Colombia - Third Small Scale Industry Project

Document of The World Bank FOR OFFICIAL USE ONLY FILE COPY Report No. 2626b-CO STAFF APPRAISAL REPORT COLOMBIA THIRD SMALL-SCALE INDUSTRY PROJECT March 28, 1980 Projects Department Latin America and the Caribbean Regional Office This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENTS (as of December 31, 1979) Currency Unit = Colombian Peso US$1 = Col$43.85 Col$l = US$0.0262 ACRONYMS ACOPI - Asociacion Colombiana Popular de Industriales (Colombian Association of Small Manufacturers) BR - Banco de la Republica (Colombia's Central Bank) CAJA - Caja de Credito Agrario Industrial y Minero (Agricultural, Industrial and Mining Bank) CFP - Corporacion Financiera Popular COLCIENCIAS - Fondo Colombiano de Investigaciones Cientificos y Proyectos Especiales DANE - Departamento Administrativo Nacional de Estadistica (National Department of Statistics) DFC - Development Finance Company FEDESARROLLO - Fundacion para la Educacion Superior y Desarrollo FICITEC - Fundacion para el Fomento de la Investigacion Cientifica y Tecnologica (Foundation for the Development of Scientific and Technological Research) FFI - Fondo Financiero Industrial (Industrial Financing Fund) FINANCIACOOP - Instituto de Financiamiento y Desarrollo Cooperativo FIP - Fondo de Inversiones Privadas (Private Investment Fund) 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 - Regional Advisory Board SENA - Servicio Nacional de Aprendizaje (National Vocational Training Organization) SMI - Small- and Medium Scale Industry SSI - Small-Scale Industry UNDP - United Nations Development Programme UNIDO - United Nations Industrial Development Organization USAID - United States Agency for International Development FISCAL YEAR January 1 - December 31  FOR OFFICIAL USE ONLY COLOMBIA THIRD SMALL-SCALE INDUSTRY PROJECT STAFF APPRAISAL REPORT TABLE OF CONTENTS Page No. I. SMALL- AND MEDIUM-SCALE INDUSTRY (SMI) IN COLOMBIA ..... 1 A. Economic Setting .................................. 1 B. Structure and Characteristics of SMI .... .......... 2 C. Growth and Development Constraints .... ............. 5 D. Government Policies ............................... 6 E. Prospects ......................................... 7 IIr FINANCIAL AND TECHNICAL ASSISTANCE TO SMI .... .......... 8 A. The Financial Sector and Policy Framework .... ..... 8 B. SMI Financing ..................................... 10 C. Institutional Credit for SMI ...................... 11 D. Sources of Technical Assistance to SMI .... ........ 13 III. CORPORACION FINANCIERA POPULAR ......................... 15 A. Basic Structure ................................... 15 B. Systems and Procedures ............................ 17 C. Resources ......................................... 18 D. Financial Position and Results .................... 20 E. Operations .............. 22 F. Financial Projections ............................. 24 IV. THE PROJECT.25 I. TEPOET............................................ 2 A. Experience under Previous Bank Loans .... .......... 25 B. Project Objectives ................................ 28 C. Project Cost and Financing ........................ 28 D. The Proposed Loan ................................. 29 E. Benefits and Risks ................................ 32 V. AGREEMENTS REACHED AND RECOMMENDATIONS .... ............. 34 This report is based on the findings of an appraisal mission which visited Colombia during April/May 1979. The mission comprised Messrs. Bentley, Nogales (LCPI2), Bolte (Bank/UNIDO Cooperative Programme), and Berry (Consultant). Mr. Ishaq (LCPI2) assisted in the preparation of Annex 2 of this report. Thk document ha a restricted distribution and may be used by recipients only in the performance of ttWir oficd dutie. Its contents may not otherwise be disclosed without World Bank authorization.  TABLE OF CONTENTS (Continued) LIST OF ANNEXES 1. Comparison of Features of the Past and Proposed Small-Scale Industry Loans. 2. Analysis of Characteristics of Subborrowers and Subprojects under Loan 1451-CO. 3. Supporting Tables and Charts. T-1. Assorted Growth Rates, 1967-75. T-2. Establishments and Employment, SMI, by Industry, 1975. T-3. SMI's Participation in the Factory Manufacturing Sector, by Various Indicators, 1975. T-4. Estimates of Capital Productivity and Net Social Benefit per Unit of Capital, by Plant Size, 1956-67. T-5. Growth of SMI Compared to Large Industry, 1956-75. T-6. Estimates of SMI Employment by Industry: 1956, 1966 and 1975. T-7. Institutional Financing. T-8. Fondo Financiero Industrial. T-9. FFI's Portfolio. T-10. Caja Agraria. Tables on Corporacion Financiera Popular: T-11. Distribution of Ownership Over the Period December 31, 1974 to December 31, 1978. T-12. List of Board of Directors and Alternates as of May 1, 1979. T-13. Evolution of the Staff at Headquarters and in the Regional Offices between January 1, 1977 and December 31, 1978. T-14. Organization Chart as of June 30, 1978. T-15. Summary of Supervision Activities during 1977. T-16. Summary of Supervision Activities during 1978. T-17. Summary of Resource Position as of December 31, 1978. T-18. Balance Sheets as of December 31, for the Period 1972-78. T-19. Summary of Past Operations from 1970 through 1978. T-20. Sources and Applications of Funds for the Period 1977-78. T-21. Income Statements for the Years 1972-78. T-22. Administrative Expenditures of Regional Offices 1977-78. T-23. Key Financial Indicators for the Years 1976-78. T-24. Aging of Arrears - Evolution between 1973 and 1978. T-25. Analysis of Loan Approvals from Inception through 1978. T-26. Analysis of Loan Approvals for 1978. T-27. Equity Investments Held as of December 31, 1978. T-28. Lending Program. T-29. Projected Sources and Uses of Funds for the Years 1979-82. T-30. Balance Sheet Projections for the Years 1979-82. T-31. Statement of Income and Expenses for the Years 1979-82. T-32. Key Financial Indicators for the Years 1979-82. 4. Estimated Quarterly Disbursement Schedule for the Bank Loan. 5. Related Documents Available in the Project File. MAP  COLOMBIA STAFF APPRAISAL REPORT THIRD SMALL-SCALE INDUSTRY PROJECT I. SMALL- AND MEDIUM-SCALE INDUSTRY IN COLOMBIA 1/ A. Economic Setting 1.01 Colombia has made substantial progress during the past two decades in the transition from a predominantly rural and agricultural economy made up of largely self-contained regions to a more integrated urban industrial economy. The productive base of the economy has been widened appreciably, and there has been substantial diversification of production in both the agricultural and industrial sectors. The country has become less dependent on coffee as a source of foreign exchange earnings, and fluctuations in domestic economic conditions resulting from unpredictable shifts in world coffee prices, while still considerable, have become more manageable. Colombia's population is becoming increasingly urbanized (66% in 1978), and the value added of the manufacturing sector is now about 20% of GDP. GDP grew at an annual average rate of about 6.0% over 1967-78, while the manufacturing sector expanded at a rate of around 7.0%, which was satisfactory but not exceptional either in absolute terms or in relation to the GDP growth rate as a whole. After a seven-year period (1967-74) of fast overall growth of 6.6% p.a. (Annex 3, Table 1), with manufacturing growing at 8.3% p.a., world recession and domestic stabilization policies slowed GDP growth to 4.4% p.a. 2/ over 1974-77, and manufacturing growth decelerated to 4.0% p.a. Both rates rebounded to about 8.5% in 1978. Capacity utilization began to rise in 1976 and reached 86% in 1978. Information on investment in manufacturing is not yet available for 1977 or 1978, but it is likely to show a strong resur- gence, as domestic demand has continued to expand rapidly. The current high capacity operations in much of the sector indicate a need for a substantial surge in investment over the next few years. Manufacturing employment has also grown satisfactorily (about 4.5% p.a. over 1964-73), with its share in total employment rising from 12.4% in 1964 to 15.5% in 1973, the latest year for which complete figures are available. Factory employment (i.e., in plants of 5 or more workers) grew at 5.2% p.a. over this period and increased from 6.2% to 8.1% of total employment. Indications are that the share of manufac- turing employment in total employment has increased further since 1973. Despite the progress made in industrialization during the past two decades, Colombia's economy is still based on a large traditional sector on which a relatively small modern sector is superimposed. 1/ See para. 1.04 for the definitions used in this report for small-scale and small- and medium-scale industry, respectively. 2/ Gross national income was growing much faster than the 4.4% estimated for GDP, due to rapidly growing incomes from illegal exports and improved terms of trade of coffee. Real income appears to have risen as much as 7.8% per year during 1974-77, resulting in heavy demand pressures and contributing to inflation.  -2- 1.02 The dramatic expansion of manufactured exports that began in 1967 ended in 1975 with the world recession and has since been reversed, as the unusually large inflows of foreign exchange due to higher coffee prices and receipts from illegal exports have led to government policies, resulting in a sharp appreciation of the Colombian peso in real terms. Rapid growth in coffee exports in the 1975-78 period, together with strong domestic demand and lower incentives for exporting, reduced the share of manufacturing exports in total exports to about 14% in 1978. The trade-weighted real effective exchange rate index for non-coffee exports dropped from 100 in 1970 to 79 in 1978, with much of the decline occurring in 1976 and 1977. As a share of manufacturing output, exports reached 6.5% in 1974, but were down to 4% by 1977. It is estimated that manufactured exports decreased 10% in real terms from 1974 to 1978. 1.03 Relatively uneven income distribution, low productivity and wages, and scarcity of permanent jobs traditionally have been problems in Colombia. Along with accelerating inflation, income distribution may have worsened during the early 1970s as real wages in many sectors of the economy, including manufacturing, stagnated or fell at a time when average income per capita was rising. Increases in employment were probably not sufficient to result in significant improvements in family incomes at the lower end of the scale. Since about 1975, however, some recovery may have occurred, especially in the rural sector, as high coffee prices have boosted incomes of producers and workers, and on average more members of family units have been employed. Unemployment has generally decreased since 1976, and in early 1979 it was less than 10%, i.e., lower than at almost any time during the last decade. There are a number of indications that Colombia's traditional excess supply of labor is lessening, due mainly to a sharp decline in the population growth rate, a flow of emigrants to Venezuela and elsewhere, and the rapid expansion of the economy. B. Structure and Characteristics of SMI 1.04 Economic impact. Small- and medium-scale industry (SMI) plays an important role in Colombia's industrial sector and deserves special atten- tion from an economic point of view. There is no generally accepted defini- tion for either small- or medium-scale industry in Colombia. For the purpose of this sectoral analysis, SMI is defined as manufacturing enterprises employ- ing 5 to 99 workers, i.e., the definition adopted by the National Department of Statistics (DANE) and upon which the national statistics are compiled. On the other hand, the frequently used Bank definition of small-scale industry (SSI) is US$250,000 or less in fixed assets. While the two concepts are not identical, they are similar, and the SMI sector analyzed in this chapter is representative of the SSI target group of the project (para. 4.13). To analyze the performance of SMI, a dynamic analysis, tracking the same set of firms classified as SMI in the initial survey to see how they perform over time, is preferable to the typical static analysis of comparing firms classi- fied as SMI or large industry in different years. Unfortunately, the tracking. data necessary for the dynamic analysis have rarely been collected; except as noted in para. 1.08, available comprehensive data on SMI in Colombia (and most other countries) permit only the static analysis.  - 3 - 1.05 In 1978 SMI employed some 45-50% of the labor force in factory manufacturing, 1/ produced about 25-30% of manufacturing output, and employed 20-25% of the manufacturing sector's assets. Although widely distributed among all industrial subsectors, SMI is particularly important in the produc- tion of non-durable consumer goods, and to a lesser extent in the manufacturing of intermediate goods for large-scale enterprises. Food (15%), apparel and footwear (16%), and metal products industries (excluding machinery and equip- ment) (11%), accounted for over 40% of the SMI employmenlt in 1975 (Annex 3, Tables 2/3). SMI accounts for over 20% of factory employment in all major industrial categories except the production of beverages, tobacco, textiles and petroleum/coal. 1.06. In industries where technical economies of scale are not overly dominant, SMI has some distinct advantages for a developing economy such as Colombia's. Average labor intensity and capital productivity of SMI in Colombia are relatively high 2/ (Annex 3, Table 4), as its investment cost per job is only about one-third that of large-scale manufacturing (plants of 100 workers and up). Further, the type and location of the employment generated differ, with SMI using a greater proportion of unskilled workers; a higher share of SMI is found in rural areas and small urban centers, which serves well Colombia's rather segmented local markets. 1.07 SMI makes other less direct contributions to industrial development. It plays a significant role in the formation of entrepreneurial talent, providing both a training ground and a context in which underutilized skills can be put to work. The majority of large firms in Colombia have previously been among the ranks of SMI. The presence of SMI helps to increase the effi- ciency of the entire manufacturing sector, 3/ due to both the complementary relation it has to some large-scale producers and the competition it creates for others. Large industry and SMI both benefit when the latter provides intermediate inputs for the former under subcontracting or other arrangements. While many large firms earn high returns on capital in Colombia, 4/ the presence of smaller competitors has tended to improve efficiency and thereby reduce product prices in many sub-sectors. Some larger firms reportedly were induced to export in the early 1970s by the difficulty of competing with SMI 1/ These data on factory manufacturing exclude productive units of less than five workers, and household industries. In 1973, this "cottage-shop" sector accounted for nearly half of all manufacturing employment. 2/ Based on 1976 data for Colombia, the ratio of market price to opportunity cost (shadow price) is greater for labor than for capital, implying that capital is "underpriced" in the market relative to labor. 3/ Capital productivity of large plants was observed by Todd to be higher in those three-digit industries in which a significant share of output came from SMI. (John Todd, Efficiency and Plant Size in Colombian Manufactur- ing, Yale Ph.D. dissertation, 1972.) 4/ See A. Berry (editor), Essays on Industrialization in Colombia, Chap. 2, forthcoming.  - 4 - in the domestic market. 1/ Promotion of SMI can also be an effective means of increasing output in the short run, as the lead time required by SMI to increase production is generally shorter than for larger firms. 1.08 Performance. The growth performance of Colombian SMI firms during the 1970s was reasonably good. The increase in employment in SMI enterprises be- tween 1970 and 1975 amounted to 3.2% p.a., while value added in constant prices rose 2.5% p.a. The corresponding figure for larger industries was 7.5% for employment and 4.4% for value added (Annex 3, Table 5). However, these figures may be misleading, as the contraction or expansion of individual firms can lead to a change in their classification over time, so that the enterprises making up the SMI or larger industry group in one year may be markedly differ- ent in subsequent years. A study by the Colombian statistical office, DANE, of the growth patterns of a sample of firms defined as SMI in 1970 indicated that through 1975 there had been an increase in employment of these firms of 4.5% p.a. For a sample of larger firms in the study their rate of increase in employment was only 3.3% p.a. These data suggest that the higher growth rate for the entire group of larger firms in the economy in those years was due to the appearance of new firms in this category, most of which came from SMI enterprises "graduating" into the larger group. In general, SMI has been in a good position to take advantage of the rapid growth of domestic demand during the 1970s due to SMI's flexibility in modifying output, and the fact that it is less constrained by national market size. Aggregate SMI employment and output growth has been most rapid in the metal working industries (Annex 3, Table 6), with the average annual increase in employment above 6% in recent years. Wooden furniture, leather, and chemicals also recorded rapid employment growth. 1.09 Average labor productivity 2/ in SMI, while above that of the economy as a whole by about 45% in 1975, rose little during the decade 1966-75; as a result, increases in output were associated with increases in employment rather than productivity gains. This is a desirable result for a labor surplus economy, ceteris paribus, as it implies a greater relative use of abundant labor than scarce capital. Between 1966 and 1975 the annual growth rate in labor productivity was 1.8%, which is relatively low for a country that has been growing as rapidly as Colombia and lower than the 2.8% of the previous decade (1956-66). The rapid growth of factory employment in the last decade was fostered by stagnant real wages and the expansion of export-oriented industry. After allowing for differences in labor force composition, labor productivity is probably 20-35% lower in SMI than in large industry, a differ- ential which appears to have been rather stable over the years. A wage differential of 10-30% 3/ between SMI and large industry exists, due mainly to (a) the effects of unions in larger plants, and (b) greater efforts by larger industry to decrease job turnover by paying higher wages. 1/ Based upon a survey of Colombian SMI, including some large firms that compete with SMI, by the Bank's Development Economics Department, report forthcoming. 2/ Defined as value added per worker. 3/ Based on F. Bourguignon, "Poverty and Dualism in Developing Countries: The Case of Colombia," mimeo, 1978.  -5- 1.10 The nature of growth in SMI firms is likely to change in the future, however, as it appears that the labor surplus condition typical of the Colombian economy may be much less pronounced in the future than it has been in the past. In this case, higher labor productivity will then become necessary for continued success, which will depend on the ability, especially of smaller firms, to increase their capitAl and modify their production processes. Adequate access to credit and to technical assistance will become increasingly important. If the experience follows the pattern of other countries (e.g., Japan and Taiwan) which have passed through this transition, real wages will advance rather quickly. Although this transition is likely to be gradual, it is nevertheless important that the requisite institutional structures to aid SMI be strengthened in advance of the tightening of the labor market. C. Growth and Development Constraints 1.11 While many efficient SMIs have thrived, others which also have good potential are hampered by a number of obstacles. Inadequate access to credit on terms suitable to finance either fixed asset purchases or working capital seems to be among the most important of SMIs' development constraints. Due to their weaker bargaining position, many SMIs have to finance sales for 90 days, yet are forced to pay cash for raw materials. With the prevailing tight credit situation in Colombia (para. 2.02), a manufacturing enterprise often has to pay its suppliers in advance and/or make credit available to its customers in order to increase its sales and capacity utilization rate. Moreover, the problem is compounded by reluctance on the part of financial institutions to lend to small industry because of (a) greater risks of failure; (b) lack of acceptable guarantees; (c) higher administrative costs in process- ing and handling relatively small loans; (d) lack of reliable accounting information; and (e) limited possibilities for additional income from other banking transactions of SMIs. Recent surveys I/ have shown that the institu- tional credit/value added ratio was 0.11 for SMI as against 0.32 for large industry. On average, credit provided about 70% of the sources of funds for large industry, but less than 45% of the funds of SMI, indicative of the easier access to cre'dit of large firms. 1.12 Inadequate and expensive raw materials supply affects product quality and capacity utilization, and weakens SMI's ability to compete with larger industry. The purchasing capacity of SMIs is often too small for effective negotiations with raw material producers and usually compels them to purchase through middlemen charging mark-ups of 25-40%. Raw material availability problems also reflect both disorganized and monopolized markets. Conglomerates of large input and final goods producers has been able, in some cases, to squeeze out SMIs with a view to consolidating their'market control. Liberalization of raw material import restrictions and support for SMI in input producing industries could help to maintain competitive markets for those products and thus contribute to the development of SMIs. 1/ By Corporacion Financiera Popular and FEDESARROLLO (a Colombian economic research association).  -6- 1.13 On the other hand, several innate advantages have helped a number of SMIs in several subsectors to compete effectively with larger industry. Lower labor costs give SMI some advantage, as does the ability to adapt low-cost techniques to their needs. In some industries, the ability to interact closely with clients assists productive innovation and adaptation. Capacity to react quickly to evolving market demand has been important in recent years. Finally, some SMIs are protected by regional market segmentation in Colombia which raises distribution costs for large-scale producers. However, possible changes in policy to remove many of the above constraints to SMI growth, encourage competition and promote appropriate factor prices, could bolster SMI and likely improve overall industrial sector performance. D. Government Policies 1.14 Colombia's 1975-78 development plan gave emphasis to rural develop- ment, economic decentralization, and more equal income distribution. For the first time, the plan emphasized SMI's economic and social importance in creating employment opportunities and furthering regional development. On the other hand, policy-making mechanisms to promote SMI have not been well coordinated, and public and private institutions providing assistance to SMI have suffered from a lack of coordination and long-term planning. There is a need to institu- tionalize within the government responsibility for (i) gauging the impact of the existing and proposed government policies and programs on SMI; (ii) ensur- ing a sufficient and timely flow of resources to SMI; (iii) taking steps conducive to countrywide promotion and coordination of technical and financial services to SMI; and (iv) establishing a coordinated policy framework to promote SMI. 1.15 While the National Planning Department is charged with formulating national policies and budget priorities, it cannot be expected to undertake on a continuing basis detailed analyses of SMI policy alternatives. Corporacion Financiera Popular (CFP), with its long experience and specific objectives of supporting SMI, is now in the process of rebuilding its research department. A main topic would be to review the current practice and proposed policies regarding availability of and access to credit for SMI. Another important component of such research would be geared to a fuller understanding of the character, regional distribution, and growth process of SMI. Sufficient raw data now exist, including those of CFP, DANE, and other public entities, to permit considerable headway along these lines. Also, the impact on SMI of the issues described below will be studied on an ongoing basis by CFP, and the results of such studies will be available to the government for policy decisions. (1) Raw materials availability and price. SMI complaints about difficulties in raw materials availability and high prices are frequent. One solution, cooperative purchasing by SMI firms, has not yet had much success in Colombia. Facilitating and encouraging raw material imports for SMI would both counteract monopoly power of raw material suppliers in some subsectors and alleviate scarcities suffered by SMI.  (2) Access to improved technology nd to reasonably priced machinery. The competitive position of SMI often rests on the ability to acquire productive equipment which can be used in more labor- intensive and smaller-scale production processes than that used by large firms. Equipment is frequently second hand, adapted or renovated., Policy steps which could assist SMI in the6e respects include: (a) encouragement of information flows with respect to available technology, including courses, arranging meetings, circula- tion of published information, and organization of information for consultation; (b) support for research in relevant technological innovations and adaptations; and (c) greater technical assistance in the purchase of both imported and locally procured used machinery. (3) Government procurement. The public sector, including many impor- tant decentralized agencies, accounts for a large volume of purchases of manufactured goods, many of which can be produced by SMI (e.g., office supplies). To facilitate SMI bids, the government could (i) give wider publicity to available procurement contracts and simplify bidding procedures; (ii) reduce for SMI the typical 6-12 months delays in payment of bills that cause more difficulty for SMI than for larger industry; and (iii) focus on relatively small but still efficient contract sizes. (4) Industrial parks have received much attention as a tool of regional decentralization and SMI development. CFP plans to use an industrial condominium-type approach, whereby CFP first finances the construction of industrial parks and subsequently the purchase of individual factory shells. Industrial parks should be an efficient tool for moving industry from the major urban centers to nearby industrial zones, but are not likely to provide sufficient motivation by themselves to lure industry to the less developed regions of Colombia. E. Prospects 1.16 SMI contributed significantly to the growth of output and employment in Colombian manufacturing during 1970-78 and, if given sufficient access to credit and an adequate economic environment, can play an important role in the future. It appears that the number of small entrepreneurs increased substan- tially in the 1970s, due mainly to better formal education and the substantial experience managers and employees accumulated in both SMI and large industry, which enabled them to establish their own businesses. A continuation of the recent healthy overall economic growth is expected, and the increasing com- plexity of the economy, coupled with rising income levels, is creating signi- ficant demands for new products. The Colombian economy is expected to continue to grow at about 6-7% per year through 1983, and factory manufacturing at 7-9%. Past experience indicates that the value added of firms in the SMI size range should also be growing at about the average growth rate for the manufacturing sector. With SMI's output/capital ratio probably continuing to be somewhat above the total factory manufacturing sector's average of about 0.5, i.e., at least 0.6 to 0.7, the net additional capital 1/ required to sustain an SMI growth rate of 7-9% per year would be as follows: 1/ Not including depreciation replacement.  -8- Table 1.1: NET ADDITIONS TO CAPITAL STOCK REQUIRED a/ (Billions of 1978 Pesos) 1979 1980 1981 1982 1983 7% output growth 3.4 - 4.0 3.6 - 4.2 3.9 - 4.5 4.2 - 4.8 4.4 - 5.2 9% output growth 4.4 - 5.1 4.8 - 5.5 5.2 - 6.0 5.6 - 6.6 6.2 - 7.2 a/ Projections based on estimated value added of SMI of Col$34 billion in 1978. This implies average net investment increases by SMI of around US$105-140 million p.a. during 1979-83, about one-third of which would represent fixed asset investments, and the remainder working capital. The proposed Bank loan (para. 4.12) would finance the equivalent of 8-10% of the investment during the estimated three-year disbursement period. II. FINANCIAL AND TECHNICAL ASSISTANCE TO SMI A. The Financial Sector and Policy Framework 2.01. Colombia's relatively well-diversified financial system consists of the central bank (BR), commercial banks, development banks (financieras, including CFP), one public industrial investment bank (IFI), one public agricultural bank, one public mortgage bank, several savings and loan associa- tions, insurance companies, mutual funds and two stock exchanges (in Bogota and Medellin). The system's structure has been shaped by past government policies favoring institutional specialization. The existing 26 commercial banks (23 private and 3 public) dominate the regulated financial system, accounting for 42% of the institutional portfolio in December 1978 (Annex 3, Table 7). The savings and loan associations (17.5%) and the financieras (16.3%) rank next, followed by the mortgage bank (10.6%) and the agricultural bank (9.7%). Lending by these institutions is supplemented by BR rediscounting facilities which have been established to finance particular sectors. For industry the most important of these funds are (a) Fondo Financiero Industrial (FFI), which rediscounts loans made by financial institutions to mostly SMIs; (b) Fondo de Inversiones Privadas (FIP), which generally serves firms larger than those covered by FFI; and (c) Fondo de Promocion de Exportaciones (PROEXPO), which provides primarily export credit to a wide range of exporters. 2.02. For many years the financial system has been subject to strict government control through a complex system of reserve requirements, forced investment regulations and official interest rate ceilings that vary according to the various categories of borrowers and savers, and that are established by Colombia's Monetary Board (Junta Monetaria). In the past three years, the  -9- government has reinforced measures aimed at restricting money supply growth, principally by establishing reserve requirements for commercial banks of 46% of sight deposit levels existing on January 31, 1977, and of 100% of deposits in excess of these levels. The resulting average reserve requirements for commercial banks s presently around 80%, which has strongly limited commercial banks' ability to expand credit. Consequently, about 78% of t e portfolio increase of commercial banks in 1977 and 1978 was financed with BR loans or rediscounts from the development funds. 2.03. As a result of the tight monetary policies, institutional credit for most economic activities has been severely curtailed. While the consolidated outstanding portfolio of the financial system grew in real terms at an average rate of 4.1% p.a. in 1973-78, the industrial loan portfolio increased by only 1.8% p.a. in the period, compared to a 7.4% p.a. average growth rate of industrial value added. The private financieras have continuously increased their share of the total institutional industrial portfolio, from 42.9% in 1973 to 53.3% in 1978 (see Table 2.1 below). Still, the share of the industrial portfolio within the financial system decreased from 34% in 1973 to 30.7% in 1978, and industry has been forced increasingly to tap the extrabank (largely non-institutional) credit market. On the other hand, in February 1980 the Monetary Board removed the 100% marginal reserve requirement and increased over a 5-month period the average reserve requirement to 50%; this move should result in a more rapid expansion of commercial credit to industry and other sectors. Table 2.1: INUOSThWL VlUlIcliG (1973-78) A. rnetitutfomal Out*.ewdinp Portfolio to Indutrv (in Col$ iloim,. Current prices) Couvo,-,,.l Banks Total Groth Rates *t * On FF LTr7! Ind.strial Total lnstit. F.ude PRoExPO Tot.