FOR OFFICIAL USE ONLY. Document of The World Bank. Report No.: IMPLEMENTATION COMPLETION REPORT (LOAN 39380; 3938A; 39390; 39400) ONA LOAN

Transcription

1 Public Disclosure Authorized Document of The World Bank FOR OFFICIAL USE ONLY Public Disclosure Authorized Public Disclosure Authorized IMPLEMENTATION COMPLETION REPORT (LOAN 39380; 3938A; 39390; 39400) ONA LOAN IN THE AMOUNT OF US$ MILLION TO THE REPUBLIC OF THE PHILIPPINES FOR A Report No.: SECOND RURAL FINANCE PROJECT February 4, 2000 Public Disclosure Authorized Rural Development and Natural Resources Sector Unit East Asia and Pacific 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.

2 CURRENCY EQUIVALENTS (Exchange Rate Effective as of September 30, 1999) Currency Unit = Peso (P) 1 pesos= US$ US$ 1.00 = 40 Pesos FISCAL YEAR Government: January 1 - December 31 ABBREVIATIONS AND ACRONYMS ARF - Agrarian Reform Fund CLF - Countryside Loan Fund DENR - Department of Environment and Natural Resources DOF - Department of Finance EU - Environmental Unit FXRC - Foreign Exchange Risk Cover GOP - Government of the Philippines GRT - Gross Receipts Tax ISAP - Institutional Strengthening Action Plan LBP - Land Bank of the Philippines NG - National Government PFIs - Participating Financial Institutions RFP - Rural Finance Project RBs - Rural Banks RCF - Retail Cofmancing Fund SAR - Staff Appraisal Report TBs - Thrift Banks TRFP - Third Rural Finance Project Vice President: Country Director: Sector Director: Task Team Leader/Task Manager: Jean-Michel Severino, EAPVP Vinay K. Bhargava, EACPF Geoffrey Fox, EASRD Arie Chupak, EASRD

3 FOR OFFICIAL USE ONLY CONTENTS Page No l. Project Data 1 2. Principal Performance Ratings 1 3. Assessment of Development Objective and Design, and of Quality at Entry 2 4. Achievement of Objective and Outputs 4 5. Major Factors Affecting Implementation and Outcome 7 6. Sustainability 9 7. Bank and Borrower Performance Lessons Leamed Partner Comments Additional Information 21 Annex 1. Key Performance Indicators/Log Frame Matrix 22 Annex 2. Project Costs and Financing 23 Annex 3. Economic Costs and Benefits 24 Annex 4. Bank Inputs 25 Annex 5. Ratings for Achievement of Objectives/Outputs of Components 26 Annex 6. Ratings of Bank and Borrower Performance 27 Annex 7. List of Supporting Documents 28 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.

4

5 Project ID: P Project Name: RURAL FINANCE II Team Leader: Arie Chupak TL Unit: EASRD ICR Type: Core ICR Report Date: February 4, Project Data Name: RURAL FINANCE II L/C Number: 39380; 3938A; 39390; Country/Department: PH[LIPPINES Region: East Asia and Pacific Region Sector/subsector: AC - Agricultural Credit KEY DATES Original Revised/Actual PCD: 05/12/93 Effective: 01/08/96 04/23/96 Appraisal: 04/28/94 MTR: Approval: 09/14/95 Closing: 12/15/99 Borrower/lImplementing Agency: Other Partners: LBP/LBP STAFF Current At Appraisal Vice President: Jean-Michael Severino Gautam Kaji Country Manager: Vinay K. Bhargava Thomas W. Allen Sector Manager: Geoffrey Fox Pamela Cox Team Leader at ICR: Arie Chupak lcr Primary Author: Paul Harrison; Consultant 2. Principal Performance Ratings (HS=Highly Satisfactory, S=Satisfactory, U=Unsatisfactory, HL=Highly Likely, L=Likely, UN=Unlikely, HUN=Highly Unlikely, HU=Highly Unsatisfactory, H=High, SU=Substantial, M=Modest, N=Negligible) Outcome: S Sustainability. HL Institutional Development SU Impact: Bank Performance: S Borrower Performance: S QAG (if available) Quality at Entry: S Project at Risk at Any Time: Yes ICR

6 3. Assessment of Development Objective and Design, and of Quality at Entry 3.1 Original Objective: The objective of the project as stated in the Staff Appraisal Report (SAR) was 'to help expand the volume of medium and long term commercial credit to agriculture and rural development in the Philippines and to enhance the policy framework of the rural financial sector by: (a) supporting the development of Land Bank of the Philippines (LBP) two funds, the Countryside Loan Fund (CLF) and Retail Cofinancing Fund (RCF) to finance private sector investments in the rural areas; (b) strengthening LBP as the main wholesale financial institution serving rural areas; and (c) upgrading the operational capacity of rural cooperatives, Participating Financial Institutions (PFIs), particularly Thrift Banks (TBs) and Rural Banks (RBs), to provide financial services in the rural areas'. The project was also concerned with environmental impact. It was intended that... "sub-projects financed should be environmentally sound and comply with GOP environmental laws and regulations". These primary objectives were, and continue to be, important. Because of the undeveloped domestic market in long term financial instruments, was, and continues to be a shortage of finance for long term investments. Investment projects are financed by short term loans which are rolled over, exposing them to potential liquidity crises when financial markets tighten, as happened under the recent financial crisis. The objective of using retail as well as wholesale mechanisms reflected the fact that LBP has become an important rural credit retailer as well as a wholesaler, and the inclusion of this fund would also allow LBP to prudently expand its own long lending for rural investments. It provided the opportunity to broaden project coverage beyond that of CLF I. The strengthening of LBP and upgrading of the various intermediaries operational capacity were also important, and relevant, objectives as their achievement would result in a system which functioned more smoothly and would be better able to make sound credit decisions. The policy objectives of distancing LBP financially from the Agrarian Reform Fund, and limiting dividends to rational levels 1/ were very important. In particular because of the 1992 law requiring all State Enterprises, including Government Financial Institutions (GFIs) to pay dividends at 50% of nominal profits. These objectives are still more important now because (i) the value of outstanding Land Reform Bonds and their annual service/retirement cost are much higher than at the time of appraisal; and (ii) the financial crisis has had a negative effect on the balance sheet strength of all Philippine Banks, including LBP, thereby making conservation of real equity vital. The approach on environmental matters was in line with the Bank's OD environmental policies. However, for full compliance to have been realistic within a credit project, the tools for environmental evaluation and enforcement, (i.e.: a strong and efficient environmental setting and regulations at state level) should have been in place at the time of project inception. While GOP had a reasonable set of environmental regulations, it had not yet demonstrated the capability of implementing them effectively. The SAR did not raise concerns explicitly as to the extent that this objective could be achieved although probably it should have done so. However the fact that the Bank encouraged the establishment of the Environmental Unit (EU) in LBP indicates implicit concern about GOP's implementation capacity. 1/ There were covenants in the loan and guarantee agreements to require that exemption from the act should be obtained. These were that LBP would pay only dividends out of profit in real terms after inflation adjustment and providing additional equity to cover 1/8 of the incremental growth of agrarian lending (with the advent of the Third Rural Finance Project (TRFP) this was modified to take account of the need for paying preference share dividends and to provide for incremental equity to support CLF III incrementaloans). -2 -

7 3.2 Revised Objective: N/A 3.3 Original Components: The project would be implemented over five years, and would comprise the following components: (a) Credit: medium and long term loans to finance private investments in agricultural and other viable rural operations, (both fixed assets and incremental working capital) for a broad spectrum of agricultural and non-farm rural investment, (e.g. production, processing, marketing, transport services, storage facilities, custom service or leasing operations, and import of critical inputs). The credit operation would be determined by market forces, with no earmarked allocations by crop or type of investment and would bear a market determined interest rate. This component would include two programs. (i) (ii) CLF II: a wholesale operation to be carried out by LBP. CLF II resources would be channeled through accredited financial institutions for onlending to private investors; RCF: a cofmancing lending program for medium and long term rural investments, implemented by LBP to finance, together with other financial institutions, eligible sub-projects initiated by LBP's own customers; (b) Institutional Strengthening Programn Strengthening LBP financially and organizationally to ensure its sustainability in providing financial services to the rural population. This component would support LBP's long-term development program with concentration on private saving mobilization, agrarian lending policy, and organizational streamlining. The component would consist of implementation of an Institutional Strengthening Action Plan (ISAP) aimed at strengthening LBP's financial and institutional capabilities as wholesale financial institution serving the rural areas. An important element of this would be training and technical assistance with special attention on the management and members/staff of the cooperative, TBs, and RBs. These components were well linked to the objectives of the project, and CLF II followed on from the successful implementation of CLF under the earlier Rural Finance Project. Both ISAP and CLF II were appropriate to the implementing capacity of LBP. The initial Bank approach had been to provide LBP with medium and long-term resources entirely through CLF II. By the time of negotiations, GOP and LBP were anxious for a retail component also. The Bank yielded to this request because (i) LBP as the administrator of CLF had not been able to offer CLF funds to its own clients, thereby placing it at a disadvantage compared with CLF accredited banks, and (ii) existing rural customers of LBP were being deprived of CLF funds. However, the resultant RCF component was a compromise concept, which, although geared towards achieving Project objectives, and aimed at SMEs in the countryside, was not capable of being effectively implemented. Despite considerable preparation and promotion efforts, no lending had been made under the RCF by early It appears that under the original structure of RCF, the maximum sub-loan size was too small to attract co-financing. Shortage of long-term resources among potential co-financiers (which was not addressed by the project in this component) may also have been a contributing factor

8 3.4 Revised Components: While the CLF II component remains as it was designed at the time of the SAR, the RCF component was redesigned. This was a pro-active initiative to restructure the component before it affected the performance of the entire project. In June 1997 the RCF component was amended as follows: (i) the fund was restricted to lending to medium and small borrowers with a maximum of total assets of P60 million (US$2.3 million); (ii) sub-loans were not to exceed P4 million; (iii) financing of sub-projects under RCF were as follows: minimum 15% equity contribution by the sub-borrowers (as under the CLF II), at least 35% LBP's participation using its own resources, with the balance, not to exceed 65%, be provided out of the proceeds of the Bank's loans. Cofinancing by another financial institution was no longer required. At the same time, as the component redesign, US$40 million allocated originally to RCF was reallocated to CLF II. [Loan 3938 (CPL) US$5 million; Loan 3939 (FRSCL) US$20 million; and Loan 3940 (VRSCL) US$15 million] As a result of this reallocation total funds to be available for CLF II (under the three loans) was increased from US$90 million to US$130 million and funds available for RCF were reduced from US$60 million to US$20 million. This change in RCF and expansion of CLF II resources is expected to (i) allow LBP to accommodate its traditional customers; (ii) strengthen the CLF II and provide it with more funds to meet the strong demand (within one year of project start up, more than 66% of the initial fund allocated to the CLF II had been disbursed and an additional US$18 million of sub-loan applications were being processed); (iii) encourage more PFIs to actively participate and finance viable investments in the rural areas, thus increasing the generation of incremental gross value added and employment, and contributing to poverty alleviation efforts; (iv) enhance project ability to meet its agreed objectives; and (v) accommodate potential rural investors which otherwise may not have been supported, thus advancing project objectives. 3.5 Quality at Entry: Peer Reviewers and Bank decision makers endorsed the project design, its features and components. Therefore quality at entry was satisfactory. (QAG did not exist at the time of project approval). 4. Achievement of Objective and Outputs 4.1 Outcome/achievement of objective. There were no macro-economic policy objectives, nor broad sector policy objectives. However the narrower sector policy objectives with regard to distancing LBP from the ARF, and limiting LBP's dividends to real profits have been achieved to date. The ARF assets and liabilities are now outside LBP's balance sheet and the original Land Reform Bonds, issued in LBPs name have been converted to National Government (NG) Bonds with the same terms and conditions. Although there is still pressure from GOP to try to obtain dividends from LBP in relationship to nominal profits, the existence of specific covenants to prevent this has helped to ensure compliance with the agreement so far. Other than being confined to 'rural areas', the credit component of the project was undirected, consequently there were no specific physical development objectives under the project. Although if the achievement of levels of investment which result in expansion of physical assets are considered as general physical objectives, then the project was highly successful because the rate of investment has been more rapid than projected at appraisal. The project did not specifically set social objectives in the areas of poverty alleviation or gender concerns. However, with incremental economic activity and creation of over 10,000 jobs in the rural areas being likely results of investment under the project, it would be expected to have a positive impact on poverty -4 -

9 alleviation. By the same token, enhanced availability of medium term finance and the involvement of a large number of smaller financial institutions have both had a positive impact on private sector development. 4.2 Outputs by components: Credit. The project has been successful both in terms of the volume of lending and the proportion channeled through TBs and RBs. By March 1999, some 678 CLF loans had been made, including 451 loans through TBs and RBs. The CLF and total targets for amount lent and the amount channeled through TBs and RBs both have both been exceeded, but the RCF has failed to take off so far. Funds delivered under the project, are now expected to support investment of about US$260 million, almost identical to the level planned at appraisal, the main difference being that CLF will be substantially larger than initially envisaged and RCF smaller. The question of additionality for the credit component is open to debate. Under the project, commercial sub-project selection criteria were applied, and, as a result, the investments supported were sound. Because of this, many of the specific investments financed under the project would probably also have been made 'without project' using alternative funding. The project's additionality in this regard should be assessed not on whether specific investments were induced but rather on whether it resulted in an increase in overall investments and further that it did not 'crowd out' deposits, particularly term deposits. Macro data for the Philippine banking system as a whole indicates that over the project period, there was on aggregate a substantial expansion in intermediate and long term loans. Between December 1995 and June 1998, intermediate and long term loans of commercial banks as a whole 2/ grew by P364 billion or 320%, while the corresponding growth of short term and demand loans was 208%. The project's contribution to this growth in term loans was about 1%. Project contribution in terms of rural investments is likely to have been much larger than 1% as not all of the P364 billion total growth in lending went to rural areas. On the resources side, there was a growth in deposits of the banking system over the period December 1995 to December 1998 of 72% overall, but a 118% growth in time deposits. The absolute amount of time deposit growth was P280 billion (compared with project funds disbursed over the same period of P3.6 billion). Furthermore, over this period, Philippine banks, on average had adequate risk assets to equity ratios, consequently, the limitation to lending was availability of resources. 3/ A reasonable inference to be drawn from this data therefore is that the project contributed to the overall expansion of the availability of longer term resources and hence loans and investment. There is no evidence of 'crowding out'. Institutional Strengthening Program The institutional development objectives of strengthening LBP and the PFIs implementing the credit component of the project were largely achieved. In addition, the considerable support to co-operatives has resulted in improvements in their classification in excess of that targeted within the Key Performance Indicators (KPIs). On resource mobilization, LBP was successful in mobilizing private deposits, but, partly as a result of the financial crisis, it has not yet issued long or medium term peso debt instruments; 4/ LBP did, however, issue preference shares which partially substituted for these. Some specific but minor institutional development components, unrelated to the credit line such as the establishment of co-operatives' cross guarantee programs were not implemented mainly because it became clear, that many of these co-operatives were too weak to undertake this additional activity. Through its Environmental Unit (EU), LBP has developed the ability to ensure that sub-borrowers can secure the environmental permits required by the Department of Environment and Natural Resources (DENR). To this extent it reported that by May 1999, 55% of subprojects obtained DENR permits, - 5 -

10 compared to 70% targeted 5/ for the year end. However, spot check audits 6/ indicate that actual compliance with environmental regulations is not being achieved in some cases. This deficiency can be attributed to lack of enforcement by DENR, although it could be helped by more intense dialogue between the EU, the PFI and the borrower to ensure that DENR stipulations are actually translated into mitigation measures. Achievement of project with respect to ensuring that all sub-projects are environmentally sound can thus be rated only as partially satisfactory. 2/ This data applies to the banking system as a whole, while the project was concerned with rural investment. Nevertheless, the incremental lending and resource mobilization is so large that it is reasonable to assume that the picture for the 'rural areas' (defined under the project as the Philippines, excluding Metro Manila) would have been similar. Funds provided under the project would have still been small compared to total rural investment and time deposit mobilization. 3/ LBP's analysis of the 30 commercial banks with which it has credit lines indicates a weighted average risk assets to equity ratio of 6.2 at end 1995, 6.8 for end 1996, 7.0 for end 1997 and 6.5 for end These are all well within the BSP requirement of 10. Only at the end of 1997 were a significant number of banks (25%) having problems meeting this prudential ratio. 4/ LBP is planning to retail a floating rate five year deposit instrument, which will allow depositors to take advantage of recent tax law changes. That is, because its maturity is over five years, the instrument will be free of the 20% final withholding tax on interest. 5/ Target added during project implementation. It assumes that by year end 1999, 70% O of requests for environmental permits would have been fulfilled. Of the 30% remaining, most would be still in process and would be approved later. 6/ An audit of 18 out of 123 projects requiring environmental clearance indicated that 39% had fully complied with regulations, 44% had partially complied and 17% had not complied at all. 4.3 Net Present Value/Economic rate of return: At appraisal, no specific 7/ estimate was made of the project's likely economic rate of return because at that time it was unclear what types of demand-led investments would be made. Subprojects of above PI billion were simply required to have ex-ante ERRs in real terms in excess of 12% (15% since June 1998), and to be outside metro Manila. An analysis of the 47 projects (amounting to 52% of investment) for which ERRs were calculated, ex ante, indicates a highly satisfactory median ERR of 27%. Because the sub-projects are all quite new, an ex-post analysis of ERR would be premature at this stage. If resources could be found, an analysis of ex-post sub-project financial and economic performance would be useful in about 5-8 years time, by when, most of the sub-projects would be mature and the first round loans would have been repaid. 7/ The SAR pointed out that ex ante FRRs for the first Rural Finance Project had averaged about 30% and ERRs 35% and that similar results could be anticipated under this project. 4.4 Financial rate of return: In line with its objective, the project is contributing to raising the profitability of LBP, PFIs and sub-borrowers. Analysis of LBP's internal accounts show the net profit from CLF II operations over the first three years as follows: -6 -

11 Year P Million P Million P Million Income on CLF II Funds Interest Cost FXRC fee, Guarantee fee etc Provisions for loan losses Estimated Administrative Costs LBP Profit from CLF II The net profit at a little over 2% of the outstanding CLF II portfolio is quite robust, with administrative costs only amounting to about 12% of the gross margin. It will be important, however, that the provisions for losses remain reasonable. It is possible that over the next year or so these would have to increase substantially. If for example CLF past dues were covered 100% by provisions, loss provisions in the 1998 income and expense account would have needed to be P16 million, rather than P0.8 million, even so, LBP would still have shown a profit of P63 million for the year. The PFIs have reported operating spreads on CLF II funds of typically 3%-4% for medium - large loans by commercial banks and up to 8-12% for rural banks. No detailed analysis of the marginal profitability of PFIs from CLF II has been made, but the fact that there has been a fairly strong demand and significant repeat business indicates that PFIs view this funding opportunity as profitable. Ex ante analysis of sub-borrowers Financial Rates of Return (FRR) indicates that of those projects for which FRRs were estimated, amounting to 84% of project investment, three quarters projected FRRs in excess of 20%. The median FRR was 28%. 4.5 Institutional development impact: See Institutional development was a Project component. 5. Major Factors Affecting Implementation and Outcome 5.1 Factors outside the control of government or implementing agency: The Asian Financial Crisis - this has had the effect of slowing demand and weakening the financial sector, has reduced credit demand, rendered some PFIs no longer acceptable conduits and made achievement of LBP's ISAP more difficult. 5.2 Factors generally subject to government control: The project was prepared early 1994 and appraised in summer, 1994 and, after being updated through a desk analysis in spring 1995, was negotiated in June The loan was signed on October 10, 1995, became effective in early 1996 and started disbursing in the spring of The appraisal to negotiations delay resulted from concerns of GOP' s cabinet. The secretary of finance at that time (who was also ex officio chairman of LBP) postponed negotiations, just prior to their planned start because it was felt that the Project should have been retail oriented. This was not acceptable to the Bank. By early 1995, GOP accepted that the project would be basically a wholesale banking operation, as originally designed and negotiations were set for June However, GOP pressured LBP to request a larger 'retail element' during negotiations. As a result, the size of the proposed loan for the Retail Cofinancing (RCF) component was doubled from US$30 million to US$60 million. Further, the proportion of the cofinancing element to - 7 -

12 be sourced from LBP was to fall from 50% to 25%. This implied an increase in financing by other, as yet unidentified financiers of a factor of four. It was seen as means to allow LBP to further develop its cofinancing and syndication activities. Other factors subject to GOP control include: * The dividend policy for GFIs introduced in 1992 and the 1998 change in tax policy has reduced the potential for LBP increasing its equity from organic growth. * The long standing GOP Policy of not allowing provisions as a tax deductible reduces the incentive for making adequate provisions against bad debts. This is a particular problem when balance sheets are weakening as is the case at present ( ). * GOP's policy to keep agrarian lending rates low acts as a break on LBP expanding its agrarian lending in line with its ISAP target. * The inconsistent interest rate policy among donors, which has recently become more significant, is making wholesaling under CLF more difficult. * The moderate capacity of GOP's environmental agencies makes it difficult to ensure that investments will be environmentally sound. 5.3 Factors generally subject to implementing agency control: The project was implemented by LBP which (i) provided funds borrowed under the Bank Loans on a wholesale basis (via its CLF) to PFIs for on-lending to sub-borrowers based on a project implementation manual agreed by the WB and LBP (ii) undertook retail financing using RCF 8/ resources, and (iii) implemented an Institutional Strengthening Action Plan (ISAP) for itself. * CLF II has benefited from experience gained from ALF and CLF I and LBP's decision to manage these funds using the same staff and systems. * There seems to have been a lack of interest in and full understanding of RCF (since it has become a retail fund) by LBP's branches. This may be the result of LBP marketing too many loan products, lack of appropriate incentives for RCF lending and possibly ineffective communications between head office and the branches. * Over the past two years, perhaps as a result of organizational changes there appears to have been a deterioration in the quality of accreditation and supervision of Countryside Financial Institutions (CFIs). 8/ The initial cofinancing component did not move at all. The size of loan targeted (below US$3 million) was thought by LBP to be too small and other banks did not want to participate, consequently, in 1997 the loan agreement was amended to reallocate $40 million back to CLF. The remaining US$20 million was to be for retail lending by LBP specifically to SMEs. The initial proposals were for loan size to be less than $150,000 and total assets of the borrower less than P60 million (equivalento US$2.3 million at that time). 5.4 Costs andfinancing. As at March 31, 1999, the project cost 9/ to date, excluding minor 'second use of repaid funds' amounted to US$206 million. Estimated project cost at completion (December 2000) is almost identical to the Appraisal figure at $263 million. However, the CLF component at US$223 million is 86% larger than anticipated, and the RCF component, now estimated to be US$38 million is substantially (73%) smaller. CLF sub borrowers contributed significantly more to investments than the minimum required (30% rather than a minimum of 15%). This, together with the PFI contribution of 17% of the sub-loan amount, rather than a minimum of 13.3% helped to leverage Bank funds for CLF II well beyond the appraisal estimate (US$1 of Bank loan supported US$1.71 of investment, rather than the appraisal estimate of US$1.33). This offset the effect of a reduction in size of the RCF (in which Bank funds were planned to be more - 8 -

13 highly leveraged). The cost of the training component was in line with the appraisal estimate. Despite the devaluation, average exchange rate for the whole project - likely to be about P32 per US$ - is slightly lower than that estimated at appraisal (P33 per US$). Overall, the project has been implemented within the time frame envisaged. Delays have occurred in RCF, but CLF is well ahead of schedule. 9/ Because this is a credit project, costs are mostly induced investment by private entrepreneurs borrowing under the project. The cost figures for the CLF and RCF components represent the value of investments partly financed by project resources. 6. Sustainability 6.1 Rationale for sustainability rating: The project is highly likely to be sustainable at sub-borrower, PFI, LBP and Government levels. For the sub-borrowers, investments have been assessed as being profitable ex ante with a median FRR of 28%, well in excess of the cost of funds. The fact that loan repayment record is substantially better 10/ than that for the banking system as a whole, supports the thesis that loans under the project have been relatively good and that on average, the investments are sustainable. The PFIs under the project have set interest rates and fees at levels which make project lending profitable to them. This, together with the long maturity of project funds means that this type of lending, will be sustainable using project resources for more than another 10 years. Beyond that, other long term borrowing from development or commercial institutions, increased capitalization or the issuance of bonds would allow the levels of medium/long term investment lending initiated under the project to be sustained and expanded. LBP's internal accounts show that the project adds to LBP's net income, and so it will be sustainable provided LBP continues to be prudent in its accreditation process and the need for loan loss provisions against its lending under the project remains low (well below 1% per annum). However, if LBP becomes constrained by its risk assets to equity ratio, there is a possibility that wholesale lending could be crowvded out by retail lending with higher spreads. Because of the timing of the Asian financial crisis and the associated devaluation, this project will probably be less beneficial to GOP than RFP on which substantial fees were earned prior to the crisis, or TRFP, which is post crisis. 11/ The simple Foreign Exchange Risk Cover (FXRC) fee is unlikely to be sufficient to cover the foreign exchange loss. However, that fee represents only a small part of the direct project benefits to GOP. In addition, GOP also receives a 1% 'guarantee fee' and gross receipts tax (GRT) both on income to LBP from funds on-lent to PFIs or invested in Government Securities and further GRT on interest paid by sub-borrowers to PFIs. The incremental GOP income from these direct fees and taxes and incremental company tax stemming from incremental value added by sub-projects is estimated to more than cover the foreign exchange risk exposure. Analysis based on past and projected interest rates, exchange rates and project loans indicates that the net present value (NPV) 12/ to GOP of incremental Guarantee fees, FXRC fees and Gross Receipts Tax (GRT) less incremental foreign exchange cover costs resulting from the project is positive at US$1.2 million. If incremental income tax on estimated incremental profits of LBP, PFIs and the sub-borrowers is also taken into account, the NPV rises to US$P32 million. Environmental sustainability requires that LBP continues to insist on conformity with environmental laws -9-

14 in future lending operations. This is probable, at least in the medium term, because the Bank, and other environmentally conscious donors are continuing to fund projects through LBP and while this goes on, its EU will remain in place and be reasonably effective. Its sustainability would be sure if the 'playing field were leveled' and the code of practice presently adopted under the project were required for all investment loans by the entire banking sector. Without certainty that this will happen, there are some minor doubts as to the long term sustainability of the environmental objectives of the project. 10/ The overall level of gross past dues for the banking sector as a whole at end 1998 was 11% whilst those of CLF II sub borrowers to PFIs was 6.6%. 11/ As a result of, and in response to, the Asian financial crisis a Third Rural Finance Project (TRFP) was approved by the WB in December The TRFP's main objective is to provide financial support to the rural economy to overcome the difficulties created by the regional financial crisis. The project was designed as a wholesale banking operation, incorporating fully the lessons learned from RFP and RFP II 12/ Discounted at 10% real interest rate. Using a higher real discount rate (12% or 15%) would increase the net benefit from US$1.2 million to US$2.1 million and USS3.1 million respectively. 6.2 Transition arrangement to regular operations. The first round of fund use under the project is virtually complete. Future operations revolve around LBP continuing to collect the sub-loans from the PFIs and making new sub-loans using rolled over funds for similar purposes. This process is already in place. The detailed operational manual will continue to be used and lending procedures will be harmonized with those of the successor project (TRFP). The institutional strengthening of LBP, initiated under this project through the ISAP will be continued under TRFP's Institutional Development Plan (IDP). The performance of this will be monitored through the TRFP KPIs. 7. Bank and Borrower Performance Bank 7.1 Lending: The main elements of Project design (CLF and ISAP) were relatively simple, and took account of lessons learned under the earlier Rural Finance Project (RFP). The initially proposed RCF sub-component, was new and untested. It was intended to attract additional commercial finance to SME countryside lending as well as to provide LBP with retail funds under the Project. The appraisal process was relatively smooth and rapid. However, there was a substantial delay, caused by GOP, between appraisal and negotiations with the project being negotiated some time after the appraisal mission. The decision by the Bank at negotiations to agree to a doubling of the retail (cofinancing) element which had not been prepared in detail, but then to insist upon 75% contribution from other banks was probably strategically sound. It kept the project alive but created a component which was unlikely to (and indeed did not) move. In due course, this component was amended. The time period from negotiations to board approval was normal and the project became effective in Spring It has been the practice in some Bank credit projects to establish an environmental unit in the financial intermediary processing the loan. Ideally, it is preferred that state environmental authorities should look after monitoring and enforcement of environmental regulation. Experience in the Asia region shows, however, that strengthening the national agencies is a slow process, often inefficient, that is never available in time to serve the individual project under discussion. The Bank observed this situation in the Philippines at the time of Project preparation, and, having evaluated DENR agreed with LBP to incorporate an - 10-

15 Environmental Unit (EU) within the project structure, to enhance compliance. Because it lacked the leverage, the Bank did not try to insist that all Philippine banks operate in this way. The Bank also made a sensible decision to accept compliance with GOP regulations as the sole environmental eligibility criteria under the project. This simplified the project's environmental requirements and eliminated more demanding, complex Bank procedures. The training program incorporated in the project helped raise environmental awareness inside and outside the LBP. 7.2 Supervision: Following loan effectiveness, the Bank mounted supervision missions once 13/ a year. Most supervision missions were staffed by a financial analyst and a financial/agricultural economist, with additional support, on several occasions, from an environmental specialist. The supervision missions worked closely with LBP on resolving any problems concerned with the credit lines, including amending the size and scope of the RCF component. They also participated in the review and updating the ISAP, helped fine tune policies to ensure adequate provisioning for agrarian loan losses and maintained dialogue with GOP to ensure that the specific covenants which protected LBP's financial strength were met. On environment, the Bank's initiative has meant that investments under the project are more environmentally sound than they would otherwise have been, but it is not clear whether the improvement in environmental systems will extend any further. An important feature of this project was Bank staff and consultant continuity. This was appreciated by the borrower and led to good communications and a high degree of trust and understanding. During supervision, the Bank's missions identified a significant gap between legal eligibility to borrow (from an environmental standpoint) and actual compliance with environmental regulations. To monitor and help to close this gap, environmental audits have been incorporated in project implementation. As a result of the EU introduced under the project, investments have been more environmentally sound than they would otherwise have been. But the project has not come up with an improvement in environmental systems which is sure to be sustainable. The Bank's performance with regard to environmental matters is therefore rated as satisfactory rather than highly satisfactory. 13/ Because there were other ongoing activities with LBP during the project period involving the same task manager and consultants, closer contact was maintained with the project than the one per year supervision missions indicated here. 7.3 Overall Bank performance: Overall, the Bank's performance is rated as satisfactory on project design and appraisal and highly satisfactory during implementation. Borrower 7.4 Preparation: Preparation of the project was largely undertaken by LBP, based on the knowledge gained from the RFP. This was professionally undertaken and the main elements of the project, CLF and ISAP were harmoniously developed together with Bank support and can be considered highly satisfactory. There was little preparation undertaken on the RCF component however which is one of the reasons that in its initial form it did not move. This is considered unsatisfactory. Taken overall, the borrowers preparation performance can be considered as satisfactory. 7.5 Government implementation performance: GOP's performance has been less satisfactory than LBP's. It delayed the project process for about a year

16 by postponing negotiations, but has still ended up with a project which is pretty close to the original design. As Guarantor of the project GOP has needed to be reminded of its obligations, both in the area of dividend policy 14/, and in ensuring that ARF does not become overdrawn at LBP beyond the covenanted level of P1 billion. Overall, the performance of the project's borrower and guarantor, taken together can be viewed as satisfactory, providing GOP cancels its 1998 dividend proposals and quickly restores the ARF balance to an acceptable range. 14/ GOP's Initial proposals for dividend payments on ordinary shares out of 1998 nominal earnings were contrary to the loan and guarantee agreements in that they sought cash dividends even though LBP's earnings were insufficient to cover erosion of equity by inflation, adequate provisions for loan losses and preference share dividends. 7.6 ImplementingAgency: LBP, the principal implementing agency and the borrower has generally performed well under the project. On the credit side, CLF funds moved quickly and in line with project estimates. Targets on numbers of loans, numbers of PFIs and flows through RBs and TBs indicating diversity of coverage were all met. Collection rates and levels of past dues have been good but some problems resulting from the financial crisis appear to be surfacing. Until July 1997, there were no past dues to LBP under CLF, but now (June 1999), past dues from the PFIs to LBP amount to 0.82% - still a pretty low level. Performance under the wholesale part of the project has been highly satisfactory. The main credit problem has been the fact that the RCF did not move. Probably, the original design was insufficiently researched, although there also appear to have been problems within LBP's organization of providing appropriate incentives to lending units to make use of the funds. The program has now been redesigned somewhat and LBP branch management expect that the reduced RCF (US$20 million) will be disbursed in the next 18 months, on time for project completion. LBP has made progress with most elements related to institutional strengthening, although the KPIs which were set shortly before the financial crisis were not all met. LBP has taken the agreed actions to safeguard financial strength, but growth in real equity has been lower than planned. Lending objectives have been realistically set. Quite rationally, in view of market conditions, no long term peso instruments have been introduced, but LBP intends that these will be introduced before project closure. The costs of agrarian lending are still excessive, but a marked improvement has taken place between 1997 and Agrarian loan loss provisioning has been rationalized in line with the ISAP proposals. Cooperative strengthening has been implemented and the agrarian loan data base improved. The internally funded training and technical assistance program under the project has been implemented more or less as envisaged. Overall, LBP performance on this part of the project has been satisfactory. Project monitoring and reporting has been satisfactory, with an agreed format, which allows performance against KPIs to be monitored, having been adhered to and reports provided on time. 7.7 Overall Borrower performance: Overall, the performance of the project's borrower and guarantor, taken together can be viewed as satisfactory, providing GOP cancels its 1998 dividend proposals and quickly restores the ARF balance to an acceptable range. 8. Lessons Learned The generally successful project outcome underlines a number of positive experiences on which to draw in future and lessons learned when implementing credit projects in countries with a diverse banking system

17 Firstly, using a sound wholesale institution with a strong accreditation unit to appraise PFIs, rather than having the Bank review each can work satisfactorily in a system where there are many potential lenders. Secondly, this type of project is a suitable mechanism for disbursing relatively large loans quickly to the private sector, without placing much burden on Government. By on-lending at domestic market rates and having the Government benefit from the interest rate arbitrage, but bear the foreign exchange risk, the long term liklihood is that the Government will not lose and that domestic markets will be strengthened. Furthermore, any benefits derived from the fact that a country can borrow Bank funds more cheaply than commercial funds would accrue to Government. This type of operation is more transparent with single currency libor linked loans than with pool loans. 9. Partner Comments (a) Borrower/implementing agency: RURAL FINANCE PROJECT II (RFP 11) IMPLEMENTATION COMPLETION REPORT I. Evaluation of the Proiect A. Project Obiectives - The objective of the project which was to help expand the volume of medium and long term commercial credit to agriculture and rural development in the Philippines was highly relevant at the time of project inception considering the developmental needs of the Philippine countryside. Moreover, with the underdeveloped capital markets in the country, the supply of long-term funds was very limited. - The project aimed to fuel economic activity in the rural areas and support the Philippine government's regional dispersal policy. This is meant to achieve a balanced regional development structure with strategic growth poles in the various regions outside the National Capital Region. * As a credit program, RFP II was designed to provide long-term financing to private sector priority projects in the countryside. The focus sectors were agriculture, fishery, agro-processing, manufacturing and services. These are high impact sectors with large multiplier effects in rural areas. * RFP II also aimed at increasing the supply of long-term financing for small and medium enterprises (SMEs) and further boost the development of the SME sector. The country's vision is to develop SMEs as the principal engine of economic growth. * With regard to environment, sub-projects financed should be environmentally sound and comply with existing environmental laws and regulations of the Philippines. Such objective would safeguard the use of the fund to sub-projects which would not adversely impact on the environment and thus, promote sustainable development. * The project will continue to be significant as the long term financial market in the Philippines has not yet fully developed. Most banks do not readily lend on long term basis unless long term

18 funds are tapped from special financing programs such as the Countryside Loan Fund II. For the institutional component of the project, the principal objectives of the RFP 11 were as follows: 1. To strengthen LBP as a development financing institution and deepen its impact as the principal source of formal credit for small farmers and fisherfolk. 2. To expand LBP's role as a catalyst of countryside development. 3. To enhance LBP's financial soundness and delivery capability as well as operational viability. B. Project Design * The design of the wholesale credit component was highly effective as the project was able to build-up from the experiences, organizational set-up, and systems and procedures of the two predecessor projects which LBP successfully implemented - the Countryside Loan Fund I and the Agricultural Loan Fund. * The project proved anew that a wholesale approach, with LBP as the apex financial institution, was effective in reaching out to more beneficiaries nationwide. It was able to capitalize on LBP's financial strength and capability to implement the accreditation process and performance review of PFIs. LBP's strong presence in the rural areas, as well as the geographical location of the rural and thrift banks, and the branch network of the universal/commercial banks also contributed to the wide dispersal of the fund. * lthe project was likewise designed to respond to the needs of the countryside economy at the time of its inception. In addition to the eligible projects of CLF I, CLF II was able to assist tourism-related and environment protection projects. Moreover, CLF II loan in US Dollar denomination became available to serve the long term foreign currency loan requirements of sub-borrowers. * Same cannot be told, however, on the original design of the Retail Countryside Fund, the loan component for direct lending to LBP's clients. The initial RCF project design required the participation of a co-financier. However, the sub-borrower's asset-size limit of P 250 Million was too small to attract co-financing. The asset size likewise discouraged loan syndications as small loan size would not be cost efficient. The rural banks, on the other hand, preferred to avail CLF II under the wholesale operations rather than participate as a co-financier. The RCF features were subsequently amended with the deletion of the co-financier and increasing the maximum asset size and loan size. With such amendments, LBP management is confident that the remaining fund will be fully disbursed before the end of year * Effective systems and procedures were in place to ensure that prior to funding, sub-projects either had the necessary environmental clearances/permits or had been reviewed as non-coverage projects. However, there were weaknesses in the monitoring of the sub-borrowers' compliance to the conditionalities of the Environmental Compliance Certificate (ECC) approval, principally because of the government's weak enforcement of environmental laws and regulation. LBP addressed this thru itsown environmental audit but thecoverage of environmental audits being conducted by LBP's Environmental Unit (EU) was constrained by the limited number of EU personnel

19 C. Achievement of Obiectives Credit Component - Achievement of the project's credit objectives was rated highly satisfactory as proven by the following: The project increased the volume of medium and long term credit in the countryside with an aggregate of P3, Million for 672 sub-projects as of March 31, More importantly, the project was able to encourage the financial participation of the PFIs and sub-borrowers at a level exceeding the minimum requirement of the project. (The PFIs and sub-borrowers provided 12% and 30%, respectively, exceeding the minimum 10% and 15% participation). It could not be claimed, however, that these project investments would not have pushed through without CLF II funding as these viable projects could have otherwise tapped alternative sources of financing. * The loan releases and investment activities helped fuel regional development and contributed to the regional dispersal policy of the Philippine government. Loan releases were distributed among the various regions. The highest availment rates were in the following regions: Southern Tagalog, Central Luzon, Southern Mindanao, Western Visayas, and Ilocos. - The objective to expand long-term financing for SMEs was adequately met. SMEs accounted for the bulk of the loans granted with a 67% share or P2.68 Billion. In fact, for "small enterprises", the total loans granted amounted to P1.81 Billion representing 45% of total loans. * Loan drawdowns had always been ahead of the 5 year drawdown schedule. After less than 3 years from effectivity date, 82% of the US$150 Million had been drawn from the Loan Account. * The project's key performance indicators which included total project investments, number of loans, amount and number of loans released thru thrift and rural banks and employment generation were all surpassed. * The project contributed positively to the overall financial performance of LBP, the PFIs and the sub-borrowers. Aside from increased economic activity in the countryside, the project created socio-economic impact through employment generation and revenues to the government. * All sub-projects had been reviewed and endorsed as environmentally acceptable by LBP's EU prior to CLF II funding. The project likewise (a) promoted awareness of the PFIs and sub-borrowers with the prevailing environmental laws and regulations; and (b) enhanced the PFI's environmental due diligence through the conduct of relevant trainings and seminars. Institutional Strengthening Action Plan (ISAP) - In 1994, LBP prepared an ISAP for ISAP was a medium-term business plan which outlined the main actions which were to be taken by LBP to maximize its economic development impact while maintaining a high level of financial health. The achievement of the ISAP and RFP II institutional objectives was highly satisfactory. The strategic impact of the actions taken is considered extensive. The benefits registered in the plan period 1994 to 1998 shall continue to be recorded in the coming years