Turkey - Prospects for small-medium scale industry development and employment generation (Vol. 2) : The main report

FILE COPY Report No. 2913-TU Turkey - Prospects for Small-Medium Scale Industry Development and Employment Generation (In Three Volumes) Volume II: The Main Report September 10, 1980 EMENA Projects IDF Division FOR OFFICIAL USE ONLY Document of the World Bank 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 Turkish Liras per US$ 1959-70 - average 09.00 1970 - average 11.50 1971 - average 14.90 1972 - average 14.15 1973 - average 14.15 1974 - average 13.90 1975 - average 14.45 1976 - average 16.05 1977 - before March 16.05 - March - September 17.50 - September - December 19.45 1978 - January - March 19.45 - March - December 25.25 1979 - January - April 25.25 - April - June 26.35 - June - December 47.10 1980 - January to May 70.00 - June to July' 78.00 - August to date 80.00 ABBREVIATIONS AND ACRONYMS DESIYAB Devlet Sanayi ve Isci Yatirim Bankasi DYB Devlet Yatirim Bankasi HB Halk Bank IED Department of Organized Industrial Zones and Small- Scale Industrial Estates at Ministry of Industry and Technology MIT Ministry of Industry and Technology PENDIK Leather Research and Training Center SBAG Small Business Advisory Group SEE State Economic Enterprises SIDO National Small Industry Organization SMI Small-Medium Scale Industry SPO State Planning Organization SSI Small Scale Industry SYKB Sinai Yatirim ve Kredi Bankasi A.0. TBTAK Marmara Institute for Science and Technology (at Gebze) TSKB Turkiye Sinai Kalkinma Bankasi A.S.  FOR OFFICIAL USE ONLY This report is based on the findings of a mission composed of Messrs. G.C. Maniatis, C. Goderez, E. Kepper and S. Banerji, which visited Turkey for approximately four weeks in November-December 1978, and subsequent extensive work at headquarters. The mission visited industrial estates and establishments in Ankara, Istanbul, Izmir, Gaziantep and Kayseri, and had discussions with Government officials and the business and banking communities. The cooperation and assis- tance of all persons and agencies contacted is gratefully acknowledged. The report draws on a number of studies, referred to in the text. This report has been discussed with, and cleared for released by, the Government. This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contenst may not otherwise be disclosed without World Bank authorization.   TURKEY PROSPECTS FOR SMALL-MEDIUM SCALE INDUSTRY DEVELOPMENT AND EMPLOYMENT GENERATION VOLUME II TABLE OF CONTENTS Page No. I. INTRODUCTION. ........... .................................. 1 Scope of the Survey ...... ............................. I Definition ........................................................ I Data Base. . .. ...... . 1 A. THE ENVIRONMENT II. THE URBAN SETTING ....... .................................. 2 Urbanization Trends ... ................................ 2 The Spatial Incidence of Urban Poverty ................ 3 Urban Unemployment ...... ................ 3 Policies to Foster Urban Employment Generation ........ 5 Measures to Promote Industrial Employment in Regional Urban Centers .. 5 III. THRUST OF THE INDUSTRIALIZATION EFFORT . . 7 Role of Industry in the Economy .. 7 Industrial Policies... . .... 7 New Policy Initiatives. ........ .. 7 IV. INSTITUTIONAL FINANCING ................................... 9 Structure of the Banking System ................. 9 The Role of Commercial Banks .......................... 9 Financial Assistance to Small-Medium Scale Industry ... 10 The Halk Bank ....... .................................. 11 SYKB ................... 13 TSKB .................................................. 14 B. MORPHOLOGY OF SMALL-MEDIUM SCALE INDUSTRY V. STRUCTURAL FEATURES OF SMALL-MEDIUM SCALE INDUSTRY ........ 15 The Pattern of Ownership ..... ......................... 15 Size Distribution ...... ............................... 16 Sectoral Distribution ...... ........................... 17  TABLE OF CONTENTS (Continued) Page No. Regional Distribution - .. ... . ........... so. . ............... . 19 Distribution of Employment .......................... 21 Distribution of Value Added ......................... 24 Capital Intensity o.s...o..................................o .... 25 Labor and Capital Productivity ...oo ........ ...... .. 26 VI. ORGANIZATIONAL AND OPERATIONAL FEATURES OF SMALL-MEDIUM SCALE INDUSTRY .o..o.o .... . ........ ....*.*..*.*.*....... ...*... * .......*. *...* 27 Form of Ownership ..o ........ .. . .... .. .......... 27 Origin of Entrepreneurship .......................... 27 Specialization of Management Functions .............. 28 Labor Skills and Training ........................... 28 Forms of Employment .... . ... . .. ... . . . . . . .................... . 30 Labor Mobility .............. .......0......................a........... 30 Wages .. ......................................... 30 Labor-Management Relations s.ooo. . . 32 The Role of the Guilds ............... o.... o. 32 Equipment .................................... .................... 33 Workshops so....................... ................ o...... o..o....so 33 Marketing and Distribution ................... ......... 34 Profitability .......... ............................................ 34 Cooperatives .............................. . . ......... 35 C. THE INSTITUTIONAL AND POLICY FRAMEWORK VII. GOVERNMENT PROGRAMS IN SUPPORT OF SMALL-MEDIUM SCALE INDUSTRY ................................................ .......... 36 Role of the Government ................. .............. 36 The Industrial Estates Development Program ......... 37 The SIDO Program .................................... 39 Industrial Training ..... .................. .......... 41 Term Credits to Manufacturing Cooperatives .. . 42 VIII. POLICY INSTRUMENTS AND THEIR IMPACT ON SMALL-MEDIUM SCALE INDUSTRY ........ ................................ 43 Investment Incentives ............. . .................. . 43 Export Rebates ...................................... 44 Export Financing . ......................... ... ....... 45 Procurement ......................................... 45 Subcontracting so..................................................... 45  TABLE OF CONTENTS (Continued) Page No. D. SECTORAL ISSUES AND PROPOSED REMEDIAL ACTION IX. PRO'BLEMS HBMPERING THE EFFICIENCY AND GROWTH OF SHALL- 46 NEDIUM SCALE INDUSTRY . Internal Constraints ....46 External Constraints .... 49 Impact on SMI .....49 X. PROPOSED TECHNICAL ASSISTANCE PROGRAMS ... . 50 Technical Assistance to the Halk Bank for Institutional Building ... . 51 Extension of Technical Assistance to Garment! Making-up Subsector .. ........ 51 Extension of Technical Assistance to Small-Medium Scale Industry .................................... 51 Technical Assistance for Development of a Subcontracting Exchange . .52 Potential ................................. . 52 Conceptual Framework . .53 Approach .................................... 53 Development of an Industrial Research Service . . 54 E. SURVEY OF SMALL-MEDIUM SCALE INDUSTRY IN PRINCIPAL SUBSECTORS XI. THE FOUNDRY INDUSTRY .................................... 55 Overview of the Industry ..... ....................... 55 Organization ..................................... ....... 55 Employment .......................................... 57 Wages .... .. . . ...................... 58 Performance ....... 58 Exports ...60 Problem Areas ........... .... 61 Small Foundries . ...... 61 Medium and Large Foundries .62 Prospects.. 63 Strategy for Subsector Development . .64 Concluding Remarks ...65 XII. THE ENGINEERING INDUSTRIES.. 66 Role of Private Sector ... 66 Structural Trends ....67 Value Added and Productivity . . .69 Investment ....71 Capital Intensity and Capital Productivity .72 Exports ....................... 72 Technological Considerations. 73 The Automotive and Consumer Durable Industries 73  TABLE OF CONTENTS (Continued) Page No. The Automotive Industry ............. .. .............. 73 Industry Profile ...... ......................... 73 SMI Potential .................. 74 The Refrigerator Industry ..... ...................... 75 Industry Profile . .. ............. 75 Implications for the SMI Sector ............. ... 77 Recommended Export Expansion Program ........ ... 77 Plan Objectives in Engineering Industries ........... 77 Prospects for SMI Development in the Engineering Industry ... 78 Promising Product Areas for SMI Involvement ......... 80 1. Low Value--Low Technology Standard Products ..... 80 2. High Technology--High Unit Value Products ....... 80 3. Products Manufactured under Ancillary and Sub-Contracting Relationships .............. ... 81 4. Production of Selected Capital Goods ............ 82 Strategy and Policies to Promote SMI in the Engineering Sector . ........ . ...................... 83 XIII. THE LEATHER PROCESSING AND FOOTWEAR INDUSTRIES .. ........ 85 The Leather Processing Industry ......... .. .......... 85 Overview ....................................... 85 Problems Encountered in Raw Hide and Skin Production ..... ......................... 86 Production and Consumption of Hides and Skins ...................................... .. 86 Leather Production ..... ........................ 87 Leather Production Facilities ......... ......... 88 Technology ................. .. .................. 89 Leather Products .............. .. ............... 90 Leather Garments .............. .. ............... 90 Prospects and Strategy for the Development of the Leather Garment Industry .............. 92 The Footwear Industry .............. .. ............... 93 The Industry ................... ................ 93 Structure and Performance ......... .. ........... 93 The Domestic Market ............ .. .............. 96 Exports .................... .................... 97 Export Incentives .............................. 97 Development Strategy for the Footwear Industry.. 97 Government Policy .......................... 98 Action Program for Restructuring the Footwear Industry ...... .................. ............. 99 XIV. THE WOOD-BASED INDUSTRIES ............................... 100 Introduction ........................................ 100 The Resource Base ................ .. ................. 100 Sectoral Overview ................ .. ................. 100  TABLE OF CONTENTS (Continued) Page No. Sawn-wood Industry ................................. 102 Structure and Performance ....................... 102 Supply and Demand ............................... 103 Prices .......................................... 103 Strategy and Policies ........................... 104 Wood-based Panels .................................. 105 Structure and Performance ....................... 105 Market Aspects .................................. 106 Furniture Industry ................................. 108 State of the Industry ........................... 108 Organization and Performance .... ................ 108 Policy and Recommendations ......................... 109 XV. THE GARMENT INDUSTRY ................................... 111 Organization ....................................... 111 Production and Demand .............................. 112 Wages .......... .................................... 113 Exports and Export Potential ....................... 114 Sectoral Issues and Problems ....................... 116 Technical Assistance Requirements .... .............. 117 Subsector Strategy ................................. 118 XVI. PROMISING SUBSECTORS FOR SMALL-MEDIUM SCALE INDUSTRY DEVELOPMENT .......................................... 120 F. STRATEGY FOR THE DEVELOPMENT OF SMALL-MEDIUM SCALE INDUSTRY XVII. OBJECTIVES AND STRATEGY - A SUGGESTED APPROACH .122 Contribution of SMI to Industrial Development 122 Urban Employment Generation .123 Efficiency Improvement and Growth .125 G. RECOMMENDATIONS XVIII. SUGGESTED ACTION .128 Policy Measures .128 Technical Assistance .130 Studies .130   I. INTRODUCTION Scope of the Survey 1.01 This survey attempts to establish the prospects for growth and the attendant potential for urban employment generation of the small and medium scale industry (SMI); to assess the policy framework and its impact on the development of the SMI; to ascertain the adequacy of the existing infrastruc- ture (e.g., industrial estates, technical assistance delivery systems); to identify basic technical and financing problems hampering the efficiency and growth of the SMI; to suggest suitable financial intermediaries for possible Bank support; and to formulate a technical assistance package that could be part of a DFC operation. Definition 1.02 For the purpose of this report, small scale industry is defined as comprising establishments employing up to 50 workers; medium scale between 50-200; and large scale 200 or more workers. The only demarcation in manufac- turing in Government statistics is that between the organized sector (large industry) which comprises establishments employing 10 or more workers, and the unorganized sector (small industry), which includes establishments employing up to 9 persons. Such a broad classification of manufacturing industry is patently inadequate and provides very limited information on the SMI. Adoption of a more operational definition of small and medium scale industry to reflect present realities undoubtedly would be more serviceable. Data Base 1.03 Available information on SMI, a segment accounting for some 99% of the total number of establishments, 55% of employmcat atd 41% of value added in the manufacturing sector, is out of date and fragme.ntary. The more syste- matic, though far from complete, source of data is the Census of Manufacturing and Business taken in 1970. The lack of comprehensive and sufficiently detailed time series by size of establishments and subsector on important economic indicators (e.g., investment, production, value added, exports, capital/labor ratios, labor and capital productivity, profitability, capital structure, etc.) makes the study of trends in the SMI sector and the assessment of the performance of the SMI extremely difficult. Building up a reasonably adequate and reliable data base and ensuring a constant inflow and timely processing of information would enable policy-makers to better appreciate the contribution and prospects of SMI and would greatly facilitate policy formulation to realize its potential. Until a comprehensive census is taken, our knowledge of the vital statistics of the SMI sector is bound to remain incomplete, our analysis limited and inadequate, and our approach largely impressionistic.  - 2 - A. THE ENVIRONMENT II. THE URBAN SETTING Urbanization Trends 2.01 Turkey is rapidly becoming a predominantly urban and industrial country. Urban population has been growing at over 5% per annum, or double the national average, with the urban portion of total population increas- ing from 18% in 1950 to 35% in 1975. The increase is primarily due to the continuing high rural-urban migration. 1/ The seven largest urban centers (over 250,000 population) increased their share in the national population from 35% to 41% between 1965-75, particularly at the expense of the smaller urban centers in the 10,000 - 100,000 size category, suggesting that rural- urban migration has tended to bypass the smaller centers in favor of larger cities. 2/ These deep-seated urbanization trends can be expected to persist. Urban population growth is projected at 5.5% per annum for the next decade and it is anticipated that 75% of the population might be urban residents by 1995 (Appendix Tables 1.1-1.7). 1/ The rapid increase in urbanization is the result of the structure of Turkish economy and social changes. A combination of "push" and "pull" factors have contributed to the sustained increase in rural-urban migra- tion since 1950. The "pull" factors appear to be: (a) higher wages paid by industry to urban workers; and (b) job opportunities,for migrants from rural areas as well as smaller towns, reinforcing the income-pull factor. "Push" factors include: (a) the persistent rural unemployment and under-employment; (b) continuing farm mechanization and consolidation of holdings, which tends to exacerbate rural unemployment; (c) very limited availability of non-farm job opportunities in rural areas (only 12% of rural workers are employed in manufacturing); (d) the impetus from Government industrial investment in secondary urban centers, and the development of road networks and bus transportation with low passenger fares. Finally, the highly skewed income distribution pattern in rural areas (income in over 50% of agricultural households is less than half the national average), better living conditions, and educational possi- bilities for offspring in the cities, help enlarge the pool of potential migrants. These are all potent forces and the present trends can be expected to continue. Despite recent improvements in inter-regional income distribution following the Government's decentralization policies and emphasis on regional equality, average household incomes continue to be substantially higher in the three larger urban centers. The recent decline in external emigration is unlikely to slow down perceptibly rural-urban migration as long as the other "push" factors persist. 2/ Istanbul, Ankara, Izmir, Bursa, Eskisehir, Adana and Gaziantep.  - 3- The Spatial Incidence of Urban Poverty 1/ 2.02 The incidence of urban poverty by urban size category is shown in Table 1. Table 1: INCIDENCE OF URBAN POVERTY Population in No. of Poor Poverty Group People Urban Size Category % (000) Istanbul 5.2 132 Ankara 8.0 136 Izmir 7.8 50 Centers 100,000 persons or more 13.9 561 Centers between 50,000 - 100,000 persons 16.8 268 Centers between 25,000 - 50,000 persons n.a. n.a. Centers between 10,000 - 25,000 persons 18.7 84 Source: 1975 Census data; 1973 income distribution data. The data clearly indicate an increasing concentration of urban poor in the smaller centers--or that the incidence of urban poverty is inversely related with the size of the urban center. Thus, in centers of 100,000 - 500,000 in size, the incidence of poverty is twice that of the metropolitan centers, and in smaller centers even greater. The higher incidence of poverty in secondary urban centers reflects basically the greater difficulty in securing employment due to the persistent (and widening) gap between available employment oppor- tunities and growth of the local labor force. The situation can in part be alleviated by accelerating industrial development in these regional centers, with emphasis on activities with relatively low capital intensity. Urban Unemployment 2.03 Urban unemployment exceeded 10% of the total labor force in 1979 2/, as a result of reduced external emigration, worsening economic conditions, and sluggish growth of productive employment opportunities. Registrations with 1/ The urban poverty group is defined as households with incomes less than one-third the average real national income, which in 1976 prices was about TL 31,000. 2/ This implies over 1.5 million unemployed in urban areas. Total unemploy- ment was close to 20%. Reliable and up-to-date information on the level of urban unemployment and its structure (e.g., distribution by industry, skill, sex) are not available. The timely compilation of such data in sufficient detail certainly deserves the Government's attention.  - 4 - employment exchanges (which should be viewed with extreme caution due to uneven registration practices) indicate that urban unemployment in the 67 urban centers increased on average by 17% a year during 1970-77, with annual rates ranging from -0.5% to 40.8%. The three largest cities (Ankara, Istanbul and Izmir) show below average growth rates (11%) but have high absolute levels of unemployment, accounting for 22% of total unemployment in 1977. Twenty other major regional centers 1/ account for another 53% (Appendix Table 1.8). Unemployment growth rates have been particularly sharp in the East and South- East region, followed by the Mediterranean and Black Sea regions (Appendix Table 1.9). In general, the higher unemployment growth rates in most regional urban centers partly reflect a lower base, but also the greater difficulty in securing employment for the reasons mentioned in para. 2.02 above in connection with the also observed higher incidence of poverty in secondary urban centers. Unemployment (1976-77 averages) for all Turkey appears to be particularly high among primary school graduates (54%), followed by middle level (26%) and barely literate or illiterates (19%). Unemployment among persons with higher education was only 1% (Appendix Table 1.10). It is also interesting to note that unemployment among the 14-19 years age group averaged 18% for the country as a whole. Given the shortages in particular skills, this pattern suggests in part the importance of providing a minimum number of years of schooling and of upgrading technical skills to enhance the employability of the unemployed segment of the labor force (paras. 6.06-6.09, 9.06). 2.04 Employment in manufacturing increased from about 1.0 million in 1962 to about 1.7 million in 1977, or by 81%. 2/ In contrast, employment in services increased from 1.7 million to 3.6 million, or by 110%, during the same time span. The slower growth of manufacturing employment is a reflection of the high investment cost per job as a result of industrialization policies based on the development of high technology and capital intensive industries, particularly in the public sector. It is important to note, however, that the annual growth rate of employment in services has been declining since 1972, dropping from 5.4% during 1962-72 to 4.6% in 1972-77. This suggests that employment opportunities in this sector are increasingly becoming circumscribed. With the continued faster increase of the civilian labor force than employment (Appendix Table 1.12), the inauspicious prospects for increasing external emigration, and the persistent rural-urban migratory trends, structural unemployment is likely to swell, creating a nagging social problem and present- ing policy makers with a major challenge. 1/ Adana, Afyon, Antalya, Aydin, Balikesir, Bursa, Elazig, Eskisehir, Gaziantep, Hatay, Kayseri, Kocaeli, Konya, Kutahya, Malatya, Sakarya, Samsun, Sivas, Trabzon and Zonguldak. 2/ Employment in manufacturing increased by 4.0% during 1962-72 per annum and 4.2% annually during 1972-77 (Appendix Table 1.11).  - 5 - Policies to Foster Urban Employment Generation 2.05 The Government is keenly interested in accelerating the process of urban job creation for unskilled and semi-skilled labor. In this effort, the role of the maniufacturing sector looms large, as it is expected to absorb 30-41% of the projected increase in labor force, or some 75,000 workers per year, dtring the next five years. This clearly suggests the need for greater consideration of the capital intensity in industrial investment. The challenge can be met inter alia by giving greater attention to the possibili- ties of job creation in the small and medium scale segment of the manufactur- ing industry (SMI), which has the greatest potential of creating jobs at relatively low investment outlay (see paras. 16.01, 17.01). 2.06 To foster urban employment opportunities would require, in addition, setting right factor prices which, because of the present system of incentives and Government intervention, tend to favor capital-intensive technologies and production. Subsidized interest rates and exemption of imported equipment from duties tend to lower the price of capital and the cost of machinery, 1/ while minimum wages if set at unrealistically high levels and the incidence of social benefits (about 55% of the wage base), tend to raise the effective cost of labor. Both these tendencies induce substitution of capital for labor, and are reinforced by increasing unionization and difficulties in laying off workers when activity slows down. The Government should consider the dis- continuatioa of the former incentives or at least, as a first step, grant them only for projects that demonstrably have a significant (direct and indirect) employment effect. The Government should also consider the possibility of funding social benefits out of taxes on company income or value added, in order to alleviate the impact of the cost of labor on the choice of technology, along with a policy of setting realistic minimum wages, if such policy is deemed essential. 2/ Encouragement of multi-shift work through appropriate measures also deserves full consideration. Measures to Promote Industrial Employment in Regional Urban Centers 2.07 The Government is also cognizant of the need for balanced develop- ment of large and small urban centers and has taken measures to foster regional development, but with modest success. Government industrial investment in small towns in itself is unlikely to generate sufficient employment for the urban poor. In view of the incidence of unemployment and poverty in the smaller and intermediate size urban centers, creation of low investment cost industrial jobs in these centers by marshalling private sector entrepreneurship should be the centerpiece of the Government's industrial policy. A reassess- ment, and possibly redesign, of the existing incentives scheme might also be 1/ An overvalued foreign exchange rate has similar effects. 2/ See also paras. 6.12-6.14, 8.03 below.  advisable to thwart the concentration of private industry in the major metro- politan centers. Continuation of a vigorous industrial estates program (para. 7.04) should also help promote the regional dispersion of industry, while financial intermediaries could make an important contribution to this effort by redirecting available resources. 2.08 In the special case of Istanbul, 1/ the Government should consider containing the growth of the labor force and fostering the development of skill-intensive and advanced technology oriented industries. The higher and rising level of education of Istanbul's labor force clearly suggests the need to create an increasing proportion of jobs requiring higher skills. This could be achieved by devising appropriate incentives. The process could be reinforced by discouraging industries generating low-skill employment through appropriate administrative measures (e.g., land use planning, zoning, licensing), since they encourage rural-urban migration. 2/ To ensure their effectiveness, however, the measures to be adopted should be adequate to offset the perceived advantages of locating in Istanbul by those the Government wishes to discourage from moving into the city. For instance, to stem the inflow of rural migrants into Istanbul, work opportunities with a job mix heavily weighted in favor of unskilled work should be created in alternative growth centers to attract them there (para. 2.06), regional wage differentials should be reduced, social infrastructure and educational possibilities should be improved. 1/ Greater Istanbul has about 4 million inhabitants and accounts for almost 10% of total population and 22% of Turkey's urban population. Instanbul's population has been expanding rapidly, and more than doubled since 1960, with migration from rural areas accounting for an estimated 75% of the growth. Its labor force is estimated at about 1.4 million in 1978 and is projected to exceed 2.5 million by 1995. Istanbul is the nation's major industrial center accounting for 38% of total employment and 42% of all manufacturing establishments in the organized sector. Though official figures are not available, unemployment (including underemployment) probably hovered around 8% in 1978. Istanbul's demographic, labor, and income distribution characteristics are detailed in Annex A. 2/ E.g. the textile sector could relocate in less developed labor-surplus areas. However, export oriented garment industry should be excluded because it relies heavily on the communication and marketing facilities of Istanbul for its development.  -7- III. THRUST OF THE INDUSTRIALIZATION EFFORT Role of Industry in the Economy 3.01 The growth of manufacturing industries in Turkey has been rapid through the first three five-year Plan periods (1962-77). Value added in manufacturing rose from about 13% of GDP in the early 1960s to about 20% currently, growing at the respectable annual average rate of 9.5%. The manufacturing sector absorbed some 30% of total fixed investment over the decade ended in 1975, but its share fell to 23% in 1978 largely as a result of its import-dependence and the growing scarcity of foreign exchange. With rising capital-labor ratios, its share of total employment grew only slowly to reach 10% currently. The private sector has invested mainly in the lucrative consumer goods and durables market, and only recently entered intermediate and capital goods production. It accounted for 57% of industrial investment during the last decade and now contributes about 68% of value added in manufacturing (Appendix Tables 1.8, 3.2, 3.5, 3.6, 3.19, 3.20). Small-medium scale industry accounts for an impressive 55% of total employment and 41% of total value added generated in the manufacturing sector and, despite limited support, SMI not only has been able to hold its own but even grow (paras. 5.02-5.12). Industrial Policies 3.02 Industrialization has been viewed as instrumental for the achieve- ment of high rates of economic growth, structural changes and greater self- sufficiency. Objectives such as employment generation and income distribution have been less weighty--being envisaged as the natural outcome of a rapid industrialization effort. In the pursuit of these objectives, industrial development policy has stressed import substitution, the application of advanced technology and a shift of emphasis from consumer goods to inter- mediate and investment goods industries. The main instruments to support this policy have been large investment allocations with direct planning and control in the public sector, and the stimulation of capital-intensive, large-scale private investment by generous incentives and high protection against imports. These policies have produced some benefits. The growth rates of industrial output and value added have been impressive. There has been considerable local production of basic commodities using newly introduced technologies and creating valuable labor skills. Dispersal of state economic enterprises and incentives for private firms to invest in less-developed regions have helped mitigate to some extent interregional disparities, as has a nationwide program of industrial estates. New Policy Initiatives 3.03 Turkey has a considerable natural resource base for further indus- trialization, a comparatively low-wage, skilled and hardworking labor force, a diversified industrial base, and a dynamic private sector responsive to market incentives. It also has an advantageous location near both Western and Eastern European and Middle Eastern markets. Nonetheless, the level and rate  - 8 - of growth of manufactured exports are very low. Industry is inward-looking because, until very recently, there has been no conscientious export promotion policy, and export incentives have not been sufficient to outweigh the advan- tages of effective industrial protection which, in a climate of growing domestic demand, makes domestic sales more profitable than exports. The Government has taken a number of steps to improve the export institutional framework, and proposes to take measures to achieve international competi- tiveness. 1/ But perhaps a more vexing issue with dire social consequences is the persistent high rate of unemployment, arising inter alia from earlier policies and the design of incentives, which have over-encouraged capital- intensive activities. As already alluded (para. 2.05), industry is expected to play a key role in absorbing the rising labor surplus. In this regard, the potential of SMI for employment generation and in saving/earning foreign exchange, aside from its important complementary role in the development of the industrial sector, has not been fully appreciated in the past. The SMI segment of manufacturing is endowed with necessary technical skills and entre- preneurship (para. 6.01) and, properly supported, can improve its performance and enhance its contribution significantly. Government authorities now appear to recognize the SMI's multi-faceted potential (see paras. 17.01 et seq.) and can be expected to become increasingly sensitized to its needs and prepared to take remedial action to remove impediments hampering its development. 1/ See "Turkey: Policies and Prospects for Growth," Report No. 2657a-TU, 1979, Vol. I, paras. 4.39-4.55.  - 9 - IV. INSTITUTIONAL FINANCING Structure of the Banking System 4.01 Turkey has a well-developed banking system. At the end of 1977 there were 40 banks, state or privately owned, excluding the four investment banks DYB, TSKB, SYKB, and DESIYAB; of these, 28 are privately-owned with 5 foreign-controlled. 1/ The remaining 12 are banks established and controlled by the state under special laws and serve as instruments to implement the Government's economic development objectives. Although the state-owned institutions provide regular banking services, such as holding deposits or making loans, each bank serves a specific industry or target group, e.g. agriculture (Zizaat Bankasi), small industry (Halk Bank), etc. The distribu- tion of assets and credits of the principal state-owned and private financial intermediaries is shown in Appendix Table 2.1. The private commercial banking sector is characterized by a high degree of concentration, with three banks (Is, Yapi-Kredi, and Akbank) accounting for 70% of the assets and credits in this segment of the sector. Probably 40% of total private fixed investment in industry is financed at long term by intermediaries, 50% out of internally generated funds, 8% from bond issues, and the remaining 2% through direct foreign loans or equity participations. The proportion of self-financing is much higher in small-medium scale enterprises. The Role of Commercial Banks 4.02 Private banks focus heavily on short-term lending and prefer to cater to the needs of selected companies; but since these credits are rolled- over at the maturity date, they form a longer term source of finance for the borrower. The Central Bank requires that the 20 largest private banks pro- vide at least 20% of their deposits in medium-term maturities. Most of the banks reportedly have satisfied this requirement, but it is not clear whether the increase is real or represents a reclassification of previous short-term credits, especially to the banks' more favored customers or companies in which the banks have large equity participations. Banks are extremely cautious in their lending procedures, with almost all of the credits given on a collateral and guarantee basis. 4.03 While the commercial banks have extensive branch networks, they remain uninterested in financing small and medium scale industries primarily because of the higher administrative costs and level of risk. Instead, they prefer to direct their lending almost exclusively to large, capital-intensive projects promoted by interest groups, since transaction, processing and supervision costs, and the default risk are much lower. Private commercial 1/ Foreign banks play a very minor role in Turkey, with only 6% of all assets held by commercial banks. Like the domestic private banks, they strongly depend for their funding on deposit mobilization.  - 10 - banks have sustained no losses from credit transactions and operate on a very high profit margin. 1/ Also, there is strong suggestion that they control a large share of the equity of major companies. In 1977, the private banks portfolio of shares and securities was TL 11.5 billion, up from TL 2.6 billion in 1972. Reportedly, such equity participation reflects promotion of under- takings by controlling interests or forced participation to promising enter- prises. Turnover in their portfolio is minimal. It appears that private commercial banks contribute little to help in the establishment and operation of small-medium size ventures, particularly in instances where an entrepreneur has limited collateral. Financial Assistance to Small-Medium Scale Industry 4.04 Recourse to non-institutional sources of funds (self-financing, curb market, sale or mortgage of personal property, solicitation from family members) is an important feature of the SMI segment of industry. Lack of access to institutional finance is particularly noticeable in SSIs occupying the middle of the size range and in MSI. However, the true dimensions of its extent and severity are hard to ascertain, in view of the reluctance of many SSIs to seek financial assistance from banking institutions. Lack of collateral, unwillingness to disclose the true financial position of the enterprise, fear of possible interference of the bank with operations, short- sightedness and complacency seem to be some of the reasons. Small enterprises occupying the lower end of the SSI size range and artisans are conveniently, though not adequately, served by the Halk Bank through the intermediation of some 650 guarantee cooperatives with a membership totalling 415,000. The Halk Bank remains the major supplier of short and medium term funds to SSIs (Appendix Table 2.2), but so far it has had a limited impact. Because of shortage of funds 2/ and orientation toward cooperative financing, its lending to manufacturing enterprises for working capital has been limited to about 12% of its portfolio and for machinery and equipment to only 8% (in 1977). Also, the Halk Bank has kept the credit granting authority of the branch manager to a minimum (TL 150,000 or US$6,000 equivalent in 1978). This prevents managers from developing judgement and ability to appraise potential investment risks, and has emphasized fully collateralized (or guaranteed) loans with immediate legal action (within 3 months) if repayment problems develop. These two characteristics reflect the conservatism of the Halk Bank and put the develop- ment orientation of the institution in question. 1/ The capital/asset ratio of the private banks decreased from 4.1% in 1972 to 1.2% in 1977. By decreasing capital ratios and maintaining profit margins on total assets, the private banks have raised their return on capital from a modest 23% in 1972 to 63% in 1977. 2/ The Halk Bank rejected over one-third of the applications received to finance equipment in 1977-78 due to lack of funds.  - 11 - 4.05 With the Halk Bank constrained (para. 4.04), the commercial banks reluctant (para. 4.03), and TSKB (para. 4.12) ill-equipped, 1/ even the present financing needs of small-medium size enterprises cannot be met. The need to replace obsolete equipment, less strict collateral requirements, 2/ and increasing competition can be expected to bring about a change in attitudes and to increase perceptibly the effective demand for term lending in the next several years. This demand would be further enhanced significantly by the expansion of the more dynamic SMIs, parLicularly in faster growing subsectors (see paras 16.01, 16.02). But availability of loanable funds is not the only pressing issue. The supply of equity capital must also be increased to facili- Late expansion and to ensure a sound capital structure of SMI. Insufficient equity capital to provide for start-up costs, collateral for loans, or as a cushion against business vicissitudes in dynamic enterprises, is another important matter to which the Government and Halk Bank would have to turn in the immediate future. 3/ Leasing/hire-purchase schemes, which reduce the small firm's investment needs by enabling it to rent the equipment, are instruments that also deserve full consideration. The Halk Bank 4.06 The Halk Bank, basically a commercial bank founded in 1938, is one of the largest banks in Turkey ($700 million assets in 1978) and the only institution catering to the needs of small-scale and, to some extent, tedium- scale industry. Within the framework of its statutes and the Government's Development Plans and Annual Programs, the main functions of the Halk Bank consist in extending short, medium and long-term credit to artisans and small and medium scale manufacturers for raw materials and equipment; providing commercial credits and guarantees to repairmen and tradesmen; acting as an intermediary for financing the construction of industrial estates; promoting the establishment/relocation of SSI in industrial estates; arranging for training of small industrialists (in cooperation with the Ministry of Industry and Technology, Ministry of Education, and the Confederation of Small Tradesmen and Craftsmen); assisting artisans and small manufacturers to form cooperatives, particularly in less developed regions; extending technical assistance to 1/ The actual and potential contribution of Halk Bank, SYKB and TSKB to SMI financing are discussed in paras. 4.06-4.12. 2/ Since October 1978 instructions have been issued to Halk Bank's branch managers to accept chattel mortgage on the equipment to be purchased. 3/ The difficulty in the provision of equity lies in that banks normally are not equipped to assess the risk of unproven business and that businessmen object to the idea of having to hand over a share of the ownership of management of their business to financial backers. In this regard, the concept of an SMI being able to buy out its financial backers after a period of time and of ways to limit the power of interference by providers of finance would have to be looked into.  - 12 - artisans and SSI; supporting returning workers from abroad to establish industrial enterprises. To a limited extent, it provides credit for con- struction of cooperative workshops by artisans and for low-income housing. Since mid-1977, the Halk Bank is providing loans to newly graduated profes- sionals (dentists, doctors, pharmacists, engineers, etc.). Priority in credit allocations is given to activities which are export-oriented, import- substituting, generate employment opportunities, and process local raw mate- rials, and to firms locating in industrial estates. The Halk Bank proposes to support joint ventures by SSIs for the production of basic inputs they utilize (e.g. chipboard) and to increase its role in guiding SSI on the use of appro- priate technology. Halk Bank accepts deposits (accounting for some 50% of its resources) and provides a wide range of commercial banking services. 4.07 The Halk Bank's financial position, resource mobilization effort (type of deposits and depositors, terms, etc.) and major facets of operations (types of loans, sectoral, geographic and size distribution of credits are presented in the Staff Appraisal Report, in preparation). Suffice it to mention here that the outstanding amount of loans for equipment rose from TL 200 million in 1973 to slightly over TL 1 billion by the end of 1977, involving some 2,000 individual operations. This reflects the Halk Bank's increasing emphasis in supporting industrial investment, a policy which was instituted in 1973. Approvals for equipment in 1979 amounted to TL 1,410 million (US$30 million). Food processing, metalworking, woodworking and textiles are the major borrowing industries. The Halk Bank does not partake in equity. Availability of funds could enhance substantially the Halk Bank's volume of lending to SMI. 4.08 Based on a fairly small but representative sample of 56 loans approved in 1977, ranging from TL 100,000 to TL 9,000,000 in size and account- ing for an estimated TL 51 million of new investment, some 570 new jobs would be created at an investment cost per job of TL 89,500 ($5,000), which is remarkably low. About 80% of the loans were below TL 1 million ($55,000) and only two over TL 5 million ($278,000). Of the firms surveyed in the sample, which cover the major urban areas, 4% were new and the balance expansion projects. Employment in the supported enterprises prior to expansion ranged from 5 to 75 workers. The above attest to the relatively small size of enterprises that the Halk Bank caters to and the labor-intensity of the projects it supports. 4.09 The Halk Bank's involvement in development financing so far has been rather limited. The Halk Bank's appraisals center basically on the ability of the borrower to perform satisfactorily and to repay the loan. To establish this, the Halk Bank checks on the borrower's investment and production plans to satisfy itself that the new machinery and the proposed capacity are actually needed, estimates his needs in working capital, considers the availability of raw materials, assesses his ability to fulfill contracts, estimates the value of his assets, obtains information on his reputation, etc. By and large, the Halk Bank's appraisal and lending procedures are straight forward and expedi- tious. Follow-up is routine with focus on problem areas. The Halk Bank has a number of capable staff in its Industrial Credit Department, but in order to appraise larger projects and to increase significantly the volume of project  - 13 - lending it reeds to build up its project appraisal capability at the Ankara head office and at the two regional offices in Istanbul and Izmir. Other weaknesses include: undefined lending objectives and lack of credit policy statement; absence of operating plans for the short and medium terms; overly am.bitious p ans for branch network expansion without study of their cost- effectivene<s; lending on highly secured basis; lack of well-conceived staff recruitment and training programs; and inadequate systems and procedures (e.g. deficiencies in the accounting and management information system). The proposed Bank loan aims at increasing the resource availability and at assist- ing the Halk Bank to implement a time bound action program to rectify these institutional shortcomings, so that it can perform more effectively its role as an intermediary for term financing of SMI projects. 4.10 The Halk Bank is in the best position to play a significant role, if properly strengthened, in the development of the small-medium scale indus- try for a number of reasons: (a) has established an approach and attitude to accommodate the small-scale borrower by an extensive branch network, a regionalized lending approach, lower than normal service charges, "same day" deposit withdrawal option for its depositors, and training programs for small business, while maintaining acceptable standards of financial discipline and loan collection; (b) can assess better than centralized financial intermedi- aries the personal qualities of the borrower and local business conditions; (c) the commercial banks remain basically uninterested in financing small- and medium-scale industries, preferring to lend almost exclusively to larger, capital-intensive projects promoted by small interest groups; (d) development banks also ilave a somewhat different focus and organization, with a highly trained and experienced staff all geared to the promotion and financing of large projects; in addition, they do not have the available branch network to reach the small and medium scale borrowers and, as a result, such institutions as TSKB can extend only limited support to medium size industry at the upper end of the range; however, SYKB could assist in the development of the medium- scale lending by utilizing the branch networks of the private commercial banks which are its shareholders; (e) the Halk Bank provides working capital as well as the financing of the purchase of machinery and equipment; this is vitally important as working capital credits are persistently scarce. SYKB 4.11 SYKB was established in 1963 by a group of five Turkish private com- mercial banks (Is Bank, Osmanali Bank, Turkish Guarantees Bank, Ak Bank and Turkish Vakiflar Bank) and is managed efficiently. It provides medium and long-term loans, mostly in local currency, and working capital loans. Although SYKB has been financing industries in all subsectors, its major involvement has been in the engineering, textiles, and food processing indus- tries. 1/ So far, SYKB has had limited involvement in financing medium size enterprises but is willing to increase it lending to this segment of the industrial sector. SYKB has the requisite project appraisal/supervision 1/ For details, see Staff Appraisal Report (in preparation).  - 14 - capability and can utilize the extensive branch network of its shareholding commercial banks. It should have no difficulty to adapt its organizational structure to finance medium size projects and would significantly complement the Halk Bank's role in financing SMI. A Bank loan to SYKB is under con- sideration for this purpose. TSKB 4.12 TSKB is an efficient development bank and concentrates mostly on foreign currency loans (85% of total lending). Textiles, food processing, engineering industries, forestry products, chemicals, and plastics industries are the major recipients. Over 60% of approvals in 1976 and over 50% in 1977-78 were over TL 30 million (US$1.2 million). This lending pattern reflects TSKB's limited support to medium scale industry. This is more evident if one considers the investment cost of the projects supported, as TSKB in many instances finances only part of the total project cost. This is understandable since TSKB is geared to supporting large projects, has no extensive branch network to reach SMI, and extends no loans for working capital. The capital intensity of TSKB-financed projects also is indicative of its support of large projects. The frequency distribution of the estimated capital/labor ratios of projects approved for 1976 is shown in Appendix Table 2.3. Of the projects supported, 45% had an investment cost per employee up to TL 334,000 (US$20,000), while another 41% between TL 334,000 and TL 1 million (US$60,000).  - 15 - B. MORPHOLOGY OF SMALL-MEDIUM SCALE INDUSTRY V. STRUCTURAL FEATURES OF SMALL-MEDIUM SCALE INDUSTRY 1/ The Pattern of Ownership 5.01 With the exception of steel, pulp and paper, and tobacco, manufac- turing in Turkey is predominantly in private hands, 2/ accounting for 62% of total production and 65% of total employment in the sector. 3/ The respective shares of private and public sectors in organized manufacturing are shown in Appendix Table 3.1. The share of private organized sector in total value added increased from about 43% in 1963-72 to 48% during 1973-77, while that of the public sector declined from 36% to 31% (Appendix Table 3.2). The share of the unorganized sector 4/ averaged about 20% through out the period 1963-77, with that of rural manufacturing dropping from 1.5% to 0.4%. During 1965-75, real value added in the private organized 5/ sector increased at an average annual rate of 17%, compared with about 10% in the public sector (Appendix Tables 3.3, 3.4); it slowed down to about 7% during 1976-77, due to the severe problems that plague the Turkish economy in recent years. The lower growth in the public sector may in part be attributed to price controls and other social burdens thrust upon state enterprises. The share of real investment of the private sector in total manufacturing investment increased from 56% in 1965 to 65% in 1974, but dropped to 60% in 1975 (Appendix Tables 3.5, 3.6). Since 1975, growth in manufacturing investment has been declining in both the public and private sectors, reflecting the vicissitudes of the economy. In terms of employment (1976), the private sector accounts for 83% of total in manufac- turing and 65% of employment in the organized sector (Appendix Table 3.1). 1/ For details on the structural characteristics and various dimensions of performance of the subsectors surveyed in this report see as follows: foundries (section XI), engineering (XII), leather and footwear (XIII), wood-based (XIV), and garments/made-up (XV). The salient structural features of small-medium scale industry in the Istanbul Province are reviewed in Annex B. Establishments employing 10-199 workers located in Istanbul accounted for 40% of the total number and 41% of total employ- ment of the respective brackets at the national level in 1977. 2/ The public sector is organized into 13 major holding companies controlling about 400 plants. 3/ The importance of the private sector is even greater if the unorganized sector is included, particularly with respect to employment. 4/ Establishments employing 1 to 9 workers. 5/ Establishments employing 10 or more workers.  - 16 - Finally, almost all exporting enterprises are in the private sector, which is responsible for the bulk of manufactured exports (86% in 1977). This share has been relatively stable since 1962 (Appendix Table 3.7). Size Distribution 1/ 5.02 Small (establishments with less than 50 workers) and medium scale enterprises (establishments employing 50-200 workers) are predominant in Turkish manufacturing. In 1970, there were about 175,000 establishments in the private sector, 2/ of which 487 (or 0.3% of total) employed 200 or more workers, 4,332 (or 2.5%) between 10 and 199, and the balance of some 170,000 establishments (or 97.2%) less than 10 workers. In the last category, the so-called unorganized sector, over 162,000 (or 93%) establishments employed 4 or less workers (Appendix Table 3.8). The average size per establishment is 2 (see Table 2). There is no recent statistical information available, but the situation probably has not changed perceptibly in this sector during the 1970's. The number of establishments employing 5-9 persons declined from 7,701 in 1970 to 7,333 in 1977 (Appendix Tables 3.8, 3.9), or by 5%, presum- ably as a result of the demise of a number of inefficient enterprises whose problems were exacerbated by the adverse economic conditions in recent years; some of the missing enterprises undoubtedly graduated to higher size brackets through growth or merger. 5.03 In the organized sector, between 1970 and 1977 the total number of establishments increased from 4,819 to 8,537, or by an impressive 77%, as can be seen in Table 1. The increase has been more dramatic in the medium (50-200) Table 1: DISTRIBUTION OF ESTABLISHMENTS IN ORGANIZED MANUFACTURING SECTOR (Number of establishments) 1970 % 1977 % Growth 1970-77 (%) 10-49 3,391 70 6,166 72 82 50-199 941 20 1,773 21 88 200+ 487 10 598 7 23 TOTAL 4,819 100 8,537 100 77 Source: Appendix Tables 3.8, 3.9. 1/ To bridge the hiatus due to the fragmentary information, data drawn from various sources had to be collated and spliced. In view of the limitations of the data base, some of the conclusions drawn in this and subsequent paragraphs should be treated with caution. 2/ They probably number around 180,000 currently.  - 17 - and small scale industry (10-49) sectors, where the number of establishments rose by 88% and 82%, respectively. The number of large scale establishments also increased, but by a modest 23%. In the wake of these developments, the relative importance of small and medium scale industry within the organized manufacturing sector has slightly increased--at the expense of the large enter- prises whose number declined from 10% in 1970 tc 7% in 1977. The observed growth is mostly the net result of new entry and upward inter-class shifts all the way up from the unorganized sector, as is evidenced from the crude figures shown in Table 2. Indeed, the average size of establishments in all subgroups between 10 and 199 remain relatively stable between 1970 and 1977, Table 2: AVERAGE SIZE OF ESTABLISHMENTS IN ORGANIZED MANUFACTURING SECTOR Size of Establishment 1970 1977 1-9 2 N.A. 10-49 21 20 50-99 69 63 100-199 134 138 200+ 718 855 Source: Appendix Tables 3.8-3.12. in contrast to the larger scale industry where the average size increased by 19%. Also, in 1976, about 640 new manufacturing joint stock companies were established in Turkey. This figure is likely to include a cluster of small firms which reorganized and formed larger firms. Sectoral Distribution 5.04 The sectoral distribution of establishments in the organized manu- facturing sector in 1977 is shown in Table 3. In all industries, small Table 3: SECTORAL DISTRIBUTION OF ESTABLISHMENTS /a, 1977 By Size of Firm Total Industry 10-49 50-99 100-199 10-199 200+ Food, Beverages 1,077 215 99 1,391 93 Textiles and Clothing 887 210 132 1,229 184 Leather and Leather Goods 217 32 13 262 4 Wood Products, Furniture 145 48 33 226 17 Cellulose and Paper Products 297 48 19 364 20 Chemicals 533 95 51 679 64 Plastics 459 40 21 520 22 Basic Metals 608 181 62 851 54 Metal Products 1,891 305 154 2,350 138 Other Manufacturing 52 12 3 67 2 TOTAL 6,166 1,186 587 7,939 598 /a Size of establishments expressed in terms of number of workers employed. Source: Appendix Table 3.9.  18 - and medium scale industry is dominant in numbers. Metal products, basic metals, food and beverages, and textiles and clothing, are predominant in both medium (50-199) and small scale (less than 50) industry. It is interest- ing to note that there are as many medium size firms (100-199) in the upper end of this class as are large ones (200+), and that their respective numbers (close to 600) are almost evenly distributed among industrial subsectors. Lower end medium size firms (50-99) are twice those in the upper end in almost every industry. Although there are activities in each industry which fall within the domain of large scale industry, particularly in basic metals, metal products, plastics, and chemicals, the fact remains that the entire spectrum of the manufacturing sector is heavily populated by smaller undertakings. This implies inter alia that economies of scale vary greatly within each subsector and that smaller firms are able to capture a share in the market large enough to ensure survival and even growth. The protected market is certainly an additional contributing factor. 5.05 The growth in the number of manufacturing establishments in the organized sector during 1970-77 is shown in Table 4. The most dramatic Table 4: GROWTH OF ESTABLISHMENTS IN ORGANIZED MANUFACTURING, 1970-77 Industry 1970 1977 % Increase Food, Beverages 1,229 1,484 20 Textiles and Clothing 1,038 1,413 36 Leather and Leather Goods 92 266 189 Wood and Wood Products 152 243 60 Cellulose and Paper Products 253 384 52 Chemicals 376 743 98 Plastics 202 542 168 Basic Metals /1 160 905 466 Metal Products 859 2,482 189 /1 Includes castings. Source: Appendix Tables 3.9, 3.13. increases, both in absolute numbers and percentage-wise, have been in basic metals, including foundries, and metal products, followed by leather products, plastics, and chemicals, reflecting the potential in these industries and the quick response of new entrepreneurs, as evidenced by the high incidence of new entry, in the face of rising demand for consumer and intermediate goods (backward linkages). It is also notable that in the more traditional industries such as food and beverages and textiles (excluding clothing), the growth in the number of establishments was very low. This reflects the already higher level of development of these industries (large original base) and, possibly, growth in the size of establishments (para. 5.09).  - 19 - 5.06 Data of recent vintage are not available to study the present sec- toral distribution of the establishments in the unorganized (very small scale) manufacturing sector. Impressions, nonetheless, can be formed regarding its structural features and relative importance from Table 5, based on 1970 data. The observed structural features probably are not too different from those prevailing at present. The textiles, wearing apparel and leather products group accounts for 37.6% of the total number of manufacturing establishments in the unorganized sector, while the fabricated metal products, machinery, and transport group for 29.0%. Thus, the unorganized sector is dominated (66.6%) by these two groups. Their relative importance is somewhat lower in the orga- nized sector (40.5%), though still very important. Table 5: SECTORAL DISTRIBUTION OF UNORGANIZED MANUFACTURING, 1970 Industry Unorganized % Organized % Food, Beverages 17,673 10.4 1,230 26.7 Textile, Wearing Apparel & Leather 64,196 37.6 1,130 24.5 Wood, Wood Products and Furniture 23,034 13.5 258 5.6 Paper & Paper Products, Printing & Publishing 2,505 1.5 253 5.5 Chemical, Rubber & Plastics 1,914 1.1 366 8.0 Non-metallic Minerals 4,602 2.7 318 6.7 Fabricated Metal Products, Machinery & Transport Equipment 49,607 29.0 859 18.6 Other Industries 7,034 4.2 202 4.4 TOTAL 170,565 100.0 4,616 100.0 Source: Appendix Tables 3.11, 3.12. Regional Distribution 5.07 The regional distribution of manufacturing establishments in the organized sector in 1977 is shown in Table 6. 1/ Istanbul accounts for 42% 1/ The Plan's estimate of total number of establishments differs signifi- cantly from that of the Union of Chambers of Commerce and Industry. This difference, however, does not distort the emerging picture.  - 20 - Table 6: REGIONAL DISTRIBUTION OF ORGANIZED MANUFACTURING ENTERPRISES /a (1977) Food, Minerals, Region Beverages Textiles Paper Chemicals Engineering Other Total Marmara 211 192 92 115 146 - 757 Aegean 62 116 26 71 55 - 332 Mediterranean 91 97 28 64 62 - 342 S.E. Anatolia 32 36 4 29 13 - 114 Eastern Anatolia 31 4 1 13 7 - 56 Central Anatolia 101 22 16 55 90 - 284 Black Sea 138 16 41 84 77 - 356 Istanbul 248 517 146 535 839 61 2,346 Izmir 145 85 42 89 197 5 563 Ankara 147 11 64 48 141 4 415 TOTAL 1,206 1,096 460 1,103 1,627 73 5,565 /a Establishment employing 10 or more workers. Source: Appendix Table 3.14. of the total number of establishments, followed by the Marmara, Izmir and Ankara regions with 14%, 10%, and 8%, respectively. This suggests a high regional concentration around a few major urban cities, where about 75% of total manufacturing establishments are located. Istanbul also has by far the highest concentration in all industries, especially in chemicals (62%), engineering industries (52%) and textiles (47%). In terms. of regional dis- tribution of employment and value added in the organized sector, the Istanbul province accounts (1973) for 38% and 35% of the national total, respectively (Annex B, Appendix 1). This pattern of industrial location is virtually replicated in the three other major urban centers. 5.08 The bulk of the unorganized small firms are located in the metro- politan areas (Istanbul, 1/ Izmir, Ankara) and in other major cities, e.g. Konya, Bursa, Gaziantep, etc. There are often regional concentrations of small firms in the same industries, e.g. weaving and leather products in Istanbul, metal products in Gaziantep, Bursa, Konya and Polatli. The regional distribution of establishments and employment in the unorganized manufacturing sector in 1970 is shown in Table 7. I/ See Annex B.  - 21 - Table 7: REGIONAL DISTRIBUTION OF UNORGANIZED MANUFACTURING ESTABLISHMENTS /a AND EMPLOYMENT, 1970 Number of % Number of % Region Establishments Share Employees Share Marmara 65,536 38.6 110,811 33.7 Aegean 28,471 16.7 58,731 17.8 Central Anatolia 23,192 13.6 52,190 15.9 Black Sea 21,710 12.8 41,383 12.6 Mediterranean 14,153 8.3 30,715 9.3 Western Anatolia 9,901 5.8 21,083 6.4 Eastern Anatolia 7,156 4.2 14,081 4.3 TOTAL 170,123 100.0 328,986 100.0 /a Establishments employing less than 10 persons. The Marmara region ranks well above the other regions, accounting for about 39% of the establishments in this sector and one-third of employment. The pattern of regional concentration appears to correlate with the degree of industrial development of a region and the size of population -- with the more advanced regions attracting a larger number of small scale industry. This appears to be in line with observed urbanization trends (para. 2.01, footnote 1). Data are not available for recent years. However, the regional distribution is unlikely to be substantially different. Distribution of Employment 5.09 The distribution of employment by size of establishment in the organized manufacturing sector is shown in Table 8. Employment increased by 55% during 1970-77, growing at an annual average rate of 6.5%. The growth rate has been much higher (8.0-8.5%) in the medium and small compared to the Table 8: DISTRIBUTION OF EMPLOYMENT IN THE ORGANIZED MANUFACTURING SECTOR Size of Establishment Year Growth Rate, 1970-77 (No. of Persons Employed) 1970 % 1977 % Annual Average 10-49 71,681 14 123,411 16 72 8.1 50-99 41,971 8 74,318 9 77 8.5 100-199 47,156 9 80,836 10 71 8.0 200+ 349,648 69 511,328 65 46 5.6 TOTAL 510,456 100 789,893 100 55 6.5 Source: Appendix Tables 3.10, 3.12.  - 22 - large scale industry (5.6%) 1/. Enterprises in the 10-200 workers range created 118,000 new jobs between 1970 and 1977, or 17,000 a year, which is quite remarkable when compared to 161,000 jobs, or 23,000 annually, created by the 200+ firms during the same period. The increase in employment has been the result of the growth in the number of establishments, primarily in the lower end of the spectrum, as well as of the growth in the size of firms, measured by the number of persons employed, particularly in the upper medium and large scale industries (para. 5.03). Table 8 indicates that in 1970, employment in large enterprises accounted for 69% of total in establishments with 10 or more workers, and that this share declined to 65%, while that of the medium (50-199) and small scale industry (10-49) each edged up by 2%. In reality, however, the share of large scale industry in total manufacturing employment is much less, if employment in the unorganized sector (1-9 workers) is included. Thus, in 1970, a year for which data are available (Appendix Table 3.8), 328,586 persons were employed in this category, reducing the share of large scale industry total manufacturing employment to 42%. Probably, its share is not very different in recent years. The persistence and continued growth of the medium and small scale industry clearly attests to its vitality and dynamism and reflects a continuing inflow of new entrants (possibly some returning workers from the EEC) and high inter-class mobility. 1/ To the extent that underreporting by smaller establishments for a variety of reasons (e.g. evasion of social security and tax payments) takes place, the actual level of employment is underestimated. So are perforce levels of production and value added.  - 23 - 5.10 The sectoral distribution of employment in the organized manufac- turing sector is presented in Table 9. In the small-medium scale industries (10-199), employment increased by a respectable 78% to 279,000 during 1970-77, Table 9: SECTORAL DISTRIBUTION OF EMPLOYMENT IN PRIVATE SMALL-MEDIUM SCALE MANUFACTURING, /a 1970 AND 1977 % % Growth Industry 1970 Share 1977 Share 1970-77 Food, Beverages and Tobacco 36,139 23 50,236 18 39.0 Textiles and Clothing 32,461 21 52,775 19 62.6 Leather and Leather Goods 2,495 2 8,233 3 230.0 Wood and Furniture 8,229 5 10,530 4 28.0 Cellulose and Paper Products 8,126 5 12,552 4 54.5 Chemicals and Related Products 22,113 14 33,974 12 53.6 Basic Metals 6,005 4 27,182 10 352.7 Metal Products 34,191 22 81,024 29 137.0 Other Manufacturing /b 6,808 4 2,059 1 -230.6 TOTAL 156,567 100 278,565 100 77.9 /a Enterprises of 10-199 employees. /b Part of the decline probably is due to the lack of comparability in the industry elassification in the two sources used. To that extent, the growth of the other sectors is somewhat overstated. Source: Appendix Table 3.10, and Table 6.4 from World Bank Report No. 1359-TU, 1977. rising at an annual average rate of 8.5%. The growth has been particularly high in basic metals, metal products and leather goods, raising substantially their respective shares in total employment in this segment. Metal products, textiles and clothing, food and beverages, chemicals and basic metals accounted for 88% of total employment in the small-medium size group. This same group also accounts for 94% of total employment in large scale industry (Appendix Table 3.10). As stated earlier (para. 5.09), the growth in employment was effected through both the establishment of new enterprises and the growth of existing firms. 5.11 A number of industries within the unorganized sector make a particu- larly high contribution to employment and output as can be seen in Table 10. Footwear and clothing, a range of metal products, wood products, furniture and fixtures, and leather products account for between 70% and 93% of the employ- ment in their respective industries, and for 22% of total employment in private  - 24 - Table 10: UNORGANIZED MANUFACTURING INDUSTRIES WITH HIGH SHARES IN TOTAL MANUFACTURING EMPLOYMENT AND OUTPUT, 1970 Share of Share of Number of Employment to Output to Persons Total of Total of Industry Employed Industry Industry Footwear and clothing 80,342 93 80 Furniture and fixtures 13,573 83 73 Wood and wood products 33,139 78 59 Leather products (except footwear) 7,194 72 42 Metal products (except machinery and transport) 49,616 70 22 TOTAL 183,864 22 9 Source: Table 6.5 from World Bank Report No. 1359-TU, 1977. manufacturing industry (1970). The same remarks can be made for the contri- bution of the unorganized segments of their industries to total output. Distribution of Value Added 5.12 The percentage distribution of value added in the manufacturing industry is shown in Table 11. During 1963-77, the respective shares of establishments in the 1-9 and 10-199 groups averaged around 20% each, while that of large scale industry (200+) rose from 21% to 27% at the expense of the public sector. Since real value added during this period grew at an impressive 13-14% per annum, the fact that SMI, despite many shortcomings (paras. 9.01-9.07), maintained its share is indicative of its dynamism and important contribution. The observed constant decline of the share of the rural sector in manufacturing during the same period attests to the strong pull of the urban centers as loci for industrial activities. Table 11: PERCENTAGE DISTRIBUTION OF VALUE ADDED IN MANUFACTURING INDUSTRY, 1963-77 (In 1968 prices) Public Private Sector Rural Year Sector 200+ 10-199 /a 1-9 Sector Total 1963-67 35.8 21.6 20.0 21.4 1.2 100.0 1968-72 36.8 23.3 20.5 18.6 0.8 100.0 1973-77 31.4 27.4 21.0 19.7 0.4 100.0 /a Mission's estimate. Source: Appendix Table 3.2.  - 25 - Capital Intensity 5.13 The average capital-labor ratio 1/ for all industries increased from TL 50,000 to TL 93,000, or by 86% during 1965-75, at an annual average rate of 7% (Appendix Table 3.15). In the private organized sector, the overall capital- labor ratio increased from TL 26,000 to TL 66,000, or by 154A, during the same period, at an annual rate of 9% (Appendix Table 3.16). Yet, in absolute terms, capital intensity on average is by about 50% higher in public sector enterprises (1972-75) and, in reality, even higher due to overstaffing. In very general terms, this attests to the lower capital intensity of private firms, in part due to their greater involvement in light and non-process industries. Unfortunately, there is no information on private sector capital/labor ratios by class of enterprise size and major product categories, to establish in a more definitive manner its capital intensity. Nonetheless, using as a proxy the investment cost per job created as estimated in the Encouragement Certificates granted during 1/1/1979-6/60/80, it can be argued convincingly that small and medium size enterprises (projects involving investment up to TL 160 million) tend to be less capital intensive compared with large firms. As can be seen in Table 12, the cost per job created (on average) in 70% of the planned investment projects undertaken by SMI is US$23,000 or less, whereas the figure for larger projects escalates to US$73,000. It is noteworthy that the average investment cost per job in the upper end of SMI (US$23,000) is about one-half of the average of the Table 12: INVESTMENT COST PER JOB BY SIZE OF MANUFACTURING PROJECT /1 Average Investment Total Investment Cost Per Job Number of in Mach. & Equip. Employment ('000 (US$1= Project Size /2 Projects ('000) Generation TL) TL60) /3 Below TL40 million 129 3,523,465 4,896 720 12,000 TL40-TL80 million 146 8,435,001 8,979 939 15,650 TL80-TL160 million 108 12,267,042 8,868 1,383 23,050 TL160-TL320 million 69 15,940,000 8,466 1,883 31,380 TL320-TL640 million 44 19,540,407 9,947 1,964 32,730 TL640-TL1,000 million 18 14,451,318 4,779 3,024 50,400 Over TL1,000 million 28 103,793,432 23,677 4,384 73,060 TOTAL 542 177,950,665 69.612 2,556 42,600 /1 Based on projects which received Certificates of Encouragement during 1979- June 1980. /2 Investment in machinery and equipment. /3 Average for the period 1/1/1979-6/30/1980. Source: State Planning Organization. 1/ Computed on the basis of total capital stock/total labor in each sector in constant 1965 prices.  - 26 - total population of the projects (US$42,600). Also, SMI projects are expected to generate 22,743 new jobs or one-third of the total at 14% of the total capital cost. Table 13 provides a ranking of various industries in both private and public sectors in terms of capital-labor ratios. Engineering, wood and furniture, textiles, and food industries rank lower in terms of capital intensity compared to intermediate product industries. There is a fairly close correlation between the industry rankings in both sectors with the notable exception of metal prod- ucts, possibly stemming from differences in production scale and technology employed in the two sectors. Table 13: RANKING OF INDUSTRIES BY K/L RATIOS (1975) Private Public Basic metals 1 3 Petroleum refining and coal products 2 1 Non-metallic minerals 3 6 Chemicals 4 4 Paper and paper products 5 5 Textiles and clothing 6 12 Transport equipment 7 11 Wood and furniture 8 9 Food and beverages 9 7 Printing 10 8 Machinery 11 13 Electrical machinery 12 10 Metal products 13 2 Tobacco 14 14 Source: "Capital and Labor in the Turkish Manufacturing Industries", by K. Ebiri et al.: SPO, Turkey 1977. Labor and Capital Productivity 5.14 During 1965-75, average labor productivity (value added per worker) in the private sector increased by approximately 9% per annum compared with 7% per annum in the public sector (Appendix Tables 3.17, 3.18). Private sector industries with labor productivity growth rates above the sector average were engineering, tobacco processing, paper, rubber and petroleum. The overall growth rates were, however, lower than the corresponding growth rates of investment during the same period. Capital productivity, on the other hand, has shown a consistent decline over the same period. The gross capital-output ratio for the manufacturing sector increased from 0.49 in 1965 to 0.67 in 1975. The capital/value-added ratio in the public sector increased from 1.28 in 1965 to 1.87 in 1975 (46%) and in the private sector from 1.14 to 1.98 (74%) during the same period, which can be attributed to the rapid increase in capital intensity of investment, particularly in the private sector (Appendix Tables 3.19, 3.20). Again, there is no breakdown by enter- prise size to assess factor productivity in the SMI segment of the private sector.  - 27 - VI. ORGANIZATIONAL AND OPERATIONAL FEATURES OF SMALL-MEDIUM SCALE INDUSTRY 6.01 The principal organizational and operational features of SMI based on plant v. its by the mission and other studies 1/ are described below. Form of Ownership 6.02 The most common forms of ownership in the SMI segment of industry are the individual proprietorship and partnership. Joint stock companies, to be found among the larger of medium size enterprises, are the exception-- and by and large family-controlled. Enterprises employing 9 workers or less, which constitute the bulk of manufacturing establishments in Turkey, remain predominantly (88% in 1970) single proprietorships. Establishments employing between 10-25 workers are more frequently organized as partnerships, usually between a usta (master craftsman) and a trader who provides marketing ability, trade contacts and capital. The responsibility which partners assume in such a kollectif sirket resembles the situation where individual partners are financially responsible for all of each other's dealings. In a limited sirket, mostly used by firms employing 25-200 persons, such financial obligations are circumscribed. Medium sized enterprises employing 100 workers or more are increasingly getting organized as joint stock companies. In 1977 alone, 640 joint stock companies were set up in the private manufacturing sector. In some instances, this form of ownership involved reorganization or horizontal integration of several smaller firms. Appreciation that the joint stock com- pany provides greater access to institutional credit and continuity than individual family businesses seems to have prompted the adoption of the cor- porate form of business organization among medium scale establishments in recent years. Origin of Entrepreneurship 6.03 Most owners of workshops employing up to 10 workers have originated from the shop floor. This is especially true in the metalworking, shoemaking and woodworking industries and, to some degree, in the garment industry. In some other branches of the textile industry, chemicals, plastics, rubber and some branches of the food industry where the required initial investment is larger, the entrepreneur is generally a trader who has commercial acumen and has accumulated capital from wholesaling or retailing operations. In the case of medium size enterprises, entrepreneurs are usually businessmen or profes- sionals with considerable industrial experience. However, many small enterprises have been able to graduate into medium size through successive 1/ L. Van Velzen, Peripheral Production in Kayseri, Ajans-Turk Press, Ankara, 1977; World Bank, Export-Oriented Industries and Small-Scale Industries in Turkey, Report No. 1359-TU, 1977; SPO/DSP Study in five cities on occupational and operational characteristics of the unorganized manufacturing sector.  - 28 - expansions. A more recent development has been the promotion of medium size joint stock companies by young university trained engineers with equity provided by Turkish workers abroad. These tend to be in products such as electrical cables, urethane foam, etc., which rely on imported technology. 6.04 The mission was favorably impressed with the quality of entrepreneur- ship in many SSIs. Owners of establishments appeared energetic, resourceful, hardworking, and good at improvising and copying. They displayed initiative, deep knowledge of their trade, confidence and, in certain instances, consider- able ingenuity in the development of machines. Small manufacturing firms have demonstrated an ability to improvise and add to their capital stock by manu- facturing machinery in their own machine shops or by reconditioning old equipment. In certain other respects, however, they remain traditional, they are less amenable to new ways of doing things, have a paternalistic attitude toward workers, and are unappreciative of good housekeeping. Specialization of Management Functions 6.05 A characteristic of the small scale manufacturing sector is the concentration of the managerial and production functions in the person(s) of the owner(s). The proprietor carries out all the activities--except the actual production operations--involved in the running of his business. 1/ Obviously, these individuals have neither the expertise nor the time to carry out adequately all the required tasks. In most cases they have risen from the shop floor and have had only primary education. It is only natural to concentrate on the production side and pay scant attention to the rest. The critical point in the development of an industrial establishment is reached when it becomes essential to delegate managerial functions. Many entrepre- neurs are incapable or, at least, find it very difficult to take this step. In consequence, the growth potential of their establishments is limited. It is difficult to help such individuals, as it is necessary to overcome entrenched attitudes, and they have neither the time and patience nor the realization of their need for assistance. If follows, therefore, that the class of establishments with a realizable potential for growth, and conse- quently for continuing viability, 2/ is likely to be the one in which there is some degree specialization of management functions (see also para. 17.07). Such specialization in Turkey has been observed to commence when an enterprise reaches the 15-20 employment bracket, depending on the industry and technology employed, and increases progressively with the size of establishment. Labor Skills and Training 6.06 Entrepreneurs in traditional workshops (usta) generally start their careers through the guild system. They begin working at age 10-16 as 1/ In the artisanal field he frequently engages in the production process as well. 2/ For an array of other factors contributing to viability and growth, see para. 16.01.  - 29 - apprentices (cirac), and after gaining experience they become journeymen (kalfa). Thereafter one attains a status of master (usta) and goes into business on his own. There are no formal demarcations between successive phases of the learning process. An alternative approach would be for somebody to ritad ; niche -s a workshop owner. After completing secondary school (orta okull), v-cational training is available at the vocational school (sanat okul) where stLdy and work experience is combined. The sanat okul graduate can then st rt his business, but the bulk of the throughput from vocational schools finds its way to the larger SSI and MSI. There is impressionistic evidence that graduates of the sanat okul are responsible for innovative developments in SMI workshops. In some of the larger industrial urban centers, e.g. Kayseri, there are technical colleges (urksek sanat okul), where a lycee diploma (lice) is a requirement for admission. Instruction, however, tends to be more theoretical and less practical, and graduates seldom establish their own business. Workers from larger enterprises also enter the small enterprise sector after accumulating savings to purchase machinery. This was particularly true in the 1950's when large numbers of workers employed in military estab- lishments were demobilized. Higher requirements for initial capital to start workshops and the possibility of external emigration have somewhat slowed down the movement of skilled workers to the SSI sector. The reverse movement from smaller to larger enterprises has also been at work. 6.07 The traditional system of training manifested in the usta-kalfa- cirac relationship, however, is increasingly coming under strain because of the changing structure of the economy. Production in workshops is confined to a limited number of goods. Within this production system the training of craftsmen is at best inconsequential. Furthermore, the advent of uniform production has reoriented demand toward semi-skilled workers. For example, there is little need for master cabinet makers specializing in hand carving in the furniture industry as the demand has been redirected for carpenters who can work on a more streamlined process for mass production of furniture made from vaneer, block board, and formica. The traditional form of training persists in the SSI. 6.08 In some branches the general level of skills is low. Tanning and the manufacture of footwear are examples. This probably is due to the persistence of old-fashioned and "rule of thumb" methods in tanning and to the failure to introduce modern equipment and methods in shoemaking. The percentage of illiterates in small scale manufacturing, although decreasing, is still high. This situation tends to inhibit the introduction of new techniques. There is need for skill upgrading in tasks requiring high pre- cision work. For instance, in the engineering field, there is need to train fitters and machinists to become capable of making press tools, dies, jigs and fixtures; in furniture making, skills are needed to operate machines such as large lathes, copying lathes, multi-purpose milling machines, etc. as well as to inculcate working to tolerances. 6.09 Plan projections for skill requirements are confined to university degree holders. Requirements should also be identified at the vocational training level, and steps should be taken to expand existing or institute new programs to meet the demand. See also paras. 2.03 and 9.06.  - 30 - Forms of Employment 6.10 Employment in SSI comprises mainly (i) self-employed proprietors and partners in small workshops and factories; (ii) dependent workers such as apprentices, journeymen, etc.; and (iii) seasonal laborers with fixed-term contracts. In the upper end of SSI (establishments employing 20-50 workers) and in medium size enterprises (50-200) workers are hired with work contracts (isci); seasonal and casual labor is also employed. Workers in larger enter- prises are usually members of labor unions, have set working hours, and are covered by social security. Casual labor is recruited on the open market (amele-pazar) at daily wages for incidental work, such as loading trucks. Casual labor is drawn from a pool which would be best described as unemployed. Family members, paid or unpaid, can be found in all but the largest indivi- dually owned enterprises. Female participation is generally low, with the ready-made garment and weaving industrial branches the main employers of female labor. Child employment is common, particularly in SSI. Labor Mobility 6.11 Upward mobility is the basis of the apprentice system as apprentices become journeymen who, upon reaching a certain age and skill level, will set themselves up as independent craftsmen/masters. This route, however, is approaching its limits (para. 6.07). It involves low wages for the apprentice and in the long-term is more costly for the employer. Increasingly, journey- men have turned to positions in the larger of small and the medium size enterprises competing more effectively with the unemployed at the amele-pazar because of their skills/work experience acquired as apprentices. Inter-city mobility among skilled workers does not appear to be widespread, but external migration, at least in the past, appears to have been more common. Wages 6.12 In the private organized sector wages are determined by prevailing supply-demand conditions. 1/ Due to strong demand and increasing unionization, 1/ Nevertheless, wage levels in smaller enterprises tend to be much lower than in larger firms, suggesting the existence of a dual labor market, in part stemming from the prevalent apprenticeship system and the circum- scribed absorptive capacity of large firms. Discrimination due to age, sex and location of the establishment also plays a role. Furthermore, low-wage firms tend to be concentrated in typical small-scale industries, e.g. apparel, leather products, furniture and fixtures, lumber and wood products, etc., and geographic locations. Composed of numerous and small firms producing products or services that are quite similar, such industries are highly competitive and tend to have below average profit levels. Also, low-wage firms tend to have a high ratio of wages to total production costs and, because of their size and number, they do not dominate their labor markets. The above seem to suggest that, although "exploitation" and discrimination (e.g. sex, age) may have something to do with the observed low-wage levels in SSI, genuine inability of these firms to pay higher wages, due to strong competitive market pres- sures and a major percentage of costs represented by wages, may well be another important contributing factor.  - 31 - real wages rose dramatically between 1965 and 1971 - at 8.5% per annum on average, and remained relatively stable during 1971-75 (Appendix Table 3.21). They increased by an additional 15% in 1976, but declined by 2% in 1977 and 5% in 1978 (Appendix Table 3.23). 1/ In the public sector, real wages increased constantly throughout 1965-75, at an average annual rate of 5.8% (Appendix Table 3.22). In both sectors nominal wage increases outpaced productivity growth, and this appears to have contributed to the higher capital intensity of investment observed during this period. 6.13 Social benefits, which are proportional to wage levels, add an esti- mated 55% in 1976 to the basic wage. 2/ This incidence, coupled with labor laws which make laying off of workers very difficult or set high severance payments and the frustrating experiences from management/union confrontations, provides an inducement to substitute capital for labor. Consideration should be given to the possibility of reducing the effective cost of labor by funding such social benefits (or a major part thereof) out of taxes on company income or value added. 6.14 For social reasons, the Government's role in regulating wages - setting minima - has been growing in significance. Minimum daily wages rose by 3.2% per annum during 1970-1976, but appear to have declined in recent years (Appendix Table 3.23). However, the net effect of the Government's intervention appears to have been at best mixed, in the face of an excess supply of unskilled labor and the fact that the system is difficult to police except in larger establishments. Reportedly, only a fraction of even union- ized workers earn the minimum wage. Compliance is further discouraged because social security payments are based on the wage level. 3/ Aside from the issue of workability of the system in the light of the surrounding circumstances, circumspection should be exercised in setting minimum wages, since unreal- istically high levels, inducing further union demands to maintain relative wage levels, would reinforce tendencies to substitute capital for labor, thereby thwarting employment creation and exacerbating the unemployment situation. 1/ A survey conducted by the Employers' Confederation shows an average annual increase of 28.5% in nominal and 6% in real wages during 1973-77, while productivity of the non-agricultural sectors increased by 2.8% per annum. According to other sources of data (Social Insurance Insti- tute and Annual Survey of Manufacturing Industries), annual real wage increases were about 2.5%-3% during 1965-76. See World Bank, Turkey: From Crisis to Growth, 1979, Table 14, p. 187. 2/ Turkish Industrialists' and Businessmen's Association, The Turkish Economy: Prospects for Growth Within Stability, 1978, p. 212. 3/ Many workers in smaller establishments are not covered by social security despite the fact that coverage is mandatory. This occurs through collu- sion between the employer and the worker, since the practice is perceived to be to their mutual advantage.  - 32 - Labor-Management Relations 6.15 The majority of workers in medium and large scale industry belong to trade unions. Labor contracts are negotiated collectively and establish a set of rights (wages, working hours, social security, etc.) for officially contracting workers. Threatened by excess labor supply and in their drive to increase real wages in the face of rapidly rising prices, unions have become increasingly militant in recent years. Furthermore, there is proliferation of trade unions; for example, in Kayseri there are 31 trade unions. This is attributed to the fact that unions are organized by sector or craft, and that workers also belong to the local branch of one of the three nationally organized unions. 1/ The coexistence of many unions results in clashes and frustrating experiences for both management and workers since, to negotiate new contracts with management (once every 2 to 3 years), workers must elect their representative only from one union. Often, enterpreneurs get involved in this conflict in their interest to the election of a union representative with whom they feel they can work more effectively. Labor-management disputes have frequently resulted in protracted work stoppages and lock-outs, and consequent disruptions in production and loss of workdays. Persistence of such union-management practices will undoubtedly have deleterious effects on industrial performance and on the development of workable industrial relations. Changes in the trade union and collective bargaining laws are under consideration to the end of instituting the principle of a single union for each plant and industry, and increased government control over the union organs. The Role of the Guilds 6.16 The guild system perhaps is the most important institution binding together the large number of small workshops. At the base is the dernik which is the craft or vocational union. Membership, which is required by law 507 of 1965, signifies that one is qualified to pursue one's profession. Proof of membership is required to borrow, although that in itself is not sufficient to assure a loan. Proof is also necessary to register for the social security fund for self-employed (bagkur) or for applying for membership in a building cooperative. Derniks act to promote a variety of interests of its members, as pressure groups, to secure scarce raw material supplies, etc. They are instrumental in setting up cooperatives for building industrial estates. In some instances, they are involved in training members of their profession through sanat okul. Derniks are organized at the regional level in birliks, or regional associations, and represent all crafts and professions. The birliks act as channels of communication, passing on lists of raw materials and equipment needed by the derniks' members to Government. Also, the foreign exchange quota imports of machinery and raw materials is distributed through the birlik. At the national level, all the regional birliks form a Confedera- tion. The Confederation also includes an intermediate organization--the Association, which is formed by at least 50 derniks of the same craft joining together. 1/ Turk-Is, political neutral; DISK, more to the left politically than Turk-Is; and MISK, newly organized by the extreme right.  - 33 - Equipment 6.17 Both small and medium scale enterprises in all industries utilize a range of technologies. The origin of equipment in SSI varies considerably-- from locally made, old imports purchased on secondhand markets, to modern. In general, a good part of the machinery employed, especially in the leather, metalworking, woodworking and textile industries, is outdated, affecting adversely productivity levels. 1/ However, to a considerable extent, the observed low level of productivity is attributable either to frequent break- downs or to the machines being no longer capable of consistently turning out acceptable work. Both are due to lack of adequate maintenace. Bearing in mind the difference in the cost of labor between Turkey and the industrially developed countries, there is still a place in Turkey for the utilization of less sophisticated equipment. Serious losses of productivity are attributable to lack of attention to the lay-out of machinery and to good housekeeping. There is much "lost motion" in that machines are not correctly located for sequential operations. Admittedly, many of the small-scale manufacturing units are housed in commercial premises so structurally subdivided as to make a rational work flow impossible. In such instances, relocation to industrial estates appears to be the only solution. 6.18 In certain fields the technical processes used are outdated. For fine machining the lathe is still employed. There is very little equipment for plain, external and internal cylindrical grinding in the SSI. Wood turners who may be producing thousands of identical items (e.g., tapered legs for furniture, bobbins for the textile trade) have neither copying lathes nor even taper turning attachments. In the production of metal hollow-wares spinning has not replaced deep-drawing. The tanning methods used in the private sector are unscientific and not quite effective. 6.19 In contrast, most medium scale establishments and the larger SSI employ fairly up-to-date technology and equipment, a very high percentage of which is imported. However, a great deal of copying of imported equipment is apparent, not only of individual pieces but of an entire production line, which is remarkable. Examples include an automatic extrusion unit for manu- facture of aluminum channels, Jacquard type looms for producing woolen "oriental" carpets, equipment for manufacturing steel cans, etc. All of the medium scale establishments employ engineers with university training. Workshops 6.20 SSI workshops which have not moved from traditional locations in central city areas operate under very unsanitary, cramped, badly lit and 1/ The mission's impressions from visiting a good number of workshops are further detailed in the sections reviewing the foundry (XI), engineering (XII), leather processing and footwear (XIII), wood-based (XIV), and garments-making up (XV) industries. See also Annex F in World Bank Report, No. 1359-TU, 1977.  - 34 - congested conditions. 1/ As a result many derniks have taken the initiative to establish industrial estate construction cooperatives and relocated members to new locations outside the city. Many of these new sites, however, have within ten to fifteen years come to resemble the old workshop locations since little attention had been paid to planning for expansion. In Kayseri, the SSI estate is a typical example. With increasing business, most workshops are now overcrowded, with machinery and production tending to spill over into roads in front of the workshops. It would be highly desirable for SSI owners to seek planning assistance during relocation to industrial estates. A good example of what can be achieved with planning is the automobile repair estate in Istanbul which is extremely well laid out with excellent facilities and infrastructure and room for future expansion. Marketing and Distribution 6.21 Most small establishments of the workshop type serve local markets. When not catering directly to their clientele, marketing is handled by whole- sale traders who place orders for specific quantities with the workshop. As such, the small scale manufacturer is forced to accept a lower margin of profit and often wait for payment until the goods are disposed of by the trader. Larger establishments which operate on regular production plans use distributors and large wholesalers to distribute their products on a regional or national basis. Often, these products are made to specifications of the buyer for marketing by chain stores. This is particularly true in garments, footwear and textile products. Most sales of SSI are to the local market except in special product areas such as leather garments. Profitability 6.22 Profitability in SMI, even among establishments in the same subsector varies widely. Official data by size of enterprise are not available. Spot checks in enterprises employing 10-200 workers reveal gross profits as percent of sales ranging from 5% to 20%, which is reasonably satisfactory. In estab- lishments employing 1-9 workers, a recent study 2/ estimates that gross profits as percent of total cost of production (including depreciation) were 27% in 1970 compared with 18% in establishments employing 10 or more workers. The percentage drops to 12% when the imputed cost of the services rendered by owners and unpaid family members--calculated on the basis of the going wage rate in this segment of industry--is taken into account; profitability becomes negative if the imputed costs of such services is based on wages prevailing in large scale manufacturing. This implies that the remuneration of the owners of very small establishments vacillates somewhere between that 1/ There is no information on the extent to which working space is owned or rented. In Kayseri, of the 1,500 workplaces, 758 (or 66%) were rented, 220 (or 18%) owned, 181 (or 16%) unknown. This pattern may not be atypical for the country as a whole. 2/ Gurel Tuzun, A Study on Small Scale Manufacturing in Turkey, 1979.  - 35 - of wage earners in small and in large scale industry. However, due to poor bookkeeping practices and underreporting, these estimates must be treated with caution. The evidence though corroborates the belief that the survival of the very small enterprise depends largely on the hard work of the owner(s) and his family. Cooperatives 6.23 The better organized cooperatives appear to be the 650 guarantee cooperatives, with a membership reaching some 415,000, which assist their members (mostly artisans, SSI and service industries) to obtain credit through the Halk Bank (para. 4.04). Yet, the financial soundness of many of these cooperatives is doubtful. A step in the right direction has been the estab- lishment in 1978 of a Fund to guarantee losses from defaults on loans. Cooperatives have made little progress in entering production (see para. 7.16), marketing and procurement. Since cooperatives are called upon to play an important role in fostering the reorganization of small producers in many subsectors (see Sections XI-XV), SPO should take the lead in commissioning a study which (a) would assess the present system and identify its shortcomings; and (b) would make specific recommendations as to how such a reorganization could be approached and implemented. The study should also consider other complementary measures that could help reorganize the SSI sector (e.g. mergers), including the requisite policy framework in the light of Turkish circumstances. The potential role of financial intermediaries (e.g., the Halk Bank) in this effort should also be examined.  - 36 - C. THE INSTITUTIONAL AND POLICY FRAMEWORK VII. GOVERNMENT PROGRAMS IN SUPPORT OF SMALL-MEDIUM SCALE INDUSTRY Role of the Government 7.01 The Government's objectives in the SMI sector, enumerated in the First Plan (1963-67) document, were to: (a) provide better and safer working conditions for SSI; (b) promote cooperative effort among small firms; (c) bring scattered firms and small workshops together in order to facilitate the provision of cormnon facilities; (d) encourage the decentralization of industry; (e) promote and facilitate subcontracting; (f) stimulate the integration of existing small firms rather than encourage the establishment of new ones; (g) extend technical assistance to small firms; (h) relieve congestion and disloca- tion of traffic in the cities. Subsequent Plans, including the 1979-83, have reaffirmed these policy objectives. The Government's objectives, however, have yet to be fully articulated in specific measures, and the results so far have been inconsequential. Implementation has been slow in providing training, technical assistance, marketing services, etc. to SSI by public entities as the program of the Small Industry Development Organization (SIDO), intended to provide industrial extension services on a national basis, has made a belated and slow start. Financial support, directly or through the Halk Bank, has been very modest. A notable exception has been the subsidized program of industrial estates. This tardy implementation of Plan objectives reflects lack of commit- ment and vacillating views of past Governments as to the potential role of SMI in industrial development, and insufficient budgetary appropriations. 7.02 Responsibility for the implementation of the technical/financial assistance programs and the various administrative regulations affecting SSI 1/ rests with the Ministry of Industry and Technology (MIT). Admin- istrative procedures per se do not seem to be particularly cumbersome. Rather it is the processing phase and the manner of conduct which create inconvenience and delays, and efforts to improve efficiency in these areas would be beneficial. 1/ MIT defines SSI as establishments with machinery and equipment valued up to TL 10 million.  - 37 - 7.03 The Ministry administers the following programs 1/ supportive of SMI development: (a) industrial estates; (b) the National Small Industry Development Organization (SIDO) program, including the pilot service center known as the KUSGEM project at Gaziantep; (c) industrial training; and (d) subsidized term credits to manufacturing cooperatives for machinery and working capital. An assessment of these programs follows. The Industrial Estates Development Program 7.04 Formal Government action to promote industrial estates began under the First Five-Year Plan (1963-67). Earlier, in the mid-1950s, a few guilds located in decaying urban centers had sponsored, financed and built estates with their own resources in response to the desperate need for more adequate factory buildings and infrastructure. The idea spread and, stimulated by the availability of Government credit and technical assistance, today there are 195 SSI estate projects (including 19 classified as municipal projects) - 52 operating, 59 under construction, 11 being tendered, and 73 in early phases of study. Organized Industrial Zones for larger enterprises number 20 in varying stages of completion and occupancy. MIT's budget for support of industrial estates rose from TL 10 million in 1964 to TL 1.2 billion (US$44 million) annually in 1977 and 1978, for a total of TL 3.5 billion. In addi- tion to these MIT/Halk Bank financed projects, it is estimated that some 100 privately financed SSI estates are in existence, completed or under construc- tion. There are no reliable data on total and sub-sector occupancy, employment, output, etc. The mission estimates that, when all current projects are completed by the mid or late 1980s, employment in the estates could reach 600,000, or about 25% of total labor force in manufacturing. By any measure, the industrial estate program is massive and promises to maintain its momentum. 7.05 Responsibility for the industrial estates program rests with the Department of Organized Industrial Zones and Small-Scale Industrial Estates in the MIT. This Department, with a staff of about 30 land surveyors, engineers, 1/ With the exception of industrial estates for which the Department of Organized Industrial Zones and Small Scale Industrial Estates is in charge, the other programs are administered by the Department of Small Scale Industries and Handicrafts. The range of SSI activities covered by the latter department includes: (a) Small Industry Development - training, extension services and common workshop facilities; (b) Pro- curement - alloting import quotas and assisting cooperatives in the procurement of raw materials, semi-finished goods and machinery; (c) Small Industry Cooperatives - credit facilities and manufacturing cooperative organization and monitoring; (d) Industrial Fairs; and (e) Handicrafts Training. The approved and budgeted staffing pattern provides for 20 professionals - engineers, economists, business admin- istrators - headed by a Director General, of which 12 are working and 8 are being recruited. The staff appeared qualified. In each province there is an MIT field office to provide liaison and on-the-spot coordina- tion with other agencies and the private sector.  - 38 - economists and technicians, has a single clearly defined purpose - to assist organized groups of small-medium size enterprises 1/ in obtaining modern fully-serviced factory premises at reasonable cost through the construction of industrial estates. To this end, it provides technical and financial assis- tance to three types of project sponsors: (a) SSI construction cooperatives; (b) municipalities; and (c) associations of small to large scale firms, i.e. local Chambers of Commerce or Industry. Projects carried out by SSI coopera- tives and municipalities are designated as Small Industry Estates; these provide individual shop units of 50-400 m2 and occasionally up to 600 m2. Projects sponsored by Chambers are designated as Organized Industrial Zones and provide only improved land and services; the individual factories are built by the owners to their own specifications and range in size between 800 and several thousand square meters. The primitive conditions under which SMI traditionally has operated--dangerously ramshackle and overcrowded buildings; inadequate lighting, ventilation and sanitary facilities; and congested, sometimes temporarily impassable, transport facilities--need hardly be empha- sized. The industrial estate program has effectively addressed the infrastruc- ture problem and must be recognized as one of the pillars of industrial development policy in Turkey not only because it benefits directly thousands of small-medium size enterprises, but also because it fosters regional indus- trial growth and urban decongestion. The institutional arrangements, proce- dures, financial arrangements, technical aspects, industry mix, geographic distribution, cost and benefits, etc. of the industrial estates program are discussed in detail in Annex C. In what follows, the main features and achievements of the program are highlighted. 7.06 Partly because of the technical assistance provided by MIT, but also by a process of "learning by doing", earlier errors in site location, 2/ building design, road and utility network specifications, inadequate expansion areas, failure to take into account project related housing, transport and waste disposal demands, are being eliminated. The newer projects evidence an improved level of professional competence in design and execution. Increas- ingly, the projects are also expanding their scope; many now plan common laboratory and shop facilities for raw materials testing, precision machining, tool and die making, product quality control, and special processing needs such as heat treatment and dynamic balancing of rotary units, meeting rooms and facilities for training workers, technicians and managers, technical reference libraries, etc. These facilities and the concentration of large numbers of SMI engaged in related activities at specific locations will enhance the impact and efficiency of MIT's technical assistance programs as well as the expanded credit and business advisory services of Halk Bank to be initiated under the proposed Bank loan. 1/ A precondition for Government assistance. 2 Many earlier projects have already been engulfed by urban growth and its occupants are planning second stage projects further out.  - 39 - 7.07 Undoubtedly, the dynamism and strength of the industrial estates program derives from the private sector--the collective perception of need and the will to initiate and execute projects cooperatively by groups of entrepre- neurs. This is not to denigrate the important role of the public sector, particularly by providing funds at low interest rates and long maturities. Indeed, the public/private sector "partnership" and its success in providing modern factory premises and infrastructure offer a model that should be studied by other developing countries. A unique feature of the Turkish model is that, by the simple expedient of requiring the future occupants to form a cooperative to act as the executing agency, to plan, finance, build, occupy, and manage their own project, a major uncertainty regarding future occupancy which usually plagues most industrial estates in other countries simply disappears. 7.08 There can be no question that the Turkish industrial estate program is a growing success and is making an important contribution to industrial development. The factors contributing to this favorable result are many, including the stage of industrialization which the country has reached, the fact that so many Turkish workmen have had the chance to work and gain skills in Western Europe, the policies of the Ministry of Industry and Technology and, not least, the character of the Turkish small entrepreneur himself. But perhaps the most important factor has been the fact that estates are set up spontaneously by groups of industrialists who clearly see them as beneficial, can construct them according to their specific requirements, and are willing to contribute financially to their establishment. Government support supple- ments private effort and financial means. The SIDO Program 7.09 A National Small Industry Development Organization (SIDO) program has been evolving since 1968 with UNDP/UNIDO finance and technical assistance. In the first phase (1968-1978), a regional pilot program was implemented at Gaziantep (the KUSGEM project) which included the following elements: (a) a model industrial estate; (b) common workshop and laboratory testing facilities; and (c) technical consulting services and training in project evaluation and credit applications, factory design and operations, product improvement and new product development. The services are made available on-site at the KUSGEM center and as extension services. Construction of the Gaziantep facilities has not been completed and there are various budget, staffing and functional problems adversely affecting its operations. 1/ 7.10 MIT, with UNIDO assistance, proposes to organize a system of tech- nical extension services and common facilities for SMI on a national scale, 1/ The background, objectives, organization, and implementation of the KUSGEM project, as well as the mission's evaluation of its performance are discussed in Annex D.  - 40 - with headquarters in Ankara and offices in initially half a dozen and ulti- mately 14 major regional centers. The Gaziantep project will serve as a model, with modifications as appropriate to take into account regional needs and priorities. The stated objectives of the SIDO program include inter alia: (a) provision of technical, management and marketing consulting services to new and existing small-scale units in different regions throughout the country; (b) assessment of the performance of industrial cooperatives; (c) promotion of subcontracting; (d) carrying out subsector reviews, regional industrial surveys, market studies, etc.; (e) conducting training courses in different technical skills; (f) identifying needs and assisting the SSI sector in obtaining finance and in improving financial planning; (g) preparing feasibility studies for prospective entrepreneurs; (h) devising a strategy for export promotion and rendering assistance and advice to SSI units on export marketing; and (i) studying the legal, economic, financial and technical problems of SSI and recommending suitable policies and programs. In addition, to assist the approximately 110 small foundries in the Ankara region, a particularly high priority target group, a specialized service center is proposed to be developed under the program which will provide technical extension services, common facilities such as material testing, special machining and heat treatment, quality control laboratories, etc., and training facilities to upgrade the skills of foundry workers. 7.11 The specifics of organization, staffing and administration are still under study. But it is of cardinal importance for the success of the program that SIDO function as an autonomous institution under MIT's aegis. By not being part of the civil service system, conditions of employment and compen- sation of SIDO's staff could be set at levels that should attract and retain qualified personnel from the private sector; services, technologies and other inputs could be better adapted to the needs of prospective users; businessmen's fears of Government prying into their affairs would be allayed. These features should certainly enhance SIDO's effectiveness and performance. MIT would provide policy guidance, budget control and also monitor performance. 7.12 The SIDO program is envisaged by MIT as the basic delivery system for all forms of non-financial services needed by the small-scale industry sector throughout the nation. At the present time, certain complementary activities are stilled managed by Ministry staff (e.g. some types of training, mobile shops, special credits for cooperatives), but if SIDO grows as planned, it is likely that responsibility for many of these complementary activities will be taken over by SIDO. However, the experience already accumulated at the Gaziantep pilot project suggests that implementation of this ambitious expansion program would be very difficult, particularly within the assumed time frame (1981-85). In view of the limited results to date at Gaziantep, it would be prudent for the Ministry to critically review its current proposal to establish a SIDO program of the scale envisaged. There is no assurance that the staffing and functional deficiencies of the pilot project will not be repeated in the national program and on a larger scale. A more rational approach would be to complete the Gaziantep project and provide adequate funding for the full staff and operational level originally planned, in an effort to achieve a more effective pilot operation. Among the measures needed are: (i) improved levels of compensation to attract and retain qualified staff; (ii) expanded budgetary allocations to cover non-salary operating expenses; (iii) a more businesslike approach to operating the pilot industrial  - 41 - estate, e.g., immediate negotiation of sale and rental contracts upon comple- tion of a cluster of buildings; (iv) a revision of pricing policies to reflect market rates to the extent possible (some small element of subsidy may be needed in disposing of buildings over the near-term); and (v) more intensive promotion to increase the use of idle plant and common service facilities with service charges applied where feasible (at least to larger firms). Once the Gaziantep project achieves more visible success in assisting small-medium scale enterprises on a reasonable cost/benefit basis, a national SIDO program would be much more easily replicable. In the light of the above, the Govern- ment should perhaps consider supporting at this stage a scaled down, appro- priately phased, and thereby more manageable and affordable SIDO program. 7.13 Specifically, a more realistic approach would be to establish initially I or 2 regional extension service offices in major urban centers, without the huge investment in physical facilities carried out in the KUSGEM project. Each office could have a professional staff comprising 3 to 4 mechanical and electrical engineers, 2 financial analysts, I economist and 1 marketing specialist. These regional teams would need to be sufficiently mobile to visit individual establishments to diagnose problems and to provide assistance. In addition, each regional office would be responsible for undertaking market studies, industrial surveys and subsector reviews on a regional basis. A small group of technical and management staff could admin- ister the program from Ankara. Such a SIDO program would complement the Business Advisory Services to be provided by the Halk Bank and could draw on private consultants. This pilot program could be gradually replicated to cover additional regional centers. The Government should also reconsider the actual need and ponder the problems likely to arise from establishing the proposed foundry extension center at Ankara. In this regard, it might be advisable for SIDO to investigate the possibilities of arranging training for workers from small foundries at large private foundries and subsidizing the costs involved. The mechanical and metallurgical testing facilities already available in large foundries could probably be utilized by small foundries with the Government subsidizing the costs. After an evaluation of the experi- ence to be gained from this effort, a decision could be made as to whether or not to establish the proposed foundry extension facilities at Ankara--although a specialized extension service team can probably be justified. Industrial Training 7.14 The program of skill upgrading for industrial workers supervised by MIT started in 1964 with USAID financing through a 3% interest spread on a loan to the Halk Bank. The fields of training include welding, technical drawing, furniture making and woodworking, machining and sheet metal work. Trainees are semi-skilled workers from the private sector. Until 1970, owners also participated in the program. Classes are held in the evenings and during summers in high schools and vocational training facilities in selected urban centers. The duration of the courses is about 10 weeks, totaling 144 hours of instruction. Most of the training involves practical work. A committee of four members, with representatives of MIT, Ministry of Education, the Union of Guarantee Cooperatives, and the Halk Bank administers the program. This committee selects the locations for holding the training programs and also  - 42 - approves the annual budget prepared by the Halk Bank. The selection of candidates is undertaken by a sub-committee consisting of the local Halk Bank branch manager, the manager of the school and the head of the local "Birlik" (association of unions of cooperatives). There is no charge to the partici- pants attending the course. 7.15 The number of persons that received training under the program and the number of centers utilized are indicated below: Year No. of Trainees Number of Centers 1972 1,452 9 1973 1,210 8 1974 708 6 1975 218 3 1976 516 6 In 1977, no program was offered because instructors were unavailable. As the AID fund is gradually decreasing, the program is being curtailed because of inadequate funding. In 1978, the Ministry of Finance approved TL 750,000 for the program. Halk Bank's own funds are not used. In 1979 the program called for the training of 400 students in 5 urban centers. During 1974-75, training programs had also been developed over a wide range of subjects, technical and managerial, for small industrialists by the Small Industries Development Center (KUSGEM) at Gaziantep. It appears that they have been discontinued. In the light of the needs for training (para. 9.06), the present programs appear inadequate and steps should be taken to rectify the situation. Term Credits to Manufacturing Cooperatives 7.16 MIT is responsible for providing Government term credits to manu- facturing cooperatives for machinery and working capital. Since 1964, MIT and Halk Bank have collaborated effectively in administering a special fund to encourage and support SSI manufacturing cooperatives. 1/ To date, 67 manufac- turing cooperatives--with membership ranging from as few as 10 to over 100-- have been granted loans. In 1978, TL 75 million was allocated for such lending and TL 150 million was budgeted for 1979. The program is modest in scale and most loans are actually for working capital. The practicality of extending loans at highly concessionary terms should perhaps be assessed, and the policy reconsidered (see para. 8.03). 1/ Procedurally, in a typical machinery or working capital loan application, the Halk Bank prepares a short "feasibility" analysis and transmits it to the Ministry. The application then goes back to the Halk Bank which, acting as the financial agent (for which it collects a management fee), disburses the approved amounts and follows up on interest and principle repayment. Loans for equipment are 3.5% per annum for five years with two years of grace, and for working capital at 6% with one year grace period. Cooperatives do the procurement.  - 43 - VIII. POLICY INSTRUMENTS AND THEIR IMPACT ON SMALL-MEDIUM SCALE INDUSTRY 8.01 Until very recently, the SMI sector has had very limited access to Government incentives (exemption or deferral from customs duties, tax holidays, interest rate subsidies, export rebates and financing, etc.), since the criteria of eligibility and the administration of the scheme were designed to support large scale operations. Inadvertently, this policy framework dis- criminated against SMI. In 1979, in recognition of the potential contribution of SMI to industrial development and employment generation, the Government took steps to rectify the situation (see para. 8.02). Nevertheless, there remain certain other areas where the concern of the Government is invited, which include the existing arrangements concerning export rebates and financing and the institution of appropriate procurement and subcontracting policies. Such policy instruments and their impact on SMI, as well as suggested correc- tive action, are discussed briefly below. Investment Incentives 8.02 As conceived and administered in the past, the eligibility criteria for investment incentives (e.g. scale and technological requirements) ruled out all small and many medium size establishments. 1/ Inconvenience and delays during the processing of applications also tended to discourage smaller applicants. However, in 1979, though effectively since March 1980, the newly established Incentives Department at the Office of the Prime Minister began issuing Letters of Encouragement for projects in priority sectors with an investment cost not exceeding US$750,000 2/, thereby providing incentives that otherwise were available only under the Certificate of Encouragement. During the year ended in June 1980, about 50 small projects have received Letters of Encouragement. Reportedly, it takes about four weeks to process an application for a Letter of Encouragement. If organized and implemented effectively, this newly introduced system of Letters of Encouragement could be an important step toward putting SMI on an equal footing with large enterprises. In addition, in a proposed Bank loan to SYKB, the Government intends to provide preferential 1/ Eligibility thresholds vary among industries, and are usually set administra- tively at high levels. In general, projects with an investment cost less than US$1 million were only sparingly considered for a Certificate of Encouragement which entitled the beneficiary to a number of incentives. In the decade ended in 1979, some 5,200 certificates have been granted mostly to large (public and private) enterprises for an estimated fixed investment exceeding TL 200 billion. The actual number of recipient firms, however, is much smaller because of the issue of repeater certifi- cates to the same enterprises. 2/ In this connection, the Government may wish to consider assigning eventually to SIDO the task of evaluating such applications (para. 7.10).  - 44 - treatment to SSI 1/ by agreeing to assume the foreign exchange risk, appreciat- ing that small entrepreneurs are particularly affected by the uncertainty associated with devaluations, since they have great difficulty in hedging or planning against foreign exchange rate changes and in absorbing sudden losses. 8.03 The Government nevertheless might wish to reconsider the expediency of subsidizing interest rates and the importation of machinery and equipment for both small and large enterprises, along the lines suggested in para. 2.06. There is no hard evidence that the inclusion of an interest subsidy in the incentives package acts as a stimulant to investment and as an effective device to direct investment in desired areas. Yet, subsidized interest rates tend to distort the true price of capital, introducing a capital-intensive bias and thereby influencing the choice of technology. Besides, if there is strong conviction about the need for an additional inducement, it could be provided in a different form--e.g. by increasing tax credits. As for the interest subsidy to SSI cooperatives, it is more likely that availability, suitable credit maturities, and less strict security requirements are of much greater importance than the price of capital. By the same token, exemption of imported capital goods from duties also results in reduced cost of capital and tends to promote the use of capital-intensive techniques. Deferment of duties should perhaps suffice; but, with persistent high inflation rates, deferment becomes tantamount to exemption. Export Rebates 8.04 As presently designed, 2/ export rebates discriminate against smaller exporters who fail to qualify as bona fide traders. The incentive could be granted to smaller firms which combine in order to reach some minimum absolute value of exports--perhaps using an export agent. More generally, the present system could become more effective by providing greater rewards to better export performers and inducing backward integration, e.g. by instituting a greater degree of progressivity based on higher value added. There could be an additional export stimulus by extending the incentive to immediate inter- mediate suppliers of exporters to induce them to upgrade quality and thereby facilitate the development of new export outlets. Export rebates for indirect exporters have been used in a number of countries, most notably in Korea. Admittedly, its administration is more complex than for direct exports but is workable. 1/ Defined as projects with an investment cost for machinery and equipment up to US$350,000. 2/ The export rebate regime provides cash payments to successful exporters, calculated as a percentage of the selling price according to a published rebate schedule which changes from time to time. Large volume exporters (TL 1.8 million or more in a given year) are entitled to an additional 5 percentage points.  - 45 - Export Financing 8.05 The Government has taken courageous measures inter alia to improve the export institutional framework (e.g. streamlining of export procedures and licensing, easing of imports for production of export items, etc.), and to increase export credit availability. Export financing 1/ could become more effective for SMI if: (a) the credit is made available on a constantly replenishable basis rather than on the basis of the current "annual cycle" which requires commitments to be made by exporters on the basis of the Govern- ment budget cycle rather than on their own selling cycle; (b) it is made available more expeditiously; for instance, shorter deadlines (say, 7 days) should suffice for the Ministry of Commerce to grant approval (or disapproval); (c) it is made more accessible to smaller exporters (publication of a guide on how to obtain this credit would be helpful); (d) exporters are allowed to utilize their local banks to process applications, thereby minimizing contacts with Government agencies. Procurement 8.06 Within the present industrial structure of Turkey, there is con- siderable potential to expand markets for SMI through access to public sector procurement. The various schemes whereby the Government can foster the develop- ment of SMI through public sector purchasing contracts and tenders are discussed in Annex E. The Government should identify particular items suitable for procurement by SMI and adopt appropriate programs (e.g. improved information about public procurement opportunities, set aside programs, etc.) to ensure SMI participation. The Department of Small-Scale Industries in the Ministry of Industry and Technology, properly strengthened, seems to be the most appropriate to administer this activity. Subcontracting 8.07 The importance of developing commercial linkages between SMI sup- pliers and medium to large scale producers of finished goods is overriding. 2/ Aside from the typical advantages of subcontracting to the principal firms discussed in para. 12.21, footnote I and Annex F, para. 5, recent trends in 1/ Qualifying firms, i.e. that can guarantee a certain level of exports (usually 20%), are eligible for Central Bank funds for working capital at interest rates that are 7-8% below the prevailing rate. 2/ The lines along which subcontracting, as an instrumentality to facili- tate the growth of SMI, could be promoted through a concerted effort by private and Government initiatives are suggested in Annex F. The discussion includes inter alia types of subcontracting, conditions for success, means to encourage subcontracting in the private and public sectors, potential benefits to the parties involved, and the major problems frequently encountered.  - 46 - Turkey tend to enhance the attractiveness of subcontracting arrangements. Manufacturing enterprises are required to increase each year the proportion of locally produced parts and components. The increasing unionization of the labor force in larger enterprises has brought along a constant demand for higher wages, fringe benefits, and severance payments, including substantial loss of man-hours in strikes and low utilization of specialized equipment. These circumstances have fostered an attitude opting for farming out the production of components. The process could be reinforced if the cost of investible funds were to be raised by disallowing interest rate subsidies and exemption of imported equipment from duties as suggested in para. 8.03. The key constraint is availability of efficient and reliable SMI suppliers. Nevertheless, some activity is going on in the private sector, particularly in the engineering industries. Specific measures to promote subcontracting arrangements between large firms and SMI are outlined in paras. 10.06-10.09, 12.32-12.34. D. SECTORAL ISSUES AND PROPOSED REMEDIAL ACTION IX. PROBLEMS HAMPERING THE EFFICIENCY AND GROWTH OF SMALL-MEDIUM SCALE INDUSTRY Internal Constraints 9.01 Fundamental internal and external constraints are affecting SMI, creating conditions which are inimical to efficient operation and which, as a result, deserve immediate attention. Major internal problems adversely affecting productivity, quality of output or service and, by extension, constraining the growth of SMI and particularly of SSI, include: (a) out-of- date machinery; (b) deficient technical and management methods (product design and new product development, process engineering, production planning and work methods, raw and intermediate materials selection and procurement, quality control, preventive maintenance, etc.); (c) poor working conditions and housekeeping, in part due to shortage of suitable factory premises and infra- structure; and (d) marketing problems. Annex G provides in a tabular form an assessment of the deficiencies in five important industries--engineering, woodworking and furniture, leather processing and footwear, chemical, plastic and rubber products, and garments/made-up goods. 1/ Not all of these problems necessarily apply to every enterprise. Their relative importance varies with the particular industrial activity, size of establishment, and the surrounding 1/ For industry-specific problems see also: foundries (paras. 11.11-11.14); engineering (para. 12.21); leather processing and footwear (paras. 13.02, 13.08, 13.15, 13.19, 13.22); wood-based industries (paras. 14.05, 14.07, 14.11, 14.13, 14.16); and garments (para. 15.8). See also paras. 6.07, 6.08, 6.15-6.21.  - 47 - circumstances of each enterprise. Nor do they imply that the small, and possibly medium scale, enterprise is inherently inefficient. Their existence rather attests to a built-in weakness, namely the lack of specialized mana- gerial, technical and marketing expertise within a small organization. To rectify the situation, a comprehensive program of technical assistance is called for--aimed primarily at the SSI sector. 1/ 9.02 At this juncture, the "technological gap" and the attendant need for "vertical" as opposed to "horizontal" growth of production units pose a more intractable problem. For SMI using traditional skills to produce basic consumer goods such as furniture, garments, footwear, simple building hardware and household utensils for a relatively undemanding market, this is not a serious problem. Productivity and quality in those subsectors can usually be upgraded without too much difficulty by introducing and financing better tools and simple machines (hand operated or powered) and demonstrating their ad- vantages on-the-spot or in formal shop practice courses. Industrial develop- ment in Turkey, however, has reached the point where future growth of the SMI sector must be based increasingly on modern, efficient production techniques and a concomitant shift of the product mix to more sophisticated designs and special emphasis on the provision of parts, components and sub-assemblies for large-scale factories (subcontracting or ancillary linkages). In this respect, by far the most important sub-sectors are in metalworking--foundries, forges, die-casting and machining; cutting, forming and welding of sheet metal; electroplating and other finishing processes--and capital goods production. 9.03 To make the transition, however, Turkey's endowment with a broadly based class of skilled craftsmen does not suffice. To be sure, the pool of this traditional craftsmanship has been considerably enlarged in recent years as thousands of Turkish workers and technicians have returned from Europe with new skills, familiarity with more sophisticated tools and machinery, and at least an awareness of the need for and competitive advantages of "scaling up". Also, there is no doubt that as a result of this technological input, certain industrial sectors--including the all important metalworking and capital goods subsectors--are already experiencing added growth and diversification. Nevertheless, there are problems that must be addressed in order to channel growth into more productive and efficient lines. Herein lies the challenge to policy makers and institutions. 1/ Medium scale establishments have less need for general institutional technical assistance as in-house technical expertise is usually adequate. Areas requiring outside assistance refer primarily to marketing, finan- cial planning, management functions, and to technical problems which are specific to the industry and require specialized know-how (e.g. metallo- graphic analysis in foundries, quality and process control in chemicals). Such types of technical assistance are usually provided by suppliers of machinery, engineering consulting firms and universities (e.g. Middle- Eastern Technical University). Financial intermediaries could help diagnose problems, extend assistance, or refer firms to appropriate delivery agencies.  - 48 - 9.04 A major impediment to growth under the present institutional set up is the prevalence of the master craftsman (usta) who is still wedded to the guild concept of industrial production, i.e. local associations of micro- factories owned and managed by craftsmen/entrepreneurs and staffed by appren- tices and journeymen numbering, typically, less than ten workers. The numbers and output of the shops associated in each guild are more or less in balance with demand. Under the circumstances, growth is "horizontal"--proliferation of micro-factories as newly certified master craftsmen are admitted to guild membership in response to growing demand. The system works well in tradi- tional activities and isolated regions where the role of the micro-factory sector serving local markets is historically dominant. 9.05 For modern SMI, on the other hand, considerations involving product design and quality, operating efficiency and the importance of meeting promised delivery schedules, market expansion nationally and internationally, dictate size of factory and management sophistication above the micro-range. Adequate production facilities, specialized staff and management skill (financial, tech- nical and commercial) are critical to the success of the enterprise, whether producing finished products or components for larger factories. Modern small-medium scale enterprises are, of course, emerging spontaneously as the more entrepreneurial-minded "ustas"--especially noticeable among those that have been exposed to Western European influences--alone or in groups, seize the more obvious opportunities. 1/ But the trend is constrained by their inadequate business/technological talents, inappropriate business organization forms (para. 9.04), and insufficient assistance and capitalization (paras. 9.06 and 9.07). The constraints are especially critical in the 10-100 workers enterprise size range where the potential for growth is greatest. 9.06 The introduction of more sophisticated techniques and specialized machines is creating a demand for such highly skilled craftsmen as designers, tool and die makers, tool setters and precision machinists in the engineering industries. The requirements of the plastic branch of the chemicals industry must be met from the engineering trades, particularly by upgrading the skills of fitters and machine men. In the furniture making branch contemporary designs are needed. There is lack of experienced operators of shoe making machinery and in the woodworking trades. Insufficient supply of such skills may prove a restraining factor on the growth of these industries. To remedy the situation, there is need to train more skilled workers and to upgrade the skills of those employed in these crafts. The ready-made garment industry is able to train quickly the workers it requires (though not at the supervisor- foreman level), but the other sectors of SMI rely mainly on the traditional apprenticeship system to train workers. This system is rather inelastic; it involves low wages for the apprentice and in the long-term is probably costlier to the employer. It could be supplemented by more formalized apprenticeship schemes based on explicit contractual terms, accelerated training and skill upgrading programs, and greater reliance on formal vocational training. The 1/ Individuals with technical or business background is another important source of SMI entrepreneurship in Turkey.  - 49 - issue deserves to be reviewed from a longer term perspective by Government Authorities in collaboration with the respective associations of the various trades, in order to ascertain structural trends in demand and supply in the relevant range of skills and possible need for reforms of the existing train- ing systems, to identify technical assistance and finance requirements, and to map out a strategy, requisite policies and a timetable for imp>-mentation. External Constraints 9.07 External problems tend to compound the internal ones. Limited access to institutional finance and technical assistance make it difficult for smaller firms to improve their operations. These deficiencies are insti- tutional in origin, reflecting past ambivalence regarding the development potential of the sector and inadequate allocation of credit and technical assistance by intermediaries supposedly responsible for assisting SMI. 1/ The incentives system appears to be biased in favor of large-scale industries but, more importantly, administrative complexities (e.g. minimum project size, minimum number of participating firms) make it difficult for smaller firms to take advantage of such benefits as are available (paras. 8.02-8.05). Shortage of foreign exchange severely affects the operations of those sub-sectors that are dependent on imported raw materials, spare parts, components or capital goods. Low capacity utilization, which was observed by the mission in a substantial number of shops visited, can be directly attributed to delayed imports as one of the major, if not the single most important cause. 2/ Continued power cuts also have a disruptive effect. Finally, the purchase of inputs from inefficient or dominant domestic producers at prices well above world market levels due to import quotas or prohibitions erodes significantly profit margins, given that SMI operates in a fairly competitive environment. The Government's determination to rationalize the system of protection should help alleviate the situation. Impact on SMI 9.08 Despite the problems enumerated above and the attendant low produc- tivity, SMI generally remains viable, is retaining its relative importance and is actually increasing it in some industries, e.g. foundries, light metal fabrication, machinery, sawmills and furniture, rubber and plastics, 3/ where there has been a noticeable increase in the number and size of firms. The 1/ On the role of commercial banks and the Halk Bank in SMI financing see paras. 4.02-4.10. 2/ E.g., one completed automobile valve factory in Gaziantep was unable to start operations for lack of imported special alloy steel. 3/ On the other hand, the footwear industry is lagging the general trend and is in urgent need of revitalization (see paras. 13.27-13.30).  - 50 - explanation for the seeming paradox--encouraging performance in the face of serious sectoral problems and constraints--lies in the remarkable initiative, character and skills of the master craftsman, and more often than not, the entrepreneur/manager of the shop, whose individual capabilities are reinforced by pooling of resources in guilds, cooperatives and industrial federations of all types. Such associations of small businesses are able to exert the necessary leverage in obtaining a share of available term-credit for equipment and working capital, construction loans for industrial estates and technical and procurement assistance, to the extent available, for its members to a degree that could not be achieved by the individual firm. Additional factors accounting for SMI's survivability and self-sustained growth include: low overhead, long work hours by the owner, payment of low wages, employment of child labor, use of scrap material (e.g. in plastics), flexibility in meeting local needs promptly, production of low-priced items ensuring a market niche, involvement in activities with low scale economies. Operating in a protected market is a further advantage, though shared with firms of all sizes. X. PROPOSED TECHNICAL ASSISTANCE PROGRAMS 10.01 Technical assistance delivery agencies catering to SSI are virtually non-existent. To help the small scale manufacturing sector alleviate the problems currently impeding its full development, and in conjunction with the suggested financial support of SMI, a program of technical assistance is proposed aiming to reach primarily the target group (paras. 16.01, 16.02). The suggested programs would complement on-going or proposed related activi- ties in support of SMI. 1/ Efforts will be directed to: (a) raise the productivity of enterprises, by improving manufacturing processes and exten- sion of assistance to solve problems in areas of production management and engineering; (b) remedy deficiencies in the commercial aspects of management and facilitate project preparation (technico-economic aspects); and (c) to develop a Subcontracting Exchange, in an attempt to forge commercial linkages between SMI and large manufacturing enterprises (private and, hopefully, public). The suggested program is outlined below (para. 10.03 et seq.). 1/ It is essential that such non-financial services as the SIDO program and MIT's industrial estates operations be closely coordinated with the financial operations of the Halk Bank for maximum overall efficiency. The coordinating mechanism could be a formally constituted Working Committee with representatives from the relevant operating departments of both institutions, including a representative from the Ministry of Finance. Meetings should be held at least monthly, more often if neces- sary, to exchange information, deal with procedural problems, reach consensus on regional and sub-sectoral priorities, monitor progress reports, clarify responsibilties and eliminate duplication of effort and, generally, examine ways and means of improving cooperation and performance. Organizational and procedural details will be worked out during project preparation.  - 51 - 10.02 The relatively advanced stage of industrial development in Turkey dictates fundamental differences in the formulation of technical assistance programs compared with other developing countries. Thus, it is important to appreciate that there is no dearth of small entrepreneurs; the need is rather to redirect an increasing share of industrial growth into modern more efficient small-mediun Jize enterprises. Institutional assistance in identifying, evaluating, financing and implementing suitable projects should, therefore, be strongly oriented to promoting partnership, corporate or cooperative ownership. Similarly, improvements in project design, plant layout and equipment selection, factory operations, necessitate a more sophisticated level of technical assistance. Typically, extension service personnel should be able to assist in the solution of the more routine problems, but contracted consultants would be called on to assist in the design and production of products to rigid quality control standards such as special alloy castings, parts and sub- assemblies for the vehicle and other capital goods industries, electrical components (thermal relays, starters, switches, indicating instruments and controls), and other technologically more advanced product lines. 1/ Technical Assistance to the Halk Bank for Institution Building 10.03 The weaknesses of the Halk Bank and the need to devise and implement a time bound action program to redress these shortcomings to enable it to discharge more effectively its role in the development of SMI, were discussed in para. 4.09. The Bank is prepared to assist the Halk Bank to strengthen its project appraisal/supervision capabilities, internal operations, and overall management. Terms of reference for the proposed review of Halk Bank's organi- zation, systems and procedures are presented in Staff Appraisal Report (in preparation). Extension of Technical Assistance to Garment/Making-up Subsector 10.04 The technical assistance and training requirements of the garment industry are discussed in paras. 15.08 and 15.09. The pilot technical assis- tance program included in a recent Bank loan, with TSKB as a delivery agency, proposes to foster a local capability to meet some of these needs (para. 15.10 and Staff Appraisal Report No. 25256-TU, July 24, 1979). Extension of Technical Assistance to Small-Medium Scale Industry 10.05 At present, the Halk Bank provides rudimentary technical assistance to its clients on an ad hoc basis rather than as part of an organized program. Moreover, there is virtually no other agency to deliver such assistance. The proposed project would include a component for setting up a Small Business Advisory Services Group (SBASG) in the Halk Bank to provide a broad range of 1/ Factories producing these items were observed by the mission to be suffering from various problems related to product design, raw material quality, production control, finished product quality, cost accounting and marketing.  - 52 - services to its existing and prospective SMI clients as follows: (a) assist entrepreneurs in preparing projects for setting up new SMI or expanding existing ones, in choosing appropriate technologies, processes and equipment, and in promoting partnership, corporate or cooperative ownership; (b) assist small and medium enterprises to resolve technical, quality, production and management problems during the operational phase of their projects; (c) provide assistance in domestic and export marketing; (d) help SMI to enter into sub-contracting relationships with larger enterprises and to expand such links where they already exist or where they face problems. The Staff Appraisal Report (in preparation) contains detailed suggestions on the organization, location and staffing of the proposed units, training of staff, funding, as well as on other aspects of the SBASG program. Also, SYKB has agreed under the proposed Bank loan to extend technical assistance to its SMI borrowers on a cost sharing basis (80:20) with the beneficiary enterprises. Activities to be financed include: (a) consultants fees and other services to sub-borrowers for assisting them in solving technical, management and marketing problems relating to project preparation, implementation and operation; (ii) training of enter- prise staff both at the enterprise and at outside facilities, in specific areas relating to the operations of the enterprise; and (iii) training of SYKB's staff on subjects related to SYKB's objectives and operations. Technical Assistance for Development of a Subcontracting Exchange 1/ Potential 10.06 The Chamber of Commerce in Istanbul, in cooperation with the Chamber of Industry, has been involved in the promotion of sub-contracting. A special unit (YASTAB) was organized with UNIDO support in 1972 to assist in establish- ing both international and national business linkages leading to supply contracts, licenses, joint ventures and other commercial arrangements. The program has met with very limited success in part due to inadequate funding and staffing, and in part because it was lodged in the wrong place--no large manufacturing firm belonged to the Chamber of Commerce under whose auspices YASTAB was established. The program has been practically inoperative since 1976. 10.07 There is considerable potential for a properly organized and sup- ported Subcontracting Exchange that could actively promote linkages in direct consultation with production/procurement personnel of large (buyer) and smaller (supplier) firms and follow-up as needed to ensure successful con- tract negotiation. The fact that the Istanbul-Izmit-Bursa region is the most highly industrialized in Turkey with the highest concentration of metal- working industries and large engineering firms suggests that Istanbul should be the focal point of the proposed sub-contracting program. Similar units could be replicated in other major regions in the future. The Istanbul Chamber of Industry and the Union of Chambers of Commerce and Industry in Ankara, however, do not appear to share this view, in the belief that large 1/ The importance and prospects of sub-contracing in Turkey were discussed in para. 8.07 and references therein.  - 53 - firms are already successfully sub-contracting to smaller firms and that no assistance is needed. This is a narrow approach, since it does not take adequately into account the potential for further development of sub-contracting and, in particular, the need to build up the capability, capacity and contacts of the SMI. The possibility of promoting subcontracting arrangements in the public sector and at the international level also deserves serious considera- tion. Conceptual Framework 10.08 Essentially, the task is to identify those parts and components that can be efficiently produced in SMI enterprises and then establish contact between potential buyers and qualified suppliers leading to direct business negotiations. Successful performance by the institution acting as the sub- contracting "exchange" depends on the capabilities of the staff specialists who must gain familiarity with the practices of the larger factories that may be producing or importing parts that can be made competitively in SMI shops, as well as of the existing or attainable capabilities of numerous smaller enterprises. The major functions of a sub-contracting exchange therefore would be: (a) to develop and maintain an up-to-date information storage and retrieval system containing data on firms interested in sub-contracting, e.g. list of machinery, capacity, skills, record of performance, etc., for evaluat- ing the capability of firms for fulfilling orders with facility for quick access to specific data needs; and (b) to promote sub-contracting opportuni- ties between large and small firms by actively contacting large firms to encourage purchase of products, components and sub-assemblies from smaller firms on a sub-contracting basis; providing information to large firms on capabilities, competitiveness, delivery dates, etc. of potential suppliers; assisting large firms to purchase their requirements through competitive procurement practices; assisting small and medium size firms with techno- logical and management problems, and referring smaller firms intending to invest in equipment to financial intermediaries. Approach 10.09 TSKB, in cooperation with the Middle East Technical University in Ankara and TBTAK, is working on a data bank to record the available capa- cities and capabilities of SMI to perform different types of technical work. When the work is completed, probably late in 1980, a great deal of informa- tion will become available on the sub-contracting capacity and capability of the small and medium industry. TSKB, TBTAK, and the Middle East Technical University should then contact directly large industrial (private and public) firms as well as the Halk Bank to determine how this information can best (a) be used to expand the volume of sub-contracting; and (b) be regularly updated, in order to maintain its utility to potential users. The ultimate aim should be to draw up an Action Plan to organize a Subcontracting Exchange Unit, housed and funded appropriately. The Bank is prepared to support this effort. Technical assistance for training staff of the Exchange, if needed, could be obtained from agencies with relevant experience, such as the Irish Industrial Development Authority. Although the program initially could focus on the engineering industries, it should be expanded to include other indus- tries, e.g. furniture.  - 54 - Development of an Industrial Research Service 10.10 Extension of technical assistance needs to be complemented by an industrial research service. The practical problems raised by SMI entre- preneurs and the experience gained in the course of delivery of technical assistance provide an invaluable feedback and the right focus for applied research activities. To meet the evolving needs of SMI manufacturers, such research should encompass: (a) economic research, to assess market prospects in a wide range of products, costs, and location factors 1/; (b) engineering studies to screen known technology for adaptation to local resources, factor prices, scale factors, etc.; and (c) management studies, to integrate tech- nical and economic factors and to adapt accounting, marketing and distribu- tion as well as other related management techniques to the needs of the SMI. Within the applied research program, establishment of a technical informa- tion service is of paramount importance. The availability of an SMI indus- trial research service would serve not only SMI but also policy-makers. In view of the importance of developing such a facility, the Government should consider the possibility of creating an Industrial Research Institute, preferably as part of the SIDO program. The important point to stress is that such studies should have an aura of continuity, following up on develop- ments in a range of important manufacturing subsectors and ascertaining SMI needs. Assuming that SIDO will operate as an autonomous entity, the active participation of financial intermediaries (TSKB, SYKB, Halk Bank and, possibly, commercial banks), the Middle East Technical University, the Marmara Institute for science and Technology, appropriate Government units (e.g. SPO, MIT) and the end users of the facility must be ensured. The Institute should develop a core of competent staff, which should be complemented by local or foreign experts. To a considerable extent studies could be farmed out. Funding arrangements would have to be worked out, with the ultimate aim to make the Institute self-supporting. The mission suggests that a working group composed of representatives from SPO, Ministry of Industry, TSKB, Halk Bank, the private sector, or other groups that the Government would deem desirable, be set up to work out a plan of action. 11 Specifically, economic research should focus on competitive conditions and anticipated market growth in important small plant products, cost factors and cost-reducing methods, minimum economical size of plant, break-even projections, potential for development links between SMI and large enterprises, and an assessment of the industrial potential in particular industries, regions and locations.  - 55 - E. SURVEY OF SMALL-MEDIUM SCALE INDUSTRY IN PRINCIPAL SUBSECTORS 1/ XI. THE FOUNDRY INDUSTRY Overview of the Industry 11.01 The foundry industry in Turkey has made substantial progress in the past fifteen years as supplier of intermediate inputs to engineering industries, serving principally the domestic market. The industry has a comparative advan- tage in terms of both labor and transport costs and the domestic market for casting generally has not been overly protected. Imported castings are specialty items (auto engine blocks, machine tool components, etc.) which exceed the technical capability of local firms. The Government, however, has a policy of encouraging gradual substitution of imports in automobile and other engineering industries which has led to local manufacture of more complex castings. The imports of these complex products have been restricted to afford infant industry protection. 11.02 The increasing technological sophistication, product mix and output of the engineering industries 2/ has resulted in substantial growth of the number and size of foundries. But equally important has been the increase in their technological competence, enabling them to produce castings of greater precision, complexity and quality for a variety of end uses, utilizing a variety of metals. Major materials which account for the bulk of castings produced in Turkey are ferrous-based-grey iron, malleable iron, nodular iron (spheroidal) and steel. Non-ferrous metals, light or heavy, represent a smaller share of the total sector output. In 1976, of the total output of ferrous castings, grey iron accounted for 83%, malleable iron for 4% and steel for 13%. Organization 11.03 The foundry industry is predominantly in private hands, and has shown dynamism, versatility, and receptiveness and adaptability to growing technological requirements. Many of the now larger private foundries started 1/ The subsectors surveyed in this report are not the only ones where SMI has a significant potential. The food processing, plastic, rubber and paper products industries all have a high and growing population of SMI (Appendix Table 3.9) and deserve in depth review. Relative importance, staff and time constraints, and data availability in effect dictated subsector selection for review. 2/ Of the engineering industries, transport equipment (automotive vehicles, railway equipment), agricultural machinery and vehicles, construction equipment, pipes, fittings and valves, iron and steel industry equipment, industrial machinery, wood machinery, electrical equipment and consumer durables, are the major consumers of castings in Turkey.  - 56 - as small establishments 15 to 20 years ago and went through successive stages of expansion. Table 1 provides a breakdown of foundry smelting capacity for key product groups by plant size, ownership, and end use of output. The foundry industry comprises 440 firms, of which 400 are producing castings of grey iron, Table 1: NOMINAL CAPACITY OF TURKISH FOUNDRY INDUSTRY, 1975 /a ('000 tons) By Size of Firms By Ownership By End Use Medium /b /c Large Small Public Private Jobbing Captive Capac- Capac- Capac- Capac- Capac- Capac- No. ity No. ity No. ity No. ity No. ity No. ity Grey Iron 53 268.6 349 90.5 10 86.2 392 272.9 390 201.6 12 157.5 Steel 31 45.9 3 1.4 4 8.9 30 38.4 34 47.3 - - Malle- able 4 14.0 - - - 4 14.1 2 3.0 2 11.0 Total 88 328.5 352 91.9 14 95.1 426 325.4 426 251.9 14 168.5 /a Based on smelting capacity. 7W With capacity over 500 tons/year. 77 With capacity below 500 tons/year. Source: Tumas, Study of the Casting Industry, 1976, commissioned by TSKB. 4 of malleable iron, and 36 of steel. Of these firms, 88 (or 20% in number) are mostly medium (with a capacity of 500-5,000 tons/year) and some large (over 5,000 tons/year) in size, whereas the remaining 80% are small establish- ments (with capacity of 200-300 tons/year). Most foundries, and certainly all small ones, are of the jobbing type. 1/ The public sector accounts for about 23% of total smelting capacity and includes four large steel casting mills and ten large establishments producing castings of grey iron. 11.04 Large firms account for about 23% of existing capacity, medium for 55%, and small for 22%. The average capacity of small foundries is about 260 tons/year. Only three small foundries produce steel castings which require more sophisticated technology. There are no small foundries producing malle- able iron castings; but they produce a small quantity of steel castings. 1/ Foundries are distinguished between (a) captive, belonging to firms which utilize all foundry output in-house; and (b) Jobbing, supplying castings to other firms on the basis of contractual arrangements.  - 57 - Small foundries producing castings for machinery and transport equipment industries generally employ 20 or more workers and usually subcontract for large foundries. Small grey iron foundries, using simple equipment, produce a variety of standard final products such as weights, cooking vessels, sanitary pipes, agricultural implements, space heater parts, etc., which do not require dimensional accuracy or careful control of physical properties -f the metal. In 1974, the average price for grey iron castings of small foundries was TL 6.6/kg compared to TL 7.5/kg for simple castings of large foundries, reflect- ing the higher quality of the latter's products. Medium size foundries, all in the private sector, account for 75% of the grey iron casting capacity and almost all of steel capacity. The average size for large grey iron foundries is slightly over 5,000 tons/year which compares favorably with grey iron foundries in the U.K. (4,240 tons/year) and Germany (5,800 tons/year). Steel foundry capacity averages 1,500 tons/year. These figures suggest that the large foundries approximate internationally competitive size. Employment 11.05 Total emloyment in the foundry industry (1974) is estimated at about 12,400. Employment in smaller foundries (less than 500 tons/year capacity) is estimated at 3,650 employees or 30% of the total, at an. average of 10 workers per establishment. The larger foundries employed on average about 124 persons. Table 2 portrays the employment distribution in the industry by size of estab- lishment. About 80% of the foundries in a sample of 70 surveyed with capacity over 500 tons/year employ less than 200 workers, attesting to the preponderance of medium scale establishments. In terms of annual production capacity, the larger foundries employed 27 employees per 1,000 ton/year production capacity compared to 40 employees per 1,000 tons/year for small foundries. This re- flects the greater degree of mechanization and capital intensity of larger foundries. The skill and geographical distribution of employment in small foundries, based on a sample survey, is shown in Appendix Table 4.1. Two- thirds of those employed are skilled and highly skilled persons, indicating the relatively high skill-intensity of the industry. Major concentrations of small foundries are in such urban centers as Ankara, Istanbul, Izmir and Bursa. Table 2: EMPLOYMENT DISTRIBUTION BY SIZE OF ENTERPRISE /a Size of Establishment 10-50 50-100 100-200 200-500 500 and over Total No. of Foundries 17 19 20 11 3 70 % of Total 24.3 27.2 28.6 15.6 4.3 100 /a Expressed in number of workers employed. Source: Tumas, Study of the Casting Industry in Turkey.  - 58 - Wages 11.06 Average wage levels (TL/day) in the foundry industry during 1970-1975 are shown in Table 3. The largest increase during this period occurred in the Table 3: WAGE LEVELS Average Annual Increase % 1970 1973 1975 (1970-75) Foreman 85.0 127.9 151.7 18.6 Skilled labor 36.8 57.4 73.0 27.0 Unskilled labor 22.3 42.0 54.4 29.5 wages of unskilled labor. These wage levels are high compared to the average wage of TL 25/day in 1975 prevailing in the industrial sector. In terms of international competitiveness, average gross wages in Turkish foundries in 1974 compare favorably with industrial countries as Table 4 indicates. How- ever, the steep increase in wage levels in Turkish industry between 1975-1978 (30-35% per annum) as a result of social policy and increasing labor union Table 4: INTERNATIONAL COMPARISON OF WAGE LEVELS /a Country Gross Wage Index Rate of Wage Increase (1974-75) W. Germany 100.0 17.1 Italy 61.8 20.7 France 58.5 22.0 England 49.6 20.6 Spain 42.5 34.3 Turkey 30.0 29.5 /a Source: Association of European Foundries No. 5 (CAEF), reproduced in Tumas, Study of Casting Industry in Turkey. pressure may erode this advantage, particularly vis-a-vis some of the lower cost countries such as Spain. Performance 11.07 As a supplier of intermediate inputs to the engineering industry, growth of the foundry industry is conditioned by the growth of the output of the engineering sector itself, particularly those engineering products which use castings, and the rate of import substitution of imported castings, e.g. automobile engine blocks. Since the Government's policy has been to increase progressively the domestic value added of engineering industry products on a  - 59 - regular annual basis, 1/ the growth rate of foundry industries on average has been higher than that of the engineering sector. During 1975-78, total foundry capacity increased by about 128,000 tons or by 29%. About 60% of this increase is estimated to have occurred from the establishment of 14 new foundries and the balance from the expansion of 24 firms. Table 5 provides a breakdown of the increas_ in foundry capacity during this period by product categories. Grey iron capacity increased by about 100,000 tons, or 8.5% per annum, while steel castings capacity by some 25,000 tons, or about 15% annually. Expansion in malleable iron has been negligible. This growth in capacity reflects the underlying pattern of the demand for castings. Table 5: GROWTH OF NOMINAL CAPACITY OF FOUNDRIES, 1975-78 /a ('000 tons) /b _c Type of Castings 1975 1978 Annual Growth (%) Grey Iron 359.1 457.5 8.5 Steel 47.3 71.6 14.8 Malleable Iron 14.0 14.5 1.2 Total 420.4 543.6 9.0 /a Based on smelting capacity. lb Estimates of Tumas Study. Ic 1979-83 Plan estimates. 11.08 Output figures of foundry industries for selected years are given in Table 6. The overall real growth in output of foundries between Table 6: PRODUCTION OF GREY IRON AND STEEL CASTINGS Grey Iron Steel Total Tons Value Tons Value Tons Value Year (000) (TL million) (000) (TL million) (000) (TL million) 1972 219 1,752 23 345 242 1,994 1974 184.8 - 27.3 - 212.2 - 1977 348 2,784 63 945 441.0 3,729 1977 Ia 349 2,792 50 750 399.0 3,542 /a Third Plan targets. 1/ For example, in the case of passenger cars, every year imports of certain components are prohibited based on local availability. Currently, imports account for 22% of the ex-factory value of a car.  1972-77 has met or surpassed the Third Plan targets. The average annual output growth for steel castings was 22% and for grey iron castings about 10%. 11.09 Table 7 gives the structure of production of ferrous castings by plant size, end use of output, and ownership, a well as the capacity-utilization in 1974. Capacity utilization of grey iron foundries in 1974 was relatively low, both in the private and public sectors, due to.the deterioration of the Table 7: PRODUCTION AND CAPACITY UTILIZATION By Plant Size By End Use By Ownership Medium/ LarRe Small Jobbing Captive Public Private Production (000 tons) Grey Iron 131.0 44.7 88.1 92.6 43.6 132.1 Steel 26.8 0.5 27.3 - 4.8 22.5 Malleable 9.1 - 0.4 8.6 - 9.1 Total 166.9 45.2 115.9 101.2 48.4 163.7 Capacity Utilization (%) Grey Iron 61 61 55 73 63 60 Steel 65 42 64 - 60 65 Malleable 72 - 72 - - 75 economic conditions in the country and the consequent lower growth and capacity utilization in the engineering industries. The situation has further worsened in recent years. As expected, capacity utilization rates in captive foundries are higher (73%) than in jobbing foundries (55%). In the grey iron area, both large and small foundries show similar capacity utilization rates. Exports 11.10 It is difficult to asses the export performance of the industry as castings are intermediate inputs to engineering products. Exports of castings (pipes, handtools, machinery parts) amounted to $1.5 million in 1973, $2.05 million in 1974 and $2.25 million in 1977. Steel castings usually comprise one-third of the total value of exported castings. Exports in 1978 are projected at a low level because of reduced output, stemming from shortages of raw materials and power. In addition to exports of castings directly, there have been considerable indirect exports of castings in the form of machinery. The Plan forecasts exports of castings to increase to $20 million by 1983. This is an optimistic though attainable target, provided the industry can gear  - 61 - up and modernize. Major markets for castings so far have been the Middle East and some European countries (West Germany, Italy). The EEC holds considerable potential as a direct and indirect export market for Turkish castings. Problem Areas 1/ Small Foundries 11.11 A major problem which small foundries face is the lack of knowledge of foundry technology 2/, particularly as it relates to modern techniques for producing quality castings. Furthermore, the situation is exacerbated because process equipment is antiquated and operating procedures are not conducive to production of high quality castings. Sand processing equipment, core manufac- turing facilities, and in-house pattern shops are unavailable. Small foundries use cupola smelting, though three produce steel castings and have induction furnaces. Since almost all small foundries are jobbing foundries producing a variety of castings of different sizes and weights, there is often a mismatch between the available equipment (cupola, moulding shop facilities), and the product mix (size and weight of castings), leading to poor machinery utiliza- tion and low productivity. Small foundries do not have laboratory facilities for quality control of raw material inputs (coke, pig iron, steel scrap, moulding sand, core sand, resin, etc.), process control (checking carbon and silicon levels of molten metal), and quality control of finished castings (Brinell hardness, transverse testing for measuring breaking load and defec- tion, metallographic testing and microstructure). In addition, there are problems with raw material supply. Domestic coke is considered unsuitable for cupola operations. Electric power is in short supply so that switching to induction furnaces may not provide a solution. Synthetic moulding sand which is required to produce high quality castings is also in short supply because of lack of sand processing equipment in most foundries. Domestic prices of some inputs are also higher than import prices (e.g. about 40% higher for steel scrap and 15% for pig iron in 1975). Shortage of spares due to unavailability of foreign exchange has also become a problem. 11.12 A survey of 158 small foundries enumerated only 18 engineers in the work force (Appendix Table 4.1). Shortage of technically experienced production workers occur particularly in pattern making and moulding operations. Small foundries suffer from the same types of management deficiencies as are observed in other small industries. Lack of suitable accounting system for costing and financial control and marketing problems are important deficiencies. Production planning and inventory control are difficult in jobbing foundries, 1/ For a summary see Annex G. 2/ There are of course exceptions. The mission visited a small foundry owned by the head of the foundry dernik in Ankara who, without having formal training, has written a basic textbook on foundry technology and has prepared the entire technical design for a large foundry he is constructing near Ankara. Incidentally this is another example of internal growth and transition from small to medium scale plant.  _62 - particularly in uncertain economic conditions, and need considerable improve- ment. Access to term finance is also limited. Medium and Large Foundries 11.13 The technology employed by large foundries is summarized in Appendix Table 4.2. It is common for foundries to employ more than a single technology/ machine type in any process 1/, as well as different technologies 2/. The Tumas survey shows that, in general, the technology in use is relatively labor-intensive, and modern technology is confined to a small percentage of foundries 3/. For example, the use of automated sand processing equipment is available ln 35% of plants, centrifugal casting in 7%, continuous casting in 2%, resin process for core manufacture in 41% and non-destructive testing in 12%. Only three firms had formal licensing agreements with international firms. While it may be necessary to import technology, particularly for precision casting and improved metallurgical control, there is need for greater diffusion of modern technology already available in Turkey among foundries to upgrade the general level of technology, by fostering linkages through subcontracting, creation of an extension service center for foundry technology, etc. 11.14 Larger foundries have raw material problems similar to those of small foundries which include poor quality of coke for use in cupola furnaces and shortage of synthetic moulding sand and electric power. The larger jobbing foundries have problems of matching their production facilities (type of equip- ment, technology) with the product mix required by the market, which results in inefficiencies. The present weighted average defect levels is 17.5% and 9.5% for grey iron and steel castings, respectively, which is high. The auto- motive and machinery industries--the major users--have yet to set stringent quality standards (dimensional tolerance, metallurgical standards) and cast- ings of uneven quality are accepted. The trend of higher quality require- ments of both domestic and export purchasers suggests that the quality of casting needs to be systematically upgraded by adopting modern technology throughout the industry. 1/ A moulding shop may employ hand moulding on the floor for large castings and several different moulding machines for different casting sizes and production volumes (jolt-squeeze pin-lift moulding machine for moulds over 750 mm, sand slingers for medium size moulds at high production levels, etc.). 2! E.g. sodium silicate and organic binders for non-bake moulding. 3/ However, laboratory facilities are widely available. About 76% of large foundries have chemical laboratories, 70% have sand testing facilities, 64% mechanical testing laboratories and 12% facilities for non-destructive testing.  - 63 - Prospects 11.15 Forecasts of demand and required capacity for various types of foundry products are shown in Table 8. The demand for all types of foundry products grew at 17% per annum during 1974-78. However, the growth of demand has been arrested due to the adverse economic conditions in 1979 and 1980. Thus, the demand for grey iron products is now projected to grow only at 5% during the 1978-83 period. With forthcoming capacities to the tune of 40,000 tons, it should be possible to satisfy this demand, including an estimated 10,000 tons for exports, without further investments until 1983. Supply can also be increased by improving levels of utilization, particularly by the larger foundries. Demand for steel and mallable iron castings is estimated to increase by about 4% and 9% respectively during the same period. Similarly, existing capacities appear sufficient to meet such demand requirements in the next few years. Table 8: FORECASTS OF DEMAND Effective /a Foundry Demand Capacity Difference Products Year ('000 tons) ('000 tons) ('000 tons) Grey Iron / 1977 285 360 +75 1978 340 370 +30 1979 320 370 +50 1980 330 370 +40 1981 385 390 + 5 1983 430 410 -10 /c Steel- 1977 40 45 +5 1978 44 63 +19 1979 42 63 +21 1980 42 63 +21 1981 48 63 +15 1983 53 63 +10 Malleable Iro n 1977 14 13 -1 1978 16 14 -2 1979 15 14 -1 1980 15 14 -1 1981 18 14 -4 1983 24 14 -10 /a TSKB and mission projections. The capacity figures in the Table represent adjusted figures in which nominal smelting capacity has been reduced by 20% for iron castings and 10% for steel castings to account for casting defects and work stoppages. /b Projected demand growth rate 5% per annum. /c Projected demand growth rate 4% per annum. /d Projected demand growth rate 9% per annum.  - 64 - The structure of demand for castings (Appendix Table 4.3) indicates that the major increase would be for sophisticated end use products such as automotive equipment, construction machinery, iron and steel works and agricultural machinery which require complex technology only available in larger foundries. Small foundries which are sub-contractors to large foundries and have acquired or are able to acquire the requisite technological capabilities are likely to expand to meet this demand. But the majority of small foundries are not likely to grow without some degree of integration or reorganization and technical assistance. Strategy for Subsector Development 11.16 To survive and grow, small and medium foundries need to improve significantly their technological capabilities and productivity. Improved technological facilities would be essential for the manufacture of more com- plex castings required for sophisticated end use products for which demand is projected to grow rapidly, while improved production management, includ- ing rationalization of production, is essential to improving efficiency. An approach for improving technological capabilities would be for the Government to provide to a group of small foundries clustered in close proximity a common service facility which could include laboratory equipment for chemical, mechanical and metallographic analysis, 1/ a sand recycling facility, pattern shop for both wood and metal patterns, core manufacturing shop using modern technology and, possibly, a machine shop. However, the possibility of obtaining such services from large foundries on a contractual basis should be investi- gated before establishing common service facilities (see para. 7.13). Invest- ment in such common service facilities could also be undertaken by the small firms themselves on a cooperative basis, provided the mutual benefits are brought to their attention. Financial intermediaries can organize the setting up of such common facilities. The Government has provided incentives (custom duty deferals, medium term loans, etc.) to encourage small firms to establish common facilities but so far the response has been limited (see para. 8.02). Technical assistance to small foundries to help them establish such services would also be necessary. 11.17 An alternative approach to establishing common service facilities would be for several small firms clustered in a location to integrate and to establish a medium size jobbing foundry designed to produce more complex prod- ucts on a rationalized production system. Since more than one type of tech- nology can coexist in an individual foundry, initially some existing equip- ment can be utilized efficiently. These new foundries would utilize modern machinery to produce castings of the required quality and complexity but would still be within the medium size range. Financial intermediaries can also pro- mote the horizontal integration of small foundries through their business advisory services. 1/ For a list of equipment for sand, physical and metallurgical labora- tories, see UNIDO, Guidelines for Establishing a Demonstration Foundry in a Developing Country, Vienna, 1976, Annex III.  - 65 - 11.18 Aside from the growing domestic market for iron and steel castings, there is considerable potential for exports of both rough and machined cast- ings, provided technological and raw material problems of the industry are resolved. The motor vehicle, machine tools, construction machinery, materials handling equipment and heavy electrical machinery industries are potential sources of demand for export castings. The medium scale foundries are poten- tial sources of demand for export castings. The medium scale foundries have the largest potential for exports, assuming they develop the technical and management capabilities to produce complex and high quality castings on a consistent basis. The way to penetrate foreign markets would be through development of international sub-contracting linkages with European and Middle-Eastern firms. Small foundries, properly reorganized and assisted, could in turn cater to the needs of an expanded domestic (and indirectly export) market for a significantly larger range of products. A list of typical items suggested for manufacture by small foundries is given in Annex H. Concluding Remarks 11.19 Efficient production of standardized, precision, high quality castings requires use of basic machinery which cannot be economically operated by existing small foundries. The domestic market for more complex and higher quality-castings is expanding and smaller foundries will have to satisfy the product specifications of the future market in order to survive and to expand. The technological capabilities of smaller foundries could be enhanced either by providing use of sophisticated machinery through a common service center or by fostering integration of smaller foundries. Subcontracting arrangements on a more extensive basis beween large and smaller foundries would assist the latter to improve their efficiency and the quality of their output. Small foundries may be induced to make the transition by appropriate incentives to foster mergers, use of common facilities and subcontracting. Technical extension services to promote all these forms of reorganization are also vital. Financial intermediaries can play an important role in promoting the establishment of common facilities through cooperative arrangements as well as horizontal integration. Larger foundries, though operating at an efficient scale, are limited in their use of modern foundry technology. There is a need for greater adoption of modern technology and production management methods to improve quality and productivity.  - 66 - XII. THE ENGINEERING INDUSTRIES Role of Private Sector 12.01 Reflecting Turkey's relatively advanced stage of industrial develop- ment, the engineering industries 1/ as a group is one of the most important manufacturing sectors in terms of investment, employment and value added, with shares amounting to 18%, 28% and 25% respectively in 1976. The private sector (organized) dominates the engineering industries with 88% of production, 81% of investment, and 75% of employment, as shown in Table 1. If the unorganized sector is included, the importance of the private sector is even greater (see para. 12.03). Only in the transport equipment industry is the public sector Table 1: RELATIVE SHARES OF PRIVATE AND PUBLIC SECTORS IN ENGINEERING INDUSTRIES - 1975 Share of Share of Share of Production (%) Employment (%) Investment (%) Industry Public Private Public Private Public Private Fabricated Metal Products 3 97 3 97 19 81 Non-electrical Machinery 26 74 29 71 10 90 Electrical Machinery 2 98 6 94 4 96 Transport Equipment 14 86 45 55 36 64 Scientific Equipment - 100 - 100 - 100 Average 12 88 25 75 19 81 Source: Turkish Industrialists' and Businessmen's Association, The Turkish Economy, 1978, and Appendix Tables 3.5, 3.6. relatively important, employing 45% of the work force but accounting for only 14% of output. However, eight private enterprises which are exclusively engaged in automobile production are among the first fifty of the largest industrial firms in Turkey, with aggregate sales of some TL 12 billion and 1/ They include the following five product lines: fabricated metal products (ISIC 381), non-electrical machinery (ISIC 382), electrical machinery (ISIC 383), transport equipment (ISIC 384), and professional and scien- tific equipment (ISIC 385).  - 67 - employment of 13,500 workers in 1976. SMI engages in the fabrication of a great variety of sheet metal products, castings and forgings, engineering supplies, machine products, replacement parts, welded products, and in minor repair operations. Structural Trends 12.02 The relative importance of the organized and unorganized segments of the engineering industries--as well as of the respective production lines--in terms of number of establishments, value added and employment is evidenced in Table 2. The unorganized sector, with 98.3% of total establishments, employs Table 2: DISTRIBUTION OF SIZE, EMPLOYMENT AND VALUE ADDED IN ENGINEERING INDUSTRIES - 1970 /a /b Organized Sector Unorganized Sector Number Value /c Employ- Number Value /c Employ- Industry of Firms Added ment of Firms Added ment Fabricated Metal Products 361 1,240.2 33,712 26,122 549.1 49,602 Non-electrical Machinery 215 1,148.8 20,656 4,115 112.3 9,819 Electrical Machinery 132 390.7 9,960 4,384 108.2 7,528 Transport Equipment 151 880.8 33,047 14,986 332.0 30,051 Total 859 3,659.8 97,375 49,607 1,101.6 97,000 - =r =D = - _ ,... / a Establishments employing 10 or more workers. 7W Establishments employing 1-9 workers. 77 In TL million. Source: Appendix Table 3.12. one-half of the engineering sector's work force and produces about 23% of the total value added. The average employment in the organized sector is 114 com- pared with 2 in the unorganized. This suggests that a large part of the organized sector comprises small-medium size firms (i.e. establishments employ- ing up to 200 workers), which underlines the importance of the SMI group. In the unorganized sector, 84% of the number of establishments and 82% of employ- ment are in metal fabricated products and transport equipment, while in the organized the respective figures are 60% and 69%, with a significant share (25%) in non-electrical machinery. 12.03 Current statistics on the structure of the engineering industries in the unorganized sector are not available, and this hampers the analysis of structural trends for this size group. Table 3 shows developments in the  - 68 - number of establishments and employment during 1963-70. The period is char- acterized by a small increase in the number of establishments (15%) but by a much greater increase in employment (47%), primarily due to the fast growth of the transport equipment industry. If past trends have persisted as is likely, then both the number of firms and total employment have probably increased during the 1970's. Table 3: ESTABLISHMENTS AND EMPLOYMENT IN THE UNORGANIZED ENGINEERING SECTOR, 1963, 1970 /a 1963 1970 Number Number Branch of Firms Employment of Firms Employment Fabricated Metal Products 29,416 40,423 26,124 49,600 Non-electrical Machinery 1,310 1,700 4,115 9,820 Electrical Machinery 3,162 4,894 4,384 7,530 Transport Equipment 9,271 18,814 14,986 30,050 Total 43,159 65,831 49,609 97,000 /a Employing 1-9 including owners, partners and unpaid family workers. Source: Appendix Table 3.11 and Census of Manufacturing and Business, 1963. 12.04 The structural trends of the sector during 1970-77 are shown in Table 4. The number of firms in all industries of the sector taken together Table 4: STRUCTURAL TRENDS IN ORGANIZED ENGINEERING INDUSTRY, 1970-77 1970 1977 Change 1970-77 Number of Firms 859 2,488 190% Employment 97,375 200,430 106% Source: Appendix Tables 3.9, 3.10, 3.12. tripled between 1970 and 1977 while employment doubled, reflecting the rapid growth of the sector during this period. Table 5 shows the size distribution of firms and employment in the organized engineering sector in 1977. It is noteworthy that 76% of all establishments fall in the 10-49 size bracket and only 5.5% in the 200+. In terms of employment, however, the respective figures are 20% and 60%. Table 6 provides insights in the structural developments of employment between 1970 and 1977. Employment in both the SMI (10-199) and large scale industry (200+) increased at the same rate and thus retained their shares, 40% and 60%, respectively. The observed pattern of development in size distribution of establishments and employment suggests a high rate of growth of individual enterprises and resultant shifts between size brackets  - 69 - Table 5: SIZE DISTRIBUTION OF FIRMS AND EMPLOYMENT IN THE ORGANIZED ENGINEERING INDUSTRIES - 1977 /a Number of workers 10-49 50-99 100-199 200+ Total Number of Firms 1,891 305 154 138 2,488 Percent 76.0 12.3 6.2 5.5 100.0 Employment 39,791 20,250 20,983 119,406 200,430 Percent 19.9 10.1 10.4 59.6 100.0 /a Data on the number of firms are not compatible with those of the State Institute of Statistics. The latter's figure for the number of engineer- ing firms with 10 or more workers is 1,356 in 1976 and appears to seriously underestimate the number of firms. Source: Appendix Tables 3.9, 3.10. throughout the spectrum, a hypothesis strongly supported by interviews con- ducted by the mission, and new entry. Growth through merger appears to have been less prevalent. It also demonstrates the persistent importance of the SMI segment in the development of the engineering sector. Table 6: EMPLOYMENT STRUCTURE IN ORGANIZED ENGINEERING SECTOR - 1970, 1977 Change 1970 1977 1970-77 10-199 39,617 81,024 105% 200+ 57,758 119,406 107% Total 97,375 200,430 106% Source: Apendix Table 3.10 and Table 6.4 of World Bank Report No. 1359-TU, 1977. Value Added and Productivity 12.05 Trends in real value added and productivity -n the engineering industries during 1965-1975 are shown in Appendix Tables 3.3, 3.4, while annual average growth rates in Table 7. Public sector electrical and non-electrical machinery industries experienced high rates of growth, whereas fabricated metal products a fairly large negative growth rate. In the private sector, transport equipment and non-electrical machinery industries have shown strong growth, whereas fabricated metal products and electrical machinery a modest one. The growth was fueled by strong demand, protected environment and, to some extent, incentives.  - 70 - Labor productivity growth, with the exception of non-electrical machinery, has been consistently higher in the private sector. In absolute terms, labor pro- ductivity in the engineering industries has been substantially higher in the private compared to the public sector (Appendix Tables 3.17, 3.18)--by three and four times in the electrical machinery and transport equipment industries, respectively. This reflects superior managerial capability in the private sector and the severe problems, including overstaffing, that plague public sector enterprises. Table 7: ANNUAL AVERAGE GROWTH RATE IN VALUE ADDED AND LABOR PRODUCTIVITY, 1965-1975 (Percentage) Value Added Labor Productivity Industrial Branch Public Private Public Private Fabricated Products -15.1 15.6 2.0/a 8.0 Non-electrical Machinery 28.5 19.7 6.7 5.3 Electrical Machinery 30.8 12.0 4.1 9.5 Transport Equipment 11.4 32.2 7.5 9.8 /a Average 1972-74 since the figure in 1975 is unreliable. Source: Appendix Tables 3.3, 3.4, 3.17, 3.18. 12.06 The change in the respective shares of individual branches in total value added in the engineering sector during 1965-75 is indicated in Table 8. During this period, there has been a dramatic decline in the relative contri- bution of fabricated metal products to value added in the public sector, accompanied by an equally dramatic increase in the share of non-electrical machinery. There has been no perceptible change in the share of the transport equipment industry. Small gains in electrical machinery are also observed. This structural change in the public sector appears to reflect the emphasis of public investment and the fact that, in Turkey's present level of industrial development, fabricated metal products is an activity better performed in the private domain. In the private sector, a small decline in the shares of electrical and non-electrical industries and a sharper drop in the fabricated metal products industry is noted. On the other hand, there has been a substan- tial rise in the share of transport equipment. Several reasons account for the observed structural changes. In the first place, a large part of the out- put in the fabricated metal products industry is produced by a large number of small firms which tend to have lower value added. Fabricated metal product industries generally have a lower value added per unit of output compared to other engineering industries. Furthermore, the technology employed in the manufacture of fabricated metal products (e.g. sheet metal products, pressed steel components, welded items of sheet plate such as containers, pressure vessels, etc.) is relatively simple and hence involves relative low capital investment. By contrast, the manufacture of non-electrical machinery, elec- trical machinery and transport equipment involve more sophisticated and  - 71 - relatively more capital-intensive technology (sophisticated designs, use of special alloy steel and non-ferrous alloys, use of precision metal working machinery to achieve close tolerances, and in the case of mass production items, such as automobiles, highly automated transfer machines which provide economies of scale). These features, coupled with the much faster growth and investment outlays, particularly in the transport equipment industry, (see Appendix Tables 3.5, 3.6, 3.12, 3.13), explain the large decline in the relative share of fabricated metal product industry. More importantly, the structural shifts observed suggest a trend to introduce more sophisticated technology in the engineering sector and to produce higher value added products. This is an important sectoral development which deserves to be further en- couraged. Table 8: STRUCTURAL TRENDS OF VALUE ADDED SHARES IN ENGINEERING INDUSTRIES (Percentages) 1965 1975 Industry Private Public Private Public Fabricated Metal Products 40.4 45.2 21.0 4.0 Non-electrical Machinery 23.0 9.2 19.5 46.9 Electrical Machinery 26.3 0.4 24.5 3.0 Transport Equipment 10.3 45.2 35.0 46.1 100.0 100.0 100.0 100.0 Source: Appendix Tables 3.3, 3.4. Investment 12.07 Real investment in the engineering industries has increased at an average annual rate of 15.5% in the private sector and 14.5% in the public sector during 1965-75 (Appendix Tables 3.5, 3.6). The organized private sector's share in total investment in the engineering industries has been maintained at about 80% during this decade. Although real investment growth rates in individual branches of the engineering industries have fluctuated from year to year, the overall pattern is one of strong growth. The distribu- tion of investment by branch of industry for 1975 is shown in Table 9. The bulk of public sector investment has been in transport equipment (railroad locomotives, aircraft repair, shipbuilding), whereas the emphasis of the private sector has been on the capital goods industries (electrical and non-electrical machinery) and transport equipment.  - 72 - Table 9: DISTRIBUTION OF INVESTMENT IN ENGINEERING SECTOR - 1975 (Percentages) Public Private Fabricated Metal Products 12.4 12.4 Non-electrical Machinery 14.8 30.9 Electrical Machinery 4.3 28.2 Transport Equipment 68.5 29.5 100.0 100.0 Capital Intensity and Capital Productivity 12.08 Capital/labor ratios for public and private enterprises are shown in Appendix Tables 3.15, 3.16. Capital intensity has been rising in all segments of the engineering industry, and more so in the private sector. 1/ In the latter, the capital/labor ratio is highest in the transport equipment industry, because of the need to realize economies of scale, and in the non-electrical machinery, because of the need for large capital investments in manufacturing machine tools and similar precision machinery. In the public sector, capital intensity is generally lower compared to the private sector, except in fabri- cated metal products. This may reflect sheer overstaffing. On the other hand, the higher capital intensity observed in fabricated metal products appears to arise from the type of products manufactured. In public enter- prises, they include heavy structural steel and plate products, water tube boilers and process plant which require large investment. By contrast, private firms produce simple items of sheet steel which require simpler presses and welding equipment and thus are relatively less capital intensive. It is noteworthy that in spite of the higher levels of capital investment in private sector engineering industries, capital productivity (inverse of capital/value added ratio) is much higher in private enterprises - except for non-electrical machinery industries (Appendix Tables 3.19, 3.20). This anomaly probably arises either from the poor quality of the data or from the nature of products manufactured rather than intrinsic differences in efficiency. Exports 12.09 Exports of engineering industry products totalled US$26 million in 1977, up from US$2 million in 1970. Non-electrical machinery and transport equipment (including components) accounted for about three quarters of total exports of engineering products. This is a little over 4% of total industrial exports, and a modest figure for Turkey's level of industrialization, but still is indicative of future potential. 1/ The figures for metal products and non-electrical machinery in the public sector are not meaningful.  - 73 - Technological Considerations 12.10 The technological capability of engineering firms varies with the size of establishment and products manufactured. The trend has been to introduce more sophisticated technology and to produce higher value added products. The larger firms in transportation, electrical and non-electrical machinery and in consumer durable industries are generally equipped with imported modern machinery, and employ production engineers who have the skills and training to operate the equipment efficiently. However, because of limited local research and development, there is little innovation in terms of adapting and improving production processes (tooling, adaptation of product design and machinery, production planning and workshop layout, etc.) to local conditions. In the smaller firms, much of the equipment is home-made or second hand. Such equipment tends to be old or obsolete with limited capabili- ty to produce at required tolerances and rated capacity. Production processes often are wasteful of raw material and fail to produce goods of consistent quality. Quality control standards and procedures are inadequate. There is generally a lack of modern measuring tools and procedures. However, the overall picture which emerges is that of an industrial sector based on a well- diffused technological capability in both the SMI and large scale segments, with excellent prospects for growth and improved efficiency. The Automotive and Consumer Durable Industries 12.11 Before discussing sectoral problems and needed corrective measures, including possibilities of enhancing the role of the SMI sectors, in the fol- lowing paragraphs we will review in some detail the automotive and refrigerator subsectors, which are suggestive of the potential they afford for the develop- ment of ancillary and sub-contracting relationships between SMI and large firms. The automotive industry has been selected for more extensive discussion because it was the only engineering category in which the mission was able to check available data on structure, performance and prospects through field visits, interviews and personal observation. In many aspects, it appears to be typical of the sector as a whole. The refrigerator subsector is repre- sentative of the important durable consumer goods category. The Automotive Industry Industry Profile 12.12 The development of the Turkish road transport industry is a recent phenomenon. In 1955, two plants (TOE and Turk Traktor) began producing trucks and tractors under foreign license. In the following ten years, nine more companies began to manufacture trucks, tractors, cars, buses and mini- buses, vans and motorcycles. Between 1965-70, import restrictions were imposed and domestic production expanded rapidly; in this period, seven more companies began operations and another five between 1971-77 for a total of 23. Efforts to increase the domestic content of the products were intensified and the industry progressively converted to local manufacture as opposed to simple assembly of imported components. Output grew remarkably between 1970-77, with  - 74 - production of vehicles of all types increasing from 23,857 units in 1970 to 136,701 units in 1977. Installed capacity is estimated at about 180,000 units per annum, implying an industry-wide capacity utilization factor of 75%. 12.13 Shortages of foreign exchange in the 1970's pushed forward import- substitution. It is estimated that currently 70-80% of automotive product value is of domestic origin. Direct employment in large public sector assembly factories totalled 23,000 in 1975 (Appendix Table 3.24), in the private organized sector 32,000 in 1976 (Appendix Table 3.13), and total employment in the sector may be as high as 100,000 if we include in the estimate the thousands of small shops in the SSI sector. 1/ A study of a sample of 900 firms in the private organized sector shows that over 75% are machine shops employing 10-50 workers, 20% employ 51-250 and only 5% are relatively large, partially integrated producers of finished vehicles employing more than 250. The industry is highly concentrated regionally, with about 75% of the production facilities located in and near Istanbul and the remainder in Izmir, Bursa and one or two other cities. Production is fragmented among six types of cars, 16 of minibuses, 20 of trucks and vans, 4 of buses and 21 of tractors. 2/ SMI Potential 12.14 The Otosan factory in Istanbul -- reputed to be the largest privately- owned company in Turkey -- produces cars, trucks and mini-buses. Its produc- tion capacity was stated to be 42 cars, 35 trucks and 24 mini-buses per day or, assuming 250 working days per year, about 25,000 per year on a single shift basis. In 1976, the plant actually produced 22,000 units but, at the time of the mission visit in December 1978, production was running at an annual rate of 12,000-15,000 units, implying that only 60% of capacity was being utilized. The work force had also dropped 23% from 3,000 in 1977 to 2,284 of which 600 were on the assembly line and the balance (1,684) were engaged in manufacture of components. The number of outside suppliers totalled 210, down from 360 a few years earlier. The attrition was explained as resulting from growth and merging of small shops into larger firms thereby permitting Otosan to place higher subcontracting orders with fewer firms. Because of the efficiencies attainable in larger enterprises, quality and delivery had improved. Otosan exports cars and components to nearby Middle East countries but the amounts are inconsequential. 3/ 1/ It is impossible to fully reconcile statistics from various sources. The unorganized sector, in particular, is amorphous and includes large numbers of repair shops. Many of them also manufacture components for the replacement market (Appendix Table 3.11). 2/ Depending on the definition of "type" or "model", some estimates run as high as 40 for cars, 25 for buses and 30 for trucks and vans. 3/ For the entire Turkish auto industry, exports were 75 cars in 1974, 415 in 1975 and 153 in 1976.  - 75 - 12.15 Otosan management provided the following breakdown between imported and domestic source procurement: SOURCES OF PARTS AND COMPONENTS - OTOSAN Domestic Imported (%) (%) Passenger Cars 78 22 Buses 70 30 Trucks 65 35 The above figures are probably typical of the industry as a whole. Major components that are still being imported include engine blocks, fuel system, crank shafts, transmissions and the steering box. Under the "domestic" heading, Otosan produces the complete body and chassis (including fiberglass shells), electrical harnesses, hubs and flywheels in its own shops. All other parts and components are purchased from SMI suppliers. Taking into account the more or less 30% imported component, there appears to be ample scope for expansion of domestic procurement, a large portion of which could be channelled to outside SMI suppliers who are able to handle short-run production more efficiently than large factories such as Otosan. The Refrigerator Industry 1/ Industry Profile 12.16 The Turkish refrigerator industry began in 1958 and experienced rapid expansion after 1962. By 1976, annual production was 570,000 units (compared to 200,000 units in 1972) manufactured by 6 firms of which the two largest accounted for 85% of total output. 2/ Growth since 1976 appears to have slackened -- although still healthy at 10-15% per annum -- partly because of slower growth in demand but also because of problems affecting the availability of both domestic and imported raw and intermediate materials (steel sheet, aluminum, chemical refrigerants) and components manufactured by small and medium-scale subcontractors. Despite the supply constraints, the 6 large producers of finished refrigerators are operating at about 90% of capacity and expanding their facilities in step with demand. The price 1/ Some information is available on the durable consumer goods sector com- prising refrigerators, washing machines, vacuum cleaners, radios and television sets, among others, and generally classified under electrical machinery. This brief review of the refrigerator industry provides in- sights that may be generally applicable to the entire sector. 2/ This suggests a very high concentration of the industry which, at this stage of development, is acceptable and even desirable.  - 76 - structure also appears to be favorable, reflecting the balanced supply/demand pattern of the market and improved production efficiency. Between 1970 and 1976, the manufacturer's price index for refrigerators increased only 50%, while the wholesale price index by 173%. By 1982, sectoral capacity according to present plans is expected to be about 1,300,000 units annually, or more than double the 1976 level. Domestic demand in 1982 is projected at about 1,000,000 units and, if this conservative estimate proves to be correct, the industry will be faced with either the need to cut back on expansion plans or to seek wider export markers, or both. 12.17 Exports are small, ranging between 2.5% to 6.0% of total production and from 10,000 to 22,000 units a year in absolute figures, mainly to Middle East countries. Exports of 22,000 units in 1976 earned a total of US$3.9 million in foreign exchange. Although exports are still a relatively modest activity, the fact that Turkey is able to export at all in the highly competi- tive international market is significant and augurs well for the future. As the domestic industry grows and achieves economies of scale -- as well as improved quality and design -- it should be able to capture larger shares of the international market. In view of the planned expansion of productive capacity, a major export promotion campaign is needed and should be instituted immediately. Government support through priority licensing of needed imports, fiscal incentives, export credits, promotional assistance, rationalization of freight rates, etc. would be an important factor in the success of the export development program. Concessional trade agreements with the Arab bloc -- i.e., Turkey's natural market area -- should also be negotiated. An export target of 100,000 units by 1982 is not unreasonable if the measures suggested above are adopted. 12.18 On the whole, the industry exhibits a favorable growth rate based on the expanding market and continually improving productive efficiency. Further gains in efficiency, cost-competitiveness and domestic added value will depend on replacing imports of sheet steel, plastics, enamel and components such as thermostats, themorelays and compressors with domestic source substitutes of equal quality and price. As the 6 manufacturers expand production from the 1976 average of 100,000 more or less to the projected 200,000 plus in 1982, and higher in future years, 1/ it will become increasingly feasible to develop efficient import-substitution ancillary industries. 2/ 1/ In more advanced countries, factory output ranges between 300,000 - 1,500,000 units per annum. Clearly, Turkey is rapidly approaching such levels of production with all that implies in achieving economies of scale. 2/ Basic information is gleaned from TSKB's study: The Refrigerator Industry, 1977.  - 77 - Implications for the SMI Sector 12.19 The SMI sector, consisting of hundreds of small shops producing parts and components for the 6 large factories, is essential to the healthy development of the industry. But to realize this potential, the SMI sector must improve productivity and quality and to achieve such improvement, finan- cial and technical assistance are required. Very small shops should be merged or form cooperatives, thereby increasing their production capacities to handle larger orders and making it more feasible to channel credit, technical advice assistance. A number of large firms, recognizing the important sub-contracting role of the SMI community, including the ability of SMI firms to produce many items more cheaply than large firms, are providing assistance in the form of finance of raw materials, leasing special equipment to selected small shops and technical and quality-control supervision. A concerted effort to expand sub-contracting and ancillary relations is called for. Recommended Export Expansion Program 12.20 There can be little doubt that the Turkish refrigerator industry is inherently capable of rapidly expanding exports not only to nearby Middle East and African countries but over the long-term to Europe and elsewhere in direct competition with long established exporters in Italy and Spain. Some growth will occur spontaneously but the process can be accelerated and rational- ized through careful planning and coordination of the efforts of both the public and private sectors. To start, an international market survey should be undertaken to identify the promising target countries -- over the short- and long-term in an order of priority -- and to formulate the optimum institu- tional structure, resource needs and other measure needed to seize the available opportunities. Plan Objectives in Engineering Industries 12.21 The 1979-83 Plan assigns high priority in the development of the engineering industries. Within this category, the manufacture of capital and intermediate goods has been singled out for special attention, relegating consumer goods to a somewhat lower priority. 1/ Nevertheless, the growth and performance of the durable consumer goods sector (typified by the refrigerator industry discussed in paras. 12.16-12.20) should not be unduly constrained as 1/ Investment targets for 1979-83 are as follows: electrical machinery and electronics TL 9.5 billion, up by 50% compared with the realized levels during the Third Plan; non-electrical machinery TI. 21 billion, an increase of 141%; transport equipment TL 14.5 billion, up by 38%; and agricultural implements and machines TL 7.7 billion, an increase of 217%. There is no breakdown between private and public sectors. The Plan targets appear ambitious in the light of available domestic and external financial resources. Nonetheless, SMI can be expected both to contribute to and benefit from the increased investment.  - 78 - a consequence. Clearly, the consumer goods sector, which is dominated by pri- vate firms has a built-in momentum. Though the major problems of the engineer- ing industries 1/ -- low capacity utilization, inadequate product quality, lack of standardization and, in the automotive industries in particular, need for rationalization of production models and standardization of spare parts 2/ -- have been identified, the Plan does not outline a coherant strategy for tackling these issues -- with the possible exception of the automobile indus- try. The Plan indicates that the public sector would provide assistance to small firms in the non-electrical machinery manufacturing industry provided the small firms reorganize themselves through mergers or cooperative produc- tion relationships. However, there is no indication of specific industries in which small firms would be encouraged nor of supportive measures to develop the SMI sector. Perhaps the major weakness in the Plan is the excessive reliance on public sector manufacturing entities which have not been success- ful in fulfilling assigned roles, and the failure to chart the future course of development of the private sector. Prospects for SMI Development in the Engineering Subsector 12.22 Regarding the SMI segments of the industry, there can be little doubt of the huge potential for growth. Of all the manufacturing sectors, the engineering industries provide the most fertile area for developing ancillary relationships or subcontracting linkages between large-scale manufacturer/assemblers and small-to medium-scale suppliers. Such a division of labor comes about because of the short production runs typical of many processes. Also, since engineering goods in general are skill-intensive and Turkey has a supply of labor with basic industrial skills, engineering products can be produced efficiently at small and medium scales and at varying levels 1/ The problems of SMI engineering industry are summarized in Annex G. 2/ Annex I lists engineering industry products typically produced by small and medium scale establishments in industrially developed countries.  - 79 - of technology. 1/ The low investment requirements would permit greater private sector participation and better prospects of efficient management of these firms. For instance, the Otosan factory -- with almost three times as many in-house workers engaged in component manufacture as in assembly of finished products -- would operate much more efficiently if it could shift a larger share of manufacture to outside suppliers capable of producing relatively small quantities of specific items at lower cost than in a large factory. Its procurement personnel are constantly seeking out and evaluating new ancillary sources of inputs, but the problems of quality, tolerances, timely delivery and working capital finance inherent in the SMI sector are enormous. Technical assistance programs (paras 10.01-10.08) should be designed to provide the needed technical and finance assistance to enable the SMI engineering firms to respond more effectively as sub-contractors to large firms as well as to 1/ Unlike "process" (e.g. steel) industries, most of the equipment employed in the "non-process" (e.g. engineering) industries is multi-purpose, i.e. capable of producing a much wider variety of different products. In addition, equipment of different capacities and embodying different techniques can easily be substituted at the level of individual process stages, while most plants produce an assortment of end products and/or components rather than a single undifferentiated output. These features give rise to strong interdependencies in the sense that the optimal means to produce an intermediate mechanical engineering product (component or sub-assembly) is dependent not only upon its unit cost but also on the volume of demand arising from the output of higher stage products. In turn, the demand for higher stage products depends upon their production costs, which are in large part determined by the costs of intermediate inputs. Therefore, unless the demand for a particular end product is sufficiently large and stable, sub-contracting and the external purchase of components (semi-finished as well as finished) and sub-assemblies assume critical importance. On the other hand, smaller firms concentrat- ing on a few operations or components common to a larger number of producers are able to fully utilize special purpose (and expensive) equipment as well as to reap the benefits of specialization in a narrow area. Other factors supportive of ancillary or sub-contracting arrange- ments include: (a) given that many individual processes within the engineering industries are labor-paced and controlled rather than machine- paced and controlled, the skill of the laborer perforce determines the time required to set up a machine for a production run just as it deter- mines the tolerance achieved or the proportion of rejects; and (b) the rapidity of technological change in many branches of the engineering sector. The available range of raw maerials with highly differentiated, specific properties is constantly expanding; new and improved processes are continually being introduced; and product designs keep changing, sometimes to the extent that completely new products are introduced. Although not all of these changes would rate as dramatic, they nonetheless affect both production costs and the service characteristics of end products and tend to increase the risk of unexpectedly rapid obsolescence. For details see L. E. Westphal, The Methodology of Investment Planning in the Non-process Industries, World Bank Paper, 1976.  - 80 - expand into primary production of certain types of less-complicated finished equipment. As already mentioned (paras. 12.17-12.20), there appears to be a promising potential for rapidly expanding exports in the durable consumer goods sector. Since a large part of appliances in aggregate could be sub- contracted to SMI enterprises, export expansion would provide a strong addi- tional demand for parts and components produced by SMI. Promising Product Areas for SMI Involvement 12.23 The mission has made an attempt to identify product areas where SMI can make a potential contribution and to suggest supportive measures needed to achieve these goals. The growth potential of SMI broadly appears to exist in the following distinct product areas: low value--low technology standard products, high technology--high unit value products, items manufactured under ancillary or subcontracting relationships, and production of a carefully selected range of capital goods. 1. Low Value--Low Technology Standard Products 12.24 Engineering industry products are highly differentiated by price, quality and technical specifications, and thus provide opportunities for SMI to operate in market niches without engaging in direct competition with larger firms. There is a wide range of industrial intermediate and final goods for consumer markets which generally includes low-quality and low-priced standard items as distinct from products with a brand name. Examples of final products include non-electric stoves, builders' hardware, hand tools, cutlery, agri- cultural hand implements, metal furniture, safes, small household appliances, etc. The same considerations apply to industrial items such as fasteners (springs, screws, bolts, nuts, etc.), electrical fittings, industrial heaters, dies, tools, jogs and fixtures, etc. where the low price (and low quality) create a ready market for these low-priced, technologically unsophisticated products. Markets for such items often tend to be fragmented and small, requiring capability to manufacture a wide range of items in short production runs. Also, many of these items are bulky and costly to transport. The large and growing domestic market for such items in Turkey, given inter alia the skewed income distribution (which defines a diversified pattern of demand for both final consumer and intermediate goods) and the regional concentration of the population, offers considerable potential for SMI development. I/ 2. High Technology--High Unit Value Products 12.25 There are many products at the opposite end of the technological spectrum which have a high unit value, and cater to the needs of sophis- ticated, high technology users. Examples of such products include precision measuring instruments, scientific and laboratory instrtmentation, medical equipment, process control equipment (flow control, pressure control and temperature control devices in chemicals, food and allied industries), and 1/ The growth of SMI, on the other hand, by raising the level of employ- ment and income, should begin to have a beneficial effect on income distribution.  - 81 - electronic and electro-mechanical telecommunication products. Because of the high level of product differentiation (often items are custom-made for specific applications), lot sizes are small and manufacture car. be carried out in small-medium scale establishments. 1/ Certainly, this is the case in developed countries. With the large supply of highly trained technicians and engineers in Turkey and the growing investments in process industries (food, chemicals, rubber, plastics, synthetic fibers) and in infrastructure projects (power, telecommunication), considerable potential exists for the development of high technology, small and medium scale industries. This is an area which merits further detailed study by the Government. The Marmara Scientific and Research Institute at Gebze has qualified engineers who could investigate and develop a specific plan for this subsector taking into account the structure and growth of demand. 12.26 A related potential area is the assembly of electronic consumer items for export. The availability of relatively low cost semi-skilled and unskilled labor is a major requirement for setting up such industries. How- ever, in view of the high and rising minimum wage levels in the industrial sector, a careful study of the relative labor costs and comparative advantage of Turkey vis-a-vis other low-wage exporters should be undertaken before any projects are taken up. 3. Products Manufactured under Ancillary and Sub-contracting Relationships 12.27 As already discussed (paras. 12.14, 12.15, 12.18, 12.22), this is an area where SMI has the greatest potential for growth. In most developed countries, small and medium scale firms have production and marketing relation- ships with large firms through ancillary and subcontracting linkages; however, they are more prevalent in the engineering industry. Industries where such linkages have been highly developed include automobiles, aircraft, railway, shipbuilding, machine tools, construction machinery, textile machinery, conveyors and elevators, electrical distribution equipment, electric motors and generators, and consumer durables. The automotive sector (cars, trucks, and tractors) in Turkey has approximately 850 SMI firms as ancillary suppliers, manufacturing a variety of components and sub-assemblies. 2/ The consumer durables industries (refrigerators, washing machines, stoves) also have a large number of ancillary suppliers -- PROFILO and ARCELIK claim to have contacts with 1,200 and 200 SMI firms, respectively. However, outside the transport and consumer durable industries, there is little sub-contracting. 1/ Such products, however, often require components (e.g. semi-conductor diodes, transistors, etc.) which can be produced efficiently in large scale factories. These components would have to be imported since their manufacture may not be economically justified in Turkey. 2! E.g. sheet metal products, castings, rubber, plastics, chemicals, glass, electrical components, batteries, springs, radiators, shock absorbers, pistons, valves, extensive machine shop work, etc.  - 82 - So far, the private sector encourages and relies on ancillary suppliers to a far greater extent than state economic enterprises. This stems from the generally difficult labor situation with strong unions and high minimum wages which have affected private firms more adversely than public sector enter- prises. It may also have to do with the arduous task of lining up suitable and reliable suppliers, and the lack of appreciation for the economic and financial importance of sub-contracting to SMI and the development of effi- cient ancillary suppliers. Suggestions to increase public sector involve- ment are offered in para. 12.33. 4. Production of Selected Capital Goods 12.28 Capital goods production evokes the image of large, highly mechan- ized plants, with large scale economies. What is not fully appreciated is that the capital goods sector comprises many industries which are fairly skilled labor intensive, and that many product lines of the industry, even in the U.S., are characterized by a relatively large number of medium-sized firms (up to 250 workers), and make significant contributions to value added and employment (see Appendix Tables 3.28, 3.29). This pattern has persisted throughout and can not be attributed to factor market imperfections. The wage rate paid by the smaller firms is roughly similar to that of large firms and it seems unlikely that the firm structure would be maintained in the face of low returns on capital. Moreover, average labor product is fairly similar among firms of varying size class and it seems unlikely that the smaller firms exhibit a lower average product of capital. These observations imply that total factor productivity is not likely to be lower in the smaller companies. 1/ Assuming availability of a substantial supply of skilled workers at a fraction of the wage paid to their counterparts in developed countries and not significantly lower productivity, wage levels should confer a cost advantage. Also, relatively lower remuneration is not confined only to skilled operatives but applies equally to engineers and technicians. To the extent that the sector requires substantial inputs of the latter, its compara- tive advantage is further enhanced. Thus, a priori, wage advantages need not be offset by relatively small size. An array of capital goods that could be produced in Turkey by medium scale industries is shown in Appenxix Table 3.28. Some of them are already produced, while other items could be identified by commissioning studies to qualified institutions, such as the Marmara Institute for Science and Technology. 1/ The existence of a spectrum of firms of varying size may partly reflect the production of different commodities within the 4 or 5 digit indus- trial classification, each individual product being produced by a narrower range of firms (by size) than appears in the broader grouping. If this is the case, the potential comparative advantage in developing countries would be among those products in which average firm size is small. For details see H. Park, Fostering the Capital-Goods Sector in LDC's: A Survey of Evidence and Requirements, World Bank Staff Working Paper, No. 376, 1980, pp. 1-4.  - 83 - Strategy and Policies to Promote SMI in the Engineering Sector 12.29 The strategy to develop SMI's potential should be to direct resources and effort to support expansion of the activities outlined in paras. 12.22- 12.28. Encouraging the growth of consumer durable industries would stimulate indirectly the growth of SMI, not only in the engineering but also in other sectors (see para. 12.26, fn 1), provided measures are taken to increase their efficiency and reliability. The potential of SMI should not be viewed narrowly as confined to saving foreign exchange, but also as capable of earning foreign exchange through direct or indirect exports. To this end, efforts toward standardization and quality control are crucial. 12.30 The problems of SMI in the erigileering industries have been pre- sented in paras. 12.21 and 4.04. To encourage SMI to develop its full poten- tial both in the areas of standard and higi technology products, there is a need for the Government to ensure that technical assistance is provided to SMI in such areas as product development, improved designs, improved nanafacturing technology, improved production management and utilization of raw materials, etc. The KUSGEM program at the Gaziantep center was a pilot step in this direction which unfortunately has not yet achieved fully its objectives. The current initiatives through the SIDO program and that of the Half Bank with the cooperation of private sector engineering consultants deserves strong support by the Government. In addition, the Government should ensure that the banking system provides the necessary term finance to SMI engineering firms for acquisition of plant and machinery and for working capital needs. 12.31 Some of the problems in the automotive ancillary industry stem from the proliferation of vehicle models which undermines efforts to improve effi- ciency. The problem is compounded by the existence of a large number of very small units, undercutting each other in prices, so that stable ancillary relationships are more difficult to establish. In addition, lack of product specialization among ancillary firms leads to higher costs. These problems could be effectively tackled through standardization of models and components in the domestic market and an expansion of exports, thereby creating a large market for standard vehicle components. In this regard, ancillary industries should be encouraged to export components directly to automobile manufacturers in the EEC. The Plan has stressed the objective of making the automotive in- dustry more export-oriented, and this is a move in the right direction. How- ever, a strategy and the requisite policy framework have yet to be developed (para. 12.21). Finally, exports of consumer durables for which there appears to be potential in the Middle East and African countries would also benefit significantly ancillary industries supplying components to the large consumer durable manufacturers. 12.32 To support the development of ancillary industries in general and vehicle component suppliers in particular, the Turkish Standards Institute should assist the industry to develop product standards and to grant a quality seal of approval on products meeting its minimum standards. This would be similar to the approval of the Underwriter's Laboratories (UL) in che USA given to electrical and mechanical products. The formulation of  - 84 - standard specifications -- materials, sizes, tolerances, performance para- meters, quality -- should encourage SMI to upgrade the quality of products since greater market acceptance would exist for products carrying the Insti- tute's mark. 12.33 The extreme shortage of foreign exchange has affected raw material availability to ancillary producers who are often forced to purchase imported raw materials at premium prices from the open or black market. Furthermore, though ancillary producers provide indirect inputs to exports, they are ineligible for any tax subsidy, exchange retention or foreign exchange alloca- tion for purchase of raw materials and spares. To remedy these problems, the Government could develop a system of foreign exchange allocation based on the contribution to value added by both final exporters and contributing suppliers. At the same time, the Government could encourage the large firms to share their allotment of improved raw materials with their ancillaries. In this regard, the Chambers of Industry who are responsible for the allocation could be instructed to devise a workable scheme. 12.34 The public sector, which has a poor record of sub-contracting, should be directed by the Government to develop such relationships. Monitor- ing of investment decision in the public sector and encouraging purchase rather than in-house production of every nut and bolt which is uneconomical, could lead to increase in subcontracting. A unit in the Ministry of State Economic Enterprises could be established to promote this important task. Public sector factories could also be encouraged to establish industrial estates for ancillary firms in close proximity to the main factory. Indus- trial sheds and infrastructure could be developed by the large firm in the estate and rented out to smaller firms. In addition, designs, blueprints, tooling (jigs and fixtures) and other forms of technical assistance could be provided by the large firms to the ancillaries to assist them to develop into efficient suppliers.  - 85 - XIII. THE LEATHER PROCESSING AND FOOTWEAR INDUSTRIES The Leather Processing Industry Overview 13.01 The leather processing industry is virtually in private hands. The number of establishments in the organized sector increased from 92 in 1970 to 119 in 1976, while employment expanded from 2,729 to 4,239, or by 7.6% annually, accounting for about 1% of the total manufacturing labor force (Appendix Tables 3.12, 3.13). In 1970 there were about 4,400 establishments in the unorganized sector, employing about 7,200 (Appendix Table 3.11). Their number and employment have probably declined since 1970 due to increasing competition from more modern plants in the organized sector. Investment in real terms averaged TL5.6 million annually during 1966-70, but increased to an impressive TL 50 million during 1971-75 reflecting the trend toward larger and modern units. Capital intensity has tripled between 1965 and 1975, with the K/L ratio averaging TL 50,000 in 1973-75 (Appendix Table 3.16). As expected, capital productivity has declined considerably during 1973-75 (Appendix Table 3.19), also mirroring the shift toward larger plant scale. Real output quadrupled during 1965-75, rising from TL 140 million to TL 550 million. Data on value added show an erratic behavior and their reliability is questionable (Appendix Table 3.3); it probably increased between 1965 and 1975, reflecting improvements at the processing stage by the more efficient modern firms established in recent years. Because of the unreliability of value added time series, labor productivity trends are not very meaningful (Appendix Table 3.17). Despite the fact that the industry has been growing, real wages have marginally declined during the same period, while for the manufacturing sector as a whole they grew at 5.5% per annum (Appendix Table 3.21). This probably indicates availability of cheap labor easy to draw on. Output and investment growth slowed down and employment declined in 1976, and probably also during 1977-78, as a result of the adverse conditions that have beset the Turkish economy and which have particularly affected the smaller firms. Although the industry has made considerable progress in the past decade, it is still troubled by structural and production problems that inhibit it from reaching its full potential. The ensuing discussion highlights some of these problems, as well as those of the footwear industry, and suggests courses for remedial action.  - 86 - Problems Encountered in Raw Hide and Skin Production 13.02 Despite the impressive resource base, 1/ the industry has not realized its full economic potential because of poor product quality. Inadequate feed and diseases as well as damage from thorns, faulty skinning and poor handling after slaughter, especially in rural areas, all contribute to off-quality hides and skins. Spoilage during transport and in storage also occurs because of inadequate control of ambient temperature and humidity, parasites and rodents. It is estimated that up to 65% of all hides and skins are damaged, often to the extent of being unusable to produce even low-quality leather, as the result of the factors mentioned above. 2/ Efforts are underway to ameliorate the adverse conditions affecting product quality but much more remains to be done, particularly in improving pastures and feeding practices (e.g., feed-lot fattening), establishment of better managed and more efficient abattoirs, tighter government inspection, and improved handling, storage and transport facilities. Notwithstanding the implied costs of carrying out a national improvement program, the potential benefits in expanded production and higher quality leather for the domestic and export markets are substantial and there can be little doubt that the investment would be cost-effective over the long term. Production and Consumption of Hides and Skins 13.03 The projected growth of output of hides and skins is shown in Table 1. The small but steady incremental growth in both hides and sheepskins reflects the expected results of various programs sponsored by the General Directorate of Animal Husbandry supported by loans from the Agricultural Bank, the World Bank and others. Presumably, the decline in goatskin availability is directly related to a shift of demand for meat for human consumption from goats to sheep now taking place. Raw product quality should experience a parallel improvement and finished leather production can be expected to grow even more rapidly as damage and deterioration of hides and skins are reduced. 1/ Hides and skins are, essentially, by-product raw material and the supply is directly dependent on the animal husbandry industry whose growth, in turn, depends on the economic, political and natural forces affecting the primary products -- milk, wool and meat. Turkey's livestock wealth makes it one of the important nations in this respect--first in Europe (excluding Russia) in its sheep and goat population, and fourth in goats, sixth in sheep and fifteenth in cattle on a world scale. In recent years, the value of raw hides has averaged about TL 1.3 billion annually, representing 7% of the total value of the end products of the cattle industry (meat, offal and hides). Although statistics vary depending on the source, there is consensus that annual sheep and goat skin production amounts to about 21 million and cattle hides to 2.5 million (1975 figures). 2/ Some 750,000 skins (4% of production) every year are left to decompose in remote rural areas because of preservation and transportation difficulties.  - 87 - Table 1: HIDES AND SKIN PRODUCTION, 1976-1983 (Million Pieces) Annual Growth 1976 1977 1978 1979 1980 1981 1982 1983 % Cattle Hides 2.9 3.1 3.3 3.5 3.7 4.0 4.3 4.6 7.0 Sheep Skins 15.8 16.4 17.0 17.6 18.2 18.8 19.5 20.2 3.5 Goat Skins 5.4 5.2 5.0 4.8 4.6 4.4 4.2 4.0 -4.0 Total 24.1 24.6 25.1 25.6 26.1 26.6 27.1 28.8 Source: TSKB and mission projections. 13.04 All cattle hides are consumed domestically mainly for handbags, luggage, and, most importantly, shoes. Turkey imports small quantities of higher quality leather for special uses and heavy leather for soles, averag- ing about US$1 million annually during 1973-77. PVC and other synthetic substitutes are making inroads. In reality, however, Turkey should be capable of supplying its entire domestic demand and still have a surplus for export. Leather Production 13.05 Data on leather production are insufficient and often contradictory. TSKB concluded 1/ that although apparently 19 million skins were processed in 1974, i.e. the net domestic availability after deducting exports, the quantity of finished leather from skins corresponded to perhaps 13 million skins. The difference, about one-third of total, is assumed to be the amount discarded because of damage to and deterioration of the raw material. Leather produced from hides in 1972 amounted to 450 million dm2 (4.5 million m2) of "calfskin", the softer product used in clothing and shoe uppers, rising to 600 million dm2 in 1974 and to 800 million dm2 in 1977. 2/ 3/ This implies an annual growth rate of 12%. Production of upper leather is proj- ected in the Plan to reach 2,250 million dm2 by 1983, or to grow at 19% per annum, which is overly optimistic. Sole leather production increased from 5,700 tons in 1972, to 6,000 tons in 1974 and 8,000 tons in 1977, 2/ or at annual rate of 7%. The Plan projects an annual growth of about 30% to 29,000 tons, which is unrealistic even if dramatic improvements are made in quality, because of constraints in the poduction of hides. 1/ Leather and Leather Products Industry, 1976. 2/ 1978-83 Plan estimate. 3/ Since each hide measures about 2.5 m2, the figures on hide leather production indicate, as in the case of skins, a very high attrition between raw hides and finished leather.  - 88 - Leather Production Facilities 13.06 Tanneries in Turkey are classified under three headings -- traditional, mechanized and modern. 13.07 Traditional Tanneries. They include small, family-owned plants using only manual labor and vegetable tanning materials, often operated seasonally. These are slowly disappearing as demand for their lower quality products declines and competition from larger more efficient tanneries increases. Of an estimated 1,600 units in 1958, only 350 were thought to be operating in 1974. In 1976, TSKB estimated that traditional tanneries processed only about 4% of raw hides and skins (mainly the latter) and that few if any will have survived early in the 1980's. 13.08 Mechanized Tanneries. About 50% of the mechanized tanneries employ less than 10 workers and about 90% less than 50. They use simple labor- intensive machines and are generally established in old, unsanitary buildings with inadequate water and other services. Technology is primitive. Located within the city's perimeter, they cause air and water pollution, and pose an acute problem of relocation. The largest concentration of this "intermediate" sub-sector is in Kazlicesme-Istanbul where 60% of Turkey's leather is produced and, to a lesser extent, in Izmir and Bursa. Despite the plant and infrastructure problems 1/, this group of tanneries appears to be surviving and even grow -- in 1974, there were about 500 vs. 434 in 1958 -- and their share in total production is preponderant as shown in Table 2. Table 2: PRODUCTION CAPACITY OF MECHANIZED TANNERIES --1974 Type Unit of Measurement Capacity Skins Pieces 19,000,000 Hides: Calfskin dm2 650,000,000 Sole Leather tons 7,000 Even though some units are not operating at full capacity, this segment offers stiff competition to modern plants, 2/ which is suggestive of its importance. However, many of these firms turn out leather of inferior quality and over- priced, and have great difficulties in marketing their product. This derives 1/ The problems and technical assistance requirements of the industry are summarized in Annex G. 2/ TSKB had made six loans to private companies to introduce modern leather processing technology, but the companies involved are experiencing difficulties competing with the smaller traditional tanneries. The development loans had been made on the basis of an understanding that the traditional leather-working establishments in the Kazlicesme district of Istanbul would be gradually closed down. These measures were not, however, taken by the government probably for social reasons.  - 89 - in part from the fact that these tanneries are located away from the sources of raw material, thereby incurring higher transport costs and waste due to deficient preservation of skins. At least the smaller of these plants will find it increasingly difficult to compete with existing and new large plants and, unless they reorganize to reach economical size and modernize their chances of survival appear to be slim. It is estimated that a plant operates at optimum capacity when it processes 2,000 sheep and goat skins and 400 cattle hides a day. This implies a plant employing some 100-120 workers. But attempts to organize these smaller plants in larger units and to relocate them near the source of raw material so far have not been successful 1/. 13.09 Modern Tanneries. Three modern tanneries have come into existence since 1972 when "Certificates of Encouragement" were first made available as a means of fostering modern, efficient plants of economic size capable of competing internationally. Total capacity is 2.7 million skins and 180,000 hides, corresponding to 13% and 8% of the national skin and hide supply, respectively. A further 21 certificates have been granted, of which 14 will produce leather garments also. If all are implemented, the toal capacity of the modern sub-sector will be 11.4 million skins and 1 million hides. This expected expansion of large-scale tanneries, even if not entirely realized, has serious implications for the "mechanized" sub-sector. Inevitably, many of the less efficient units will disappear; but given the potential economic and social contributions of efficient small to medium scale tanneries, organized efforts should be made to assist them through mergers, the formation of cooperatives, technical assistance, the development of special product niches, promotion of subcontracting and other forms of market expansion, and the provision of factory buildings and infrastructure (industrial estates) in suitable sites. The long-term strategy should be two-pronged -- the simultaneous strengthening of the mechanized (small to medium scale) and the modern (medium to large scale) sub-sectors as complementary components of the sector in aggregate. The proposed study for restructuring the footwear industry would also suggest a course of action concerning the restructuring of the leather tanning segment, including realistic options for possible relocation. Technology 13.10 The introduction of electronically controlled machinery per se apparently contributes very little, if at all, to quality improvement. Under these circumstances, and given the availability of relatively cheap labor in Turkey, adoption of sophisticated labor-saving techniques does not appear justified. Furthermore, because the raw skins produced in the country do not come in standard size, continued use of high labor inputs is necessary--which dinimishes the potential advantages of highly capital- intensive technology. Use of "appropriate" leather processing technology is likely to serve both the private and public interest. 1/ Such relocation would also alleviate the acute pollution problems.  - 90 - Leather Products 1/ 13.11 In 1970, there were 4,400 establishments in the unorganized sector producing leather products excluding footwear, with total employment of about 7,200, or an average of 1.6 persons, which reflects the minuscule size of these enterprises (Appendix Table 3.11). The number of establishments was about 10% lower than in 1963, suggesting a declining population. Recent statistics are not available, but probably the trend has continued. In addition, there were 92 establishments in the organized sector employing about 2,700 workers. Their number increased to 119 in 1976 while employment expanded to 4,240, suggesting a rise in average employment and, by extension, in enterprise size from 30 to 36 persons (Appendix Tables 3.12, 3.13). Leather Garments 13.12 In 1971, there were 1,500 shops producing leather garments employing an estimated 8,000 workers and fixed assets amounting to TL 100 million. Most of the shops are typically very small scale, employing only 3-4 people estimated and processing less than 20,000 skins per annum. Units processing 20,000-50,000 skins a year are classified as medium scale; those processing 50,000-100,000 a year or more large scale. Of the 1.8 million leather gar- ments produced in 1974, it was estimated that 40% were produced in the 15 medium to large scale plants located in Istanbul (9) and Izmir (6). The small scale shops are also clustered mainly in Istanbul with smaller numbers in Ismir, Bursa and Ankara. Since 1970, the trend appears to be toward larger production units which have a comparative advantage in their ability to meet the exacting standards of the export market where over three quarters of the production is sold. Sometimes, expansion of these more competitive factories takes the form of setting up a new branch or affiliated plants at other locations. 13.13 Up to 1967,an estimated 200,000 skins were processed annually into garments; but since that time expansion of the industry has accelerated as greater shares of worldwide production have shifted from the developed to the developing countries. The trend is illustrated in Table 3. Production rose from 442,000 pieces in 1969 to an estimated 2.3 million in 1977 2/, which is remarkable. 1/ Under this heading are included the two major leather consuming industries -- garments (coats, jackets, skirts, hats and gloves) and shoes. Miscellaneous products of lesser importance include, luggage and handbags, saddlery, furniture, belts, wallets, bookbindings, desk accessories, sporting goods and souvenirs of all types. 2/ The 1978-83 Plan figure is 2 million pieces, which is probably an underestimate.  - 91 - Table 3: PRODUCTION OF LEATHER GARMENTS, 1969-1977 ('000s of pieces) /1 Year Production Exports Quantity (%) 1969 443 148 33 1970 541 248 45 1971 709 409 58 1972 911 586 64 1973 1,337 937 70 1974 1,840 1,390 76 1977 2,300 /2 n.a. n.a. /1 Export figures may be understated, as domestic demand (production less exports) also includes sales to tourists. /2 Mission's estimate. Source: State Planning Organization, adjusted by TSKB. Plan projections raise production by 100% to 4.6 million pieces in 1983, at an investment cost of TL 840 million (US$ 33 million) in 1978 prices. The Plan also provides for an investment of TL 3 billion during 1978-83 to improve leather processing. Such production targets are feasible provided there are dramatic improvements in processing efficiency and quality (see para. 13.15). 13.14 The value of output in the organized sector rose from TL 250 million in 1970 to TL 1,204 million in 1976 (Appendix Tables 3.12, 3.13). The manu- facture and export of leather garments, stimulated by liberal incentives, has increased rapidly in recent years. As shown in Table 4, exports of leather Table 4: EXPORTS OF LEATHER PRODUCTS /1 (US$ million) Total Value of Manufacturing Year Exports Exports % 1970 4.6 588.5 0.8 1971 10.6 676.6 1.6 1972 21.5 885.0 2.4 1973 45.3 1,317.1 3.4 1974 74.0 1,532.2 4.8 1975 64.9 1,401.1 4.6 1976 59.9 1,960.2 3.1 1977 52.0 1,753.0 3.0 /1 Includes hides. Source: State Planning Organization.  - 92 - products rose from US$4.6 million in 1970 to US$74.0 million in 1974 (4.8% of total exports of manufactures). They have been declining every year thereafter to US$52.0 million in 1977 (3% of total manufactured exports), reflecting the decline in the export of raw and pickled skins diverted for local processing, the slow-down of economic activity in Western Europe, and rising domestic demand (in part from tourists). West Germany takes about 80% of leather gar- ment exports with the balance distributed among the USA, Holland, Switzerland, France, Austria and other smaller markets. The Plan projects exports of leather garments at US$187 million by 1983, up from US$52 million in 1977--an overly ambitious target. Prospects and Strategy for the Development of the Leather Garment Industry 13.15 Although Turkish production is quite well-established in world markets, there are quality and delivery problems. If these problems are not addressed, rising competition from the Far Eastern and Latin American regions is likely to place severe constraints on future growth. Nevertheless, given the history of growth and that forecasted for the world market, mainly in the EEC countries, continued expansion of exports by Turkey should be attainable even though, perhaps, at a slower pace than in the early and mid-1970's. To meet the market challenge it will be necessary first to improve raw material quality (see para. 13.02). Accessories (buckles, buttons, zippers, linings, etc.) must also be upgraded and, importantly, delivery schedules maintained rigorously. Consistent failure to deliver as contracted really reflects the full spectrum of sectoral deficiencies. Beside the overriding problems arising from poor leather quality, performance is also hampered by the large number of small inefficient producing units, shortages of qualified labor, and management and technical weaknesses. Government strategy to improve the sector places primary emphasis on promoting large scale firms. But it would be prudent to concentrate attention also on the SSI subsector. In view of the important complementary role of SMI, the strategy should have the parallel aim (a) to facilitate the expansion of large scale firms and the adoption of appropriate technologies (see para. 13.10); and (b) to strengthen the small- medium scale firms, to the end of developing a functionally integrated sector. SSI can be upgraded as sub-contractors to larger factories as well as manufac- turers of special lower-demand products through the provision of adequate physical facilities (industrial estate), term credits for appropriate types of modern equipment (many of simple but efficient design which could be produced in Turkey), technical and marketing assistance and training. Of fundamental importance would be incentives to foster cooperative arrangements and mergers. Larger, but still small-scale, production units--say in the range of 10 to 40 workers each--working in close harmony with large production/marketing would contribute effectively in maintaining and enhancing Turkey's international competitiveness. Specific recommendations would be one of the objectives of the proposed sector study (para. 13.30).  - 93 - The Footwear Industry The Industry 13.16 There were 37 establishments in the organized private sector in 1976, employing some 2,100 workers (Appendix Table 3.13). No reliable data on the number of establishments in the unorganized sector are available. It is estimated that probably 6,400 workshops were in operation in 1975, employing 20,000-25,000 workers, and accounting for some 90% of the total leather shoe production. Two state factories account for 50% of the total footwear produc- tion in the organized segment of the industry. Istanbul and Izmir are the major areas of concentration. Investment in the organized sector increased from less than TL 5 million in 1970 to TL 39 million (in current prices) in 1976, suggesting increasing outlays for modern equipment and larger plant scale. Output and value added registered a remarkable six-fold increase during the same period (Appendix Tables 3.11, 3.13). The sector faces the same structural and production problems as the leather processing industry, as discussed below. Structure and Performance 13.17 In the early 1950's, there were very few shoe factories producing for the general market. Most of the demand was met by one- and two-man cobbler shops who produced for individual consumers on order. Because of the rapid urbanization, the rising standard of living and the emergence of chain stores, the demand has since shifted dramatically from custom-made to ready-made shoes. Fashion and design have become much more of a factor. The artisan-cobbler has been replaced by mini-factories (selling through middlemen) and, since the early 1970's, a trend to larger factories has developed. Three categories of factories are recognized according to the production process used: (a) factories using a conveyor belt system; (b) mechanized factories with no conveyor belt system; and (c) small workshops -- assembly factories based on piece work. 13.18 Factories Based on Conveyor Belt Production. There are three establishments in existence in the private sector with a capacity of 2,000 pairs/8 hours each, and a state-owned enterprise with an installed capacity of 6,500 pairs/8 hours. Capacity utilization rates in the private sector are reportedly satisfactory. The state factory and two of the private ones are located in Istanbul. 13.19 Mechanized Factories with No Conveyor Belt System. In this less- modern lower-capacity category, there are 16 factories, of which 14 are located in Istanbul and one each in Ankara and Izmir. Installed capacity totals 6,630 pairs/8 hours, ranging from 200 to 1,000 pairs per factory with an average of slightly over 400. One million pairs were produced in 1975 compared with a nominal single shift capacity of 2 million for a 300-day production year, implying that this sub-sector worked at only 50% capacity that year. Various reasons are advanced for low capacity utilization, including: (a) several factories were still installing or starting-up newly  - 94 - acquired modern machinery in 1975; (b) inadequate buildings and inefficient layout of equipment; (c) congestion in and around the "old quarter" in Istanbul where many of the factories are located, interfering with deliveries and shipments; (d) damaged motors due to frequent power cuts and low voltage conditions; (e) haphazard delivery and off-quality (rejected) inputs from ancillary suppliers; (f) market problems -- the domestic market being geared to a wide range of styles and low-volume production runs, necessitating frequent changeovers which reduce productivity. 1/ 13.20 Workshops. Although 90% of shoes with leather uppers are produced in this sub-sector, statistics are fragmentary and unreliable. One source ("Bag Kur" -- an insurance institution serving the footware industry) lists 23,232 enterprises (1975), about half located in Istanbul and Izmir and the balance in some 35 other cities and towns. However, the listing does not differentiate between cobblers, manufacturers, traders in raw material, last makers, other ancillaries, etc. According to a limited survey conducted by TSKB, only about 6,400 of these insured units can be classified as "work- shops", i.e., mini-factories where a number of workers are employed producing (assembling) finished shoes. Approximate production by region for 1975 is shown in Table 5. Estimated production accounts for 91% of national leather shoe production. Table 5: FOOTWARE PRODUCTION IN WORKSHOPS, 1975 Province Workshops /1 Production (Millions) (%j Istanbul 2,536 23.1 52 Izmir 2,707 13.0 29 Gaziantep 350 3.7 8 Ankara 800 2.5 6 Other -- 1.9 5 6,393 44.2 100 /1 TSKB's estimate. 13.21 Typically, a workshop is owned by a primary school graduate, occupies rented quarters in an old rundown building, uses little or no power, and employes one or two apprentices; occasionally, a "larger" workshop will also employ one or two qualified workers. Materials are supplied by a wholesaler and finished shoes are delivered to the same middleman who markets the product, paying minimum prices. Thus, financing of working capital by the middleman permits any qualified worker to set up his own shop if he can in some way acquire a bench, simple tools and a work area. There is no doubt that the middleman serves a needed function, but by the same token the system insures a permanent proliferation of small shops, cutthroat competition and marginal incomes for owners, qualified workers and apprentices. Product 1/ See also Annex G.  - 95 - quality is acceptable for the local market at the relatively low prices paid for workshop produced shoes but, generally, it cannot meet the standards of upper income national buyers and could not possibly be exported. Because of the low level of education of workshop entrepreneurs -- apparently the foot- ware industry attracts the residual elements of each upcoming generation of apprentices who fail to break into higher paying sectors -- it is particu- larly difficult to impart technical and management assistance to them, of which there is little available through Government institutions. Signifi- cantly, despite the widespread growth of regional construction cooperatives by the SSI of many industrial sectors who finance, build and occupy indust- rial estates, there has been virtually no initiative for such self-improve- ment among the shoe workshops. In part, the reason may be that shoemakers are explicitly excluded from Government-supported industrial estates in Turkey. In Konya (Central Anatolia), an estate has been set up by some 300 shoemakers without Government aid. 13.22 Workshop costs are higher than in mechanized factories by about 30%, even though labor and overhead in the former are lower. Raw materials cost per unit of finished product are also higher in the workshop partly because of middlemen profit taking, the small volumes involved which preclude alternative sources of procurement, but most importantly because of wastage in the inefficient, poorly managed operations. Based on TSKB data, the cost comparison between the workshop and the larger factory is shown in Table 6. Table 6: COST STRUCTURE - WORKSHOP VS. FACTORY Item Workshop Factory (TL/Pair) (% ) (TL/Pair) ( % ) Materials 105 68 90 75 Labor 45 29 20 17 Amortization 1 -- 5 4 Overhead 4 3 5 4 Total 155 100 120 100 The normal mark-up for the manufacturer is 20-25% and for the retailer 35%. Wholesalers' profits are 15% and, whenever they intermediate, their margins cut into the manufacturer's profit. Besides the production cost advantage factories produce better quality. 13.23 A cost comparison between Turkey and U.K. shows that, given improve- ments in raw material handling, introduction of more efficient leather processing technology, 1/ appropriate training, proper location, modernization and attain- ment of minimum economical scale of production, Turkey can have a substantial comparative advantage. 1/ However, see para. 13.10.  - 96 - Table 7: COST STRUCTURE (Percentages) Turkey U.K. Material 75 45 Labor 17 34 Overhead 4 18 Amortization 4 3 Total 100 100 Source: P.N.E. Wrights, The Problems of Modernizing the Footwear Industry in Developing Countires, quoted in TSKB, the Turkish Footwear Industry, p. 28. The Domestic Market 13.24 Production of footware in all its forms must be considered not only as a down-stream industry and major consumer of leather but of several other basic raw and intermediate matrials including canvas, yarn, plastics, rubber, glues and chemicals, nails and other hardware. The demand for foot- ware in Turkey is entirely met by domestic production. Approximately 150 million pairs of shoes were produced in 1975 as shown in Table 8. The figures are impressive but there is still room for significant expansion. 1/ Consump- tion of shoes with leather uppers, for example--the largest component of production--works out to 1.21 pairs per capita per annum, compared to 1.54 pairs in England, 1.78 in France and probably about 2.0 in the USA and other high income countries. Over time, Turkish demand will increase as population growth continues at high levels (2.5% per annum) and the general standard of living rises--probably around 9% per annum in the next five years. However, if the sector is to meet the need in future years, particularly in leather shoes, and to avoid the paradox of a major leather-producing country such as Turkey having to resort to importing shoes, a substantial effort must be taken to expand production and improve quality. The Plan provides for an investment of TL 1.4 billion (US$54 million) in the foot-wear industry during 1978-83. This is a somewhat unrealistic target. Three shoe factories are now under construction in the public sector (Sumerbank) at Kars, Sinop and Van, at an estimated cost of US$23 million. They will have a capacity of one million pairs each annually of leather shoes with leather and rubber soles. The plant at Kars, in addition, will produce one million PVC boots. Production is intended for the domestic market. The factories are expected to start opera- tion early in 1982. Nevertheless, due consideration should be given to the suggested restructuring of the footwear industry (see paras 13.27, 13.28). 1/ The Plan's actual production figures for 1972 and 1977, and projections for 1983 for various types of footwear are shown in Appendix Table 5.1. The Plan figures appear to grossly underestimate past production levels.  - 97 - Table 8: FOOTWEAR PRODUCTION, 1975 (Million pairs) Type Shoes Slippers Total Leather 48.7 11.6 60.3 Textile 5.4 -- 5.4 Artificial Leather 0.5 25.0 25.5 Rubber 20.0 -- 25.5 Plastic (Inj. Molded) 34.0 -- 34.0 "Tokyos" 5.1 -- 5.1 Total 113.7 36.6 150.3 Source: TSKB, The Turkish Footwear Industry, 1977. Exports 13.25 Exports have been inconsequential except for 1974/75 when one of the modern large factories delivered an army order to Iraq--US$2.3 million and US$3.3 million in 1974 and 1975, respectively. Since then, exports have dropped back to normal levels of about US$100,000 per year. However, with rising internal demand for better quality shoes and the potential for export which undeniably exists, a number of investors have come forward and have been granted project permits. Currently there are 14 valid permits and TSKB expects that several of these will be implemented over the near to medium term when financing problems are resolved. Export Incentives 13.26 The only incentive for shoe exporters is the tax rebate. Export earnings of up to US$1.8 million (f.o.b.) annually qualify for a 15% rebate; above this limit exporters receive a flat 20%. Export licences are granted to bona fide traders by the General Director of Assessment at the Ministry of Commerce. In practice, very few manufacturers qualify for export licenses. Development Strategy for the Footware Industry 13.27 As with tanneries and leather garments, Government strategy to expand and improve the shoe industry emphasizes the promotion of the large scale segments of the industry. This is an approach deriving from the convic- tion that there is an excessive number of basically non-viable small-scale enterprises and that the future of the industry would be best served by fostering larger, efficient and internationally cost competitive producers. As far as it goes, there can be no argument with the premise. However, unlike the tannery subsector, efficient shoe production (and leather goods generally) can be based on linkages between well-managed SSI units and larger factories. In adopting a strategy which recognizes the important potential of SSI, sectoral performance will depend on the success integration of SSI production  98 - capacity with the production and, more importantly, the marketing capabilities of the larger scale enterprises. Certainly, the SSI subsector--which currently accounts for about 90% of national leather shoe output--cannot be ignored with- out serious social and economic effects. Employment considerations alone would justify the rational nurturing of small labor-intensive but efficient workshops. In short, since the thousands of workshops cannot be replaced by even the most optimistic anticipated growth of the large scale modern sub-sectors over the medium term, development strategy must perforce contemplate the allocation of capital and technological resources to assist the SSI sub-sector in fulfilling an appropriate role in future growth and rationalization of shoe production. 13.28 Corrective action will be needed on many fronts in an integrated national program to achieve: (a) institutional strengthening of both public sector agencies concerned with the shoe industry (e.g. PENDIK) and the private associations of enterprises; (b) expansion of production and/or improvement of quality of leather and other material inputs including textiles, yarn, hardware, etc.; (c) improved plant and infrastructure, e.g. relocation on industrial estates; (d) term credits and technical assistance in project analysis equipment selection and procurement, production line layout and operations, machine maintenance, quality control, etc.; (e) formation of larger and more efficient producing units through, e.g., mergers of micro-shops and organization of cooperatives; (f) promotion of sub-contracting and other marketing assistance, including organized procurement by public agencies and export development; (g) management training in financial planning and management and other business practices; (h) skilled worker training and upgrading. Government Policy 13.29 The Goverment's Policy on the leather industries is set forth in the 1979-83 Plan. The aim is to: (a) improve the quality of hides and skins; (b) rationalize the industry; (c) re-organize and strengthen PENDIK (the technical assistance agency for the leather industry in Istanbul, established with assistance of UNIDO); (d) ban exports of raw and semi- processed hides; (e) establish standards for leather and promote quality control programs in firms; (f) establish a leather stock exchange (commodity  - 99 - exchange) to facilitate the sale and purchase of leather; (g) encourage export of leather products and footwear. Although specific measures are not spelled out, the Government's policy is in the right direction. Action Program for Restructuring the Footwear Industry 13.30 The singular problems of the footwear industry distinguish it from other subsectors--its structural, technical, financial and marketing problems are so deeply entrenched and all-pervasive that they clearly point to the need for a comprehensive program of development as a new point of departure. For this most disadvantaged though promising sector, therefore, the mission proposes a basic study leading to the formulation of an integrated national program of rehabilitation and reorganization of the sector. Much of the needed sectoral data and analysis already exist in past studies. The new study would be oriented more to the preparation of a comprehensive Action Plan of institutional reorganization and strengthening, technical assistance, and credit aimed at transforming the industry into an efficient, cost-competitive activity. The participation of the thousands of independent shoe craftsmen, now threatened to be displaced, would be fundamental to the success of the integrated program envisaged, although it is likely that somewhat larger (but still small-scale) enterprises will have to be encouraged, whether as coopera- tives, partnerships or corporations. These more efficient production units, besides supplying local demand as in the past, would be capable of sub-contract production for large-volume assembly and distribution organizations--e.g. chain stores, a relatively recent but growing phenomenon in Turkey. In time, a share of the large and expanding international market for shoes could be captured. The study should be carried out by an internationally recognized consulting firm, under the auspices of TSKB/Sumerbank/Ministry of Industry.  - 100 - XIV. THE WOOD-BASED INDUSTRIES Introduction 14.01 There is a multitude of primary and secondary manufactures classi- fied under the heading of forest-based industries including logs, sawn-wood, cellulose, parquet, various types of processed board or panels (plywood, veneer, particle board, fiber board), boxes and furniture. Of these, the mission has selected sawn-wood, wood-based panels and furniture as priority target industries because of the importance of small and medium scale enter- prises in their sectoral structure, output and performance. Rational develop- ment of the SSI subsector is especially critical in two of the industries -- sawn-wood and furniture. For technological reasons, wood-based panel produc- tion tends to be concentrated more in the medium to large scale factory range, but as the basic supplier of raw material to the furniture industry, this sub-sector must also be given consideration in an effort supportive of SMI. The Resource Base 14.02 The country's total forest area is estimated at 20 million hectares, with the Black Sea and Mediterranean regions each accounting roughly for 25%, and the balance evenly distributed among the Marmara, Aegean and Anatolia regions. Production of industrial roundwood rose from 2.4 million m3 in 1963 to 6.2 million m3 in 1973, or at an average annual rate of 10%. The Forestry Plan 1973-95 projects potential production of industrial roundwood at 14 mil- lion m3 by 1985 and some 20 million m3 by 1995. Industrial roundwood demand is projected to rise from 6 million m3 in 1975 to 27 million m3 by 1995. Although Turkey has major reserves of unutilized wood, the country is a net importer of forest products to the tune of about US$19 million in 1977. The Government, with Bank support, is introducing a comprehensive forest management system to expand log production and attain self-sufficiency in industrial roundwood by the year 2000. 1/ Sectoral Overview 14.03 The wood-based industries sector is dominated by very small enter- prises. It comprises some 23,200 establishments (about 6,200 in furniture making), of which only about 240, or 1% of the total, employ 10 workers or more (Appendix Tables 3.9, 3.11). The size distribution of firms in the organized sector is presented in Table 1. Only 50 firms employ 100 or more workers. The sector also includes 16 large public sector saw mills. Total 1/ Turkey, Northern Forestry Project, Staff Appraisal Report No. 1870a-TU, March 12, 1978.  - 101 - fabie 1: SIZE DISTRIBUTION OF ESTABLISHMENTS - 1977 Size /a Number of Establishments 10-49 145 50-99 48 100-199 33 200+ 17 Total 243 /a Expressed in terms of workers employed. Source: Appendix Table 3.9. employment in the sector is estimated (1977) at 42,000, of which about 19,000, or 45%, in establishments with 10 or more workers (Appendix Table 3.10). In 1975, employment in private sector enterprises with 10+ workers accounted for 58% in wood products and 60% in furniture (Appendix Table 3.1). Employment in wood-based industries accounts for 2.5% (1975) of total manufacturing in enterprises with 10 or more workers. Employment in the sector rose substan- tially in the past decade or so. In the private sector, it increased from 4,150 in 1965 to 9,750 in 1975, or by 9% per annum, while in the public sector from 3,900 to 7,100, or by 6.2% a year (Appendix Tables 3.24, 3.25). Employ- ment in the private sector (establishments with 10-199 workers) continued to grow at the same pace through 1977, reaching 10,530 workers. It is noteworthy that SMI (10-199) accounts for 55% of employment in the organized sector. Within SMI, 45% of employment is in the 100-199 size bracket (Appendix Table 3.10). 14.04 Production in the wood-based industries (10+ workers) rose constantly during 1965-75 from TL 188.4 million to TL 583.2 million in real terms, or by 12% per annum, while in the public sector from TL 149.5 million to TL 304.0 million, or by 7.4% (Appendix Tables 3.26, 3.27). During the same period, value added increased by 10.5% per annum in the private sector and by 6.7% in the public (Appendix Tables 3.3, 3.4). Real investment increased from TL 10.8 million to TL 133.3 million, or by 28.5% annually in the private sector, and from TL 13.2 million to TL 35.7 million, or by 10.5% in a year in the public sector (Appendix Tables 3.5, 3.6). The rise in investment was accompanied by a doubling in capital intensity (Appendix Tables 3.15, 3.16). Real wages doubled between 1965 and 1975 in the private sector and rose by 75% in the public sector (Appendix Tables 3.21, 3.22). Despite the high rate of growth of value added, investment and employment, statistics show (Appendix Tables 3.17, 3.18) that labor productivity in establishments employing 10 or more workers in both the public and private sectors has stagnated. Persistent low capacity utilization and continued employment of outdated equipment probably account for the observed performance in the private sector. Over- staffing and other management problems seem to be at the root of public sector performance.  - 102 - Sawn-wood Industry Structure and Performance 14.05 The most striking feature of the sawn-wood subsector is the large number of private SMI units using simple low-capacity tools, equipment and technology, existing side-by-side with a small number of much larger, rela- tively modern public sector enterprises. In 1977, it was estimated that there were about 6,000 private sawmills with an installed capacity 1/ of 8.7 million m3/yr of logs -- an average of 1,450 m3/yr/mill. At the same time, there were 16 public sector sawmills -- seven of which are also integrated with parquet production -- with a total installed capacity of almost 524,000 m3/yr for an average capacity of 33,000 m3/yr/mill. Thus, almost 95% of sectoral processing capacity lies in the private SMI sector. However, partly for reasons connected with uneven supply of raw material and marketing procedures which favor the public sector, 2/ capacity utilization in the private sector is 44% compared to 88% in the public sector. Old equipment subject to breakdown may also explain in part the low capacity utilization of the SMI sector. 14.06 Geographically, the private sawmills tend to be located in the major consumption areas such as Adana (382), Ankara (300), Istanbul (479), Izmir (370), Bursa (545), Antalya (356), and raw material must be shipped in from the forested regions. The public sector plants, on the other hand, are all located in the Zonguldak-Bolu region close to major sources of logs. Thus, the private sector is subject to a transport cost penalty on log input which represents 50% higher weight than the finished product. 14.07 There are no data available on relative costs of production in the private and public sectors but it is likely that the differences, if any, are not large. This assessment derives from the fact that, despite low capacity utilization and raw material cost penalties, the SMI sector has survived and still accounts for almost 90% of sawn-wood production -- a clear case of low-technology labor-intensive small-scale enterprises competing successfully with the more modern, capital intensive medium- to large-scale sector. It is reasonable to postulate, however, that the SMI sector is sustained in large measure by paying low wages and earning marginal profits, a situation that can be improved through technical assistance and better equipment to enhance operating efficiency. 3/ 1/ Installed capacity refers to the amount of raw material (as logs) capable of being processed to sawn wood on a one-shift basis. Normally, the out- put of sawn-wood is 65% of the log input. 2/ Production and marketing of logs is controlled by the General Directorate of Forestry. 3/ Deficiencies and technical assistance requirements of the industry are summarized in Annex G.  - 103 - Supply and Demand 14.08 Available studies indicate that 73% of sawn-wood is used in con- struction, 15% in furniture, 10% in packing materials and 2% miscellaneous. Past supply and consumption statistics and the projected consumption for 1976-1983 are tabulated in Table 2. Internal production and consumption were in balance between 1970-75 growing at a rate of 1% annually. Since 1976, Turkey has been importing sawn-wood. A TSKB study projects an increase in consumption of 6.4% per annum during 1976-83, and a growth rate of 6% in production through planned expansion of public sector capacity (log input) and improved capacity utilization and efficiency in the SSI private sector. According to these projections, there would be no need for imports from 1980 on. The 1978-83 Plan projects an average annual growth both of production and demand of 7.5%, which is optimistic. 1/ Due to the prevailing adverse economic conditions, however, an annual increase of about 4% for both consumption and production would be more realistic. Existing surveys of forest resources indicate that there is no constraint on potential log production in the fore- seeable future. Table 2: SAWN-WOOD PRODUCTION/CONSUMPTION, 1970-1983 /a (000's m3) Year Production Domestic Consumption Exports 1970 2,861 2,805 53 1971 2,865 2,800 65 1972 2,910 2,850 58 1973 3,100 3,040 60 1974 2,969 2,900 69 1975 2,990 2,970 20 1976 - 3,080 - 1977 - 3,270 - 1978 (est) 3,250 3,480 - 1979 " 3,380 3,400 - 1980 " 3,515 3,500 - 1981 " 3,655 3,640 - 1982 " 3,800 3,800 - 1983 " 4,000 3,950 - /a TSKB, Sawnwood and Wood-based Panels Industry, 1977 and mission's projections. Prices 14.09 The picture that emerges from the foregoing discussion is one of negligible growth, if any, in the early 1970s and only modest growth projected 1/ 1978-83 Plan, pp. 396-7. As a result, production and consumption are projected to rise to 5,455,000 m3 by 1983. This is about 9% higher than TSKB's projections.  - 104 - between 1976-83, accompanied by low capacity utilization in the SSI sector. Implicitly, there is a large unsatisfied demand for sawn-wood and this assess- ment is supported by the price trend tabulated in Table 3. The sharp price jump of about 185% between 1973 and 1976 assuredly reflects a growing shortage of logs and this adverse impact on production costs carries through to the wood-based panel industries discussed later. Table 3: SAWN-WOOD PRODUCER PRICES Year TL/m3 1965 470 1970 570 1973 700 1974 1,000 1975 1,400 1976 2,000 Strategy and Policies 14.10 Government policy has strongly favored the creation of publicly owned large-scale modern sawmills. Even today, promotion efforts are con- centrated mainly in the expansion of the public sector and little organized assistance is available to the private sector. Presumably, the rationale is that large modern mills are more efficient through economies of scale. Yet, the SMI private sector still accounts for an overwhelming proportion of both production and employment and is competitive under adverse conditions-- though incomes of both the workers and the enterprises are probably close to subsistance level. Certainly, the planned 50% expansion of public sector capacity 1/ and the implied investment in high-cost imported equipment can be strongly challenged in view of the 56% unused capacity in the SMI sector. 14.11 The strategy for the development of the sawn-wood industry and for alleviating existing constraints would involve: (a) Government Policy Reorientation - towards a more balanced approach to raw material allocation between the two sectors in recognition of the important role of the private sector; (b) Production and Marketing of Logs - measures must be taken to expand production and maintain a more even flow of raw material in line with projected consumption demands; this will also tend to constrain future price increases; 1/ From 524,000 m3 to 757,000 m3 during 1976-81.  - 105 - (c) Term Credit and Technical Assistance - to the SMI sector in support of re-equipping mills with more efficient tools and appropriately-sized machines; (d) SMI Subsector Reorganization - fostering of company nergers and cooperatives in order to eliminate the smallest mar- ginal producers in favor of larger (but still small-scale relative to the public sector) more efficient mills. Wood-based Panels Structure and Performance 14.12 The wood-based panel industry in Turkey started with the production of plywood in 1930 and has since expanded to include the production of fiber board and particle board. Sharp increases in production capacity have occurred since 1965, especially in particle board, as shown in Table 4. Table 4: PRODUCTION CAPACITY OF WOOD-BASED PANELS, 1965-76 /a (000's m3) Year Plywood Fiber Board Particle Board Total 1965 36.0 30.0 6.0 72.0 1969 50.0 36.0 60.0 146.0 1971 50.0 42.0 160.0 252.0 1973 60.0 48.0 195.0 303.0 1976 81.5 76.6 361.5 519.6 /a Source: TSKB, Sawnwood and Wood-based Panels Industry, 1977. 14.13 Plywood production is split between a group of 11 "large" factories averaging about 6,500 m3 p.a. (the largest capacity is 9,000 m3 p.a.) and 9 small-scale units averaging 1,000 m3 p.a. Only one of the larger more modern factories, located in Ankara, is in the public sector; the remainder, including the small-scale more primitive factories, are privately owned. Thus, the private sector is dominant in plywood as it is in sawn wood. The industry is located in the centers of consumption and is virtually uninte- grated. The only exceptions are a public firm in Ankara and two of the larger private enterprises, which are integrated with particle or blockboard plants. Location away from the sources of supply and lack of integration result in inability to produce high quality plywood due to damage in the logs transported over long distances and the lack of proper storage facilities, increased cost of raw materials, and high rate of waste. The smaller plants employ antiquated equipment and production techniques which, coupled with the irregular supply of raw materials, result in the low productivity and capacity utilization -- around 50%. There are 4 plants producing fiber board ranging in (single shift) capacity from 5,000 m3 to 9,500 m3, three using the wet process and one the dry process. These plants, however, are operating on a 3-shift basis in order to meet the annual demand of 77,000 m3.  - 106 - Market Aspects 14.14 In Turkey, as has occurred elsewhere, wood-based panel production is growng much more rapidly than sawn wood production. Table 5 illustrates Table 5: MARKET SHARES OF SAWN-WOOD AND WOOD-BASED PANELS, 1965-1975 (Percentages) Sawn Fiber Particle Year wood Plywood Board Board Total 1965 97.3 1.4 0.9 0.4 100.0 1975 90.1 1.3 1.4 7.2 100.0 the trend to substitution of wood-based panels for sawn wood in 1965-75, while Table 6, within this market-shift pattern, shows the strong bias in favor of particle board. Table 6: RELATIVE SHARES OF WOOD-BASED PANELS, 1965-1975 (Percentages) Fiber Particle Year Plywood Board Board Total 1965 53.8 34.1 12.1 100.0 1975 14.0 14.1 74.0 100.0 14.15 Sawn wood and the three basic types of panels are in large measure interchangeable in specific end uses. However, for various reasons (price, aesthetics, ease of handling, etc.), end-use industries tend to favor one over the others. Thus, 80% of particle board output is used in the furniture industry and 20% in construction as internal wall panels, doors and ceilings. The comparable split for fiber board is 70:30, although much smaller in absolute terms than particle board. Past consumption and projected demand are summarized in Table 7. In comparing actual consumption figures with  - 107 - Table 7: WOOD-BASED PANEL CONSUMPTION, 1965-1983 (000's m3) Particle Fiber Year Board Board Plywood Total 1965 6 17 27 50 1967 22 30 34 86 1969 45 38 55 138 1971 61 38 38 137 1973 152 40 45 237 1975 241 43 42 326 1977 463 77 43 583 1978 505 80 48 633 1979 (est.) 550 82 52 684 1980 (est.) 600 83 52 735 1981 (est.) 654 85 54 793 1982 (est.) 713 88 56 857 1983 (est.) 780 /a 90 /b 58 /c 928 /a, /b, /c Indicated growth rate of 9%, 3%, and 4.4% respectively, during 1978-83; mission's projections. installed capacity in the early- to mid-1970's it is clear that supply and demand were essentially in balance. 1/ With respect to particle board, the figures indicate that demand has continued to exceed supply. However, by 1980 the number of firms in operation reached 20 (up from 8 in 1976) with a rated capacity (3-shift, 300 days/year) of 821,000 m3/year. Certificates of Encourage- ment have been issued to 23 firms for an estimated capacity of 1.3 million m3/year, of which four plants are under construction with a rated capacity of 200,000 m3/year. 2/ As a result, production should suffice to meet demand growth until the mid-1980's. It is therefore unlikely that many other projects will be implemented in the very near future considering the projected growth of demand. The expansion of production of particle board should have a stimulative impact of the furniture industry (paras. 14.16-14.20) in two ways: (i) by meeting suppressed demand; and (ii) by putting downward pressure on prices which appear to be about double the European prices. The demand for plywood, on the other hand, is estimated to increase by a modest 4.4% per annum during the 1980's. The General Directorate of Forest Industries is presently building three new plywood mills at Kastamonu, Vezirkopru, and Bolu each with a rated capacity of 9,000 m3/year. With these forthcoming capacities, further invest- ments in plywood do not seem warranted. Finally, the plant under construction at Vezirkopru will also include 27,000 m3/year of fiber board capacity which meets domestic market requirements for the next several years. It is noteworthy that these appear to be export possibilities to the USSR and the Middle East of fiber board and particle board, which need to be exploited by the industry. 1/ No significant imports or exports appear to have been made. 2/ SYKB, Forest Products Processing Industry in Turkey, 1980.  - 108 - Furniture Industry State of the Industry 14.16 Furniture manufacture is an old tradition in Turkey. Hand-made fur- niture decorated with wood carvings or inlays is an important type of product still today. Most furniture falling into this category is reproduction of classic styles for the local market. Simple woodworking machines like circular saw, band saw, surface and thickness planes are used to separate components from lumber and to carry out the pre-machining. Mortising machines for paint- ing are common. All finishing, including wood carving, inlay, and sanding is done by hand. The final surface finishing is done either with shellac (manu- ally) or cellulosic lacquer (spraying). The number of similar products in one batch is small, usually only a few pieces. Production is very labor-intensive. Kiln-drying and seasoning of lumber is not widespread. In modern furniture, usually copies of Scandinavian or West European designs are used. Knockdown constructions which are necessary in export are very little known. The modern type of dimensioned working drawings in reduced scale are not used in the factories. This makes the machine set-up difficult and lowers the dimensional accuracy. Machinery and equipment in some factories is very good. Typical problems are: unsuitable work places and poor plant layout; scrap wood col- lecting leaves much to be desired; chip and dust extraction network exists only in few factories; tool and machine maintenance is not properly carried out; quality control is not systematically organized. Also there is scarcity of qualified designers who could create designs in accordance with the needs of foreign markets, and who could also help reduce transport costs to foreign markets. Finally, the quality of accessories (e.g. hinges, handles, locks) is poor. Annex G summarizes the problem areas and the type of assistance required. Organization and Performance 14.17 There were 106 establishments employing 10 or more workers, or 1.7% of the total number of establishments, with total employment of 2,800 (Appendix Table 3.12) and average employment of 26 workers in 1970. According to the Annual Manufacturing Survey of 1976, the number of establishments employing 10 or more workers stood at 68 and total employment at 1,970 (Appendix Table 3.13). Average employment was 29 workers in 1976. Since the bulk of the 10+ enterprises employ 10-15 workers, the observed decline in the number of larger establishments between 1970 and 1976 probably reflects a slip to the below 10 worker group or elimination of some of these more marginal producers. Invest- ment during 1970-76 increased from TL2.5 million to TL31 million, which coin- cided with the appearance of a few (though still small in size) factories employing 25-100 workers, mostly through the expansion of smaller establish- ments. Gross product value in 1976 was TL354 million (US$22 million). In general, furniture factories are typical SSI units employing 2-10 workers, 1/ 1/ The mission visited one furniture factory on an estate in Izmir employ- ing 35 and using modern machinery; however, this unit was atypical.  - 109 - housed in old congested quarters (except for those operating on industrial estates), using simple tools and machines and catering to neighborhood markets only. They are clustered in and around eleven of the largest cities where demand is most concentrated; of these, the three most important centers are Ankara, Istanbul and Izmir. There were 6,046 furniture (including fixtures) establishments in the unorganized sector in 1970, with total employment of about 13,600 (Appendix Table 3.11). There may have-been some attrition in the number of establishments employing less than 10 workers in the 1970s, through the disappearance of inefficient producers. 14.18 Exports in 1978 were US$4.6 million, up from US$2.6 million in 1967, and were destined mainly to Middle East countries. The 1978-83 Plan projects exports of furniture to reach US$16 million by 1983, which is feasible. Two factors appear to be instrumental in promoting the modest, but still signifi- cant, increase in Turkish furniture exports. First, a number of Turkish engineering civil works contractors have won contracts in nearby Arab coun- tries and as a consequence have been able to foster furniture exports. Secondly, a relatively large private firm -- ORSAN, employing 200 workers -- has been successful in developing export sales through sophisticated inter- national marketing techniques. Not only has it exported its own products but also those of other furniture factories for which ORSAN functions as inter- national marketing consultants and agents. The success of such private initiatives is attributable in part to Government policy support in the form of tax rebates on exports of 30% value (formerly as high as 75%). Such a high level of subsidy is necessary because of the high price of particle board (para. 14.14) which cannot be totally neutralized by low wages. 1/ Policy Recommendations 14.19 The forest resources of Turkey, properly developed and managed, are sufficient to support a major expansion in the efficient production of logs and the intermediate and finished products derived from logs. As a first step, the General Directorate of Forestry, the monopoly producer, needs to take steps to expand production and improve its marketing system. The ensuing benefits will be felt by the down-stream sectors (sawn wood, panels and furniture) as a whole -- but especially so within the important SSI segments of these sectors -- in terms of raw material availability, quality and price. 14.20 The importance of the SMI sectors in the sawn wood, wood-based panel and furniture industries is self-evident. Any organized program of sectoral expansion and operational improvement must be based not only on assisting the medium to large scale firms but also on strengthening the SSI segment. SSI growth and efficiency can be promoted through increased credit availability and technical assistance. Mergers and cooperatives 1/ Prices are high not only because of overpriced wood (para. 14.09) but also because of the very high cost of urea-formaldehyde resin which represents about 30% of the cost of panels.  - 110 - should also be encouraged to eliminate or reorganize very small marginal producers in favor of the scale economies achievable by larger (but still small-scale) production units. In the marketing area, ancillary linkages and subcontracting can be fostered through a subcontracting exchange to take advantage of the distribution capabilities, nationally and inter- nationally, of larger manufacturing and trading companies. In view of the export potential of the furniture industry, an in-depth study of this industry is warranted.  - ll - XV. THE GARMENT INDUSTRY Organization 15.01 Textiles is predominantly in private hands. In 1976, the organized textile sector (including clothing) accounted for 20% of total manufacturing investment, 16% of production and value added, 45% of exports, and 24% of employment. The public sector comprises 44 large scale enterprises under Sumerbank's control and accounts for 19% of the investment in the textile sector, 14% of production, 17% of value added, and 20% of employment. In 1977, there were some 1,400 textile establishments in the private organized sector, up from 950 in 1970, employing close to 200,000 workers, or about 50% above the 1970 level. In addition, there were an estimated 8,000 weaving establishments in the unorganized sector 1/ operating some 19,000 looms, or 27% of the total weaving capacity compared with 59% of the organized sector, and employing probably another 80-90,000 persons. In the organized cotton sector probably 155,000 are employed in spinning, weaving, and finishing, and about 40,000 in the garments/making-up and knitting segments. A considerable part of the sector (one-third) is well on its way toward integration (spinning/ weaving/finishing); but only a very small number of these integrated firms extend to making-up. About one-half of the textile enterprises in the orga- nized sector are located in the Istanbul area. 15.02 Unlike the other segments of textile sector, the garments/making-up industry is undeveloped, in the sense that about 80% of total production takes place largely in households and in the unorganized sector, by units mostly of less than 4 employees. Subcontracting is widespread in the garment industry, as many small firms with sales outlets complement their production by asso- ciating themselves with a number of very small producers, supplying them (wholly or partly) with the requisite raw materials and/or advancing part of working capital requirements. Survival of the multitude of small units in many instances depends on these subcontracting arrangements, in part due to their inaccessibility to short-term credit. Garment making is largely concen- trated in Istanbul--with more than 80% of the organized sector firms, including most of the largest ones. Ankara and Izmir account for perhaps an additional 12-14% and the small remaining balance is found mostly in Bursa and Adana. The Government's policy to decentralize development has succeeded in promoting new garment firms in the cities of less industrialized areas such as Edirne 1/ Most of these establishments employ up to 4 workers. A "quasi modern- ization" process goes on in the unorganized sector, in the sense that scrapped equipment by modernizing firms is bought at very low prices and used, particularly in the less developed areas, to replace more antiquated equipment. The individual circumstances of the entrepreneur (e.g. working together with his family long hours, nominal overhead) and local conditions (e.g. disregard for quality, low wages) appear to make possible an extension of the economic life of the otherwise "out-of-date" equipment and the convenient supply of local markets with products in demand.  - 112 - and Trabzon but the potential for such development is limited principally to standardized, relatively low fashion items in which rapid communication and transport access are less important or where the volume of production can justify services, such as fashion designing, production engineering, and marketing, on an "in-house" basis. Factory production of men's garments is relatively more important than women's garments because they are more standard- ized, but penetration of the total market is probably 15-25%--at most. The organized sector has been fast growing--from only 25 registered establishments in 1963 to an estimated 600 at the present time, 1/ employing some 40,000 workers 2/. Two-thirds of these establishmetns employ 10-20 workers, and the balance 20 or more, with only 10 employing over 100 workers. 3/ Estimates about the number of establishments and employment in the unorganized sector are more unreliable. TSKB reports some 4,400 units employing about 23,000, based on social security records. However, given that many establishments are not registered with the social security for a variety of reasons, actual figures are probably a multiple of those recorded. Production and Demand 15.03 Estimates of capacities and production levels of various types of garments in the organized sector in 1976 are shown in Appendix Table 6.2. Despite problems in procuring various inputs, capacity utilization ranges from 70%-89% which, under the circumstances, is satisfactory. Appendix Table 6.2 also provides estimates of domestic demand. The observed excess demand for most of the items listed was satisfied mostly by the production of the un- organized sector and marginally by imports, given the very low--only 10%-- penetration of the domestic market by the organized sector as illustrated in Table 1. 1/ TSKB's estimate from study: The Textile Sector, 1978. TSKB's estimate is well above that of the Statistical Service (119 establishments in 1976) and is based on information from the Clothing Manufacturer's Association, according to which the number of establishments is 480. This estimate is then revised upward to include many new establishments in the 10-25 worker range which apparently have failed to register. The public sector, through the Summerbank complex, is also involved in the production of garments, but accounts only for 2% of total production and employment. 2/ This figure is substantiated by the fact that, according to TSKB's estimates, the sewing machine pool in the 600 establishments was about 25,000 in 1976. 3/ See Appendix Table 6.1. The Annex also provides information on the distribution between knit and woven garments.  - 113 - Table 1: EXTENT OF PENETRATION OF DOMESTIC MARKET BY ORGANIZED SECTOR /a lb Domestic Market- Production - Penetration (Thousand Tons) (Thousand Tons) % 1977 Woven Garments 133 5.5 4 Knit Garments 42 10.5 24 Household Articles 86 8.0 9 2I T --g-_ 1983 Woven Garments 153 13.4 9 Knit Garments 58 18.1 31 Household Articles 102 12.7 12 313 44.2 14 /a Organized and unorganized sectors. 7F Organized sector. Source: Staff Appraisal Report: Project to Promote Modernization, Efficiency, and Export Capability of the Private Textile Sector, No. 2525-TU, 1979. However, it can be expected that the factory production of garments in the organized sector will continue to erode the large share of the clothing market now held by the less efficient home and artisan production. With the rapid urbanization and industrialization that is taking place in Turkey, and the economies of labor input that organized production of garments ensures, growth of the domestic market will continue at a fast pace in spite of a slowdown in overall economic activity. During 1977-83, domestic demand (in volume) is projected to increase at an annual average rate of 3%, while production in the organized sector by 11% annually (Table 1 and Appendix Table 6.3), with the greatest part of the increase in demand to be met through the expansion of the organized sector. The average penetration of the domestic market by the organized garment sector is projected to rise to 14% by 1983, a gain of 4 percentage points over the 1977 position. The foreign exchange requirements for equipment to meet domestic and export demand during 1977-83 are estimated at US$98 million (Appendix Table 6.4). Such an investment would create about 24,900 new jobs at a cost of US$4,000 per job created, which is very low. Part of the required investment will be financed by the Bank's loan to TSKB and SYKB. Wages 15.04 In 1976, the average daily wage was TL 86.3 (or $5.18 at the then exchange rate). This represents only 75% of the average wage prevailing in manufacturing and can be explained by the predominance of female employment in garment/making-up. Although high, 18.9% annually between 1969 and 1976,  - 114 - wage increases in the garment sector have been somewhat lower than in the other textile activities (20.0% per annum). Real wages rose by 2.2% per annum during 1969-76. Part of the increase seems to have been absorbed by the manufacturers, as wholesale and retail price increases of textile products were lower (18.3% and 16.3% per annum, respectively). Still, labor costs in Turkey compare favorably with those prevailing in all European countries, even if lower productivity levels are taken into account. However, Turkish labor costs are well above those of Brazil, Korea, Egypt and Pakistan as shown in Table 2. Table 2: LABOR COST PER OPERATOR-HOUR IN THREE-SHIFT IN TEXTILE PRODUCTION IN SELECTED COUNTRIES, 1976-1979 Country Labor Cost/Hour Country Labor Cost/Hour Belgium $8.27 France $3.76 Netherlands 7.28 U.K. 3.14 Sweden 7.20 Greece 2.23 F.R. Germany 6.90 Mexico 2.21 Denmark 6.68 Turkey 1.46 Norway 6.21 Portugal 1.45 Italy 5.15 Brazil 0.92 Canada 4.92 Korea 0.45 U.S. 4.40 Egypt 0.36 Pakistan 0.28 Source: Werner International, Report on the Portuguese Textile and Make-up Industries, prepared for the World Bank, June 1977, p. 111. This underscores greater reliance on other non-labor competitive advantages (e.g. proximity) and need for sustained growth of productivity, if Turkey is to capture a share of the world market. In this regard, the steep wage increases across industry in recent years and fast rising fringe benefits may erode Turkey's export possibilities irreversibly and should be followed closely (see para. 15.07). Exports and Export Potential 15.05 Garments and made-up articles represent a small, though fast grow- ing share of Turkish exports. Starting from token levels in 1970, exports increased to 3,000 tons or US$26 million in 1974, to reach 5,500 tons or US$60 million in 1977 (Table 3), which is 26% of the sector's export value. Since Table 3: GARMENT EXPORTS FROM TURKEY, 1974-1977 (in tons) 1974 1975 1976 1977 Knitted Garments 158.6 109.1 381.2 1,480 Woven Garments 1,028.5 1,149.3 3,505.2 2,373 Sub-total 1,187.1 1,258.4 3,886.4 3,853 Household Articles 1,221.3 2,633.5 1,556.0 1,638 Total 2,408.4 3,891.8 5,442.0 5,491  - 115 - 1977, exports have stagnated due to the growing domestic demand and the uncer- tainty about export opportunities following the import restrictions imposed by the EEC. 1/ Nevertheless, there is potential for further expansion of garment exports within existing market constraints. 15.06 By contrast with textiles, there have been only modest changes in production technology for clothing; also, labor productivity in developed countries grew only at 0.9%/year during 1965-1974--less than in any other major group of industries. Turkey's lower wages, therefore, afford Turkish clothing industry a clear advantage over developed countries, provided that productivity approaches European levels, except in product lines where fre- quent fashion and style changes require small units in immediate proximity to markets. Its overland accessibility to Europe allows transit times of two to three days at only 5-10% of total product costs--an imporLant advantage over many clothing exporting countries with far lower wages. Though exports of garments are still only about 5% of total production and knowledge about export requirements is generally sketchy, a nucleus of exporters has recently developed in response to Government export incentives and established links to European marketing channels. Such exporters have acquired practical experi- ence in exporting and are therefore cognizant of the export markets' stringent requirements for quality and timely delivery. Now organized into a Government- sanctioned Exporter's Union, the industry has a potentially effective institu- tion for concerted action to promote quality control, to intensify education and training, and to monitor export quotas. The Union has launched a trade fair promotional program and, in concert with the Ministry of Education, has recently established a Trade School for Garments in Istanbul. Provided that export incentives are set at a level to offset import duties on inputs, indirect domestic levies, and the adverse impact of lagging foreign exchange adjustments due to high rates of domestic inflation, and assuming availability of technical assistance in production, quality control and export marketing, the basis for a major, near-term expansion of this industry exists. 15.07 A major constraint to rapid export growth of garments is the current restrictions to trade, particularly in the EEC, which accounts for 70-75% of present Turkish exports by weight and about four-fifths in value. Virtually all exports of made-up articles from Turkey are classified in either the sensitive or highly sensitive categories. Restrictions have been imposed on four items which comprise more than 50% of Turkey's exports to the EEC, though they represent less than 3% of total EEC imports in the categories concerned. 2/ Even though neither industry nor the Turkish Government has 1/ A list of garments and made-up articles for which EEC proposes quantita- tive restrictions for Turkey in the medium term is shown in Appendix Table 6.5, while Turkey's share of EEC's imports and degree of penetration of Turkish exports into the EEC market in Appendix Tables 6.6 and 6.7, respectively. 2/ The restricted items are: knit shirts; woven's blouses; knit jerseys, pullovers, etc. for women; and bed sheets. Allowable growth rates are 1-1/2% year on blouses, 4%/year on all others.  - 116 - accepted the application of quotas to the garment industry, the EEC is likely to enforce growth rates of between 1-1/2% to 4% on the restricted items. Turkey's share of EEC imports on which no restrictions to Turkey have been placed averages about 0.7%. 1/ Growth rates of 24% per annum as envisaged by the Fourth Five-Year PLan, therefore, do not appear to be likely without major changes in EEC policies. A more conservative rate of 10% per annum, given Turkey's very low initial base, seems more realistic and unlikely to be viewed as disruptive by the EEC. Appendix Table 6.8 shows garment export projections for 1977-83. At a 10% rate of expansion, garment exports by the end of 1983 would account for more than a third of total textile exports. Sectoral Issues and Problems 15.08 Although the organized garments/making-up sector, with 80% of its firms founded in the last 10 years, has relatively modern equipment, its labor productivity on the average is estimated at only about one-half of European standards. There are a few recently established firms with technical assis- tance from marketing partners whose standard times for typical relatively high volume items such as men's trousers are nearly equal to an average modern producer in Europe; but most are far below. In part, this is due to short production runs resulting from lack of specialization. It is not uncommon to find firms producing both men's and women's wear and from both knit and woven fabrics. It is also the result of serious and pervasive deficiencies stemming from unsuitable and cramped premises, poor plant layout, organization and work methods, lack of cost controls--deficiencies which not only lower productivity but quality as well. Furthermore, there is a shortage of skilled workers, especially technicians, high turnover in part due to high female participation rates, low level of mechanization, poor utilization of labor and machines, lack of incentives for harder work, absence of standards on which to base financial controls, lack of knowledge of modern management techniques by the great majority of managers-owners, and lack of concern for the workers. There is, therefore, scope for considerable improvement in efficiency for this segment, primarily in areas relating broadly to industrial engineering and production management, as shown in Annex G. More attention should also be paid to personnel management. Financial institutions can play an important role by brining to management's attention such deficiencies in performance, by advising them or referring them to technical assistance delivery agencies, and by assisting with their own staff or in the hiring of experts. Another problem hampering efforts to improve quality in the unavailability of imported accessories (lining materials, zips, sewing thread, buttons, etc). However, the recently reinstated foreign exchange retention scheme (50% of f.o.b. export value) should alleviate considerably this problem. Also, measures to simplify procedures of imports for re-export and to remove hurdles to export 1/ This is high enough under current EEC guidelines to trigger possible additional quotas if sustained expansion took place. For highly sensi- tive products the threshold at which surveillance and possible quotas are triggered is 0.2% for EEC imports from "low cost" sources (developing, centrally-planned, and Southern European countries); for sensitive ones the threshold is 1.2%.  - 117 - licensing and registration, should facilitate exports. Finally, lack of access to financing, particularly of the smaller firms, results in borrowing from the curb market or in inferior bargaining position in sub-contracting arrangements and a squeeze of profits. Here also the role of financial intermediaries, and particularly of the Halk Bank, looms large. In this regard, availability of funds at reasonable collateral requirements deserves full consideration. Technical Assistance Requirements 15.09 In the Turkish garment industry most training is on the job not only at the operative level but also at the supervisor/foreman level as well. Training in basic management and production/industrial engineering skills is, therefore, usually inadequate and the bad practices that exist tend to be perpetuated. At the present time, the best single source for upgrading quality and improving productivity of garment making is the training and technical assistance provided by foreign marketing affiliates of some of the exporting firms. Not only do these affiliate firms give guidance in designs and pattern and levels of quality required for their specifications, but they typically provide a practical consulting and diagnostic service. These ser- vices are one of the important benefits that come from expansion of exporting but they are necessarily limited to firms that already have a basic capability to export and have been able to find a marketing affiliate willing and able to work with them. Their scope usually ensures only that output will meet buyer's specifications and Turkish firms necessarily pay for the services indirectly through price concessions or exclusive dealerships. These services are not, in any case, intended to provide broader types of training particularly in the industrial and production engineering field, which would enable garment technicians to solve many of their own problems rather than relying on outside consultants. In addition to the above, there are a textile high school in Istanbul operated by the Ministry of Education which graduates lower level technicians, and a recently established school under the auspices of the Ministry of Education and the Garment Exporters' Union to train garment operatives. 15.10 In light of the above, it is clear, that there is need for practical training and technical assistance at the middle management and supervisory level involving broadly technical and industrial/production engineering fields which is not presently supplied locally by existing institutions. The (pilot) technical assistance component proposed for the Bank's loan, with TSKB as a delivery agency, is a step forward in remedying this, by fostering a local capability to meet some of the needs. For the garment/making-up subsector the component provides for (a) a single industrial engineering consultancy, training and follow-up service, involving a more complete consultancy and training service than for textiles, because garment industry middle management is likely to be less specialized; and (b) an export marketing service tied closely to the Garment Exporters' Union, utilizing a resident marketing advisor supported by short-term specialists to help focus the Union's export promotional efforts and provide practical marketing assistance to individual  - 118 - firms. Hopefully, this pilot technical assistance will be expanded to increase its coverage, particularly for smaller units. Subsector Strategy 15.11 The Government is eager to expand exports of finished goods such as made-up articles, knitted goods, carpets, etc. To this end, investment in intermediate goods and auxiliary materials used as inputs in textile manufac- turing and in short supply will be encouraged. This approach is in the right direction. But there is also a growing potential to supply the domestic ready-made market through subcontract production for large volume distribution firms, such as chain stores. Basically, the strategy to achieve these objec- tives should be to improve productivity and to upgrade the quality of output in the industry. This would involve in particular: (a) modernization in certain segments of the sector (especially dyeing, printing and finishing where there is substantial obsolete equipment) to ensure the right quality of fabrics used in garments; (b) extension of technical assistance and training; and (c) fostering new entry, the growth of existing establishments to attain efficient sizes, and sub-contracting. 15.12 Structural changes are not difficult to achieve in this segment of industry, because economical production scales vary widely and hover within relatively low levels of output. By extension, scale economies can be achieved by medium size firms. It is estimated that, on average, and depending on the type of product, establishments employing 65 to 175 workers could operate at optimum capacity (Appendix Table 6.9). Also, being a highly labor-intensive industry, capital requirements are relatively much lower and, as a result, more affordable by less well-off industrialists. Promotion of cooperative arrangements, particularly with respect to procurement and marketing, creation of trading companies which could stimulate the development of an extensive network of sub-contractors and thereby link and help upgrade the performance of the small units, and encouragement of weaving and knitting establishments to integrate forward to garment/making-up, deserve full consideration. Encouragement of garment exports is an important objective not simply for increasing value added and foreign exchange earnings but also because of its impact on sectoral development as a whole. 1/ Through upstream linkages to the textile industry it transmits export standards for quality and price and therefore helps to upgrade the supplying industry's performance. As one of the first industries to sell a product involving more rigorous consumer-oriented standards of quality and delivery, it provides a valuable learning experience which for a country still in an early stage of export development, is broadly applicable to other industries as well. Finally, being one of the most promi- sing exporting industries and highly labor-intensive, its growth should expand urban employment. 1/ Faster growth of garment exports could come about if the industry is free to import fabrics when local fabrics are unavailable at a satis- factory price or quality.  - 119 - 15.13 The rising share of the organized sector can be explained on ef- ficiency grounds and the trend will undoubtedly continue. Home and artisan production would gradually fade out. Some of these "independents" will grow in size but the bulk most likely would have to seek "factory" employment. The major initiative in improving sector performance rests primarily with private entrepreneurs themselves. Nonetheless, development banks could play an important role in identifying and fostering priority projects, the reorganiza- tion of individual firms, promoting cooperative arrangements, and in delivering technical assistance. The Government's role should be directed primarily at improvement of the industrial policy framework, particularly for exports, but it can also help to ensure that the necessary training, technical assistance, and export promotion facilities are available.  - 120 - XVI. PROMISING SUBSECTORS FOR SMALL-MEDIUM SCALE INDUSTRY DEVELOPMENT 16.01 Subsectors which appear most promising in terms of potential growth of SMI comprise the foundry (Section XI), engineering (fabricated metal products, electrical and non-electrical machinery, transport equipment, and professional and scientific equipment - Section XII), food processing, ready- made garments (Section XV), footwear and leather products (Section XIII), wood-based and furniture (Section XIV), plastics, and construction materials. Aside from growing domestic demand and export potential, factors reinforcing the potential growth and viability of the SMI in these industries include: relatively low scale economies in a wide range of activities in each industry; locational factors, such as local processing of bulky raw materials (saw mills) or local production of bulky products (construction materials); comple- mentarity with larger industry, involving manufacturing operations in which the processes are readily separable (e.g. producing specialized machine products, components and tools), craft or precision handwork, simple operations of assembly, mixing, or finishing, provision of services and repairs to other industries; production of differentiated products for specialized markets, or low-priced mass consumption goods. Engineering is perhaps the most inviting of all subsectors, and its development should therefore command higher priority, as it combines virtually all the above mentioned features: it envelops numerous industrial intermediate and final goods which can be produced ef- ficiently at small-medium scale; it is relatively less capital-intensive and highly skill-intensive; entails relatively low capital requirements; has strong backward and forward linkages; demand is income elastic in many segments; there are substantial possibilities for replacement of imported parts and components; and it fosters a link between small-medium-large scale industry through ancillary or sub-contracting relationships. In general, all industries mentioned above are "non-process" and belong to the so-called "modern" SMI sector but, for a wide range of products, they require small-medium scale of plant (and size of firm) and low capital intensity. 1/ Their expansion should inter alia enlarge employment for the urban poor. Evidence from industrialized- countries (e.g. Japan and other countries of similar level of industrialization (e.g. Korea, Taiwan) reinforces their view. 16.02 In the absence of a sufficiently detailed data base and subsector studies, it is extremely difficult to forecast the investment requirements of the SMI. It can be expected that about 40% of the investment outlay in the next several years would be for replacement of old equipment probably with zero employment effect; 2/ some 55% would derive from expansion projects and 1/ The kind of activities suggested here are known to be dominated by SMI and expressely exclude the traditionally large scale, capital-intensive "process industries" (steel, aluminum, pulp, fertilizers, etc.). See also para. 5.13. 2/ It is noteworthy that replacement of obsolete equipment helps preserve existing jobs, which otherwise might have been eliminated.  - 121 - 5% from new entry, both with a positive employment effect 1/. The investment per job created for expansion and new projects would vary among subsectors but is likely to average around US$10,000. Thus, if it is only to maintain the historical level of employment generation (17,000 jobs a year), organized SMI would require about US$850 million (in foreign and local currency in 1979 prices) during 1979-83. However, the need for new productive employment opportunities at relatively low capital/labor ratios is much greater. In light of SMI's multi-faceted potential as outlined in paras. 16.01 and 17.01, support of SMI in the suggested judicious and eclectic manner is an efficient way to use scarce resources. Within the promising subsectors mentioned above, target groups could be identified by the financial intermediaries in the major urban centers, preferably from among those dynamic SMIs which have developed some degree of specialization in the management function (para. 6.05) and possess some of the features mentioned in para. 16.01--characteristics which are likely to enhance their potential for continued viability and self- sustained growth. 16.03 All subsectors and enterprise sizes, although to a varying degree, would greatly benefit from well-designed technical and finance assistance and infrastructure (industrial estates) programs, and the involvement of the Government and the Bank can have considerable impact on the development of the SMI (see Section X). However, an attempt to extend the scope of technical assistance to encompass a large number of establishments in all subsectors, though desirable, might be impractical at this initial stage. This implies that, aside from referrals to such public sector delivery agencies as the SIDO facilities in Gaziantep, TBTAK in Istanbul, PENDIK in Ankara, the National Productivity Center in Ankara, the proposed Business Advisory Service within the Halk Bank, and the pilot technical assistance scheme for garments to be launched by TSKB, there will be need for greater reliance on private consulting firms. The Halk Bank has undertaken to draw up a list of such firms which can provide assistance to its borrowers in the production engineering, marketing, administration and management fields. Similar referrals can be expected of SYKB. 1/ Since probably 30-35% of the required equipment can be produced locally, modernization and growth of the SMI would stimulate the development of the country's capital goods industry and thereby further enhance the creation of new employment opportunities.  - 122 - F. STRATEGY FOR THE DEVELOPMENT OF SMALL-MEDIUM SCALE INDUSTRY XVII. OBJECTIVES AND STRATEGY - A SUGGESTED APPROACH 1/ Contribution of SMI to Industrial Development 17.01 The importance of the SMI sector to the Turkish economy, in terms of employment, output, processing locally produced materials, and saving/earning foreign exchange, remains unequivocal. There can be no doubt that a modern SMI sector can contribute considerably to the country's industrial development, can promote structural shifts (e.g. by enhancing the share of intermediate and capital goods in industrial output), while active support of small-medium scale but viable manufacturing enterprises could also have other tangible socio-economic advantages (e.g. greater scope for initiative and the development of technical and managerial skills; supply of a good part of low-priced mass consumption goods; adaptation of production techniques to the country's particular circumstances; promotion of more equitable distribution of income; reduction of regional imbalances and urban concentration; etc.). Despite the problems enumerated earlier, 2/ SMI generally remains viable, is retaining its relative importance and is actually increasing it in many industries, as is evidenced by the noticeable increase in the number and size of firms in the last decade or so. 17.02 It should be stressed in this connection that scale economies, often viewed as a factor debilitating SMI efficiency, have been taken unwittingly 1/ For objectives, strategies and policies to facilitate the development of SMI in the subsectors reviewed in this report, see: foundries (paras. 11.16-11.18); engineering (paras. 12.22-12.34); leather processing and footwear (paras. 13.15, 13.27-13.30); wood-based (paras. 14.10, 14.11, 14.19, 14.20; garment (paras. 15.11-15.13). 2/ The range of problems in the SMI sector has been summarized in paras. 4.04, 4.05, Section IX, and Annex G.  - 123 - or uncritically for granted. 1/ Observations of the size distribution of many "non-process" industries in developed countries indicate that a large diversity of sizes exists at any one time and has persisted over time, suggesting that there is no unique optimum size for an "industry". This can be attributed to the multitude of individual activities in each subsector and the particular circumstances of each firm, as determined by technology and adaptability of equipment (e.g., use of multi-purpose machines), varying factor proportions (e.g. through longer shifts or more shifts), degree of vertical integration, managerial capability, structure of demand and product- mix, factor prices, location, or particular stength compensating for dis- advantages in other areas (e.g. greater flexibility and adaptability to market changes). In view of the uniqueness of the surrounding circumstances, scale economies can be meaningfully ascertained only at the product level. Certainly there are, say, engineering products that require high levels of production and large size of firm. But the decision whether or not more or less capital-intensive processes are entailed, or whether larger or smaller scale is required can be resolved only by considering individual products and not on a priori determination. This implies further that minimum economical size and overall project viability (in economic and financial terms) should be established case by case through project appraisal by competent financial intermediaries. Moreover, even where scale economies are important (e.g. the automobile industry), this does not mean that the large firm produces every single part or component that goes into the final product. Normally, a major percentage of the components is subcontracted to smaller firms because, inter alia, they can be produced more efficiently. Urban Employment Generation 17.03 In view of SMI's multi-faceted potential as outlined in paras. 16.01 and 17.01, support of SMI in the suggested judicious and eclectic manner 1/ Optimal plant scale arguments in "non-process" industries tend to be specious. It has been demonstrated that manufacturing costs often make up only about one-third of the final cost (which includes distribution and overhead) and that the total production cost curve often becomes flat after a relatively small volume. Even then, the cost penalties asso- ciated with building plants that are one-third of the minimum economical size can result in unit cost increases of about 10%. In general, the size of plant required in most industries is not very large (this is particularly true for consumer goods industries--SIC Number 2000-3900) as is evidenced in Appendix Table 3.30, which shows the distribution of manufacturing plants in the U.S. by employment size in 1970. The distri- bution is highly skewed: 88% of all units employ fewer than 100 persons, although those same units in the aggregate employ only 24% of all manufac- turing employees. The largest plants tend to be concentrated in certain industrial sectors, involving large assembly-line operations (e.g. automobiles and heavy electrical equipment). B.A. Stein and M.B. Hodax, Competitive Scale in Manufacturing--The Case of Consumer Goods, Center for Community Economic Development, Cambridge, Mass.  - 124 - is an efficient way to use scarce resources. In devising a strategy and the requtisite policy framework for SMI development, two major objectives, with divergent implications for job creation and use of capital, stand out: (a) employment generation in the urban areas, an arduous task whose accomplishment is heavily conditioned on economizing capital in the process of industrial investment (para. 2.05); and (b) alleviation of the problems constraining the growth and development of SMI, the latter amounting essentially to concerted efforts at modernization and improving efficiency, involving capital outlays with virtually no immediate impact on employment generation. Unemployment, including underemployment, in Turkey is reportedly high. In addition, the manufacturing sector is expected to absorb 30%-40% of the projected increase in the labor force in the next five years, or some 75,000 workers a year, and probably more so subsequently. In the light of these circumstances, it would appear that creation of employment opportunities in the face of a youthful and growing population and urban labor force becomes a singular and overwhelming issue. Emphasis on relatively less capital-intensive and more on skill- intensive technologies, to a considerable extent suitable for SMI, should greatly serve both objectives. A policy framework to foster urban employment, including the decentralization of industrial location, was suggested in paras. 2.05-2.08. 17.04 The strategy should be to achieve an appropriate balance between larger and smaller (but efficient) establishments, more and less capital- intensive techniques, in accordance with the country's factor endowment--a symbiotic relationship in which the activities of small and large firms conform to their respective comparative advantages 1/. It is noteworthy that in the 1960's the proportion of employment in establishments with 200 or less workers was 51% in France, 53% in Sweden, 54% in Japan, 58% in the Netherlands, 61% in Switzerland, and 66% in Italy. 2/ This attests to the viability, complementary role in industrial development, and fitting position of SMI in every country's industrial structure. The historic lesson is that a bimodal industrial structure composed of mutually supportive large and small firms has worked, has become highly productive, and is economically and socially desirable. 17.05 For the optimum large/small, capital/labor-intensive mix to evolve, however, policies to avoid or minimize biases that tend to distort the alloca- tion of resources (e.g. that underprice capital equipment, social measures that overprice labor) would not suffice. Since large firms possess certain inherent advantages by and large unrelated to their economic efficiency (e.g. they are more immune from risk failure, have more market influence and bargaining power, have readier access to Government and financial institutions), there is need for Government intervention and assistance in several forms, such as access to institutional credit, extension of technical assistance, promotion of coopera- tion between small and large enterprises (e.g. through sub-contracting or marketing arrangements, etc.). 1/ The issue is not promotion of small versus large establishments, or indiscriminate support of SMI for the exclusive purpose of alleviating unemployment at any cost. -2/ For other major developed countries, see Appendix Table 3.31.  - 125 - Efficiency Improvement and Growth 17.06 An equally important objective is the need to improve efficiency and to foster the growth of SMI. This can be achieved primarily through modernization of equipment; improvement of the performance of small-medium manufacturing establishments through well-conceived and properly tailored financial and technical assistance programs, including skill upgrading and accelerated training; 1/ and by inducing structural changes (e.g. cooperative arrangements and consolidation of smaller enterprises as appropriate), export orientation, and adaptation to new circumstances, rather than protecting inefficient firms and out-of-date methods of production from the competition of more modern methods. Industrial development in Turkey has progressed to the point where future growth of SMI depends on increasing adoption of modern efficient production processes, specialization in production, improved product designs and higher product quality, modern management methods, and developing closer linkages with larger manufacturing firms either as sub-contractors or ancillaries for the provision of parts, components and sub-assemblies or with large scale trading establishments serving either the national or export markets. 17.07 In light of the above, the introduction of more sophisticated and specialized equipment at this stage of Turkey's development should not be discouraged. However, this does not mean to imply an indiscriminate choice of highly capital-intensive techniques and equipment. Given the scarcity of capital and foreign exchange and the comparatively lower wages in Turkey by US or European standards, a conscious effort should be made to adopt, to the extent possible, appropriate technologies suitable to the local circumstances. Appropriate technology should be viewed rather as a complement to, and not a substitute for, modern technology. The concept basically implies utilization of technologies that economize on the use of scarce resources--capital, foreign exchange, highly trained personnel; that are intensive in the use of abundant factors--labor; and that they exploit domestically produced inputs-- cotton, wood, etc. It may also mean scaled-down plant and levels of output. Appropriate technology (= a set of inputs and processes) is premised on the existence of a spectrum of technologies either on the shelf or in the form of possibilities for adaptive development. Although appropriate technology could conceivably be imported, the emphasis is on building up an indigenous tech- nological innovative capability. To ensure diffusion, appropriate technology options should display engineering efficiency, economic viability, and social acceptability. This would involve for the Government: (a) to allow factor (i.e. capital and labor) prices to reflect their true costs; and (b) to take upon itself the task of ascertaining, promoting, demonstrating and disseminating appropriate technologies. Financial intermediaries could also play a useful role in guiding prospective borrowers. In this regard, the Ministry of Industry and Technology, properly strengthened and assisted, could be assigned the task to advise on matters of technology, inter alia, in the course of approving Certificates of Encouragement and Letters of Encouragement. 1/ The youthful structure of the labor force should imply greater occupa- tional and geographical mobility and readiness to acquire higher skills.  - 126 - 17.08 The transformation and growth of the small workshop into modern factory is important for the dynamic development of the industrial sector. I/ Support to SMI would enable the most efficient and dynamic enterprises to grow in size and to graduate to higher size brackets which are now thinly populated. It is notable that much of the increase in the size of SMI establishments in Turkey, as well as in other countries, derives from the internal growth of enterprises attained in phases of successive expansions and less from mergers. Although mergers and other cooperative arrangements should be encouraged, every attempt should be made to facilitate the growth of the more dynamic and promising SMIs. Much small firm growth is likely to result from second generation owner-managers as they succeed the family firm, which is a continuing process. The latter, contrary to their predecessors with an "artisan" identity, are more likely to have a conventional middle-class background and education and, therefore, a more modern "managerial" view of economic activity and acceptance of the dominant business ideology which stresses growth and efficiency. Imbued with such principles, they are likely to be more receptive to such ideas as the expansion of the firm, the professionally managed business, external borrowing; would probably display less strong element of paternalism which minimizes dependence on others inside or outside the firm; and would be more appreciative of the administrative necessity to use more consultative leadership style to delegate and to negotiate with professionally trained people brought in from outside. 17.09 The effort to develop SMI should have an export-orientation bent. The presumption seems to be that exporting firms should be large, entrenched entities. However, in the light of Turkey's own experience in garments and leather, this is not a condition sine qua non. The evidence in Korea also indicates that many of the presently large exporting firms were small not many 1/ It is interesting to note that in the present state of development of the Turkish economy, with a well-developed transport system and the presence of modern advertising, local markets are no longer as isolated as in the past and are increasingly becoming integrated with the national market. Under such circumstances, small enterprises relying on small protected local markets face an uncertain future. These are industries where relative efficiency increases with firm size and are found among pro- cessors of dispersed resources and service industries (wheat milling, beverages, wooden furniture, etc.). Some will continue to flourish in industries where transport costs are large (building materials, food processing, etc.) or goods are perishable. Small firms will also con- tinue to predominate in simple assembly, mixing or finishing operations, since in these industries smaller firms likely to be equally productive compared to large firms. However, in the course of the industrialization process, the competitive base of small firms tends to shift to separable manufacturing operations (engineering industries is perhaps the best example), as the experience in other rapidly industrializing countries (e.g. Korea, Taiwan) indicates. Moreover, very small firms (with less than five workers which constitutes the bulk of small establishments in Turkey) tend to move out of manufacturing operations into repair and service type activities--unless they are able to graduate into larger size categories.  - 127 - years ago and that direct exports (i.e. not through intermediaries) by medium size firms have grown very rapidly, particularly in such items as textiles/ garments, leather products, and shoes--apparent areas of Turkish comparative advantage. Of course, efforts to develop trading companies which would act for a collectivity of SMIs or would sub-contract to SMI for exports would be highly desirable. 17.10 In essence, the strategy should be to direct private investment and to extend support 1/ primarily to SMIs with relatively high probability of survival and success and conducive to promoting the gradual integration of the SMIs with the rest of the industrial sector. Viability and self-sustained growth of SMI are most likely to be attained, it would seem, by focusing on industrial activities with promising growth potential and characterized by rather low economies of scale, catering to specialized or low-priced market niches, or displaying complementarity with activities undertaken by larger firms in the form of ancillary or sub-contracting arrangements, or provision of services. At the enterprise level, the establishment with a realizable potential for growth, and consequently for continuing viability, is likely to be the one in which there is some degree of specialization of management functions. Such a conscious approach, effected through the appraisal and lending procedures of financial intermediaries, would ensure acceptable economic and financial rates of return on investment, that SMIs fill inter- stices, command an interlocking position within the industrial structure, and that they constitute vital parts of a functionally integrated whole. To this end, the pointers furnished in this report, hopefully, will help the responsible authorities to work out realistic solutions and to devise suitable policy instruments. I/ SMI's role should be perceived as complementing that of the large private enterprises. Therefore, extension of technical and financial assistance and elimination of policy and institutional biases, should in effect be construed as putting the SMI on an equal footing with the large enter- prises and as establishing a policy framework providing for equal oppor- tunities.  - 128 - G. RECOMMENDATIONS XVIII. SUGGESTED ACTION Policy Measures 18.01 In the past, measures supportive of SSI were prompted primarily by social considerations (e.g. concessionary loans to industrial estates and cooperatives). The Government now may wish to take a more positive, develop- mental attitude and to begin addressing more vigorously problems constraining the gowth of SMI, by devising and implementing realistic policy measures. An array of issues that need to be acted upon have been identified and suggestions for the Government's consideration as to how they could be approached have been made earlier. Certain measures, mostly within the domain of the adminis- trative decision-making process, which command relatively higher priority are summarized below for convenience. Data Base (i) Impove the system of data generation, processing, and analysis by: (1) devising a more operational definition of SMI in the light of present realities (para. 1.02); (2) build up the requisite data base on SMI at a sufficiently disaggregated level, and ensure a constant flow of statistical information (para. 1.03); 1/ take a well-designed census of the SMI to estab- lish the present status of the sector (para. 1,03, 2.03, fn 1). Employment Generation (ii) To accelerate the rate of job creation in manufacturing: (1) foster the development of viable SMI, which is character- ized by relatively low capital/labor ratios (paras. 2.05, 16.01, 17.01-17.04); (2) adjust factor prices to reflect better the relative scarcities of capital and labor to facili- tate the adoption of appropriate technologies, by discontinuing interest rate subsidies and the exemption of imported equipment from duties (or at least confine such incentives to projects which demonstrably have significant employment effects), and by exercising circumspection in setting minimum wages (para. 2.06); and (3) assess and redesign the existing incentives in order to foster low-cost private investment to the small and intermediate urban centers where the incidence of unemployment is higher (para. 2.07) 1/ Among other data, the compilation, for instance, of information on mortality rates of enterprises (by industry, size, years of operation, cause of demise, etc.) as well as on new entrants could help identify . patterns of SMI behavior.  - 129 - (iii) To foster the adoption of appropriate technologies, in addi- tion to adjusting factor prices as suggested in (ii) above: (1) strengthen the Ministry of Industry to advise on matters of technology, inter alia in the course of approving Certifi- cates of Encouragement and Letters of Encouragement; and (2) encourage financial intermediaries to inform and guide prospective borrowers (para. 17.07). Export Promotion (iv) Restructure the present export incentives system to provide greater rewards to superior export performers; modify eligibility criteria to include smaller firms, provided they combine to reach a minimum absolute value of exports; extend the incentive to intermediate suppliers of exporters; maintain a realistic foreign exchange rate and review periodically the level of export rebates and incentives to ensure that they are maintained at competitive levels for efficient producers (para. 8.04). (v) Improve the effectiveness of export credits by (1) making them available more expeditiously and on a constantly replenishable basis rather than on an "annual cycle"; (2) making them more accessible to smaller exporters; and (3) allowing exporters to utilize local banks to process applica- tions to minimize contacts with Government agencies (para. 8.05). Promotion of Subcontracting (vi) To promote ancilliary and sub-contracting arrangements between SMI and large engineering firms, particularly in the automotive and consumer durable fields: (1) institute appropriate measures to ensure standardization of models and components (para. 12.31); (2) direct the Standards Institute to develop product standards and to grant a quality seal of approval (para. 12.32); (3) develop a system of foreign exchange allo- cation based on the contribution to value added not only of the final exporter but also of the indirectly contributing suppliers (para. 12.33); (4) instruct the Chambers of Industry to devise a workable scheme for allocating imported raw materials between large firms and ancillaries (para. 12.33); (5) adopt measures to secure the involvement of the public sector as suggested in para. 12.34. Miscellaneous (vii) In the sawn-wood industry, take steps inter alia to increase the availability of raw materials to the private sector (paras. 14.05, 14.10, 14.11, 14.19, 14.20).  - 130 - (viii) Devise appropriate measures: (1) to encourage investment and commercial banks to increase financing to medium scale manufac- turing enterprises, particularly in smaller regional urban centers (para. 2.07); and (2) consider ways to increase the supply of equity capital, including the advisability of setting up leasing/hire-purchase schemes which reduce SMI's capital requirements (para. 4.05). (ix) Project requirements of vocational skills as well as needs for skill upgrading through the 1980's, and expand existing programs or institute new ones to meet these needs (see paras. 6.09, 7.15, 9.06). Technical Assistance 18.02. At this stage, the Government's efforts should concentrate on improving the effectiveness of the pilot program at Gaziantep and the estab- lishment of 1 or 2 regional extension offices, along the lines suggested in paras. 7.11-7.13. Studies 18.03 It is suggested that the following studies be commissioned. (a) In view of the need to foster the reorganization of very small producers by encouraging them inter alia to establish coopera- tives, SPO should take the lead in commissioning a study which (1) would assess the present system and identify its short- comings; and (2) would make specific recommendations as to how the proposed reorganization could be approached and implemented. The study should also consider other complementary measures that could help reorganize the SSI sector (e.g. mergers), including the requisite policy framework in the light of Turkish circumstances. The potential role of financial inter- mediaries should also be examined. (b) Commission carefully designed studies to identify capital goods and high technology items that could be produced by SMI (paras. 12.25, 12.28). The Marmara Institute for Science and Technology could be entrusted with the task under the direction and support of SPO and the Ministry of Industry and Technology (para. 46). (c) Undertake a basic study of the footwear industry, under the auspices of TSKB/Sumerbank/Ministry of Industry, to the end of formulating an integrated national program of rehabilitation and reorganization of the sector, and preparing a comprehensive Action Plan (paras. 13.27-13.30). The study should also suggest a course of action concerning the restructuring of the leather tanning segment of the industry, including realistic options for possible relocation (para. 13.09).  - 131 - 18.04 TSKB undertakes -industry studies in connection with its lending operations. It would be useful if such studies paid greater attention to the potential and problems of SMI, as appropriate. In general, such studies should take an holistic approach. in the sense that they should examine the potential role of firms covering the entire spectrum of sizes, i.e. large, medium, and small, promoting the development of the industry under review, and should also suggest ways to forge links among them.