP Government of the Philippines JICA Japan International Cooperation Agency KPI Key Performance Indicator LA Loan Agreement LBP Land Bank of the Philippines LGU Local Government Unit LGUGC LGU Guarantee Corportation MDFO Municipal Development Fund Office NEDA National Economic and Development Authority ODA Official Development Assistance OP Operational Policy PAD Project Appraisal Document PD Program Development PDF Philippines Development Forum PDO Project Development Objective PFI Private Financial Institution PMT Project Management Team RIGP Regional Infrastructure for Growth Project SLA Subproject Loan Agreement SPMT Sub-Project Management Team SSWSPs Small Scale Water Providers WSP Water and Sanitation Program Regional Vice President: Axel van Trotsenburg, EAP Country Director: Motoo Konishi, Philippines Senior Global Practice Director: Ede Jorge Ijjasz-Vasquez, GSURR Practice Manager: Abhas Jha, GSURR Task Team Leader: Christopher T. Pablo, GSURR NCO Team Leader: Christopher T. Pablo PHILIPPINES Regional Infrastructure for Growth CONTENTS Data Sheet A. Basic Information B. Key Dates C. Ratings Summary D. Sector and Theme Codes E. Bank Staff F. Ratings of Program Performance in ISRs 1. Context, Project Development Objectives, and Design ......................................... 1 2. Post-Approval Experience and Reasons for Cancellation .................................... 4 3. Assessment of Bank Performance ........................................................................... 8 4. Assessment of Borrower Performance .................................................................. 11 5. Lessons Learned ...................................................................................................... 12 Annex 1. Bank Lending and Implementation Support/Supervision Processes ..... 14 Annex 2. List of Supporting Documents .................................................................. 16 A. Basic Information Country: Philippines Project Name: PH – Regional Infrastructure for Growth Project ID: P108904 L/C/TF Number(s): IBRD-81190 NCO Date: 12/10/2015 Lending FIL Borrower: REPUBLIC OF Instrument: THE PHILIPPINES Original Total USD 50.00M Disbursed Amount: USD 0.12M Commitment: Revised Amount USD 0.13M Environment B Category: Implementing Agencies: Development Bank of the Philippines (DBP) Co-financiers and Other External Partners: N/A B. Key Dates Process Date Process Original Date Revised /Actual Date(s) Concept Review: 06/26/2009 Effectiveness: 05/02/2012 10/11/2012 Appraisal: 04/15/2010 Closing: 11/03/2016 06/19/2015 Approval: 12/01/2011 C. Rating Summary Performance Rating by NCO Outcomes: Not Applicable Risk to Development Outcome: Not Applicable Bank Performance: Moderately Unsatisfactory Borrower Performance: Moderately Unsatisfactory D. Sector and Theme Codes Original Sector Code (as % of total Bank financing) General agriculture, fishing and forestry sector 21 General industry and trade sector 21 General water, sanitation and flood protection 21 sector Rural and Inter-Urban Roads and Highways 21 Sub-national government administration 16 Original Theme Code (as % of total Bank financing) City-wide Infrastructure and Service Delivery 20 Natural disaster management 20 Other Private Sector Development 20 Regional integration 40 E. Bank Staff Positions At NCO At Approval Vice President: Axel van Trotsenburg James W. Adams Country Director: Motoo Konishi Bert Hofman Practice Manager/Manager: Abhas Jha Mark C. Woodward Project Team Leader Christopher T. Pablo R. Mukami Kariuki NCO Team Leader Christopher T. Pablo F. Ratings of Project Performance in ISRs Actual Date ISR No. DO IP Disbursements Archived (USD millions) 01 04/11/2012 Satisfactory Satisfactory 0 Moderately Moderately 02 05/26/2012 0 Satisfactory Satisfactory 03 12/16/2012 Satisfactory Satisfactory 0 Moderately Moderately 04 06/26/2013 0 Unsatisfactory Unsatisfactory Moderately 05 12/29/2013 Unsatisfactory 0 Unsatisfactory Moderately 06 08/06/2014 Unsatisfactory 0 Unsatisfactory Highly Highly 07 04/01/2015 0 Unsatisfactory Unsatisfactory 1. Context, Project Development Objectives, and Design Country and sector background 1. The average annual economic growth rate of the Philippines over the period 2000- 2008 was 2.8 percent, quite low compared to the 8.7 percent average growth rate of other East Asian countries. The share of the population below the poverty line increased in 2003 and 2006, income inequality remained high, and the country risked missing the Millennium Development goals on education and maternal health. At the time, the external environment threatened to dampen growth prospects, including the possibility of facing fiscal pressures. In addition, weak governance was recognized as a constraint to sustained growth and poverty reduction. The challenge of inclusive growth required a sustained effort from both the public and private sector agencies involved in development. 2. The Local Government Code of 1991 provided local government units (LGUs) increased local autonomy and accountability and empowered LGUs to improve public services and stimulate development. The Code gave LGUs the mandate to raise additional funds for investments in infrastructure and services through taxation and borrowing. However, progress in generating own source revenue and sourcing credit financing from the local capital markets was slow. As a result, many LGUs were plagued with inadequate access to basic services such as water and electricity, declining literacy rates, environmental degradation, high unemployment, lack of low-cost housing, and inadequate services. Limited access to affordable finance, a high degree of fragmentation of LGUs, and weak coordination between the national government and LGUs in local development planning contributed to generally poor LGU performance. 3. To strengthen LGUs and support the implementation of the Code, the Government of the Philippines (GoP) established the Working Group on Decentralization and Local Government under the umbrella of the Philippines Development Forum (PDF).1 This Working Group brought together national government oversight agencies, LGU leagues, civil society organizations and donor partners to identify and tackle a common reform agenda. Three key support areas were identified: (i) expanding access to finance by introducing reforms that would enable LGUs to have greater access to the capital markets; (ii) promoting inter-LGU cooperation, including cooperation-enabling frameworks and stimulating joint investment financing across multiple LGUs; and (iii) improving planning by streamlining and harmonizing procedures for planning, budgeting and resource mobilization at the local level, developing and 1 The PDF is the primary Government mechanism for policy dialogue among stakeholders in the country’s development agenda. It also serves as a process for developing consensus and generating commitments among different stakeholders toward critical actionable items of the Government’s reform agenda. 1 disseminating tools that enable joint planning across LGUs; and building capacity for strategic development planning. Rationale for Bank assistance 4. The GoP requested Bank assistance in designing and financing a project that would support the reforms identified by the PDF by expanding access to finance through the Development Bank of the Philippines (DBP), a Government Financial Institution (GFI). In addition to enabling LGUs to identify and finance investments that would promote physical and economic integration, the said financing project was envisioned to encourage LGUs to work with private partners to accelerate the pace of service provision and support Private Financial Institution (PFI) lending for local public infrastructure and services by piloting a wholesale lending facility to on-lend to LGUs, public utilities and private sector providers. The project would also expand credit access to smaller LGUs (third or fourth income class), which comprised 75 percent of all LGUs. Thus, the Regional Infrastructure for Growth Project (RIGP) was conceived. 5. At the time of appraisal, the Bank had been supporting local government reforms in the Philippines, and had a comparative advantage in its long involvement in policy dialogue, technical assistance and LGU lending. The Bank had been playing a strategic role in supporting GoP in LGU finance since the mid 1980s, and could bring global knowledge from similar projects to bear on the design of RIGP. The project would allow the Bank to continue supporting the development and refinement of the LGU financing framework and enable DBP to expand its capacity to finance a wider spectrum of infrastructure and services, and to target a broader range of public and private actors. Also, in the aftermath of the October 2009 disaster brought by Typhoons Pepeng and Ondoy, the Government requested DBP to finance the rehabilitation of infrastructure and services identified in a Post-Disaster Needs Assessment prepared jointly by the Government and Development Partners. The project would therefore provide an avenue for the Bank to support rehabilitation and reconstruction efforts of LGUs affected by natural disasters. Finally, the Bank’s access to global knowledge was important for piloting wholesale lending by PFIs willing to on-lend for local public infrastructure and services. 6. The Project was to contribute to the overall objective of the Country Assistance Strategy’s (CAS) for the period FY2010-20122 of achieving more inclusive 2 World Bank. 2009. Philippines - Country assistance strategy for the period FY2010- 2012. Washington, DC: World Bank. (Project ID: P113433) 2 growth in the Philippines. The Project would particularly support the CAS strategic objective of Improved Investment Climate through (i) increased and improved delivery of infrastructure, (ii) increased investments and employment opportunities in urban and rural areas, and (iii) increased delivery and access to financial services. Project description and costs 7. The project development objective (PDO) in the PAD was: “to stimulate integrative investments within regions by improving access to finance for local public infrastructure and services, and to contribute to the economic recovery of participating LGUs by improving access to local public infrastructure and services.” The formulation in the Loan Agreement (LA) was very similar: “(i) to stimulate integrative investments in the territory of the Guarantor by improving access to finance for local public infrastructure and services; and (ii) to contribute to the economic recovery of LGUs participating in the Project by improving access to local public infrastructure and services.” Key performance indicators (KPIs) were: (i) number of eligible integrative investments included in Provincial, Regional or Sector Development Plans; (ii) percentage of participating LGUs are third and fourth income class; and (iii) percent of participating LGUs have increased or sustained local public infrastructure. 8. The Project was designed as a Financial Intermediary Loan (FIL) with the following components. Component 1: Support for Regional Infrastructure Investment (IBRD $40.0 million, Total $40.0 million). This component was to support improved physical and economic integration within participating provinces or regions by increasing access to finance for an open menu of local public infrastructure and services. Financing was to be provided through either: (i) the Retail Lending Window, to LGUs, the private sector, and public utilities proposing priority subprojects identified by the provincial development council or other regional body and that should contribute to infrastructure and service delivery objectives identified in sector, regional, provincial, city or municipal development/investment pans and (ii) the pilot Wholesale Lending Window that would extend long-term financing to accredited PFIs, including thrift and rural banks, for on-lending to eligible sub- borrowers and subprojects. Component 2: Support for Post Disaster Reconstruction (IBRD $10.0 million, Total $10.0 million). This component was to provide support for post- disaster reconstruction needs of participating LGUs. Post-disaster reconstruction support was to be made available throughout the project period but limited to rehabilitating and reconstructing local public infrastructure and services in LGUs that declare a state of calamity following a natural disaster. 3 Component 3: Institutional Support and Strengthening of Local Service Providers (DBP $0.5 million, IBRD-Water and Sanitation Program $0.4 million, Total $9.0 million). This component was to support the management and supervision of the project, including the provision of technical assistance to eligible local service providers, concerned national agencies, and DBP personnel to enhance prioritization and planning of investments as well as improve efficiency in subproject implementation. Implementation arrangements 9. DBP, a development finance institution with a mandate to support GoP’s development agenda, was designated as the executing agency. DBP had two Program Development (PD) departments that, in coordination with other DBP units, were responsible for overseeing implementation of foreign-funded projects such as RIGP. The PD was organized into project teams whose main responsibility included reviewing project proposals generated by Account Officers. For RIGP, DBP designated a Project Management Team (PMT), composed of staff drawn from DBP’s PD units. The PMT was responsible for the day-to-day implementation of the entire project, including collaborating with other units in DBP to: (i) support wholesale lending; (ii) monitor fiduciary and safeguard compliance; and (iii) monitor overall project implementation progress. Depending on the type of subproject, DBP would identify and bring in internal and external specialists to support the PMT when needed. 10. The DBP marketing units, mostly based at the local level, were the point of contact for most clients. For RIGP, the marketing units were responsible for promoting the project to potential clients, and assisting them in the preparation and submission of subproject proposals and in meeting fiduciary and safeguard requirements. Each sub-borrower (i.e. local service provider) was required to establish a Sub-Project Management Team (SPMT) primarily responsible for supervising and monitoring progress of subproject implementation. The PMT was to provide offsite and onsite assistance to the SPMTs. For the wholesale lending facility, DBP was responsible for reviewing subproject proposals submitted for approval by the PFIs, and monitoring the PFI’s overall condition. 2. Post-Approval Experience and Reasons for Cancellation 11. The Loan was approved by the Board on December 1, 2011, signed on February 2, 2012, and made effective on October 11, 2012. It took the project about three years from concept review stage to effectiveness (See Table 1 below): Table 1. Project Milestones Milestone Date Concept Review Meeting June 26, 2009 Authorization to Appraise/Negotiations May 06, 2010 Bank Approval December 01, 2011 Loan Signing February 02, 2012 4 Effectiveness Date October 11, 2012 Mid-Term Review June 19 to 27, 2014 Cancellation June 19, 2015 12. The factors that contributed to the lack of progress and eventual cancellation are discussed below. a. Implementation delays. From concept review stage, it took about 30 months for the RIGP to be approved. This was mainly on account of the series of reviews of the PAD (i.e. the initial proposal was for a $100 million loan but DBP management decided at the later stage of the appraisal to scale down the loan to $50 million, prompting another round of reviews) and delays in securing the preconditions to negotiations (i.e. GoP Monetary Board Opinion, Special Authority from the President) and presentation to Bank board (i.e. DBP Audit Report for 2010). The project was declared effective (after three loan effectiveness date extensions) 10 months after Board approval despite the fact that there was only one effectiveness condition i.e., the Borrower (DBP) had to adopt the Operations Manual, in form and substance satisfactory to the Bank. b. The Bank conducted the first supervision mission in April 2012, after loan signing in February 2012, to address remaining concerns (i.e. submission of the Operations Manual) and worked with DBP to jumpstart project implementation (i.e. developing an updated pipeline, technical assistance for capacity development). On September 25, 2012, nearly eight months after project signing, the Borrower submitted the Operations Manual, for which the Bank provided its no objection on October 10, 2012. Effectiveness was declared on October 11, 2012. The Project subsequently encountered major delays in launching the facility. In the May 27, 2013 Management Letter, the Bank called DBP’s attention to the little progress made in completing preparatory activities. Six months after effectiveness, DBP had yet to launch RIGP and orient its Marketing Units and prospective sub-borrowers on project requirements and processes. c. Lack of disbursement. DBP failed to develop a viable subproject pipeline and no single project had been approved under RIGP. While DBP provided the Bank a list of subprojects, these lacked the supporting information (e.g., technical studies, feasibility studies, detailed engineering design) necessary to assess eligibility, viability and readiness for implementation. The delay between appraisal and effectiveness saw many prospective LGU sub- borrowers (including those that were affected by Typhoons Pepeng and Ondoy during appraisal) going to other sources of funds to meet their investment or post-disaster reconstruction needs. The local elections in 2013 also contributed to delays in pipeline build-up. The RIGP was to be presented to newly-elected officials after the local elections. 5 13. Underlying reasons for lack of project progress. The lack of disbursement and delays are due to fundamental implementation weaknesses: a. Lack of full-time Project Management Team and supervision consultants. DBP never established a full-time PMT even when the need for one was critically underscored due to lack of project progress. Instead of a dedicated PMT, DBP insisted on a PMT with an ad-hoc composition of personnel that also performed non-project functions in other DBP units. Focus on project concerns was compromised as the efforts of the PMT personnel were divided between the RIGP and their other tasks. DBP was not agreeable to Bank’s recommendations to hire technical consultants to augment the capacity of the PMT even when a credible pipeline of subprojects failed to materialize. b. Change in key counterpart staff. A number of key DBP officials and staff involved during project appraisal and approval either retired or were reassigned at project commencement. This resulted in a PMT that did not fully appreciate the context of the agreements reached during project appraisal and negotiations and contributed to significant delays as it spent considerable time clarifying and contesting the provisions of the PAD and the Loan Agreement (such as those pertaining to the Subproject Loan Agreement) rather than focusing on building a robust pipeline of integrative investments. This also affected the project implementation approach by DBP. Support for integrative investments was a key project feature that was highlighted and agreed with GoP and DBP during design and approval. However, this feature was not pursued in the PMT identification of subprojects as they reverted to the traditional LGU lending business model, i.e., DBP processing much localized LGU investments. c. Exogenous factors. The three project preparation years saw market interest rates declining, especially at the time of project effectiveness. By that time, RIGP’s on-lending rate could not compete with the market and DBP’s internal funds (the average lending rate at the time of implementation was around 4.5%). This was due to the fixed fees (around 4% for sovereign guarantee and exchange rate cover) that GoP imposed that add to the Bank's lending rate (around less than 2%). These make up the base cost upon which DBP adds a spread (about another 2-4%). The non-competitive rates generally overshadowed the Project’s advantage of providing fixed long-term funds and affected project demand.3 It also created an implicit disincentive for DBP to use the Project for its retail and wholesale on-lending operations (because DBP’s internal funds are cheaper and thus, easier to promote). Subprojects in 3 GFIs also price their interest rates based on the risk and credit rating of LGUs. Hence, those with lower credit rating get higher than market rates – this is where RIGP could have competed. As a development Bank, DBP could have introduced enhancements to move RIGP - there are LGUs (lower income class) that appreciate fixed interest-long term credit over low-variable interest credit because the former provide regularity in financial planning. 6 the initial Project pipeline were delisted as the proponent LGUs found lower rates, among other reasons (DBP letter to the Bank dated September 24, 2013). Also, efforts to pilot the wholesale window to supply credit to Small Scale Water Providers (SSWSPs) were faced with fundamental constraints – lack of scale, capacity and information – that underpin the lack of interest of PFIs to lend to SSWSPs. No financial structuring or innovation can truly and sustainably unlock the credit market for SSWSPs without addressing these weaknesses. 14. Steps taken to resolve problems. The following measures were carried out: a. An implementation support mission was undertaken prior to effectiveness to assist DBP in developing the Operations Manual. b. The Bank conducted an orientation workshop on Bank operational policies and project development processes to build the implementation capacity of the PMT. c. Since the start of implementation, the Bank repeatedly communicated to DBP the need to establish a full-time PMT and recruit the necessary implementation consultants to help in developing a robust pipeline of integrative investments. However, DBP maintained its decision that the PMT set-up was sufficient and that consultants would be hired only as needed. d. The Bank provided support to the PMT in formulating a Subproject Loan Agreement (SLA) that was consistent with the requirements of the LA. However, the PMT contested the inclusion of several provisions from the LA into the SLA and requested revisions to the LA instead. The Bank was considering the request but emphasized the need to build the pipeline before modifying the LA. The Bank planned to include the proposed LA modification in the processing of DBP request for restructuring to use project funds to support post Typhoon Yolanda reconstruction efforts (see item “g”). e. The Bank mobilized support from the Bank’s Water and Sanitation Program (WSP) to help DBP build a pipeline under the RIGP wholesale facility, including drafting a wholesale operations manual. A WSP grant amounting to $32,000 was mobilized to support preparation of proposals from small scale water service providers (SSWSPs). The Bank led DBP to possible client LGUs and facilitated discussions with the Department of Interior and Local Government (DILG) that was leading a program for LGUs pursuing water projects. f. On October 22, 2013, the Bank proposed to conduct an implementation support mission to assess the status of subproject pipeline development, agree on a revised implementation plan to address the lack of progress, and to further discuss the changes to the sub-loan agreement. However, on October 29, 2013, DBP requested to defer this mission. g. When super typhoon Yolanda (International Name: Haiyan) struck the country in November 2013 and devastated the central Philippines, discussions shifted to restructuring the Project to utilize the project funds to assist the Typhoon Yolanda-affected LGUs. On February 3, 2014, the Department of Finance (the Guarantor) wrote to the Bank endorsing the DBP’s request for 7 such a restructuring. The Bank saw this as a window to jumpstart project implementation especially since the proposed financing arrangement for the restructured project would involve grant financing from GoP (making RIGP funding attractive as it addresses the issue of RIGP’s uncompetitive lending rates. The Bank agreed to the request to restructure and started preparing the restructuring package, which would have entailed changing the PDO, and amending the Loan Agreement and the SLA. In a letter dated November 11, 2014, the Department of Budget and Management (DBM) confirmed GoP’s commitment to provide grant funding through the Project to finance up to 90% of the cost of LGU reconstruction subprojects. On the basis of this commitment, the Bank encouraged DBP to step up its pipeline generation efforts and suggested certain performance targets be included as covenants in the restructured Project. However, DBP would not agree to the inclusion of any covenant in the Loan Agreement that obliged the Borrower to meet suggested performance targets by certain dates. The restructuring was never approved and, finally on June 19, 2015 DBP requested to cancel the Project. 15. Implications of Cancellation. RIGP was designed to demonstrate a more efficient use of municipal finance through support to integrative investments (e.g. cross-border infrastructure). The cancellation represents a missed opportunity in this regard. Also, the RIGP would have been a natural follow-up engagement with Philippine LGUs. The Bank is supporting a similar operation with another GFI (Land Bank of the Philippines’ Support to Strategic Local Development and Investment Project (Ln. 4833) closing in April 2016), which supported about 75 LGU subprojects since becoming effective on February 28, 2007. It also financed earlier operations implemented by the Department of Finance (i.e., the Local Government Finance and Development Project, Ln. 4446 and the Community Based Resource Management Project, Ln. 4299) which introduced matching grants for LGU subprojects. It should be noted that there are other financing programs for LGUs offered by other Development Partners (e.g. KfW, ADB and JICA, among others). As a result of the RIGP cancellation, financing support from the Bank for municipal infrastructure (through LGUs) came to a halt. 3. Assessment of Bank Performance 16. Quality at Entry was Unsatisfactory. At appraisal, the Bank identified appropriately the core issues that contributed to the slow process of development at the local level: (i) limited access to affordable financing, (ii) high degree of fragmentation of LGUs and (iii) weak NG-LGU coordination in local development planning. The PDO was appropriate as it aimed to support investments that would benefit several LGUs or integrate local economies, address key gaps in local infrastructure provision, and offer support for reconstruction needs in the aftermath of two devastating typhoons that hit the Philippines at the time of appraisal. While the project was designed as an innovative intervention to target integrative investments, it faced challenges on the “demand side” given the institutional and political set-up of LGUs in the 8 Philippines and the conventional approach to local development planning that focuses on individual political jurisdictions rather than broader complementation among adjacent LGUs. Though innovative, targeting integrative investments is a challenge given the institutional and political set-up of LGUs in the Philippines. There was a disconnect between the Project’s focus on the integrative-type and (i) the open menu approach, and (ii) the target LGUs (3rd and 4th class municipalities that, given their income status and capacity, could hardly afford lumpier investments). “Integrative” investments would have been more appropriately proposed at the provincial level (in that a number of municipalities make a province). What constitutes an ‘integrative investment’ was not adequately defined in the PAD and was therefore subject to interpretation during implementation. For example, a majority of subprojects in DBP’s initial project pipeline were water systems that catered to only one LGU; the pipelined subprojects were not different from subprojects supported by the Bank’s other municipal finance operations (e.g. those funded by the SSLDIP of LBP). 17. The project was designed as a FIL, which was appropriate considering DBP's role as a financial intermediary. Key risks were identified during appraisal but additional mitigating measures could have been made to cover two important risks, as follows: a. Market risk. The PAD discussion on the wholesale lending window recognized the market liquidity overhang but indicated the possibility that demand could pick up once market liquidity was reduced (PAD Annex 4, paragraph 15). The appraisal team also expected DBP to fine-tune its wholesale loans to ensure attractiveness to PFIs. But excess market liquidity persisted throughout implementation, driving domestic market rates lower and making the Project’s lending rates uncompetitive. The low domestic market rates likewise affected the demand for the retail window. The possibility of a project operating in a low interest rate environment should have been identified as a key risk, and the project could have been designed to offer options for subproject financing under such market conditions.4 b. Institutional readiness of DBP. Given the open menu approach, the PAD rightly assessed that DBP had limited technical capacity in implementing the project. As a mitigating measure, a PMT would be set up to provide implementation support and DBP would assign internal or hire external specialists as the need arose. The staff assigned to the PMT however did not serve on a full-time basis. Also, despite being encouraged to do so, DBP did not hire technical consultants to augment PMT’s capacity during implementation. Stronger measures could have been identified at appraisal to 4 During implementation, when Typhoon Yolanda affected a number of LGUs in 2013, the Government proposed to restructure the project to provide financing to affected LGUs with GoP counterpart grant funding. The provision of the grant eventually addressed the issue of uncompetitive lending rates (see paragraph 14 item g) and thus opening a window for the project to progress despite the low interest rate environment. 9 mitigate capacity risks. These measures include requiring DBP to commit to a timeline for the establishment of a full-time implementing unit or PMO or the hiring of full-time technical consultant after a certain period after loan effectiveness). This should have been agreed given previous Bank operations under DBP were rated unsatisfactory. 18. Bank’s Supervision and Implementation Assistance is Moderately Satisfactory. The Bank conducted a supervision mission immediately after loan signing to help the Borrower in meeting the effectiveness condition and developing a pipeline of subprojects. The Bank had identified key implementation bottlenecks at the outset and repeatedly brought the need to address performance shortcomings to the attention of PMT and DBP. Considering that the PMT members were mostly not involved in project design and appraisal, the Bank immediately organized and conducted an orientation workshop to support their understanding of the project Measures were suggested to remedy the lack of progress, including the hiring of full-time consultants to augment PMT’s capacity, and the possibility of including performance targets as a covenant during the restructuring (see para 14 items a to f). 19. The Bank was candid with the Borrower with respect to evaluating progress towards the achievement of the PDO and implementation progress during implementation review missions. Aside from the missions, the Bank was in constant coordination with the Borrower to catch-up on the progress of agreed actions. It also extended significant efforts (e.g. conducted an orientation workshop to build implementation capacity, tapped WSP funding to try piloting wholesale lending, etc., see par14) to help the Borrower resolve implementation issues and develop a robust pipeline. 20. Considering the project's long preparation period, changes in leadership and staffing at DBP, as well as changing market conditions, the Bank could have proposed to restructure the project earlier, i.e., after effectiveness. However, the Bank deemed it extremely necessary to put in place the fundamental elements to implement the project prior to any restructuring: (i) preparation of a sub- loan agreement that fully complies with RIGP Loan Agreement, (ii) establishment of and strengthening of a project implementing unit, and (iii) development of a subproject pipeline. The Bank provided necessary support to DBP in these aspects. The Bank further felt that restructuring would be more meaningful if it is informed by lessons generated from on-the-ground implementation of sub- projects.5 When DBP continued to have difficulties in building a credible pipeline a year after project effectiveness, the Bank downgraded the project’s implementation rating from “moderately unsatisfactory” to “unsatisfactory,” 5 Note also that there was a change in DBP top management (a new DBP President was appointed) hence it was necessary to allow the new management the opportunity to turn the project around. Therefore it did not seem right to restructure the project without giving the new management this chance. 10 flagging RIGP as a problematic project. The Bank provided options for DBP to pursue, including (i) restructuring the project to provide support to Yolanda- affected LGUs, and (ii) project cancellation when it was clear that the restructuring was unlikely to move forward. As the Bank could not cancel the loan unilaterally, it considered suspending the project by a certain timeline and pursue cancellation after six months of the suspension date if no progress was made. But this course of action became irrelevant as DBP eventually decided to cancel the Project. 4. Assessment of Borrower Performance 21. Performance of the Government is Moderately Satisfactory. GoP was supportive during preparation and implementation. It pursued discussions with the Borrower to resolve disbursement and implementation issues as part of monitoring of foreign-funded projects. The monitoring, however, seemed to have fallen short in terms of proposing measures to the Borrower and the Bank to address the implementation bottlenecks. 22. Performance of the Implementing Agency is Unsatisfactory: Loan effectiveness took much longer than expected due to delays in obtaining final government clearances for the loan and finalizing the Project OM. Subsequent delays were encountered by DBP in launching the product due to slow preparatory activities. Despite the support provided by the Bank and WSP to DBP in developing a pipeline of subprojects, no investment subprojects with sufficient information to determine viability and eligibility for funding under the Project’s requirements were put forward by DBP for Bank review. A major bottleneck in project implementation stemmed from the lack of a full-time PMT that could focus on moving the project forward. DBP also did not recognize the need to hire implementation consultants to augment capacity despite the apparent difficulty in generating a credible, duly appraised pipeline. Despite repeated communication of the need to address this issue, especially when there was the possibility of RIGP increasing in size to support Yolanda reconstruction activities, DBP maintained its position that the existing set-up was sufficient to implement the project. 23. Though the change in financial market conditions during implementation affected the competitiveness of RIGP, DBP did not offer concrete actions to enable it to adapt to these market realities. DBP could have suggested to restructure the project, assess what adjustments could be made and offer a more appropriate lending instrument to the LGUs and local service providers, and review whether the wholesale window was still viable or, as suggested in the PAD, adjust its lending rates or spread so that PFIs could find the lending facility attractive. 24. Due to the lack of proactive measures by DBP to address implementation bottlenecks, the implementing agency performance is rated unsatisfactory. 11 5. Lessons Learned 25. The project offers important lessons that could be useful in design and supervision of municipal finance operations: a. Realistic assessment of implementation capacity and Borrower commitment is key. Though the implementation capacity risk had been identified during appraisal, the mitigating measures identified in the PAD did not concretely identify actions that could adequately address the gaps in implementation capacity relative to the risk of DBP not having a full-time and dedicated Project unit. Establishment of project management offices fully staffed with the required skills would be an important condition for effectiveness. Recruitment of technical consultants at project commencement is another important condition. Unwillingness of DBP management to recruit full time staff and implementation support consultant should have been a concrete sign of fundamental disagreement between the Bank and the Borrower on how the project was to be implemented and an impetus for assessing Borrower commitment to the project design. b. Incentives for performance (and sanctions for non-performance) are still important features in project design. There are simply no such features considered in project design. Because the PMT was comprised of composite staff who are remunerated regardless of involvement in the project, the project staff are not compelled to generated business out of the project (unlike in cases where project staff are full time hires and are normally compensated based generally on project performance, or that there is a stake in project success or failure). To compound the problem, the commitment fee on the RIGP loan (having been negotiated and signed after September 27, 20076) was waived and any delay in implementation was not seen as having any cost implication to the Borrower. This lack of financial implication to the Borrower does not compel even to restructure or cancel the loan even if it is not performing. c. Greater engagement of national government in oversight of problematic projects. Involvement of oversight agencies in proactive monitoring of problematic projects would be extremely helpful as it would allow early dialogue on contentious issues (such as project staffing and management). A mechanism can be supported that will strengthen national government role to evaluate progress and proactively propose measures that the Implementing Agency should undertake to address performance shortcomings. 6 http://siteresources.worldbank.org/PROJECTS/Resources/40940- 1250176637898/Engl.pdf 12 13 Annex 1. Bank Lending and Implementation Support/Supervision Processes (a) Task Team members Responsibility/ Names Title Unit Specialty Lending Christopher Casuga Ancheta Sr Sanitary Engineer GWADR Municipal Engineer Rosemary Mukami Kariuki Lead Water and Sanitation Specialist GWADR Urban Specialist EASSD - Kamran M. Khan Program Manager HIS Social Safeguards Victoria Florian S. Lazaro Operations Officer GSURR Specialist Procurement Rene SD. Manuel Senior Procurement Specialist GGODR Specialist Urban Mgmt. Lawrence Tang Urban Mgmt. Spec. GSURR Specialist Environmental Maya Gabriela Q. Villaluz Senior Operations Officer GENDR Safeguards Specialist Supervision Christopher T. Pablo Senior Urban Specialist GSURR Task Team Leader Artessa Saldivar-Sali Municipal Engineer GSURR Municipal Engineer Catherine Vidar DRM Specialist GSURR DRM Specialist Social Safeguards Victoria Florian S. Lazaro Operations Officer GSURR Specialist Environmental Maya Gabriela Q. Villaluz Senior Operations Officer GENDR Safeguards Specialist Procurement Rene SD. Manuel Senior Procurement Specialist GGODR Specialist Financial Aisha Lanette De Guzman Financial Management Specialist GGODR Management Specialist Kristine May SJ Ante Program Assistant EACPF Program Assistant Mari Trillana Team Assistant EACPF Team Assistant Municipal Finance Rhoel Bernardo Consultant GSURR Specialist Marilyn Martinez Consultant GSURR Urban Specialist Rosanna Martin Consultant GSURR Project Specialist Abegyl Albano Consultant GSURR M&E Specialist M&E/Project Hope Gerochi Consultant GSURR Specialist 14 (b) Staff Time and Cost Staff Time and Cost (Bank Budget Only) Stage of Project Cycle USD Thousands (including No. of staff weeks travel and consultant costs) Lending FY 2008 26.54 Total: 26.54 Supervision FY 2014 50,555.67 FY 2015 68,716.42 FY 2016 7,918.36 Total: 127,190.45 15 Annex 2. List of Supporting Documents 1. Technical Preparation Mission Aide Memoire, February 1-12, 2010 2. Project Appraisal Document, November 02, 2011 3. World Bank Implementation Review Mission Management Letter and Aide Memoire dated: a. May 15, 2012 b. June 26, 2013 4. World Bank Medium Term Review Mission Management Letter and Aide Memoire dated July 14, 2014 5. DBP Letters dated: a. October 29, 2013 (Implementation Support Mission deferment) b. April 5, 2013, September 24, 2013, Dec 9, 2013 and February 24, 2015 (on proposed Loan Agreement revisions, project pipeline and revised operations manual) c. July 15, 2015 (confirming DBP Management consent to restructure for Yolanda) d. July 28, 2014 (Response to findings of the Medium Term Review mission) e. May 6, 2015 and June 19, 2015 (Proposal to Cancel) 6. DOF letter dated February 3, 2014 endorsing request to restructure RIGP for Yolanda 7. DBM letter to the Bank dated November 11, 2014 confirming commitment for national government grant infusion 8. WSP (2014, draft note). Lessons on Credit Financing for Water Supply Projects in the Philippines: A Case Study on an Attempt to Introduce a Water Credit Facility in a World Bank Financial Intermediary Lending Project for Local Infrastructure Development. 16 IBRD 33466R4 I Ilocos X Northern Mindanao 1 2 3 Ilocos Norte Ilocos Sur La Union 58 59 60 Bukidnon Camiguin Lanao del Norte Basco Batan Islands PH I L I PPI NES 4 Pangasinan 61 Misamis Occidental 62 Misamis Oriental 11 SELECTED CITIES CAR Cordillera Admin. Reg. 20ºN 5 Abra XI Davao Reg. Luzon Strait PROVINCE CAPITALS 6 Apayao 63 Compostela Valley 7 Benguet 64 Davao del Norte REGION CAPITALS 8 Ifugao 65 Davao del Sur 9 Kalinga 66 Davao Oriental 10 Mountain Province NATIONAL CAPITAL Babuyan XII SOCCSKSARGEN Islands II Cagayan Valley 67 North Cotabato RIVERS 11 Batanes 68 Sarangani Babuyan Channel 12 Cagayan 69 South Cotabato MAIN ROADS 13 Isabela 70 Sultan Kudarat 14 Nueva Vizcaya RAILROADS 15 Quirino XIII Caraga Laoag City 1 6 71 Agusan del Norte 12 III Central Luzon Kabugao PROVINCE BOUNDARIES 72 Agusan del Sur 16 Aurora 73 Dinagat Islands Bangued 17 Bataan 74 Surigao del Norte Tuguegarao REGION BOUNDARIES Vigan 5 9 18 Bulacan 75 Surigao del Sur 19 Nueva Ecija CAR Tabuk II INTERNATIONAL BOUNDARIES 20 Pampanga ARMM Autonomous Reg. in Bontoc Ilagan 21 Tarlac Muslim Mindanao I 2 10 22 Zambales Lagawe 13 NCR National Capital Reg. 76 77 Basilan Lanao del Sur San Fernando 8 Luzon 125ºE 78 Maguindanao** 3 La Trinidad Cabarroguis 79 Sulu Baguio Bayombong IV-A CALABARZON 80 Tawi-Tawi 7 14 15 23 Batangas Lingayen 24 Cavite 25 Laguna ** Shariff Aguak (Maganoy) and 16 Sultan Kudarat serve as co-capitals 4 26 Quezon of the province. Baler III 27 Rizal Palayan Tarlac 19 IV-B MIMAROPA Iba 21 PHILIPPINES 28 Marinduque 22 San Fernando IV-A 29 Mindoro Occidental 20 18 Polillo 30 Mindoro Oriental Malolos Islands 31 Palawan* Balanga Quezon 17 Pasig 32 Romblon NCR MANILA Antipolo Trece Martires 27 34 V Bicol 24 Santa Cruz Daet V 33 Albay Calamba25 26 34 Camarines Norte 23 35 Camarines Sur Lucena 35 36 Lubang Batangas Catanduanes 36 Catanduanes Islands Pili Virac 37 Masbate Calapan Boac 38 Sorsogon 28 Mamburao Marinduque 30 Legaspi VI Western Visayas 39 Aklan Mindoro Sibuyan 33 Sorsogon Phil i ppi n e Burias 38 40 Antique 29 Sea 41 42 Capiz Guimaras M ind Romblon Ticao Catarman Sea or Tablas Masbate 43 Iloilo o Str 32 Sibuyan 52 Samar 44 Negros Occidental Busuanga 37 ai t Masbate VII Central Visayas 53 Semirara Visayan 45 Bohol Culion Islands Kalibo Catbalogan VIII 46 Cebu Roxas City Sea 47 Negros Oriental Linapacah 39 Naval 49 50 Borongan 48 Siquijor 41 Panay Tacloban Cuyo VIII Eastern Visayas Islands 40 43 51 Leyte 49 Biliran Jordan Cebu Leyte Gulf 50 Eastern Samar San Jose de Iloilo 51 Leyte Buenavista Bacolod 42 46 52 Northern Samar Dumaran 53 Samar Cebu 54 Dinagat 54 Southern Leyte Maasin IV-B VI Negros VII 73 San Jose 10ºN IX Zamboanga Peninsula 44 45 Bohol Siargao 55 Zamboanga del Norte Puerto Princesa Tagbilaran Surigao 56 Zamboanga del Sur 31 74 47 57 Zamboanga Sibugay Mindanao --- Zamboanga City Palawan Dumaguete Siquijor Sea Mambajao 71 XIII 48 Siquijor 59 Camiguin Tandag * Executive Order 429, May 23, 2005, 75 provides for the transfer of Palawan X Butuan Cagayan province (#31) from Region IV to Region VI; Administrative Order 129 holds EO429 in abeyance until an Sulu Sea IX Dipolog Oroquieta de Oro Prosperidad implementation plan is approved 62 72 by the President. Bugsuk 61 Iligan 55 Malaybalay 58 60 Marawi Balabac Pagadian Ipil Tubod 77 Mindanao 63 57 56 64 Nabunturan Tagum Cotabato 67 66 Cagayan Davao Sulu ZAMBOANGA CITY Moro 78 Mati XI Shariff Aguak Kidapawan Isabela Zamboanga Gulf (Maganoy) Digos City ARMM Isulan Davao Basilan 76 70 Koronadal Gulf This map was produced by 69 65 the Map Design Unit of The Jolo Sulu General Alabel World Bank. The boundaries, MALAYS IA M A LAYS IA XII Santos 68 colors, denominations and any other information shown on this map do not imply, on 79 the part of The World Bank 0 50 100 150 Kilometers Tawi-Tawi Group, any judgment on the Sarangani 80 Panglima Celebes Sea legal status of any territory, or any endorsement or 5ºN Sugala acceptance of such 0 50 100 Miles 125ºE boundaries. 120ºE NOVEMBER 2012