Philippines - Cottage Enterprise Finance Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 15172 PROJECT COMPLETION REPORT PHILIPPINES COTTAGE ENTERPRISE FINANCE PROJECT (LOAN 3312-PH) DECEMBER 14, 1995 Infrastructure Operation Division Country Department I 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.  CURRENCY EOUIVALENTS Currency Unit = PesoP US$1.00 =P 24.5 (at Appraisal, July 1990) US$1.00 =R 24.3 (Average 1990) US$1.00 =P 26.7 (Average 1991) US$1.00 =P 25.0 (Average 1992) US$1.00 =P 27.8 (Average 1993) ABBREVIATIONS CEFP - Cottage Enterprise Finance Project DBP - Development Bank of the Philippines GOP - Government of the Philippines IGLF - Industrial Guarantee and Loan Fund KfW - Kreditanstalt fuer Wiederaufbau MGA - Mutual Guarantee Association PFI - Participating Financial Institution SBGFC- Small Business Guarantee and Finance Corporation SMI - Small and Medium Industries TA - Technical Assistance FISCAL YEAR GOP - January 1 to December 31 DBP - January 1 to December 31  FOR OFFICIAL USE ONLY THE WORLD BANK Washington, D.C. 20433 U.S.A. Office of Director-General Operations Evaluation Decmnber 14, 1995 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Completion Report on Philippines Cottage Enterprise Finance Project (Loan 3312-PH) Attached is the Project Completion Report for the Philippines Cottage Enterprise Finance Project (Loan 3312-PH, approved in FY91) prepared by the East Asia and Pacific Regional Office. Part 11, contributed by the Borrower comments on project implementation and lessons learned. The objective of this US$ 15 million loan was to channel funds to cottage industries, essentially micro enterprises, with major employment generation capabilities (including women) outside the Manila area. The Development Bank of the Philippines (DBP) was chosen as the apex institution for onlending through participating financial institutions (PFIs), mostly commercial banks. This objective was only marginally achieved. At the request of the Government, US$ 13.5 million (90% of the loan amount) were canceled in August 1993, about two years after approval. The cofinancier, KfW, canceled its loan at the same time. Most of the loan was not disbursed, mainly because the commercial banks were not interested in lending to cottage industries. The banks took issue with the high costs of processing the subloans and continued to regard cottage enterprises as high risk borrowers. Other circumstances that had a negative impact on the project included: (i) an increase in subsidized loans to the sector from other sources, and (ii) the creation of a new government guarantee program which further reduced the need for the project. Bank and Borrower performances were wanting in several respects. The project design included cumbersome legal arrangements for onlending, and the Bank overestimated the willingness of commercial banks to participate in the project. The Borrower showed a high degree of cooperation in general, but proved reluctant to initiate a complicated arrangement without a pilot project. The project outcome is rated as unsatisfactory, its sustainability as unlikely, and its institutional development impact as negligible. The PCR is of reasonable quality but does not include the views of the cofinancier. Part 11 from the Borrower rightly focuses on fundamental issues in project design, asking whether banks are the right financial intermediaries for lending to cottage industries. An audit is planned in parallel with an audit of a Financial Sector Adjustment Loan (Loan 3049- PH). Attachment This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents nmay not otherwise be disclosed without World Bank authorization.   FOR OFFICIAL USE ONLY PHILIPPINES COTTAGE ENTERPRISE FINANCE PROJECT (LOAN 3312-PH) PROJECT COMPLETION REPORT TABLE OF CONTENTS Preface .......................................................... Evaluation Summary .............................................. Part I: Project Review from Bank's Perspective ....................... 1 Project Identity ................................................. 1 Project Background ............................................... 1 Project Objectives and Description .............................. 2 Project Design and Organization ................................. 2 Project Implementation ........................................... 3 Project Results .................................................. 5 Project Sustainability ........................................... 6 Bank Performance ................................................. 6 Borrower Performance ............................................. 7 Project Documentation and Data ................................... 7 Main Lessons ..................................................... 7 Part II: Projec4 Review from Borrower's Perspective .................. 9 Part III: Statistical Information ................................... 19 Table 1 Related Bank Loans ......................................... 19 Table 2 Project Time Table ......................................... 20 Table 3 Loan Disbursements ......................................... 21 Table 4 Use of Bank Resources and Missions ......................... 22 Table 5 Status of Legal Covenants .................................. 23 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. l   - i - PHILIPPINES COTTAGE ENTERPRISE FINANCE PROJECT (LOAN 3312-PH) PROJECT COMPLETION REPORT Pretace The Cottage Enterprise Finance Project (Loan 3312-PH) was approved by the Board on March 26, 1991, and is the subject of this Project Completion Report (PCR). While the Bank has made a number of Loans in support of the industrial sector, this is the first Loan specifically for cottage industry, and uses a concept that has not previously been adopted by the Bank in its lending to cottage and micro firms. The Loan funds were channeled through the Development Bank of the Philippines (DBP) for relending to participating financial institutions to be loaned to cottage industry entities. The Loan is guaranteed by the Government and has a scheduled closing date of June 30, 1996. There was a single disbursement of US$1.5 million made on August 3, 1992 under the Loan and the remaining US$13.5 million was canceled on August 11, 1993 at the Government's request. The Borrower's repayments are current in accordance with the amortization schedule. The co-financier was the Kreditanstalt fuer Wiederaufbau, who provided DM 20 million (US$13.4 million) of which DM 17.5 million was earmarked for lending to cottage industry and DM 2.5 million represented a grant to fund technical assistance. In line with the Bank Loan, the co- financier also canceled its Loan in August 1993. The PCR (Preface, Evaluation Summary, Parts I and III) was prepared by the Infrastructure Operations Division of Country Department I, East Asia and Pacific Regional Office. DBP, the apex institution, provided Part II and data for Part III. The PCR draws on the Staff Appraisal Report; the Loan, Guarantee, and Japanese Grant Agreements; Supervision Reports, correspondence between the Bank and DBP, internal Bank memoranda and economic/sector reports, supplemented by the staff interviews; DBP's financial and audit reports; and subproject data provided by DBP.   - ii - PHILIPPINES COTTAGE ENTERPRISE FINANCE PROJECT (LOAN 3312-PH) PROJECT COMPLETION REPORT Evaluation Summary 1. The Board approved a loan of US$15 million on March 26, 1991. The Kreditanstalt fuer Wiederaufbau (KfW) provided co-financing for DM 20 million (US$13.4 million) of which DM 17.5 million was on terms similar to IDA and DM 2.5 million as a grant to finance technical assistance. The objective of the project was to channel funds to cottage industries, using the Development Bank of the Philippines (DBP) as the apex institution for onlending through participating financial institutions (PFIs) (para 3.1). 2. The project contained three basic components: (a) subloan financing for cottage enterprises belonging to MGAs available through accredited PFIs; (b) technical assistance in the supervision and training of MGAs, the implementation of pollution abatement measures, and training of PFIs in evaluating cottage enterprise projects; and (c) formation of guarantee funds comprising subscriptions by MGA members, supplemented by matching loans from DBP (para 3.2). 3. While the project was designed essentially to surmount obstacles on collateral faced by cottage firms and improve their access to bank credit, the project design was flawed, inevitably cumbersome and involved considerable inter-institutional collaboration. Legal documentation within the project was overly complex and time consuming (paras 4.1-4.4). 4. Project implementation was problematic. Of the total US$15 million loan approved for the project, US$13.5 million or 901 was canceled. The principal bottleneck proved to be the overall attitude of the PFIs towards cottage enterprise lending. The MGAs were not all effective. There were also extraneous circumstances which had a significant impact on the project. First, there was an influx of new funds to the cottage and small and medium industries (SMI) sectors as government financial institutions began funding directed credit (subsidized) programs and as banks. Second, the establishment of a new government institution, the Small Business Guarantee and Finance Corporation (SBGFC), replaced the underlying need for the project. Project appraisal failed to take into account the impact of this GOP guarantee program on the project (paras 5.1-5.6). 5. The principal objective of channeling funds to cottage firms was marginally achieved. By the end of 1993, the Bank had disbursed onlyP 41.9 million, a quarter of theR 176.4 million that had been expected at appraisal. A total of 912 subprojects amounting toP 72.6 million were financed, of whichP 62.6 million (86%) went to the cottage sector andP 10 million (14%) to small enterprises. About 48% of project lending went to women entrepreneurs. Some 7,805 jobs were created as a result of this lending. Total estimated project cost amounted toP 3.5 billion orP 448,000 (US$16,000) per job. Although the number of MGAs which were established is in some sense satisfactory when compared with the project  - iii - goal, their strength and performance were not adequate. By September 1993, when the registration of new MGAs ceased, 39 MGAs had been established, against the 4-year target of 60 MGAs by the end of 1994 (paras 6.1-6.5). 6. With the cancellation of the project, DBP's involvement in the cottage industry sector is likely to be phased out fairly rapidly. The key factor will be the lack of funds accumulating in DBP's credit risk protection fund (protecting DBP against the credit risk arising from its matching loans to MGAs), which was to be funded through the interest rate differential on the KfW loan. With the termination of the KfW loan, the DBP protection fund cannot increase and, as a result, DBP is unlikely to provide additional matching loans. In any event, the function of these loans has been superseded by the SBGFC. To date, discounting of the PFIs' cottage industry loans by DBP has already ceased and the DBP component of the project has been terminated. The sustainability of MGAs' operations is now uncertain as the needed institution building is not forthcoming and since their loan guarantee role has also, to some extent, been replaced by the public sector SBGFC (paras 7.1-7.2). 7. There were lessons learned from the project. First, thorough project preparation reduces the likelihood of difficulties during project implementation by anticipating some of the problems that may arise and devising methods to deal with them. In this project, an in depth review of MGAs would have revealed their lack of sophistication and training and the need for lengthy institutional building. A review of the registration procedures for MGAs would have revealed that substantial simplification would be essential. An analysis of financing project cost, in terms of commitment charges, staff time and travel expenses, and the use of higher priced consultants would have revealed the inappropriateness of Bank lending for very small projects in the informal sector. In addition, the institutional arrangements with the financial intermediaries should have been established and reliably worked out well in advance. The project could have benefitted from assurances obtained from the PFIs, while the project was in the preparatory stage, of their firm commitment and close partnership (para 11.2). 8. Second, realizing the complexities of the new approach, the Bank should have insisted on the run-through, first of a pilot project, regardless of the resulting delay in getting a full-scale project underway (para 11.3). 9. Third, a project of this type, with cottage entrepreneurs, the informal sector, a new organizational structure, and a whole new approach to lending requires on-going, intense and effective TA (para 11.4). 10. Finally, given the scarcity of trained staff, managerial skills, and administrative capacity of MGAs, special efforts to simplify project objectives and to select technologies and approaches that have a high probability of working in the particular environment would have paid off better for the project. The use of grass roots and already existing organizations such as NGOs in the field and the provinces would have been a more effective approach in implementing the project. The MGA concept adopted was an innovative lending approach and in the long run, through the use of more effective conduits, would have been more beneficial to the cottage industry than the public sector approach adopted by the GOP (para 11.5).  PHILIPPINES COTTAGE ENTERPRISE FINANCE PROJECT (LOAN NO. 3312-PH) PROJECT COMPLETION REPORT Part I: Project Review from Bank's Perspective 1. Project Identity Project Name Cottage Enterprise Finance Project (CEFP) Loan No 3312-PH RVP Unit East Asia & Pacific Country Philippines Sector Industry Subsector Private Investment 2. Project Background 2.1 The Philippine industry had grown under an import substitution regime, marked by protection, heavy foreign borrowing and by infusions of creaits and transfers from the Government of the Philippines (GOP). This uncompetitive structure could not be sustained and, as a result, industrial output declined by 25% between 1982 and 1986. Although the GOP began a process of gradual removal of major distortions in 1980, the economic and political downturn of 1982-86 gave a strong impetus to the pace of these reforms. By 1990, major progress had been achieved in the key areas of tariffs, incentives, foreign exchange management and privatization of public industrial assets. The degree of transformation represented a remarkable economic achievement for the country. 2.2 The cottage industry segment of the cottage, small-scale and medium-scale industries in the Philippines is defined as enterprises with total assets between P 50,000 and P 499,999 (about US$2,000-20,000), or with less than 10 employees. This segment, which excludes the large informal household sector, is particularly important in terms of employment; accounting for over one million individuals or 40% of formal non-agricultural employment. Another positive feature is the segment's countrywide dispersal; about two-thirds of the industries' value added is derived from outside Metro Manila, compared to only one third for larger industries. The cottage industry, therefore, has the potential for improving the regional dispersal of industries and employment.  - 2 - 3. Proiect Obiectives and Description 3.1 Project Obiectives. The Board approved a loan of US$15 million on March 26, 1991. The Kreditanstalt fuer Wiederaufbau (KfW) provided co-financing for DM 20 million (US$13.4 million) of which DM 17.5 million was on terms similar to IDA and DM 2:5 million was given as a grant to finance technical assistance. The' objective of the project was to channel funds to cottage industries, using DBP as the apex institution for onlending through participating financial institutions (PFIs) accredited by DBP. The project was targeted at: (a) establishing 60 Mutual Guarantee Associations (MGAs) with about 4,200 member enterprises; and (b) providing funds with a wide geographical distribution, and to a significant proportion of women entrepreneurs. 3.2 Project Description. The project contained three basic components: (a) subloan financing for cottage enterprises belonging to MGAs available through accredited PFIs; (b) technical assistance in: (i) the supervision and training of MGAs; (ii) the implementation of pollution abatement measures by MGA member cottage enterprises; and (iii) training of PFIs in evaluating cottage enterprise projects by DBP; and (c) formation of guarantee funds comprising subscriptions by MGA members, supplemented by matching loans from DBP also to be used for guaranteeing MGA member subloans. 4. Project Design and Organization 4.1 To surmount obstacles on collateral faced by cottage firms and improve their access to bank credit, the project aimed at setting up guarantee funds to be derived from two sources. First, groups of cottage firms (each group consisting of 40 to 100 members) would join together to establish MGAs. The groups' equity subscriptions, averaging R 20,000, would be pooled as guarantee funds. Second, these pooled funds would be supplemented by matching loans to the concerned MGAs by DBP, at an initial ratio of three times the amount of each particular MGA's guarantee fund. The matching loans would be held by DBP while the MGA funds by the PFI selected by the MGA. The MGAs would be able to issue guarantees covering up to 80% of each loan, with the total amount of MGA guarantees limited by a ratio of total loans to guarantee fund, agreed between the MGA and the PFI. The arrangement was expected to attract PFIs since the guarantee funds would be on their books, and could be drawn upon as needed without complex recourse procedures. Since members would be liable if a guarantee were called, MGAs were expected to exercise considerable moral suasion on delinquent members. 4.2 A particular feature of the project was the emphasis on women entrepreneurs who dominate the cottage sector. The MGA management unit  made specific efforts to promote the project to women entrepreneurs and managers. 4.3 Overall supervision of the project was undertaken by a review committee in DBP, consisting of GOP agencies concerned with cottage industry and the PFIs' representatives. DBP acted as a wholesale bank, rediscounting subloans made by accredited PFIs and monitoring their performance according to agreed criteria. DBP also financed and administered the guarantee fund of about US$10 million provided by MGAs' subscriptions and DBP's matching loan fund. The MGA management unit monitored the MGAs and reported to the review committee. 4.4 The design was inevitably cumbersome and involved considerable inter-institutional collaboration. Legal documentation within the project was overly complex and time consuming, consisting of separate agreements between DBP and the PFIs, the MGA management unit and the MGAs. Legal documentation for incorporation of MGAs and their registration with the Securities and Exchange Commission, legal agreements between MGAs and their selected PFIs, including the trust and guarantee agreements, subsidiary loan agreements between DBP and the PFIs, and the memoranda of agreement between DBP and the MGAs for regulation of matching loan funds were also required during project implementation. Given the education level of cottage industry entrepreneurs, the legal arrangements were complex and difficult to understand. DBP estimated that the process of full Securities Echange Commission's registration took about a year and, in retrospect, realized that considerable simplification would have been possible and would have benefitted the project. 4.5 The technical assistance (TA) component, considered vital to the project's success, was channeled through DBP to a central MGA management unit responsible for implementing the program and assisting individual MGAs. The TA funds were made available for four years from loan effectiveness date (August 1991). 5. Proiect Implementation 5.1 Of the total US$15 million loan approved for the project, US$13.5 million or 90% was canceled. During appraisal, US$50 million was initially planned for the project, subsequently reduced to US$25 million, and finally to US$15 million because of GOP's concern over the commitment fee for the uncommitted portion of the loan, and because, according to Bank staff, the borrowing capacity to the cottage enterprises did not warrant more than US$15 million Bank loan. The reduced loan amount was agreed upon between the Bank and GOP on the understanding that a follow-up loan could be made after the initial Bank/KfW loans had been fully committed. 5.2 Project implementation was problematic. At the outset, both DBP and the Bank were dissatisfied with the performance of the initial foreign consultants. Although a change was made at the early stage of project implementation, the change resulted in a delay of about one year in establishing and registering MGAs. The performance of the subsequent consultants was also not satisfactory.  -4- 5.3 The principal bottleneck, however, proved to be the overall attitude of the PFIs towards cottage enterprise lending. Not only was it difficult to select qualified PFIs, but once accredited, these PFIs acted with great conservatism, applying exacting credit criteria for financing cottage loans. Many branches of accredited PFIs had no cottage lending staff, for the most part, these branches functioned only to accept deposits. The PFIs also took issue with the high cost of processing cottage subloans and proposed that the MGAs undertake more of the administrative work on subproject processing, reducing PFIs' time and cost. Therefore, despite the guarantee mechanisms in place, the PFIs continued to regard cottage enterprises as high risk borrowers and did not vigorously pursue cottage borrowers. 5.4 Realizing this deficiency in the lending link between DBP and the final borrowers during project implementation, several alternatives were considered. The idea of using rural banks, who have close affinity with the cottage industry, as active collaborators with nFIs, was considered. Another alternative was to have a two-tiered system in which PFIs would be responsible to DBP, but essentially on-lent through rural banks. Both alternatives were not considered because rural banks were regarded as financially weak institutions. However, some rural banks were eventually accredited by DBP in utilizing their own funds for cottage enterprise lending. 5.5 The MGAs were not all effective. They displayed a lack of homogeneity and cohesiveness and, in the process, diluted their basic advantage--the force of peer pressure, an essential element in the regulation of member firms. In order to increase the MGAs' cohesiveness, various initiatives were undertaken such as grouping MGAs by no more than three industries in each MGA; splitting large MGAs; drawing in more sophisticated, educated leaders to provide better management and attract smaller entrepreneurs; and increasing the minimum asset size from P 1.0 million to P 5.0 million. 5.6 There were also extraneous circumstances which had a significant impact on the project. First, there was an influx of new funds to the cottage and SMI sectors as government financial institutions began funding directed credit (subsidized) programs and as banks, in accordance with legislation in 1991, were required to allocate 10% of their loan portfolios to the cottage and SMI sectors. Second, the establishment of a new government institution, the Small Business Guarantee and Finance Corporation (SBGFC), replaced the underlying need for the project. SBGFC, whose general objective is to promote, develop and assist small- and medium-scale enterprises, was established by law in January 1991, prior to loan signing for the project. Project appraisal failed to take into account the impact of this GOP guarantee program on the project. With guarantees from the SBGFC, cottage entrepreneurs could borrow directly from PFIs without the need for MGAs. The creation of SBGFC therefore represented a simpler (although also probably riskier --its structure was criticized by the Bank) alternative to the complicated MGA structure. Consequently, the project's actual contribution to the cottage sector became negligible. 5.7 Although the Bank made constructive suggestions to allow for more flexibility in subproject lending, the project was designed so that  - 5 - each MGA became a small, intermediate institution in the process of developing expertise, relationship, and credibility with the PFIs. While the cottage sector benefitted from the MGA concept, it became clear that developing successful cooperatives or MGAs requires time, experience, and training. It became apparent that a more successful approach would have started with already existing organizations,, in which members knew each other and had worked together. It also became apparent that TA to the MGAs should have been provided by local (regional and provincial) non- governmental organizations (NGOs). Using higher priced (even though local) consultants located in Manila prevented the TA from being cost effective. The long term nature of successful institution building was not recognized early enough. Also, in project formulation (in spite of interviews with bankers during the appraisal process) bankers' risk aversion and distaste for non-collateralized lending was underestimated. In fact, some PFIs went so far as to insist that their lending to cottage firms only be on the basis of a 1:1 ratio with the guarantee fund. As a result, the risk-sharing objective of the project was suObstantively diluted. 6. Project Results 6.1 The principal objective of channeling funds to cottage firms was not sufficiently achieved. By the end of 1993, the Bank had disbursed only P 41.9 million, a quarter of the P 176.4 million that had been expected at appraisal. A total of 912 subprojects amounting to P 72.6 million were financed, of which P 62.6 million (86%) went to the cottage sector and P 10 million (14%) to small enterprises. About 48% of project lending went to women entrepreneurs. Some 7,805 jobs were created as a result of this lending. Total estimated project cost amounted to R 3.5 billion or R 448,000 (US$16,000) per job. All loans were made for working capital resulting in an expansion of the sector's inventory and production. While the collection rate of 77% was undesirable, the default rate involving drawdowns on the guarantee and matching loan funds remained at a moderate 4.3%. The situation indicates that most of the arrears (18.5%) should be collectible, and represents slow repayments rather than eventual losses. DBP paid the PFIs i 1.5 million as its share of defaults by certain MGA members. The amount for the defaults was drawn from the KfW interest rate subsidy; DBP's matching loan funds for principal, however, remained intact. Although the losses are not likely in any event to be serious, the PFIs and DBP continue to monitor the situation. 6.2 Although the number of MGAs which were established is in some sense satisfactory when compared with the project goal, their strength and performance were not adequate. By September 1993, when the registration of new MGAs ceased, 39 MGAs had been established, against the 4-year target of 60 MGAs by the end of 1994. The established MGAs are dispersed over seven regions with 19 of them receiving matching loans from DBP, while other MGAs' applications are still being processed. Apart from two MGAs, the rest were able to contain loan defaults at modest levels. The overall default ratio, as of December 31, 1993, stood at 4.3%. Nevertheless, in spite of the apparently satisfactory numbers, the MGAs were not strong enough to help members with loan applications and legal documentation, and to convince the PFIs to lend to them.  - 6 - 6.3 DBP accredited 18 financial institutions under the project. Although difficulties were experienced in securing suitable PFIs, those accredited have not suffered any loss on their lending to cottage firms. 6.4 DBP adequately performed its role in project implementation. It accredited PFIs in accordance with established procedures and carried out the required semi-annual reviews, managed the Bank and KfW,loans, undertook the oversight of the TA component efficiently, and conducted the necessary verification of subloans. 6.5 Despite the hiatus in the TA component due to a change from the originally-appointed consultants, the consultants gave important assistance to the formation of the MGA scheme. Although the consultancy services were terminated on September 30, 1993, such training as had been provided to the management unit had been of appreciable benefit to its management and staff. 7. Project Sustainability 7.1 With the cancellation of the project, DBP's involvement in the cottage industry sector through the MGA approach of this project is likely to be phased out fairly rapidly. The key factor will be the lack of funds accumulating in DBP's credit risk protection fund (protecting DBP against the credit risk arising from its matching loans to MGAs). These funds were to be funded through the interest rate differential on the KfW loan. With the termination of the KfW loan, the DBP protection fund cannot increase and, as a result, DBP is unlikely to provide additional matching loans. In any event, the function of these loans has been, to some extent, superseded by the SBGFC with the GOP now taking the risk on cottage enterprise loans. Indeed, discounting of the PFIs' cottage industry loans by DBP has already ceased and the DBP component of the project has been terminated. 7.2 Although it is possible that successful MGAs could continue operations, if they are satisfactorily integrated into the new arrangements designed for the cottage industry, they are no longer necessary under the SBGFC scheme. Therefore, the MGA's overall prospects appear dim. The sustainability of MGA operations is now uncertain as the needed institution building is not forthcoming and since their loan guarantee role has also, to some extent, been replaced by the public sector SBGFC. 8. Bank Performance 8.1 Overall, the Bank's performance was found wanting in many respects. While it was especially keen on the innovative nature of the project (lending to cottage firms using the MGA concept and an apex institution for distribution of Bank funds) and exercised flexibility during project implementation in constructively addressing the difficulties with the PFIs, potential problems with project design and issues that may impact the project were not sufficiently anticipated. The Bank was also overly optimistic and aggresively marketed the project. Greater consultation with other institutions during appraisal would have  - 7 - been beneficial to get a better reading on their attitude towards this innovative lending approach. 9. Borrower Performance 9.1 There was a high degree of cooperation, during project preparation and implementation, between the concerned GOP agencies, DBP, KfW and the Bank. DBP kept itself closely involved with the PFIs and MGAs, and contributed significantly to the progress of the project. 9.2 While initially there was strong support for the project from the PFIs, DBP and the concerned GOP agencies, there was also some reluctance on the part of the GOP to initiate a complicated arrangement without a pilot project preceding the main lending thrust. However, the opinion at that time was that running a very small pilot project would have derailed the Bank project, as nervousness over the difficulties needed to be overcome might have rendered the institutions reluctant to commit themselves to the later, full-scale project. According to Bank staff, there was a sense that the TA required could ably be justified if the project were done on the scale proposed. With hindsight, it would have been preferable to confront problems head-on in a smaller pilot project. 10. Project Documentation and Data 10.1 All required papers were available for the assessment of the project. The periodic reports prepared by DBP and the consultants were found to be helpful in following the developments of the project. The reports and data recorded as a result of the Bank supervision missions were helpful and formed an adequate basis for the assessment of the project. 11. Main Lessons 11.1 The main lessons learned are: 11.2 Thorough project preparation reduces the likelihood of difficulties during project implementation by anticipating some of the problems that may arise and devising methods to deal with them. In this project, an in depth review of MGAs would have revealed their lack of sophistication and training and the need for lengthy institutional building. A review of the registration procedures for MGAs would have revealed that substantial simplification would be essential. An analysis of financing project cost, in terms of commitment charges, staff time and travel expenses, and the use of higher priced consultants would have revealed the inappropriateness of Bank lending for very small projects in the informal sector. In addition, the institutional arrangements with the financial intermediaries should have been established and reliably worked out well in advance. The project could have benefitted from assurances obtained from the PFIs, while the project was in the preparatory stage, of their firm commitment and close partnership.  -8- 11.3 Realizing the complexities of the new approach, the Bank should have insisted on the run-through, first of a pilot project, regardless of the resulting delay in getting a full-scale project underway. 11.4 A project of this type, with cottage entrepreneurs, the informal sector, a new organizational structure, and a whole new approach to lending requires on-going, intense and effective TA. 11.5 Given the scarcity of trained staff, managerial skills, and administrative capacity of MGAs, special efforts to simplify project objectives and to select technologies and approaches that have a high probability of working in the particular environment would have paid off better for the project. The use of grass roots and already existing organizations such as NGOs in the field and the provinces would have been a more effective approach in implementing the project. The MGA concept adopted was an innovative lending approach and in the long run, through the use of more effective conduits, would have been mo-e beneficial to the cottage industry than the public sector approach adopted by the GOP.  Development Bank of the Phil- WHOLESALE 30MNKrING GROUP } - 'May 4, 1994 Ms. REBECCA CANDOY-SBKSE I & E Operations Division DopartmQnt I, East Asia Region THE WORLD BANK li8e H. Street, N. W. Washington, D. C. 20433 Dear Me. Sekse: SubJect: Cottage Enterprise Finance Project (CEFP) - Loan No. 3312 PH We are transmitting herewith two (2) copies of the Proj,ct Completion Report, Part 1I for subject loan. '4e shall be pleased to provide additional information and/or cl.arification, if necessary. Our bect regards. Very truly yours, E* +DO f P GAJ^ Son or Vlice P cstdont Enclosed a a/S 0N4h l MMC 66JA%&7I. CW"OMua MP#~ j ~ ", Q. orE ir oW. |1 Z_ ms amv E 9 7* Nt&11.N _N SNI _I  - 10 - Development Bank of the Philippines COTTAGE ENTERPRISE FINANCE PROJECT PROJECT COMPLETION REPORT Part II I, GENERAL EVALUATION OF THE PROJECT Tho CEFP has a very good obJective: to assist the informal sectors of the Philippine economy (the cottage and the lower segment of small enterprises) which can not gain access to the lormal banks and avail of their credit facilities lor lack of collateral. The success of the project hinged on the guarantee mechanism to be established by the cottage entrepreneurs/enterprises themselves. Despite its good Intentions, the projoct sulffred certain flaws and setbacks as follows: 1. Only few PFIs participated in and accepted the project; 2. Slow formation and operationalization of the MGAs -- too many activities and documentation to accomplish before a MGA can operate; 3. There was pressure to form the MGAs simply for the purpose of getting loans and providing guarantee; 4. Improper promotion of the CEFP to intended beneficiaries which led to defaults of some MGA members. The above factors led to the cancellation of the undrawn balances of the CEFP loans which we foresaw can not be fully disbursed until the closing date on June 30, 1996. In spite of the said cancellations, the balance of the Special Accounts and second Generation funds are considered enough to continue with the project.  - 2.1- CEFP Project Completion Report, Part II II. PERE0RMANCE OF AGENCIES CQNCERNED A. I B R D Project Evolution CEFP was initially designed utilizing the services of Garson & Partners, a French-domiciled firm, which was responsible for gathering data in the Philippines ne-eded for the project design and the initial p-omotion and formation of the Mutual Guarantee Associations (MGAs). This was in Juno, 1989 until November$ 1990. The consultancy services was financed through an IBRD- administered grant to the Department of.. Trade and Industry (DTI), which was then the main proponent of the project. The Bank's intention in launching the project is to answer the needs of the informal sectors (cottage enterprises) which are not covered by other wholesale landing programs such as IRP, JEXIM, IGLF, and other -tail lending program such as DTI'S Micro Financing Program for micro enterprises. The project was later transferred to DSP as the CEFP loan borrower. Being the original proponentp DTI was made a member of the CEFP Review Committee (CEFP-RC), the policy-making body for the project, with its Undersecretary as Chairman of CEFP-RC. The MGA Management Unit (MGAMU), manned by DSP and DTI officers, was created to cversee day-to-day operations of the MGAs. The MGAMU reports to the CEFP-RC. As IBRD was aggressively marketing the CEFP, it requested the DSP In late 1990 to already implemont the project prior to the project approval by NEDA and the loan approvals by IBRD and KfW. ProJect Imolementatign IBRD very closely supervised the CEFP. Ms. Margaret Hanson Costan was on top of the project and regularly came over to the Philippine- to discuss and consult with the officers of DSP, DTI, .PFIs, MGAs, and other institutions involved in the CEFP as well an with the CEFP consultants on the progress and problems confronting the project. The Bank was also supportive in approving changes in the project to simplify some policies and procedures of the project.  - 12 - CEFP Project Completion Report, Part II ----_---------------------------------- ZBRD, however, did not agree to the request of PFIc to grant subloans to the MGA, instead of the individual members, to simplily PfIrm administration of the loan. This is understandable since the projects objective is to establish the banking relationship and credit worthiness of individual members with the PFI which will not be achieved by lending to the MGA. On the other hand, IBRD's emphasis was always on the slow disbursement of the CEFP loans. IBRD was also in constant communication with the D8P, DTI, and KfW to follow up on the project implementation and was quick to react and give its views o'n some issues confronting the project. S. DEVELOPMENT BANK OF THE PHILIPPINES tD1Pl Prolect SvolutnZL DTI was initially the implementing agency of the CEFP. Later on, IBRD saw it fit for DBP to be the CEFP loan borrower. DBP agreed to take on the project in view of its developmental impact ano support for the small, informal, and marginalized entrepreneurs/enterprises. DDP's wholesale lending programs were made complete by CEFP as these now covered the whole spectrum of borrowers from cottage to large enterpricos. To show its support for CEFP, the DBP Board approved an allocation of P100 MM for rediscounting loans to PFI1 and back-up for the matching loans to ba granted to the. MGAs while there was no accrual yet o-f the interest differential from the KfW lo#ns. Although the loan agreements with IBRD and KfW were respectively signed in April, 1991 and Septemter, 1991, the loan agreements were declared effective only in July, 1992 and the first disbursements were made in August, 1992. In order to sustain the momentum of the project as some MSAs were already formed and have their PFIs, DUP started granting the matching loans to MGAs and the rediscounting loans to PFIs in September, 1991 using DBP's own funds. Before the first IBRD/KfW loan disbursements were made, total rediscounting loans of P24.9 MM were grantod to 7 PFIs for 12 MGAs and total matching loans of P13.36 MM to 14 MGAs. As of 31 March 1993, total matching loan. granted to 24 MGAs amounted to 019.4Z MM although the back-up IDF is only P1.54 mm. Of thi- accrued IDF, an amount of P1.47 MM was paid to PFIs due to defaults of some MSA members lleaving a balance of only P70,000.00 in the IDF).  - 13 - CEFP Project Completion Report, Part II --------------------------------------- Project Imolementation In order to oversee the CEFP implemehtation, DBP assigned three (3) officers to the MGAMU on a full time basis for the following functions: 1. Regroup the MGA% formed by Sarson consultants as the core group beneficiaries for the initial implemontation of the CEFP. 2. Conduct orientation briefings to the MGAs concerning the pollcies, mechanics and procedures uf CEFP. 3. Assist the MGAs in accomplishing documents for registration with the Securities and Exchange Commission (SEC), looking for and matching them with prospective PFIs, licensing by the CEFP-RC as a MGA, and granting the matching loan fund (MLF). 4. Promote the proJect to as many PFIs as possible. 5. Prepare the rediscounting procedures and the operating procedural guidelines for DBP's diflerent units involved in CEFP. 6. Managa/monitor the consultancy services with Africa Asien Bureau (AAS). 7. Handle tho rediscounting loan and matching loan applications of PFIc and MGAs, respectively. e. Assist the MGAs in resolving problems within the association or with their PFIs. 9. Act as secretariat of the CEFP-RC. 10. Coordinate the activities of the WB-KfW supervision missions, arranging for meetings and random subproject visits undertaken by the mission representatives. C. DEPARTMENT OF TRADE AND INDUSTRY CDTI) Prolect ImoleMentati2n DTI was very supportive of the CEFP. It assigned 2 full- time staff in its head office and I staff each in the regional and provincial offices to the CEFP. Their valuable assistance are in the following" l. Identification of groups of entrepreneurs for the formation into MGA%l  CEFP Project Completion Report, Part Ir --------------------------------------- 2. Formation of the MGAc and assistance in accomplishing documentation for the MGAs; 3. Coordination of meetings with the cottage entrepreneurs/ enterprises during promotion of the project; and 4. Follow up supervision of operating MGAs. The DTI Undersecretary as Chairman of the CEFP-RC was also very active in giving advice% and finding ways to improve the project and solicit PFIs participation. D. CONSULTANTS GARSON & PARTNERS As mentioned above, IBRD and DTI hired the services of Garson & Partners for assistance in the project design. The consulting services of Garson were found to be wanting in many aspects. The Garson consultants were not able to anticipate potential problems which might arise during the project implementation, for which they could have recommended measures to reduce or eliminate said problems. They did not even review the initial project concept and prepared a critique on it. The initial promotion of the CEFP by the Garson Consultants was focused on the amount of loan the MGA members can get compared to their initial contribution - -- P90,000 loan for P10,000 contribution. Evan those who have no businesses yet during, the MGA formation were accepted although the CEFP required that members should have businesses existing already for at least I year. The consultants were not very conscious of the project's eligibility criteria and had no selection criteria in choosing the MGA members. AFRICA ASIEN BUREAU (AAB) AADBs Project Manager for CEFP, Mr. Jan D. Bol, came to the Philippines on August 10, 1992. It took time for him to familiarize himself with the-CEFP concept and the Filipino culture in order to work with the MGAs. Mr. Sol introduced some selection criteria in the formation of succeeding MGAs. The local consultants assisted in the development of the CEFP system for information monitoring and the training courses for the MGAS. They also assisted the MGAMU in training the MGA offieors/members and supervision/monitoring of the MGAs.  CEFP ProJect Completion Report, Part II Although AAS consultants had done their best in improving the CEFP implementation, the project was still hampered by PFIs participation and MSA formation. III. PROBLEMS ENCOUNTERED Two (2) major problems were identifieds a) PFI participation and b) MGA formation. A. PFI PARTICIPATION Many banks were unwilling to participate in the CEFP for the following reasons: 1. Subloan sizes of P90,000 to 0270,000 were considered too retail, especially by the commercial banks. 2. High administrative costs for individual lending considering the subloan size.. 3. Lack of personnel to handle many but small accounts which involve considerable man-hours. 4. Provincial branches of many PFIs are only deposit takers and have no lending functions; thus, some MGAs can not find PFIu to service them. Out of 39 MGAc formed and registered with SEC, only 20 became operational due to this problem. 5. Many smaller banks (thrift and rural banks) which are expected to be the natural conouits of CEFP failed to hurdle the DBP's pro-qualifying requirements to become. PFIs because of their high past due levels (over 1/% of total portfolio) and/or inadequate paid-up capital (below P5 0 to P10.0 MM for rural and thrift banks, respectively). 6. PFIs are not comfortablo with the concept of credit gearing ratio, i.e., granting subloans at certain multiples of the MGA's Guarantee Fund (GF) which-was initially set at 3x. While the MSA is allowed to guarantee up to a maximum 80Y of the members' subloans, the GF in affect covers only 33% of the subloans while 47'. is clean at the maximum gearing ratio of 3x. 7. News of default from some MGAs discourage other PFIs in participating or continuing their participation with the CEFP.  - 16 - CEFP Project Completion Report, Part II --------------------------------------- B. USA FORMATION / MEMBERSHIP Problems inherent in the formation/organization of the MGAs are as followst 1. No selection criteria. The formation of MGAs was focused on quantity instead of quality since the consultants were required to establish certain numbers of MGAs with certain numb-r of entrepreneurs per Terms of Reference (TOR). i 2. Lack of cohesiveness. Members of the MSA came from a wide area, i.e., from different municipalities in a province. Ideally, members should come from tho same community and know each other so that peor pressure will work. 3. Long gestation from promotion to operationalization of the MGAs. It took as long as one (1) year with a lot of documentation to accomplish before a MGA can be operated. 4. Formation of MGAs in areas .where PFI participation is doubtful or obviously absent. 5. The project was improperly promoted, i.e., emphasizing on the amount of subloan a member can obtain (9x his initial contribution). Initially, this resulted in several defaults (members were looking at short-term gains). 6. Albeit some members have valid reasons for their defaults, many member-subborrowors are of the usual thinking that government-assisted projects such as CEFP are dole-outs and need not be repaid. This thinking was aggravated by a wrong promotion: "a P90.0 M loan can be obtained for a PlO.OM investment" which was emphasized and repeated too many times. 7. MGAs were formed simply to get loans and provide guarantee instead of emphasizing long-term band its to the association, other than getting loans, such as assisting each other in materials procurement and product marketing.  CEFP Project Completion Report, Part II IV. LESSONS LEARNED 1. The project design glossed over the PFI's conservatism, especially the commercial banks which wQer the initial targeted PFIs. Their reluctance especially in accepting the credit gearing ratio was not foreseen. This aspect should have. thoroughly been discussed with them before the project was implemented, i.e., "pre-selling" the idea first. 2. Smaller banks which are the natural conduits of this kind of financing were also overlooked in the study. 3. The long gestation in the MGA formation and operationalization should also have been considered. In order lor the MGA to be successful, they should have gone lirst to a lengthy institution building to make them cohesive and institutionalize controls and poer pressure before the subloans were given. This could not be done in a very short period of time. 4. Emphasis in MGA formation should be more on quality and not quantity to establish easier management and quality accounts. 5. The creation of other guarantee corporations such as the Small Business Guarantee and Finance Corporation (SBGFC) which overtook the MSA guarantee scheme, was not anticipated. V. IBRD - DSP RELATIONSHIP The relationship between IBRD and DBP was very cordial and frank. Everything was done in transparent manner. There was constant communication between the 2 financial institutions. Problems and changes/amendments sought were brought to the attention of the funders who promptly gave their comments/advices on such matters. The rapport established with the funders gave an enriching experience for the DSP staff. VI. SUGGESTIONS If we have to do another CEFP, we suggest the followings 1. The closing date of the loans should be longer and the commitment fees waived to remove pressure on the loan disbursements.  CEFP Project Completion Report, Part II 2. The project should be supported with more technical assistance (TA) in the MGA formation. In this connection, only local consultants may be considered to maximize utilization of the TA grant. Besides, tho local consultants are already familiar with the local conditions of the country. 3. Expand the coverage to eligible members of existing producer associations, NGOs, cooperatives, etc. and increase the maximum subloans for the project to be more attractive to PFIs. 4. Institutional strengthening of the M6As should be given more time before the members can avail of tho CEFP credit facilities.  - 19 - PHILIPPINES COTTAGE ENTERPRISE FINANCE PROJECT (LOAN 3312-PH) PROJECT COMPLETION REPORT PART III. STATISTICAL INFORMATION Table 1. Related Bank Loans Year of Project Description Approval Status 0998 Industrial Investment Credit Project 1974 Closed 1120 Small & Medium Industries Dev. Project) 1975 Closed 1190 2nd Industrial Investment Credit Project 1975 Closed 1572 3rd Industrial Investment Credit Project 1978 Closed 1727 2nd Small & Medium Industries Dev. Project 1979 Closed 1903 Structural Adjustment Loan 1980 Closed 1984 Industrial Finance Project 1981 Closed 2127 Textile Sector Restructuring Project 1982 Closed 2169 3rd Small & Medium Industries Dev. Project 1982 Closed 2266 2nd Structural Adjustment Loan 1983 Closed 2787 Economic Recovery Program Loan 1987 Closed 3038 4TH Small and Medium Industries 1989 Closed Development Project 3049 Financial Sector Adjustment Program 1989 Closed 3084 Manila Power Distribution Project 1989 Active 3123 Industrial Investment Credit Project 1989 Closed 3287 Industrial Restructuring Project 1991 Active 3523 Telephone System Expansion Project 1993 Active  - 20 - PHILIPPINES COTTAGE ENTERPRISE FINANCE PROJECT (LOAN NO. 3312-PH) PROJECT COMPLETION REPORT Table 2. Project Time Table Item Date Planned Date Actual First Presented to the Bank May 1989 Preparation March 1990 Appraisal Mission July 1990 Negotiations October 1990 December 1990 Board Approval December 1990 March 1991 Loan Signature April 1991 Effective Date January 1991 August 1991 Project completion August 1995 Loan Closing June 1996  - 21 - PHILIPPINES COTTAGE ENTERPRISE FINANCE PROJECT (LOAN NO. 3312-PH) PROJECT COMPLETION REPORT Table 3. Loan Disbursements (US$ million) Annual Basis Cumulative Basis Estimated Actual Estimated Actual 1991 1.5 1.5 - 1992 1.2 1.5 2.7 1.5 1993 3.6 6.3 1.5 1994 4.2 10.5 1.5 1995 2.9 - 13.4 1.5 1996 1.6 -15.0 1 1.5  - 22 - PHILIPPINES COTTAGE ENTERPRISE FINANCE PROJECT (LOAN NO. 3312-PH) PROJECT COMPLETION REPORT Table 4. Use of Bank Resources A. Staff Inputs (Staff Weeks) TASK FY86 FY87 FY88 FY89 FY90 FY91 FY92 FY93 FY94 TOTAL Project Pre- Appraisal 24.5 24.5 Project Appraisal 26.7 26.7 Negotiations 10.3 10.3 Supervision .3 3.0 3.8 8.4 2.0 17.5 TOTAL .3 24.5 39.9 3.8 8.4 2.0 78.9 B. Missions Stage of Month/ Number Days in Staff Performance |_Pro.lect_Cvcle Year of Persons Field Days Ratinv Through Appraisal 03/87 1 2 2 06/89 3 1 3 10/89 2 2 4 03/90 2 12 24 Appraisal through Board |Approval 07/90 3 14 42 Supervision 10/91 l 3 3 2 09/92 1 5 5 2 05/93 1 8 8 2 J ~~~~~ ~ ~ ~ ~~08/93 1 1 1  - 23 - PHILIPPINES COTTAGE ENTERPRISE FINANCE PROJECT LOAN 3312-PH Table 5. Status of Legal Covenants Agreement Description of Covenants Status and Proposed Section Action WITH DBP Loan Agreement Section Adherence to the Operating Policy Guidelines, Compliance. 1.02Cc) and Schedule 4, which incorporate agreements reached with the para. 3 Bank regarding project structure, legal arrangements and guidelines for project operation and management. Loan Agreement Section Surplus funds to be used as agreed, for Compliance. 3.04 further lending to eligible subprojects. Loan Agreement Section Be audited yearly by independent auditors, Compliance. 4.02 including audit of Special Account and Statement of Expenditure, and submit audit to Bank within six months of year end. Loan Agreement Section Require PFIs and cottage enterprises to Compliance. Schedule 4, paras. 1 and maintain operating and financial records. 2 Loan Agreement Section Effectivity of the KfW loan agreement with Compliance. 5.01 DBP. Loan Agreement Section Agreement on a mechanism for offsetting the Compliance. Agreement 4.03 foreign exchange risk assumed by Government. on the FX premium to be paid to Government was reached in mid-1991 and approved by the Bank. WITH GOVERNMENT Guarantee Agreement Adherence to the Policy Manual, which Compliance. Section 3.02 incorporates agreements reached with the Bank concerning the operations of the Review Committee and the MGA Management Unit. Guarantee Agreement Agreement on a mechanism for payment from DBP Compliance. Agreement Section 3.01 to offset the foreign exchange risk assumed was reached in mid-1991 by Government. and approved by the Bank.    IMAGING Report No: 15172 Type: PCR