Philippines - Third Urban Development Project

Document of The World Bank FOR OFFICIAL USE ONLY r Report No. 7897 PROJECT COMPLETION REPORT PHILIPPINES THIRD URBAN DEVELOPMENT PROJECT (LOAN 1821-PH) JUNE 30, 1989 Infrastructure Division Country Department II Asia Region This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  ABBREVIATIONS AND ACRONYMS CURRENCY EQUIVALENTS Currenc, Unit - Peso *P) US$: - P21.0 (comnletion) US$1 P7.4 (apn)raisal) WEIGHTS AND MEASURES 1 meter (m) - 39.37 inches (in) 1 square meter (sq m) - 10.8 square feet (sq ft) 1 cubic meter (cu m) - 35.3 cubic feet (cu ft) 1 kilometer (km) - 0.62 mile (mi) 1 square kilometer (sq km) m 0.386 square mile (sq mi) I hectare (ha) - 10,000 square meters (sq m or 2.47 acres (ac) ABBREVIATIONS AND ACRONYMS BLISS - Bagong Lipunan Improvement of Sites and Services CIF - Capital Improvement Folio COA - Commission on Audit DBP - Development Bank of the Philippines ESC - Environmental Sanitation Center LOI - Letters of Instruction MCB - Management Coordination Board NHS - Ministry of Human Settlements MMINUTE - - Metro-Manila Infrastructure, Utilities and Engineering Program MMA - Metro Manila Area MMBIDP - Metro-Manila Barangay Industries Development Program MKC - Metro Manila Commission NMTA - Metro Manila Transport Authority MOB - Ministry of Budget MOF - Ministry of Finance MPH - Ministry of Public Highways MPW - Ministry of Public Works MWSS - Metropolitan Water and Sewerage System NACIDA - National Cottage Industry Dev4Iopment Authority NEDA - National Economic and Development Authority NHA - National Housing Authority NHMFC - National Home Mortgage Finance Corporation NMYC - National Manpower and Youth Council PBSP - Philippine Business for Social Progress PD - Presidential Decree PROGRESS - Program for Removing Se,age from Streets RAD - Research and Analysis Division SIR - Slum Improvement and Resettlement Program SSE - Small Scale Enterprise TRC - Technology Resource Center TUVD - Technology Utilization Ventures Division UNDP - United Nations Development Program UPISSI - University of the Philippines Institute for Small Scale Industry ZIP - Zonal Improvement Program FISCAL YEAR January 1 to December 31  FOR OMCIAL USE ONLY THE WORLD BANK Washtngton. DC. 20433 U.S.A. OMice d4 DOwectt(.ClhI -, Operatmm EvoakiatKi June 30, 1989 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT STThJECT: Project Completion Report _n Philippines - Third Urban Development Project (Loan 1821-PH) Attached, for information, is a copy of a report entitled "Project Completion Report on Philippines - Third Urban Developmrat Project (Loan 1821-PH)' prepared by the Asia Regional Office with Part II of the report contributed by the Borrower. No audit of this project has been made by the Operations Evaluation Department at this time. Attachment I This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  FOR OMCIAL USE ONLY PROJECT COMPLETION REPORT PHILIPPINES THIRD URBAN DEVELOPMENT PROJECT (LOAN 1821-PH) Table of Contents Preface .................................................... i Evaluation Summary ......................................... ii I. Part I Project Identity ................................... 1 Background ......................................... 1 Project Objectives and Description ................. 1 Project Highlights ................................. 1 Lessons ............................................ 4 Remaining Issues ................................... 8 Attachment 1: Table of Key Indicies ................ 12 Attachment 2: Inflation ............................ 14 II. Pa ft II The Borrower's Project Completion Report ..... 15-74 III. Part III Summary of Statistical Data Table 1: Related Bank Loans .76 Table 2: Project Timetable .76 Table 3: Loan Disbursements .77 Table 4: Project Implementation (Key Indices) 78 Table 5: Project Costs and Financing .78 Table 6B: Economic Benefits ...................... 78 Table 6D: Studies ................................ 80 Table 7: Compliance with Loan Covenants ......... 81 Table 8: Use of Bank Resources .................. 82 ATTACHMENT - Comments from the Borrower ..... ............... 84 This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  PROJECT COMPLETION REPORT PHIL'PPINES THIRD URBAN DEVELOPMENT PROJECT (LOAN 1821-PH) PREFACE This is the Project Completion Report (PCR) for the Third Urban Development Project in Philippines, for wh!.ch Loan 1821-PH in the amount of US$72 million was approved on March 23, 1980. The loan was closed on December 31, 1987, three years behind schedule. During implementation, US$1.0 million was cancelled in November 1985. The last disbursement was made on July 11, 1988 and the undisbursed loan balance of US$3,895,100.12 was cancelled effective July 12, 1988. The PCR was jointly prepared by the Infrastructure Division, Country Department II of the Asia Region (Preface, Evaluation Sulmmary, Parts I and III) and the Borrower (Part II). Borrower provided comments on the PCR (Attacbment). Preparation of this PCR was started during the Bank's final super- vision mission of the project in July 1988, and is based, inter alia, on the Staff Appraisal Report; the Loan, Guarantee, and Project Agreements; supervision reports; correspondence between the Bank and the Borrower; and internal Bank memoranda.  - ii - PROJECT COMPLETION REPORT PHILIPPINES THIRD URBAN DEVELOPMENT PROJECT (LOAN 1821-PH) EVALUATION SUMMARY Introduction 1. The project was the third Bank urban development operation in the Philippines. It represented the continuation of Bank assistaAce to the urbaa sector initiated in 1976 with the purpose of providing basic shelter improvements to the urban poor, primarily through slum upgrading and sites and services. The Bank loan of US$72 million to the Government of the Republic of the Philippines was approved on March 23, 1980 and was closed on December 31, 1987, three years after the original closing date. Obiectives 2. The objective of the operation was to continue assistance to the Philippine Government in developing affordable solutions to the problems of poverty, shelter and environmental sanitation in the Manila Metropolitan Area (Part II, para. 1.03). Implementation Experience 3. Project implementation encountered severe economic and political turbulence resulting in three one year extensicas (Part I, para. 4). Total project costs were estimated at appraisal to be US$120, and while the final dollar equivalent was only marginally above this figure, project costs in pesos rose from the P800 million expected at appraisal to P1,804.7 million at closing. This increase was due to a combination of domestic inflation and the significant devaluation of the peso relative to the dollar during the implementation period. The near equivalence of dollar costs reflects the programmatic nature of the operation, as the quantity of physical works was expanded to maximize utilizatioui of the dollar value of the loan (Part I, para. 5). Out of the original Bank loan, nevertheless, US$1.0 million was cancelled in November 1985 and another US$3.9 million was cancelled at the time of last disbursement in July 1988. Results 4. Despite implementation delays, project physical achievements for slum upgrading and sites and services subprojects eventually reached targeted levels and, in the case of several components, significantly exceeded them (Part I, Attachment 1). The small business component was also largely successful, while the level of up front subsidies implicit in the below cost pricing of project-generated shelter solutions was significantly decreased vis-a-vis the experience under the two previous projects (Part I, paras. 6- 11). On the negative side, however, serious problems with lot sales and  - iii - mortgage and home improvement loan collections persisted (Part I, paras. 38- 40). Furthermore, the project proved to be too complex in design, having eight distinct components involving five different agencies, and overly ambitious in its technical assistance and institutional strengthening objectives (Part I, paras. 12-19). Sustainability 5. While benefits deriving from improved shelter and access to urban services on the part of many low-income families as a result of the project are likely to be sustatned, the poor cost recovery performance under this and the two previous uroan projects calls into question the ability of the National Housing Authority to replicate sites and services and, especially, squatter upgrading efforts on a broad scale. Findings and Lessons 6. This project, as one of a series of very ambitious urban shelter and service improvement operations financed by the Bank in the Philippines over the past decade and a half was prepared at a time when the institutionally demanding, multi-component approach to urban development was still in vogue and before the Bank had become fully aware of the institutional and financial -- especially cost recovery -- difficulties experienced by Urban I and II. As a result many of the problems observed in the two earlier operations also occurred in the present project. Urban III, nevertheless, made several notable contributions, including the consolidation of the relatively cost effective and socially less lisruptive slum upgrading/sites and services approach to the provision of housing and basic services for the urban poor in metropolitan Manila and, in the process, benefitted large numbers of low-income people and attained effective involvement of the private sector. 7. The principal lessons of this institutionally complex and financially disappointing urban operation are the need to make such projects more manageable by following a subsectoral approach (Part 1, para. 14) and to focus greater attention on the establishment and administration of effective cost recovery mechanisms (Part I, para. 39). A related lesson is the need to rationalize the regulatory and bureaucratic framework in which urban land development and shelter provision presently takes place in the Philippines (Part I, paras. 31-32).  PART I 1. PROJECT IDENTITY >oject Same: Third Urban Development Project Loan Number: 1821-PH RVP Unit: Asia Country: Philippines Sector: Urban Development Subsectors: shelter, small business lending, infrastructure municipal development II. BACKGROlnND 1. Sector Background. Develoament Obiectives. and Policy Context: The project was intended as a multi-faceted urban development project designed to contii.ie rnew low cost solutions to shelter problems (upgrading through both ZIP and MMINJTE, and public sector serviced sites) and test a number of innovative new ideas (private sector serviced sites, commercially viable micro- enterprise lending schemes, internal MMC solid waste operations), as well as significantly strengthen the newly created Metro Manila Commission (MMC) whose job it was to coordinate development in the metronolitan area. See paras 1.01 -1.07 of Part II. III. PROJECT OBJECTIVES AND DESCRIPTION 2. Project Objectives and Components: The project objectives are spelled out in para 1.03 of Part II. The main project components were: (a) ZIP upgrading with tenure, (b) MMINUTE upgrading without tenure, (c) Public Serviced sites by NHA, (d) private sector serviced sites utilizing construction loans by PSSP, (e) small business loans by TRC's ULFP, (f) Solid Waste Improvement by MMC, (g) Taxmapping by MMC and Local Governments, and (h) a complex set of technical assistance and studies for the above agencies plus several others. See para 2.01 of Part II. IV, PROJECT HIGHLIGHTS. LESSONS AND ISSUES 3. The attached PCR prepared by the Borrower (Part II) fully discusses project results, and distills a number of useful lessons emerging from the experience of Urban III. It is generally appropriately focused on policy issues and future directions. This section (Part I) will therefore highlight a few of the more important or interesting features of the project, cover any key points omitted, and note the Bank's perspective where it differs from the that of the Borrower. Comments have been received from the Borrower and incorporated as appropriate in all three sections of this report. Hip hlights 4. #1: Project Restructuring and Physical Expansion. Project implementation encountered severe economic and political turbulence, requiring three extensions to the closing date. However, physical achievements eventually reached targetted levels, and in the case of several components, significantly exceeded them. The Key Indices Table (Attachment 1) shows NHA's ZIP upgrading program serviced about 26,000 plots, 53% more than the SAR target of 17,000; 6,000 sites were developed in Pasig vs. 3,300 targeted (nearly  -2- twice); the PSSP serviced over 4,000 lots vs. 2,800 targeted (a 43% increase); and the Taxmapping pr,.'ect, which achieved a remarkable 15 month cost recovery based only on partial completion, was added to the project. These accomplishments were possible, of course, because of the dramatic depreciation of the peso from 7.4 at appraisal to 21.0 at completion, and greater buying power of the project's dollars. The unprecedented inflation (Attachment 2) which triggered what theoretically should have been approximately equivalent devaluation, actually resulted in additional purchasing capacity of the project's dollars. 5. The use of dollar savings was endorsed by the Bank through a formal restructuring of the project in 1985, which acknowledged the need for an additional two years for completion and, in effect, shifted the project more toward programmatic support of serviced sites, upgrading and small business lending. Thus the lengthened implementation has yielded additionality in project benefits and should not be characterized strictly as a time overrun. vlork was completed in nine rather than five years, an extension of 80%. The nine year actual implementation period is about average for Asia Region Urban projects (8 vears). The initial four year implementation schedule was therefore unrealistic for such a complex operation; it represented a target figure. The close achievement of dollar costs ($120.4 vs. $120.0 at appraisal) also reflects the programmatic nature of the oneration more than any great accuracy of either the initial estimates or implementation achievements, as physical works were expanded to target and mop up all the available dollars under the loan. 6. #2- Success of Small Business Lending Scheme. The Urban Livelihood Financing Program (ULFP) component which was designed to test whether lending operations to small scale informal sector entrepreneurs could be done on a financially viable basis, appears to have succeeded. The .onger term objective was to ultimately bring commercial banks into this segment of microenterprise lending operations. The Socialized Credit Scheme (SCS) made credit available to groups rather than individuals, collateralized with small personal goods with resale value such as TVs, refrigerators, bicycles, etc.. Fluctuation in individual monthly incomes were covered by the group, since default by the group resulted in termination of the credit line for all individuals. In addition to SCS, the ULFP program experimented with equity positions and lending to selected medium size businesses. A total of $6.5 aillion of the Bank's loan was invested in ULFP. 7. Despite a very rocky start, and five changes in leadership, this component appears to have achieved its financial objectives. In 1986 and 1987 the ULFP reported a real return (after taxes and inflation) (f between 1%-2%. This is generally considered sufficient to attract private capital. This is partly due, of course, to the unusually low inflation rates of these two years (0.8 and 3.8% respectively), and to the earnings being realized on ULFP's high liquidity (due to short term lending). Early equity investments and a few large, insufficiently secured loans turned sour and are presently being liquidated, thus the overhang of past due arrearage and foreclosed assets is high. However, net of these old arrears, collections are now an encouraging 97% of current amounts due. Together with ULFP's loan loss reserve (presently 6.8% of outstanding loan principal) this would be sufficient to sustain profitability.  - 3 - 8. Rapid inflation in early 1988 (from 3.8% to 9% in 3 months) required a sharp upward adjustment in rates in mid year for ULFP to remain profitable. Year-end results are not yet available. However, if the program's financial success can be sustained, there is re&son to hope that the ULFP can and will be replicated in larger scale by commercial banks. As ULFP grows, such a move should be encouraged, since this operation is essentially a banking function within TRC and is not presently subject to any kind of regulatory inspection or control. 9. #3: Lowered Subsidies (InMroved Pricing by NHA). Cost recovery involves three features: pricing, sales and collections. While the latter two remained problematic in Urban III (see paras 18 and 22 for local government performance, and para 35 for NHA performance), improvement occurred over Urban 1 and II in pricing. Direct charging of development costs comprises a key element in helping ensure greater replicability and cost recovery, since any capital costs not directly charged to recipients either (a) increases the subsidy burxen on government, or (b) is recovered through indirect means such as MWSS tariffs which have proven problematic and elusive in actually achieving full recovery. 10. Annex 6 of Part II shows a distinct increase under UrbFn III in NHA's direct allocation of development costs to beneficiaries. Under previous projects, sometimes ad-hoc pricing decisions placed a major burder of cost sharing on subsidy (NHA, local and national governments) and on indirect recovery through MWSS tariffs. Because the MWSS' tariff structure is cross- subsidized, full recovery from low income groups is not generally possible without a restructuring of the tariff. Under Urban I only 37% of development cost was directly charged to beneficiaries, with 35% charged to subsidy (NHA 17% and other line agencies 18%, and the balance charged to MWSS (28%). Under Urban II direct ch.rges increased to 50%, with subsidies still bearing over a third (37%) (NHA 3% and other line agencies 34%) and MWSS shouldering about 13%. Following continuing improvements in pricing policy, Urban III achieved direct recovery of 74% of total development costs, with subsidy of only 10% or less than a third of that under Urban II, and indirect charges through MWSS still about 16%. ZIP sites were best, with Phase I sites achieving 71% direct recovery from beneficiaries, and Phase II fully 84%. Pasig sites and services trailed significantly with only 64% directly recovered, but still well ahead of the NHA's first major serviced site of Dagat Dagatan, which achieved only 37%. 11. The best performance with regard to cost allocation, however, was delivered by PSSP, under which fully 94% of costs were recovered through direct charges to beneficiaries. The balance 6% was subsidized. Such a charging structure implies that the use of private sector developers capable of utilizing smaller infill pockets of land (which require little expensive trunk infrastructure) can and will significantly reduce public sector expenditures for shelter development. Direct Charge Subsidy Indirect Charges to B .nficiaries ---------------------------------- tbrough NMA LGUs Nti Govt Total MWSS Urban I (tnol Tondo) 37S 172 --- 182 35X 28X Urban II 50S --- 32 34Z 372 132 Urban III (NBA) 742 22 22 62 lO 16S POOLg SI$ 64X --- --- 112 11X 25X ZIP Phase I 71X 12 62 62 132 162 ZIP Pbase II 842 5S --- 3X 8S 8S PSSP 942 --- --- 62 62  -4. Lessons 12. Lesson #1: Project Complexity. "overnment's PCR ciearly outlines the problems which arose from attempting such a complex project with eight quite distinct components involving five different agencies (plus the newly created Management Coordination Board (MCB)). All of the componen:s were complex programs in themselves (four involving essentially banking operations). Several represented quite innovative experiments requiring constant reassessment of ongoing experience and adjustment of approach during implementation. To manage this process, the MCB was established with its own Technical Secretariat (TS) which gradually inflated to fully 120 people at its peak, spending a total of $2.1 million. None of the project components were linked and each functioned essentially independently of one another. While the MCB Board did provide a useful forum for interagency dialogue, expenditures on the Technical Secretariat (which were not financed by the Bank) comprised an unproductive use of scarce government funds. 13. The establishment of the MCB/TS was essentially for the convenience of the Bank, in managing what was otherwise an unmanageable project. Fully 32 reports required review each year (seven individual quarterly reports plus four annual audit reports). The SAR estimated a requirement of 25 staffweeks (SW) of field effort for the first two years, and 15 SW per year thereafter, totalling 285 including office time through the end of 1984, four years later, when completior was expected. Actual SW inputs over the project life from 1980-1988 tota:'.Led only 161 or about half. Viewed another way, the SAR planned for 57 SW per year for loan supervision, and subsequent budgetary realities delivered 18, or less than a third of that anticipated. Under ideal circumstances, this would have been a difficult project. The subsequent collapse of the Philippine economy during the heart of the implementation period from 1983-1986 (para 3.01 of Part II) made the task yet more problematic. 14. As noted in the PCR on Urban II, subsequent Bank operations in the urban sector have been structured along narrower sub-sectoral lines, allowing greater focus on sectoral policy and institutions. The recently approved Housing Sector Loan and Municipal Development I and II projects offer examples. Such operations have not required government to establish artificial coordinating structures with full time staff to manage project complexity, and should hopefully prove more compatible with the tighter Bank supervision coefficients of recent years. The price of greater focus has been the elimination of support for some of the activities (e.g. TRC's ULFP small business loan program discussed above) which proved viable as an experiment and merit eventual expansion. 15. Lesson #2: Studies Overload. Part III, Table 6D lists the 13 studies undertaken as part of Urban III. As a general point, this was far too many studies to be well supervised by the Bank and well absorbed by the Borrower. Fully 90 staff years of TA was programmed at appraisal, 551 and 529 manmonths of foreign and local advisory assistance. All required the Bank's normal rcview of TOR, procurement and contracting which proved time consuming. The appraisal promised an exceptionally wide array of institutional strengthening, particularly with regard to MMC, nearly all of which require complex behavioral  - 5 - and systemic change within institutions which occurs only slowly and only through intensive, consistent and committed effort by all concerned. With the benefit of greater experience as well as hindsight, the project's institutional TA objectives are now recognized as overly ambitious. 16. A number of the studies have proven useful. The Construction Industry Study, Taxmapping study/TA, Solid Waste Master Plan Update, TEAM Traffic Management, and project preparations for the Municipal Development I (PREMIUMED) and Regional Cities Development Project (Urban IV, RCDP) both of which resulted in Bank loans, all achieved identifiable results and are generally considered useful. The Urban V preparation, involving another complex multi-faceted urban project, was aborted and replaced as noted above, with the separate housing sectot and municipal loan streams. 17. Lesson #3: Overly Ambitious Institutional Strengthening Obiectives for MMC. The remaining 6 studies targeted on strengthening MMC's capabilities in municipal management, on which about $1.4 million was spent, ultimately produced little lasting impact, due largely to the MMCs increasing politization during the Marcos years and eventual dismantling under the new governrment. These studies were all contracted in the first three years of the project, and included (a) a Management Review to identify areas where MMC should build capabilities, (b) Physical Planning to develop a physical framework plan, (c) Financial Systems to improve fiscal management, (d) Capital Investment Folio to establish a rational basis for project prioritization and investment, (e) Urban Land Management to study land utilization and regulation, and (f) Physical Survey (MMAPS) to establish a master reference system for development projects. 18. The effort to strengthen MMC was well intentioned and probably appropriate at the time of appraisal, when the Commission had just been established and mandated with a clearly critical coordination function among the 17 existing local governments comprising Metro Manila. The fundamental problem lay in the nature of MMC. The Commission's fatal flaws were embedded from inception in its legal framework as a Presidential Commission of 5 individuals, appointed by and accountable only to the President, with full taxation powers. MMC structurally was unresponsive to and unrepresentative of the city's large and fairly sophisticated populace, who had no voice in its actions. Not surprisingly, it was resented and resisted from the outset by the local governments, and their elected mayors. When the change of government occurred in early 1986, MMC was one of the first targets for reform. The new constitution, approved in early 1987, formally dissolved the Metro Manila Commission. The current government recognizes the need for metropolitan wide coordination and management, and alternative structures based on more direct representation of the population have been under consideration by Congress. 19. In 1985, as conditions deteriorated, the Bank withdrew from further funding of M1C initiatives. The proposed Organization and Management Study was shelved, pending the outcome of internal organizational reviews of MMC. However, in view of the serious metropolitan-wide issues still unresolved (e.g. traffic congestion, solid waste management, environmental pollution), the solution of which urgently require effective coordination of the many local governnent's individual initiatives and energies, future assistance might again be contemplated to MMC's eventual successor at some point in the future.  - 6 - Meanwhile, attention has been focused on more direct assistance to local government units (LGUs). 20. Lesson #4: Weak Control of Proiect Costs. Lack of control over total commitments under technical assistance and studies was a major cause of restructuring the project. By early 1984 it was clear the total contractual obligations for studies of about $11 million significantly exceeded the $7.35 million allocated at appraisal. Since the Bank disbursed 100%, and TA funds were to be provided as grants to agencies, discipline was lax in their use. Bank "no objectionsw had beer given based essentially on procurement, TOR and contractual signoffs, but withiout any internal monitoring of commitment or expenditure control. Final expenditures for studies totalled $11.4 million, an overrun on appraisal figures of over 50%. Following project restructuring, a control system was set in place to ensure expenditures did not exceed those agreed. To the credit of all agencies, these ceilings were thereafter strictly observed. No agency exceeded its allotted loan amount, despite the lack within the Bank of any way to enforce such a limit, since six different agencies were drawing against a single disbursement category for Technical Assistance and Training. To have established separate categories for each agency and each category would have been difficult and unwieldy for the Bank. Termination of such institutional complexity in subsequent projects, as well as improved internal monitoring and controls on the part of both the Bank and the Borrower have resolved this problem. 21. Lesson #5: Local Government (LGU) Particigation in ZIP Mortgages. Local government performance under the project was disappointing, as it was under Urban II with respect to their proposed role under the shelter components. The project envisioned that completed ZIP sites (both assets (mortgages) and liabilities (peso loans from Govt and NHA )) would be turned over to local governments upon completion for maintenance and cost recovery. NHA indicates it opposed this arrangement at appraisal on the grounds that local government units (LGUs) would not vigorously pursue cost recovery. As it turned out, the Authority was right. Collection on the first two ZIP sites which were turned over to the respective LGUs upon completion, is now reported at or below 5% (compared with 46% for NHA sites). 22. This parallelled experience under the Urban II SIR component, where cities took over completed sites with a subsequent virtual collapse of mortgage sales and collections. Of the 10,000 completed upgrading .nd serviced site units, as of 6/88 only 16% were even reported to be under collection in Cebu. The figures for Cagayan and Davao are little better at 35% and 38% respectively. These low figures reflect the political realities of local government pressures; collection even on property taxes is estimated at around only 50%. 23. As a result of this experience, the Bank agreed in 1984 to abandon plans to transfer mortgages to LGUs. Collection was left for the moment with NHA. Though below ideal levels, the 77% sales record on ZIP Phase I sites (meaning 77% of completed units are under collection), and 46% collection efficiency is well above the levels achieved by local governments under both Urban II and III.  - 7 - 24. Under the subsequent Housing Sector Loan it was agreed with Government that mortgage servicing would be done preferably by a housing finance institution such as NHMFC, or for existing mortgages (until they can be sold), NHA. Though they may in future act as housing developers, local governments are not mandated to function as mortgage servicing institutions. 25. L.Qsson #6: More Direct Local Government Role under MHINUTE. Even under the MNINUTE component, local government's did not realize the intended cost recovery through property reassessments which was to be the expected vehicle. Funding was passed from the Bank and Government directly to DPWH/MMINUTE which implemented the project. The subsequent planned transfers of assets to both MWSS and MMC/LGUs did, in fact, occur toward the end of the project. However, the associated debt remained with the national government. Now that the benefits have been delivered, and the assets transferred, it will be exceedingly difficult to separately effect a post-facto transfer of the liabilities to these local governments. It was perhaps rather unrealistic, even at appraisal to expect this could be easily accomplished after completion of construction. Since the debt obligation was not on the books of the local governments, none felt compelled to increase property assessments in the affecteA areas. Attempts were made by GOP itself to update/undertake reassessment in 1985 and again in 1987, but were deferred under political pressure related to elections. Another attempt is expected in 1989/90. 26. The subsequent NMINUTE II (Municipal Development II) project, now being appraised by the Bank, has corrected this project design problem by ensuring a direct flow of the Bank loan through the Municipal Development Fund to Local Governments, who will be the borrowers. LGUs will also bear direct responsibility for project implementation, with technical guidance from the MMINUTE staff. This will not, in itself, guarantee the subsequent increases 'n property assessments needed to achieve indirect cost recovery, but together with close monitoring by both government and the Bank, should enhance its prospects. 27. A closely related point is that the MMINUTE program was designed specifically to complement, and not compete with, NHA's ZIP upgrading. The distinctions between ti.e two programs are three: first, ZIP delivers tenure through title which MMINUTE does not since it is liiited to already individually tenured areas; second, ZIP delivers a comprehensive package of services while NMINUTE is intended to provide much more limited services to a much larger number of people with lower subsequent per unit/dwelling investment; and third (and most important), under ZIP, cost recovery is through direct monthly charges, while MMINUTE appears 'free" to residents, with capital recovery only indirectly through increased property tax assessments, which, as noted above, never actually occurred. 28. Tlhus to avoid undermining NHA's cost recovery efforts, it is important that NMINUTE improvements be kept relatively small on a per dwelling basis, and limited to tenured sites. If implemented in untenured sites (e.g. public rights of way) cost recovery could not take place since the property would not be reflected on the tax rolls. If implemented in large single holdings, where invasion has occurred, and which typically count among the 50% of properties not paying property taxes anyway, the program is likely to undermine and compete with NHA's upgrading program by providing the appearance and reality of "free' services. (Owners who do not pay tax will not pass on this tax.) Such  cases represent somewhat of a grey area, and if they arise, should probably be discussed with NHA before implementation on a case-by-case basis. 29. Once NMINUTE investments have occurred, residents feel secure that government will not uproot or destroy these investments. Thus, the distinction between formalized tenure (a papered title) and the sense of security which results in a community by government recognition and investment is narrow, at best. Asked to choose between P20,000 of charged services with a title and monthly payments, and P20,000 of "free" services without the paper title, residents might well choose the latter. Under the current highly politicized conditions in the Philippines, those beeiag askged to sign mortgages by NHA under the ZIP mortgage program might refuse, feeling unfairly treated. Since NHA is achieving only a 50% rate on signature of mortgages in Bank-assisted upgraded areas, a problem definitely exists though it is not clear that the MMINUTE program is a primary, or even contributory cause. Close attention is therefore needed in the design of follow-up investments to keep from undermining and eroding NHA's already troubled cost recovery efforts. 30. The principles agreed under MMINUTE I of non-competition with ZIP programs, and the two distinctions noted above of small per dwelling investments, and restriction of investments to tenured sites have been structured into the followup Project (Municipal Development II) recently appraised by the Bank. However, if local government investments ultimately undermine cost recovery under mortgage upgrading operations because of weak enforcement of indirect cost recovery by LGUs, further investment in one or the other may have to be cut back or terminated. In view of the urgent and pressing demands for improved services, this would be unfortunate. Remaining Issues 31. Issue #1: Regulatory Bottlenecks. Government's PCR (Part II para 8.11) outlines the serious problems encountered by several components in obtaining subdivision approval and individual titling. Protracted delays of over two years have occurred in some cases. Lack of individual titles, in turn, has obstructed signature of mortgages in the ZIP program as well as both the NHA and PSSP serviced site programs, seriously impeding cost recovery. Once occupancy has been tolerated for an extended period without payment, cos- recovery becomes a virtual impossibility under the politicized post-revolution Philippine environment. Thus such delays carry high, though indirect, costs to both the public and private sector. As problems with bureaucratic obstructions have crystallized, fears have grown that the investment component of the Housing Sector Loan (US$80 million) could encounter significant delays attributable to this cause. 32. Government's PCR calls for a national program to improve service delivery and streamline bureaucratic procedures involved in subdivision of titles, r, gistration/annotation of deeds, government permits and licenses, mortgage processing procedures and inspection procedures. Though some of these concerns will be addressed in a preliminary way under the Housing Sector Loan, it may be worthwhile to explore with Government more direct and more extensive means of addressing these obstructions which are convoluted, time-consuming to understand, well entrenched, and difficult to reform because of the endemic petty graft which often underlics tham.  -9- 33. Issue #2: Delayed Occupancy in Serviced Sites. The PCR also clearly notes the problems being encountered in occupancy of both the Cielito Homes (PSSP) and Pasig (NHA) sites, having only respectively 28% and 11% of units occupied (paras 5.26-5.32). A similar problem is being experienced in the Kauswagan site in Cagayan de Oro, developed by NHA under Urban II where occupancy is about one third. Cielito suffered from innumerable startup tribulations and was problematic from the beginnirg. Its present problems involve legal snarls under litigation, and are therefore more easily understood. Pasig, however, has remained unoccupied for two years and represents a serious and puzzling issue. 34. The cause of the problem is under investigation by NHA. Fully 81% of lots are sold, and collections are reported to be an impressive 90% (net of arrears) on these lots. Thus plotholders appear to be aware of the value of their property and anxious to retain it. The PCR outlines a number of possible explanations, including (a) allocation (via lottery) to moderate income households who are unwilling or unable to complete dwellings by self-help (an explanation which is strengthened by the use of Pasig as a test site for mortgage sales to NHMFC, and thus the acceptance of a proportion of formally employed applicants), (b) lack of titles to secure construction loans by these moderate income households, (c) lack of building materials on site, and (d) lack of standardized building plans, help with building permits, and technical assistance in house construction. An additional explanation may be that lottery winners are speculating on thes. well located, and valuable properties. NHA has not yet enforced sanctions against beneficiaries who have failed to initiate development within six months, as st!pulated in the sale agreement. 35. Widespread vacancy in this large site in which nearly $20 million has been invested, conveys the impression of weak demand i:or self-help -erviced sites. The appearance of disinterest, however, markedly contradicts the strong market resporse when the first 700 lots in Pasig were made available in early 1985. Wichin six days, 20,000 applicants responded, of whom 17,000 were financially qualified. Such strong demand, together with the relatively high payment rate would further point to possible speculation by moderate income, employed owners. 36. A further facet of the problem is that while NHA is getting paid, MWSS is not. Recovery on water and sewer investments was to occur through tariff payments. However, with occupancy rates of only 11%, MWSS is realizing very little return on this expenditure, representing about 25% of total costs or $5 million. Maintenance of the system with low water flow will be problematic. 37. It is likely that a variety of factors is at play in Pasig, rather than any single cause. In November NHA formulated an Action Plan based on the findings of the survey noted in the PCR as being underway, to begin to enforce sanctions, cancel sales of non-payers, develop standard building plans and technical assistance, and help community groups engage contractors and labor to initiate construction. Meetings have taken place with groups of plotholders to either rescind the sale or begin construction. Hopefully close followup by NHA management will yield results. However, this remains one of the primary concerns upon project coLpIation, and will be the subject of followup under the Housing Sector Loan.  - 10 - 38. Issue 03: NHA Cost Recovery (Sales and Collections). Government's PCR echos continuing and justified concerns regarding the financial viability and replicability of the upgrading and serviced site program. Cost recovery involves the compounded effects of (a) pricing policy, (b) sales, and (c) collection efficiency, since expenditures not allocated to beneficiaries cannot be recovered, units not sold cannot be collected, and sums not collected are lost to NHA. 39. Based on sales and collection performance to date on Phase I only of the ZIP program, overall cost recovery is only about 39% of total expenditures (assuming 100% recovery by M?SS of water/sewer costs charged off to that agency). About 11% of total cost is allocated as grant funds for schools, clinics, roads and writeoffs, and the balance 50% is lost to the compounded effect of slow sales (23% not yet sold) and weak collections (46%). Of the two, collections are the most pressing problem. Poor collection performance is either attributable to weak administration by NHA, to unwillingness by recipients to pay for the services and tenure provided, or to politization of payments. Whatever its cause, this low level of recovery raises serious questions about the financial viability of squatter upgrading in the Philippines over the longer term. The Shelter Sector Loan assumed all but about 15% of recoverable investment would be renaid. The ZIP program so far is showing a loss of fully 56% of recoverable investment. Either lot pricing will have to be significantly increased to add a sizable margin for such losses, or major government subsidy will be required to sustain the program. 40. Pasig serviced sites are somewhat more encouraging with 72% recovered (assuming 100% recovery of MWSS costs), 11% grant funding, and 17% lost to the effects of slow sales and collections. Of the 89% recoverable, 19% is still being lost, however, primarily to delays in titling which make sale difficult. On a site of this size, this represents a loss of nearly $4 million which would otherwise be available for reinvestment. Table 37.1 on the following page summarizes compounded cost recovery in the NHA ZIP and Pasig sites. Miscellaneous Observations/Recommendations a. Special Account. The PCR (para 6.03) blames the lack of a Special Account for the decline in disbursements which occurred from 1984 through 1986. Elsewhere, however, it notes clearly the more probable cause which was the economic and financial decline which occurred during those years and the consequent contracting, budgeting, procurement, and payment problems which resulted. These problems were followed by extensive institutional uncertainties and changes following the change of government in early 1986 which further slowed contract awards and disbursement (see paras. 3.01 and 5.04 in particular). b. Greater use of SOE procedures to reduce Procurement Reviews. The Special Account proved very effective, however, and could only be improved by ensuring in future that SOE procedures are utilized for small civil works contracts. Under this project SOEs were used only for sub-loans. All civil works contracts had to be reviewed on either a pre- or post-award basis, leading to an overload of review requirements on the part of Bank staff for quite small civil works and survey contracts often amounting to only a few thousand US dollars.  - 11 - c. MKINUTE Emphasis on Roads/Drainage. As shown in Figure VI.2 of Part II, the MNINUTE component eventually developed substantially more roads and drainage facilities than planned at appraisal (75% of total costs vs. 61% at appraisal, translating in physical terms into 153 kms vs. 64 kms planned), at the expense of water supply and sanitation (11% vs. 29% of planned expenditures). The reasons quoted for this distinct shift of emphasis involved difficulties in obtaining land and rights of way, and are doubtless true. In addition, political pressures were often intense for road/drainage improvements which had more immediate impact on property values. Table 37.1 NU COMPOUNDED COST RECOVERY Pricing PolLcy Total -------------------------------------- Sales Collection TOTAL Expenditure Allocated to Allocated to Subtidized Efficiency RUHE) D4eficiaries MWSS 2IP PROGM 1001 77X 121 lX ------------------------------------------------------------- Rcoveary by NRA 77% 77Xx77 - 59 46Xx59 - 27 27Z (35X) (Phase i) (Phase I) Recovery by MWSS 121 assume 1001 assume 100S 121 (100X) SUMMARY: TOTAL RECOVERED 391 =------ TOTAL SUBSIDIZED 11X TOTAL LOST TO SALESICOLL 501 TOTAL INVESTMENT 13OX SMMVICED SITES 1001 641 25X llX Recovery by NUA 64 81Xx64 - 52 90Sx52 - 47 471 (731) Recovery by MWSS 25 assume 1001 assume 1002 252 (1002) SUMMARY: TOTAL RECOVERED 72Z TOTAL SUBSIDIZED llX TOTAL LIST TO SALESICOLL 17i  - 12 - Attachment 1 TABLE OF KEY INDICES CoMPnt SAR Target Actual I of SAR Zonal a. Improvement of 13 sites .... Improv nt of 31 sites +138X Improvement b. Servlce to 160,000 people ........ Service to 350,000 ppl +118S Program (ZIP) a. Service to 17,000 plots .........Servlce to 25,882 plots (elsewhere +53X shown as 26,075) (15,781 & 10,294) d. Unit Sale Price P12,261 .Average sale prlce P25.500 (Annex 7 ---- of Part 'I)s P19,500 Phs Is and P27,300 Phs II (Annex 6-Cost Alloc) a. Unit cost $1,768 average ..... Unit cost $1,690 average -41 ($43.7 costl25,882 plots) P24,500 average unit cost (P635.5125,882) f. Affordability to 10th ile .. Average affordability to te 10th 100X percentlle (Annex 7 of Par' II). However, several sites exceed limits, raising questions of suitabilLty for ZIP (West Crame: only 35th to 55th Zile can affords Sitio Matang: 23rd Xlles and Albina Altura: 35th lile) g. Investment of $43.3 m .. Investment of $43.7 m 100X h. 100X completion .1002 of original sites completed, and 802 of extended sites. Weighted 87X total: 87X as of 6188 i. 1002 sales .12,088 signed mortgages of the 15,781 Phase I completedt Phs I... 771 0 signed of the 3,115 completed under Phase II Phs II. .O J. 1002 collections ................. 462 collections on units under collection (range from 132 to 1001) 462 Metro Manila a. Service to 11 LGUs .Serrvice to 17 LGUs +54X Infrastructure, b. Service to 300,000 persons . Service to 867,000 people (direct) +189X Utilities, and and 1,400,000 indirect per PCR of Engineering Government (MIInUTz) c. Unlit Cost Lower than ZIP ........ Uuit cost P3,980 ($190) vs. P25,500 OK (aprox P7,200 ($974) for ZIP or 16X of ZIP (P345.1m/86,700 dwelling units e 10 ppl/unit) d. Tenured sites only ............... Tenured sites only OK e. Physical works: -- 64 kms of road impvmts ........ 153 1m& of road lmpvnts +139X -- 200 kms of drainage impvmt .... 226.4 kms total, comprising 39.2 +13X major and 187.2 minor vorks -- 86 units of community ......... abandoned (1 built) due to technical --- septic tanks and social reasons -- 60 public toilets 1/ .......... 41 publlc toilets 681 -- 365 public water faucets ...... 203 public water faucets 562 f. Financial turnover of liabil- .. . .P1Om or 31 of total expenditures ities to MWSS and MffC turned over to MWSS as of this date 3X g. Cost recovery through indirect mans of increased property . Not achieved --- values Private Site a. Development of 2,840 lots ........ Development of 4,063 lots +432 and Services b. Affordability to lower income.... Based on two completed sltes and OK (PSSP) for at least 702 of units projections for five sites: 31X of unlts affordable to 20th lile & under 402 of units affordable to 40th lile & under 701 of units affordable to 50th 2ile & under c. 1002 completion. 75X completion (3029 units) 75S d. 100 sales .572 sold (2,316 units) 57X e. 1002 occupancy ................... 211 occupancy (846 units) 212 f. 100X mortgage sales to NHMFC ..... 172 sold (706 units) 172 g. 1OO2 recovery of construction .... Data not yet available MA loan P & I Public Sites a. development of 3,300 lots . Development of 6,002 lots +82X and Services b. Sale Price P9,700-P32,000 (NM) weighted average P 14,200 . Sale Prlce average P25,600 (Annex 6) +80S Avg Sale Price in $1,920 . Avg Sale Price $2,061 e 12.42 excb rt + 7S average for Public SS component 1-Dtafo------oumnsan-i-o---prt----T-tafiniae-6pulc-o-t 1/ Data fromt Working Documents and Villoria Report. MMINUS staff indLcate 16 public toilets  - 13 - Attacbment 1 p.2 Component SM Target Actual 2 of SAR c. Affordability to lower income .... Sale prices range from P16,000 to for at least 70X of units ........ P38,000 affordable to households from the 18th Sil* to the 20th lile (i.e. OK affordoble to 802 of MM). Distribution not available. d. 10OX ceompletion . .0........... lOS physical completion except for 10lX individual titles e. 100l sales . .............. Of 6,002 lots, 4,850 sold (81Z) x81 f. 1OOZ occupancy . ................. Of 6,002 lots, 687 occupied IIl g. 1002 collections ... ........... 90S collections (net of arrears) 901 net data not provided inel arrears TRC Urban a. Generation of 6,000 new jobs ..... TRC reports 28,471 jobs generated +374Z Livelihood at a cost of $657 per job ........ at a cost of $250 per job, but the 381 Flnancing assumptions and basis of this Program calculation is unclear and probably (ULIP)JTRC different from the SAR b. 1002 collectionsicost rec ........i nitially large arrears and fore- closed assets now being liquidated, collections now 972 of amount due 972 net net of arrears; 461 including the 462 arr large initial arrears a. Total beneficiaries not .......... 30,541 total borrowers --- Identified d. Return on equity not identified ..Following startup period, ULIP --- but targeted to be positive ..... achieved a real return after taxes and inflation in 1986 of 22 and 1987 of 12 Solid Waste a. Drain cleaning and support ....... Abandoned Improvement for waste recycllng of NNC Component b. Plot to test whether MHC ........ Pilot carried out with about 5,000 HB (SWIP) internal operations were more but efficiency of MRC operations were efficient Involving 5,000 HE never properl- eoste and tested before expansion c. Extension phase to service ....... Extension to service 110,000 RH; no data 120,000 RH in lowest income to indicate exactly which areas are areas and markets being serviced with this equipment d. Acqulsition of 25 trucks ......... Acquired 16 trucks and 140 refuse 641 handc-rts containero, followed immediately by Japanese grant donation of an addLtional 116 trucks for NMC operation TaxMapping a. Identlfleation of additional ..... 65,000 additional properties identlfied, by NMC properties not Included in 552 assessed for added value of P67 m (CORPTEP) tax roles (no specific targets) to tax roles; and expected collection of about 502 efficiency for a yield in the first year of about P33.5 m. b. Rapid recovery of costs .......... P 43 m total cost expected to be recovered in about 15 montba. Actual expenditures to date of P32 m shonld be recovered wLthin the flrst year c. Completion of computerization .... Not achieved, still underway before June 1988 and remain- ing valuations on 451 of identified properties Technical a. Significant strengthening of ..... Not achieved. See Part I discussion. Assistance & the metropolitan planning Studies capacity of MMC General a. Implementation ln 4 years ........ IplementatIl in 7.5 years from +88S from 6180-6184 6180-12187 .th full cempletion of l lagging component ln 3190  - 14 - Attachment 2 INFLATION 'I _. Project Implementation o" it" ton) at 43)t" im au a