Colombia - Fourth Small-scale Industry Project

Document of The World Bank FOR OFiICAL USE ONLY Lk). -2-cj- C Report No. 4929b-CO STAFF APPRAISAL REPORT COLDMBIA FOW-RTh SMALL-SCALE INDUSTRY PROJECT June 14, 1984 Projects Department Latin America and the Caribbean Regional Office This domment has itrstrcted bstibution and ny be ued byr recipients only in the performnce of ttbi'r olicial dufies. Its contents way not otherwrise be disclosed witbout World Bak satborlzstion.  CURRENCY EQUIVALENTS (as of May 31, 1984) Currency Unit = Colombian Peso US$1 = Col$98.47 Col$1 = US$0.0102 ACRONYMS ACOPI - Asociacion Colombiana Popular de Industriales (Colombian Association of Small-Scale Industrialists) BCE - Banco Central Hipotecario (Housing Bank) BR - Banco de la Republica (Colombia's Central Bank) CA Consejo Asesor de la Politica para la Pequena y Mediana Industria (Advisory Council for Small- and Medium-Scale Industry) CAJA - Caja de Credito Agrario Industrial y Minero (Agricultural, Industrial and Mining Bank) CAVIs - Corporaciones de Ahorro y Vivienda (Savings and Loan Associations) CDT - Certificado de Deposito a Termino (Term Desposit Certificate) CESO - Canadian Executive Service Overseas CFCs - Companias de Financiamiento Comercial (Trade Finance Companies) CFP - Corporacion Financiera Popular COLCIENCIAS - Fondo Colombiano de Investigaciones Cientificas y Proyectos Especiales (Colombian Fund for Scientific Research and Special Projects) DANE - Departamento Administrativo Nacional de Estadistica (National Department of Statistics) DFC - Development Finance Company DNP - Departamento Nacional de Planeacion (National Planning Department) FEDESARROLLO - Fundacion para la Educacion Superior y Desarrollo (Foundation for Higher Education and Development) FFI - Fondo Financiero Industrial (Industrial Financing Fund) FICITEC - Fundacion para el Fomento de la Investigacion Cientifica y Tecrologica (Foundation for the Development of Scientific and Technological Research) FINANCIACOOP - Instituto de Financiamiento y Desarrollo Cooperativo (Institute for Financing and Development of Cooperatives) FIP - Fondo de Inversiones Privadas (Private Investment Fund) FNG - Fondo Nacional de Garantia (National Guarantee Fund) IDB - Inter-American Development Bank IFI - Instituto de Fomento Industrial (Industrial Development Institute) IIT - Instituto de Investigaciones Tecnologicas (Institute of Technological Research) KfW - Kreditanstalt fuer Wiederaufbau PROEXPO - Fondo de Promocion de Exportaciones (Export Promotion Fund) RAB - Junta Asesora Regional (Regional Advisory Board) SENA - Servicio Nacional de Aprendizaje (National Vocational Training Organization) SIP - Secretaria de Integracion Popular de la Presidencia de la Republica (Secretariat of Popular Integration of the Presidency of the Republic) SKI - Small- and Medium-Scale Industry SSI - Small-Scale Industry UNUP - United Nations Development Program UNIDO - United Nations Industrial Development Organization USAID - United States Agency for International Development FISCAL YEAR January 1 - December 31  FOR OFFICIAL USE ONLY COLOMBIk STAFF APPRAISAL REPORT FOURTH SHALL-SCALE INDUSTRY PROJECT TABLE OF CONTENTS Page No. LOAN AND PROJECT SUMMARY ......................................... (I) I. SMALL- AND MEDIUM-SCALE INDUSTRY (SMI) IN COLOMBIA . ......... 1 A. The Industrial Sector Framework ...........se. ..........-. I B. Structure and Characteriatics ........................... 2 C. Growth and Development Constraints ............. ....... 3 D. Government Objectives and Policies ..................... 3 II. FINANCIAL AND TECHNICAL ASSISTANCE TO SKI ................... 4 A. Industrial Credit .... .................................. 4 B. SMI Financing ....* .................................... 5 C. Technical Assistance to SKI . .......... ............. ... 6 III. CORPORACION FINANCIERA POPULAR .................... .... 7 A. Basic Structure 7........................................ B. Systems and Procedures * ............................ 9 C. Operations .......... 11 D. Resources .................................. 14 E. Financial Position, Results and Projections ............ 15 IV* THE PROJECT 19 A. Experience Under Previous Bank Projects ................ 19 B. Project Objectives .......... ....00.00.00 ... ........... 21 C. Project Cost and Financing ............................. 22 D. Project Components ..................................... 23 E. Procurement and Disbursement............. ................ 25 F. Accounting, Auditing and Reporting ...... ............... 27 G. Benefits and Risks *****.*.............................. 27 Vs RECOMMENDATIONS *******..........*********** *.............. 28 This report is based on the findings of an appraisal mission which visited Colombia during November 1983. The mission comprised Messrs. Alonso, Voljc, and Wogart and Ms. Press (all LCPI2). This document has a retricted distribution and may be used by recipents only in the perfonnanc of their official duties. Its contents nmy not otherwise be disckoed without World Bank authorizaton  TABLE OF CONTENTS (Continued) LIST OF ANNEXES Arnex 1: Comparison of Features of the Past and Proposed Small-Scale Industry Loans. Annex 2: Institutional Characteristics of Financial and Technical Assistance to SNI. Annex 3: Supporting Tables and Charts. T-1 Industrial Development and its Determinants 1970-82. T-2 Distribution of Employment According to Size of Enterprise. T-3 Establishments, Employment, Value Added of SMI, 1976-81. T-4 SMI's Participation in Manufacturing Sector and Subsectoral Share. T-5 Capital and Labor Productivity of SMI, 1976 and 1980. T-6 Remuneration in Manufacturing, 1972-80. T-7 Sources and Uses of Investment Financing, 1978-82. T-8 Institutional Credit to SMI, 1970-81 T-9 Fondo Financiero Industrial Approvals, 1969-1982. T-10 Share of Principal Intermediaries in FFI Approvals, 1970-82. Tables on Corporacion Financiera Popular T-11 Distribution of Ownership, 1979-83. T-12 List of Board of Directors and Alternates as of November 1, 1983. T-13 Organization Chart as of November 1, 1983. T-14 Number of Staff at Beadquarters and in the Regional Offices, 1979-83. T-15 Summary of Supervision Activities, 1979-83. T-16 Analysis of Loan Approvals, 1979-83. T-17 Principal Characteristics of Bank Financed Subloans, 1976-83. T-18 Balance Sheets as of December 31, for the Period 1970-83. T-19 Income Statement, 1970-83. T-20 Sources and Applications of Funds, 1980-83. T-21 Inflow of Funds, 1973-83. T-22 Summary of Resource Position as of June 30, 1983. T-23 Key Financial Indicators, 1972-82. T-24 Aging of Arrears, 1979-83. T-25 Administrative Expenditures, 1979-82. T-26 Lending Rates as of November 1, 1983. T-27 Average Subproject Impact - Survey I. T-28 Average Subproject Impact - Survey II. T-29 Employment Generation and its Costs for an Average Subborrower. T-30 Sales and Productivity Growth in an Average Subborrower. T-31 Lending Program, 1984-87. T-32 Balance Sheet Projections, 1984-87. T-33 Projected Sources and Applications of Funds, 1984-87. ~ - T-34 Statement of Income and Expenditures, 1984-87. T-35 Key Financial Indicators, 1984-87. ,Annex 4: Estimated Quarterly DisbuLsement Schedule for the Bank Loan. Annex 5: Related Documents Available in the Project File. - MAP  - iII - Estimated Disbursements: (US$ millions) IBRD FY85 FY86 FY87 FY88 FY89 Annual 2.0 7.2 18.0 8.4 4.4 Cumulative 2.0 9.2 27.2 35.6 40.0 Rate of Return: The financial rates of return on the subprojects are expected to average about 15S and be no lees than 12Z. The economic rates of return are, bowever, likely to be higher, primarily because prices of input saterials are somewhat Inflated as a result of the protectlon Involved while, on the other hand, most suall-scale industry products are openly traded so that market prices reflect economic values. As a norm, therefore, the typical products of small-scale industries have lower protection rates than the corresponding inputs. Appraisal Report: Report No. 4929b-CO, dated June 14, 1984.  COLOMBIA FOURTH SMALL-SCALE INDUSTRY PROJECT LOAN AND PROJECT SUMMARY Borrower: Banco de la Republica (BR) Guarantor: Republic of Colombia Beneficiary: Corporacion Financiera Popular (CFP) Amount: US$40 million equivalent, including capitalized front-end fee of US$99,751. Terms: 17 years, including four years of grace, at the standard variable interest rate. Relending Banco de la Republica would relend the proceeds of the loan Terms: to CFP in pesos, and CFP would make service payments to Banco de la Republica whenever this shall be required to meet service payments on the Bank's loan. Any net losses to Banco de la Republica as a result of covering the foreign exchange risk would be charged by Banco de la Republica to the Govern- ment's Foreign Exchange Special Account administered by Banco de la Republica. Banco de la Republica would charge interest: (i) on the credit component of the project, at rates 4.5 or 5 percentage points lower than the interest rates on subloans made by CFP for investment projects located, respectively, in the major metropolitan areas or outside these areas; and (ii) on the technical assistance component of the project, at 12% per annum. CFP would on-lend the proceeds of the loan for 4 to 10 years, including 1 to 3 years of grace, at a rate representing the weighted average of a variable and a fixed interest rate as follows: for each sub-loan, 15% of the principal would bear interest at the rate payable on 90-day certificates of deposit plus 3 percentage points, and 852 of the principal would bear interest at a fixed rate to be agreed semiannually with the Bank (initially 262 p.a. for subprojects in the metropolitan areas and 24Z p.a. outside the larger centres) and established by Banco de la Republica. At the end of the first full year of operation of the line (December 1985) BR, CFP and the Bank would review the variable/fixed proportions of the interest rates with a view to increasing to 30% the portion of the principal of each sub-loan subject to variable  - ii - interest rates, thus reducing commensurately the portion under fixed rates. The new percentages would apply to sub-loans approved after January 1, 1986. A slilar review 12 months thereafter would seek to increase further the proportion of market-deteruined rates in the weighted average interest rates of the sub-loans. Project The proposed project would consist of the provision of: (a) Description: medium- and long-term credit through CFP to small and medium-scale industries for fixed investment and permanent working capital; and (b) technical assistance to CFP and small scale industriallsts. Project Risks: Even though the project involves no special risks, Insufficient Government equity contributions to CFP, delays in strenthening CFP's regional offices, slower than expected recovery of small-scale industry, and fluctuations in availability of local working capital might affect project implementation. However, agreements reached with the Government regarding increases to CFP's equity base, together with the Action Program to be carried out bs CFP In close collaboration with the Bank, should help ensure achievement of project objectives. Estimated Cost:1/ (US$ million equivalent) Local Foreign Total Construction and installation 8.9 2.2 11.1 Imported mchinery (CIF) - 7.5 7.5 Imported machinery acquired locally 10.6 13.7 24.3 Locally anufactured mchinery 13.2 6.1 19.3 Working capital 45.6 5.0 50.6 Technical assistance 0.7 0.4 1.1 Front-end fee 0.1 0.1 79.0 35.0 114.0 Financing Plan: (US$ million equivalent) Local Foreign Total z CFP 4.5 - 4.5 4 Enterprises 30.5 - 30.5 27 Otber local sources 39.0 - 39.0 34 IBRD 5.0 35.0 40.0 a/ 35 79.0 35.0 114.0 100 a/ Includes capitalized front-end fee of US$99,751. 1/ Taking account of Bank estimates of Inflation and peso depreciation, as of January 1984.  COLOMBIA STAFF APPRAISAL REPORT FOURTH SMALL-SCALE INDUSTRY PROJECT I. SMALL- AND MEDIUM-SCALE INDUSTRY (SMI) IN COLOMBIA A. The Industrial Sector Framework 1.01 Colombia's manufacturing sector has expanded rapidly over the past two decades. As a result, the share of manufacturing in GDP rose from 16% in 1960-1964 to about 22% in 1980-1982. However, growth rates for the sector have been quite variable, achieving an average of 8.6% in 1967-1974, decelerating to 5.2% in 1974-1978, and falling to an average of only 1% in 1978-1982. During the last 20 years, the structure of the sector has changed considerably, with consumer goods declining from about four-fifths to one-half of the total and intermediate goods increasing to account for about 40% in 1982. Capital goods make up the remaining 10% in total production. However, as output and employment growth declined in the last few years, con- sumer goods industries once again grew in importance relative to industries producing intermediate and capital goods. 1.02 Policies favoring import substitution in the 1950s and 1960s, followed by export promotion measures in 1967, stimulated the growth of the manufacturing sector. While less important in absolute terms than the expansion of internal demand, the dynamic growth of manufactured exports was the principal impetus for the rise in industrial output and employment between 1967 and 1974. After rapid export development in the early 1970s, the Colombian industrial sector catered again to the domestic market during the latter part of the last decade, mainly because of the spillover from the boom in traditional exports and an overvalued exchange rate (Table 1). However, during the last few years, competition from contraband, weakened domestic and international demand, lower international prices for major agricultural exports, and substantial devaluations by Colombia's most important Latin American trading partners and export competitors have resulted in decreasing industrial production and exports. 1.03 Reacting to the recession abroad and at home, the Government reverted to protection through shifting a large share of imports from free to previous licenses and increasing external tariffs. At the same time, incentives to exporters have been strengthened, first, through increased tax rebates for exports and, in 1983, through the acceleration of the rate of devaluation. The objective of the export-oriented strategy is to stimulate production in the industrial and agricultural sectors. While most small- and  - 2 - medium-scale industries (SMIs) 1/ do not export directly, they are expected to benefit indirectly, especially from the expansion in agricultural activities. If the current export incentive system can be maintained in the next two years, export competitiveness of the Colombian industrial sector should be restored in 1985, allowing for phasing out of most of the import licenses by the second half of the 1980s. The authorities have also initiated a strategy of promoting the construction industry as an additional help to stimulate economic recovery and to generate industrial growth and employment. The Government also intends to increase the share of capital goods industries in manufacturing value added. Again, SMIs from several subsectors should benefit from the above-mentioned policies. However, with the expected recovery of the industrial sector, a renewed foreign exchange bottleneck (characterized by a current account deficit of US$2.2 billion and considerable capital flight in 1983) is likely to persist in the near future. B. Structure and Characteristics 1.04 The SMI subsector plays a significant role in Colombia's industrial development process, employing about two-fifths of the industrial labor force and producing close to one-fourth of manufacturing value added. Most important is its impact upon employment generation. Recent studies have shown that, compared with large-scale manufacturing, SMIs have generated up to about three tim,es as many jobs per peso of invested capital. They also absorb a higher share of unskilled workers than do large enterprises. At the same time, they also seem to use capital as effectively as larger firms (Table 5). Two further characteristics of SMls are their lesser dependence upon imported inputs and the fact that they are much more evenly distributed throughout the country than industry as a whole. SMIs' economic importance, particularly in smaller centers, is, in part, attributable to the relatively high transportation costs and the distinct regionalization of markets in Colombia. Finally, SMIs provide a productive outlet for private savings. These firms are self-financed to a much greater degree than large industry, partly because of their more limited access to funds from the organized financial system. Since 1978, over 50% of the total financing to SMIs came from the small industrialists' own resources, and 29% from commercial sources, compared with 25% and 48%, respectively, for industry as a whole (Table 7). 1/ Colombia does not have a single official definition for SMI. The National Department of Statistics (DANE) defined SKIs, for purposes of data collection and tabulation, as firms with 5-99 workers, and small- scale enterprises (SSI), as Eirms with 5-49 workers. With the rise of micro-enterprises, DANE has recently increased the minimum number of workers for SSIs from 5 to 10. Currently, firms with total assets of less than Col$ 60 millicn (some US$675,000 equivalent), regardless of the number of workers, qualify as SMIs for loans from the Fondo Financiero Industrial (FFI) and Corporacion Financiera Popular (CFP).  -3- C. Growth and Development Constraints 1.05 Available statistics indicate that, during the mid-1970s, SMIs' output grew about 6.6% p.a. in real terms as compared to 5.2% for the manu- facturing sector as a whole. As in the rest of the industrial sector, SMI growth rates have been declining since 1980, with a sales growth of only 1.3% reported for 1982. In spite of the fact that their exports grew faster than those of larger-sized firms between 1978 and 1981, SMIs, on the average, exported less than 2.5% of their output. Moreover, their exports decreased sharply in 1982, as large devaluations in Venezuela and Ecuador reversed trade flows. On the other hand, employment in the SMI subsector was still growing as recently as 1982, when most of the larger firms had to dismiss workers. 1.06 The relatively satisfactory performance of the SMI subsector has been achieved despite some difficulties that still face SMIs, such as limited access to credit, insecure supplies of raw materials, limited technical assistance, and inappropriate technology/management systems. Inadequate access to institutional credit has been a consistent constraint to SMIs' development. By virtue of their characteristics and geographical dispersion, SMIs frequently lack information concerning available sources of institu- tional finance. This problem is compounded by reluctance on the part of commercial banks and private DFCs to lend to small industry, which may be a result of the higher risks and administrative costs of such lending and the low intermediation margins offered under the Central Bank's (BR) rediscount- ing facility. D. Government Objectives and Policies 1.07 Government industrial policies have focused on the sector as a whole without differentiating between SMIs and larger enterprises. Neverthe- less, both general industrial and macro-economic policies have affected SMIs less than larger enterprises because of the former's greater self-sufficiency and larger involvement in the informal part of the Colombian economy. Speci- fic Government policies toward small- and medium-scale industry have not been clearly defined or well coordinated, except for the authorities' support to CFP and other official sources of SMI financing as well as to a few institu- tions providing technical assistance to SMIs. This situation, however, has changed recently, partially as a result of the Bank dialogue with the Govern- ment over the past years, with support coming from both a more coordinated effort by the authorities to design specific policies for SMIs and from recently announced policies in such fields as the capital goods industry and assembly operations, which are expected to benefit SMIs. A more coordinated policy approach to address the specific needs of SMIs is being encouraged further. The Advisory Council for Small and Medium Industry (CA) has been established, consisting of upper level representatives of private and public sector organizations concerned with SMIs, such as the Ministry of Economic Development, the National Planning Department (DNP), the National Training Service Organization (SENA), Corporacion Financiera Popular (CFP), the Export Promotior. Fund (PROEXPO), the Industrial Development Institute (IFI), the National Guarantee Fund (FNG), the Secretariat of Popular Integration of the ?rmsidency of the Republic (SIP), and the Colombian Association of Small- Scale Industrialists (ACOPI). CA is currently developing, through its work-  - 4 - ing groups and the Industrial Sector Planning Division of the Ministry of Economic Development which serves as CA's technical secretariat, guidelines for financial and technical assistance to the SMI subsector, including a revised definition of SMIs2/. In addition, the impact on SMIs of policies regarding subcontracting, zovernment procurement, and industrial parks, as well as five SMI subsectors (food, printing, metal-mechanics, plastics and garments) will be analyzed, under CA coordination, for policy formulation purposes. Because of their significant impact upon employment, the Govern- ment also intends to assume a leading role in a program to assist very small firms from the informal sector of the economy (so called "micro- enterprises"), a program which has been carried out successfully by private sector foundations (para. 3.14). 1.08 According to DNP's projections, which are made taking into account the Government policies discussed in paragraph 1.03, the Colombian industrial sector is expected to recover gradually, achieving estimated growth rates of 4% p.a. in real terms between 1984 and 1987. Once the overall industrial growth in Colombia resumes its earlier pace, expansion of SMIs may proceed even faster if the experience of the 1970s is repeated. To raise the sub- sector's contribution to employment and to value added, the coordinated efforts of both the private and public sectors would have to be strengthened further, and additional financial resources and technical assistance services would have to be mobilized. The proposed project would assist the Government in its efforts to increase the availability of term credit and technical assistance to small-scale industrial enterprises. II. FINANCIAL AND TECHNICAL ASSISTANCE TO SMI A. Industrial Credit3/ 2.01 Colombia has a relatively well-developed and diversified financial sector comprising the Central Bank (Banco de la Republica - BR), about 25 coumercial banks and 25 corporaciones financieras (DFCs-investment banks), a central mortgage bank (Banco Central Hipotecario--BCH), ten savings and loan associations (Corporaciones de Ahorro y Vivienda--CAVIs), nearly 40 trade finance companies (Companias de Financiamiento Comercial--CFCs), more than 70 insurance companies, several mutual funds and two stock exchanges (at Bogota and Medellin). The commercial banks have the largest branch network, with more than 1,800 offices distributed throughout Colombia, followed by the CAVIs (400 branch offices in total), BCH (75 offices) and CFs (over 100 branch offices). Commercial banks represent the most important segment, accounting for more than one-third of the total outstanding credit of the institutionalized financial system in recent years. 2/ In principle, the definition of SMIs under preparation would include three criteria: fixed asset size, sales value and number of employees. 3/ Report No. 4274-Co, The Colombian Investment Banking System and Related Financial Sector Issues, dated August 1, 1983, presents detailed information about Colombia's financial sector.  -5- 2.02 Direct lending by banks and DFCs is supplemented by rediscounting funds of BR funded in part out of the reserve requirements of the banking system. For industry, the mDst important of these funds are: (a) the Industrial Financing Fund (FFI), which rediscounts loans made by commercial banks and financieras to small- and medium-scale industrial firms; (b) the Private Investment Fund (FIP), which rediscounts loans for firms larger than those covered by FFI; and (c) the Export Promotion Fund (PROEXPO), which provides credit to non-traditional export activities (mostly industry), using mainly funds derived from a 5% import tax. The Caja Agraria (state-owned Agricultural, Industrial and Mining Bank), once a significant source of institutional credit to SSI, now offers term financing to only the smallest agro-industrial firms.4/ (Table 8). B. SMI Financing 2.03 Because of SKI's traditional lack of ready collateral and the high cost of SMI subloan processing, the sector has considerable difficulty in gaining access to conventional commercial sources of financing, particularly for medium- and long-term purposes. Consequently, the bulk of the SMI loar. portfolio of commercial banks and financieras is now financed by FFI or other special credit lines. Corporacion Financiera Popular (CFP), as the Government specialized SMI lending institution, provides credit for fixed investment and working capital. It accounts for some 30X of total SMI credit, thus providing the bulk of the medium- and long-term credit to SKIs (paras. 3.11, 3.30-31, and 4.03). However, while, in aggregate, the amount of public funds available for lending to SKI has increased in real terms since 1974, it has, in the past, been subject to fluctuations that have not always been conducive to the steady institutional development of SMI lenders. Availability of official funds for SMI financing has only recently become more stable. Access of SSIs and micro-enterprises to commercial credit has traditionally also been constrained by their inability to provide sufficient collateral. The Government has, therefore, recently established a National Guarantee Fund (FNG) which should help to alleviate this problem by providing guarantees. 2.04 There are considerable variations in the cost of industrial credit. Most lending rates in the formal financial market were liberalized in 1980, but effective rates have remained high in real terms (e.g., 17Z p.a. for commercial bank and DFC loans in 1982), mainly because of distor- tions of the financial sector and the overvalued exchange rate. Medium- to long-ternm rates from FFI and CFP ranged from 18% to 26% p.a. in nominal terms (Table 26), with most loars at about 2% p.a. in real terms in 1982. During 1983, CFP and FFI rates were maintained unchanged in nominal terms, thus yielding higher real rates as inflation decliced. In December 1983. the Monetary Board approved a new interest-rate system for FFI loans, designed to further expose SKI borrowers to free-market interest rates. Instead of bear- ing a fixed interest rate, individual subloan3 made by intermediaries to sub-borrowers now bear a fixed rate on the portion redlscounted with FFI, with the remainder bearirg a rate of 3 points above the floating 90-day 4 Anntlex 2 includes further details on institutions providing official credit to SKI.  - 6 - certificate of deposit (CDT) rate. A similar rate structure will be used for CFP lending rates under the proposed loan (para. 4.12). It is expected that this arrangement will introduce SMIs to receiving credit at interest rates closer to prevailing commercial rates and that commercial banks and private DFCs may thus find lending to SNIs more attractive as a result of the in- creased intermediation margins available to them. C. Technical Assistance to SMI 2.05 Provision of technical assistance to SMIs remains rather weak and poorly coordinated compared to the financial assistance. In total, up to some 700-800 SMIs are receiving some kind of technical assistance each year from the various institutions discussed below. These firms represent less than 5% of the total population of SKIs in Colombia. Moreover, technical assistance services have been largely fragmented, with little cooperation among institutions at the regional level, and no joint focus or definition of priorities despite the existence of various cooperation agreements at the national level, in particular between the Corporacion Financiera Popular (CFP) and most of the below-mentioned technical assistance institutions. To some exteut, this problem reflects the difficulty of determining, at the national level, an appropriate program to be followed in all regions given the diversity of activities engaged in, and the problems faced, by small- scale industry in different areas. CFP, CU and some agencies providing tech- nical assistance are now making special efforts to coordinate and streamline technical assistance to SMI at the regional level (paras. 3.04, 3.17-3.19). 2.06 Technical assistance to Colombian SMI is available from a number of sources.5/ The largest program is that of SENA (Servicio Nacional de Aprendizaje), a Government institute which has vocational training facilities throughout the country and provides free technical assistance services cover- ing the industrial, commercial and services sectors. SENA has recently decentralized its technical assistance to SMI, placing at the regional level the selection of priority SKI subsectors which receive technical assistance. Some 60 SENA professionals are currently involved in providing technical ass-istance to SMIs in different regions, benefiting up to 500-600 firms nationwide each year. 2.07 FICITEC (Fundacion para el Fomento de la Investigacion Cientifica y Tecnologica) is a small, private non-profit institution which assists, on a subsidized basis, some 60 small firms annually; FICITEC also presents semi- nars on topics of interest to small-scale industry. In addition, FICITEC is the Colombian contact for CESO (Canadian Executive Service Overseas), which has been helping about 50 small- and medium-sized firms per year to solve technical and technological problems. IIT (the Government's Instituto de Investigaciones Tecnologicas) provides some technical assistance with regard to production processes, largely to firms in the food and chemical indus- tries; about 10-20 small firms a year have received such assistance. PROEXPO (Fondo de Promocion de Exportaciones) provides technical assistance to both actual and potential exporting firms in the areas of quality control and 5/ For details about institutions providing technical assistance to SKI in Colombia, see Annex 2.  packaging, as well as broader assistance to selected subsectors felt to have good potential for increased exports. Another Government institution, COLCIENCIAS (Fondo Colombiano de Investigaciones Cientificas y Proyectos Especiales), does not itself provide assistance, but has funds (among oLhers, from an Inter-American Development Bank loan) available to finance certai.L kinds of technical assistance. In addition to the above-mentioned national institutions, universities are increasingly establishing small-scale local technical assistance programs, often in conjunetion with one of the national institutions. Private consultants are also available, but are generally perceived by small-scale firms as costly. III. CORPORACION FINANCIERA POPULAR A. Basic Structure 3.01 Gwnership. Corporacion Financiera Popular (CFP) was founded as an autonomous institution in 1967 by Banco Popular, a Government-owned commer- cial bank, to assist in the promotion and development of small-sized produc- tive enterprises6/ by providing medium- and long-term financing. Since commercial banks and private financieras focus mainly on medium and large firms, CFP's role is to develop, and help to provide, a range of services tailored to the characteristics and needs of small-scale industries (SSIs). To achieve these goals, CFP is governed by the -Estatutos- and Operational Guidelines, which have been amended from time to time to reflect the evolu- tion of CFP's lending operations. The present 'Estatutos- were approved in July 1975 and the Operational Guidelines in February 1982 (with some subse- quent modifications) and are expected to guide CFP's operations under the proposed project. In line with previous practice, agreement was reached during negotiations that CFP would not modify its 'Estatutos and Operational Guidelines in any way that would materially and adversely affect its finan- cial condition or operations without the Bank's prior approval. 3.02 Under Loan 1834-CO, the Government, through the Ministry of Economic Development, contributed Col$ 304 million in CFP equity between 1981 and 1983. Direct Government participation, as of December 31, 1983, accounted for over 46% of CFP's total equity (compared to 12% at year-end 1980); PROEXPO holds about 32% of CFP's equity, with Banco Popular and its subsidiary, the Corporacion de Ferias y Exposiciones, holding 16.5% and 4% respectively. The remaining 1.5% was held by the Instituto de Fomento Indus- trial (IFI) (Table 11). In practice, the Ministry of Economic Development, to which both CFP and PROEXPO are administratively attached, has played an important role in guiding CFP's Institutional development and its lending operations. 3.03 Management, staffing and organization. CFP's principal governing body is its eight-member Board of Directors (Table 12), which is headed by the Minister of Economic Development (who normally appoints as his repre- 6/ CFP's main target group consists of small-scale manufacturing enterprises, but it can also lend to small-scale agroindustrial and mining enterprises and cooperatives.  - 8 - sentative a high-ranking official of the Ministry of Economic Development) and which has a broad representation from other Government agencies, finan- cial institutions, and private sector and industrial associations. Over the past years, in line with the discussions between the Bank and CFP under pre- vious Bank loans, the Board has delegated increasing authority to CFP's management in (a) establishing creditworthiness criteria for borrowing; (b) directing supervisory and portfolio management activities; and (c) selecting key management staff. Besides approvals of CFP's lending operations, the Board, Which meets twice a month, has been focusing increasingly on financial, trade, investment, technical assistance and other policy matters relevant for SSIs, dedicating one sessiou each month to those aspects since late 1983. The Board also appointed a four-member Loan Committee which has responsibility for approval of medium-sized loans (currently up to Col$ 25 million, some US$285,000), while the Board approves most of the larger loans itself.7/ 3.04 Since 1976, CFP's 14 regional offices have been governed by Regional Advisory Boards (RABs) which were established to increase CFP's linkage with other Goverament institutions and the business communities in individual regions. The RABs are normally composed of five members and five alternates who, apart from the regional manager of CFP's office, include the manager of the local Banco Popular and PROEXPO office, a local representative of the Colombia Association of Small Manufacturers (ACOPI), and representa- tives of SENA, Chamber of Commerce or local industrialists and bankers. Their first-hand knowledge of the local environment and entrepreneurs has enabled RABs to be quite effective in their role of reviewing and approving credit applications. Currently, their approval limits are Col$ 6 million for the four larger regional offices (Bogota, Medellin, Cali and Barranquilla), Col$ 4 million for the three medium-sized offices (Bucaramanga, Pereira, and Armenia) and ColS 3 million for the rest. Also, the RABs' close contact with small-scale entrepreneurs helps CFP's management to receive feedback from clients on the quality of services and on SSIs' financial and technical assistance needs. The RABs, which have already improved coordination among regional CFP offices and other local entities assisting the SSIs, are ex- pected to play an important role in the development of the regional SSI technical assistance programs (para. 3.18). 3.05 During 1979-1983, CFP's top management underwent frequent changes. A general manager, who had been appointed in September 1978, moved the insti- tution toward greater cost-consciousness and profitability, halted staff turnover and improved staff morale. However, because of differences of opinion with the Minister of Economic Development, he resigned in October 1981 and another manager was appointed in December 1981. Differences of opinion soon emerged between the new general manager, who espoused an ambi- tious program of CFP involvement in industrial parks, and staff at different levels, affecting the institution's financial performance in 1982. By early 1983, however, another general manager was appointed who has succeeded in 7/ The maximum loan amount that CFP can theoretically approve to one client is limited to 7% of CFP's paid-in capital and reserves (currently limiting one single loan to some Col$ 70 million, some US$790,000 equivalent).  -9- establishing good working relations with CFP's submanagers and improving the overall working environment. 3.06 CFP's institutional capabilities have improved over the past years, in part as a result of previous Bank loans. Also, CFP's staff has developed adequate knowledge, skills, and experience to carry out their day-to-day operations effectively. However, strengthening is still required regarding institutional strategy, portfolio management, supervision activities and coordination of the technical assistance offered to SSIs. Improved salaries and a policy of promotion from within have resulted in most department man- agers having at least four years of experience with CFP. The institution has consolidated responsibilities under three major areas focusing on financial management aad programing, lending operations, and general administration including personnel management. Since 1975, CFP maintained the 14 regional offices to support the Governmentts industrial decentralization policy, adding three promotional offices (Popayan, Florencia and Villavicencio) in 1983 (Table 13). CFP believes that this network of regional offices is adequate to enable it to serve the majority of actual and potential SSI clients economically. However, while CFP's overall operating efficiency has continued to improve and its total staff has been relatively stable in the past four-five years, too high a proportion of staff is still concentrated in CFP's headquarters (general management, supporting services and operational control). In 1983, 41X of the total staff were at headquarters (Table 14), while staff constraints in some regional offices were interfering with CFP's lending operations. CFP's management has recognized the need to achieve a better balance between headquarters and regional office staffing. Agreement between the Bank and CFP was reached during negotiations on general guidelines for a CFP reorganization study and on 2n action plan to strengthen the operating capabilities of its regional offices (para. 3.26), both of which are part of the overall 1984-1987 Action Program (para. 3.33). B. Systems and Procedures 3.07 Loan approval and supervision. The Bank has been providing consid- erable support to CFP regarding its subproject appraisal and supervision systems, monitoring of regional offices, internal auditing, and financial management. Also, a UNDP technical assistance program was carried out between July 1978 and March 19818/. Appraisal quality (which had varied during 1976-1978 because of staff turnover) has improved since 1978, thanks in part to the preparation of a credit manual and some training of CFP. analysts; the thrust of subproject appraisal changed gradually from an exces- sive emphasis on the history of the enterprise to the viability of the pro- posed subproject. Further improvement, mainly regarding the market and repayment capacity analysis of the subprojects and/or enterprises, is needed. 8/ The UNDP-financed and UNIDO-executed technical assistance program was designed to assist CFP to strengthen its operational capabilities, providing experts in cooperatives, marketing, subproject appraisal and supervision, financial systems, industrial parks and information systems. However, because of problems during the implementation stage, the end result of the program was modest compared to initial expectations.  - 10 - 3.08 Between 1978 and 1981, CFP gradually improved its subproject super- vision system by putting more emphasis on enterprise, subproject and subloan supervision activities, defining frequency and scope of supervision visits9/ and making regional offices responsible for supervisory work. Since 1982, however, the organization, focus, and quality of supervision activities have deteriorated and the number of supervision visits declined from 52% of CFP's total clients in 1981 to only 13% in the first half of 1983 (Table 15). This was partly because of CFP's focus, during 1982, on lending to industrial parks and partly because of staff constraints in regional offices and lack of headquarters' control over regional managers' supervisory activities. Inade- quate supervision by CFP has been identified as one of the reasons for grow- ing arrears in its portfolio during 1983 (para. 3.27). By October 1983, CFP's management had taken action, hiring six professionals and reassigning eight additional staff to improve its superr-'sion work. By year-end 1983, 41% of CFP's clients had been visited and supervised. In addition, during the second half of 1933, Regional Supervisory Committees were established and given responsibilities for monthly reviews of supervision work and recom- mendations of action to be taken. During project appraisal, insufficient and inadequate communications were found to exist between two currently special- ized types of operational professionals, i.e., analysts and supervisors. It was, therefore, agreed that CFP would take steps to assemble a unified pro- fessional staff to deal with CFP's lending operations (subproject promotion, preparation, appraisal and supervision) and help to correct the weaknesses in this area. The proposed steps are included in CFP's overall 1984-1987 Action Program (para. 3.33). 3.09 Information system. Due, in part, to the impact of Bank reporting requirements established under the three previous loans, the quality and scope of statistical information on CFP's operations has improved substan- tially. Nevertheless, while ample information exists on subprojects' finan- cial situation, status and results, both ex ante and from supervision reports, it has not yet been processed and used systematically for evaluating CFP's impact or for providing some general patterns for subproject apprai- sal. In addition, analytical feedback between headquarters and regional offices, as well as use of information flows for managerial purposes, needs to be developed further. CFP is currently preparing a study aimed at stren- gthening and broadening its electronic data processing system, taking into account different types of information (for both managerial and statistical purposes) available and needed for its operations. The study forms part of CFP's overall 1984-1987 Action Program (para. 3.33). Under the proposed loan, financing would be made available for the purchase of computer terminals for regional offices to complement the data processing equipment financed under the third loan, as appropriate (para. 4.16). 3.10 CFP's internal auditing procedures are adequate, and its annual accounts are reviewed by Board-appointed independent auditors. After frequent changes of CFP's external auditors during its earlier years, the 9/ Three types of supervision activities were introduced: (i) control of investment (following the last disbursment of stibloan); (ii) followup on individual subprojects (at least once a year), and (iii) supervision of subloans in arrears.  - 11 - same auditing firm has been used consecutively since 1976. The external audit reports called for under Bank Project Agreements have been acceptable and have contained unqualified opinions. However, there have been delays of about two-three months in their submission to the Bank. CFP has stated that it would submit the 1983 audit report to the Bank by June 30, 1984, as required by the Project Agreement. C. Operations 3.11 Lending operations. In line with its statutory objectives and lending policies, CFP has focused primarily on lending to smaller industrial firms without alternative sources of term credit. During 1979-1983 (first semester), an average 70% of CFP loans (accounting for some 38% of total lending) went to firms with less than Col$ 5 million (about US$90,000 equivalent) in total assets (Table 16). The number of subprojects financed grew by 59% between 1979 and 1982, when 3,023 subloans were approved, falling to 3,009 in 1983. The average subloan size during the period was Col$ 1.1 million (about US$20,000 equivalent). The food, beverages and tobacco, apparel and footwear, wood and furniture, paper and printing, and chemicals, petroleum and coal derivatives subsectors accounted for almost two-thirds of CFP's lending. The average maturity of CFP's subloans fell from 3.3 years in 1979 to 2.7 years in the first half of 1983, mainly because of a growing share of working capital lending in CFP's total portfolio (61% in 1979, 77% in 1983).10/ In compliance with its objective of lending to a broad and regionally well diversified range of SSIs, CFP increased lending outside the three major departmen,s; of Cundinamarca, Antioquia, and Valle del Cauca from 50% of its total lending in 1980 to 56.5% in 1982. New clients, including newly created firms as well as first-time borrowers, accounted for about 14-19% of the total number of enterprises financed. 3.12 The recovery of funds under previous Bank loans, other external sources and local financing contributions connected to these loans allowed CFP to develop its )wn specialized lending programs. Apart from free-stand- ing working capital financing, which CFP offers to SSIs out of its own funds (accounting for some 13-15% of CFP's total lending in the past two years), a Special Artisan Credit Program was introduced in 1978, which finances mostly working capital requirements of very small labor-intensive firms. During 1982, Col$ 28 million were lent under the Artisan Credit Program. Also, under the third loan (1834-CO), US$1.7 million equivalent were used for this purpose. 3.13 In light of the difficult financial situation faced by many of its clients in 1983, and in order to improve its portfolio quality (para. 3.28), CFP introduced a special program to assist those SSI clients with liquidity problems whose operations were found to be viable. Given the program's recent initiation, data on its implementation and impact are not available. However, CFP is currently analyzing its experience under this program. 10/ As a result of the deteriorating economic situation in Colombia during that period, SSIs were requiring more working capital rather than fixed asset financing. Uader the third Bank loan, average maturity of subloans was 4.4 years, with only 12.5% of total lending for permanent working capital (Table 17).  - 12 - 3.14 Micro-enterprise financing. Another innovation in the 1979-1983 period was CFP's lending to very small industrial firms in the informal sector of the economy, micro-enterprises (ME). This new approach is a com- prehensive program, including (a) initial training in general administrative aspects and entrepreneurial development; (b) additional training and tech- nical assistance, in accordance with the characteristics and/or needs of the enterprises; and (c) financial assistance. Once the entrepreneurs are enlisted in the program, they have to go through the initial training in order to have access to the other two parts of the package. The program, which focuses upon enterprises with less than 15 employees and total assets not exceeding 200 minimal monthly salaries (currently equivalent to some US$22,500), was initiated by Fundacion Carvajal, a private voluntary organi- zation funded by the Carvajal industrial group in Cali, in 1977. Currently, there are some 12 private foundations (Fundaciones) established in various regions which have, to date, trained and assisted an estimated 7,500 micro- entrepreneurs, of whom some 1,500 also have obtained loans. 3.15 Since 1982, CFP has been providing financing to the micro-enter- prises, mainly through the Fundaciones, earmarking Col$ 70 million of own funds for this program in 1983 (and an additional Col$ 100 million for CFP's direct lending to MEs and artisans). The Fundaciones bear the credit risk, and some administer the loans and keep a financial spread of 6 percentage points on CYP's funds. The training and technical assistance, which are provided by the Fundacionest staff and some small industrialists, are financed almost entirely by the Fundaciones.11/ Recently, SENA also, in coordination with the Fundaciones, has been providing technical assistance to MEs. As yet, firm conclusions cannot be drawn on the merits of the ME program, although the initial indications (including the very low - US$1,500 cost per job created by MEs) appear promising. One possible benefit of this approach is that it provides a way to create new small enterprises that start their existence with the basic administrative tools, now often lacking in Colombian SSIs. Currently, the Government is initiating a nationwide ME program with Inter-American Development Bank (IDB) financing, which would include participation of both private Fundaciones and Government agencies such as SENA and CFP. Given the high administrative costs involved (the average subloan size to MEs is some US$1,500 equivalent), CFP should limit its direct lending to MEs in the informal sector (to not more than 2% of CFP's annual lending during 1984-1987) and concentrate its main assistance to MEs in continuing to finance, through Fundaciones or other financial institutions, which should assume the credit risk and/or administrative costs of lending activities, portfolio management and supervision. Agreement between the Bank and the Government/BR/CFP on this matter was obtained during loan negotiations. 3.16 Special Government programs carried out by CFP. In addition to the above-described special programs, CFP is also carrying out two special Gov- ernment programs which are only partially related to CFP's principal target 11/ Under current arrangements, entrepreneurs are charged a fee which covers some 10-15% of the cost of the training provided. The bulk of the Fundaciones' operating costs are financed with grants from larger private enterprises.  - 13 - group (with funds provided by the Government): (a) "Programa de Integracion Social de la Presidencia de la Republica," offering financing to SSIs in so-called Neighborhood Development Centers in several secondary cities in Colombia (funded in part by Bank Loan 1558-CO); and (b) "Programa de Rehabi- litacion," which provides investment credit to recently amnestied political activists. Col$ 114 million and Col$ 50 million, respectively, were assigned to the two programs for 1984, and CFP assumes their administrative costs. During negotiations, the Bank and the Government/CFP reached agreement that: (a) CFP would maintain, on behalf of the Government, separate financial records and accounts for the Government special programs, and would undertake such programs only if it is assured of adequate and timely compensation to cover the corresponding administrative costs to them (this arrangement should be introduced and implemented starting in 1985); and (b) lending for Government special programs would not exceed 2.5% of CFP's total lending for each year during 1984-1987. 3.17 Technical assistance to SSIs. From 1978, when CFP discontinued the bulk of its direct technical assistance activities, replacing them with systematic supervision of its clients, CFP's focus has been on (a) supervi- sion and referral services to other more specialized sources of technical assistance, liriting its assistance mainly to the identification of clients' technical assistance needs; (b) coordinating, with SENA and other institu- tions, their provision of direct technical assistance and training to SSIs; (c) lending to SSIs to finance technical assistance needs;12/ and (d) sup- plying some technical information. This approach reflected CFP's experience that effective direct technical assistance to a large number of individual firms required more staff time and expertise than CFP could afford to allocate.13/ The results of this approach, which was adopted under the third project, were modest compared with expectations, mostly because of the lack of a clearly planned technical assistance program and because of CFP's orientation to coordinate, at the national level, the activities of special- ized institutions providing technical assistance services to SSIs, for which it has limited institutional capabilities and resources. 3.18 In line with the findings of the Project Completion Report (PCR) for the second project and in view of the preceding comments, CFP's future focus is expected to be on coordinating technical assistance (TA) for SSIs at the regional level (where its relationship with the institutions providing TA and its ties with SSIs are stronger than at the national level) in addition to any other effort at the national level. CFP presented for discussion, during negotiations, three TA programs at the regional level, which form part of CFP's overall 1984-1987 Act'on Program (para. 3.33) and which would start 12/ Between 1979 and 1983, CFP approved 62 technical assistance subloans (0.6% of the total number of subloans) for a total of Col$ 25 million. 13/ CFP's technical assistance advisors, who are also responsible for subproject and enterprise supervision, do not have the qualifications or experience to perform in-depth consultancy on specific problems of SSIs, whereas there are now several public and private institutions (SENA, FICITEC, universities and other entities- paras. 2.06 and 2.07) and consultants who can provide such assistance.  - 14 - to be implemented during the first year of the project. The programs were prepared for a period of at least one year, by a working group comprising the TA agencies operating in the corresponding region, with CFP coordinating the effort. The regional TA programs spell out: (a) the technical assistance needs of SSIs in the region and corresponding priorities; (b) the target groups for TA; (c) the supply of TA services in the region; (d) the type of TA services to be provided under the program by each agency; (e) the terms and conditions of the TA; (f) the modus operandi of the working group/TA program, including the coordination aspects; (g) the corresponding followup system; and (h) the type of agreements between CFP and the TA agencies at the regional level. CFP also stated, during negotiations, that it expects to initiate the above-mentioned coordination in the other regions in which it is operating during 1985. 3.19 The efforts at the regional level would be complemented with CFP's efforts to coordinate TA at the national level (as under the third loan) through continued collaboration between CFP and the institutions providing TA to SSIs. In this regard, CF? presented a TA plan at the national level which was discussed during negotiations. CFP is including, in this plan, the required TA for CFP to coordinate and monitor the regional programs. The results of the three regional and national TA programs would be reviewed, and appropriate modifications, consistent with the review, would be introduced within 18 months of the programs' inception. Additionally, CFP should strengthen its TA unit at headquarters to facilitate appropriate coordination of the TA activities, in accordance with the organization study to be carried out by CFP (para. 3.33). Agreement between the Bank and the Government/CFP was reached on the above matter during negotiations. D. Resources 3.20 CFP's overall resource mobilization during 1979-1983 was adequate, in quantitative terms, to support the credit demand (Table 20), although it experienced some fluctuations in resources available from different sources. Inflow of funds grew some 53% in re.l terms between 1980 and 1982, but fell by 6.3% in 1983. While domestic borrowings grew more than 4.5 times between 1979 and 1983, their share as a percentage of CFP's total liabilities and net worth fell from 38% in 1979 to 33.6% at the end of 1983. Despite a Col$ 304 million increase in paid-in capital, the share of CFP's net worth was reduced from 28.5% of total liabilities and net worth in 1979 to 24.2% at year-end 1983. CFP's foreign sources grew 4.7 times during 1979-1983, increasing the institution's foreign outstanding debt/total assets ratio from 27.4% to 38.2%. This result reflects the fact that foreign resources have consider- ably longer maturities than domestic funds. Taken as a flow, CFP's foreign resources represented, on average, 16% of the total inflow of funds during 1979-1983, 92% of which were Bank funds (the remaining 8% came from KfW). Under the third loan, CFP's indebtedness to the Bank grew from 24.6% at the end of 1980 to 33.4% at the end of 1983, but, taken as a flow, Bank resources represented only 14.8% of total resources used by CFP during the period. 3.21 One of the objectives of the previous Bank loans was to assist CFP in building up a diversified and sound resource base for its operations. Nevertheless, CFP remains dependent upon three main sources of financing: FF1, PROEXPO, and the Bank. Taken as a flow, those sources together  - 15 - accounted for almost 90X of CFP's total borrowings and some 44% of its total sources during 1979-1983 (Table 21). FFI has been the most important seurce of local borrowing (45X), providing funds on terms and conditions suitable for the bulk of CFP's clients. The remaining local borrowings were mainly from PROEXPO resources (43Z of total domestic borrowing), which were used predominantly for short-term export financing to larger-sized SMIs (i.e., with total assets exceeding Col$ 60 million). 3.22 While, on the whole, CFP has fared adequately in terms of obtaining resources for lending, it has become excessively dependent upon a few sources (para. 3.21). CFP intends to seek non-traditional permanent sources of local funds (possibly through mobilization of passbook savings resources) to com- plement the funds obtained from FFI and PROEXPO. Also, based upon CFP's expected lending operations (para. 3.30), incremental funds in the form of paid-in capital will be required in the 1984-1987 period. At negotiations, agreement was reached between the Bank and the Government on the need of additional equity contributions to allow for the growth in CFP's lending operations and to maintain the debt/equity ratio at acceptable levels (para. 3.32). E. Financial Position, Results and Projections 3.23 Asset growth, portfolio growth and profitability. Despite some fluctuations in its resources, CFP's total assets and portfolio experienced a steady growth between 1979 and 1983. Total assets increased from Col$ 2,134 million at the end of 1979 to an estimated Col$ 7,350 million at the end of 1983, a real growth of over 10% p.a. CFP's portfolio experienced dynamic growth between 1980 and 1982 (more than 12% p.a. in real terms) and then slowed down in 1983 (a 4.7% real increase), reaching an estimated Col$ 6,420 million at year end 1983 compared to Col$ 1,841 million at the end of 1979 (Table 20). Equity increased by about Col$ 675 million, of which Col$ 304 million were paid-in capital as required under the previous two Bank loans. CFP's debt/equity ratio rose from 2.5:1 in 1979 to 4.8:1 in 1982, but then dropped to 4.1:1 by year-end 1983 as a result of the capital provided by the Government. Recent levels have continued to be substantially below the maximum debt/equity ratio of 6.0:1, as spelled out by the Project Agreement for Loan 1834-CO. 3.24 Since 1979, both CFP's gross income and financial expenses have increased faster than its asset growth. However, while income growth only modestly surpassed that of CFP's assets (10.7Z p.a. in real terms compared to 10.3%), financial expenses almost doubled in real terms in four years (a 19% real annual growth) and CFP's financial spread decreased from 10.8% of total assets in 1979 to 8.9% in 1982, indicating that its on-lending rates did not increase as much as the cost of its financial resources. Consequently, CPP's profits as a percentage of total assets fell from 2.1% in 1979 to 0.8% in 1983 (Tables 18 and 19). Also, profits did not compensate for the erosion of  - 16 - CFP's eauity by inflation during 1979-1983, and the profit/equity ratio declined from 7.4% in 1979 to 4.5% in 1983.14/ 3.25 CFP's administrative costs. The level of CFP's administrative costs has been a major concern of the Bank and CFP's management. Administra- tive costs have been somewhat high in relative terms. Although CFP managed to reduce its administrative costs/average assets ratio from 7.3Z in 1979 to 6.7% in 1983, the latter was higher than the 6.0% ratio required for 1983 under the third project (Table 25). The higher-than-anticipated administra- tive costs/assets ratio has been due in part to: (a) slower-than-expected growth in CFP's portfolio and (b) the high relative share of administrative costs of headquarters staff (43% of total staff and 49% of total administra- tive costs in 1982) as compared to costs in regional offices where CFP's income and p_ofits are generated. 3.26 As a result of CFP's expected incremental lending, strengthened supervision, technical assistance coordination and staff training activities, CFP's administrative costs must be expected to increase in absolute terms in the future.L5/ In order to control those costs _nd to improve CFP's operating efficiency and profitability, CFP is currently preparing an action plan to ensure a sound portfolio growth at lower unit cost. This plan would focus on reviewing CFP's organization with the view of strengthening the operating capabilities of its tegional offices, which are CFP's primary lend- ing units. This would be achieved by gradually converting the regional offices into profit centers and grouping them into different categories in terms of size, number of operations and approval limits. In this connection, CFP would review regional offices' operating policies, organization and staffing, the composition of their Regional Advisory Boards and their authority of subloan approval and would initiate the implementation of any proposed modifications required prior to September 30, 1985. The action plan would also identify the needs for modification and streamlining of CFP head- quarters" organizational units, to be implemented not later than Septem- Ler 30, 1985. As a control mechanism, an administrative costs/average assets ratio of no more than 6.7% for 1984, 6.5Z for 1985 and 6.0% for 1986 and onward has been set. Agreement between the Bank and CFP on the action plan and target levels for CFP's administrative costs/average assets ratio, which form part of CFP's overall 1984-1987 Action Program (para. 3.33), was obtained during negotiations. 3.27 Portfolio quality. Another area of concera of the Bank and CFP has been the quality of CFP's portfolio. While, from 1975 to 1982, notable achievements had been made and portfolio affected by arrears fell steadily 14/ Although established by CFP as a long-term goal, it is not realistic to expect CFP to make sufficient profits to compensate for erosion of its equity through inflation, especially given its developmental objectives. 15/ The higher administrative costs must be considered as the price of operating an institution supporting only small-scale industries and, at the same time, using adequate standards of subproject appraisal, supervision and technical assistance.  - 17 - (from 20% of total portfolio in 1975 to 12% in 1982), serious deterioration in portfolio has occurred since early 1983. As of December 31, 1983, arrears stood at 17.6% of CFP's total portfolio (Table 24). In addition to the over- all difficult economic situation and consequent financial weakness of the SSI subsector, the following main factors causing arrears have been identified: (a) extended payment schedules from SSI clients, coupled with prompt payments required from SSIs by their suppliers, and difficulties and delays in obtain- ing credit for such working capital; (b) delays in completion of investment projects; (c) inadequate and limited portfolio supervision activities by CFP in 1982 and early 1983; and (d) CFP's lack of firmness in collecting overdue subloans, divided responsibility regarding portfolio administration and slow subloan collection and legal procedures. 3.28 CFP took action in the second half of 1983 to improve its portfolio quality. As initial steps to solve the problem, it has (a) established a special refinancing program (para. 3.13); (b) started to strengthen its sub- project supervision activities by hiring additional supervisors, increasing the number of supervision visits and intensifying their scope, and setting up regional supervision committees (para 3.08); and (c) given specific respon- sibilities to regional managers to improve their respective portfolios. The Bank has asked CFP also to prepare a portfolio recovery plan which should include: (a) recovery targets; (b) collection policies and procedures; (c) a supervision system including the use of the regional supervision committees; (d) clear and concrete assignments of portfolio administration responsibili- ties; (e) the corresponding operating/organizational adjustments; and (f) a review of the above-mentioned refinancing program with regard to expected duration, funding requirements, access terms and results obtained. Agreement between the Bank and CFP on the portfolio recovery plan, which forms part of CFP's overall 1984-1987 Action Program (para. 3.33), was reached during negotiations. 3*29 Financial projections. Under previous Bank loans, CFP's financial projections have improved substantially in quality and scope and have become an important planning tool in balancing operational targets with resource availabilitylb/. During appraisal of the proposed project, CFP prepared detailed firnancial projections (Table 31) that assume: (a) disbursement of the proposed Bank loan over three years; (b) a 42% p.a. nominal increase (some 18% p.a. in real terms) in own resources; (c) a 36% p.a. nominal growth (13% in real terms) in withdrawals from FFI funds; and (d) a 38% p.a. nominal growth (15Z in real terms) in withdrawals from PROEXPO funds. 3.30 With the expected economic recovery and a projected average SMI subsector growth of some 4% p.a. during 1984-1987, which is acceptable compared to historical growth rates (para. 1.05), and given CFP's target of increasing the coverage of SKI lending from an estimated 16% of total SMI financing in 1983 to 20Z in 1987 (an estimated 40% of total credit to SIs), CFP projects its annual lending to rise from Col$ 4,340 million in 1982 to Col$ 13,500 million in 1987, at a 40% p.a. nominal growth, or about 16% p.a. in real terms (Table 31). This lending program appears somewhat ambitious, 16/ Total assets, income and expenses projections, made under the third loan, have been within 10% of the actual results achieved.  - 18 - surpassing historical growth rates (e.g., 12.5% p.a. in real terms in 1979-1982). It may, nevertheless, be achieved if the SMI subsector recovers as expected and provided that CFP strengthens its regional offices, allowing it to finance a larger number of clients than in the past. 3.31 Assuming that the lending growth mentioned in paragraph 3.30 can be achieved, the nominal growth rate in lending will mean a nominal asset growth rate of some 35Z p.a. (some 13X p.a. in real terms), from Col$ 7,350 million in 1983 to Col$ 24,206 million in 1987. More than two-fifths of CFP's asset growth is expected to be financed by external sources of credit, about one- third from local credit lines, some 10Z from increases in net worth and the remaining 15% from increases in other liabilities. FFI is expected to account for some 40% of the funds needed for CFP's 1984-1987 lending program, followed by the Bank with 17% and PROEXPO with 16% of total funds. CFP's own funds--which include loan recovery, increase in paid-in capital, retained earnings and mobilization of savings-are expected to represent about 20% of CFP's total lending during 1984-1987. CFP's projections show that income should grow 37% p.a. in nominal terms during 1984-1987, total expenses by some 36.5% p.a. and administrative expenses by 31% p.a. Net profits are expected to grow three times between 1983 and 1987 and would represen!. between 4.5 and 9% of average equity (Table 34). 3.32 Given the preceding lending program and the projected sources of funds to support it, CFP's debt/equity ratio is expected to increase from 4.4:1 in 1983 to over 9.0:1 by 1987 (Table 32). This ratio is considered too high for the level of CFP's portfolio in arrears and its relatively low level of profits. A considerable increase in CFP's net worth would, therefore, be needed to maintain CFP's debt/equity ratio at an acceptable level, similar to that under the previous Bank loans. Agreement was reached during negotia- tions between the Bank and the Government/CFP that: (a) CFP's maximum debt/ equity ratio would be maintained at 6:1; and (b) the Government would make arrangements to provide CFP, annually, with additional budgetary resources and/or cause CFP to be provided with funds from other sources to increase CFP's paid-in capital to levels consistent with the above-mentioned ratio and the lending targets specified in CFP's Action Program (para. 3.33); the exact amounts of the annual budgetary allocation would be determined by the Government by September 30 of each year, after prior consultation on CFP's projected needs with the Bank, BR and CFP. 3.33 Action Program. CFP has prepared an Action Program to strengthen its institutional capabilities, which would be carried out during 1984-1987. The Action Program deals in a concrete way with areas that need strengthen- ing, which have been identified mainly by the PCR of the second project and during supervision of the third project. The Action Program includes: (a) guidelines for a CFP reorganization study and for an action plan to strengthen its regional offices (paras. 3.06 and 3.26); (b) a plan to strengthen its subproject appraisal and supervision systems (para. 3.08) and information system (para. 3.09); (c) technical assistance programs at the regional and national levels (paras. 3.18 and 3.19); (d) guidelines to carry out a study to mobilize resources from the market (paras. 3.22 and 4.12); Ce) a plan to ensure a portfolio growth at a lower unit cost (para. 3.26); and (f) a portfolio recovery plan (para. 3.28). The Action Program also includes CFP's projected 1984-1987 lending targets. Agreement between the Bank and  - 19 - CPP was obtained during negotiations on the Action Program as a whole aAd on the individual components, as well as on the schedule, to carry out the corresponding actions. As a condition for effectiveness of the Loan, CFP *hould present the Action Program, in terms satisfactory to the Bank, approved by its Board of Directors. IV. THE PROJECT A. Experience Under Previous Bank Prolects 4.01 Bank lending to support the Colombian industrial sector has been wall diversified, encompassing eight loans totaling US$493.0 million to finance medium and large industry through private development finance compa- nies (DFCs); three loans totaling US$52.5 million to support SSI; a loan component of US$19.0 million for agroindustry; a loan of US$15.0 million for an export processing zone in Cartagena; a loan component of US$2.0 million for very suall enterprises in target areas of the Intermediate Clties Urban Project; and a loan of US$80 million for the Cerromatoso nickel project. The loans have helped to support a balanced Industrial development strategy, including assistance to smaller, labor-intensive enterprises in areas where Colombia's traditionally abundant, relatively low-cost labor gives these enterprises a competitive edge, as wall as to more capital-intensive enter- prises In product lines with significant economies of scale. Past Back support to SSI, while still small in comparison to total industrial lending, bas become a significant factor in the development of SSI in Colombia. 4.02 Bank funds have been vital in providing long-term financing through CFP for fixed assets to the SSI subsector. Loan 1071-CO, approved in January 1975, for US$5.5 million equivalent, was the Bank's first SSI operation in Latin America. The second, loan 1451-CO for US$15.0 million, was approved in June 1977. Those two loans have been fully disbursed. The third, Loan 1834-CO for US$32.0 million, approved for CFP in April 1980, is now fully committed and over 90% disbursed. 4.03 In total, CFP has financed, through the above-mentioned loans, about 2,300 subprojects with an aggregate investment value of some US$130 million (i.e., 2.5 times the funds lent by the Bank). The average subproject Investment cost has been US$56,000 equivalent and the average subloan size with Bank loan resources, about US$22,800, with only about 70 subloans being for more than US$100,000 equivalent. About two-thirds of the subloans have been for five years or less, indicating some reluctance on the part of SSIs to incur longer-term debt, and, at times, a CFP tendency to under-estimate subprojects' payback periods. About three-fourths of the subborrowers had total assets below US$100,000 equivalent. The geographical distribution of subloans has been fairly even, with 67Z of the subloans and 56Z of the subloan amounts going to firms located outside the metropolitan areas of Bogota, Medellin and Cali. SSI distribution has also been spread evenly across subsectors, with only one subsector (food and beverages) accounting for 20% of the subloan amounts, and two subsectors (printing and chemicals) for 11Z each. Finrs with registered exports accounted for only 2.3% of total firms financed. New enterprises represented 15% of the total. About 87% of the amount of the subloans financed with loan resources were for fixed assets  - 20 - and about 13% for working capital. The average investment cost per job created has been about US$6,900, US$13,500 and US$15,000 (in current US dollars) with about 2,100, 3,300 and 4,600 new direct jobs generated uader the first, second and third projects, respectively, compared with US$45,000 for medium and large industry. Smaller and new firms had a lower average cost per job. A project completion report (PCR) for the second project (Loan 1451-CO), distributed on Hay 10, 1983, includes a detailed discussion of the economic impact of Bank-financed subprojects. Tables 27, 29 and 30 of Annex 3 also show the impact of subprojects financed under the third project (Loan 1834-CO). 4.04 The main conclusions from the experience with the previous SSI projects, as mentioned in the OED report for the first project, in the PCR for the second project, and in Bank supervision reports for the third project, can be summarized as follows: (a) the Government, through CFP, has succeeded in increasing term lend- ing to SSIs, in achieving a satisfactory economic impact at regional level, in gradually developing CFP's lending capabilities, and in introducing a system for supervising CFP's clients; (b) CFP's performance in holding administrative costs down has been only partially satisfactory; during implementatioa of the third project, it became apparent that, to achieve better administrative costs/assets ratio, CFP should focus its attention on obtaining a sound portfolio growth at a lesser unit cost; thus, increasing the role of the regional offices in the coming years seems necessary to reach the projected growth of CFP's lending activities at a lower lending cost; Cc) CFP's profitability has improved gradually, but it remains inade- quate to enable CFP to maintain the real value of its equity. Hoe- ever, when account is taken of CFP's focus on the smaller SSIs and of its developmental activities in support of the subsector, for which it is not remunerated, CFP's financial performance can be judged to be reasonably satisfactory; (d) CFP's approach of concentrating on coordinating the technical assistance services provided to SSIs by other agencies, rather than on providing such assistance directly, was introduced under the third loan; while such an approach was viewed as a positive step toward future development of the SSI subsector, the results have not met the expectations, in part because of CFP's limited ability to influence the technical assistance activities, at the national level, of other institutions such as SENA, which has broader objectives than just serving SSIs; (e) CFP's resource mobilization has been adequate in quantitative terms. However, its continuing dependency upon the Government and public sector agencies for domestic resources has, in the past, led to significant fluctuations in resource availability. Also, Bank resources have become increasingly important as a source of foreign funds to CFP; and wr  - 21 - {f) the projects' contribution in inducing the Government to establish a policy framework for the SSI subsector development has been limited. B. Project Objectives 4.05 In line with sectoral objectives, the proposed project would build and expand upon the accomplishments of the first three SSI projects, taking Into account the experience gained through them. Four specific project objectives merit special attention: (a) contributing to provide the SSI subsector with an adequate amount of mediumr- and long-term investment credit; (b) helping to strengthen the quality and delivery of technical assistance services to SSIs; (c) strengthening institutional, technical and operating capabili- ties of CFP, particularly by its regional offices; and (d) assisting the Governmentts efforts in designing and coordinat- ing policies affecting the SSI subsector development. 4.06 Bank involvement in the proposed project is important because it would provide effective support to the objectives mentioned, including the introduction of a more focused and practical approach to the design of tech- nical assistance to SSIs and a market-related CFP onlending interest rate system, as weli as a continuation of the dialogue with the Government on policies affecting the SMI subsector. Bank involvement is also important because CFE (Colombia's only financial intermediary specialized in term lend- ing to SSIs) has hardly any alternative sources of long-term foreign exchange. 4.07 The possibility of having the proposed loan structured as a two- tier mechanism, using, in addition to CFP, DFCs and/or commercial banks for SSI financing, has also been analyzed. However, it was concluded that, currently, the only practical way for channeling loan resources to SSIs in Colombia is through CFP because of: (a) the recently initiated financial sector reform which appears to support an increased degree of specialization of commercial banks into short-term lending institutions and DFCs into medium- and long-term lenders; (b) DFC's limited network of branch offices, particularly outside the main metropolitan areas; (c) DFCs' and commercial banks' reluctance to finance smaller SSIs (which account for a large majority of firms expected to be financed under the proposed loan) because of higher administrative costs and risks involved; and (d) the fact that the inclusion of other finaancial intermediaries under the Bank loan, notably Caja Agraria and Caja Social de Ahorros, would not be feasible because of the former's discontinuation of lending to SSIs (Annex 2, para. 4) and the latter's li-ited size and experience in SSI lending.  - 22 - C. Project Cost and Financing 4.08 Based on expected demand for CFP lending related to Bank-financed subloans during September 1984-September 1987, the estimated total project cost would amount to US$114.0 miliion equivalent (at a projected Col$ 130 to US$1 average exchange rate for 1984-1987). Table 4.1 following presents the estimated project costs and financing plan. Since FFI credit for working capital would complement Bank resources, it is expected that Bank participation in the total project cost would be lower than in the third project (35% compared to 54%). Local sources (FFI, commercial banks, suppliers, extra bank market) would finance about 34% (of which about two-thirds are expected to come from FFI). Enterprises' own funds, mainly from capitalization and internally generated resources, would finance about 27% of total project cost (about the same as under the third project). Taking into account FFI lending and current working capital needs of SSIs, working capital financing is expected to equal about 44% of total project cost (including up to US$5.0 million from the proposed loan for permanent working capital). As a result of CFP's policy to limit its lending for industrial construction as a share of its total portfolio, less than 10% of the financing under the project is expected to be used for such type of investment. Under the proposed project's financing structure, Bank funds would provide 55% of the fixed assets financing required. Out of total project cost, 65% would come from local sources. Table 4.1: ESTIMATED PROJECT COST AND FINANCING PLAN (in US$ equivalent) Uses Fixed Working Technical Sources Assets Capital Assistance Total % World Bank 34.6 a/ 5.0 b/ 0.4 40.0 35 CFP Own Resources - 3.8 0.7 4.5 4 Other Local Sources c/ 15.7 23.3 - 39.0 34 Firms' Own Resources 12.0 18.5 - 30.5 27 Total 62.3 50.6 1.1 114.0 Z 55 44 1 100 100 a/ Includes about US$0.1 million capitalized front-end-fee. b/ Permanent working capital. c/ Includes CFP credit with FFI resources. 4.09 Taking into account (a) CFP's historical credit demand and projected 1984-1987 lending program (paras. 3.23 and 3.30) and (b) CFP's lending/operational capabilities under the previous three projects, a US$40.0 million Bank loan is proposed for the fourth SSI project. The proposed loan is expected to cover the equivalent of the foreign exchange costs of CFP's total lending program during the above-mentioned period. This level of Bank  - 23 - assistance is also determined by its concern not to increase further the relative share of Bank resources in CFP's total funding (para. 3.20). The proposed loan would be made to BR with the guarantee of the Republic of Calrbia. Banco de la Republica would bear the interest risk on the loan and the responsibility for the foreign exchange risk, including the cross-currency risk. BR would charge any net losses as a result of covering the foreign exchange risk, to the Government's Foreign Exchange Special Account administered by BR (in accordance with Decree #73 of January 13, 1983 and Agreement between the Government and BR dated April 25, 1983). BR would relend the loan proceeds in Colombian pesos to CFP under terms and conditions discussed in paragraphs 4.10 to 4.16. The loan would consist of a US$39.6 million component to assist CFP to finance the investment needs of SSIs and a US$0.4 million technical assistance component designed to finance data processing equipment for CFP, as appropriate. 4.10 Since the expec.ed large number of subloans would make a composite amortization schedule impractical, the loan would be repaid in equal princi- pal installments over a fixed 17-year term, including a four-year grace period. The prevailing Bank interest rate, as well as commitment and front- end fees, would apply. Any surplus of repaymeats from subloans to CFP over repayments due by CFP to BR and, correspondingly, by BR to the Bank, that would result from the use of a fixed amortization schedule, would permit CFP to build up term resources for lending to SSIs. D. Project Components 4.11 Credit component. Based on experience under the previous projects, the lending component would be used to finance a wide range of small manufac- turing firms, with total assets of less than US$750,000 equivalent. The Bank loan would finance subproject fixed investment needs such as purchase of machinery, equipment, spare parts, and services; construction of industrial facilities and purchase of industrial buildings in selected industrial parks; civil works; and related permanent working capital. 17/ Working capital financing would be limited to the initial stock of raw materials and supplies for the commencement of, or the increase of, such stock needed for the expan- sion, of an SSI firm. Subloans would have repayment periods of four to ten years, including one to three years of grace. CFP would perform a cash flow analysis and calculate the financial rate of return for all subloans above Col$ 2.0 million to determine the appropriate subloan repayment and grace periods. As under the third loan, the limit for subloans not requiring prior Bank approval (the free limit) would be US$100,000 equivalent, which would result in the Bank reviewing an estimated 35 subprojects covering about 10% of the proposed loan amount. In addition, selected subloans below the free limit would be reviewed on a subproject post-appraisal basis by field super- vision missions. 17/ Under the third Bank loan, CFP was allowed to finance free standing working capital with loan resources. However, the major part of loan resources made available for such financing was used by CFP for fixed investment financing, using mainly local resources to finance free standing working capital.  - 24 - 4.12 Under the credit component, CFP would onlend to ultimate subbor- rowers in local currency, and the interest rate to be paid by them would be based on two parts for each subloan: (a) 15X of the principal would bear interest at the rate for 90-day certificates of deposit (CDTs), as such rate would be established through an index to be calculated and published by BR, plus 3 percentage points, during each 90-day interest payment period; and (b) 85% of the principal would bear interest at a fixed rate to be proposed by BR to the Bank and to be agreed pursuant to the procedure described in paragraph 4.14. BR, CFP and the Bank would review the above percentages no later than December 31, 1985 (the end of the first full year of operation of the line) with the view of increasing to 30Z the portion of each subloan sub- ject to variable interest rates, reducing the portion subject to fixed inter- est rates accordingly. This review should take into account the prospects of demand for industrial credit (specifically demand for FFI and CFP lending) and the prevailing levels and structure of interest rates. Based on the con- clusions of such review, the parties would try to reach agreement on the com- position of the interest rates referred to above, which would be applied on subloans made by CFP starting January 1, 1986. If no agreement is reached, the Bank may withheld further commitments under the loan after such date. A similar review would be held by December 1986 as appropriate, when the par- ties would exchange views on the feasibility of making further increases in the percentages of the variable portion of the interest rates applied on subloans. The fixed rates initially proposed would be similar to the ones applied by FFI, i.e., 26% p.a. in the case of investment subprojects located in the metropolitan areas of Bogota, Medellin and Cali, and 24% p.a. for investment subprojects located outside these areas. The rates applicable on the 85% of the principal would be reviewed and modified, if necessary, for all subloans to be presented to the Bank (para. 4.14) but, once applied to a given subloan, they would not change during the life thereof. Applying cur- rent CDT and FFI rates, weighted onlending rates for subborrowers under this arrangement would currently be about 26.9% p.a. nominal for firms located in Bogota, Medellin and Cali and about 25.2% p.a. nominal for firms located in other parts of the country. Thus, about one-seventh of the principal of each subloan would bear a variable, market-determined interest rate, while the remaining portion of the principal would bear a rate which, although required to be positive in real terms at the time of subloan approval, would be fixed for the duration of the loan. To uaify, as much as possible, CFP's onlending rates and, in line with the results of the feasibility study, to mobilize resources from the market (para. 3.33), CFP would also attempt to apply the above-mentioned onlending interest rate system for its 'Credito para la Pro- duccion- and Capital de Trabajo' programs, for which it uses own resources. 4.13 The proposed onlending rates represent relatively high levels of real interest since inflation declined in 1983. However, the current pro- posed rates are in line with (a) those charged by other local financial institutions and (b) the alternative rates available outside Colombia, taking into account the faster rate of miai-devaluations, which had reached 26.3% in 1983. Once the Colombian peso reaches an adequately competitive level regarding major foreign currencies, the pace of devaluation should be more closely aligned with the difference between external and local inflation rates and should result in lower real interest rates for Colombian peso loans.  - 25 - 4.14 The proposed fixed portion of the onlending rates would be reviewed semi-annually by the Bank and the Borrower/CFP, mainly with the objective of maintaining a rate that is compatible with other local term interest rates, in particular with the FFI onlending rates, and positive in real terms. BR would, within eight days after the end of the corresponding semester, notify the Bank of its proposed rate, and the Bank would have 30 days to respond. Should the Bank object to the proposed interest rate, and agreeement with BR on a new rate not be reached within 60 days, the Bank could refuse to make further subloan commitments at the interest rate so objected. In essence, this arrangement gives the Bank the right, during the commitment period of the loan, to insist upon onlending interest rates that are both adequate and positive in real terms. 4.15 The nominal spread on CFP subloans made with Bank loan resources would be (a) 4.5 percentage points for subborrowers located in Bogota, Medellin and Cali; and (b) 5.0 percentage points for enterprises located in other parts of the country (the average spread would be somewhat lower than the nominal average 5.0 percentage points allowed under the third loan). The proposed spread would be achieved by means of BR charging CFP, in every case, a rate which is 4.5 or 5.0 (as the case may be) percentage points lower than the weighted nominal average interest rate discussed in paragraph 4.12. Based on experience under the previous SSI loans, this spread is adequate to cover the higher administrative costs related to making SSI subloans, provide for portfolio losses and generate small profits (paras. 3.24-3.26). Unlike CFP, BR does not charge interest quarterly in advance; consequently, the effective spread for CFP would be higher than the nominal spreads. However, this effective spread could be reduced if oulending rates go dowa (as may be the case in the near future). Agreement between the Bank and the Government/BR on the above-mentioned spreads was obtained during loan negotiati ons. 4.16 Technical assistance component. The project's TA component would provide direct TA to CFP and to the regional TA programs for SSIs (para. 3.18), with an estimated total cost of about US$1.1 million equiva- lent. The loan component for TA would provide US$0.4 million to finance the purchase of computer terminals for CFP's regional offices to complement the data processing equipment financed under the chird loan, as appropriate. CFP would pay to BR an interest rate of 12% p.a. for resources used for this component. The Bank would review and approve in advance any contracts to be financed under this component. CFP would provide the additional estimated US$0.7 million to finance its other technical assistance needs as established in the Action Program (para. 3.33), including consulting services to support CFP's efforts in coordinating and monitoring the initial regional technical assistance programs for SSI. Agreement between the Bank and CFP on the above matters was obtained during loan negotiations. E. Procurement and Disbursement 4.17 Procurement. Local suppliers of imported machinery and equipment are adequately represented in Colombia, competition among them is keen, and their services are reasonable; only in special cases would SSIs find it cheaper to procure machinery directly from abroad. The construction industry in Colombia is quite efficient and competitive, and small industrialists make  - 26 - great efforts to obtain the lowest prices possible. The survival of small firms is heavily dependent upon their ability to purchase materials at the lowest possible prices; procurement of raw materials can thus be expected to be satisfactory. CFP would satisfy itself in all cases that goods and services procured are competitive in quality and price and adequately meet the needs of ultimate beneficiaries. Whenever justified, items would be procured on the basis of at least three different quotations. Since it may be difficult, or relatively costly, to obtain quotations from several suppliers in the case of small orders, the solicitation of three offers would be mandatory for single items exceeding US$25,000 equivalent or for procurement from a single source exceeding such an amount. All subproject appraisals would include a discussion of procurement procedures used, responses received, prices quoted and criteria for selection of suppliers, as appropriate. 4.18 Disbursement. The final date for submission of subloan proposals to the Bank would be December 31, 1987, and the closing date for disbursements would be December 31, 1988 (Annex 4). This schedule is based on average disbursement schedules under the previous three loans and is somewhat shorter than the average disbursement profiles for IDF projects in the LAC Region. As under the previous SSI Loans, disbursements of Bank funds would be made for: %a) up to 100% of CFP's subloan amount to cover direct foreign expenditures for goods and services imported directly by the subborrower; (b) up to 1OOZ of CFP's subloan amount, provided it would not exceed (i) 90% of expenditures for investment items procured localLy cequipment, machinery, industrial construction, permanent working capital, services) or (ii) 70% of the purchase prices of finished industrial buildings, excluding the price of land in selected industrial parks (Rio Negro or any other to be agreed between the Bank and CFP); and (c) up to 100% of expenditures for direct technical assistance to CFP. Although, in principle, Bank funds could finance as much as 100% of CFP's subloans, CFP rarely lends more than 60% of total subproject cost and has finnmced, o. average, about one-half of subproject costs under the first three loans. Il an effort to facilitate rapid project execution, a Special Account would be established at BR and would be used to finance the Bank's share of subloans made by CFP. The account would be established and maintained in US dollars, into which the Bank would make an initial deposit of US$4.0 million. Each withdrawal application submitted to BR would be accompanied by a certified statement of expenditures from CFP. BR would be required to submit a monthly statement of the transaction of the Special Account. 4.19 Because of the large aumber of subloan requests, CFP submit.- sub- projects below the free limit for Bank authorization on a monthly basis, while subloans above the free limit are submitted individually for Bank review and approval. All disbursements would be made against a certified statement of expenditures. Detailed documentation for subproject expenditures would be retained by CFP for inspection by Bank supervision missions and by CFP's external auditors. 4.20 While the Bank normally reimburses for subproject expenditures made less than 90 days prior to the receipt of the subloan request, this limit was raised to 180 days under the first project to compensate for the typical lack  - 27 - of financial and investment planning by SSIs, which frequently leads to sub- mission of subloans at a very late stage. A review of subprojects led to revising the limit under the second and third loans to 135 days prior to the date on which CFP approves the subloan. This limit is appropriate and would be maintained for the fourth loan. F. Accounting, Auditing and Reporting 4.21 CFP would maintain records adequate to reflect its operations and financial situation, in accordance with accounting principles consistently applied and in a form satisfactory to the Bank. CFP would continue to main- tain separate project records as established for the first three SSI pro- jects. CFP would also, on behalf of the Government, establish separate records and accounts for the Government's Special Programs (para. 3.16). CFP's accounts, including the Special Account for the proposed project, would be audited annually by independent auditors acceptable to the Bank. CFP's annual financial statements would include a balance sheet, a statement of income and expenses, a statement of sources and uses of funds, and the trans- actions of the project accounts. The audit report would express an opinion regardlng the reliability of the statements of expenditures to support claims for disbursements and, specifically, whether such claims are supported by adequate documentation and properly reflect the expenditures eligible for financing under the loan agreement, as well as regarding compliance with loan conditions and the adequacy of CFP's internal controls. The audit reports, with scope and format satisfactory to the Bank, would be submitted to the Bank not later than four months after the end of each CFP fiscal year. CFP would forward to the Bank periodic reports on its financial situation and operations, based on its accounting and on its information system. G. Benefits and Risks 4.22 In addition to supporting the Government's development policies, the proposed project would help to fill a gap of medium- and long-term foreign exchange resources needed for the financing of efficient SSI sub- projects in Colombia. Also, the project would provide credit to a group of enterprises that has generally had inadequate access to institutional financ- ing, introduce a new approach for providing technical assistance to SSIs, contribute to the broadening of business ownership, and achieve a more balanced regional development. An estimated 70% of subloans by number and 60% by amount is expected to be for firms located outside the three major cit.es. The proposed project would provide support to the Government's efforts regarding development policies for SSIs and would also strengthen CFP's institutional development initiated under the previous SSI loan by: (a) focusing on increasing CFP's operational efficiency and portfolio growth at a lower unit cost, mainly by strengthening its regional offices; (b) streamlining some of its organizational units; (c) completing an adequate information system; and (d) improving CFP's portfolio quality. 4.23 Based on the results of the first three projects, between 1,500 and 1,700 small enterprises are expected to benefit from the proposed loan, with an average subproject investment cost of US$65,000 equivalent in 1983 prices. Bank financing, it is expected, would foster 5,500-6,500 direct new job opportunities, with another 2,000 created indirectly through forward and  - 28 - backward linkages. The average cost per job is expected to be a relatively low US$16,000 in 1983 prices. Both the expected employment creation and the average capital cost per job are in line with the corresponding results under the second and third projects (para. 4.03). A wide variety of industries would benefit from the project, with none of the ten major SSI subsectors likely to receive more than 20% of the subloan amount. Although most subpro- jects would be for capacity expansion and mainly produce for local markets, it is expected that some 15% would be for new firms and up to 10% would involve exports. Since the resources generated by the project would accrue mostly to urskilled workers, small-owner operators and the Government, the project's income distribution impact would be beneficial. The typical sub- project's financial rate of return is expected to be satisfactory, with most subprojects' financial rates of return about 15%, and in no case be lower than 12%. Because of the small size of the subprojects, calculation of the economic rate of return is not justified. However, the economic return on the projects is likely to be higher than the financial one because prices of input materials are somewhat inflated as a result of the protection involved while, on the other hand, most SSI products are openly traded so that market prices reflect economic values. As a norm, therefore, the typical products of SSI's have lower protection rates than the corresponding inputs. Also, labor costs are estimated to be higher in financial than in economic terms, as reflected by the prevailing unemployment rates. 4.24 The proposed project, as conceived, does not involve any unusual risk. However, some risks do exist: (a) excessively frequent changes of CFP management could hinder its institutional development; and (b) delays in strengthening of CFP's regional offices or shortage and/or fluctuations in availability of funds to CFP, mainly from FFI to finance subprojects' short term working capital needs, and/or from the Government as equity contribu- tion, could affect CFP's operational efficiency negatively and delay project implementation. CFP managers may continue to change relatively often, but efforts over the past several years to institutionalize operating policies and procedures should help to minimize the impact of those changes. To reduce the risks, the Bank and the Government/BR/CFP have agreed upon an Action Program to be carried out by CFP, and on the Government causing to provide or providing funds to increase CFP's paid-in-capital. Thus, it is not likely that these factors could significantly impede attainment of the project's objectives. V. RECOMMENDATIONS 5.01 During loan negotiations, agreement was reached on the following points: With the Government, BR, and CFP on: (a) CFP's limiting its direct lending to micro-enterprises from the informal sector to 2% of CFP's total annual lending during 1984- 1987 and arrangements for CFP's indirect lending to micro- enterprises (para. 3.15); and (b) amount, allocation and relending terms and conditions of the pro- posed loan (paras. 4.09-4.21).  - 29 - With the Government and CFP on: (a) arrangements on: (i) CFP's maintaining, on behalf of the Govern- ment, separate financial records and accounts, and (ii) limiting CFP's lending for those programs to 2.5% of its total annual lend- ing during 1984-1987 (para. 3.16); (b) technical assistance programs to SSIs (para. 3.19); and (c) CFP's maximum debt/equity ratio of 6:1 and the concomitant amount, terms, conditions, and timing of CFP's incremental paid-in capital (para. 3.32). With CFP on: (a) CFP's 'estatutos" and operational guidelines not to be modified in any way that would materially and adversely affect its financial condition or operations (para. 3.01); and (b) CFP's Action Program to strengthen its institutional capabilities (para. 3.33), including: (i) action plan to strengthen CFP regional offices' operating capabilities and guidelines for a CFP reorganization study (paras. 3.06 and 3.26); (ii) steps to strengthen CFP's subproject evaluation and supervision system (para. 3.08); (iii) program regarding strengthening of CFP's information system (para. 3.09); Civ) preparation and execution of annual regional technical assistance programs and the technical assistance plan at the national level; their review with the Bank and appropriate modifications, and strengthening and upgrading of CFP's tech- nical assistance unit at headquarters (paras. 3.18 and 3.19); (v) target levels for CFP's administrative expenses and action plan to ensure a sound portfolio growth at lower unit cost (para. 3.26); (vi) CFP's portfolio recovery plan (para. 3.28); (vii) guidelines for the study to mobilize resources from the market (para. 3.33); and (viii) consulting services for the regional technical assistance programs (para. 4.16).  - 30 - 5.02 As a condition for effectiveness of the Loan, CFP should present the Action Program, in terms satisfactory to the Bank, approved by its Board of Directors (para. 3.33). 5.03 The proposed project would be a suitable basis for a Bank loan of US$40 million to BR, for a term of 17 years, including four years of grace. June 14, 1984  ^ 0 ! 31 | ;Wi I iJ Ii1 iiG I 1143 t h t e e~ i l RI} ild 3R l M ! X . ii. 1 , 1 '1 1 I 1i 3181 ii 0  ANNEX 2 - 32 - Page 1 of 5 COLOMBLk STAFF APPRAISAL REPORT FOURTH SMALL-SCALE INDUSTRY PROJECT Institutional Characteristics of Financial and Technical Assistance to SMI A. Institutions other than CFP Providing Credit to SMI 1. Fondo Financiero Industrial (FFI), was formed in 1968 by the Monetary Board as a second-tier mechanism for financing working capital and investment needs of SMI. There are various criteria for eligibility of firms to receive FFI financing, including maximum total asset size (up to ColS 60 million, some US$680,000 equivalent). The location of a firm determines the interest rate applicable, with higher rates for firms based in the Bogota, Medellin and Cali metropolitan areas. The average term of FFI loans has been around four years since 1979. 2. In 1982, FFI approved almost 1,400 loans totaling Col$ 3.4 billion (approximately US$38 million equivalent). After having reached Col$ 2.3 billion in 1978, total value of approvals dropped by over 50% in real terms between 1978 and 1981; 1982 was the first year since 1978 in which there was real growth in either total value or average value of FFI approvals. Reduc- tion in FFI lending between 1978 and 1981 in real terms does not necessarily reflect a decline in SMI credit demand, however; it is, rather, the result of a significant drop in the effective intermediation margin received by finan- cial intermediaries after 1976. It is partly in response to the reluctance of many intermediaries to place FFI loans or otherwise make longer term cor- mitments, thus curtailing credit supply to SMI, that the Monetary Board approved a new interest rate scheme for FFI and other rediscount funds in December 1983. Under the new scheme, firms no longer pay a fixed interest rate. The rediscounted portion of a subloan (an average 80X of total sub- loan) wil. continue to bear a fixed rate of interest, and the remainder (from the financial intermediary) will be adjusted quarterly, bearing interest at 3 percentage points above the floating CDT rate (based on an index computed and published by BR). BR hopes that the margin provided by this system will be sufficient to encourage term lending on the part of private DFCs and comrer- cial banks. 3. Another objective of the new system is to expose SMI gradually to a more market-oriented interest rate system by introducing flexible rates. If successful, it may be expected that FFI's own rates may eventually become flexible in order to reflect changes in the cost of funds to FFI. FFI receives its resources from several sources, primarily from forced invest- ments: in 1982, over 90% of FFI resources came from forced investments by  ANNEX 2 - 33 - Page 2 of 5 financial institutions under Monetary Board Resolutlon 39J1978, which bear an Interest of 25Z for the Investor. Since the FFI rediscount rate has been lower (20-22Z in 1983) than the average cost of FFI resources since 1979, it faces a negative spread on Its lending operations. Moreover, the spread Is more strongly negative on marginal operations, so that the average negative spread increases as FFI's operating volume Increases. 4. Caja de Credito Agrario, Industrial y Minero (CAJA) is a state- owned development bank set up in 1933 which provides credit mainly to the agricultural sector. While it makes some subloans to small-scale Industrial enterprises, it was decided in 1977 to disband CAJA's industrial lending staff. CAJA now makes some SSI subloans which it redlscounts wich FFI: only Col$ 1 million each (US$12,500 equivalent) during the first half of 1983. In 1983, CAJA accounted for 16X of total FFI financing operations but less than 6Z of total value of loans, mainly for small agroindustries. CAJA also makes loans out of its own resources, but borrowers must have total assets (includ- ing personal assets) of not greater than Col$ 6 million (US$68,000 equiva- lent), and maximum subloan size may not exceed Col$1 million. Because of CAJA's many branches In rural areas, it has reached small agroindustrial firms outside the areas covered by other financial Institutions. 5. A number of other public financial institutions act as discounting facilities for CFP and other direct SKI lenders. The Instituto de Fomento Industrial (IFI), the public industrial Investment bank, does not lend directly to SSI, but lends funds to CFP and other intermediaries for onlending. These funds come from various forms of forced investment, al of which are declining and are expected to remain low in the future PROEXPO, the Government export promotion institutlon, provides credit at 18t p.a. nominal rate to exporters through financial intermediaries However, since few SMIs are exporters, it has only a marglnal impact on the sector in terms of direct lending. As one of the principal shareholders of CFP, it can exercise an important influence on SKI financing. 6. The Instituto de Financiamiento y Desarrollo Cooperativo (FINANCIACOOP) is a financial insticution whlch was established In 1969 to meet the needs of Colombian cooperatives. It makez the FFI credit lines available to its agroindustrial member cooperatives and is also the executive agency (since 1982) of the marketing subprogram of the Integrated Rural Development (DRI) Program. 7. DFCs and commercial banks also provide sLae credit, mainly for working capital to the SMI sector, primarily through FFI loans. In the first half of 1983, private DFCs accounted for 42 of the .iolume and 8% of the value of FFI approvals, with loams averaging Col$ 5.4 million (some US$60,000 equivalent). Conmercial banks, during the same period, accounted for over 9% of the volume and almost 22% of the value of FFI approvals, with subloan size averaging ColS 6.4 million (some US$72,000 equivalent). Both of these types of Institutions concentrate on the larger SMI firms because they consider the  ANNEX 2 - 34 - Page 3 of 5 administrative handling costs too high to justify smaller loans (by compari- son, CFP's average loan size through FFI for January-June 1983 was Col$ 2.1 million, some US$23,000 equivalent). It is expected that the recent changes in FFI lending rate structure and levels would result in an increased will- ingness of commercial banks and DFCs to provide loans for smaller firms. 8. Fondo Nacional de Garantias (FNG), the Government guarantee fund, was established in 1981 by IFI and CFP to guarantee loans to small manufac- turing and mining firms with total assets not exceeding Col$ 60 million (US$680,000 equivalent). In addition to IFI and CFP, the following institu- tions are also shareholders of FNG: PROEXPO, CARBOCOL and ACOPI. FNG's total paid-in capital amounts to Col$ 100 million. FNG, which started its operations in 1982, guarantees up to 80% of the subloan approved by a finan- cial intermediary. As of November 1983, it had approved some 230 guarantee certificates, totaling Col$ 140 million. More than two-thirds of all its operations, however, have been with SMI subloans made by CFP. FNG is cur- rently endeavoring to attract commercial banks and DFCs to provide more loans to small firms. B. Institutions Providing Technical Assistance to Colombian SKI 9. Technical assistance to Colombian SKI is available from a number of sources, the most important of which are discussed below. However, SNIs have suffered because there is no one coordinating service responsible for identi- fying the SMI sector's technical assistance needs and ensuring that the various institutions respond to these needs, either at the national or regional level. While there has thus been no inventory taken of these needs, they are known to include: accounting and financial control, planning, mar- keting, production processes and quality control, general management develop- ment, and adequate information services. 10. Programs for technical assistance to SKI at some of the Government-supported institutions have contracted, rather than expanded, in the past few years. This has been attributed largely to the lack of a coordinated Government policy toward SMI. Under the new Government, there has been a notable improvement in this regard, with strong support emerging for comprehensive policies directed at SKI. 11. Servicio Nacional de Aprendizaje (SENA) was established by the Government in 1957 to provide vocational training services in Colombia. SENA is financed by a national 2% payroll tax and is thus well funded. SEIA operates in 18 regions of Colombia, with at least one office in each of these regions. It undertakes many different activities, with classroom vocational, tecbnical and professional training predominating, but including technical assistance to enterprises. SENA's programs for enterprises are structured according to their size, with separate programs for: (a) large-scale -.- -- prises, (b) small- and medium-scale enterprises, and (c) microenter.ls,s Each classification embraces the agricultural, industrial, commerce ^-4 service sectors indiscriminately.  ANNEX 2 - 35 - Page 4 of 5 12. SENA's program for small- and medium-scale enterprises, centers upon providing 'integral- technical assistance, which includes management development, industrial relations, accounting and finance, training and development of supervisory staff, assistance with credit requests, financial resource management and technological assistance. SENA has a staff of about 60 persons involved in this program nationwide, down from some 80 in 1975. Up to 500-600 SMI firms receive assistance each year. Almost all of these firms have at least 20 employees since SENA prefers to work with firms that have at least two management levels. Recently, SENA decentralized the setting of annual goals, including the choice of priority subsectors to receive technical assistance, to the regional level. 13. Fundacion para el Fomento de la Investigacion Cientifica y Tecnologica (FICITEC) was set up in 1971 as a non-profit organization by the Colombian National Coffee Growers Association. Today, only 1OZ of FICITEC's annual operating budget (Col$ 30 million in 1983) comes from this group, with about 23Z coming from KfW funds and 20% from own resources. The remaining 47% comes from client's fees. FICITEC now provides in-depth technical assis- tance to only 5-10 large firms annually (mainly firms in the Coffee Growers Association, such as Banco Cafetero), which pay the fuli cost of the services provided, but it also works with some 60 smaller firms each year, which pay approximately half of the full cost.l/ In addition, FICITEC organizes and gives seminars to groups of SKIs. FICITEC undertakes these activities with a small professional staff, currently numbering 14. While FICITEC's offices are in Bogota, FICITEC is better known in the provinces, with only 20% of its clients in the capital. In addition to its own activities, FICITEC coordi- nates CESO's (Canadian Executive Service Overseas) operations in Colombia, which assists some 40-50 SMI firms annually throughouit the country. 14. Instituto de Investigaciones Tecnologicas (IIT) is an autonomous Government institution established in 1958 to undertake applied research, especially that related to industrial uses of agricultural products. The three operational areas of IIT cover, respectively, research into new pro- ducts and processes, industrial services such as quality control and develap- ment of pilot plants for new products and processes, and consulting on planat and process design with machinery evaluation, among other services. Over 50% of IIT's work focuses on the food subsector, and about 20% each on chemical products and metal mechanics. About 80% of IIT's clients are in Bogota, where it has its facilities. It has helped about 30 SMI firms a year in the past, subsidizing this technical assistance with KfW funds. 1/ The subsidy is generally allocated against the KfW funds, which results from an interest rate differential on KfW funds and is to be used only to help pay for technical assistance to smaller firms.  ANNEX 2 - 36 - Page 5 of 5 15. Fondo Colombiano de Investigaciones Cientificas y Proyectos Especiales (COLCIENCIAS) was set up in 1969 as a decentralized dependency of the Ministry of Education, with the mandate to promote science and technology in Colombia. COLCIENCIAS operates as a financial fund to promote research activities. Its activities in the industrial sector consist mainly of pro- moting technological innovation. A US$20 million loan was made by IDB in 1983 to be channeled through COLCIENCIAS for scientific and technological research. In addition, COLCIENCIAS has promoted, since the early 1970s, the creation of information centers for science and technology which have tended to be associated with local universities. Today, in cooperation with the National Association of Industrialists (ANDI), it is striving to shift the focus of its activities to the productive sectors. A further service pro- vided by COLCIENCIAS, which is of interest to some SMIs, is assistance in obtaining patents for new products and processes. 16. Fondo de Promocion de Exportaciones (PROEXPO) was established by the Government in 1967 to promote non-traditional exports. In its technical assistance activities, it offers support to SNIs with export potential, fre- quently in coordination with other agencies such as SENA, FICITEC, IIT or by external assistance. PROEXPO usually focuses its assistance on manufacturing subsectors, such as furniture, leather processing, food processing, orna- mental plants, and garments. It identifies their export potential and assists groups of small- and medium-sized producers to develop and market their products abroad. Some 100-150 SMIs are assisted each year.  - 37- ANNEX 3 COLOMBIA T-1 STAFF APPRAISAL MEPORT FOURTH SMALL-SCALE INDUSTRY PROJECT Table 1: INDUSTRIAL DEVELOPMENT AND ITS DETERMINANTS, 1970-1982 Changes in Determining Factors of Output Change Value Gross Internal Import Export Added Output Demand Substitut. Development Total Manufacturing 1970-74 10.1 9.0 7.0 1.0 1.0 1974-79 3.7 4.0 4.1 -0.5 0.5 1979-82 -0.6 -0.3 2.3 -2.3 -0.3 Consumer Goods 1/ 1970-74 9.3 6.2 5.5 -2.3 0.8 1974-79 3.4 4.3 4.5 -0.1 0.5 1979-82 -0.4 0.1 1.0 -.05 -0.3 Intermediate Goods 1970-74 10.0 12.3 9.4 1.4 1.4 1974-79 3.4 0.3 2.8 0.1 0.4 1979-82 - 0.3 2.8 -2.3 -0.2 Capital Goods 1970-74 17.0 14.7 4.5 9.3 1.0 1974-79 6.6 5.8 7.6 -2.6 0.8 1979-82 -5.1 -5.4 5.0 -10.2 -0.2 1/ Excluding coffee production. SOURCE: FEDESARROLLO,Coyuntura Economica,-XIII. 3 LCPI2 December 1983  - 38 - COLOMBIA ANNEX 3 T-2 STAFF APPRAISAL REPORT FOURTH SMALL-SCALE INDUSTRY 'ROJECT Table 2: DISTRIBUTION OF EMPLOYMENT ACCORDING TO SIZE OF ENTERPRISE Year/Country Cottage Shop SMI Large Industry ( 5 workers & less) (5-99 workers) (100 workers & more) Colombia 1964 51 26 24 1970 54 22 25 1978 43 28 28 Other LDCs: Taiwan 1966 4 39 57 1977 3 33 64 Korea 1975 36 17 47 Philippines 1967 78 7 15 1975 66 8 26 India 1973 60 18 22 Indonesia 1975 76 12 12 Nigeria 1972 59 15 26 ADVANCED COUNTRIES U.S.A. 1967 1 22 77 Canada 1959 2 32 66 Japan 1975 19 36 44 Source: h. Cortez, et al.. "What Makes for Success in Small and Medium Scale Enterprises: The Evidence of Colombia" mimeo, Washington, The World Bank 1983. LCPI2 December 1983  ~ 39 ~ ANNEX 3 T-3 COLOMBIA STAFF APPRAISAL REPORT FOURTH SHALL-SCALE INDUSTRY PROJECT Table 3: ESTABLISHMENTS, EMPLOYMENT AND VALUE ADDED OF SMALL AND MEDIUM-SCALE INDUSTRY, 1976-1981 1/ Year Establishments Employment of SKI Value Added Z No. Z of total in thousand Z of total in million Z of total Growth 1976 5446 85 157.7 34 20.3 19 - 1977 5664 85 162.8 33 25.8 17 27 1978 6055 91 160.9 32 33.4 18 30 1979 5681 84 164.7 32 .41.8 16 23 1980 5771 84 163.7 32 54.9 16 31 1981 5743 85 163.6 33 65.0 16 18 Source: DANE 1/ SKI defined as firms having between 5-99 workers. LCPI2 December 1983  - 40 - AME 3 T-4 COLOMBIA STAFF APPRAISAL REPORT FOURTH SMALL-SCALE INDUSTRY PROJECT Table 4: SMI'S PARTICIPATION IN MANUFACTURING SECTOR AND SUBSECTOR SHARES (in %) 1976 and 1983 No. of SMI Enterprises SMI Employment Subsectoral ISIC Subsectors as X of Total Employment Shares 1976 1983 1976 1983 1975 1983 Food 92 94 49 53 15.5 14.5 Beverages 54 55 12 12 1.0 3.9 Tobacco 69 78 18 19 .2 .4 Textiles 81 84 16 23 6.7 9.7 Clothing/Shoes 94 96 52 67 17.6 15.5 Wood/Cork 97 98 57 70 3.1 1.5 Furniture 98 98 77 83 10.9 2.9 Paper 84 87 33 36 2.1 1.6 Printing 95 96 52 60 4.5 4.7 Leather 90 94 50 52 1.5 2.0 Rubber 90 95 28 57 1.1 1.2 Chemicals 81 87 32 35 5.2 6.4 Oil & Coal Derivatives 85 96 9 43 .3 NM Minerals 92 92 42 44 12.2 5.9 Basic Metals 91 95 36 47 4.4 2.6 Metal Products 93 94 52 57 9.5 10.1 Non-Electrical Machinery 96 96 67 61 5.8 2.1 Electrical Machinery 94 96 51 59 2.7 4.9 Transport 91 98 n.a. 69 4.3 4.5 Other 92 94 51 57 5.8 5.6 Total 92 94 42 52 100.0 100.0 Source: Instituto de Seguro Social, Anuario 1976, and unpublished statistics. LCPI2 April 1984  - 41 - COLOMBIA ANNEX 3 STAFF APPRAISAL REPORT T-5 FOURTH SMALL-SCALE INDUSTRY PROJECT Table 5: CAPITAL AND LABOR PRODUCTIVITY IN SMI, 1976 AND 1980 Plant Size Value Added Value Added Value Added (No. of Workers) Book Value of Fixed Consumption of Ism pJ.o y e d Wo-rce'r Amount & Inventory Energy in thousand pesos 1976 1976 1976 2 - 5 .93 25.6 5 - 9 .95 21.3 79.3 10 - 14 ;1.00 Z6.3 85.8 15 - 19 1.10 29.4 90.0 20 - 24 *1.54 35.7 130.4 25 - 49 1.11 $4.5 118.2 50 - 74 1.00 29.4 128.1 75 - 99 1.33 22.7 199.4 100 - 199 1.19 41.2 216.8 200 '1.09 21.2 305.1 TOTAL 1.11 24.4 224.65 Small Industry 1.14 32.3 100.74 Medium Industry 1.15 25.6 163.75 Large Industry 1.10 23.8 232.5 Small & Medium Industry 1.15 28.6 .118.74 Source: Cortez, et al. "What Makes for Success in Small and Medium Scale Enterprises: The Evidence of Colombia" mimeo, Washington, The World Bank 1983. LCPI2 December 1983  - 42 - COLOMBIA ANNEX 3 STAFF APPRAISAL REPORT T-6 FOURTH SMALL-SCALE INDUSTRY PROJECT Table 6: REMUNERATION IN MANUFACTURING 1972-1980 (in Thousands 1970 Col$/pe- employee) 1972 1980 Salary Soc. Sec. Total Salary Soc. Sec. Total Small Firms 11.0 2.8 13.8 11.6 4.1 15.7 Medium Firms 15.8 5.8 21.6 14.6 7.4 22.0 Large Firms 22.8 11.5 34.3 20.6 16.0 36.6 TOTAL 18.4 8.2 26.6 17.3 11.4 28.7 Ratios LI/MI 1.4 2.0 1.6 1.4 2.2 1.7 LI/SI 2.0 4.1 2.5 1.8 3.9 2.3 MI/SI 1.4 2.1 1.6 1.3 1.8 1.4 SOURCE: DANE,"Pequena, Mediana y Gran Industria en Colombia 1970-1980',' Boletin Mensual de Estadistica,No. 373, August 1982. LCP12 December 1983  - 43 - ANNEX 3 T-7 COLOMBIA STAFF APPRAISAL REPORT FOURTH SMALL-SCALE INDUSTRY PROJECT Table 7: SOURCES AND USES OF INDUSTRIAL FINANCING, 1978-1982 (in %) 1978 1979 1980 1981 1982 A. SMALL ENTERPRISES SOURCES Own Resources 25.5 28.1 51.9 49.7 45.8 CFP 12.5 28.7 12.6 12.6 14.0 Commercial Banks 9.2 18.2 15.6 17.3 14.6 Financieras 15.8 7.2 4.9 3.8 2.3 Supplier 15.5 12.0 9.5 8.1 8.8 Parallel Market 17.3 3.6 5.0 3.1 6.0 Other 4.2 2.2 0.6 5.4 3.4 USES Machinery & Equipment 33.1 27.5 20.6 18.9 17.0 Buildings 15.5 9.6 4.8 5.3 3.6 Working Capital 38.2 58.5 70.9 70.2 76.8 Other 13.2 4.4 3.7 5.3 2.7 B. LARGE ENTERPRISES SOURCES 1971 1975 1979 Retained Profits 13.6 10.8 9.1. Depreciation 10.6 15.2 10.8 Equity 10.2 4.0 5.2 Loans 65.3 70.0 74.9 USES Physical Investment 62 61 49 Financial Investment 38 39 51 Source: CFP Surveys 1979-83 and Superintendencia de Sociedades Anonimas LCPI2 December 1983  COLOMBIA STAFF APPRAISAL REPORT FOURTH SMALL SCALE INDUSTRY PROJECT Table 8 : CREDIT TO 5MI (Col$ millions) -SMI Credit Total Credit FFI Approvals CFP CAJA to Industry SMI Credit Year (1) (2) Banks CFs Total (percent) (percent) 1970 89.4 132 87.3 62.1 370.8 6.2 55.6 1971 182.7 172 89.7 54.0 498.4 7.0 52.0 1972 180.2 186 135.7 90.0 591.9 7.8 55.5 P 1973 274.4 465 323.3 114.8 1,177.5 12.6 49.7 1974 331,0 891 494.2 130.5 1,846.7 19.5 41.8 1975 466.5 598 274.2 124.9 1,463.6 11.3 37.8 1976 808.3 470 435.2 202.6 1,916.1 11.8 44.8 1977 971.4 838 985.4 407.0 3,201.8 15.4 55.7 1978 1,072.5 1,863 1,195.9 640.2 4,771.6 16.9 49.1 1979 1,301.3 2,724 1,286.5 888.2 6,200.0 20.7 43.4 1980 2,061.0 n.a. 1,324.8 383.1 3,768.9 8.5 60.3 1981 3,074.9 n.a. 1,148.9 482.6 4,706.4 6.5 72.4 n.a Not available. (1) Corporacion Financiera Popular (2) Revista Banco de la Republica, Annual Report. SOURCE: Banco de la Republica w LCPI2 January 1984  - 45 ANNEX 3 T-9 COLOMBIA STAFF APPRAISAL REPORT FOURTH 5(ALL SCALE INDUSTRY PROJECT Table 9: FFI APPROVALS 1969-1982 Average Value of Approvals Approvals Real Approvals 1/ Nominal Real Year Number Col$ millions Col$ millions CO1$000 CO1$000 1969 426 160.9 173.2 378 407 1970 576 206.3 206.3 358 358 1971 1,017 259.4 232.4 255 229 1972 980 328.6 249.2 335 254 1973 1,166 585.7 347.2 502 298 1974 1,110 772.2 336.5 696 303 1975 1,119 552.8 192.1 494 171 1976 1,140 858.3 242.7 753 213 1977 1,765 1,785.1 398.3 1,011 226 1978 2,101 2,345.0 444.8 1,116 212 1979 1,948 2,691.5 339.4 1,382 205 1980 1,624 2,280.3 272.5 1,404 168 1981 1,251 2,213.6 211.9 1,769 169 1982 1,396 3,406.4 261.1 2,440 187 1/ Deflated using Wholesale Price Index, base 1970-100. SOURCE: Banco de la Republica LCPI2 January 1984  COLOHBIA STAFF APPRAISAL REPORT FOURTH SMALL SCALE INDUSTRY PROJECT Table 10 : SHARE OF PRINCIPAL INTERMEDIARIES IN FFI APPROVALS, 1970-1982 1/ (Percentages) BANKS CFP CAJA Private Mixed & Official CFs Year No. Value No. Value No. Value No. Value No. Value 1970 44.8 22.3 8.5 5.3 23.1 38.5 1.9 3.8 21.7 30.1 1971 71.9 41.1 3.9 3.5 13.9 33.0 0.9 1.6 9.4 20.8 1972 54.8 29.8 2.3 1.5 25.4 35.6 2.5 5.7 15.0 27.4 1973 51.4 23.9 1.5 1.3 31.8 46.3 4.3 8.9 11.0 19.6 1974 51.8 18.9 0.3 0.2 31.0 55.8 4.2 8.2 12.7 16.9 1975 57.2 23.6 7.3 4.2 20.7 44.3 2.4 5.3 12.4 22.6 1976 50.4 21.3 13.7 4.4 22.1 45.7 1.5 5.0 12.3 23.6 1977 52.8 18.7 9.3 3.3 17.5 41.5 6.1 13.7 14.3 22.8 1978 46.3 14.3 23.7 7.4 12.1 39.2 8.3 11.8 9.6 27.3 1979 39.0 12.6 28.1 6.6 14.2 31.7 9.9 16.1 8.8 33.0 1980 42.8 25.1 32.3 10.4 10.3 26.6 9.4 21.1 5.2 16.8 1981 44.7 26.3 32.8 11.2 7.4 16.6 8.8 24.1 6.3 21.8 1982 45.9 32.2 25.5 10.7 12.1 23.9 11.1 19.8 5.4 13.4 1/ Up to November 1982. SOURCE: Banco de la Republica. LCPI2 January 1984  00L@UIA STAFF APPRAIAL SR T FOURTH WIALL-SCLE INDUTSRT PROJECT COaW IaON FIIICIERA UMA Table 11: DISTRIBUTION oF OWERSIIP OVER THE PERIOD IECEWR 31, 1979 TO SEPllYEER 30, 1953 (Coll theusand; Par Valuet 1 share e ColOO) D enber 31, 1979 Decmber 31, 1960 December 31, 1961 D_ecbwr 31, 1982 5aptetbr 30, 1963 Sharebolders Subscrlbed Pald In Subscrlbed Pald In Subscribed Pald In Subscrlbed Paid In Subseribed PaId In Pubile Ownrship 483,281 483,261 483,261 463,281 563,281 5S3,281 633,261 633,261 7316701 707,007 ODrect Gcvernient Partllpatlon 60,000 60,000 60,000 60,000 160,000 160,000 210,000 210,000 364,526 364,526 State-owned Equity Partlcpation 423,261 423,261 423,261 423,261 423,261 423,261 423,261 423.261 423,261 423,2t1 -Banco Popular 129,70 129,706 129,706 129,706 129.70S 129,705 129,708 129".06 129,706 1291703 -Ccrporacion do relras y 32,636 32,638 32,638 32,636 32,636 32,436 32.638 32,636 32,638 32,630 Exposcliones -IFI 10,936 10,936 10,936 10,936 10,936 10,936 1O,,36 10,936 10,936 10,936 -Proexpo 250,000 250,000 250,000 250,000 250,000 250,000 250,000 250,000 250,000 250,000 Private Own r shIp 7.5 7.5 6.3 .3 9.2 9.2 9.5 9.5 9.5 9.5 TOTAL U43,219 463,25i 403,289 483,269 583,290 583,290 633,291 633,291 737,617 767.617 Sawres CFP LCP12 j O einber 1963  - 48 nNNEX 3 T-12 COLOMBIA STAFF APPRAISAL REPORT FOURTH SMALL-SCALE INDUSTRY PROJECT CORPORACION FINANCIERA POPULAR Table 12: LIST OF BOARD OF DIRECTORS AND ALTERNATES AS OF NOVEMBER 1, 1983 President Delegate 1. RodrTgo Marnn Bernal Jaime Ardila Gomez Minister of Economic Development Secretary General Mllstry of Economic Development Principal Directors Alternate Dlrectors 2. Morris Harf Meyer Luls Delfin Borrero Cabrera Manager Chlef, Special Studles Unit Nueva Frontera National Planning Department 3. Mauricio Fernandez Fernandez Ernesto Gomez Buftrago Assistant Director General Assistance Director Flnance PROEXPO PROEXPO 4. Francisco Ortega Acosta Ernesto Merlano Marln Assistant Director Assistant Director IFI 5. Pedro Javior Soto Sierra Romulo Orjuela Bernal Carrera 7a, No. 15-56 Manager, Headquarters Banco Popular 6. BenJamin Lopez Arcinlegas Hernando Cosas Rico Manager, Bogota Zone General Manager Banco Popular WIMPY 7. Francisco-de Paula Ossa Uribe Carlos Dreszer Winer National President General Manager ACOPI COLAF LCP12 December 1983  COLOMBIA STAFF APPRAISAL REPORT FOURTH SH4ALL-SCALE INDUSTRY PROJECT CORPORACION FINANCIERA POPULAR Table 13: ORGANIZATION CHART AS or OCTOBER 1 1983 Ministry of Economic Development Supcrlntondcy of Bnk's Corporacion Flanclera Populer, S.A. s Q~~~~~eneral Assmbly ts 1Comptroller t - Board of Dlrectors It Camel ttee 1Management |I 2ommltt General Manager Legaf-_Audtors_of_ __ . 1 Administratit've Admintstrav! Fln elal per an 3 Mnagor Commt t eanegr Nancg_r H ncecl G T B udgatn Ior Special | roJ eF iSarvies | DOcpartment Dept. Ocptcnd rlncnTd Studies Programming DOpt. DCpt. DOpt Organlsatlon S*ryste Portfolio Acco u torel and Development ODpartoent DOpt. Dot,pt. Ammatane Promotlon Oepartoent p Dept. Offleo Pcrnonn I Insurce Tax Lbr | election end |Unit Unit | Reglonal Advisory Board R _ cas Barranquilla - M-de iln - ll Preira - Bogota - Bucaramnega - Ibague Cartagena - Armenia - Mahnlaies - Cucut - Nava - Pasta - TunJc w Pcpcrc Promotion Office  -50- ANNEX 3 T-14 COLOMBIA STAFF APPRAISAL REPORT FOURTH SMALL-SCALE INDUSTRY PROJECT CORPORACiON FINANCIERA POPULAR Table 14: NUMBER OF STAFF AT HEADQUARTERS AND IN THE REGIONAL OFFICES. 1979-83 (year-end data) Profes:lonals Others Total No. % No. S No. S 1979 Headquarters 53 39.3 120 49.0 173 45.5 Regional Offices 82 60.7 125 51.0 207 54.5 TOTAL 135 100.0 245 100.0 380 100.0 1980 Headquarters 58 39.7 122 49.0 180 45.6 Regional Offices 88 60.3 127 51.0 215 54.4 TOTAL 146 100.0 249 100.0 395 100.0 1981 Headquarters 53 36.8 114 47.7 167 43.6 Regional Offices 91 63.2 125 52.3 216 56.4 TOTAL 144 100.0 239 100.0 383 100.0 1982 Headquarters 50 36.2 110 46.6 160 42.8 Regional Offices 88 63.8 126 53.4 214 57.2 TOTAL 138 100.0 236 100.0 374 100.0 1983 (Sept. 30) Headquarters 51 34.2 112 45.0 163 41.0 Regional Offices 98 65.8 137 55.0 235 59.0 TOTAL 149 100.0 249 100.0 398 100.0 Source: CFP LCP12 December 1983  - 51 - ANNEX 3 T-1 5 COLOMB IA STAFF APPRAISAL REPORT FOURTH SMALL-SCALE I NDUSTRY PROJECr CORPORACICN FINANCIERA POPULAR Table 15: SUMK%RY OF SUPERVISION ACTIVITIES, 1979-0 8/ Total Cllents Loans of Supervision Activities Regional Offices Year Clients In Arrears Doubtful Collection No. of Visits VisTts as % of Total Clients Bogota: 1979 1,004 117 172 397 40 1981 775 125 226 426 56 1983 775 141 199 129 16 MedeliIn: 1979 720 24 54 229 31 1981 290 16 50 174 60 1983 339 45 32 30 8 Bucaramanga: 1979 572 61 32 246 43 1981 492 80 49 163 33 1983 507 99 47 50 9 Ibaque: 1979 470 34 14 111 24 1981 39D 45 26 150 38 1983 553 122 40 54 9 Call: 1979 417 37 88 177 42 1981 219 30 95 146 67 1983 387 47 92 30 7 Others: 1979 1,950 394 416 845 43 1981 1,816 262 259 993 55 1983 1,/35 500 284 353 20 TOTAL 1979 5,133 633 776 2,005 39 1981 3,982 558 705 2,052 52 1983 4,296 954 694 646 13 a/ Data for 1983 Include activites during the first six months. Source: CFP LCPI2 Oocember 1983  - -12-~- -~~~wf- Is I IdOl- C s - k * t 3110A tm Ms NA 3" .8. 1A m MA 1 J. 0 1 W w~~~~~ m so w J . J . IN. NM NA $OA r~~~~~~e t. 93 510$ 1.30 MA .110316143- "I'a IIMA 'A ", n . 11 1. J 1 - m -. i .. inw l " RA Mail '.5l5s116 . A0