Philippines - Second Small and Medium Industries Development Project

Document o The World Bank FOR OFFICIAL USE ONLY Report No. 7940 PROJECT COMPLETION REPORT PHILIPPINES SECOND SMALL AND MFDIUM INDUSTRIES DEVELOPMFNT PROJECT (LOAN 1727-PH) JUNE 30, 1989 Asia Region Country Department II Industry and Energy Operations Division This document has a restrieted distribution and may be used by recipients only in the perfornance of their official duties. LIs contents may not otherwise be disclosed without World Bank authorization.  PHILIPPINES SECOND SMALL AND MEDIUM INDUSTRIES DEVELOPMENT PROJECT (LOA" 1727-PH) PROJECT COMPLETION REPOR: AB'REVIATIONS ADB - Asian Development Sank BOI - Board of Investment CB - Central Bank of the Philippines CEO - Chief Executive Officer (of IGLF) CSMI - Commission on Small and Medium Industries DBP - Development Bank of the Philippines DCP - Design Center of the Philippines DLC - Department of Loans and Credit (of CB) DOF - Department of Finance DTI - Department of Trade and Industry IGLF - Industrial Guarantee and Loan Fund KKK - Kapisanang Kabuhayan at Kaunlaran MASICAP - Medium and Small Industries Coordinated Action Program NACIDA - National Cottage Industries Development Authority NBFI - Non-Bank Financial Intermediaries OIA - Office of the Internal Audit (of CB) PDB - Private Developmenz Banks PDCP - Private Development Corporation of the Philippines PISCO - Philippine Investments Systems Organization PNB - Philippine National Bank RC - Review Committee (of IGLF) SBAC - Small Business Advisory Centers SMI - Small and Medium Industries STD - Special Time Deposits UPISSI - University of the Philippines Institute for Small-Scale Industries USAID - United States Agency for International Development COUNTRY EXCHANGE RATES Name of Currency (abbreviation) Peso Year: Appraisal Year Average 1978 Exchange Rate: US21 = 7.37 Intervening Yearo Average 1979-82 US31 = 8.00 Completion year 1983 USS1 = 11.JO  KO OFF'ICAL US ONLY Tl WORLD BANK Washanton. DC 2041 USA Opkw di DtmiNw June 30, 1989 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Completion Report on Phi.ippines Second Small and Medium T-iustries Development Project (Loan 1727-PH) _ Attached, for information, is a copy of a report entitled "Project Completion Report on Philippines Second Small and Medium Industries Development Project (Loan 1727-PH)" prepared by the Asia Regional Office. No audit of this project has been made by the Operations Evaluation Department at this time. Attachment 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.  i-4 CILt. ,ONY PHILIPPINES SECOND SMALL AhD MEDIUM INDUSTRIES DEVELOPMENT PROJECT (LOAN 1727-PH) PROJECT COMPLETION REPORT Table of Contents Page No. PREFACE. i BASIC DATA SHEETS .............................................ii HIGHLIGHTS ...................... ....................... iv I. INTRODUCTION ............................................ 1 II. THE SMI SECTOR. 3 III. PROJECT PREPARATION AND APPRAISAL. 3 Project Objectives. 5 Project Description. 5 Covenants. 5 IV. PROJECT IMPLEMENTATION. 6 Loan Effectiveness and Start-up. 6 Loan Allocation. 7 Disburse ments. 7 Subproject Characteristics and Performance. 7 Sectoral Distribution of Subprojects. 8 Regional Distribution of Subprojects. 8 Subproject Firm Size . Operational Results of Sample Projects. 9 Economic Performance of Sample Subprojects .10 Status of Subloans .10 Institutional Sponsorship .10 Reporting .11 Procurement .11 V. OPERATING PERFORMANCF .11 Overall Operations .11 IGLF Operations .12 Regional Development Under Loan 1727-PH .13 Interest Rates and Foreign Exchange Risk .14 Diversification of IGLF Portfolio .15 Experience with Cottage Industry .15 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.  Table of Contents (cont'd.) Page No. VI. FINANCIAL P2UFORMANCZ OF IGLC ........................... 15 Financial Position . ..................................... 15 Capital Structure and Liquidity ........................ . 15 Profitability ......................... 16 Guarantee Schen ............ ....................................... 16 Audit ................................................... 16 VII. INSTITUTIONAL PERFORMANCE AND DEVELOPMENT ....... ........ 17 Institutional Aspects . . . 17 Management ... 17 Organization ... 17 Training ................................................ 17 Procedures and Standards . . . 18 Project Appraisal . . . 18 Project Supervision . . . 18 VIII. BANK PERFORMANCE ........................................ 19 Project Justification and Overall Performance ........... 19 Supervision ............................................. 19 Working Relationship .................................... 19 IX. SUMMARY AND CONCLUSIONS .. 20 ANNEXES 1. Structure of Interest Rates - 1974-84 ................... 22 2. Schedule of Estimated end Actual Disbursements .......... 24 3. List if Subprojects Financed ............................ 25 4/4-A Analysis of Subprojects Financed ........................ 34 5. Characteristics of Subprojects Financed ................. 39 6. Economic Indicators of Subprojects Financed ............. 42 7. Financial Indicators of Subpro3ects Financed ............ 45 8. Reasons for Delay in Completion of Subprojects Overruns in Excess of 15? of Estimated Project Cost ... 48 9. Organization Chart of the IGLF Unit ..................... 50 10. Summary of Operations, 1978-84 .......................... 51 11. Summary of Operations by Industry Size, 1978-84 ......... 52 12. Characteristics of STD Approvals by Sector, Region, Size, Purpose, and Maturity, 1978-84 .................. 59 13. Summarized Balance Sheets, December 31, 1978-84 ......... 65 14. Summarized Income Statements, 1978-84 ................... 66 15/15-A Projected and Actual Income Statements, 1978-84 ......... 67 16. Collection Performance by Financing Institutions Under the IGLF Scheme ................ ............. 69 17. Status of Subprojects as of December 31, 1984 ........... 70 17-A Status of Subprojects Affec_.ed bv Arrears as of December 31, 1984 ................ 75  Table of Contents (cont'd.) Page No. 18. Comparativ- Statement of Arrears Position btween Participating Institutions and End-Users of IGLF, December 31. 1978-84 . .. .. ...... 77 19. List of Accredited Financial Institutions . . 82 20. List of IGLF TrainLng Programe and Seuinnrs ............. 83 21. IGLF Review Camittee as of December 31, 1984 . . 84 22. Fschange Rate Movements, 1975-84 . . 85 23. Borrower Comients .................... 86  PHILIPPINES SECOND SMALL AND MEDIUM INDUSTRIES DEVELOPMENT PROJECT (LOAN 1727-PH) PROJECT COMPLETION REPORT Preface This report covers the Second Small and Medium Industries Development Project in the Philippines bupported by Loan 1727-PH. The loan of $25.0 mil- lion to the Government of the Philippines was approved on June 12, 1979, became effective on November 9, 1979 and was closed on June 30, 1983 aq originally planned. $84,465.33 was cancelled effective July 8, 1983, leaving the total amount disbursed $24,915,534.67. This report was prepared by the Asia Region Country Department II Industry and Energy Division, based on a project completion report by IGLF and other information in files at the Bank. This PCR was read by the Operations Evaluation Department (OED). The draft PCR was sent to the Borrower for comments and they are attached to the report as Annex 23.  - ii - PROJECT COMPLElTION REPORT BASIC DATA SHET. PHILIPPINES: SECOND SMALL AND MEDIUM IMDUSTRISE DEVELOPMENT PROJECT (LOAN 1727-PM) KEY PROJECT DATA Appriasl As.wtu'. or Item *t ei X^ curr6at *bp_t_ t_ Loan/Credit Amount (USS mlllion) 25.00 - Disbu sed ) 2401 Cance ll d ) 0.0 Re aId to ) 7/31/93 5.68 Outstanding to ) 7/J1/99 19.09 Proportion of Time Underrun or Overrun (N) - Financial Perforoance (SCLF) Satiofactory Satisfactory Institutional Performnce (ILf) Satisfactory Satisfactory PROJECT DATES Or;ginal Actual or Item Plan Revision* Est. Actual First Mention in Filos or Ti mtable 04/14/76 Negotiations 04/28/79 Board Approval 06/12/79 Loan/Credit Agroement Date 06/27/79 Elfectivenoos Date 1109/79 Closing Date 6/30/63 06/30/33 STAFF SNPUT FY75 FY77 FY73 FY79 FY3O FY61 FY32 FY6S FY64 FYOB FY37 FYO3 TOTAL Proappraisel 1.0 0.2 2.4 3.6 12.4 Appraisal 27.0 27.0 Negotiation 6.4 6.4 Supervision 0.2 3.2 22.7 2.1 0.8 6.5 1.5 0.8 37.2 Other 0.2 0.2 Total 1.0 0.2 2.4 42.5 3.2 22.7 2.1 0.3 0.2 6.5 1.5 0.6 63.2  - Lii - No. of No. of Dot. of Item Monti Y Ver Peen P-r ono Rno ort Idetification eG/7G 1 1 1 07/21/76 Preeppralol 06/n 4 12 06/21/73 Appr.sa" 12/78 3 & 3 06/13/79 Superiselo I 0/U0 2 3 6 10/0*/90 Suprviul II 11/91 4 4 16 04/18/02 Tot l 18 44 VN PO CT DATA Borroe-r Gov.rnmemt of the Phi lippines Executing '9ency Industrial Guaranteo and Loan Fund, D*pem t of Loans and Credit, Contral Bank Fiscal Yoar of Borror January 1 - Decembr 31 Follow-on Project Nao Third Small and MIdium Industries Dvoxlo_nt Project Loan/Credit Numbr 2169-PH Amount (USS million) 3112 million (inclusivo of a front end fee of 32 million) Looan/Credit Agreemmnt Date Juo 80, 1922 DATA Oi SWPOJECTS Appraisal *etiate. Actual. Increnental employment 11,500 10,497 Inv,.t"nt cost per job P 4, 17 P s OOO Incremental saleo (million) P271.9 Incromental profit (million) P 10.4 Incre_mntal export. (mlillo) P 70. Incremntel *xporte as X of Incrementel sales 19i Incremental vale added (illilon) P 142.5 Arrears from subborrwers to Mi (under both aceroditetion and spo _ership schemes) as a per oet of IGIF-funded loan ojittandikn (1964) 4.0 Arroars from subborr.,era to PFIs as a percent of IGLF-funded loane outstanding (aceredltetlon echoem only (1964) s.5X Arr re e rm PWI* to ILF an a pereetb of IOLF loam aetet"eein (low) 3  PHILIPPINES SECOND SMALL AND MEDIUM INDUSTRIES DEVELOPMENT PROJECT (LOAN 1727-PH) PROJECT COMPLETION REPORT Hiahlidhts InstiLutional Framework i. The Small and Medium Industry (SMI) Sector in the Philippines is an important contributor to the manufacturing sector, accounting for 212 of value-added and 432 of employment in 1983. SHI firms are important in a wide range of manufacturing subsectors contributing over 302 of value added in food processing, machinery and equipment, chemicals, rubber, printing, plastic produc.ts, furniture and wood products, and wearing apparel. Within these subsectors, about half of SMI activity is concentrated in two areas, food processing and chemicals. ii. During the implementation period of Loan 1727-Ph, the Government operated a number of specialized loan programs channeled through financial institutions to hel.p SMI firms with entrepreneurial and management skills but short of capital. Pre-eminent among these programs was the Central Bank's Industrial Guarantee and Loan Fund (IGLF) to which the bulk of Loan 1727-PH funds were extended. Other programs directing credit to SKI firms during the project period included the Development Bank of the Philippines' Small and Medium Industries Lend3ng (SMILE) Program, the agri-business loan programs of both the Central Bank (CB) and Development Bank of the Philippines (DBP), the Philippine National Bank's Small and Medium Scale Industries Financing Scheme and the Kapisanang Kabuhayan at Kaunlaran (KKK). Most of the programis (other than IGLF) have in the past few years been discontinued or sharply curtailed. iii. The Industrial Guarantee and Loan Fund (IGLF), created in 1952, is a rediscount and guarantee mechanism operated by the Central Bank (CB) under which term loans (fixed asset and permanent working capital) made to eligible SMI firms by accredited financial institutions (commercial banks, rural and thrift banks, private development banks and nonbank financial institutions) are rediscounted. IGLF also offers guarantees to cover credit and collateral short risks taken by the participating financial institutions (PFIs) on IGLF rediscounted loans. IGLF has sourced its funds primarily from funds transferred from the Government budget and from World Bank borrowings. iv. The Second Small and Medium Industries Development Project is a follow-on project from ShI I (Loan 1120-PH). Under that loan, the Bank for the first time lent for funding the IGLF and, in the context of the loan, made major changes to an on-going program. Specifically, under SMI I, IGLF implemented the accreditation prog& m under which financial institutions were approved for participation and eligible for IGLF rediscounting of loans to SMI firms. The reformulated IGLF program expanded under both SMI I and II with the infusion of IBRD loans and Go%ernment counterpart funds.  Proiect Obiectives and Goals v. An important objective of the project was to ensure that long-term resources were available for the long-term financing needs of both the small and medium scale industry sectors. In this regard IGLF was to be a vehicle for providing _mloyuent opportunities. improving income distribution and regional developmes.t, and a catalyst for promoting general economic growth and exports. It was also Intended to expand and upgrade the network of financial institutions through vhich IGLF's credit was provided, and in particular to promote expertise in project evaluation and preparation. Project Description vi. The loan consisted of two parts: (a) the financing of the Industrial Guarantee and Loan Fund (IGLF) for on-lending through accredited institutions to SMI for fixed assets and permanent working capital; and (b) the strengthening of the Department of Industry's Medium and Small Industries Coordinated Action Program (MASICAP) in providing assistance to entrepreneurs in project preparation. All but 22 of the loan proceeds were allocated for the IGLF financing. Overall Operations vii. T?e objec ives of the project were largely fulfilled. Loan 1727-PH was disbursed qui(,ly, with the IGLF component of the loan ($24.5 million of the total $25.0 million) being fully utilized within two years of loan effectiveness. The loan was used to finance 264 IGLF-funded projects with an average subloan size of P 1.0 million. The fast rate of disbursement was due to the rclatively long maturities offerea by IGLF, the active participation of an increajed number of accredited financing institutions, the inclusion of medium scale industry under the IGLF eligibility criteria, the low interest rates charged by IGLF and the increase in spreads allowed the PFIs (increased from 72 to 82 for small industry loans and from 52 to 6? for medium industry loans). Disbursements under the MASICAP component ($0.5 million) were slower than anticipated because of (i) the merger of the Small Business Advisory Centers (SBAC) and MASICAP programs; (ii) a Government-imposed freeze on hiring by Ministries; (iii) bureaucratic disbursement procedures between Ministries; and (iv) Government-issued procurement guidelines which restricted the purchase of vehicles and computers. $84,465.33 originally allocated to the MASICAP technical assistance program was cancelled effective July 8, 1983. IGLF Operations viii. The loan enabled the lGLF program, which already was operating well, to expand and improve its successful performance. The loan clearly fulfilled its primary objective of channeling long-term funds to the SMI sector. By 1984, IGLF had become such an important source of term finance for SMI firms in the Philippines that it represented 62 of all long-term loanb outstanding in the country. When compared to the SMI share (21?) of value added in manufacturing, it is evident that IGLF represented a substa,itial portion of long-term credit extended to SMI in the industrial sector. During the implementation period of Loan 1727-PH, the IGLF expanded to finance medium  - vi - scale firms. Loans approved increased by almost 50Z a year (242 in real term.), from 1 47 million in 1978 to P 435 million in 1984, with projects financed by those loans generating about 10,497 new jobs. iX. The project also realized its goal of expandAng and upgrading the network of financial institutions through which IGLV cLedit was provided, and specifically of helping the institutions develop expertise in project *valuation and preparation. By December 31, 1981, 32 financial institutions were accredited under IGLF, as compared to 20 accredited institutions at the time of the 1978 project appraisal. Training pregrams organized by IGLF for PFI staff helped to upgrade appraisal and supervision capabilities of the financial institutions. The resulting improvements in portfolio performance during the project period can be seen in the collection ratios between participating financing institutions and end-lasers, which improved steadily and substantially between 1978 and 1984. Whereas in 1978 collections had amounted to only about 171 of principal and interest falling due and past due, they had improved to 542 by 1984. The proportion of PFI portfolios affected by arrears from end users decreased substantially from 42Z by amount and 51Z by number in 1978 to 122 by amount and 222 by number in 1984. At year end 1984, arrears amounted to only 4.52 of all IGLE' loans outstanding, and were even lower (3.5Z) for institutions participating in the program under the accreditation scheme. In addition, collection performance between participating institutions and IGLF improved substantially. Loan arrears to IGLF in 1978 represented 162 of the loan portfolio while in 1984 arrears accounted for only 2.12 of the portfolio. The introduction of the accreditation scheme (under SMI I), the increase in average subloan size with the new eligibility of medium-scale industry (under SMI II) and the increased attention to training, all contributed to the better results. X. The one explicit project objective that was not fulfilled under the loan was channeling of a majority of project funds outside Metro-Manila. The proportion of project funds lent to the regions was significantly lower than expected (392 vs 60Z), with Metro-Manila absorbing a greater than anticinated share of the total IGLF credit. The failure to achieve the desired regional disbursement is due largely to the fact that most SMI firms are located near Manila, which is the country's major port and largest consumer market ind has easy access to communication and transportation. The proportion of subloans that went to the regions was similar to the 1983 share of value added in the SMI industrial sector originating in the regions (42?). xi. Organization. IGLF was reorganized during the project period. In 1979, there were only two divisions involved in IGLF operations--the Induc- trial Loans Division and the IGLF Accouiting, Securities Control and Collec- tion Division--and the total personnel complement of the organization was 48 people. With the approval of 18 new positions for IGLF by the Monetary Board of the Central Bank, the IGLF organizational structure was reorganized into four divisions, as follows: (a) the Industrial Projects Division, responsible for processing IGLF financing applications, guarantees, and requests for deferment or restructuring; (b) the IGLF Examination Division, responsible for evaluating potential institutions for accreditation, for reviewing already accredited institutions, and for performing end-use verification surveys of projects assisted; (c) the IGLF Accounting, Securities Control and Collection Division, which handles the accounting, securities control and collection  - vii - functions of IGLF; and (d) the IGLF Special Studies Staff, which handles research and technical information activities and various reporting requirements. xii. Trainins. Because of the clear importance within both IGLF and the PFIs of analyzing credit risk and assessing projects and institutions, IGLF initiated a number of programs under Loan 1727-PH to further develop the experience of its staff and that of the PFIs. In addition, the staff of IGLF and the PFIs participated in relevant local training programs and traveled abroad to acquire additional knowledge on small enterpr4se financing and related fields. Lessons to be Learned xiii. Several lessons which may be learned from the Bank's and IGLF's experiences under Loan 1727-PH are outlined below: (a) There are inherent difficulties in pricing subloans to fully reflect all costs, including risk of potential foreiga exchange losses, in economies with regulated interest rates. At the outset of SMI II, IGLF funds to the PFIs were priced at an average rate of 8.6Z, which was 0.7Z higher than the 7.92 interest rate charged by the Bank on Loan 1727-PH. While this spread enabled World Bank financed projects to contribute to IGLF's administrative costs, it did not provide a premium to cover the foreign exchange risk on the World Bank loan. When the peso was devalued after disbursement of the loan (declining from an average disbursement rate of P 7.7 to the dollar, to the mid- 1988 rate of P 21.0 to the dollar), the Government was left with a large foreign exchange loss towards which the accredited financial institutions and SMI borrowers had made no contribution. Experience with the loan indicates that loans to apex institutions for onlending in regulated financial sectors should be tied to policy reform and deregulation of the financial sector. Once financial markets are deregulated and interest rates reflect expected inflation and foreign exchange adjustments, then apex program lerding rates (to both participating financial institutions and end-users) should be market- based. Such a rate should reflect the premium for foreign exchange as perceived by the market and should provide adequate cover to the Government. A proposed 4th SMI World Bank loan to the Philippines takes these recommendations into account. (b) During the implementation period of Loan 1727-PH. IGLF did not limit the participation of individual financial institutions in order to ensure that the IGLF portfolio as a whol. was balanced or diversified. The weakened financial situation of some of the PFIs during 1980 to 1984 period suggests that this failure to diversify was unwise. In 1981, after the bankruptcy of one PFI holding a large portion of the IGLF portfolio, IGLF adopted a policy limiting the participation of individual PFIs and specifying that a PFI's  - vili - portfolio of IGLF rediscounted loans may not exceed that institution's total net worth. (c) The experience of IGLF under Loan 1727-PH suggests that the apex arrangements which are appropriate for lending to small and medium firms may be inappropriate for effectively reaching cottage firms. IGLF's record vith cottage industry during the project period showed poor repayment and suggested that making and supervising cottage industry loans was overly cumbersome and expensive for the financial institutions participating in the IGLF program. The data suggest that a different vehicle for financing cottage industry should be created and that IGLF should not be pushed to fill the financing gap to that sector.  PdILIPPINES INDUSTRIAL GUARANTEE AND LOAN FUND PROJECT COMPLETION REPORT - LOAN 1727-PH I. INTRODUCTION 1.1 This Project Completion Report (PCR) on Loan 1727-PH (Loan Agreement dated June 27, 1979) reflects the utilization of that loan for strengthening and expanding the IGLF lending facility to small and medium industries and for an SHI technical assistance component. The IGLF component amounted to $24.5 million of the total IBRD loan of $25.0 million, extended to the Republic of the Philippines. The remaining $0.5 million represented a technical assis- tance program to strengthen the Medium and Small Industries Coordinated Action Program (MASICAP), a Department of Industry program Intending to stimulate investment in small and medium industries in regions outside Metro-Manila. Loan 1727-PH was the Bank's second loan to IGLF and accordingly this is the second completion report on lending to that institution. To date, IGLF has received three loans from the World Bank totaling US$99.5 million; the third loan of US$63.0 million was made on June 30, 1982. Information on the three loans is given below. A fourth SMI project, which would include an IGLF component of approximately $60 million, is currently under consideration. Year of effect- Loan IGLP Term Interest iveness Amount Portion (Years) Rate (C) Loan 1120-PH 1975 30.0 12.0 11-1/2 8.5 Loan 1727-PH 1979 25.0/a 24.5 14-1/2 7.9 Loan 2169-PH 1982 132.0/b 63.0 14-1/2 11.6 /a $84,465.33 was cancelled effective July 8, 1983. /b $64.6 million of the loan was subsequently cancelled due to financial troubles of DBP and the unattractively high interest rate on the loan.  -2- 1.2 The basic data concerning Loan 1727-PH are presented in the introduc- tory basic data shorts. Specific data for the IGLF portion of the loan (98? of the total) are as follows: Loan amount US$24.5 million Maximum subloan size 1 2.5 million Free limit No required prior review; subloans over P500,000 were to be reviewed on a post-approval basis Start of implementation February 1, 1980 Date of full utilization of September 10, 1981 IGLF portion of the loan 1.3 Although the fourth SMI project, currently under consideration, is expected to include funding from the Asian Development Bank (ADB) and possibly the ASEAN Japan Development Fund (AJDF), IGLF previously depended solely on the World Bank in raising funds from foreign sources. The IBRD loans were made to the Government of the Philippines, which bore the full foreign exchange risk. Under Loans 1120-PH and 1727-PH, the World Bank reimbursed 75? of the total loans released and the remaining 25Z came from the counterpart funds contributed by the Philippine Government. For Loan 2169-PH, 60? was reimbursed by the World Bank and 40Z came from the Philippine Government. 1.4 The following data base and approach was used in drafting this PCR: (a) The time period 1978 (the appraisal year) through 1984 was chosen as the key period for assessing results of the loan. Although the loan clo ed in 1983, results of subloans made prior to 1983 were felt after the closing date and are reflected in IGLF's information for 1984. While the nature of the IGLF program has not changed substan- tially since the closing of Loan 1727-PH, some alterations were instituted under SMI III so that data after 1984 would be less relevant to an analysis of SMI II. This PCR focuses largely on the 1978-84 time period but draws implications for subsequent years. (b) For the implementation period of the loan, data on all IGLF subpro- jects were analyzed, even though some of the projects fmnded by IGLF during the period were not financed by the World Bank but rather through internal IGLF or Government counterpart funds. The informa- tion assessed and presented reveals the overall health and contribu- tion of the IGLF program. (c) In conRidering the allocation of loan proceeds, a full listing of the 264 sub-projects financed by the loan was made (see Annex 3) and analyzed. Releases to firms were evaluated against the objectives of the loan.  - 3 - (d) IGLF staff collected and analyzed detailed data on a representative sample of projects constituting about 291 of the total loans released. The diverse group of sample projects was chosen after consideration of industry and regional classification, size and purpose of the loans, and sponsoring financial institutions. II. THE SMI SECTOR 2.1 The Small and Medium Induszry (SMI) Sector in the Philippines is an important contributor to the manufacturing sector, accounting for 21? of value-added and 432 of employment in 1983. SMI firms are important in a wide range of manufacturing subsectors contributing over 30? of value added in food processing, machinery and equipment, chemicals, rubber, printing, plastic products, furniture and wood products, and wearing apparel. Within these subsectors, about half of SMI activity i3 concentrated in two areas, food processing and chemicals. 2.2 The Government included the promotion of SMI as one of its major industrial objectives in the 1978-1982 Five-Year Development Plan. During the implementation of Loan 1727-PH, various Government agencies, including the Department of Trade and Industry (DTI), the National Economic and Development Authority (NEDA), the Department of Finance, and the University of the P ilippines' Institute for Small-Scale Industries (UP-ISSI), coordinated their efrorts to help change the policy environment in the Philippines to promote SHI growth. They initiated efforts to abolish inappropriate cottage industry incentives, restructure BOI incentives, and increase the registration of SMI firms for incentives. Also, the Government sponsored a range of technical assistance and training programs for SMI firms. 2.3 In addition, the Government operated a number of specialized loan programs channeled through financial institutions to help individuals or firms with entrepreneurial and management skills but short of capital. Pre-eminent among these programs was the Central Bank's Industrial Guarantee and Loan Fund (IGLF) to which the bulk of Loan 1727-PH funds were extended. Other programs dira;cting credit to SHI firms during the project period included the Development Bank of the Philippines' Small and Medium Industries Lending (SMiILE) Program, the agri-business Loan programs of both the CB and DBP, the Philippine National Bank's Small and Medium Scale Industries Financing Scheme and the Kapisanang Kabuhayan at Kaunlaran (KKK). III. PROJECT PREPARATION AND APPRAISAL 3.1 The Second Small and Medium Industries Development Project is m follow-on project from SMI I (Loan 1120-PH). Under that loan, the Bank for the first time lent for funding the Industrial Guarantee and Loan Fund (IGLF) and, ir the context of the loan, made major changes to an on-going program. Specifically, under SMI I, IGLF implemented the accreditation program under which financial institutions were approved for participation and eligible for IGLF rediscounting of loans to SMI firms. The reformulated IGLF program expanded under SMI I with the infusion of IBRD loans and Government  -4- counterpart funds. Loan 1727-PH continued this financing and facilitated the rapid expansion of the program which occurred between 1979 and 1984. A Bank mission visited the Philippines in June 1976 to identify the SMI II project. The Bank pre-appraisal mission visited the Philippines in June 1978. The project was appraised in November/December 1978. Negotiations were held the week of April 23, 1979 and the loan was approved by the Bank Board on June 12, 1979. The Loan Agreement was signed on June 27, 1979 and became effective on November 9, 1979. 3.2 The Industrial Guarantee and Loan Fund (IGLF) was created in 1952 for the purpose of encouraging the establishment and expansion of economically sound SMIs that would contribute to the Philippines' economic development and to a more equitable distribution of income among the people. IGLF is a rediscount and guarantee mechanism operated by the Central Bank (CB) under which term loans (for fixed assets and permanent working capital) made by accredited financial institutions (commercial banks, rural and thrift banks, private development banks and nonbank financial institutions) are rediscounted. IGLF also offers guarantees to cover credit and collateral short risks taken by the participating financial institutions (PFIs) on IGLF rediscounted loans. IGLF has sourced its funds primarily from funds transferred from the Government budget and from World Bank borrowings. 3.3 IGLF is owned by the National Economic Development Authority (NEDA) and administered by the Department of Loans and Credit (DLC) of the Central Bank of the Philippines (CB). The fact that IGLF operates as an apex unit, with the basic objective of providing resources to financial institutions for onlending to SMI, enables the program to have a wide geographical reach, since many of the institutions have an extensive regional branch network. Decisions on policy issues affecting IGLF's operations are made by an interagency Review Committee (RC) which includes representatives from NEDA, CB, Departme-t of Ttade anc Industry (DTI), Department of Finance (DOF) and UP-ISSI. 3.4 Under the IGLF accreditation scheme implemented in 1976 under SMI I, IGLF approves participation in the program of specific financial institutions and then delegates all subloan appraisa' and supervision responsibilities to those accredited financial institutions. Before granting accreditation status, IGLF undertakes a thorough review of the institution to determine its overall financial viability and the adequacy of its staff to implement an SMI lending program. The performance of each accredited institution is reviewed annually by IGLF and the Bank is informed of the findings of these reviews. 3.5 Although only accredited institutions are eligible for Bank funding, IGLF also has a 3mall sponsorship program for rural banks, under which it becomes more actively involved in the credit evaluation of the end-user. While the participating financial institutions under the sponsorship scheme do take the credit risk, their inadequate analytical capabilities necessitate a detailed review of the subloans by IGLF staff prior to rediscounting. Since the rural banks would not otherwise qualify for the IGLF program, the sponsor- ship scheme enables IGLF to promote industrial growth outside metro-Manila and to contribute to development in rural areas. The Aponsorship scheme has not been financed by the World Bank under any of its projects, including SMI II.  - 5 - 3.6 The participating financial institutions (PFIs) are free to select their own projects for financing, with IGLF financing assured if the project meets broad eligibility criteria (appropriate asset size, acceptable financial position and credit-rating, etc.). Full repayment rests with the financial institution unless the loan also carries an IGLk guarantee. Accordingly, the risk to IGLF is limited to insolvency or inability to repay by the participat- ing financial institution. Default by the end-user is borne by the PFI and not by IG_F. Proiect Obiectives 3.7 IGLF operates with the basic objective of providing resources to participating financial institutions for relending to SMIs. In this regard an important objective of the project was to ensure that long-term resources were available for inv-stment financing needs of both the small and medium scale industry sectors. Under the project, IGLF was intended by the Government and the Bank to be a vehicle for developing the SMI sector and thereby providing employment opportunities, improving income distribution and regional development, and providing a catalyst for promoting general economic growth and exports. It was also intended to expand and upgrade the network of financial institutions through which IGLF's credit was provided, and to promote expertise in project evaluation and preparation. Project Deacription 3.8 The loan consisted of two parts: (a) the financing of the Industrial Guarantee and Loan Fund (IGLF) for on-lending through accredited institutions to SMI for fixed assets and permanent working capital; and (b) the strengthen- ing of the Department of Industry's Medium and Small Industries Coordinated Action Program (MAP:CAP) in providing assistance to entrepreneurs in project preparation. 98% of the loan proceeds were allocated for the IGLF financing. Covenants 3.9 In addition to the Bank's standard covenants, the Loan Agreement for 1727-PH required the Government (as Borrower) to adopt formally a new IGLF Policy Statement which included the following major operational policies, outlined in the Staff Appraisal Report: (a) inclusion of medium-scale industries in the IGLF program by raising the maximum loan size rrom the former limit of IF 500,000 to P 2.5 million; (b) increasing the effective interest rate to the end-user of IGLF from 13.2Z to 14.7-15Z; (c) increasing the interest spread allowed to the participating institu- tiorns from 5Z to 82 on small industry loans and from 5Z to 6Z on medium industry loans; (d) adopting a new guarantee scheme; and (e) introducing a penalty charge on defaults by end-users.  It included an agreement with the Central Bank to furnish the Bank with a periodic summary of loans made by IGLF along with copies of appraisals of subloans exceeding R 500,000, and arrears data on IGLF funded loans between financial institutions and end-users. In adeition, the Central Bank and/or the Government agreed to the following: (f) further delegation of authority to the CEO of IGLF regarding the handling of routine matters; (g) the reorganization and appropriate staffing of the IGLF unit; (h) the calculation of ERR and FRR in appraisal of subloans over 1 1.5 million; (i) the preparation of a revised policy manual for IGLF; S (j) an exemption of IGLF financed loans from the requirements set out in CB/Government circulars and regulations governing money market operations; (k) an on-going monitoring of the performance of accredited institutions by IGLF; (1) a program to review the IGLF's portfolio of participating financial institutions; (m) the adequacy of CB's fee for administering IGLF; (n) the reorganizatiin of MASICAP; (o) the submission to the Bank by BOI of quarterly reports on MASICAP; and (p) the use of at least 802 of the IGLF loan component for financing subprojects with: a cost per job created no greater than the Bank's urban poverty guidelines; or with fixed assets at the time of loan application not exceeding $250,000. These covenants were met during the implementation period of the loan. IV. PROJECT IMPLEMENTATION Loan Effectiveness and Start-Up 4.1 Conditions for loan effectiveness included approval by the Government of the draft IGLF policy statement which had been agreed at negotiations, along with approval of the Loan Agreement by the Monetary Board of the Philippines. These conditions were met by October 1979, which was later than expected because evidence that the IGLF Policy Statement had been adopted was not received by the Bank as quickly as originally anticipated. The loan was declared effective on November 9, 1979.  4.2 The project started very smoothly. IGLF's actual performance for 1980 exceeded its operational targets substartially with IGLF approving 170 projects amounting to 1 147.2 million during 1980. By year end 1981, the IGLF outstanding loan portfolio to the PFIs totalled 1 415.8 million, compared with the appraisal estimate of 1 361.1 milliot. Loan Allocation & 4.3 Under the loan, $0.5 million was earmarked for technical assistance to strengthen the Department of Industry's Medium and Small Industries Coordinated Action Program (MASICAP) in providing assistance to entrepreneurs in project preparation. The remainder ($24.5 million) was allocated to the Industrial Guarantee and Loan Fund (IGLF) for on-lending through accredited institutions to SMI for fixed assets and permanent working capital. IGLF used its full allocation, while MASICAP cancelled about $85,000 of its funds. No reallocation of the loan by categories was made. Disbursements 4.4 Annnex 2 shows the estimated and actual quarterly disbursements for the $24.5 million IGLF portion of the Loan. IGLF's utilization of the loan was satisfactory, with disbursements ahead of the estimated schedule at appraisal. The MASICAP component of the loan disbursed more slowly than expected at appraisal. Delays were caused by (i) a merger of the Small Business Advisory Centers (SBAC) and MASICAP programs; (ii) a Government- imposed freeze on hiring by Ministries; (iii) bureaucratic disbursement procedures between Ministries; and (iv) restrictive Government-issued procurement guidelines regarding the purchase of vehicles and computers. Subproiect Characteristics and Performance 4.5 The table below summarizes data for subprojects financed under Loan 1727-PH and compares the estimates at the time of appraisal with actual performance. Annexes 3 to 7 present detailed information on the characteristics of these subprojects.  Table 4.1: SUMMARIZED DATA ON SUBPROJECTS Appraisal Actuals based on estimates projects surveyed No. of subprojects 510 264 Average subloan size (million) P 0.5 /a P 1.0 Incremental employment 11,500 10,497 Investment cost per job P 34,167 P 38,000 /b Incremental sales (million) P 373.9 Incremental profit (loss) (million) P 10.4 Incremental exports (million) P 70.9 Incremental exports as Z of incremental sales l9Z Incremental value added (million) P 142.5 /a The appraisal estimate adjusted for actual domestic inflation would be P 0.8 million. /b This cost per job compared well with the overall cost per job of the Philippines' modern industrial sector, which was P 53,000. 4.6 Sectcral Distribution of Subpro4ects. The IGLF subloans made under Loan 1727-PH were extended to a wide-variety of subsectors within the Philippine manufacturing sector. Construction, tourism and service industries were also subborrowers of the IGLF facility. From 1978 to 1984, the manufac- turing sector accounted for over 952 of approvals of total subloans by amount, the more important subsectors being metal products (20Z), food and food products (18Z); footwear, apparel and garments (112), and plastic products (see Annex 12). The sectoral distribution was good and in line with the distribution of SMI assets and value-added within the industrial sector. 4.7 Regional Distribution of Subproiects. The regional distribution of subprojects financed under Loan 1727-PH is presented in Annex 4. The share of subprojects outside Metro Manila totaled 47Z by number and 391 by amount, as shown below. This did not meet the appraisal target that 602 of IGLF funds be channeled outside Metro-Manila.  -9- Table 4.2: REGIONAL DISTRIBUTION OF IGLF SUBLOANS No. of Amount projects Percent (1 million) Percent Metro Manila 140 53 153.9 61 Outside Metro Manila 124 47 99.3 39 Total 264 100 253.2 100 4.8 Subproiect Firm Size. Medium scale industries received the bulk of the subloan proceeds, with 84Z cf the funds and 612 of the projects going to medium scale firms. The details of the funding breakout are shown below. The distribution of IGLF subloans was in line with the relative shares of value- added by small and medium scale industry generally, with medium industry providing 811 of the value-added of those sectors. Table 4.3: FIRM SIZE DISTRIW'JTION FOR IGLF SUBLOANS No. of Amount Size of industry projects Percent (P '000) Percent Medium Scale 160 60.6 211,486 83.5 Small Scale 104 39.4 41,709 16. J Total 264 100.0 2S3,195 100.0 Due to the inclusion of medium industr, loans under the prcject, the average size of IGLF loans grew steadily from P 336,000 in 1978 to P 1.b million in 1984, and averaged $1.0 million for the 1978 to 1984 as a whole. 4.9 ORerational Results of Sample Proiects. Operational. data on a 292 sample survey of subprojects are provided in Annexes 5 to 7. Cost estimates made by financing institutions proved to be realistic with an average cost overrun of about 102. Cost overruns were mainly due to changes in project design, machinery and equipment, and to extended delays in project implementation. Large underruns in some projects occurred because the projects were scaled-down after subproject appraisal had been implemented. 4.10 About 50! of the subprojects surveyed faced delays in implementation ranging from one to fourteen months. The delays were caused by changes in supplier and machine specifications, changes in project design, contractors' delays in project construction, equity deficiencies, difficulties in securing sufficient financing, changes in project site, delays in the processing of documents on imported machinery, and late compliance with pro-disbursement  - 10 - requirements. Annex 8 provides information regarding the subprojects which suffered implementation delays of six months or longer and cost overruns in excess of 152 of estimated project cost. 4.11 Estimated and actual sales data -were avalable for the 76 subprojects surveyed. As shown in Annex 6, actual total sales for 1982 for those projects amounted to 732 of appraisal estimates. Actual profits were significantly less than estimated for the sample group as * whole, although profits earned by 17 projects exceeded estirates for 1982 operations. Significantly, how- ever, out of the 76 projects, only seven incurred losses of varying deg ees. To some extent, actual profits were understated because they were based on surveys done in 1982, which was only the first or second year of initial or expanded operation of subprojects financed under the loan. Furthermore, in light of the increase in oil prices, rising inflation and deteriorating busi- ness climate, the project period was a more difficult time to do business in the Philippines than had been expected at appraisal. 4.12 Economic Performance of Sample Subproiects. Annex 6 presents the details of the economic performance of the subprojects surveyed. Based on data available for 76 subprojects, it is calculated that 10,497 jobs were created under the project as a whole (compared to the appraisal estinate of 11,500). The average cost per job was P 38,000, compared to the estimated average of P 34,167. The increase in cost per job resulted in the financing of some capital intensive subprojects such as Baguio Summit Inn (construction of tourist inn), Aracelli Plastic Products (factory building), Sy's Bros. Printing Press (machinery/equipment) and Pagsanjan Aggregates (machinery/ equipment). The subprojects surveyed suggest that the total project ccntri- buted P 373.9 million in incremental sales. While export sales were projected to represent 29X of total sales, according to the survey data they actually amounted to only 192 because the projects were operating in a recessionary period during which export demand was not very strong. The overall performance of the subprojects was satisfactory, especially given the difficult economic euvironment in which the project was inplemented. 4.13 Status of Subloans. As of December 31, 1984, 34 out of 264 loans under Loan 1727-PH had been either repaid or prepaid in full. Arrearages of the PFIs to IGLF were 2.1? of all loans outstanding, while arrearages of end- users to participating institutions amounted to 4.5? of total loans outstand- ing. Of the 76 subprojects sampled, 20 end-users were affected by arrears totalling P 2.8 million or 17.2Z of total loans affected by arrears. 4.14 Institutional Sponsorship. Releases under Loan 1727-PH were coursed through the following financial institutions:  - 11 - Table 4.4: IGLF ACCREDITED FINANCIAL INSTITUTIONS No. of PFIs Projects Amount 2 to total Commercial banks 10 45 1 5e,01O 22.9 Nonbank financial inter- mediaries 7 202 181,856 71.8 Private development banks 2 8 3,950 1.6 Savings and mortgage banks 2 9 9,379 3.7 Total 21 264 1 253.195 100.0 Nonbank financial intermediaries led the other financial institutions in IGLF financing both by value and iiumber of projects, accounting for P 181.9 million or 71.8Z of total credits. The remaining 1 71.3 million or 28.2? were sponsored, in descending order, by commercial banks, savings and mortgage banks and private development banks. Reporting 4.35 IGLF kept the Bank informed of major developments of its organization and maaagement, operational and financial targets and achievements adequately and in a timely manner. Procurement 4.16 Of the 76 projects surveyed, 57 projects utilized the loan proceeds for the acquisition of machinery and equipment. The suppliers were chosen on the basis of several factors: suitability of equipment, availability of spare parts, efficiency of service, reliability of supplies, compatibility with present equipment, and price. Goo-ds for 34 subprojects were procured locally while 23 subprojects directly imported items under proprietary contracts. V. OPERATING PERFORMANCE Overall Operations 5.1 The objectives cf the project were largelv fulfilled. Loan 1727-PH was disbursed quickly, with the IGLF component of the loan ($24.5 million of the total $25.0 million) being fully utilized within two years of loan effec- tiveness. The loan was used to finance 264 IGLF-funded projects with an average subloan size of 1 1.0 million. The fast rate of disbursement was due to the relatively long maturities offered by IGLF, the active participation of an increased number of accredited financing institutions, the inclusion of medium scale industry under the IGLF eligibility criteria, the attractive interest rates charged by IGLF and the increase in spreads allowed to the PFIs (increased from 7? to 8? for small industry loans and from 5? to 6? for medium  - 12 - industry loans). Disbursements under the MASICAP component ($0.5 million) were slower than anticipated because of (i) the merger of the SBA- and MASICAP programs; (ii) a Government-imposed freeze on hiring by Ministries; (iii) bureaucratic disbursement orocedures between Ministries; and (iv) Government-issued procurement guidelines which restricted the purchase of vehicles and computers. Effective July 8, 1983, $84,465.33 originally allocated to the MASICAP technical assistance program was cancelled. Under the MASICAP component, resources were provided to hire permanent staff and upgrade the facilities of thb regional SBAC/MASICAP offices. IGLF Operations 5.2 The loan enabled the IGLF program, which already was operating well, to expand and improve its successful performance. During the implementation period of Loan 1727-PH, the IGLF expanded to finance medium scale firms. Loan 1727-PH clearly fulfilled its primary objective of channeling long-term funds to the SMI sector and the IGLF provided a major portion of total long- term credit borrowed by the Philippines' small and medium industry sectors. Loans approved increased by almost 50Z a year, from P 47 million in 1978 to P 485 million in 1984, with projects financed by those loans generating 10,497 new jobs. By 1984, IGLF had become such an important source of term finance for SMI firms in the Philippines that it represented 6Z of all long-term loans outstanding in the country. When compared to the SMI share (21Z) of value added in manufacturing, it is evident that IGLF represented a substantial portion of long-term credit extended to SMI in the industrial sector. 5.3 The project realized its goal of expanding and upgrading the network of financial institutions through which IGLF credit was provided, and specifi- cally of helping the institutions develop expertise in project evaluation and preparation. By December 31, 1981, 32 financial institutions were accredited under IGLF, of which 14 were commercial banks, 12 were nonbank financial intermediaries (NBFIs), 2 were savings/mortgage banks and 4 were private development banks (PDBs). At the time of the 1978 appraisal of the project, there had been 20 accredited institutions, of which 10 were commercial bankr. 9 were nonbark financial intermediaries and 1 was a savings/mortgage bank. Accordingly, the program increased in size 60? during the disbursement period in terms of the number of participating financial institutions accredited. Furthermore, training programs organized by IGLF for PFI staff helped to upgrade appraisal and supervision activities of the institutions. Although the percentage of total funds allocated to areas outside Metro-Manila did not equal the minimum 60Z projected at the time of appraisal, the 39Z of all funds which did go to firms outside the ce?ital compared well with the share of SMI value-added contributed by firms outside Metro-Manila (42Z). 5.4 That the goal not only of expanding but also of improving the IGLF Program was realized during the project implementation period can be seen in the collection performance between participating financing institutions and end-users, which improved steadily and substantially during the implementa- tion period of the loan (see Annex 16). In 1978, collections amounted to only about 17Z of principal and interest falling due and past due, improving to 372 in 1980, to 50Z in 1981 and 1982, and to 54Z in 1984. The proportion of PFI portfolio affected by arrears from end-users decreased substantially from 422  1 13 - by amcint and 512 by number in 1978 to 121 by amount and 221 by number in 1984. At year end 1984, arrears amounted to only 4.5! of all IGLF loans out- standing, and were even lower (3.5) for institutions participating in the program under the accreditation scheme. A comparative statement of arrears position and an analysis of loans in arrears between participating institu- tions and end-users of IGLF from 1978 to 1984 are shown in Annex 18. In addi- tion, collection performance between participating institutions ar.d IGLF improved substantially. Loan arrears to IGLF in 1978 represented 16! of the loan portfolio while in 1984, arrears accounted for only 2.1! of the 1984 loan portfolio. Several changes in IGLF policies brought about these improvements. First, a major change had occurred in 1976 when, under the first Bank loan for IGLF, the accreditation scheme was introduced. Previously, all subprojects had been apprais i and supervised by IGLF, which had neither the experience nor the branch network reqaired for proper review and control. Under the accreditation scheme, IGLF adopted a rigorous appraisal system to approve specific financial institutions for participation in the program; the PFIs were then given re3ponsibility for project approval and supervision. In addition, the arrears ratio improved under the loan because average subloan size increased substantially with the new eligibility of medium-scale inCustry, and the larger industries showed a better repayment rate. 5.5 Information collected on a sample of IGLF projects financed under- Loan 1727-PH shows significant contributions made to sales, profits, exports, import substitution and employment. This is consistent with the positive economic impact of IGLF throughout its history. The total number of incremen- tal jobs generated by IGLF projects financed under Loan 1727-PH is estimated at about 10,497. The IGLF projects funded under the Loan were expected at the time of appraisal to generate P 5,675 million in sales, of which about 11Z would come from exports. Although actual figures obtained for a sample of projects financed under Loan 1727-PH indicate that these estimates were opti- mistic, the net impact was still quite Gubstantial. In addition, the fac that the overwhelming majority of firms financed by the loan were profita. e and operational as of end-1984 shows the success of the program. A sample of subprojects reviewed revealed that only about 10! of operations funded we-e incurring losses at the time of evaluation in 1982. 5.6 Regional Development Under Loan 1727-PH. The one explicit project objective that was not fulfilled under the loan was channeling of a majority of project funds outside Metro-Manila. The proportion of project funds lent to the regions was significantly lower than expected (39Z vs 60Z), with Metro- Manila absorbing a greater than anticipated share of the total IGLF credit. The failure to achieve the required regions-to-capital ratio is attributable in part to lack of interest of the participating financing institutions, which considered solicitation, approval and supervision of projects in remote areas as impractical owing to the high cost of maintaining these accounts. But more importantly, it was a result of the fact that most firms prefer to be located near Manila, since it is the country's maior port and largest consumer market, with easy access to communication and transportation. Indeed, the proportion of subloans that went to the regions was similar to the share in 1983 of value added in the SMI industrial sector originating in the regions (42z). Additionally, it may be that the failure of IGLF to support more projects in rural areas relates to IGLF's focus away from cottage industry. The experience of Loan 1727-PH suggests that rather than compelling the banks or  - 14 - providing subsidies or even undertaking direct IGLF lending to stimulate lending in the regions, the issue of regional dispersion probably should be addressed through specially-designed micro-enterprise projects since a major part of the problem is the IGLF's inability to reach cottage firms, the vast majority of which are located in the regions. 5.7 The experience of Loan 1727-PH suggests that several lessons may be learned from the implementation of the project, particularly with regard to ibsues of (i) interest rates and foreign exchange risk; (ii) proper diversifi- cation of the apex lending institutions portfolio to the financial institu- tions; and (iii) the appropriate vehicles for lending to cottage industries. Aspects of the project relating to these three element, are discussed in paras. 5.8 to 5.13 below. 5.8 Interest Rates and Foreign-Exchange Risk. The loan shows the difficulties of lending through financial institutions in an economy with a regulated interest rate structure such as that of the Philippines at the time of the loan. The overall interest rate structure was set artificially low and even within that long-term rates were set particularly low. In order to ensure disbursement, the IGLF rates were set consistent with this structure. However, this meant that (i) the rate to the final beneficiary was significantly less than market conditions would have determined (on-lending rates were about 152 while inflation was in the range of 13-18z); and (ii) in order to give the PFIs adequate spreads to engage in long-term lending, they were advanced the IGLF money at a rate much less than they could borrow domestically (IGLF lent to the PFIs at 8.6? while 90 day deposit rates were 11.0?). 5.9 Since the regulated rate structure generally failed to reflect the economy's perception of foreign exchange risk and domestic inflation, the long-term rates charged by IGLF to the PFIs notably incorporated little or no premium to cover foreign Pxchange risk. The 8.6? average rate charged for IGLF funds at the outset of SMI II was 0.7? higher than the 7.9Z interest rate charged by the Bank on Loan 1727-PH. While this spread enabled the Bank- financed projects to cover a portion of IGLF's total administrative costs (which amounted to 0.9Z of average total assets), it was not enough to allow it to compensate the Government for taking the foreign exchange risk on the World Bank loan. Since the peso was devalued substantially against the currencies in which the loan was disbursed, the Government incurred significant foreign exchange losses on the Bank loan. The average rate of exchant for disbursement against the dollar was P 7.73, while the mid-1988 rate is about P 21. 5.10 By 1982, when interest rates in the Philippines had begun to be dere- gilated and ceilings on savings deposit rates had been removed, the IGLF prices were re-examined and adjusted upwards so that the PFIs paid 11? for small industry loans and 16? for medium industry. These increased interest costs were closer to the still-administered cost of time deposits. However the $24.5 million portion of Loan 1727-PH allocated to IGLF had been fully disbursed by that time. 5.11 The experience of Loan 1727-PH suggests that in such environments, apex lending should be linked to financial sector reform in which interest  - 15 _ rates are liberalized. Once rates are market-determined, interest rates from apex institutions to PFIs and to end-users can include a premium for foreign exchange risk. A oroposed fourth SMI World Bank loan which would provide further funding to IGLF will be made in the context of financial sector reforms and will include a provision that subloans be priced at market rates and hence provide reasonable cover to the Government for the foreign exchange and interest rate risks. 5.12 Diversification of IGLF Portfolio. Of the 32 accredited institutions as of December 31, 1981, 11 were particular.y active and accounted for 942 (by amount) of IGLF loans outstanding. Two institutions accounted for 54Z of the portfolio as of year end 1981. Over 30Z of the portfolio was held by a single institution, Manphil Investment Corporation, which later went bankrupt. Another 231 was held by the Private Development Corporation of the Philippines, which also experienced financial problems in the mid-1980's. The failure during the loan implementation period to highlight the excessive domi- nance of these two firms put IGLF in a potentially dangerous situation. Fortunately, Manphil's IGLF portfolio was reduced substantially before the default, but IGLF was left with sizeable arrears from that institution. Since the Manphil default, IGLF has imposed a limit on loans to any single institu- tion at 100Z of that institution's total capital, net of reserves. Such a restriction provides reasonable protectionL to IGLF in the case of default by individual PFIs. 5.13 Experience with Cottage Industry. IGLF's experience during the implementation of Loan 1727-PH suggests that the p;ogram is not an effective vehicle for extending credit to cottage or micro-enterprises. No cottage industry projects were financed under the lrn. For those subloans to cottage industry which were in repayment during 1980 and 1981, 88Z were affected by arrears of more than three months (compared to 14Z for small and medium indus- try loans during the same two-year period). The high costs of administering cottage industry loans makes them inappropriate for such a program. Since lending to cottage industries is incompatible with conventional banking operating practices, they cannot be administered by a majority of financial institutions in a cost-effective manner. The project illustrates that the apex arrangements which are appropriate for lending to small and medium firms may be inappropriate for effectively reaching cottage firms. VI. FINANCIAL PERFORMANCE OF IGLF 6.1 Financial Position. IGLF's projected and actual balance sheets for 1978 through 1984, presented in Annex 15, reflect the high growth in IGLF's loan portfolio under Loan 1727-PH. Total assets grew at an average rate of 37Z annually (see Annex 13), which growth was almost entirely financed by drawdowns on the two IBRD loans and Government counterpart funds. As of December 31, 1984, total assets stood at P 1,088 million. The loan portfolio had grown by about eight times over the period 1978 to 1984 and, as of the end of 1984, amounted to P 1,037 million. 6.2 Capital Structure and Liquidity. IGLF's net worth increased by more than three times between 1978 and 1984, growing from P 75.4 million in 1978 to P 333.7 million in 1984. The increase was due to the accumulation of  - 16 - counterpart funds from the Philippine Government plus retained earnings during the seven year period. IGLF's total debt Lo c-;1ity ratio fluctuated from 1.2:1 to 2.5:1 over the period. IGLF's liquidity position was erratic, moving from a high of 12.3 in 1978 to a low of 0.3 in 1982 and 1984. The tight liquidity position was largely due to additional funding requirements for term loans and to the increase in principal and interest payments on IGLF's long- term loans. 6.3 Profitability. IGLF's income statements for the period 1978-84 (see Annexes 14 and 15-A) show that IGLF's total operating income increased at an average of 442 per annum over the seven years under review and stood at P 89.5 million in 1984. The sources of income were interest on loans, interest on temporary investments and guarantee fees. Interest income from term loans as a proportion of total income increased from an average of 73Z for 1978-83 and 87Z for 1984. IGLF's return on assets fluctuated during the project years from 1.82 in 1978 to 3.1Z in 1984. Its return on equity improved from 3.71 in 1978 to 8.52 in 1980; it declined in 1981 and 1983 and later improved again to 10.4Z in 1984. Financial expenses as a percentage of gross income steadily increased from 512 in 1978 to 69Z in 1981 and decreased to 57Z in 1984. The decrease was partly due to the marked increase in interest income from term loans with only a slight increase in interest expenses on IBRD loans. 6.4 Because the Government absorbs the full foreign exchange risk on the World Bank loans, the financial losses resulting from deterioration of the peso do not affect IGLF. If IGLF were required to cover its foreign exchange losses, the financial picture during the 1978 to 1984 period would have been significantly less healthy. 6.5 Guarantee Scheme. One of the IGLF features embodieu under Loan 1727-PH is the optional guarantee coverage wherein the IGLF shareg in the losses with a financing institution in case of ultimate failure by the borrower to pay. Two kinds of guarantee coverage were offered. Under the credit risk guarantee (up to 602 of the loan for small industry and 402 of the loan for medium industry), the 22 guarantee fee is absorbed by the financing institution while under the collateral-short guarantee (maximum 25Z of the loan), the 22 guarantee fee is passed on to the borrower. A reserve account of P 1.0 million was set up in 1975, in accordance with the Master Agreement, to take care of losses on guaranteed loans. As of year-end 1984, the balance of the Provision for Losses on Guaranteed Loans amounted to P 11.6 million, which was intended to cover the estimated total losses of P 7.4 million, based on claims submitted by the various sponsoring financing institutions. Based on guarantee fees collected during the period, the amount of P 4.7 million was set up for possible losses in 1984. Losses on 18 foreclosed loans totalling P 3.7 million were written-off in 1984. 6.6 Audit. IGLF is audited yearly by the Commission on Audit and the Central Bank is provided with a copy of the audit report including any comments and findings of the auditor. The Office of the Auditor issued its report for 1984 without qualifications.  - 17 - VII. INSTITUTIONAL PERFORMANCE AND DEVELOPMENT Institutional Aspects 7.1 Manaaement. Policy issues affecting IGLF's operations during the implementation period of the loan were decided by a Review Committee composed of the deputy ministers from NEDA, DTI, the Department of Finance, a deputy governor of CB, the director of the UP-ISSI and the director of the Department of Loans and Credit at the Central Bank, as ex-officio member, and chaired by the Minister of the Budget (see Annex 21). The Technical Advisory Groip (TAG), composed of all agencies forming the IGLF Review Committee, exercised recommendatory and advisory powers dealing with policy and operational matters. The TA' was under the supervision of the Chief Executive Officer for IGLF. 7.2 Organization. IGLF was reorganized during the project period. In 1979, there were only two divisions involved in IGLF operations--the Indus- trial Loans Division and the IGLF Accounting, Securities Control and Collec- tion Division--and the total personnel complement of the organization was 48 people. With the approval of 18 new positions for IGLF by the Monetary Board of the Central Bank, the IGLF organizational structure was reorganized. The organization was changed to its present :tructure of four divisions (see Annex e) witsi cht Lr,-L,,I-iZg res;-~n.ibilities: Industrial Proiects Division - responsible for the desk processing/ evaluation of applications for IGLF financing under the accredita- tion/sponsorship schemes, the servicing of guarantees, requests for deferment/restructuring, and portfolio review of accounts in arrears. IGLF Examination Division - primarily responsible for evaluating potential institutions for accreditation and for the annual institu- tional performance review of accredited institutions, as well as end- use verification surveys of projects assisted. IGLF Accounting, Securities Control and Collection Division - handles all the accounting, securities control and collection functions of IkLF IGLF Special Studies Staff - handles research and technical informa- tion activities, and the reporting requirements to the World Bank, NEDA, CB and the Review Committee. 7.3 Training. Because of the clear importance within both IGLF and the PFIs of analyzing credit risk and assessing projects and institutions, IGLF initiated a number of programs under Loan 1727-PH to further develop the experience of its staff and that of the PFIs. In addition, during the project years, the staff of IGLF and the PFIs attended relevant local training programs, and 13 IGLF staff were sent abroad to acquire additional knowledge on small enterprise financing and related fields. They attended such training programs as Small Industry Promotion in Developing Economies, and Small Industry Financing Courses in India; Advanced Training on Financial Management in the Netherlands, and training at the Federal Business Development Bank in Vancouver and Montreal, Canada. Programs attended arE listed in Annex 20.  - 18 - Procedures and Standards 7.4 Proiect ADDraisal. Because of the importance of the project appraisal to the success of the program, IGLF developed and refined specific standards and procedures for appraisal under Loan 1727-PH. Under the accredi- tation scheue, project appraisal is the responsibility of the accredited financing institutions, which undertake the necessary detailed research and evaluation of the project, assess the appropriateness of the proposed financ- ing package based on the real needs of the borrower, and determine the overall feasibility of the project. The appraisal follows preliminary analysis of the relevant industries, as well as of the marketing, technica:, management and financial aspects of the projects. Plant visits provide first-hand views of the operations. Project feasibility studies or project evaluation reports are prepared for the approval of the PFIs executive committees and are submitted usually with only minor changes to IGLF. 7.5 An analysis of reports prepared under the project suggested some deficiencies which IGLF has sought to correct through training programs initiated under both this project and the follow-on SMI III project. A review of Loaii 1727-PH financed subprojects showed that evaluation procedures of the PFIs were deficient in their lack of attention to marketing and technical aspects, and in their over-estimation of financial benefits. For example, a mere list of competitors often was the sole marketing document. There was no attempt to identify existing and possible potential customers. No study was made to establish existence of sufficient demand in the locality. Regarding technical aspects, little consideration was given to the necessity for the buildings proposed for renovation or expansion, or to the benefits of the expansion project to the borrower in terms of production and space. The PFIs were generally optimistic in their financial forecasts, and actual subproject performance under Loan 1727-PH was lower than estimated (see Annexes 6 and 7). Some accredited institutions tended to rely more on the value of collateral than on project analysis. 7.6 To strengthen the quality of project evaluation, training programs were implemented during the project time period to assist financial institu- tion staff to screen loan applications and to identify possible strengths and weaknesses of projects and potential problems. The programs helped PFI personnel to analyze project technical and marketing data in greater depth, and to adjust financial forecasts from optimistic scenarios, and to undertake sensitivity analysis. 7.7 Proiect Supervision. Most of the accredited financing institutions, although primarily responsible for project supervision, were found under the SMI II Project not to have any established, definitive system of supervision and account monitoring in their IGLF lending. Post-audit surveys of IGLF projects were not conducted by the PFIs on a regular basis. The repayment performance of IGLF borrowers was used as a gauge as to whether or not there was a need for visual inspection. Results of plant visits, if any, were summarized in post-audit inspection reports which contained limited informa- tion on the utilization of the loans and on the operating performance and financial status of the borrowers. The value of collateral was not reassessed during the project period. No records were maintained to systematically monitor the status and performance of each IGLF borrower. Credit file folders  - 19 - maintained by participating financial institutions were poorly kept and data therein were not updated. The submission of periodic financial statements by each IGLF subborrower was not strictly imposed but was enforced only during the IGLF examiners' end-use verification of selected projects. While the deficiencies could be partly explained by the limited number of personnel directly assigned to supervise these projects, IGLF has since the completion of the project stressed the importance of adequate supervision by the PFIs and has included guidelines for appropriate supervision in its training programs. 7.8 Although at the outset of Loan 1727-.?H there was no regular review conducted on the performance of participating financing institutions, ad hoc monitoring of PFIs was performed and the PFI's were required to correct or remedy deficiencies or shortcomings as they emerged. With the establishment of the IGLF Examination Division under the loan, however, institutional appraisal of financing institutions began to be conducted regularly. Reports on findings and deficiencies were transmitted to the institutions concerned and were accompanied with recommendations. Financing institutions were required to submit reports to IGLF on any action to correct or remedy these findings, and on any recommended measures implemented. VIII. BANK PERFORMANCE Proiect Justification and Overall Performance 8.1 The project achieved its basic objective which was to expand IGLF's lending to the SMI sector in the Philippines. Long-term resources not other- wise being provided within the banking system were made available for long- term financing needs of both small and medium scale industries. Furthermore, substantial progress was made in improving repayment performance, which has reached a level to effectively sustain the IGLF program. IGLF played a catalytic role in providing employment opportunities, improving income distri- bution and regional development, and stimulating exports and general economic growth. Under the project, IGLF'S network of financial institutions was expanded significantly, and expertise in project evaluation and preparation was gained. Bank involvement helped IGLF to develop its institutional frame- work, organization, and operating policies and procedures. Supervision 8.2 The Bank's supervision effort on thG project was generally adequate, although more frequent supervision, specifically a mission in 1981, may have revealed earlier that the IGLF interest rate was out of line with the general economic environment and may have resulted in an earlier change of the IGLF price charged to PFIs and end-users. Working Relationship 8.3 The Bank maintained good working relationships with the Government and IGLF. IGLF made all relevant information readily available to Bank staff and the response to inquiries was always prompt and adequate.  - 20 - IX. SUMHARY AND CONCLUSIONS 9.1 The Industrial Guarantee and Loan Fund continued under Loan 1727-PH to act as a key instrument in the Philippine Government's program to promote and develop small and medium-scale industries on a nationwide scale. During the implementation period of Loan 1727-PH, the IGLF expanded to include finance for medium scale firms, and the IGLF provided a major portion of total long-term credit borrowed by the Philippine.' small and medium industry sectors. Loans approved increased by almost 502 a year, from P 47 million in 1978 (the date of project appraisal) to P 485 million in 1984, with projects financed by those loans generating about 10,497 new jobs. Total disbursement of the loan proceeds was realized ahead of the projected schedule. Most of the 264 subprojects financed under the loan proved to be economically and financially viable in addition to being labor-intensive. 9.2 IGLF's network of accredited institutions' branches throughout all regions of the country was expanded significantly during the time period of the loan. Although the percentage of total funds allocated to areas outside Metro-Manila did not equal. the minimum 60? projected at the time of appraisal, the 39? of all funds which did go to firms outside the capital compared well with the share of SMI value-added contributed by firms outside Metro-Manila (42Z). As intended, the sectoral distribution of IGLF's loan portfolio was dominated by loans to the manufacturing sector although some service and tourism oriented industries also were beneficiaries. 9.3 During the implementation period of loan, the IGLF program substantially improved its arrears and collection ratios, which by 1984 were at satifactory levels. The proportion of the PFIs' portfolio to end-users affected by arrears was only 12Z by amount and 22? by number in 1984, and arrears themselves amounted to only 4.5? of all IGLF loans outstanding (or of 3.5? of IGLF loans made under the accreditation scheme). Loan arrears from the PFIs to IGLF amounted to only 2.12 of the IGLF's loan portfolio in 1984. 9.4 IGLF's financial position was acceptable. Return on assets fluctu- ated during the years from 1.82 in 1978 to 3.1X in 1984. Return on equity improved from 3.7Z in 1978 to 8.5? in 1980; it declined in 1981 and 1983 and later improved again to 10.4? in 1984. Financial expenses as a percentage of gross income steadily increased from 51? in 1978 to 692 in 1981 and decreased to 57? in 1984. The decrease was partly due to the marked increase in interest income from term loans with only a slight increase in interest expenses on IBRD loans. The continuous growth in total assets during the 1978-84 period was financed almost entirely through IBRD loans and Government counterpart funds. Since IGLF relied heavily on foreign borrowings and since the Philippine Government took the foreign exchange risk on the loans, the Government incurred substantial losses from foreign exchange adjustments. The peso-dollar exchange rate (see Annex 22) declined from P 7.4 in 1978 to 1 11.0 in June 1983, to P 14.0 in October 1983, to 1 18.4 in December 1984, and P 21.0 in June 1988. The proceeds of Loan 1727-PH were disbursed at the average rate of P 7.7 to US$1.00. 9.5 IGLF staff was reorganized and strengthened during the project period. IGLF expanded from two to four divisions, and staff numbers were increased accordingly. IGLF offered technical assistance and training to both  - 21 - financing institutions and end-users, to enhance staff expertise in project appraisal, project evaluation, and project supervision and monitoring. Public relations efforts to help expand the usage of IGLF were implemented and were influential in promoting the program. 9.3 Several lessons which may be learned from the Lank's and IGLP's experiences under Loan 1727-PH are outlined below: (a) In pricing subloans made by apex institutions in interest rate regulated economies, it may be difficult to ensare that end-usfrs pay an appropriate foreign exchange risk premium. The experience of Loan 1727-PH suggests that in such environments, apex lending should be linked to financial sector reform in which interest rates generally are deregulated. Once rates are market-determined, then interest rates from apex institutions to PFIs and to end-users can include a premium for foreign exchange risk. A currently proposed Bank loan which would provide further funding to IGLF will be made in the context of financial sector reforms and will price subloans at market-linked rates. (b) During the implementation period of Loan 1727-PH, IGLF did not limit the participation of individual financial institutions in order to ensure that the IGLF portfolio as a whole was balanced or diversi- fied. The weakened financial situation of some of the PFIs during 1980 to 1984 period suggests that this failure to diversify was unwise. In 1981, after the bankruptcy of one PFI holding a large portion of the IGLF portfolio, IGLF adopted a policy limiting the participation of individual PFIs and specifying that a PFI's port- folio of IGLF rediscounted loans may not exceed that institution's total net worth. (c) The experience of IGLF under Loan 1727-PH suggests that the apex arrangements which are appropriate for lending to small and medium firms may be inappropriate for effectively reaching cottage firms. IGLF's record with cottage industry during the project period showed poor repayment and suggested that making and supervising cottage ine'istry loans was overly cumbersome and expensive for IGLF. The da;a suggest that another vehicle for financing cottage industry should be created and that IGLF should not be pushed to fil] the financing gap to that sector.  PIW= OOWLVFIN iEPORr - LAd 1727-li STRn7 DO DDlTEF Rom, 1974 - 1984 (I pm' am. a 1965 1967 1969 1974 1976 1977 1979 1960 1961 /3 1982 1963 Uft Saving s?4Apoits /1 Ceircial lIs- 5-3/4 6 6 6 7 7 1 9 NMD eiling NO Ceiling NO Oeiliu MD cilingm nift BCnk 5-3/4 6 6 6-1/2 7-1/2 7-1/2 7-1/2 9-112 MD Ceiling No Oli biftiNng lb iling Ruml 3anks 5-3/4 6-1/2 6 6-1/2 7-1/2 1-1/2 7-1/2 9-1/2 ND Ceiling No Cailing No Ce ling Mb Cidling Non Stut* SIAs 7-1/2 No Ceiling No Cailing No aCilig lb ilin "NOW' wmoxats Thrift Wbi's 5 7 MD Ceiliug MD Callg o Ceidling IID CaiL SdvJnU Deptbits /2 hands with Denaia Deposi-S 9 9 No CilIng No Ceiling NO Celing lOb Celing Badw witkout xADn Deporits 9-1/2 9-1/2 No Ceiling ND Ceiling lb Cailint No Ceiling Tim Deposits 90 days 5-3/4 5-3/4 8-112 9 9 9 11 11 11 it lbN iling 180 days 6 6 6-1/2 9 9-1/2 9-1/2 9-1/2 11-1/2 11-1/2 11-1/2 11-1/2 No Ceiling 270 days 6-1/4 6-1/4 9-1/2 11-112 11-1/2 11-1/2 111-/2 No aCiling 360 days 6-1/2 6-1/2 7 10 10-1/2 10-1/2 10-1/2 12-1/2 12-1/2 12-112 12-1/2 No Oeiling 540 days 11-1/2 11-1/2 11-1/2 13-1/2 13-1/2 13-1/2 13-1/2 No CeiLing 730 days 11-1/2 12-1/2 12-1/2 12-1/2 14-1/2 14-1/2 14-1/2 14-1/2 No Ceiling Ove 730 days N o C E I L I N G Deposit SdsutiLutes 17 17 17 No Geing No Ceiling No Ceiling No GCiling AU Naturities 17 17 17 No CilUi NO Ciin No cilig No Ceiling Dank LuiJir' kates Suavised (rxdit 10 10 10 12 12 12 Im 90-2 Msupm 4ed c'di t 12 12 12 14 14. 14 m 90-2 Nw.-Tdit iiid 9 9 9 12 12 MR 90-2 f4# - 2 bwiitiwnl 12 12 12 14 14 mm ligo it o /1 ) > Nov. 10, 1965 to May 28, 1979, banks wre classified into the diftfat oateewies. f7 As of Dec. 1, 1976, banks wae classified'into those buds with mid withit d_Gd%qwits. /I )mm July 1, 1981 to present, al ceilings in deposit rates m 2KWd IpOF t tme lygo. , / n90 retenI to tht mnila referenoe mate fer 90 dAy datowwid aW wm.mred by % O cw floating ruts PIpi orH miniS a rate to be cieteumin by tte CB an the basis of the pmiliz. Ut tmti.O  1965 1967 1969 1971 1976 19771 1979 19j0 1981 /3 1962 1983 196% ILUX Int:rest Rates (CB to Pll) S lU InurBty loan 5 5 5 5 7 7 7 6.7 6.7 11 11 Medium In dustry loa 9 9 36 16 16 ICLF Intemest kites (Pfl to Erd4ker) SAll nu1try Ioan 10 10 10 10 13 13 13 13.2 13.2 18 11 23 Medium In.stry loan 1 is 15 21 21 23 MlI' Interest Rates SaoU In)ixy loan 12 12 12 12 12 12 12 14 1J la 1 Qad IW.um Inuurt r Ioan 14 1I 1 14 1% 1F 14 14 16 16 21 2L /3 1s July 1, 1981 to pre wAt, all ceilinp in deposit mtes ns rmd wampt tlve applyin4 far shaat tam liiu. 0i a'4'  - 24 - AMMx 2 PFOJECT COMPLETION REPORT - LOAN 1