Argentina - Structural changes in the industrial sector (Vol. 2 of 2) : Annexes

Report No. 1977a-AR Zlri TO LAs U FILE COPY Argentina It ETi, CEE Structural Changes in the Industrial Sector (In Two Volumes) Volume II: Annexes March 30, 1979 Projects Department Latin America and the Caribbean Regional Office 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 On January 15, 1979 the market rate for the Argentine peso (a$) was: US$1.00 = a$ 1,026 a$1,000 = US$0.9747 LIST OF PRINCIPAL ABBREVIATIONS SEDI - Secretaria de Estado de Desarollo Industrial (Secretariat for Industrial Development) INTI - Instituto Nacional de Tecnologia Industrial (National Industrial Technology Institute) BANADE - Banco Nacional de Desarollo (National Development Bank) DGFM - Direccion General de Fabricaciones Militares IDB - Inter-American Development Bank YPF - Yacimientos Petroleras Fiscales PBB - Petroquimica Bahia Blanca SOMISA - Sociedad Mixta Siderurgia Argentina INDEC - Instituto Nacional de Estadistica y Censos  FOR OFFICIAL USE ONLY ARGENTINA ANNEX I TECHNICAL DEVELOPMENT AND TECHNOLOGY EXPORTS TO OTHER LDCs TABLE OF CONTENTS Page No. I. INTRODUCTION I II. EXPORT PERFORMANCE AND LOCAL TECHNICAL DEVELOPMENT 4 III. EXPORTS OF COMPLETE PLANTS 5 IV. MACHINE TOOLS 9 1. Historical Background 9 2. Characteristics of the Industry Today 9 3. Imports and Exports 10 4. Technical Development 13 5. Managerial Development 19 V. AGRICULTURAL IMPLEMENTS 20 1. Historical Background 20 2. Characteristics of the Industry Today 21 3. Imports and Exports 21 4. Technical Development 24 VI. EFFECTS OF POLICIES ON EXPORTS AND TECHNICAL DEVELOPMENT 27 VII. CONCLUSIONS AND RECOMMENDATIONS 30 Promotion of Technical Development in Industry 34 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.   ARGENTINA: STRUCTURAL CHANGES IN THE INDUSTRIAL SECTOR VOLUME II ANNEX I TECHNICAL DEVELOPMENT AND TECHNOLOGY EXPORTS TO OTHER LDCs ANNEX II REVIEW OF THE ARGENTINE STEEL INDUSTRY Present Position, Problems, and Prospects ANNEX III SMALL AND MEDIUM-SCALE INDUSTRY   ARGENTINA: STRUCTURAL CHANGES IN THE INDUSTRIAL SECTOR VOLUME II ANNEX I TECHNICAL DEVELOPMENT AND TECHNOLOGY EXPORTS TO OTHER LDCs   STATISTICAL APPENDIX Table of Contents Tables: 1 Exportsof Relatively Sophisticated Manufactures as a Proportion of Total Industrial Exports. 2 Exports of a Selected Group of Manufacturers (1969-1974). 3 Exports of Complete Plants and Civil Works (1973-1977). 4 Exports of Complete Plants: Average Value of Plants Exported by Ownership of Firm. 5 Exports of Complete Plants by Client Sector. 6 Production,Imports and Exports of Machine Tools (1961-1976). 7 Machine Tool Producers, Size of Firms and Level of Production, 1971. 8 Exports of Machine Tools by Destination (1966-1975). 9 Latin American Imports of Machine Tools (1970-1975) 10 Argentina: Proportion of Locally Produced and Imported Machines in a Group of Machine Tool Producers. 11 Index of Physical Volume of Production of Agricultural Machinery and Implements (1974-1977). 12 Tractor Market Distribution by Companies (1970-1976). 13 Size Distribution of Agricultural Implement Producers. 14 Imports and Exports of Agricultural Implements (1956-1976). 15 Exports of Agricultural Implements by Destination (1973, 1975). 16 Latin American Imports of Agricultural Implements (1970-1975). 17 Argentina: International Comparison of Prices of Agricultural Machinery, 1977 18 Nominal and Effective Tariff Rates for Selected Products.  Graphs: 1 Trends in Production and Inventories of Agricultural Machinery 2 Trends in Prices of Agricultural Machinery in Relation to Prices of Agricultural Products 3 Sales Prices to User for Argentine Agricultural Machinery Employing Inputs at Local and International Prices  ANNEX I Page 1 ARGENTINA TECHNICAL DEVELOPMENT AND TECHNOLOGY EXPORTS TO OTHER LDCs I. Introduction 1. The objective of this annex is to explore the nature of the apparent success that Argentinian firms have had in developing and exporting appropriate technology to other LDCs. This exported technology is embodied not only in export products, but also in turnkey plants, foreign investment and licensing agreements with enterprises in other LDCs. 2. The analysis will be concerned with several related issues. First, the contribution of local technical development to export performance and the extent to which the historical circumstances of this contribution have deter- mined and will determine the conditions of supply. Second, the future trends in demand conditions (internal and external) and the extent to which these conditions are likely to be met by Argentinian producers, based on their tech- nical development in a manner congruent with the expected future demand conditions. More specifically, should technical development be supported through R&D institutions? Can it be best encouraged and promoted directly at the level of individual enterprises through financial and other forms of assis- tance? Or would a combination of both forms plus some changes in the system of incentives be appropriate? 3. In order to understand the importance, antecedents and scope of the technical development of Argentinian firms, the annex analyzes in some detail the following elements. (a) The technological characteristics of the exported products and the importance of local contributions to changes in product design and production methods. The destinations of exports and the extent to which technical modifications made them more "appropriate" to their users in other LDCs. (b) The manner in which technical development has taken place in the firms that are able to export technology. Are technical modifications made by the firms themselves based on their own technical developments or on previously imported technology? Have the modifications and adaptations been made by the firms themselves or by other agents, such as R&D laboratories and design engineering firms? In any case, what are the types and levels of skills required to make the modifications and adap- tations? (c) The effect on the firm's need for technical support of factors such as size and links with other firms (local or foreign).  -2 - ANNEX I Page 2 It is evident that policy implications are different when dealing with different types of firms. Subsidiaries of foreign firms depend for the most part on their parent companies for the supply of technical inputs. For this reason, the mere creation or development of local R&D institutions does not necessarily mean that there will be a demand for their services. Government action should be oriented towards promoting a higher proportion of research activities within the local subsidiary itself. In the case of locally owned firms, where size may be a constraint on what can be achieved, technologically, within the firm itself, there would be an argument for the support of independent R&D institutions serving a whole subsector. Similarly, the use of special lines of credit for research activites at sub- sidized rates, as well as other promotional policies such as tax rebates for research expenditures, might be justified in the case of locally owned firms, although not in the case of foreign sub- sidiaries. The same can be argued about the adoption of a scheme through which the government would bear part of the risk both of financing research activities and of lending to consultancy and engineering firms. (d) The factors that have stimulated technical changes in products and production processes in the export industries. Have these developments been stimulated by external market conditions (e.g., LAFTA agreements) or are they the result of internal circumstances? What are these market conditions and to what extent have they been affected by government policies? In other words, what are the effects of past and present economic policies, and of different institutional and organizational factors, on the technical behavior of Argentinian firms, particularly with regard to the decision to export technology? 4. In order to address these issues, the annex will first describe the export performance of a group of sophisticated industrial products on an aggregate level; secondly, analyze the evidence available on exports of turn- key plants from Argentina; and thirdly, study in detail the experience of two industrial sectors: machine tool and agricultural implements producers. The last section of the report contains the conclusions and policy recommen- dations. 5. There are several reasons for choosing exports of turnkey plants and the machine tools and agricultural implements sectors for the detailed studies. Exports of turnkey plants and process design are a recent, but increasingly important, phenomenon in Argentina. Between 1973 and 1977, Argentina exported to other LDCs 44 complete plants, with a total value of US$346 million. These exports are evidence that Argentinian firms possess the elements of technical knowledge and skills that are demanded by (and probably are particularly appropriate for) producers in countries at lower stages of industrial development. Exports of complete plants, particularly in the process industries, involve a whole range of interlinked technical  -3- ANNEX I Page 3 activities (involving different skills) which range from preliminary studies to design and construction of the plant in the recipient country. These activities can be carried out by one agent or several working in coordination, such as producers, engineering firms, equipment manufacturers, etc. The degree of division of labor is an indication of the specialization and effi- ciency achieved in the supplier country. 6. The machine tools and agricultural implement sectors have been established in Argentina for some time, in the process of import substitu- tion, and in recent years have substantially expanded their exports, particu- larly within the Latin American region. In the case of machine tools, exports now account for about 20% of output, while for agricultural machinery (ex- cluding tractors), the share is nearly 10%. Furthermore, in recent years some Argentinian producers have licensed or established joint ventures in other Latin American countries to produce the same products; these companies are exporting both technical knowledge and physical components. 7. A further reason for the selection of machine tool producers is the key role that this sector has in a country's industrial and technical develop- ment. Machine making nearly always plays a critical role in mechanical in- novations, the type most likely to occur in LDCs. 1/ Product and process innovations in other sectors are implemented through changes in production equipment. Conversely, innovations generated in the machine tools sector open new possibilities to industries using the new machines. There is increasing recognition among economists that local innovation, especially the creation of "appropriate" technology, depends on the existence of a local machine- making capacity. Following Cooper, 2/ it can be said that machine makers contribute to the development of appropriate technology by (a) creating new technical variants which are more appropriate to employment and income dis- tribution objectives; (b) reducing the investment cost of established pro- duction technology by changing the methods for making the machinery, by the materials used in their manufacture, or simply by exploiting factor cost advantages in machine making, especially low wage rates; and (c) using production techniques which in themselves are appropriate to economic and social objectives. There is evidence that machine tools producers in Argentina have made contributions of types (a) and (b). 8. In the case of the agricultural implements sector, a further reason for its selection as a case study is the evidence of early development of local technical innovations, particularly in product design. In fact, some Argentinian firms have been responsible for innovations that were afterwards adopted by producers in more industrialized countries. 1/ Charles Cooper, "The Role of Machine-Building in Innovation in Less- Developed Countries," Research Proposal presented to the World Bank, 1976. 2/ Ibid.  -4- ANNEX I Page 4 II. Export Performance and Local Technical Development 9. Since the end of the 1960s, manufacturing exports from Argentina have increased substantially. Between 1969 and 1974, manufacturing exports grew from US$507 million to US$1,867 million, an increase of 30% annually. Exports of products with relatively sophisticated technology grew from US$157 million to US$745.4 million over the same period (see Table 1). The increase of manufacturing exports from Argentina is commonly explained by three factors: (a) export promotion policies including tax exemptions, drawbacks, rebates, etc.; (b) the export strategies of a group of multi- national companies, which account for a high proportion of these exports; and (c) favorable regional factors, such as natural transport advantages in supplying neighboring countries, or complementary agreements within the LAFTA region. 10. However, as a 1976 study shows, 1/ there is evidence that the tech- nical development achieved by Argentinian industries in recent years is an important explanatory factor for the export performance of several sectors. The study analyzed the export performance of 20 large firms, which together accounted for 35% of all manufacturing exports between 1969 and 1974 (see Table 2). Most of these firms are local subsidiaries of multinational corporations. The study found that for several of these firms export per- formance was closely related to internal technological efforts. These tech- nological efforts (learning processes and R&D activities) resulted in improved international competitiveness for these firms. 11. Better international competitiveness as a result of improved tech- nical knowledge was detected in the following situations: (a) technological advantages derived from patents, inventions, know-how, etc., arising out of local development; and (b) advantages derived from the adaptation and/or improvement of foreign technical development. Examples of (a) are provided by a group of Argentinian firms (Siderca, Dalmine, Roque Vasalli, Siam di Tella) that were started several decades ago in areas of high internal demand (steel, tubes, harvesters). From the begin- ning they created plants of relatively efficient scale and diverted resources to increase their technical capability. Under conditions of protection, they soon arrived at an oligopolistic situation in internal markets, and began exporting at the beginning of the 1960s. Several of these firms have also invested abroad, taking advantage of the technical skills developed at home. 12. Technologically, these firms are at the same level as their counter- parts in the US or Europe. Particularly interesting is the case of Dalmine, 1/ J. Katz and E. Ablin. Tecnologia y Exportaciones Industriales, un Analisis Microeconomico de la Experiencia Argentina Reciente. Mono- grafia de Trabajo No. 2, Programa BID/CEPAL, Bs. As., Dic. 1976.  -5.- ANNEX I Page 5 which is doing pioneering work in the use of continuously cast rounds for seamless tube mill production. In the opinion of the mission, this is the most successful start-up yet of a direct reduction plant. 13. Examples of (b) are provided by several tractor, car and other transport vehicle producers, which are subsidiaries of multinational firms (Fiat, Ford Motor, General Motors, Mercedes-Benz). They all have large engineering teams engaged in adapting the product designs and production processes of the parent companies to local preferences and smaller scales of production. These adaptations have placed the Argentinian subsidiaries in favorable positions when the operations of the group have been expanded to other Latin American countries. 14. In addition to the two situations described, Katz and Ablin found cases in which the availability of relatively abundant and cheap qualified labor (such as technicians and engineers) has favored the use of more sophis- ticated technology in the Argentinian subsidiary than in subsidiaries in other LDCs (IBM, Olivetti, Citroen, Philips, Fapesa, Standard Electric, Cyanamid). As a result, the Argentinian subsidiaries have specialized in products that are not being produced in subsidiaries in other countries. Examples of such specialization include locally designed IBM components exported from Argentina to other subsidiaries in Sweden, Japan and Canada; and the specialization of the Argentinian subsidiary of Cyanamid in veterinary products that are exported to other Latin American countries and the US. III. Exports of Complete Plants 15. A more recent development, closely linked to technical capability, is the Argentinian export of complete plants to other LDCs. As mentioned previously, between 1973 and 1977, Argentina exported 44 complete plants for a total amount of US$346 million (see Table 3). 1/ Most of these plants were sold on a turnkey basis (i.e., the Argentinian supplier provided basic and detailed engineering, equipment, construction and start-up assistance) to other Latin American countries. A study of 34 of these plants (representing a total amount of US$340 million) 2/ shows that 28 of the plants were exported by 20 Argentinian firms, while only 6 plants were exported by the local sub- sidiaries of multinational enterprises (see Table 4). However, locally owned firms accounted for only 50% or so of the total value of exports, due to the fact that the average value of the plants exported by local producers is much lower (US$6 million) compared to the value of plants exported by subsidiaries of multinational firms (US$28.7 million). 1/ In 1976 Taiwan reportedly exported 58 plants for a total amount of US$16.3 million. The difference in the average value of the plants exported by both countries is striking. 2/ J. Katz and E. Ablin, "Export of Complete Plants from Argentina." Draft paper. Programa BID/CEPAL, Bs. As., November 1977.  -6- ANNEX I Page 6 16. The grouping of the plants sold by client sectors gives a revealing picture of the areas where local technical development has been pronounced. As could be expected (see Table 5), 17 out of the 44 plants are for food processing, a sector that was developed in Argentina at early stages of the import substitution phase of industrialization, and where there is evidently accumulated technical expertise. The number of chemical plants exported (9) can be related to the early development of the Argentinian chemical industry (non-oil based). These sectors are followed in the number of plants exported by metal industries (3 plants), construction projects (5 plants), and health services projects (3 plants). However, it is interesting to note that the most expensive plants exported are in non-manufacturing areas. They include an oleoduct and pumping station costing US$120 million sold to Peru, a turn- key airport costing US$52 million sold to Uruguay, and a turnkey hospital costing US$46 million sold to the Ivory Coast. 17. According to the same study, the export of complete plants is possi- ble due to the experience gained by the exporting firms through the building of similar plants in Argentina. Although in most cases the Argentinian plants have been built on the basis of imported technology, frequently an internal technical effort 1/ has been necessary to adapt the plant to local operating conditions and demands. It is this experience that has allowed them to solve similar problems in other LDCs. The exported plants are not based on stan- dardized designs, nor are they exact copies of local plants; they are usually designed ad hoc. An example is a plant sold to Cuba for the production of 50 tons a day of French bread. This plant is much larger than the one the exporters have in Argentina. Adaptations of the plant designs involve a fair amount of engineering work, especially in the area of detailed engineering. Thus the low salaries of engineers in Argentina, compared to those in other countries, 2/ give Argentinian firms (both locally owned and subsidiaries) a competitive advantage in this type of exports. A few of the contracts for the export of complete plants have been negotiated by engineering firms (Techint, Sei Ingenieria, Niselco) or trading companies (Phoenicia). In these cases the supplier company may or may not be the owner of the process technology. They may be licensees of other producers who do not have the resources to get involved directly in this type of venture, or of firms that have not systematized their technical knowledge in a readily trans- ferable form. It is expected that, as in the industrialized countries, 1/ This effort is related to the following needs: (a) to use more ade- quately waste or industrial by-products; (b) to substitute imported raw material; (c) to ease bottlenecks in production; (d) to provide technical assistance to suppliers and clients; (e) to adapt the product design to local requirements; etc. 2/ In Argentina one man/hour in engineering costs about US$10, whereas the man/hour in the US is about US$60.  ANNEX I Page 7 engineering firms in Argentina will become key agents in the export of complete plants, particularly in the process industries. 1/ 18. In the process industries the capacity to export technology, whether as turnkey plants, or through licensing agreement or direct foreign investment, is not confined to the largest firms. A recent study 2/ designed to evaluate this potential found that 56 out of 67 small and medium process- ing firms offer their own technology. At the time of the study, 27 of the 56 had already signed transfer operations, 18 firms were in negotiations, and 11 firms had not had previous transfer experience. Most of the firms offering technology are able to provide those elements of technical knowledge closely related to their own experience as producers, such as process technology (offered by 77% of the firms); basic engineering (offered by 71%); personnel training (offered by 82%); and technical assistance (offered by 82%). Fewer firms have the organization required to provide consultancy (64%), detailed engineering (54%) and equipment (50%) without the assistance of an engineer- ing organization. This suggests that the almost 75% of the firms that stated that they are able to offer turnkey plants underestimated the engineering effort involved in this type of operation. Nevertheless, most of the firms that have already exported technology are of medium size; and more than half of them exported turnkey plants. Prospective users of the offered technology face prices which range from US$5,000 for know-how payments to US$8 million for the installation of a complete plant. 3/ 19. Katz and Ablin also tried to evaluate the degree of originality of the offered technology. Their main conclusion is that the technology offered is based not on important innovations, but on the capacity to organize exist- ing knowledge developed as part of the production experience of these firms. In two-thirds of the firms, this capacity has been developed based on free knowledge; in the remainder it has been based on adaptations and modifications of licensed technology. The proportion of firms offering technology based on previously licensed processes is lower in this study than what might be ex- pected from looking at the plants that have been exported. This seems to indicate that the firms offering technology based on previous licenses are more likely, in fact, to carry out transfer operations, probably as a result of the higher confidence that a previous license instills in the prospective client. 1/ Local engineering firms also play a crucial role in the import of technology for the process industries. 2/ Elva Roulet, "Oferta Potencial de Tecnologie Nacional Comercializable: el caso de la pequena y mediana industria de procesos." OEA Programa de Transferencia de Tecnologia, Buenos Aires, Argentina, 1976. 3/ Five out of 24 firms offer technology worth more than US$1 million. However, most of the firms offer technology ranging from US$100,000 to $500,000.  -8- ANNEX I Page 8 20. The technical development of the small and medium firms offering process technology has taken place within the firms themselves, using, in most cases, highly qualified scientific and engineering personnel. The use of highly qualified personnel from the beginning of firm activities is more frequent in process than in non-process industries, such as mechanical engi- neering. The reason is that in the process industries technology cannot be acquired through copying a prototype or a piece of equipment. Technology is, in this case, embodied in flow charts and design specifications. Technical development has taken place in the firms without much help from outsidle agents except in the case of foreign licenses. Joint work with engineering firms, which is common for these industries in the advanced countries, is very rare. Similarly, the use of the state research facilities is very limited. Most of the contacts with Instituto Nacional de Tecnologia Industrial (INTI), Labora- tory for Material Testing (LEMIT), university research centers or other insti- tutions (Instituto del Plastico, YPF, Gas del Estado, etc.) are for routine and quality control tests. According to the firms, joint work with these institutions is limited due both to the lack of flexible mechanisms for collaborative work and to the instability of some of the government funded institutions. 21. Because of the size of most of the firms offering technology, the lack of collaboration with other agencies is likely to be a major constraint, not only on the capacity to transfer technology, but also on future technical development. To export technology, small firms may need the collaboration of engineering firms: (a) to help prepare the technology offer; 1/ (b) to serve as general contractor in a turnkey arrangement; (c) to systematize the firms' know-how in a transferable way; and, (d) to perform the detailed engineering of the clients' plant. These tasks all require a specialized engineering organization that exceeds the internal capability (in terms of personnel and financial resources) of most small and medium firms. However, an obstacle to a more efficient division of labor is that engineering as an independent activity is a relatively recent development in Argentina. 22. Further technical development, particularly the development of more important technical innovations, seems beyond the individual capacity of many small and medium processing firms. This is evidenced by the fact that all the offered technologies were found currently in operation3 with none of the firms declaring products or processes in the development stage. Similarly, the study found limited use of pilot plant work, indicating financial con- straints on research activities. Closer collaboration between process industries and the state's research facilities seems the evident solution. However, to improve collaboration, it would be necessary to design imagina- tive institutional arrangements to reduce the tensions created by the proprietary nature of the developed technology. 1/ The cost of a technology offer in the case of turnkey plants was esti- mated by the firms to range from US$20,000 to US$50,000. For each approved transaction, the firms present about 6 offers. Approximately 60% of the firms declared that they have the technical capability to prepare the offers.  -9- ANNEX I Page 9 IV. Machine Tools 1. Historical Background 23. Although production of machines in Argentina can be traced back to 1930, until the early 1940s it was a limited activity carried out by small repair shops. The industry as such emerged during the 1940s and early 1950s as a result of the difficulties of importing machinery and parts during the Second World War and recovery. By 1953 about 37% of the firms producing machine tools in 1971 had already been established. 1/ In 1960, Argentina's production of machine tools reached US$10.4 million. 2/ However, this level of production was not exceeded again until 1965, when production reached US$16.9 million (current dollars). Wide fluctuations in the level of production continued during the rest of the decade, 3/ although the value of production did not exceed the 1965 level until 1969. During the 1970s machine tool production grew more steadily, 4/ reaching about US$42 million (current dollars) by 1976 (see Table 6). This value of production was generated by approximately 110 to 120 machine tool producers. 5/ 2. Main Characteristics of the Industry Today 24. The origins of the machine tool industry in Argentina conditioned its main features today. Most firms are locally owned (by Argentinians or immigrants of European origin) and controlled by family groups. Furthermore, most of the existing firms are still controlled by their original founders, although there is evidence that in a high proportion of the older firms, the younger generation will assume control in the next few years. 25. Most machine producers are still of small and medium size. Accord- ing to one study, 6/ in 1971 62 firms out of 94 employed less than 20 workers; 1/ Approximately 72% of the firms producing in 1972 were created before 1960. 2/ In the same year 7,756 metal cutting and 1,819 metal forming machines were produced. 3/ Between 1960 and 1970, the growth of metal forming and metal cutting machines was 82% and 35% respectively. 4/ In 1971 approximately 104 firms produced 10,484 metal cutting and 3,325 metal forming machines. 5/ J. Campos, "La Actuacion Internacional de la Pequena y Mediana Empresa en America Latina, Un Estudio Empirico," INTAL, Bs. As., Argentina, April 1977. 6/ Direccion General de Fabricaciones Militares, "Estructure de la Industria de Maquinas Herramientas," Direccion de Desarrollo, Bs. As., Argentina, 1972.  - 10 - ANNEX I Page 10 20 firms employed between 20 and 45 workers; and only 12 employed more than 45 workers (see Table 7). However, the third group of firms produced 58% of the total value of production, whereas the first group contributed only 18%. This concentration in the value of production is related to the fact that the larger firms are specialized in machines of higher unit value and weight. In fact, firms with more than 45 workers produce machines with an average value of US$7,000 against an average value of US$500 for the smaller producers. These values reflect the wide differences in the technical capabilities of the various sized firms. 26. The degree of concentration in production in favor of the larger firms differs according to product. Thus, in 1971 the two largest firms (of a total of 8) produced 96% of the value of all the bending machines and guillotines, whereas in lathes the 5 largest firms (out of 22) produced 62.7% of the value of production. 1/ It should be emphasized that information about concentration in production by type of machine conceals the relative importance that some firms have in the market for machines of specific charac- teristics. This is because there is a high degree of differentiation among machine tool producers. Most firms have only one line of product and their machines usually have characteristics that differentiate them from other machines of the same type. 2/ Only the large producers have several product lines and several models within each line. 3. Imports and Exports 27. Despite the expansion and increased sophistication of machine pro- duction in Argentina, approximately 50% of the local demand is still met with imported machines (see Table 6). Imported machines (70% of which are acquired by the automobile industry, including parts producers) are not close substi- tutes for machines that are currently domestically produced. The imports are either numerically controlled precision machines or much larger models than those produced locally. Other differences in technical characteristics can be inferred from the much higher value of the imported machines. Average cost of imported machines was about US$16,000 in 1966 and US$27,000 in 1972, whereas the average cost of locally produced machines was US$1,400 and US$1,700 respectively. 3/ It is evident that further expansion of the internal market for local machine tool producers is limited, unless they 1/ Ibid. 2/ For example, none of the machines produced by 11 producers of parallel lathes has identical characteristics. E. Amadeo, F. Fernandez y R. Morales, "El Sector Maquinas Herramientas en la Argentina," CLACSO, Proyecto de Investigacion Instrumentos de Politica Cientifica y Tecnologica (STPI-IDRC), Bs. As., Argentina, 1976. 3/ Amadeo et. al, op. cit. Until recently only 20% of the imported machines were subject to import taxes because they were similar to machines pro- duced locally.  - 11 - ANNEX I Page 11 enter into the production of more sophisticated types of machines, including numerically controlled ones. Although the present structure of protection (discussed in Section VI) discourages any development in this direction, a few of the largest producers have already begun the design of more sophisticated equipment. 28. Because of its role in the investment process, the machine tool sector has always been very sensitive to variations in the level of production of its customer industries. Despite the strong cyclical conditions of their markets, machine tool producers have only recently followed a systematic export policy. Before 1970 exports had been used mainly as an alternative to local sales in periods of severely depressed internal demand. Thus, the level of exports reached during the recession of 1962-63 was not exceeded until 1970. During the 1970s, however, exports have increased significantly, reaching nearly US$10 million in 1975. Most exports go to Latin American countries (see Table 8), where they have special protection under the LAFTA agreements as well as a natural transport advantage. From 1967 to 1973 exports fluctuated between 10% and 15% of total production. Currently that figure is approximately 20%. However, spokesmen for several of the leading firms mentioned exports as high as 45% of their total production, 1/ although the same firms also showed a drop in 1977 exports. The reasons often cited for this decrease were (a) the overvaluation of the currency, and (b) internal situations in some of their traditional export markets, i.e., Mexico (drop of the peso) and Brazil (protection of local industry). 29. The character of the Argentinian exports of machine tools can be better understood if the machines are grouped into three categories of increasing degree of technical sophistication: 2/ (a) simple "universal" machines; (b) semi-sophisticated universal and special machines; and (c) nu- merically controlled and large size machines. The first group includes strong and easy to use basic machines that are appropriate for countries in early stages of industrialization. The second group consists of more advanced machines using different types of sophisticated components for measurement and control, the type of machines used by more advanced mechanical engineering industries. The third group includes the most advanced machines, which at the moment are not being produced in Argentina. 30. Argentinian firms export mainly machines of the first type; those that have benefited least from the technical development of the sector. Such machines seem to compete in quality and price with machines from other countries (aided by a 25% rebate that compensates for the high domestic price of steel). Machines of the second type are also exported, but they face stiffer competition from East European and Japanese producers. The more sophisticated 1/ Another study with a larger sample (31 firms) reports exports between 10% and 20% of total production between 1971 and 1976. Jaime Campos, op. cit., p. 58. 2/ Increasing automation, speed, precision, size and versatility.  - 12 - ANNEX I Page 12 machines of the second type, such as transfer machines, are difficult to export because the main demand occurs in countries with diversified industry, especially the automotive industry, which is usually controlled by multi- national corporations. These firms tend to buy their equipment from long established suppliers. Another factor that limits the export of technologi- cally advanced machines and components is the need for a technical infra- structure for sales, start-up, and post-sales assistance. This is presently beyond the organizational capability of most Argentinian firms. In fact, among the leading firms the mission found only two that have representatives able to provide such services in the importing countries. 31. The future evolution of the demand for the simple types of machines presently being exported from Argentina seems mixed. Although some of the smaller countries, such as Peru, that are now building up simple metal fabri- cating industries will continue to import these machines, some of.the larger client countries will advance their own production of machines of the first type. To continue exporting to the latter countries, Argentinian producers will have to specialize in the production for export of semi-sophisticated universal machines. 1/ The largest Argentinian firms are now producing this type for the internal market and they have an evident lead over other LDCs. To produce for export, however, will necessitate not only the improvement of the technical characteristics of production, but also the development of a support infrastructure to provide assistance, spare parts and services for foreign clients. 32. Faced with obstacles and uncertainties in the export markets, several firms are contemplating the transfer of production to the importing countries, either by direct investment or by sale of technology. The mission found three firms which are actually engaged in transferring production abroad. Each of these firms has created a partnership with a local firm in another Latin American country (Peru, Brazil and Venezuela), transferring production to the associated firm by stages, i.e., advancing from assembly of imported parts in the recipient firm to gradual incorporation of locally produced parts. The Argentinian firm's contribution to its partner in the recipient country consists of designs, know-how, technical assistance and some machinery, but not financial resources. In addition to these three firms, the mission found seven firms that have had similar arrangements in other Latin American coun- tries. Several of these firms expressed interest in selling their technology, but have not been able to do so because they do not have sufficient technical personnel to send to the recipient country without endangering production at home. It is evident that although the level of technical development reached by these firms is an asset marketable in countries with less experience in machine tools production, the skills involved are not in a form that makes them easily transferable. What can be sold are elements of technical knowl- edge based on ad hoc solutions and unwritten operational practices without 1/ The production of these machines is becoming less attractive to large producers in industrialized countries. They are, instead, turning towards the production of large specialized machines.  - 13 - ANNEX I Page 13 complete designs and specifications. These elements are embodied in the tech- nicians and managers of the firms and their direct participation is required to make the transfer. The few firms actively engaged in selling technology abroad are active exporters that have created important trade organizations in the recipient countries. Their interest in transferring production abroad is closely related to the problems that they have faced in their export markets (Mexico, Chile, Peru, Brazil) or their desire to enter a market which is difficult to supply through exports (Venezuela). 4. Technical Development 33. For the purpose of this analysis, the level of technical development of the Argentinian machine tools producer will be defined as a function of (a) the level of design skills reached by the firm; (b) the use in the produc- tion process of techniques that improve the quality and performance of the products (e.g., use of precision machines, heat treatment, standardization and quality control); and (c) the degree of division of labor achieved by the firm insofar as this affects the quality of its products. Although the type of production technology used is an important element in an assessment of the technological level of a producer, this will not be discussed at length in the case of Argentina because the production processes for the types of machines produced there are relatively simple and well known. 34. To evaluate the design skills achieved by Argentinian machine pro- ducers, three levels of increasing sophistication will be defined. Firms at the first level produce machines "copied" from imported equipment. This copying process is an ad hoc duplication of parts in the workshop itself. At the second level, copying becomes a more systematic activity, with drawings of blue prints and specifications produced by the firm's design team. This will be called the "imitation stage" because advanced design skills allow the firm to go beyond simple duplication and combine into one the characteristics of different machines as well as to make simpler adaptations. "Adaptation and innovation" comprise the capabilities of firms at the third level. Design skills have reached the point where a firm can incorporate into the machines technical features that cannot be easily copied without understanding the scientific principles involved. The difference between the second and third levels is the incorporation of qualified engineers into the design team; at the second level, the teams are composed solely of technicians trained in the firms themselves. At the third level, fabrication of prototypes becomes a systematic activity, and assistance external to the firms is sought to resolve technical problems. 35. Machine tool production in Argentina evolved from the repair activi- ties of small workshops or at the initiative of individual machine operators and metal workers, many of whom had been trained in army workshops or on the railways. In addition, a substantial proportion of the workers were Italian immigrants who had had previous training in metal working in their home coun- try. The machines produced in the initial stages were simple lathes, grinders, pressers, boring and drilling machines, copied from imported machines. Produc- tion was evidently artisanal, based on the technical ability of the firms'  - 14 - ANNEX I Page 14 owners to copy the machines with which they were familiar. The production technology was very simple, in many cases using machines produced by the firms themselves. Argentinian producers surpassed this technical level by the end of the 1950s. 1/ Most machine tool producers in Argentina are now at what was defined as the second level of technical development, i.e., the "imitation level." Since the 1960s, when many of the machine tool producers entered this stage, there has been a diversification of models and sizes of machines. Also, components (such as electric systems, lubrication, refri- geration, brakes, gears, etc.) have been incorporated into the machines to make them more efficient and versatile. This technical development was closely linked to the technical requirements of the machine tools customer industries. In this respect, the automobile industry, by far the sector's main client, 2/ has played an important role in demanding machines of growing precision and scope. One indication of technical advance in machine produc- tion can be given by the average value of the machines, which rose from US$1,075 in 1960 to US$1,600 in 1971 (deflated by the overall price index), whereas the average weight trebled. 3/ The increase in value is highet for the metal cutting than for the metal forming machines, seeming to indicate a greater technical advance in the former. 36. The process of gradual acquisition of design skills by the Argentinian machine tool producers has taken place with very little partici- pation of engineers and with negligible use of local and foreign consultancy groups, or R&D laboratories. Furthermore, as design skills have been acquired by copying imported machines, foreign investment and licensing has been of very little significance. It is probable that the same conditions that favored the initial development of the sector have imposed severe limitations on the capacity of many firms (which have not been able to change their tech- nical organization) to keep up with both local and foreign markets demanding machines of increasing technical sophistication. In other words, many of the small and medium firms are reaching the limits of their technical development. These need not only change in internal organization, but also a great deal of assistance from technical institutions to solve problems that demand skills and resources that are beyond individual firms' capabilities. 37. Nevertheless, there are some firms that have overcome the limita- tions that their former organizational structures placed on technical develop- ment. These firms have incorporated engineers into their design and produc- tion departments and are more receptive to technical assistance in areas where they lack expertise. The mission found some evidence that such change 1/ Amadeo et. al., op. cit. 2/ 80% of the machine tools producers interviewed for a study in 1971 declared the automobile industry (including suppliers) to be their main customer. A. Spilzinguer; Revista Equipamiento, quoted by Amadeo et. al., op. cit. 3/ Campos, op.cit.  - 15 - ANNEX I Page 15 has taken place mainly where management has been passed to the hands of a second generation, comprised of men who usually have a formal technical edu- cation. 38. It is difficult to determine how many firms have reached what is defined here as the third level of technical development. A study of the innovations made in this sector identified only eight firms that had devel- oped or adapted technical improvements that upgraded considerably the preci- sion and durability of their machines. These improvements were later adopted by other producers. 1/ The number of firms with engineers on their research teams is also an indication of technical development. In this respect, the mission found that 3 of the 15 leading firms reviewed did not have engineers on their design teams. Representatives of several of the firms with engineers mentioned that the engineers had been employed within the last five years. Amadeo's study (with a more representative sample) found that 10 firms out of 23 had no engineers in either design or production. He also found that none had more than five engineers. Even in the leading firms the number of tech- nicians in design is relatively low. In 8 of the 15 firms visited by the mission, the number of technicians in design was between 1 and 5; 3 firms reported between 7 and 8 technicians; and only 3 firms had more than 10 technicians. These results coincide with those of Amadeo's study, which found that most firms had between 1 and 4 technicians in design and that only 5 out of 23 firms had more than 5 technicians. 39. Within the group of firms with higher technical capability visited by the mission, there are three firms that are already working on prototypes of jig boring, grinding and numerically controlled machines. Of particular importance from a technical view point is the existence of three producers of specially designed automatic machinery, including transfer machines. These machines are designed to do a specific task, or series of tasks, according to the particular needs of a client. There are several externalities asso- ciated with the local production of this type of equipment. In addition to probable important savings in costs through local purchase, local production of these machines improves the efficiency of other industries by eliminating delays involved both in ordering construction abroad and in the acquisition of spare parts and repair services. These are important considerations due to the fact that, from the private point of view, their production is not very profitable. In fact, the two small producers of this type of machines visited by the mission are planning to change their lines of production in favor of universal machines made in series, for export. 40. Another important element from the technological point of view is the level of sophistication reached by local component producers. Since the end of the 1960s substantial development has been achieved in the local 1/ Amadeo et. at., op. cit.  - 16 - ANNEX I Page 16 production of components (hydraulic and pneumatic components, cutting tools, and electric and electronic elements for measurement and control). Indeed, producers of components have become the main source of technical development in the sector. l/ These producers have developed specialized skills and are, in some cases, in a position to teach machine tool producers how to adapt the technical characteristics of their machines in order to incorporate the new components. The independent development of the component producers has been favored by the fact that they produce components for a wide range of local customers, in addition to machine tool producers, as well as for export. 41. It was already noted that foreign investment and licensing have had little significance in the technical development of the sector. The mission found that only two of the leading firms had foreign equity partici- pation. In one case, the foreign partner is itself a machine producer (an Italian firm which has 13% of the equity of the Argentinian firm), which has supplied technology to the local partner. Only four cases of licensing agree- ment were found, one of which was the associate of the Italian firm. In other cases, licenses covered only one of the firm's lines of production. It is interesting to note that representatives of several of the firms that have never licensed foreign technology mentioned that they intended to do so in the future. Several of the firms looking for foreign technology are inter- ested in the production of hydraulic and pneumatic components, areas where direct copying is difficult without proper R&D activities. On the other hand, two of the firms are interested in licenses for the production of small standard machines, mainly for export. In these two cases, the trade name seems to be more important than the acquisition of technology. 42. The absence of formal agreements does not mean that foreign tech- nology has not played an important role. In addition to the copying of imported machines, the main sources of technical information are national and international fairs and exhibitions. Spokesmen for almost all the firms visited reported that representatives are sent regularly to fairs in Europe and the US. In some firms, even fairly small ones, the owner and the technicians working in design and production attend the fairs, which are seen by many as one of the main sources of ideas for new products. 43. In relation to sources of technical information, it should be remarked that there are no institutional channels in Argentina for the dif- fusion of technical information for the machine tool sector. Neither the National Institute of Industrial Technology (INTI) nor professional or industrial associations are able to provide this type of assistance. The interviews revealed a profound concern in the firms for more access to technical information. It is evident that such limited access increases the risks involved in a firm's decision to implement technical changes. 1/ Henrique Sabatte, "Evolucion de las caracteristicas y la produccion de las maquinas herramientas en la Argentina 1962-1972," INTI, Bs. As., Argentina, 1972.  - 17 - ANNEX I Page 17 44. In relation to utilization of other technical services, the mission found that consultants are rarely called upon to solve technical problems. Only three of the firms reviewed have received assistance from INTI. More- over, this assistance has been limited to the use of INTI's laboratories for materials testing, chemical analysis and mechanical tests. Use has never been made by the three firms of INTI's technical services to acquire technical information or to solve technical problems in production or design. Some of the firms that have used INTI's laboratories experienced delays in obtaining results. Furthermore, representatives of three of the larger firms mentioned that their own testing facilities were better equipped than those of INTI. The role of this institution as a source of technical services for this sector may change when CIMHER (the research center for machine tools) begins operation. This center was created jointly by INTI, the Association of Machine Tool Producers and the Association of Component Producers, with tech- nical assistance from UNIDO. In addition to providing better testing facili- ties, the center will carry out research jointly with client firms and will provide technical information. The participation of the Producers' Associa- tion indicates the need felt by the producers to have access to these kinds of technical facilities. At the moment, the most comprehensive source of tech- nical services and information for the machine tools industry is the industrial estate of San Francisco, an area where 20% of the country's production of machines takes place. This project was developed with the collaboration of UNDP which has provided, in addition to testing equipment, experts on tech- nical and organizational matters. These experts give direct assistance to the firms in the industrial estate. 45. The future export possibilities of Argentinian machine tools depend on the quality not only of their designs but of the machines themselves. It appears that the larger machine tools producers in Argentina have reached acceptable quality standards for international markets. However, there are many small and medium size firms that have not yet achieved the required levels of quality control, normalization and standardization. 46. The increase in quality standards, as well as the technical sophis- tication achieved by the leading machine tools producers in Argentina, has been closely linked to the requirements of their main customer industries. In quality, as in technical sophistication, the automobile industry (including its subcontractors) has played a very important role. It has had not only a direct impact as the main client of the sector, but also an indirect impact on the quality of foundries. 1/ It is evident, however, that the automobile 1/ Although many of these foundaries are "captive," diffusion of new tech- niques may have occurred through movements of workers and technicians. According to a recent study, since 1960 the automobile industry has been the main factor in the technical development of foundries in Argentina. In 1973 the study identified 450 foundries, 50 of them with a capacity of more than 1,000 Tn monthly. J. Sabato, et. al., "Ensayo de Regimen de Tecnologia, el Caso de la Fundicion Ferrosa," mimeo., Bs. As., Argentina, Septiembre 1974.  - 18 - ANNEX I Page 18 industry has had a greater effect on the quality standards of the producers of special machines and large machine tools than on small firms, whose pro- duction is of general purpose machines used by relatively small-scale machine shops or the maintenance departments of larger factories. 47. The evidence of quality standards is (a) the use of precision boring machines (gear grinders, honers, lapers) to achieve high tolerances; (b) the quality of the heat treatment; (c) standardization of production; and (d) quality control. Most of the leading firms studied for the present report have specialized precision equipment. The quality of the precision equipment of two small producers (77 and 47 employees, respectively) of special (in- cluding transfer) machines was remarkable. Further evidence of the use of this type of equipment is the high proportion of imported equipment used by machine tool producers (see Table 10). However, the latter can also be a sign of overcapitalization brought about by a policy of cheap capital for equipment purchase. This is corroborated by the study's finding of a high underutilization of some of these machines, particularly in the smallei firms. Comments from several producers about common procurement practices offer further evidence in this direction. They pointed out that if the quality of a local machine is not trusted, the preference is to import a machine larger than that required if the latter is not produced locally; in this way most of the tariff barriers are avoided, and only 5% ad valorem is paid. 48. All of the leading firms apply heat treatment to harden key parts of the machine tools. Most of these firms have their own heat treatment facilities and testing equipment. However, at least two of them subcontract heat treatment outside. If this service were more available, it is possible that smaller firms also would be able to subcontract heat treatment. 49. Most of the large firms carry out quality control of their inputs and finished products, although the sophistication of the quality control equipment varies from firm to firm. A study made in 1972 1/ found that only 2 among the 31 largest firms did not carry out any type of quality control, whereas about 20% carried out strict controls of finished and inter- mediate products. On the other hand, only 5 of 52 smaller firms carried out strict controls. A more recent study 2/ found that only 3 of 23 firms have control equipment above acceptable levels and specialized personnel. In these three firms, quality control covers almost 100% of the pieces of the machines. Most producers seem to be aware of the need for improvements in this area. 50. Most of the firms studied have standardized their production. In effect, one third of the leading firms producing cutting machines follow IRAM (Instituto de Racionalizacion de Materiales) standards. However, the 1/ DGFM, op. cit. 2/ Amadeo, et. al., op. cit.  - 19 - ANNEX I Page 19 study did not find any machine with this institution's official seal of con- formity to norms. (IRAM does not register norms for components.) In addi- tion to IRAM norms, there is a variety of other norms in use, such as Schlessinger, DIN, etc., and even norms created by the producers themselves. This variety limits the possibilities of further standardization and of achieving a degree of interchangeability of parts and products among pro- ducers. 51. There is evidence also of a movement towards less vertical inte- gration, although this industry has not reached the levels of specialization reached in other industrialized countries. The low degree of vertical inte- gration is evidenced by the fact that only 4 of 13 large producers have their own foundries, and in at least 2 of these cases the foundries operate as independent concerns doing work for other clients. In addition to castings, the items that are most frequently acquired from other firms are motors, gears and components. However, the present study found that only 3 firms make intensive use of subcontractors. One of these has fostered the creation of 14 small subcontracting firms among its own qualified workers, who use machines provided by the parent firm. Representatives of other firms indicated an intention of following a similar strategy to avoid labor problems. 5. Managerial Development 52. Machine tool production in Argentina was developed by craftsmen and metal workers who knew more about production than about management. As mentioned previously, most existing firms are still controlled by their origi- nal founders. In most firms, therefore, production matters have received much more attention than organizational and administrative problems. Thus, the organizational structure of many firms is inadequate to respond to the present conditions of the market, both internal and external. 53. The Amadeo study 1/ found that, with the exception of the largest, most firms have a very low level of differentiation among the various adminis- trative areas (i.e., sales, budgeting, purchasing, production planning and control, etc.). The study found that in only about half (12 of 23 firms) is sales differentiated from production. Of these 12 firms, 7 sell through distributors and only 5 have a sales department. In the other 13 firms sales are made by a person who is also responsible for other activities, such as purchases, accounts, etc. Six of these firms sell over the counter. Finally, only 50% of all the firms studied make estimates of future sales; the remaining 50% have a horizon not longer than 3 months. 54. Financing activities are also little differentiated. Only 4 of the 23 firms have finance managers, and in 3 this function is carried out by a part-time accountant who has other administrative functions. Only 9 1/ Amadeo, et. al., op. cit.  - 20 - ANNEX I Page 20 of the firms prepare financial budgets, although 19 have used BANADE's long term credit. Purchasing, on the other hand, is given more importance. Fifty percent of the firms have one or more full time persons for this task, and in 70% of the firms future purchases are planned. Production planning and control is not a differentiated activity for the smaller firms, where this function is the responsibility of the owner. Larger firms have specialized departments. Finally, only 2 firms (of the 23) knew their marginal costs and incorporated this variable in their production plans. 55. The picture that emerges is one of limited technical information within most of the firms and very low capacity to evaluate alternatives based on the information that is available. This is a situation that severely limits the effectiveness of promotion policies (credit, tariffs) that are designed for firms able to make rational decisions relevant to the advantages such policies provide, and to then incorporate these decisions in their in- vestment plans. 1/ In other words, a proper policy framework is a necessary, but not a sufficient, condition for the development of some of the small and medium size firms. Other measures involving more direct technical and managerial assistance would seem necessary. 2/ V. Agricultural Implements 1. Historical Background 56. Production of agricultural implements in Argentina started in the early 1920s in response to the demands generated by the early mechanization of agriculture. Production of more sophisticated equipment such as harvesters began in 1921, only two years after the first harvesters were imported. By 1930 approximately 70% of the firms producing harvesters in 1960 had already been created. The sector developed further during the 1940s as a result of the restrictions imposed on imports during World War II and was firmly estab- lished by 1947, when it had to face strong competition from imported equip- ment. Protection for the sector began in 1951 when it was declared of national interest. 57. By 1969 it was estimated that the number of firms producing agricul- tural implements ranged between 450 and 500. Among them approximately 380 produced only this kind of product. The specialized producers included 19 firms producing harvesters, 66 producing ploughs (and similar equipment), 38 producing silos and 4 producing tractors. About 95% of these firms were established in the provinces of Santa Fe, Buenos Aires and Cordoba, distributed among numerous small towns near the most important agricultural areas. By 1968 gross production of agricultural equipment was estimated at approximately 1/ Amadeo, et. al., op. cit., makes this point. 2/ These points will be discussed further in the concluding section of this report.  - 21 - ANNEX I Page 21 US$50 million in agricultural implements (harvesters, ploughs, silos, etc.) and US$67 million in tractors. By that year local production covered 94% of internal demand. 58. The evaluation of the volume index of production of agricultural implements in later years (see Table 11 and Graph 1) indicates that by mid- 1970 total production was roughly 50% higher than in 1960. By 1977 the volume of production had expanded further to double the 1960 level. The increase of production in 1976-77 was the consequence of the increase of agricultural prices in relation to price of agricultural equipment (see Graph 2). This situation lasted until mid-1977 when price controls on agricultural imple- ments ended. By the time of the present study, most firms were experiencing a sharp drop in demand as a result of a decline in agricultural investment. 2. Characteristics of the Industry Today 59. Although most of the firms producing agricultural implements have been in business for more than 15 years (in Provincia de Santa Fe 80% of the firms were started more than 15 years ago), they are in general small and medium size, family controlled undertakings. In Santa Fe province 74% of the firms have less than 50 employees, 1/ while in Cordoba province 33% of the firms have less than 50 employees. 2/ There are, however, 8 firms with more than 300 employees, and the largest firm (Roque Vasalli) has 650 employees. With the exception of the four tractor producers which are all subsidiaries of multinational enterprises, the firms are locally owned. The present study found only one case in which a local firm had foreign invest- ment. The market for agricultural implements appears fairly competitive and diversified. The diversity of products can be inferred from the fact that 69 firms, consulted for a study of export possibilities, reported production of 120 different products (85 agricultural machinery and 31 motorized equip- ment). 3/ Most firms produce a variety of different products, as well as all the pieces, parts and accessories of the equipment that they produce. The average imports of inputs do not exceed 10%. 3. Imports and Exports 60. Up until 1966 imports of agricultural implements exceeded exports. Beginning that year (when they totalled US$633,000) exports grew at an accele- rated rate until they reached more than US$12 million per annum (see Table 14 and Graph 3). Annual imports in recent years have amounted to approximately 1/ Camara de Fabricantes de Maquinaria Agricola, "La Industria de Maquinaria Agricola en la Provincia de Santa Fe," Rosario, Argentina, April 1977. 2/ Federacion Industrial de la Provincia de Cordoba, "La Promocion de la Industria de Maquinaria Agricola," Cordoba, Argentina, November 1976. 3/ Proyecto UGEPEK, "Estudio de la Oferta Exportable de Maquinarias Agricolas y Viales de la Provincia de Cordoba, Cordoba, Argentina, 1972.  - 22 - ANNEX I Page 22 US$2 million, although in 1975 they reached an all-time record of US$5.8 million. Exporting has become a systematic activity for a high proportion of Argentinian producers. Thus, 33% of a sample of firms in Santa Fe and 75% of a sample of firms in Cordoba have exported at some stage. 1/ Their exports range between 5% and 10% of production. However, in 1976 a group of 40 firms in Cordoba exported an average of 19% of production. Almost all exports go to other Latin American countries where they have, besides natural transport protection, special treatment under the Latin American Free Trade Association (ALALC) agreements. Within ALALC Argentinian exports must compete with exports from Brazil, which have the same special treatment. Brazilian exports are also the main competitors of Argentinian products within the Argentinian market itself, because they are tariff-free under the ALALC agreements. In addition, the Brazilian exporter receives 40% in rebates (against 25% for Argentinian firms) and can buy raw materials substantially cheaper. 61. Despite the adverse consequences of the high prices of raw materials in the cost of Argentinian production (raw materials account for approximately 42% of the selling price), there are no great price differences between the agricultural equipment produced in Argentina and that produced in other indus- trialized countries. A recent study 2/ found that at early 1977 prices, agricultural equipment produced in Argentina was on average only 3.27% more expensive than the same equipment produced in the US and Europe (at internal and not export prices) (see Table 17). The differences in prices for indi- vidual products ranged from 4.18% for ploughs and 3.19% for automated harvesters to 33.68% for scrappers. On the other hand, growers, bailers and corn col- lectors were between 20% and 50% cheaper than their European counterparts. The same study concludes that if prices of locally produced raw materials, which are between 25% and 50% higher than international prices, were lowered to international standards, prices of agricultural equipment could be reduced by about 14% (see Table 18). It is difficult to assess whether agricultural implements produced in Argentina are in any sense more technologically appro- priate to other LDCs than products produced in Europe or the US. Faced by this question, most producers answered that their products are more versatile, more robust and relatively smaller than their US and European counterparts. Versatility is particularly high in the cases of seed drills and classifiers of mill grain, whereas motorized equipment is less versatile. According to the producers, the versatility and durability are in response to the limited resources for mechanization available to agricultural producers in the country, limitations that are also present in other LDCs that acquire Argentinian equipment. There are, however, constraints on the adaptability of 1/ FIPC, op. cit. and CFMA, op. cit. 2/ Asociacion de Fabricantes de Maquinaria Agricola de la Provincia de Cordoba, "El Costo de la Maquinaria Agricola Argentina, Cordoba, Argentina, Julio 1977.  - 23 - ANNEX I Page 23 certain equipment to the soils and working conditions of the recipient coun- tries. 1/ The solution to this problem is, according to local producers, only a matter of adequate information. Several producers pointed out that because of the diversity of agricultural conditions in Argentina, designs have been developed that are easy to modify. 62. According to the findings of a recent study, 2/ export of agricul- tural implements from Argentina faces serious obstacles, both internally and externally. Two of the constraints mentioned concern transport and financing. In relation to the latter, the study points out the adverse financial effects on the firms of the time lag (60 to 90 days) between the actual export and the realization of the letter of credit. The study also found growing diffi- culties in the importing countries. Brazil, which receives 56% of Argentinian exports of harvesters and 44% of other agricultural equipment, has taken mea- sures to restrict the import of machinery. One means is to oblige Brazilian importers to make a deposit of 100% of the value of the import that can be recouped, without interest, only after a year. In Venezuela banks are reluc- tant to endorse credit letters, making their realization difficult. In addi- tion, exports to Cuba, made under a special agreement signed in 1974, have been suspended. 63. Because of constraints in their main export markets, several pro- ducers are exploring the possibilities of exporting production through foreign investment or transfer of technology. Two of the three harvester producers studied for the present report are currently selling technology to Brazil. One has licensed without participation, and two have financial interests in the licensee. Another firm has transferred technology for the production of hydraulic scrappers to a Brazilian firm, capitalizing its technical contribu- tion. Another two firms had projects to sell technology, one in Libya and the other in Venezuela. The transactions did not go through because in both cases the counterpart wanted to start production on a larger scale than that which the Argentinian firms consider viable. Finally, four other firms are negotiating the sale of technology to firms in other Latin American countries: two in Brazil (for silos and ploughs), one in Chile (for harvesters), and one in Venezuela (for ploughs). 64. The mechanism preferred by most of the firms involved in external transactions is to gain equity participation in a local producer through capitalization of royalties, and to export pieces that will be assembled by the partner according to the supplier's designs. These firms range in size from 250 to 600 employees, and all except one have at least one engineer 1/ One example of this is the use in Argentina of a scrapper that is pulled with the frame in the ground. In other Latin American countries a three- point scrapper is used. This means that locally unsold stocks cannot be exported. 2/ Federacion Industriales de la Provincia de Cordoba, op. cit.  - 24 - ANNEX I Page 24 and several technicians working in production and design. The existence within the firms of a relatively large group of engineers and technicians is a neces- sary condition for this type of external operation. Their services are needed to adapt designs to local conditions, to systematize the firm's knowledge and experience in a transferable way, and to train and provide technical assistance to the local partner. In fact, some of the firms that are presently in negotia- tions of this kind have already sent technicians to work with the prospective partner. Transfer of technology for the production of agricultural implements to other Latin American countries does not seem to be beyond the technical and organizational possibilities of many of the large firms in this sector. 4. Technical Development 65. The process of acquiring skills in this sector has been similar to the one followed by machine tools producers in Argentina. As with machine tools, technology here is incorporated mainly through product design and speci- fications. This is a free technology that can be acquired through imitation. On the other hand, production processes use well known techniques that are common to other mechanical engineering sectors. The same activities that gave rise to the skilled workers and metallurgic technicians that started machine tools production (military factories, railways), also comprised the origin of some of the people who started production of agricultural implements. How- ever, some producers started their manufacturing activities as a complement to farming activities. Here also the original acquisition of skills can be traced to the metal working backgrounds of some of the immigrants who settled in the agricultural regions. 66. As with machine tools, firms began by doing repair work and fabricat- ing spare parts for imported equipment. However, due to the fact that the designs are less sophisticated in this sector, and also more dependent on local characteristics (soils, forms of cultivation, products, etc.), the "copy" period was shorter. Also, from early in the development of the sector, local firms contributed with important adaptations and innovations in design. Thus, in 1921, two or three years after the first machines were imported, a local firm produced machines that could cut and thresh wheat and linen simulta- neously. In 1930 an Argentinian firm (SENOR) produced the first motorized harvester (cut and thresh) in the world. This was eight years before the first motorized harvester from abroad (Massey Harris from Canada) entered the country. In 1951 another firm (Roque Vasalli) produced the first corn har- vester in the world, and in 1962 a rice harvester. The innovativeness of the leading firms in this sector is evidenced by the fact that almost all the firms studied have taken patents in Argentina and other Latin American countries where they export. These patents cover key elements of the mechanisms but not the whole implement. 1/ 1/ In Santa Fe province alone, 10 of 30 firms have taken out patents in Argentina. A silo producer declared that he has 50 patents.  - 25 - ANNEX I Page 25 67. The location of these producers in small villages near the agricul- tural centers favored, from the beginning, the incorporation into the machines of the adaptations made by the agricultural workers themselves. This accounts for the versatility of Argentinian machines, which respond to the need of the farmers to maximize the use of each piece of equipment. An example of this is a locally designed harvester which, by changing some components, can be used to collect sorghum and corn. The technical development of Argentinian agricultural implements is related to the fact that almost from the beginning producers had to respond to the requirements of the export-oriented agricul- tural sector for increasingly efficient products. 68. As in the case of machine tools, design skills in the firms advanced with very little participation of engineers and outside technical help. Even today three of the eight leading firms studied for the present report do not have engineers on their design teams. (This includes the firm that made the first motorized harvester.) Only firms within the range of 350 to 650 em- ployees have engineers in design and production. Even in these big firms the number of engineers in design is fairly small. Three of the largest firms have only one engineer in design each (although two, five and six respectively in production); one has three engineers and another has four. Some of these firms have incorporated their engineers only in the last few years. Repre- sentatives of other firms mentioned that they want to hire engineers to systematize the accumulated experience. However, problems arise because the experienced technicians resent the presence of "inexperienced" engineers who usually earn higher salaries. The lack of technical personnel is a limiting factor for many of the smaller firms which "copy" the models produced by the larger ones. Without technical personnel they are unable to keep the spe- cifications of the copied implement when there are changes in the production processes or raw materials. 69. Because of the facility to copy designs, licensing of technology from abroad is insignificant. The study found only one joint venture between a local producer and a foreign technology supplier. In this case the foreign partner, which has 40% of the equity of the Argentinian firm, provided the technology for the production of hardened disks. Up until recently, this firm was the only producer of this kind of disk in the country. Now, however, Direccion General de Fabricaciones Militares produces a similar element. The same local producer who entered the joint venture is also the only one who reported production of a plough and a rotating cultivator under license. The local firm has made modifications to the original licensed design that, in turn, are being adapted by the licensor. 70. The only institutions providing technical services to the sector are IRAM (Instituto Argentino de Racionalizacion de Materiales) and INTA (National Institute of Agricultural Technology) and the testing laboratories of local universities. These services comprise, for the most part, testing of prototypes and materials. The institutions are not used in the solution of technical problems or to provide technical assistance. INTA has one testing center for prototypes in Castellar (located in the south of Buenos Aires province) and a number of experimental stations in different parts of the country. In the&Castellar center tests are carried out on prototypes,  - 26 - ANNEX I Page 26 but no certificates are issued as a guarantee of performance. Spokesmen for several firms mentioned that INTA's testing facilities are located too far from the regions where most of the equipment producers are established, and both equipment and technical personnel are inadequate. The firms that sell equipment through tractor concessionaires have their prototypes tested by the tractor producers themselves. 71. In addition to problems in technical assistance, representatives of most firms mentioned the lack of nearby high quality technical schools. Several recommended the creation of post-graduate courses for technicians with practical experience. The lack of technicians and skilled workers was mentioned by most as a major constraint to expansion. This problem is ag- gravated by firms' location in small towns. Many also mentioned lack of hous- ing as a limiting factor on recruitment of technicians from other parts of the country. 72. As in the case of machine tools, large producers of agricultural implements in Argentina have reached quality standards comparable to the standards of implements produced in other exporting countries. Quality im- provements are related not only to the demands made by the export-oriented agricultural sector, but also to the technical developments made in foundry and local machine tools production. Although production processes in the sector are relatively simple, between 5% and 20% of the total value of the equipment used is imported. (However, in the largest firm the figure is 50%.) This may be evidence that parts of the process require precision equipment. It may also indicate overcapitalization due to a previous policy of cheap loans. On the other hand, smaller producers in general use locally produced machines. In most of the larger firms, production is carried out in series or semi-series. Only one of eight plants studied still has batch production. 73. All the leading firms carry out quality control tests, and some of them have fully equipped laboratories for this task. However, according to a previous study, I/ a number of smaller firms only carry out performance tests of the finished equipment. The study mentioned found that quality con- trol is more widespread among producers of motorized equipment than among producers of other types. While with 40 of 42 motorized equipment products, quality control was carried out systematically, it was exercised with only 89 of 227 other agricultural implements. In only 27 of these 89 products were the tests conducted outside the firm (either by private laboratories, universities or ENET), Spokesmen for most of the leading firms declared that international norms are followed. These have been acquired directly because apparently IRAM does not have norms for agricultural implements. 74. As with machine tools, there is evidence of a movement towards less vertical integration. Several firms subcontract part of production 1/ Proyecto UGEPEX, op. cit.  - 27 - ANNEX I Page 27 including machining. Representatives of other more integrated firms mentioned that, due to the lack of skilled workers, future expansion will have to be based on subcontracting. The degree of subcontracting can be inferred from the fact that the value added ranges between 30% and 50% for the firms studied. Only two of seven firms still have their own foundries. The items that are most frequently acquired from other agents are components (hydraulic or pneu- matic), gears, motors, wheels and disks. (There are only two producers of disks.) VI. Effects of Policies on Exports and Technical Development 75. The first part of this section will discuss the impact that different policies and institutions have had on machine tools and agricultural equipment production, particularly with regard to technical development. In the second part, the discussion will center on the incentives to export complete plants. 76. Production of machine tools in Argentina has been stimulated by pro- tectionist policies. However, the high tariff rates and import surcharges in operation during the 1960s give an exaggerated impression of the protection granted to the sector. In fact, the 175% tariff rate of 1966 was modified several times until it was fixed in 1971 at 80%-90%. Part of the tariffs were geared not to protect, but to offset, the overvaluation of the currency. There was also a substantial redundancy in the 80% and 90% legal tariff rates operating until recently (tariff rates are now 45% and 50% for competing products and 5% for non-competing products), as can be inferred from the fact that for several types of machines internal prices were substantially lower than world prices plus tariff rates. 1/ Furthermore, Berlinski's study of effective protection in Argentina 2/ shows that although by 1976 nominal tariff rates were 25% (adjusted to reflect the price ratio between domestic and world products), the effective tariff rate was only 10% (see Table 18). Part of the difference between the nominal and effective rates of protection is due to the fact that locally produced steel has an overpricing of about 50% compared to imported steel. 3/ 1/ Valbe, H., "Estudio acerca del Gravamen Aduanero requerido por un Grupo representativo de Maquinas Herramientas," (quoted by E. Amadeo, op. cit.). Valbe found that the internal prices of parallel lathes, gun lathes and balances were lower by 31%, 18% and 81%, respectively, than foreign prices plus tariffs. 2/ Berlinski, J. and D. Schydlowsky, "Incentives for Industrialization in Argentina", mimeo, Washington 1976. 3/ Most of the large machine tools producers interviewed for the present report remarked that 45% is an adequate level of protection, one which roughly compensates for the high cost of locally produced steel. How- ever, they generally opposed any further tariff reduction.  - 28 - ANNEX I Page 28 77. The same argument can be advanced in relation to the agricultural implements sector. As mentioned previously, protection for this sector be- gan in 1951 when it was declared of national interest. Up until 1970, tariff rates for products competing with locally produced equipment ranged between 90% and 100% of CIF prices, and 20% for non-competing products. During 1970, several modifications reduced the legal tariff rate for competing pro- ducts to 60% and, for non-competing products, to 5%. Berlinski shows that the tariff rates were highly redundant: that is, if adjustments are made to reflect price ratios between domestic and foreign products, the nominal tariff rates by 1976 were about 25%. Furthermore, Berlinski shows that effective tariff rates for 1976 were strongly negative (-27%). 1/ 78. It can be argued that more important than tariff barriers in the development of both sectors has been the natural protection accorded Argentinian industry by the country's location far from the big industrial centers, as well as the size and sophistication of the internal market which demands high quality products (e.g., automobile industry and export agriculture). 79. Incentives have undoubtedly stimulated exports of machine tools and agricultural implements. Representatives of most firms in both sectors mentioned that rebates of 25% are enough to compensate for higher internal costs. However, several mentioned strong competition from Brazilian firms (and Eastern European firms) which get rebates of up to 40%. An important reason for the positive attitude of the producers towards tariff rates and export rebates is their lack of selectivity and their automatic character, which is preferable to the earlier policy in which each rebate was established after an individual analysis. Most entrepreneurs emphasized the importance of knowing the benefits beforehand and not having to depend on discretionary decision by government authorities. 80. Another policy of some impact in both sectors has been the long- term credits for equipment purchases provided by BANADE. Up until recently, these credits were highly subsidized, with interest rates well below the rate of inflation. Several of the largest firms studied for the present report have obtained long-term credits under these highly favorable conditions. These credits probably account for the overinvestment detected in a number of producers. However, the number of smaller firms that have never made use of these subsidized credits, preferring instead self finance or suppliers' credits, is remarkable. Among these firms there are two small producers of special machines which have purchased expensive precision equipment. The fact that most of the owners of these firms have the same social origin (Italian immigrants) seems to indicate that their distrust for banks may partly be a culturally influenced attitude. 81. The reaction of the entrepreneurs to the new policy of charging positive interest rates for long-term credits is mixed. Although most of the entrepreneurs agree that the interest rates should be positive, they 1/ Berlinski, op. cit.  - 29 - ANNEX I Page 29 tend to consider actual rates too high; many of them have suspended their expansion plans in the expectation that interest rates will eventually fall. In addition to complaints about interest rates, a series of obstacles and difficulties relevant to the use of BANADE's credits were mentioned. Among them was the fact that the payments, but not the amounts to be borrowed, are indexed. Thus, in the time that it takes to process a credit request, the price of the machine to be imported has increased. Other criticisms are related to what are seen as institutional inefficiencies in processing the requests. However, it is likely that the firms themselves which do not have their books in order, or are not up-to-date with taxes and social security contributions, are responsible for some of the delays. 82. None of the firms reviewed have ever used any of BANADE's special credits, such as those available for prototype development, pre-investment studies, and construction of pilot plants. Several of the producers did not know of the existence of these credits. On the other hand, the few firms that have applied for them (particularly for prototype development) have not been accepted. One of the reasons alleged is the lack of technical capability within BANADE to distinguish a prototype. Credits for prototype development are potentially very useful in these two sectors, where technical development occurs in the form of changes in design. Another policy instrument which had important potential for stimulating technological development was the tax reductions for research expenditures. This policy lasted a very short time and had very little impact. 83. Finally, the industrial promotion law, whose objective is to stimulate industrial decentralization through tax incentives, does not seem to have affected the location of firms in these two sectors. In fact, loca- tion was a key aspect in the development of these industries and they were already decentralized 1/ when legislation was enacted in 1973. Further decentralization is limited because of the difficulties that the firms face in obtaining qualified labor outside the urban centers where they are at present concentrated. Many firms, particularly in the province of Buenos Aires, face the dilemma of inability to expand in their present locations and the lack of an appropriate work force if they locate elsewhere. Under these conditions, the effect of the law can be the opposite of what was intended. An example of this is the case of a firm which, to change loca- tion, has had to automate production to the extent that its requirements of skilled labor are minimal. 84. There are several incentives for the export of complete plants from Argentina. The most important is a drawback of 40% (decree No. 2786 October 1975) of the price of the contract, excluding civil work abroad but including technical services such as design, engineering, operational 1/ Machine tool producers of Piedmontese origin settled in San Francisco, probably because of the location of the military factories (Fabricaciones Militares). Similarly, producers of agricultural implements settled near the agricultural regions.  - 30 - ANNEX I Page 30 methods and start-up assistance. Originally, this incentive did not cover non-manufacturing plants like hospitals or other civil works. (Eight of the 44 plants exported between 1975 and 1976 belong to this category.) How- ever, since May 1977 non-manufacturing projects have been included in the promotional export financing law (R.F. 21). Technical services, research and studies can be now financed up to 90% of their FOB value; the amount and the terms are decided on a case-by-case basis by the Central Bank. 85. The design of promotional legislation thus far has not included guarantees. The issue of guarantees may become an important constraint due to the small size of exporting firms in relation to the magnitude of their contracts abroad. The signature on a sales agreement implies a gua- rantee of 10% (a bid-bond or a performance bond) that must be endorsed by a local bank. 1/ Medium size firms and firms that are new to the export business may have difficulties locating a bank ready to take the risk. Other countries engaged in the promotion of this kind of exports have given much more importance to the guarantees issue. 86. A recent example is Brazil, where government-endorsed bid-bonds grew from US$22 million in 1974 to US$38.3 million in the first nine months of 1976; performance bonds that did not exist in 1974 reached US$2.7 mil- lion during the same period of 1976. 2/ Recent Brazilian legislation has other features that are lacking in Argentinian legislation. The most important is that it stimulates the integration of the various agents in- volved in the export of complete plants. Thus, if an engineering firm wins a bid in another country, the producer that sells the equipment to the engineering firm gets a rebate as if it were exporting directly (decree of October 1976). In this respect, Brazilian legislation follows the French system of promoting the creation of links among different agents 3/ (owners of technology, equipment producers, banks, engineering firms and government agencies) for the export of complete plants to LDCs. VII. Conclusions and Recommendations 87. The analysis of the experience of process industries and, in more detail, machine tools and agricultural implement producers, confirms the proposition that the technical development reached by Argentinian firms is an important explanatory factor of their export performance, not only in products but also in technology. 1/ In case of failure to deliver, or delay in the start-up time, the client retains the guarantee. 2/ Quoted by Katz and Ablin, op. cit. 3/ Judet J. Perrin, "Engineering Technologie et Fonction Economique," OCDE, Feb. 1976.  - 31 - ANNEX I Page 31 88. It is evident that industrialization in Argentina has been accom- panied by a process of gradual acquisition of technical skills and hence improvements in efficiency and productivity. 1/ Because industrialization (and learning) started in Argentina earlier than in other Latin American countries, Argentinian exports enjoy a competitive position within the region. Their competitiveness is further favored by subregional agreements and natural geographical protection. 89. The accumulated technical knowledge in the hands of Argentinian firms constitutes by itself an economic asset which has a growing market in other LDCs. The increasing number of turnkey plants exported, together with the growing involvement of local producers in foreign investment and licensing arrangements, are evidence of the potential of the market for Argentinian technology. 90. This technology seems to be particularly appropriate to the require- ments of the client countries. In addition to the fact that the demand it- self indicates it is somehow appropriate, there is evidence that the exported technology is appropriate because it has been previously adapted to the characteristics of the Argentinian economy, among them the low cost of labor and relatively small markets--characteristics which are, in many respects, shared by Argentina's clients. 91. It is interesting to note that not only the adapted technology but also the capacity to adapt have become economic assets with a market in other LDCs. This is particularly evident in the case of turnkey plants where the plants are designed according to the client's needs. Previous experience in adaptive innovation plus low costs of engineering time greatly favor the com- petitive position of Argentinian exporters of turnkey plants. 92. Local acquisition of technical skills has occurred whether the original technology was transferred through a subsidiary of a multinational firm or through licensing agreements (as is frequently the case in the process industries); whether it was "copied" or originally developed by locally owned firms (as happens in the mechanical engineering sector). In this respect, Argentina seems to have had a different experience from other LDCs, where the transfer of technology through subsidiaries or licensing agreement has had less 1/ In all the sectors analyzed in this annex, technical learning has taken place as the result of an explicit strategy within the firms to exchange resources for tasks such as technical assistance to pro- duction, "trouble shooting," product design, etc. Learning efforts are related to the need to adapt product designs and production pro- cesses to local conditions. Results of this technical effort are minor innovations that modify the original technology, making it more simple, more labor intensive or more appropriate for smaller scales of production. (J. Katz, "Creacion de Tecnologia en el sector manufacturero Argentino," Monografia de Trabajo No. 1, Programe BID/CEPAL, Bs. As, Argentina, 1976).  - 32 - ANNEX I Page 32 impact on local learning. There are several factors whose combined effect has contributed to local learning and which may not be present in other LDCs. Some of these factors are: (a) Early availability of a skilled labor force, including highly qualified engineers and technicians. The original training of this labor force (including those who later became entrepreneurs) took place either in the country of origin of the immigrant population, or in large local concerns such as the railways and military factories. Particularly important is the low cost of engineering time which makes adaptations economically meaningful. (b) Protection from competition. As was demonstrated earlier, although effective protection was not particularly high (actually negative in the case of agricultural implements), there were other protective factors, such as lack of foreign exchange for imports and the country's geographic position. (c) A relatively large internal market demanding high quality products (e.g., the automobile industry in the case of machine tools and export agriculture in the case of agri- cultural implements). 93. The main conclusion of this annex is that up to the present in Argentina the supply conditions are able to meet, to a great extent, the conditions of both internal and external demand. 1/ But what of the future? What will future conditions of demand be for the products that Argentina now exports? To what extent will Argentinian producers be able to satisfy these conditions? There are three elements to be considered in pursuing these questions: (a) tendencies of industrial production in Argentina's client countries; (b) tendencies of industrial production in developed countries; and (c) tendencies of industrial production in Argentina. 94. With respect to the first point, there is evidence that other countries in Latin America are now reaching a stage of industrialization that Argentina achieved years ago. This means that the most advanced will start production of some of the products that Argentina exports now. The growing demand for Argentinian technology in areas such as food industries, chemicals and mechanical engineering is evidence of this trend. To be able to continue exporting, Argentinian producers will have to find a niche higher up in the market, with products which are relatively more sophisticated. Also, internally the market tends to demand increasingly sophisticated pro- ducts. Thus, Argentinian firms must develop further their technical capa- bility, create better designs and improve production processes through more organized R&D efforts. 1/ As was indicated in paragraph 27, local machine tool producers are not able to produce the kind of sophisticated equipment which at present represents 50% of the local market.  - 33 - ANNEX I Page 33 95. Competition from products produced in developed countries will also oblige Argentinian firms to improve technologically. Argentinian products, such as machine tools and other capital goods now appropriate for other LDCs (because of their cost advantages or other characteristics) must be continuously improved. If they are not, developments in similar machines produced in developed countries 1/ will cause them, sooner or later, to lose their competitiveness, not only in export markets but also internally. 2/ 96. Argentinian firms, at least the larger ones, can face this challenge. They have well staffed groups in design, R&D, and engineering work. However, may of the small and medium firms are reaching the limits of their technical development. These firms need not only changes in in- ternal organization, but also a great deal of assistance from technical institutions in solving problems that demand skills and resources beyond the individual firm's capabilities. Even the larger firms need to be more receptive to closer cooperation with specialized technological institutions both inside and outside the country. 3/ 97. Argentinian producers will need technical support, not only for the "creation" of technology but also for its transfer to other LDCs, if this type of export is to increase. There are certain economies of scale associated with technology exports. Large firms may have the necessary financial and human resources; for small firms, however, transfers are limited by the fact that technical skills are usually embodied in the firm's technicians and rarely systematized in a transferable form. Any transfer operation would involve sending abroad several of the technical staff that are needed to continue production at home. To eliminate this constraint, it will be necessary to stimulate the development of independent engineering organizations able to act as links among the different agents involved in technology transfer operations. Such development could be fostered through the use of the services of local engineering firms by the government and international agencies (including the World Bank) for project design and implementation. 1/ Such improvements would include those that reduce cost or labor requirements, or increase speed, precision, versatility, etc. 2/ Howard Pack makes a similar point in relation to Korea's production of capital goods in "Report on the Korean Capital Goods Sector," (forthcoming World Bank report). 3/ As was discussed earlier with regard to the process industries and mechanical engineering sectors, technical development has taken place in Argentinian firms with negligible use of local technical agents such as research laboratories and independent engineering firms. (See paragraphs 36, 43, 44, 68 and 70.)  ANNEX I - 34 - Page 34 98. The transfer of Argentinian technology to other LDCs would also be stimulated if existing incentives for turnkey exports were complemented by a system in which official institutions were enabled to endorse the bid- bonds and performance bonds signed by technology suppliers. 99. Another measure to stimulate exports would involve the creation of export consortia, representing the interests of small and medium firms, with offices in the main export markets (Latin American and African countries). This is an area of institution building where the government should parti- cipate more actively. Promotion of Technical Development in Industry 100. A major conclusion of the present study is that local technological development will become an important factor in the growth and export performance of some Argentinian industries. To grow and compete, Argentinian firms will have to develop further their technical capability in product and proce6s technology. The potential to achieve this is present in the well-staffed groups doing design, R&D and engineering work in the most important firms in each sector. Furthermore, industry can benefit from a well-developed technical infrastructure, I/ including universities, government and private research institutes. However, it is unlikely that market forces alone will induce the kind of technological development that has been sketched in this report. In matters of learning and technical development social gains usually outweigh private gains. For this reason, some degree of government inter- vention is necessary in order to foster, promote and strengthen technical capability. 101. An important recommendation of this annex is that the promotion of further technical development in some export-oriented industrial sectors should become explicit government policy in Argentina. The main elements of this policy should be defined through close discussion between government and industry relevant to the objectives of each sector and the meaning of these objectives in terms of technological development. 102. In the case of machine tools, for example, a panel should be convened to discuss a sectoral development program. The panel would be composed of government and industry representatives (including user industries), with the participation of CIMHER as a technical adviser, and would focus on issues 1/ There are other research institutes besides CIMHER that are engaged in research in the area of machine tools. Amadeo et. al, op. cit., mention that by 1977 six universities had requested funding from the National Council for Science and Technology (CONACYT) to reinforce or initiate a total of 16 research projects in this field. They also mention other institutions, such as the Comision Nacional de Estudios Geo-Heliofisicos. More detailed studies of these and other institutions, and of ways to link them to the productive sector, should be made.  - 35 - ANNEX I Page 35 such as (a) the sector's future development in the light of present trends in local and export markets (i.e., the types of machines which should be produced for local and export markets); (b) the kinds of technical skills which should be developed to achieve sector objectives; and (c) the impli- cations of development of these technical skills in terms of training pro- grams, local R&D programs, and transfer of technology from abroad (proprietary and non-proprietary). In the light of these discussions, the panel should define priorities to guide the government's explicit technology policy in this sector. 1/ 103. The government can use a wide range of policy instruments to promote technical development in industry. Examples of direct policy instruments that might be utilized are: special credits for R&D expendi- ture, pilot plants, or prototype development; government funding of R&D in government research institutes, universities or in industry itself; direct technical assistance in the form of a technical information system, testing facilities, etc. Other policy instruments that can be applied to affect the technological behavior of enterprises indirectly are: fiscal incentives, tariff barriers, export incentives, long-term lending policies, government enterprises, procurement policies, etc. 104. Direct policy instruments to promote technical development in industry have an important advantage over indirect policy instruments. Their advantage is that they can be applied selectively, following the priorities that have been defined in each particular sector. It is evident that selectivity in the application of a policy instrument is desirable in a situation where resources are scarce, and the technological needs of pro- ducers, even within the same sector, are different. 1/ Countries such as Japan have successfully implemented an explicit policy to upgrade local production of machinery. A Law for Promoting Machine Manufacturing Industries was enforced in Japan in 1956, and renewed several times later. Through this law, the government (1) de- signated certain machinery and parts whose quality and production costs were to be improved; (2) fixed concrete targets to be attained within a given period of time; (3) provided special loans and granted taxa- tion privileges to encourage investment in order to help attain such targets; and (4) exempted from certain regulations (the Anti-monopoly Law) firms which wanted to scale up their production through mergers. Ir. addition to this type of sector-specific policies, the Japanese government has implemented other more general policies to encourage the rationalization and modernization of small firms by (1) stimulating the organization of small enterprises into cooperatives and giving them financial support; (2) giving technical guidance, free of charge (through the Diagnosis Program) and (3) giving financial support with special funds for modernization and rationalization of production facilities. (See Watanabe S., "Technological Linkages Between Formal and Informal Sectors of Manufacturing Industries," ILO Technology and Employment Programme, Working Paper WEP 2-22/Wp34).  - 36 - ANNEX I Page 36 105. Indirect policy instruments to promote technical development in industry can also be made selective. For example, special fiscal treatment or export incentives could be granted to firms that undertake certain lines of production, or devote resources to R&D above certain levels. However, to make these instruments selective would require a complicated and expen- sive infrastructure to evaluate and decide on the application of the instru- ment, practically in each case. Another disadvantage in the use of some indirect policy instruments to promote technical development (e.g., fiscal incentives, tariff barriers, etc.) is that they may make economic inefficiency profitable, thereby eliminating the basic incentives of the firms to develop further technologically. The application of this type of instrument usually involves a cost to the government in the form of loss of fiscal revenues or payment subsidies. It makes more sense to use government resources to im- prove the sector's technological capacity, while maintaining a more competi- tive market situation. The result should be to foster a more efficient sector, better able to compete in external markets. Suggestions relevant to the use of government resources to promote industry's technological development will be presented below. 106. Even if the selective use of indirect instruments to promote tech- nical development is limited, it is important to ensure that their applica- tion, in response to other policy objectives, is not an obstacle to the tech- nological development of some industries. This is an area in which more research is necessary. For example, an evaluation should be made of the extent to which the present system of policies regarding the exchange rate and tariff protection inhibits local production of certain types of relatively sophisticated machines now being imported. If the production of these machines is considered desirable, there may be an argument for finding a means to com- pensate for the disincentives generated by these policies. Similarly, it would be desirable to correct the negative effective protection that some sectors, such as agricultural implements, have at present. 107. The most direct way to promote local research in the priority areas is through government-funded R&D programs in government laboratories, universities and industry itself. In the case of the machine tools sector, CIMHER (an interesting combination of public and private interests) will probably absorb most of the government-financed R&D in the sector. It is important that this institution be provided with adequate manpower and equip- ment to accomplish the objectives for which it was created. Similar insti- tutions should be created in connection with other industries. Although mixed institutions such as CIMHER will play an important role in government- financed industrial R&D, there is scope for more government-financed re- search in industry itself, where the learning effects of the research can be most important. Several of the larger firms could, if helped, undertake more extensive work in product design, components development, improvements in metal-cutting practices, etc. Government funding of R&D in the private sector could take the form of research contracts (as in the US), government payment of salaries for some researchers in industry; and government sub- sidies to cover part of the cost of a research project undertaken by a firm or group of firms. Government-financed R&D in industry would require a  - 37 - ANNEX I Page 37 clear definition of priorities as well as some capacity on the part of the donor agent to evaluate the quality of the research. In the machine tools sector this monitoring role could be played by CIMHER. 108. Another form of assistance to industrial research is through special credits for R&D expenditures, pilot plants, prototype development and acquisition of quality control equipment. Special credit lines of this type already exist in BANADE, but they have not been utilized, apparently because the financing agent lacks the technical capability to evaluate projects of this nature. Closer collaboration between BANADE and technical agents such as CIMHER and INTA would facilitate the use of these credit lines. Preferen- tial credit terms should be given to R&D projects within the priority areas identified through close discussions between government and industry represen- tatives. 109. There is wide scope for improvements in the area of technical assistance to industry. Appropriate government or mixed institutions should provide assistance in areas such as (a) technical information, (b) technical services, and (c) industrial extension (for the smaller firms). In the machine tools sector CIMHER was designed to provide assistance in the area of technical services, in the main through testing facilities not avail- able in other institutions. However, nothing exists in the area of technical information. A technical information data system should be developed within CIMHER for the machine tool industries. Similar systems should be developed in the case of other industrial sectors (e.g., the food processing indus- tries). Using their own data banks and the network of technical information systems that already exists in other countries, these systems should be able to process technical questions posed by local enterprises. They should pro- vide translation services for journals, abstracts, technical books and other written material. However, in a situation where firms of very different technological development coexist (as is the case in machine tools and agricultural implements), the technical information system should not limit its activities to responding to interested firms. The system must play an active role in disseminating information to the small and medium producers. In this respect the system should be utilized as an important tool by industrial extension agents. 110. As copying is a very important source of technical development in the mechanical engineering sector, the technical assistance program to the sector should include technical and legal advice to firms adapting machines produced in other countries. A government institution (in the case of machine tools, probably CIMHER) should be able to buy, or collabo- rate in the acquisition of, foreign machines and components of advanced design. These would then be disassembled and studied by local producers interested in the production of machines with similar characteristics. 111. The official institutions where machines and agricultural implements are tested should be able to give certificates defining the specifications of the products tested. These certificates would be issued after the appropriate tests had been made, and would serve as a guarantee for national and international users of the products.  - 38 - ANNEX I Page 38 112. The level of technical development achieved by Argentinian industry is, to a great extent, a result of the high level of skills of its population. However, there are signs that indicate that the training institutes are not providing the kind of skills needed by industry. It would be desirable to have industry representatives participate in the design of graduate and post- graduate courses for technicians. Not only should the quality and relevance of the training be analyzed, but also the criteria for the location of technical schools. At present, it appears that technical schools are located accord- ing to density of population. However, industrial concentration in an area may not be related to the size of the population, as it is evident in the case of San Francisco. 113. Although licensing has played a very limited role in the develop- ment of the machine tools and agricultural implement sectors, it will become more important as producers enter into more sophisticated lines of production. Because of lack of experience, prospective buyers may not be aware of the costs and risks involved in the transfer. The institution which registers technology contracts should advise on these issues and guide the transfer operation in a manner which maximizes learning possibil- ities for the transferee and avoids excessive costs. 114. Special credit lines should be established to assist the develop- ment of those activities that present high externalities improving the efficiency of other industrial sectors, such as production of special machines and components in the machine tools sector. A development bank should also assume the risk of increased lending to small enterprises, not only for their employment potential, but also for their importance as training grounds for entrepreneurs, technicians and skilled workers. The experience of machine tools and agricultural implements producers is a clear example of the importance that small firms have had in the sector's learning process. The government should also study more carefully the experience of industrial parks, such as the one in San Francisco, in order to decide whether other industrial parks should be promoted to pro- vide technical support for small scale enterprises.  Table 1: EXPORTSOF RELATIVELY SOPHISTICATED MANUFACTURES AS A PROPORTION OF TOTAL INDUSTRIAL EXPORTS (%) Items 1969 1974 1969/1974 Organic chemical 8.8 4.7 5.6 Iron, steel and products 18.3 21.6 21.0 Boilers, machines and mechanical artifacts 21.4 22.1 23.7 Electric and electronic machines 4.9 5.8 5.3 Vehicles, automobiles, tractors 5.9 21.3 16.4 Total 59.3 75.5 72.0 Total industrial exports (US$ M) 166 836 2,278 Total manufacturing exports* (US$ M) 507 1,867 5,655 Note: * includes industrial exports and manufactured agricultural products. Source: Katz and Ablin, op. cit. e : PZ  Table 2: ARGENTINA: EXPORTS OF A SELECTED GROUP OF MANUFACTURERS (thol.sand of US $) Cumulated Enterprises 1969 1970 1971 1972 1973 1974 Exports 1. Olivetti 8,514 13,035 16,102 16,003 18,059 18,961 90,674 2. Acindar 5,003 4,350 3,698 7,496 38,917 21,761 81,225 3. Fiat 1,374 1,380 5,072 10,660 25,681 36,901 81,068 4. Dalmine 8,479 8,442 7,108 8,984 11,969 21,952 66,934 5. IBM 6,913 7,215 7,299 7,502 13,904 20,038 62,871 6. Mercedes Benz 863 1,586 839 11,531 19,792 23,809 58,420 7. Ford 2,694 1,441 2,301 1,218 6,642 19,408 33,704 4 8. Esso 4,800 5,029 6,698 4,195 4,192 7,437 32,351 9. Pasa Petroquimica 6,159 4,494 4,434 3,254 4,699 8,367 31,407 10. General Motors 954 1,346 2,664 3,503 4,894 8,875 22,236 11. Citroen 1,250 1,894 4,182 5,812 3,452 5,329 21,919 12. Squibb 3,216 3,049 3,207 3,674 3,066 4,753 20,965 13. Cyanamid 1,458 2,805 2,754 3,101 4,396 5,991 20,505 14. Roque Vassalli 1,268 1,676 1,169 2,917 3,490 6,361 16,881 15. Ducilo 2,568 1,790 1,673 3,024 5,178 2,159 16,392 16. Siam di Tella 1,695 2,534* 2,534* 2,534* 3,374 3,200 15,871 17. Fapesa 1,087 1,587 1,714 3,152 4,590 3,491 15,621 18. Philips 1,145 1,366 1,452 1,647 2,413 4,323 12,346 19. Standard Electric 1,369 1,230* 1,230* 1,092 2,912 4,198 12,031 20. Compania Quimica 864 1,256 1,247 1,305 1,584 3,946 10,202 TOTAL 61,673 67,505 77,377 102,604 183,204 231,260 723,623 Source: Katz and Ablin, op. cit. Note * Figures estimated by the Authors ( H x4 ZOH  - 41 - ANNEX I Table 3: AR(ENTINA: EXPORT OF COMPLETE PLANTS AND CIVIL WORKS, 1973-1977 Page 41 Value Type of Plant Destination Year (Thousands of US$) 1. Factory for vegetable oils Bolivia 1973 5,525 2. Plant for the production of cooked meat and concentrate Brazil 1973 200 3. Automatic telephone control, and external communication plant Ecuador 1973 679 4. Comprehensive communications system for public services Chile 1973 2,829 5. Meat processing plant (slaughter house and freezing) Cuba 1974 12,500 6. Integrated bread producing plant Cuba 1974 2,900 7. Glicerine plant Mexico 1974 90 8. Port warehouses and metallic structure Cuba 1974 6,775 9. Metallic structures and silos for poultry farm Cuba 1974 15,940 10. Metallic silos with integrated conveyors Cuba 1974 2,829 11. Air conditioning and control heating for bank building Paraguay 1974 90 12. Citric fruits processing plant Cuba 1974 6,200 13. Two plants for honey production Cuba 1975 1,490 14. Plant for vegetable oil production (cotton seeds) Bolivia 1975 4,000 15. Terminals for supply, distribution and pump of liquified gas Ecuador 1975 1,998 16. Pipeline and pump station Peru 1975 120,000 17. Plant for the production of antibiotics Bolivia 1975 220 18. Turn-key airport Paraguay 1975 52,000 19. Plant for water treatment for industrial use Uruguay 1975 47 20. Plant for processing citric fruits Bolivia 1976 8,810 21. Plant for production and packing of species Cuba 1976 1,441 22. Complete plant for oil solvent extraction and pelletization plant for the production of sunflowers and soja cakes Uruguay 1976 746 23. Plant for the production of lead oxides Venezuela 1976 147 24. Plant for lead melting and recouperation Venezuela 1976 106 25. Production of edstings for electric works Bolivia 2976 146 26. Bread producing Chile 1976 115 27. Poultry slaughter house and processing plant Paraguay 1976 189 28. Factory for the production of balanced powdered food Bolivia 1976 239 29. Plant for fat matter refining Chile 1976 286 30. Plant for grain processing and storing Uruguay 1976 483 31. Plant for the extraction of active components from vegetables Honduras 1976 450 32. Powdered milk plant Uruguay 1977 270 33. Pesticides plant Bolivia 1977 45,000 34. Turn-key Hospital Ivory Coast 1977 46,000 35. Plant for the production of plastic disposable syringes 1977* 296 36. Plant for urea and formaldehyde drying 1977* 540 37. Plant for oil extraction 1977* 720 38. Plant for the production of sulfuric acid and aluminum sulphate 1977* 1,006 39. Plant for the production of industrial paper trays 1977* 434 40. Plant for the processing of tubes for cordan gears 1977* 82 41. Anidine plant 1977* 330 42. Plant for the production of pharmacological drugs 1977* 43. Plant for the production of instant soluble coffee 1977* 1,570 44. Plant for the production of fish proteins - - TOTAL 345,718 Sources: Katz and Ablin;"Export of Complete Plants," op. cit. Secretaria de Comercio Exterior Note: *The date may be 1976 or 1977.  Table 4: EXPORTS OF COMPLETE PLANTS: AVERAGE VALUE OF PLANTS EXPORTED BY OWNERSHIP OF FIRM Total Average value of % Value Number Number of Contracts of Total of Plants Exporter'.s equity ownership of Firms Contracts (US $ '000s) Contracts (US $ '000s) Locally owned 20 28 168,506 49.45 6,018 Subsidiaries of multinational firms 5 6 172,236 50.55 28,706 Note: Includes the 34 first plants of Table 2. Source: Katz and Ablin, op. cit. D w