FILE COPY. BrazilFILCP. Protection and Competitiveness of the Capital Goods Producing Industries FOR OFFICIAL USE ONLY. Public Disclosure Authorized

Transcription

1 Public Disclosure Authorized Report No BR BrazilFILCP FILE COPY Protection and Competitiveness of the Capital Goods Producing Industries July 21, 1980 Projects Department Latin America and the Caribbean Regional Office FOR OFFICIAL USE ONLY Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized 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.

2 CURRENCY EQUIVALENTS Currency amounts as expressed in Brazilian Cruzeiros (Cr$) and US Dollars (US$) May 14, 1980 Selling Rate US$1.00 = Cr$ Cr$ 1.00 = US$ LIST OF ACRONYMS BEFIEX BNDE CACEX CDE CDI CIEF CONSIDER CPA EMBRAMEC FINAME FINEP FIPE IBGE TCM IPI ISIC LAFTA LDCs NMB SUDENE Special Program of Fiscal Incentives for Exporters National Bank for Economic Development Foreign Trade Department of the Banco do Brasil Conselho de Desenvolvimento Economico Conselho de Desenvolvimento Industrial Centro de Informacoes Economico Fiscais Government Steel Council Conselho de Politica Aduaneira Brazilian Mechanical Engineering Enterprise Special Fund for Industrial Financing Financiadora de Estudos e Projetos Fundacao Instituto de Pesquisas Economicas (Sao Paulo) Instituto Brasileiro de Geografia e Estatistica State Value Added Tax Federal Industrial Product Tax International Standard Industrial Classification Latin American Free Trade Association Less Developed Countries Brazilian External Trade Classification Agency for the Development of the Northeast

3 FOR OFFICIAL USE ONLY PREFACE This report is based on a mission which visited Brazil in July- August A draft of the report was discussed with Brazilian Government officials in May 1979 and, on the basis of the comments received, a revision of the draft was undertaken. In addition, the revision took into account the new policies of the government announced in January 1979 which were designed to modify exchange rates, export incentives and import restrictions over a four-year period. In December 1979, the Brazilian Government announced further changes in commercial and exchange rate policies which put into place immediately the measures which earlier had been expected to be achieved over a longer period of time. These changes are not reflected in this report. This report is now being published as redrafted in mid-1979 in the belief that the analysis undertaken and the results obtained may be of wider interest for those studying Brazilian economic development as well as for other countries considering the sectoral and subsectoral impact of commercial policies. 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.

4

5 BRAZIL PROTECTION AND COMPETITIVENESS OF THE CAPITAL GOODS PRODUCING INDUSTRIES Page No. SLIMARY OF PRINCIPAL CONCLUSIONS AND RECOMMENDATIONS... i - viii I. PERFORMANCE OF THE SUBSECTOR AND POLICY DEVELOPMENT... 1 A. Growth of the Subsector through B. Developments in the period C. Situation in D. Formulation of Policy for the Subsector... 7 II. IMPORT RESTRICTIONS AND TAX INCENTIVES A. Introduction B. Exchange Rate Policy C. Import Restrictions D. Fiscal Incentives E. Financial Incentives F. Protection of Intermediate Inputs III. ESTIMATING NET PROTECTION A. Basic Estimates B. Indications of Competitiveness IV. PROSPECTS AND PRINCIPAL POLICY RECOMMENDATIONS ANNEXES I. Some Conceptional Notes on Protection of the Capital Goods Industries in Brazil II. Notes on the Measurement of Protection III. Credit Availability and Subsidies This report is based on the findings of a mission which visited Brazil in July-August The mission comprised Messrs. Baskind, Costa and Nogales (LCPI2) and Tyler (Consultant).

6 Table of Contents (Continued) STATISTICAL APPENDICES Table I Table 2 Table 3 Table 4 Table 5 Table 6 Table 7 Table 8 Indices of Industrial Production Imports and Exports (FOB) Consumption Estimates GDP and Gross Investment Capital Goods Exports by Country of Destination 1976 Capital Goods Exports Share of Capital Goods Imports by End-Use Sectors Tariff Protection for the Brazilian Capital Goods Industries, 1974 Table 9 CACEX Participation Agreements, By Sector, 1977 Table 10 Nominal Subsidies to Capital Goods Exports, 1975 Table 11 CDI Approvals - Distribution by Amount of Fixed Investment, 1974 Table 12 Distribution by Branch of Investments Approved Under BEFIEX Program,

7 SUMMARY OF PRINCIPAL CONCLUSIONS AND RECOMMENDATIONS I. Performance of the Subsector and Policy Development i. The capital goods producing industries have played an important role in the growth of Brazil's industrial sector since the earliest stages of its modern development, beginning at the end of the 19th Century. The origins of these activities are found in the repair shops established at that time to service an extensive agricultural export economy; they ensured adequate functioning of the corresponding large transport system and simple commodity processing equipment. These shops gradually undertook the production first of spare parts for the equipment and later of basic pieces of machinery. For the most part, the growth of the subsector has not been stimulated by protection or other explicit government policies, but rather has flourished in the face of competition from imports, which during most of the period were accorded low tariffs and, at times, preferential exchange rate treatment. ii. These activities have gained in importance since the Second World War with their share in total industrial value added rising from 5% in 1949 to 15% in Domestic production accounted for about two-thirds of consumption in the former year and rose to about four-fifths in the latter. iii. The 1950s represented a period of substantial deepening of the structure of the Brazilian capital goods industries. Although some industrial policy measures date from the beginning of the 20th Century, it was not until the 1950s that a more explicit industrial policy began to emerge. Within the context of an overall macro-economic policy formulation which emphasized import substitution, priority was given to the development of specific subsectors, imports of which were considered critical, rather than to elaborating a general sectoral strategy. The principal manufacturing subsectors included passenger automobiles but otherwise were predominantly producers of capital goods. While there were major changes in import control mechanisms during this period, capital goods imports were given preferential treatment under all the systems. Moreover, the level of these imports was sustained by the availability of foreign financing. Nevertheless, with the emergence of new product lines, the share of imports in total domestic supply of capital goods declined. iv. By the beginning of the 1960s, evidence emerged of over-investment in the subsector with consequent under-utilization of capacity. This was attributed to over-optimistic expectations of the continued accelerated growth of the economy, industry in particular. With the recession in the mid-1960s, whose impact was sustained until the end of that decade, output growth of the subsector until 1970 was uneven, although continuing to show a moderate upward trend. From 1970 through 1976, expansion of the subsector was particularly rapid, well in excess of overall manufacturing growth, so that its share in the industrial sector's value added rose from 11% in 1969 to 16% in v. The expansion of the subsector during that period was linked closely to the very substantial increases in gross domestic investment which took place from the end of the 1960s. The economic reforms introduced by the

8 - ii - government in were designed primarily to open up the economy. In response to those policies and buoyant foreign demand, total exports which had been stagnant for over a decade rose rapidly, while a general political and economic climate favored inflows of foreign capital. As a consequence, liberal detrrnd management was followed and, in particular, important incentives were Jiven to investment. vi. Imports of capital goods were favored through virtually indiscriminate granting of exonerations from tariffs and from federal as well as local taxes. Domestic production was stimulated by providing accelerated depreciation to purchasers of locally produced goods and by special facilities to finance these sales. Total consumption of these items in real terms rose by a factor of more than 4 between 1965 and 1975, with imports rising by a factor of about 10 and domestic production by a factor of more than 3. vii. The increases in international oil prices in December 1973 and the world recession which followed in led to significant changes in economic performance and in policies of the Brazilian Government. With the emerging balance of payments crisis, measures were taken during the period through 1978 that directly and indirectly affected the capital goods industries. These included cutbacks in government investment, raising import tariffs and imposing a prior deposit system for many imports. In addition, the policy for liberal granting of tariff and fiscal exonerations on capital goods imports was reversed, and efforts were made to increase the share of domestically produced machinery in those projects which received incentives. At the same time, in the approval of projects, highest priority was given to the expansion of domestic capital goods producing industries, and a system of direct subsidies to promote local production of specific equipment was established. viii. As a consequence of the measures taken and the impact of external events, the expansion of economic activity, particularly gross fixed investment, slowed down. However, local production of capital goods rose to a peak in 1976, reflecting the completion of orders and implementation of investment decisions which had been made much earlier. It was not until 1977 and 1978 that output of these industries began to decline. Capacity utilization fell sharply, and the situation was aggravated by the coming on stream of new capacity resulting from the investment undertaken in the earlier periods of expansion. II. Main Policy Instruments Exchange Rate Management ix. There is evidence that the exchange rate prevailing in December 1978 was overvalued, with estimates of the overvaluation in the 25%-35% range. Until the end of 1978, the expressed official policy was to devalue the currency regularly in accordance with relative internal and external inflation, implying no change in the real value of the cruzeiro. Analysis of exchange rate movements, adjusted for domestic and international inflation, indicates only a slight real devaluation in the period from end-1973 through 1978, suggesting that exchange rate changes were not significant in that period in altering the external relative profitability and competitiveness of the Brazilian capital goods sector.

9 - iii - Import Restrictions x. While nominal tariffs on capital goods imports ranged from 7% to 205%, most items were subject to rates in the 35-55% range. Tariff rates were highest for electronic and electrical equipment and machinery, lowest for shipbuilding and agricultural equipment and tractors. In 1974, the importweighted average for capital goods was 40.8%; following slight increases in tariff levels (mainly in 1975), the average rose to 46.6% in Another reflection of growing protection was the observed changes in realized tariffs. As discussed below, tariff exemptions and reductions were important features of Brazilian commercial and industrial policy and these were reflected in a comparison of tariff revenues actually collected with the value of imports. For capital goods, realized tariff rates averaged about 10% at the end of the 1960s, fell to a low of 6.9% in 1974, and rose to 14.1% by xi. In addition to tariffs, a federal industrial product tax (IPI) and state value added tax (ICM) were imposed on domestic production and imports. But these could also be exonerated as part of incentive packages. The IPI, averaging around 10%, varied over products and its structure had no relation to tariffs, indicating that the criteria determining these rates were unrelated to commercial or industrial policy. The ICM was a uniform 14%. Finally, all imports had to pay a 6% port tax. xii. The most important changes in import restrictions since 1974 took the form of expanding non-tariff barriers. In the first place, government agencies were prohibited during that period from importing capital goods unless there were no satisfactory domestic equivalents. In addition, an import deposit system was established in 1975 requiring a prior 100% deposit for a large number of imports (including capital goods, unless exempted under incentive packages); these deposits were frozen for a one-year period with neither interest payments or monetary correction. With inflation at some 40% annually in 1977 and 1978 and nominal interest rates in the commercial banking system averaging around 50% during that time, the import deposit system was equivalent to a 50% tariff on an FOB basis, or 33% on a CIF basis. xiii. Investment incentives were provided by various agencies and sectoral or regional programs, but the principal agency concerned was the Conselho do Desenvolvimento Industrial (CDI). Established in 1964 to promote industrial growth, the main function of CDI was to review investment proposals and grant tariff and tax exonerations for imported capital equipment; approved projects could also receive other benefits, described in subsequent paragraphs. Through 1973, CDI granted these incentives virtually indiscriminately for projects of all sizes. In 1974, for the first time, it established criteria for approvals and a minimum size of project; in 1975 and 1976, it designated priority subsectors with highest priority going to the capital goods producing industries. In 1977, these enterprises accounted for 18.3% of the total value of approved investment as compared to 8.5% in In addition, local capital goods industries benefited from efforts made by CDI to increase the proportion of locally supplied equipment in all approved investment proposals; this percentage rose from 44.3% in 1974 to 74.5% in 1977.

10 - iv - xiv. The import content of approved projects also had to be reviewed by the Foreign Trade Department of the Banco do Brasil (CACEX), which was responsible for issuing the licenses which extended tariff exonerations to imported capital equipment. Under the Law of National Similars as it was administered, incentive benefits could be granted to imported equir-ent only if there were no comparable locally produced counterparts. Ho>±ver, even in cases where locally produced equipment might be available, imports of these items could be authorized if delivery of the local products involved long delays or if the latters' prices were considered excessive in relation to prices of imports. In making these comparisons, current tariff rates were applied to calculate landed CIF costs of imports. xv. In this procedure, CACEX, in consultation with domestic capital goods manufacturers and the authorities of the approved project, developed a Participation Accord which spelled out in detail the equipment to be supplied by local producers and that which would be imported. Similar to the results achieved by CDI, participation agreements authorized by CACEX in 1977 provided that 76.6% of equipment requirements of approved projects would be met by local manufacturers. xvi. Another important program providing exonerations of tariffs and domestic taxes was the Special Program of Fiscal Incentives to Exporters (BEFIEX) established in Enterprises which committed themselves to export a certain percentage of their output received these benefits for imported capital equipment. Through mid-1978, projects totalling US$3.6 billion in fixed investment costs were approved, of which US$3.0 billion was for the automobile industry. With CDI assigning low priority to industries producing consumer goods, a number of those enterprises made use of BEFIEX provisions. Fiscal Incentives for Domestic Production xvii. In March 1975, regulations were established by the Ministry of Finance for a system of fiscal incentives to domestic producers of capital goods. Purchasers of these items could be exempted from payment of IPI as well as ICM taxes. In addition, IPI fiscal credits, which accumulated when firms purchased inputs on which IPI had been collected, could serve as the basis for a direct fiscal subsidy; this incentive system was similar to that which had been established for exports. The subsidy element was estimated for capital goods industries as having been equivalent to 16-27% of ex factory prices, averaging around 20%. Unlike CDI incentives which were given to industries to be promoted, these incentives were given to specific pieces of equipment, the production of which was to be promoted. Granting of these incentives was discretionary and the criteria for application generally related to the degree of sophistication of the particular item and the period of time domestic production had been undertaken. While the exact magnitude of these subsidies is not known, they appear to have been concentrated in certain categories of equipment which represented a small share of total capital goods output.

11 v Protection to Inputs xviii. To determine the extent of protection to inputs which influenced net protection for the final products, use has been made of estimates of realized tariffs for those products. However, since one of the major groups of intermediates used in capital goods production (special steels) was subject to import quotas, this measure understates protection to inputs and consequently will ovrerstate the net protection that was provided for the final products. Financial Incentives to Local Production xix. Several subsectors of the capital goods industry benefited from substantial subsidies provided by relatively cheap credit from public agencies; the principal source was the Special Fund for Industrial Financing (FINAME) a subsidiary of BNDE, which financed local sales of domestically produced equipment. For the purchasers of that equipment, who normally would seek long-term financing for such transactions (or comparable imports), by using FINAME credits their costs on average were 8.8% lower than if the funds had been borrowed at prevailing (December 1978) market rates. This has been estimated by calculating the current value of the loan repayments (at an average 45% interest rate) when discounted at the market interest rate (50% p.a.). The subsectors that benefited most from this source of finance were essentially producers of heavy industrial machinery, electric generating and transmission equipment and railway equipment. FINAME's disbursements and commitments in 1976 and 1977 indicate that its programs covered some 10-15% of the total value of locally produced capital goods sold in the country, although in the heavy equipment branches the share was substantial. The agency also played a role in implementing the government's policy to increase the local content of capital goods output by gradually raising the minimum level of such content for eligible equipment from 50% prior to 1974 to 85% in III. Estimating Net Protection Basic Estimates xx. Measurement of the net protection that prevailed at the end of 1978 must take into account the import restrictions (mainly tariffs and prior deposits), fiscal incentives, financial subsidies, protection of inputs and exchange rate overvaluation that were in effect at that time. The approach has been to compare the possible prices of imports (landed costs) and those of domestically produced items (at the factory) after the effects of economic policies have been taken into account. There were a number of industrial policy schemes which provided tariff exonerations and/or domestic subsidies for specific pieces of equipment. Since there was a great deal of administrative discretion in the application of the measures, it has only been possible to estimate a range of net protection. Thus the present analysis provides estimates of the situation prevailing at the end of 1978.

12 - vi - xxi. Where the import of a capital good was subject to full import levies (including prior deposits) while the domestic production of that same item was free from fiscal or financial subsidies, net protection varied from 15% (shipbuilding) to 86% (motors and vehicles), with most categories between 20% and 50% These estimates of net protection apply to a considerable proportion _i the capital goods that were produced or imported, as most investment projects were ineligible for exonerations from import levies. Since the import deposit was a major component of import levies, its removal without compensating devaluation (or subsidy to domestic production) tended to leave most of the industry without protection. xxii. If levies on imported capital goods were not exonerated and the locally produced equivalents obtained production and financial subsidies, then the levels of protection for domestic production for all of the main equipment categories exceeded 200%. An opposite case would involve exonerations (partial or full) for tariffs and waiver of the prior deposit for capital goods imports, while the domestically produced equivalents would not receive any financial or production subsidies. Under these conditions, the domestic producers were disprotected (due to the overvaluition of the exchange rate and protection to inputs). However, both of these cases were extremes and unlikely to occur in practice. Normally, many types of equipment which would have had access to exonerations of import charges (because they were part of a priority project) would also have had access to fiscal and/or financial subsidies if produced domestically. xxiii. In the likely case of tariff reductions on imports combined with fiscal and financial subsidies, net protection varied from 57% to 96%; for the important category of Industrial Equipment and Machinery the estimate is 79%. In these instances, the production subsidies (including both fiscal and financial elements) were the most important forms of protection leading to relatively high levels. While there were strong arguments for some degree of protection for these industries, it is doubtful that such high levels as actually provided by the established system were needed. Also troublesome was the discretionary nature of the system which created considerable uncertainty among producers concerning eligibility of particular products. The overall operation of the system under these conditions was bound to be random and uneven. xxiv. Although the estimates of net protection indicate that it stood at relatively high levels, important qualifications suggest that a substantial portion of the subsector did not benefit from such arrangements. in particular, the very large fiscal and financial subsidies appear to have been granted to only a limited number of industries, primarily among the heavy equipment categories representing a relatively small share of total output of the subsector. In addition, protection to inputs has been underestimated, with a consequent overestimate of protection for the final products. Indications of Competitiveness xxv. As indicated in Chapter I, exports of capital goods grew significantly over the period. In 1977, for example, they accounted for about 10% of value of output of these items. While export incentives were important in

13 - vii - these sales, their magnitude offset the overvalued cruzeiro. Moreover, there were no changes in the benefits provided during the period which can explain this growth. The expansion in domestic output was also accompanied by a relative "cheapening" of capital goods vis-a-vis other manufactured products, as price indices for the former category did not rise as sharply as for the latter. These trends suggest that, as it grew within the large market offered by Brazil, and as necessary technical skills matured, the subsector managed to develop a reasonable degree of competitiveness in international markets. xxvi. Some direct evidence supporting this view is found in an analysis prepared in early 1978 comparing export price quotations (including incentives) for various Brazilian capital goods (machine tools, agricultural equipment and road building machinery) with price quotations for comparable products from major foreign sources. In the majority of cases, Brazilian quoted prices were more or less equal to or below the prices of competitors. Similarly, a study prepared in 1978 by a consultant to the staff of the Brandt Commission, covering seven relatively standardized pieces of equipment, found Brazilian export prices generally competitive with those from producers in developed as well as developing countries. However, export price quotations, particularly under prevailing international market conditions, may not have reflected actual cost-price relationships and competitiveness, and there is need for more direct analysis of these elements. IV. Prospects and Principal Policy Recommendations xxvii. At the beginning of 1979, the Brazilian capital goods industries were faced with a situation of considerable excess capacity. Given the usual lags in investment decisions and implementation, short-term prospects remained unfavorable. However, medium-term prospects appeared to be considerably brighter and there was need to focus on a number of critical issues to ensure further growth of the subsector. xxviii. Given the relatively efficient levels of output which had been achieved by a number of branches and the success already achieved in exports of certain standardized items (e.g. machine tools, diesel motors and trucks), further exports from the subsector appeared possible. For more complex pieces of equipment, which were generally sold as part of a total plant package, availability of export credit played an important role in these transactions. There was need, therefore, to expand these financial facilities. xxix. Efforts also were required to stimulate specialization and avoid dispersion of production both at the product and plant levels. To meet the government's stated objective of avoiding, on the one hand, monopoly power and on the other, an uneconomic number of enterprises in one product line, a policy framework encouraging a more open economy was considered desirable. xxx. During the years , the measures which the government undertook for the capital goods industries introduced a substantial degree of protection in contrast to the more open economic environment for the subsector which had been provided during its prior periods of impressive growth.

14 - viii - Latest indications were that the efficiency of only a small number of those industries was adversely affected, essentially the newer lines of production among the more advanced heavy equipment producers. For already existing lines, the system adopted probably raised the levels of profits. While it is unlikely that in the few years of high protection major inefficiencies could have been introduced, continuation of these policies could.,ve resulted in substantial distortions. xxxi. The principal policy conclusion emerging from the analysis of the prevailing system of import protection was that there was need for a major revision of that system. The focus would be on exchange rate overvaluation, tariffs on capital goods, prior import deposits, export subsidies and fiscal subsidies for production. xxxii. In January 1979, the Brazilian authorities announced a number of major changes in basic macro-economic policies. These included: (a) acceleration of the mini-devaluations so as to achieve 25% real devaluation by mid-1983; (b) elimination of prior deposits by 1983; and (c) the gradual elimination of the subsidy element in export incentives by mid-1983.

15 BRAZIL PROTECTION AND COMPETITIVENESS OF THE CAPITAL GOODS PRODUCING INDUSTRIES I. PERFORYiANCE OF THE SUBSECTOR AND POLICY DEVELOPMENT A. Growth of the Subsector through The industries producing capital goods 1/ have made substantial contributions to the growth of the Brazilian manufacturing sector since the earliest stages of that sector's modern development, which began at the end of the 19th Century. These activities gained particular importance after the late 1940s with their share in total industrial value added rising from about 5% in 1949 to 15% in Moreover, domestic production accounted for about two-thirds of consumption of these items in 1949 and rose to about four-fifths in There are a number of elements on both the supply and demand sides that contributed to this performance, which is rather unique among the developing countries, even the more industrialized. 2/ It should also be' noted that, for much of this century, the growth of the subsector was not stimulated by protection or by other explicit government policies but rather flourished in the face of competition from imports which during most of the period were given low tariff treatment and, at times, preferential exchange rate treatment On the demand side, Brazil, with its considerable land mass, has always been a relatively large market. At the end of the 19th Century and up through the First World War, the country exported primarily a few highly profitable basic products (coffee, sugar, cotton and rubber). Thus an extensive agricultural economy exporting substantial volumes of commodities had large transport requirements; this in turn generated demand for efficient repair installations (not only for railroads but for ships) to ensure proper 1/ The definition of capital goods used in this report is essentially producers' goods. Statistically it includes sections of the Brazilian industrial classification (similar to the International Standard Industrial Classification - ISIC) covering metal products, mechanical and electrical equipment and transportation equipment. It does not include consumer durables such as electrical appliances and automobiles. Details on definitions of categories are provided in the notes to the tables appearing in the Statistical Appendix. 2/ For example, one author notes that Brazil's ratio of local production to consumption of capital goods in was more than three times larger than that of Argentina at that time. See Nathaniel H. Leff, The Brazilian Capital Goods Industry, , (Harvard University Press 1968) p. 8. Much of the above discussion on the origins of the Brazilian capital goods industries is taken from this study.

16 - 2 - flow of the products to the ports and to the importing countries. In addition, the export products were subject to some degree of processing (e.g. coffee hulling or sugar refining) the machinery for which also required repair facilities to avoid interruptions. Given the long distances from the major sour-s of the transport and processing equipment, many of these repair shows expanded their facilities, installing foundries, and readily undertook the production, first, of spare parts and, later, of basic pieces of machinery. A number of existing Brazilian-owned major producers of capital goods had their origins in those shops. Finally, the military forces also provided a relatively important market for mechanical engineering products From the point of view of supply, in particular the factors required for production, capital was readily available from the immensely profitable exporting enterprises. Material inputs for these early simple operations also were easily accessible, the local iron ore providing the basis for ample supplies of pig iron needed for foundries Perhaps the most significant element from the point of view of the long-term development of the capital goods industries was the supply of human resources. Unlike many other developing countries, Brazil drew its technical and skilled personnel from both the indigenous population and the flow of immigrants. While gaining its wealth from essentially agricultural activities, the "Paulista Industrial Elite" very early developed an industrial mentality with emphasis on technological growth and modernization; the Sao Paulo Technical School was established in the 1890s and served as a major source of technical personnel in the early stages of Brazil's industrial growth. As regards immigration, a large number of persons entering the country at the end of the 19th Century came from northern Italy, which had a long tradition of metal-working and entrepreneurship By 1939, acccording to census data, the production of capital goods represented 4.9% of industrial sector output and provided employment to more than 21,500 workers, accounting for 3.2% of total industrial employment. The world crisis during the 1930s had brought about a cut in imports, and the domestic capital goods subsector grew at an annual average rate of 12.4% during the decade, as compared with 5.8% for the industrial sector. Transportation-related goods, accounting for nearly 60% of subsectoral output, represented the most important grouping, which was highly diversified; in addition to railways and ship repairs, it included the assembly of buses and trucks. Mechanical equipment accounted for 22% of the output of these industries with the most important products being machine tools and agricultural machinery. The balance of the subsector's output (18%) was made up of electrical equipment, mainly cables, electrical instruments, motors and transformers During the 1940s, the growth in output was constrained by the warinduced shortages of certain materials, particularly special steels required for the more complex equipment the subsector was then capable of producing. 1/ As a consequence, capital goods production expanded at about 1/ See Leff, op. cit., p. 13.

17 - 3- the same rate as the industrial sector (averaging 7-8%/year). In 1949, the share of the subsector in industrial output was 5.1%. In part due to the war, the trend toward more diversification continued; transportation's share was reduced to 44%, as both mechanical and electrical equipment increased to 31% and 24% respectively The decade of the 1950s represented a period of substantial deepening of the structure of the Brazilian capital goods industries. Output of the subsector rose by more than 11% p.a., well above the rate of growth of the industrial sector, and by 1959 its share in manufacturing production reached 9.7%. During those years, the expansion of the economy provided a growing market for domestically produced supplies Various steps were taken by the government to assist the growth of the subsector. Through the mechanism of the Executive Groups, 1/ official encouragement was given to private foreign investment and, as a result of direct negotiations, a number of major foreign equipment-producing firms established plants in the country during the decade. In addition, an accelerated program to obtain full import substitution of commercial vehicles (as well as of passenger cars) was implemented. While in the longer run these activities were designed to reduce imports of capital goods, the short-run effect was to increase demand for imported equipment. In addition, the government embarked on a substantial investment program, particularly for basic industries such as steel and petroleum refining, which had large requirements for imported capital goods Following these measures, a number of important new products began to be produced. Many, particularly from the foreign-owned enterprises, involved important advances in complexity and technological know-how and included items such as turbines, large boilers, refining and steel-mill equipment. However, transportation equipment continued to be the most important grouping, accounting for some 50% of the subsector's output in 1959, which reflected the efforts made to achieve self-sufficiency in commercial vehicles During the 1950s, import policy moved from direct quantitative controls (through 1953), to multiple exchange rates and ad valorem taxes ( ) and to a dual exchange rate combined with tariffs (after 1957). Under all these systems, capital goods imports were given preferential treatment. 2/ Among the specific measures used were providing priority in exchange allocation under a highly overvalued exchange rate, preferential rates in the multiple exchange rate systems and exemption from taxes and import duties. In addition, the level of imports of equipment was apparently sustained by the availability of foreign financing, particularly suppliers' 1/ See discussion of policy instruments later in this chapter. 2/ See Joel Bergsman, Brazil: Industrialization and Trade Policies, (Oxford University Press, 1970), Chapter 3; also Leff, op. cit., pp

18 - 4 - credits. It is estimated that almost 80% of equipment imports in the period were financed by foreign sources. 1/ Nevertheless, the share of imports in the total supply of capital goods in the Brazilian market declined during this period from about 40% in to less than 30% by the end of the decade By the beginning of the 1960s evidence began to emerge of overinvestment in the subsector and a substantial under-utilization of capacity. The situation was exacerbated by the slowdown in industrial growth beginning in Partial data available for this period suggest that output of these industries continued to rise until 1962, though at a slower rate than in the earlier decade, and fell in / At the same time, however, there was a sharp fall in equipment imports, reflecting in particular the reduction in suppliers' credits and probably in direct investment. Thus by the mid- 1960s the share of imports in domestic supplies of canital goods reached a level of around 10% The existence of excess capacity at that time was largely explained by over-optimistic expectations of the continued accelerated growth of the economy, industry in particular, with a substantial contribution from public sector investment in productive activities. Field studies indicated that excess capacity was especially acute among foreign firms while domesticallyowned firms continued to experience good demand conditions. 3/ B. Developments in the period 1.14 Output of capital goods increased more than four-fold between 1965 and 1976, at an average annual rate of about 14% (see Statistical Appendix Table 1). Expansion of the subsector was particularly rapid after 1970; average output growth was moderate in the last half of the 1960s, although fluctuating sharply from year to year, reflecting the continued impact of the prior recession period. With total industrial output rising at about 11% p.a. during this 11-year period, the share of capital goods in manufacturing production rose to about 16% in / See Leff, op. cit. p This would also include imported equipment provided as equity contributions by foreign investors. He further notes (p. 179) that imported equipment was often in product lines where domestic firms had excess capacity. 2/ IBRD, Current Economic Position and Prospects of Brazil, Volume V (Annex A). (Report No. WH-176a, July 1965). 3/ See Leff, op. cit. p. 32 and p. 37. It will be recalled that the foreign firms had been encouraged to invest in producing lines of more advanced technology.

19 1.15 Among the major groupings within the subsector, industries producing mechanical equipment were the most dynamic, with an average annual rate of growth of 16% during the period. The average annual rate of growth for production of transportation equipment was 14% while that for electrical equipment was only 7%. Data from trade associations provide some indications of the performance of individual product lines; within the mechanical grouping, production of machine tools, heavy industrial equipment and agricultural equipment appear to have increased the most. In the transportation equipment group, production of commercial vehicles and ships was up sharply The expansion of the subsector was linked closely to the very substantial increases in gross domestic investment which took place since the end of the 1960s. Following the change in government in 1964, a series of reforms of economic policies were introduced, designed primarily to open up the economy, including in 1968 the initiation of the mini-devaluation system and direct measures for export promotion. With buoyant world demand, total exports, which had been stagnant for almost a decade, began to rise sharply after /; this in turn permitted liberal demand management. In addition, the general political and economic climate favored an inflow of external capital To sustain the high rate of capital formation and, at the same time, to permit considerable increases in consumption, substantial incentives were given to investment in the period through Among the incentives granted were tariff-free imports of capital goods and exemption from federal and state excise taxes. Domestic production of these products was stimulated by granting higher rates of depreciation to investors using locally produced equipment and also through expanding activities of the Special Agency for Industrial Financing (FINAME), a subsidiary of the National Bank for Economic Development (BNDE), which had been established in 1965 to finance the purchase of Brazilianproduced capital goods The main elements of the investment boom which lasted through the mid-1970s, particularly the role of public sector investment, have been analyzed in previous Bank reports and need not be repeated here. From 16-18% of GNP in the 1960s, gross fixed investment rose throughout the period and by 1975 accounted for 25% of GDP. The sharp increase in demand for investment goods was met through both the expansion of domestic output and a rise in imports of these items which had reached relatively low levels in the mid-1960s With excess capacity characterizing the local capital-goods producing industries in the mid-1960s, the first result of the rise 1/ In current values, exports in 1967 were actually below the level reached in 1951 (at the peak of the Korean boom).

20 - 6 - in demand was a considerable improvement in capacity utilization. 1/ By the early 1970s, as expectation of sustained growth became widespread, many of these enterprises raised their own investment plans which in turn contributed to a further expansion of the investment ratio. 2/ 1.20 l the import side, the changes were particularly dramatic. From a level of around 10-11% of total consumption in the mid-1960s, imports rose sharply, increasing in real terms by a factor of more than 10 between 1965 and 1975 (falling sharply in 1976, as analyzed in the next section). By 1975, they represented about 28% of the domestic market, the highest share since the mid-1950s (Statistical Appendix Table 3) Data on the proportion of these imports financed through foreign sources, comparable to those cited earlier for the 1950s, are not available. But the overall data for foreign indebtedness and direct investment flows suggest that once more there was a link between the availability of foreign finance and the imports of capital goods. 3/ 1.22 There is further evidence that the decline in the share of domestic industries producing capital goods in total supply of these items was not a function of their inability to compete with imports. Of particular importance is the emergence in the early 1970s of relatively substantial quantities of exports of machinery and equipment. From 3-4% of output of the subsector at the end of the 1960s, external sales are estimated to have risen to almost 10% by the end of Moreover, while exports to LAFTA (a large proportion of which were in products under preferential tariff treatment) accounted for more than half of this trade in , by 1976 that share had been reduced to about 42%. C. Situation in The increases in international oil prices in December 1973, particularly important for Brazil which imports approximately 80% of its oil requirements, and the world recession which followed in 1974 have led to significant changes in economic performance and in policies followed by the Brazilian authorities, particularly those affecting the capital goods 1/ The trade association for the mechanical goods industries reported 80% capacity utilization in 1968, which rose gradually to almost 90% by These data are based on reports of the member firms reflecting their own practices and technical judgments on capacity. Most firms in this subsector, not characterized by process production techniques, prefer to work at a maximum of two shifts. 2/ An association of some of the largest equipment producers reported that its member firms had increased their capacity by 250% between 1972 and early / See William G. Tyler, Manufactured Export Expansion and Industrialization in Brazil (Tubingen, 1976), p. 50.

21 sector. In 1974 and early 1975 the dimensions of the balance of payments crunch began to emerge and steps were taken to curtail imports both indirectly by restraining overall demand, and directly through such measures as increasing tariffs and imposing a prior deposit system. In addition, government investment programs were cvt back. With imports of capital goods representing in 1973 more than one-third of total imports, specific measures were taken to cut down on those expenditures. In the first place, the previous policy for virtual indiscriminate granting of tariff and fiscal exonerations on capital goods imports was reversed and priorities among subsectors were established to govern the granting of these incentives. Secondly, highest priority within that grouping was given to the expansion of the domestic capital goods producing industries The impact of these changes on the capital goods subsector took some time to develop. While the rates of increase in overall gross investment in 1975 and 1976 were well below those of the previous years, and turned negative in 1977, domestic output of producers' goods continued to expand at relatively high rates through 1976, finally declining in Imports of these items in real terms reached a peak in 1975 but fell off sharply in 1976 and further in In the latter year, as a result of the declines in both imports and domestic production, apparent consumption of these items fell by about 10% in real terms. Some support to output was found in a further expansion of exports in real terms in 1977 after a slight decline in Partial data for 1978 indicate a recovery in output of machinery and equipment to the 1976 peak levels, an increase in the value of imports of these items to a level just below that of 1976 (consequently, substantially lower in real terms) and a continued but moderate growth in exports. Thus, apparent consumption in 1978 appears to have risen slightly over 1977 but remained well below that of Based on mission interviews in July 1978, the rise in investment reflected mainly completion of projects which had been under construction. In general, given the uncertainty which normally accompanies a change in administration, the private sector was cautious in undertaking new projects or expansions while public sector investment continued to be dampened by budgetary constraints Toward the end of 1978 utilization of capacity in the capital goods industries continued to decline slightly from the peak levels of Moreover, there was a continuing decline in orders on hand which had also begun to emerge in In addition, several major expansion projects among the equipment industries which had begun to be implemented in but yet to come on stream by the end of 1978, would create further problems of excess capacity. D. Formulation of Policy for the Subsector 1.27 Although some of the industrial policy measures date back to the beginning of the 20th Century (e.g. a form of a Law of National Similars was enacted in 1911), it was not until the 1950s that a more explicit

22 - 8 - industrial policy began to emerge. Within the context of overall macroeconomic policy formulation emphasizing import substitution 1/, priority was given to the development of specific subsectors whose imports were considered critical rather than to elaborating a general sectoral strategy. The principal manufacturing subsectors chosen were the automotive industry, shipbuilding, tractors and earthmoving machinery, heavy equi_sent and railway equipment; other related industrial activities included petroleum refining and exploration and electric energy production and distribution For the selected manufacturing subsectors, "Executive Groups" were organized with responsibility for formulating development programs; membership included representatives of key government agencies and of the productive enterprises, mostly privately owned but including some with mixed government and private ownership. The government agencies represented in these Groups were those with implementation responsibilities to ensure that the decisions were carried out. Among these agencies were the Ministry of Finance (responsible for tax concessions), the Conselho de Politica Aduaneira (CPA) and the Foreign Trade Department of the Banco do Brasil (CACEX), jointly responsible for tariff concessions for imports, and BNDE (which provided subsidized long term finance). While there was a substantial degree of coordination within each Group, there was a notable lack of coordination among the groups as well as with the overall development efforts. The pursuit of independent subsectoral development plans contributed to the excess demand and inflation which characterized the period through The priority subsectors were mainly among the producers of capital goods, the exception being passenger autos which, along with trucks and buses, made up the automotive industry. With varying degrees of success (which also varied in relation to economic optimum solutions) these Groups managed to promote the establishment of new plants and to incorporate new technology, particularly through direct negotiations with foreign enterprises With the change in government in 1964, there was a gradual revision of development strategy towards a more open economy. Rather than establishing or maintaining priorities among industrial subsectors, a broader expansionist policy was adopted which would essentially leave market forces to determine growth areas. The main exceptions were the capital-intensive intermediate goods industries such as steel, petrochemicals and pulp, in which public investment was undertaken To consider the specific problems of the industrial sector and to formulate and implement policies for its development within the established macro policy framework, the government created the Conselho de Desenvolvimento Industrial (CDI) in The Minister of Commerce and 1/ See Bergsman, op. cit., Chapter 4, and Tyler, op. cit., pp

23 - 9 - Industry was made chairman of CDI, and membership consisted of the Ministries of Planning, Finance, Interior, Mining and Energy, and the Armed Forces as well as the presidents of the Central Bank, the Bank of Brazil, BNDE, and the Industrial Federation. CDI absorbed the Executive Groups, which lost their power to implement decisions; these groups became the basis for the subsequent internal organization of CDI along subsectoral grounds The main function assigned to CDI was the granting of incentives to investment, essentially exonerations of tariffs on imported capital goods and exemptions from the federal excise tax (IPI). These were provided virtually indiscriminately from 1966 through During this period, other special programs were also in operation providing additional investment incentives; for example, the program to develop the Northeast (SUDENE) granted important income tax benefits to firms located in that region. With the essential intention of expansion, however, problems of coordination among these programs to achieve sectoral objectives were largely ignored Following the difficulties which began to emerge in 1974 and the need to reassess both overall macro-economic and sectoral policies, the government created the Conselho de Desenvolvimento Economico (CDE). Presided over by the President of the Republic, it included all of the ministries in the economic area. Its principal function was to provide general policy guidelines which in turn would be implemented by the operating agencies in the various fields. A resolution of CDE published in March 1977 set out broad objectives in the industrial area relating to the role of the private sector (including the participation of foreign investors) and assigning priority to the development of basic intermediate goods (including steel, petrochemicals and fertilizers), minerals and the capital goods industries The main elements of policy which were defined for the capital goods industries were as follows: (a) public sector agencies were to be restricted in their imports of these items; (b) in the analysis of projects, the agencies providing incentives or financial support were to orient their decisions to avoid, on the one hand, the strengthening of monopoly power and, on the other, an excessive number of enterprises producing the same product line, with a view to promotion of specialization; (c) encouragement was to be given to the progressive nationalization of production of components; (d) in public sector investment projects, an attempt had first to be made to determine equipment which could be supplied by domestic industry before undertaking international procurement; (e) in using consortia (for procurement), preference was to be given to those under the leadership of local private enterprises.

24 II. IMPORT RESTRICTIONS AND TAX INCENTIVES A. Introduction 2.01 The observed growth of the Brazilian capital goods sector has been affected by economic policy. In Brazil, as in other countries, industrial policy affects the relative profitability of certain industries vis-a-vis other industries and that of manufacturing as a whole in relation to other sectors of the economy. Since government is involved in very few directly productive activities in the capital goods sector, profitability must serve as the economic lure to attract resources to the sector. While implemented by different agencies and levels of government for widely differing objectives, there have been basically five areas of policy affecting profitability and competitiveness of the Brazilian capital goods industry: exchange rate policy, import restrictions on capital goods, tax incentives to domestic producers and users of capital goods, financial incentives and subsidies, and protection of the domestic industries supplying inputs to the Brazilian capital goods industry In this chapter each of these policy areas will be discussed separately. 1/ Subsequently, in the third chapter, an attempt will be made to assess the overall quantitative impact of these policies based on the situation prevailing at the end of The overall question is the extent to which the Brazilian capital goods industry was protected and promoted as a consequence of economic policies. Accordingly, a system of measuring net protection has been developed and applied to an examination of the competitiveness of the Brazilian capital goods sector as of the end of B. Exchange Rate Policy 2.03 Exchange rate policy has been pursued in Brazil for a number of objectives not related to industrial development. There is evidence.-athat the end of 1978 the prevailing exchange rate was heavily overvalued and, as such, it discriminated against the export sector and ceteris paribus against import substituting sectors. Estimates vary as to the extent of the exchange rate overvaluation but generally they are in the 25-35% range. These estimates are necessarily rather crude, involving, as they must, assumptions about capital flows, changes in other policies, tariff and trade restrictions external to Brazil and price elasticities Changes in exchange rate policy over time can affect the degree of overvaluation. From 1967 through the end of 1978, through the system of mini-deduction, the expressed government policy was to devalue the currency regularly in accordance with domestic inflation and inflation in Brazil's major trading partners. The objective was to maintain the real purchasing 1/ The way in which these various policies are interrelated is developed in Annex I through the use of a partial equilibrium supply and demand model. 2/ Affonso Celso Pastore, Jose Roberto M. de Barros, and Decio Kadota, "Sobre a Taxa de Cambio: Resultados adicionais e uma replica a Analise Bacha." Pesquisa Planejamento Economico, Vol. 8, No. 2, August 1978, pp

25 power parity for the currency. An analysis of Brazil's exchange rate movements (Table 2.1) indicates that the real exchange rate, adjusted for domestic and international inflation, did not undergo substantial change b.atween the end of 1973 and mid There has been observed a slight decline in the purchasing parity exchange rate index for the period, indicating that the cruzeiro was gradually depreciated in terms of other currencies. This real depreciation was not dramatic and, in fact, may have been somewhat overstated by the use of later year trade weights. Moreover, because of the important exogenous increase in petroleum import prices, it can be argued that the rate of exchange rate depreciation should have been greater. This view is suggested by the fact that import restrictions after the mid-1970s increased. Nonetheless, the degree of exchange rate overvaluation is unlikely to have changed appreciably during the final five-year period. As such, changes in the exchange rate in the final years of the period were not significant in altering the relative profitability and competitiveness of the Brazilian capital goods sector. C. Import Restrictions 2.05 Import restrictions per se incorporate both tariff and non-tariff barriers. Since they are more visible and more readily quantifiable, much of the attention devoted to Brazilian trade restrictions has focussed on tariffs. Yet tariffs, while significant, were not the principal form of restriction on imports in Brazil. An import deposit scheme, some direct import restrictions affecting government enterprises, a national origin preference for government purchases, and an involved system of fiscal incentives all served to restrict the imports of capital goods, as well as of other products. Similarly, the effects domestically were to increase domestic prices and output levels. When considered together, the indirect controls on imports constituted an implicit quota system. The determination of what could be imported, however, as will be discussed below, was not unrelated to the tariff levels. While it would have been desirable to have made direct price comparisons to assess the quantitative impact of Brazil's import restrictions, such implicit tariff calculations, on a meaningful basis, have been precluded by the non-availability of suitable data Tariffs on capital goods ranged in 1978 from 7% to 205%. The extremes, however, were limited in importance. Tariffs on the most common types of equipment fall in the 35-55% range. A simple arithmetic means for nominal tariff rates on capital goods for all such products listed in the tariff schedule was 45.8% in / Weighting the average for the same year by imports produced a mean of 40.8%.. There were growing balance of payments pressures in and subsequently there were some modest upward adjustments in tariff levels. Most of these adjustments occurred in By 1977 the import-weighted tariff average was 46.6% for all capital goods, as seen in 1/ Marcos G. Fonseca, Akihiro Ikeda, and Jose Paulo Z. Chahad, Pesquisa Complementar sobre Substituicao de Importacao de Bens de Capital, FIPE, Part A, Fevereiro 1978, pp Statistical Appendix Table 8 presents information on nominal and realized tariff rates for capital goods in 1974 according to the FIPE classification system.

26 Table 2.2. The tariff increases appear more pronounced in 1972, the year before the 1973 oil crisis, is used as a reference point. Table 2.3 presents comparison between 1972 and 1976 tariff levels for the six major capital goods groups in the Brazilian trade classification (NBM). As seen, all categories witnessed increases in nominal tariffs except the low nominal tariff shipbuilding group A similar picture of increased tariff protection is presented by the observed changes in realized tariffs. As will be discussed below, tariff exemptions and reductions have been common features in Brazilian commercial and industrial policy. These reductions are reflected in the realized tariff rates, which are computed by dividing tariff revenues actually collected by imports. In all but one category there was a reduction in the exemption levels between 1972 and 1976 (Table 2.3). Realized tariffs increased markedly, and in the case of the important machinery and electrical machinery categories, the increase was proportionately more than that of nominal tariffs An examination of total imports of capital goods over a longer period of time also reveals that there was a tightening up on tariff reductions and exemptions in later years. The realized tariff rate for all capital goods increased from an annual average of 9.6% for to 14.1% in 1977 (Table 2.4). It is also interesting to note that the increase in realized tariffs for capital goods in 1976 and 1977 was preceded by a decline in 1973 and This decline was due to a liberalization in official industrial project approvals incorporating tariff exemptions and reductions, as exercised by the government in the euphoric high growth years of 1972 and The structure of nominal, as well as realized, tariffs on capital goods can be seen in Tables 2.2 and 2.5. Both tables present a consistent picture, although the level of disaggregation is greater in Table 2.5. The latter presents, among other things, the 1976 tariff data as compiled and grouped in the format of the IBGE input-output table. Information is provided for 16 capital goods industries. Of these industries Industrial Equipment and Machinery is by far the most important. In 1976 products in this category accounted for one-third of Brazil's total capital goods imports 1/. For this industry nominal tariffs averaged 40%. While a number of other capital goods industries are seen to exhibit similar tariff levels, tariff rates were markedly higher for the electrical equipment and machinery industries, averaging 76% and 81% respectively. The highest tariffs are present in the two industry categories where there are some durable consumer goods necessarily included. With both Electronic Equipment 1/ In 1970 this industry accounted for 19% of total Brazilian capital goods production, more than any of the other 15 industries. IBGE, Matriz de Relacoes Interindustriais: Brasil, 1970 (Rio de Janeiro: IBGE, 1976). Information published in the description of the table was used to classify the import and tariff information as reported in Table 2.5, classified according to the IBGE input-output table format.

27 Table 2.1: BRAZIL'S PURCHASING PARITY EXCHANGE RATE INDEX, Brazil's Wholesale Industrial Trade Weighted Purchasing Price Index Industrial Price Parity Exchange Inflated by the Index for Brazil; Rate Index Nominal Cr$/US$ Ten Major Trading i.e., A x 100 Month Exchange Rate Partners B January June December June December June December June December June December June December June December June December June Source: IBRD staff estimates.

28 Table 2.2; Nma4IA AND RuALiZED TARBI AND INDUSTRIAL PR=DUCT TAS RASSS ON BRAZILIAN IMPORTS, BY USE OF PRODUCT, Fercmtage Tariff Rates Industrtal Product Tax (IPI) 2 Imports (CIF) Shares Rates 2 (in US Smillion) of 1977 Imports Mo inle RealizedN (2) _ () (2) M2 (7) Nondurable Consumrr Goods Durable Conu.mer Goods Combustible Fuels and Lubricants 4, Intermed. Products for Agriculture OA Intermed.Products for Industry 4, Construction Materials Capital Goods for Agriculture Machines 6 Tools Other Equipmnat Transport Equipment Capital Goods for Industry 2, Office & Scientific Machinery Tools Machinery Parts & Accessories Industrial Machinery 1, Other Fixed Equipment Transportation Equipment Parts & Accessories Moving Transport Equipment Fixed Transport Equipment Miscellaneous TOTAL CAPITAL GOODS IPORTS 4 3, S 0 TOTAL IMPORTS 13, Notes: 1. Nominal tax rates are those which would be in force if all taxes were paid. 2. Realized tax rates were calculated as actual tax collections divided by CIF isports. 3. Vehicles for personal use are included under durable consumer goods. 4 Capital goods imports include capital goods for agriculture, capital goods for industry and transportation equipment. Source: Compiled from unpublished data made available by CIEF.

29 Table 2.3: NOMINAL AND REALIZED TARIFF RATES ON BRAZILIAN IMPORTS, BY IMPORT CLASSIFICATION, 1972 AND 1976 Nominal 1/ Realized 2/ Product Tariff Rate Tariff Rate NBM Code Group (%). 84 Machinery Electrical Machinery Railway Equipment Autos, Trucks & Tractors Aircraft Shipbuilding Notes: 1/ Nominal tariff rates are those which would be in force if all tariffs were paid. 2/ Realized tariff rates were calculated as actual tariff collections divided by CIF imports. Source: Ministerio da Fazenda, Comercio Exterior do Brasil: Importacao, 1976 and unpublished data made available by CIEF.

30 Table 2.4: REALIZED TARIFF RATES FOR TOTAL IMPORTS AND CAPITAL GOODS IMPORTS, SELECTED YEARS, Capital Goods Year All Imports Imports (% ) Sources: Data for were taken from Flavio Pinheiro de Castelo Branco, Importacoes de Bens de Capital e Substituicao de Importacoes, Masters Thesis, University of Brasilia, 1976, p. 61. Data for 1976 and 1977 are from CIEF.

31 Table 2.5: TARIFF AND FISCAL SUBSIDY RATES FOR CAPITAL GOODS (1) (2) (3) (4) Fiscal Subsidy IBGE Sector Nominal Realized Nominal via IPI and ICM Classification Tariff Rate 1/ Tariff Rate 2/ IPI Tax Rate 1/ Tax Credits 3/ Code Industry tipi sipi 121 Pumps and Motors Machine Parts Industrial Equipment and Machinery Agricultural Equipment and Machinery Office and Domestic Equipment and Machinery Tractors Equipment for Electric Energy Electric Conductors Electrical Equipment Electrical Machinery Electronic Equipment Communications Equipment Trucks and Buses Motors and Vehicles Parts Shipbuilding Railway & Aircraft Equipment / Nominal tax rates are those which would be in force if all taxes were paid. Their rates are weighted by imports. 2/ The realized tariffs were calculated as actual tariff collections divided by CIF imports. 3/ The fiscal subsidies via the IPI and ICM tax credits are estimates which were made with 1975 data. Sources: Estimates for Columns 1-3 were made from CIEF data as published in Ministerio da Fazenda, Comercio Exterior do Brasil: Importacao, Column 4 is from estimates undertaken in Affonso Celso Pastore, Jose Augusto Arantes Savasini and Joao de Azambuja Rosa, "Quantificacao dos Incentivos as Exportacoes," Fundacao Centro de Estudos lo Comercio Exterior, Rio de Janeiro, 1977.

32 and Motors and Vehicle Parts it is impossible to separate out final use consumer items, for which tariffs tend to be higher. Since, unlike , the period witnessed little in the way of tariff changes, the 1976 nominal tariffs can be considered as representative under 1978 conditions In addition to tariffs, border taxes have also been charged on imports. The industrial product tax (the IPI), formally a value of product tax, was levied on the CIF value of imports plus the nominal tariff. While it varied over products, for all capital goods the IPI averaged around 10%. Tables 2.2 and 2.5 present a disaggregation of both nominal and realized IPI rates. Relatively low nominal rates were levied on tractors (5%), Agricultural Equipment and Machinery (6%), and Industrial Equipment and Machinery (8%), while the highest rates prevailed for Electronic Equipment (23%) and Office and Domestic Equipment and Machinery (19%). While the IPI rates were important in affecting net protection, the size of the IPI rates was determined by criteria evidently unrelated to commercial or industrialization policies. In addition to the IPI, imports also had to pay the state value added tax (ICM). At the close of the period, 1978, it stood at a uniform 14% over all products for most states and was levied on top of the total value of the imported item at the CIF import price plus both the nominal tariff and IPI. Port taxes were also assessed, amounting to about 6% of the FOB value of imports. Unlike the tariff, the IPI and the ICM, port taxes were never exempted. All of the border taxes--the IPI, the ICM and port taxes--while affecting relative prices, were levied for objectives outside of commercial policy. Unlike the tariffs, these taxes underwent no appreciable changes in the period While nominal tariffs were increased over the period, most of the increase in import restrictions took the form of expanding non-tariff barriers. In 1975 an import deposit system was established. This system involved a prior 100% deposit of the FOB value of the imports for a one-year period with neither interest payments nor monetary correction. 1/ Unless exempted, the import deposit was applied to all capital goods imports plus a wide range of other imports. With inflation running at some 40% annually in , the import deposit system implied an import surcharge of an amount more than 40%, depending on the rate of interest foregone. Nominal effective interest rates averaged around 50% annually in through the commercial banking system. Accepting this figure as the opportunity cost of money, the compulsory import deposit amounted to the equivalent of a 50% tariff levied on the FOB import price and about 33% on the CIF value. The compulsory advance deposit for imports formally applied equally to most manufactured products. In practice, however, capital goods imports were more likely to be made exempt from the deposit requirement than other imports which were covered by this system. 1/ In January 1979, the government announced the gradual elimination of the system; the prior deposit was eliminated altogether in December 1979 (see Chapter IV).

33 Beyond the indirect import restrictions, operating in one way or another through the market, direct controls over imporrc twere also exercised. These controls were most evident in the public sector. Gcvernment purchases constituted a powerful force for the expansion of the Brazilian capital goods sector. Consequently, changes in government purchasing policy greatly affect the sector. The public sector, including government enterprises, has accounted for over 50% of total capital formation in Brazil in recent years. 1/ In 1975 it was responsible for a large share of total imported capital goods, as a result of previously authorized and ongoing investment programs Attempts after 1974 to stem the tide of public enterprise imports, for both balance of payments objectives and reasons of promoting the domestic capital goods sector, resulted in a series of measures. The public sector, along with government enterprises, was prohibited from directly importing any consumer goods. Moreover, it was prohibited from importing or purchasing machinery, equipment and vehicles produced abroad unless there were no satisfactory domestic equivalents available. In other words, the Law of National Similars 2/ was applied to all government purchases. In international bidding a 15% preference is given to domestic suppliers. Finally, a strict import budgeting on the part of public agencies and government enterprises became standard practice. Each government organization had to submit, for approval by the President through the Planning Secretariat, an annual investment budget including a detailed justification for imports. The scrutiny of these budgets, in great part motivated by balance of payments considerations, became increasingly meticulous and intense. The import budgeting requirement and the approval for public imports amounted to the imposition of a tacit import quota system. As a result of these efforts to budget and restrict government imports, there was a decline in such imports, particularly of capital goods. A Planning Secretariat study of 69 major public enterprises shows that imports of these enterprises fell from about US$3 billion in 1975 to US$1.8 billion in The most complex part of the system of protection afforded to the Brazilian capital goods industry involved a system of incentives and benefits for industrial projects which were approved by government agencies, most notably the CDI. While these incentives varied from program to program, they included: (i) the reduction of import duties; (ii) the reduction of the border taxes due on imports, (iii) the waiver of the prior deposit on imports, (iv) access to official, subsidized credits, (v) exemption of IPI and ICM taxes for the purchase of domestically produced capital goods, (vi) the 1/ For a discussion of the magnitude of the government's impact on the economy see Werner Baer, Isaac Kerstenetsky, and Annibal V. Villela, "The Changing Role of the State in the Brazilian Economy," World Development, Vol. 1, No. 11 (November 1973). 2/ As presently administered, the Law of National Similars only applies to private sector imports which seek to obtain tariff exonerations or other benefits of incentive legislation; see para

34 granting of a fiscal subsidy, based on a credit under the administration of the IPI tax, for purchasing domestically produced capital goods, and (vii) the allowance of accelerated depreciation for income tax purposes for the case of domestic capital equipment. Since these benefits were substantial, most industrial investment projects were submitted for approval to the CDI or other agencies empowered to grant incentives. In the period, the examination of investment projects became much more thorough and there was growing reluctance to grant widespread tariff reductions on imported capital goods increased. Concerted efforts were made by government officials to reduce the import content of investment projects in the approval process As indicated in the previous chapter, in the period after its creation in 1964, CDI followed a policy of virtually indiscriminate granting of incentives. In 1974, the Council for the first time elaborated criteria for providing tariff and tax exonerations and established a minimum project size. In 1975 and 1976 it further designated its priority subsectors. As of 1978, priority industries were: (i) capital goods, (ii) basic metals and intermediate metal products, (iii) chemicals, petrochemicals and pharmaceutical products, (iv) non-metallic intermediate products, (v) automotive products, and (vi) consumer goods. Within these subsectors certain industries received more emphasis than others The changes that occurred in the final years of the period in CDI approvals are shown in Table 2.6. First, after having reached a peak in number of approvals (2,851) in 1973, there was a sharp decline in such decisions to 199 in / However, the average size of the projects which were granted incentives rose substantially, from the equivalent of US$1.3 million in 1973 to US$16.1 million in 1978 (both in current prices, relating to estimated fixed investment costs). 2/ Of particular significance is the wide range in size of "projects" which received incentives to import capital goods (see Statistical Appendix Table 11). In 1974, 3/ about three-fourths of the approvals involved total fixed investment below US$500,000 with half actually below US$100,000. Less than 10% of the projects were above US$3.0 million. Thus, there had been little attempt, at least until the end of 1974, to restrict access to relatively cheap imports of capital goods for a substantial portion of the industrial sector. 1/ In the first three years of its existence, from 1964 through 1967, CDI had approved a total of 237 projects. 2/ The high value of approvals and average project size in 1976 was due to a few very large steel projects (about US$6.0 billion of the total of US$8.5 billion). 3/ As noted earlier, it was toward the end of that year that CDI imposed a minimum size on projects which could be submitted for approval. A comprehensive analysis of recent institutional development of industrial policy is found in the essay by Wilson Suzigan "Politica Industrial do Brazil" appearing in Suzigan et al, Industria: Politica, Instituicoes e Desenvolvimento (IPEA/INPE, Rio de Janeiro, 1978).

35 Table 2.6: CDI APPROVALS, Total Value of Fixed Number of Projects Investment Approved (US$ million) n.a ,904 n.a ,851 3, ,976 4, , , , ,187.3 Source: CDI, Annual Reports for various years. a/ Cruzeiro values converted to dollars at the average exchange rate for the year As regards the subsectoral composition of the approvals, the lower priority attached to new projects among consumer goods industries after 1974 is reflected in the sharp decline in their relative share from 16.5% of the total value of fixed investment in 1974 to 5.7% in 1977 (see Table 2.7). A dramatic increase in relative importance is seen in the metals industries, reflecting the policy emphasis that was given to developing domestic productive capacity in that sector. While growing to an 18.3% share of the total in 1977, the capital goods industries fell back to 7.8% of the total in 1978.

36 Table 2.7: SUBSECTORAL COMPOSITION OF CDI APPROVALS AND SHARE OF LOCAL SUPPLIES IN EQUIPMENT REQUIREMENTS, 1974 AND 1978 (as percentage of Share of local supplies in total fixed investment) total equipment requirements Capital Goods Basic and Intermediate Metal Industries Chemicals and Petrochemicals (incl. Pharmaceuticals) Non-Metallic Intermediate Products, Paper and Cement Automotive Industry (incl. components) Consumer Goods Total: Source: CDI, Annual Reports for various years The capital goods sector was twice benefited. First, as noted, it was made a priority sector. Second, incentives granted to other sectors resulted in purchases of domestically produced capital goods. Imported machinery and equipment in 1978 accounted for 24.1% of the total approved purchases of such machinery and equipment as compared to 55.8% in Although the CDI was the agency most frequently involved with official project approval and the concomitant approval for the granting of fiscal and financial incentives, other government agencies exercised similar functions. A special program 1/, also administered by the Ministry of Industry and Commerce, permitted the granting of similar incentives, especially in the form of reduced import taxes, based on the commitment of a firm 1/ The program, entitled Beneficios Fiscais a Programas Especiais de Exportacao, or BEFIEX, was initiated in 1972.

37 to export a certain amount as a part of its project. It could grant larger import tax reductions than CDI (90% versus 80%). By 1977 total fixed investments undertaken as a part of BEFIEX projects amounted to US$3.5 billion, of which over US$3 billion was concentrated in the automotive sector. Some investing firms looked toward BEFIEX approval as an alternative to applying to CDI for new projects. This was notably the case with textile producers which received a very low priority with CDI (see Statistical Appendix, Table 12). In addition to CDI and BEFIEX, still other government agencies were empowered to grant various incentives for projects. They included, most notably, regional development institutions and sectoral development agencies A further complication in administering the policies of incentives for industrial development related to the import component of investment projects. While the CDI examined and approved the import content of its projects, subsequent analysis also occurred. After a project was approved by the CDI, its import content was further scrutinized by CACEX, the Foreign Trade Department of the Banco do Brasil. According to the Law of National Similars, a project could not receive the benefits of the incentives system unless the imports carried out under the project had no comparable domestically produced counterpart. The examination for the "national similar" was the responsibility of CACEX. 1/ This examination, providing considerable authority to CACEX, took place on several levels with respect to both the technical characteristics of the equipment in question and price considerations. If the domestically produced good was technically comparable to the prospective import, the investor could still import the product if he could demonstrate that there would be uncommonly long delays in obtaining the domestic product or if the domestic product price was excessively expensive. The initiative for such an appeal, however, rested entirely with the investor What constituted "excessively expensive" in the eyes of the policymakers, in this case CACEX officials, would be obviously crucial in any evaluation of protection. Here the nominal tariff rate for the product in question (generally capital goods) played an important role. The general rule applied by CACEX was that if the price of the imported equipment plus the tariff were lower than the price of the domestic counterpart, importation would be allowed with the full enjoyment of the permissible incentives, e.g., tariff reduction. Thus, the actual tariff entered into the decision whether a product could be imported or not The negotiating mechanism by which the examination of an industrial investment project's import content took place was, in most cases, the formulation of a Participation Agreement (Acordo de Participicao Nacional). Formalized tripartite discussions were held between the investor, CACEX officials, and the representatives of the relevant domestic capital goods 1/ Formally speaking, the Council for Tariff Policy (CPA) was responsible for setting tariff rates and granting exonerations while CACEX issued the licenses for all imports exonerated from tariffs. In order to facilitate administration, CPA delegated its responsibility to grant exonerations to CACEX in 1967.

38 producers' association. The purpose of these discussions was to examine the import content and the existence of suitable domestically produced counterparts. Once a Participation Agreement was reached concerning a project, the investor was permitted to import all the products agreed upon without further CACEX examination even though the project may have had a duration of several years The role of CACEX in the examination of the "national similar" and the formulation of Participation Agreements was critical. In granting import licenses, CACEX administers what amounts to a quota system on capital goods imports. After 1974, CACEX found itself under considerable pressure to reduce imports of foreign equipment and machinery, for both strengthening Brazil's balance of payments situation and providing greater protection for the domestic capital goods industry. Much concern had been expressed in Brazil as to the magnitude of the so-called nationalization index, or simply the proportional domestic share of equipment purchases, for investment projects. Through its control over import licenses and the formulation of Participation Agreements, CACEX had sought to increase the nationalization index for capital goods purchases. As can be seen in Table 2.8, the share of domestic capital goods purchased under approved investment projects with Participation Agreements grew from 53% in 1973 to 77% in Table 2.8: CACEX PARTICIPATION AGREEMENTS, Percentage Share of Number of Total Capital Goods Agreements & Capital Goods To be purchased To be Year Revisions Amount Negotiated Domestically Imported -- (US$ million) ( % ) , , , , (Jan. - June) Source: CACEX.

39 As noted, once imports under an incentive program were approved by CACEX, they could enter the country with tariff reductions. At the same time, the compulsory advance deposit on imports was also normally waived. In most cases foreign financing, generally accompanying capital goods imports, resulted in the automatic exemption of the imports from the import deposit. While until mid-1976 full tariff exemptions were common, from that point on CDI was limited to 50-80% deductions for capital goods imports. 1/ The permissible reductions in the IPI and ICM taxes on imports were identical. Only in the special case of presidential approval could a 100% reduction, or full exemption, of import taxes be granted. Although the intent had been to eliminate the complete exemptions, several cases were still allowed. 2/ While greatly reduced, the tariff reductions were still important and, in fact, were the general rule for the importation of capital goods. It has been estimated that about 80% of all capital goods imports entered Brazil under some sort of incentives program. This is clearly reflected in the comparison of the theoretical nominal tariffs with the tariffs actually collected. The realized rates were substantially lower than the theoretical nominal rates. As was demonstrated in Table 2.2, for all capital goods imports in 1977 the realized tariff rate, i.e., tariff collections divided by imports, was 14.1%, as compared to the theoretical nominal tariff of 46.6%. What this attested to in practice was not the lowness of tariff levels per se for capital goods but rather the way in which other mechanisms of import restrictions were employed. Tariffs nevertheless continued to exercise a role in these other mechanisms. D. Fiscal Incentives 2.25 The reduction of taxes on imported machinery and equipment was an important form of fiscal inducement to spur industrial growth. With an overvalued exchange rate, reductions or exemptions for import taxes, and waivers for import deposits, the domestic purchaser's price of imported capital goods appeared relatively cheap. This had been the Brazilian historical experience. One effect of artificially reducing relative prices of imported capital goods in the last 25 years has been to foster a capitalintensive pattern of industrial development. Consequently, technology choice and changes in industrial structure have been affected Keeping the relative prices for imported capital goods low, as a conscious objective of industrial policy, also implied that the actual and potential profitability for the domestic capital goods industry was kept low. In fact, a system of exchange rate overvaluation and tariff exemptions for imported machinery and equipment discriminated against the domestic capital goods industry. However, as poilned out in the first chapter, it was in 1/ The activities qualifying for the 50-80% reductions were listed in Portaria No. 442 of August 10, It should be noted that the products to be imported were not identified but rather the importing or using industry. For example, the production of most types of capital goods received 80% reductions on import taxes. On the other hand, no consumer goods production activity received more than 50%. 2/ In a well publicized case, the government firm Petrobras received a 100% reduction of import taxes for a project.

40 just such a general environment that the Brazilian capital goods sector developed in the postwar period until the mid-1970s. Despite the effects of discriminatory commercial policy, growth of the sector was impressive This situation was soon to change, however. By the late 1960s and early 1970s there was a growing awareness of this discriminatory treatment on the part of the increasingly vocal and politically influential Brazilian capital goods producers. There were growing pressures to redress this situation and undertake an active program of import substitution for capital goods. Various measures, described in the first chapter, were undertaken to assist the local capital goods industries. Yet, it was not until the worldwide petroleum crisis and subsequent Brazilian balance of payments problems, compounded by the swelling of capital goods imports related to CDI approvals, in the mid-1970s that further and more significant actions were undertaken. Decree Law 1335 of July 1974, implemented in March 1975, established a significant system of fiscal incentives for domestic producers of capital goods. These incentives more than overcame any discrimination stemming from an overvalued exchange rate and the apparent liberal granting of import tax reductions Decree Law 1335 provided that the purchase of domestically produced capital goods could be accompanied by exemption from the IPI tax. As a matter of course the ICM tax was waived as well. The elimination of the IPI and ICM tax liabilities for domestic capital goods purchases was necessary to put them on an even footing with imports if such border taxes had been exempted. Beyond the simple exemption of the IPI and ICM taxes, permitting a concomitant reduction in prices without changing unit profits, a direct fiscal subsidy was also allowable. The IPI tax, while technically a value of product tax, was in effect a value added tax as it was administered. Fiscal credits accumulated for a producing firm when it purchased intermediate inputs taxable by the IPI. These IPI fiscal credits could serve as the basis for a still greater reduction in tax liabilities for capital goods producers. This reduction constituted a direct fiscal subsidy. It was identical to the direct fiscal subsidy that the firm would receive if the final capital good product were exported instead of sold domestically. In fact, Decree Law 1335 simply extended the fiscal incentives for manufactured exports, based upon the indirect tax system, to the internal sale of domestically produced capital goods The quantitative importance of the Decree Law 1335 fiscal subsidy for domestic capital goods production varied over products because the IPI rates themselves varied, as did the degree of vertical integration in capital goods production. Column 4 of Table 2.5 presents estimates from a study by Pastore et al., of the D.L fiscal subsidy element for 16 capital goods industries 1/; the study was undertaken to measure export incentives, which were similar. As seen, while varying from 16% to 27%, the fiscal subsidy 1/ Affonso Celso Pastore, Jose Augusto Arantes Savisini, and Joao de Azambuja Rosa, "Quantificacao dos Incentivos as Exportacoes," Fundacao Centro de Estudos do Comercio Exterior, Rio de Janeiro, 1977, unpublished manuscript.

41 averaged about 20% of the ex factory price. The effect of the fiscal subsidies was to improve the competitiveness of domestically produced capital goods vis-a-vis imported capital goods. This was tantamount to a downward shift in the domestic supply curve for capital goods. The result ceteris paribus was an increase in domestic capital goods production The granting of the fiscal subsidies under Decree Law 1335, like so many other aspects of Brazilian industrial policy, was discretionary. The Ministry of Finance administered the Decree law 1335 regulations and, although in practice it generally agreed with recommendations made by the CDI, final authority for granting the fiscal incentives always rested with Finance. Unlike the CDI, which granted incentives primarily on the basis of industries to be promoted, the Decree Law 1335 subsidies by the Ministry of Finance focused on specific capital goods. Generally the criteria related to degree of sophistication and the period of time that domestic production had been taking place. If a product was not considered sufficiently "sophisticated" or if it had been produced in Brazil for a long time, it would not receive the protection afforded by the Decree Law 1335 incentives. In 1977 the Ministry of Finance approved approximately 200 projects, corresponding roughly to the CDI projects approved during the same year. As such, there was a maximum of about US$2.3 billion of domestic capital goods to which Decree Law 1335 incentives could have been applied. However, it is likely that the exact amount was well below that figure; no data are available on the actual amounts involved. E. Financial Incentives 2.31 In addition to the fiscal incentives allowed to the Brazilian capital goods sector, financial incentives for the purchasers of Brazilian capital goods were also provided as a part of official policy to promote the industry. Interest rate subsidies were provided through FINAME and other official financing programs. The result of these financial subsidies was to enhance the competitiveness of the Brazilian capital goods producer. It has been estimated that for 1978 these subsidies made possible an 8.8% reduction in price to the purchaser in relation to the firm's ex factory price (see Annex III). In 1977, the monetary correction index, with which all FINAME loans were then adjusted, lagged considerably behind the rate of inflation. Consequently, the financial subsidies provided to the capital goods industries were greater than they were in The 1977 credit subsidies were estimated to have amounted to 29.4% on a value of product basis The principal source of funds provided at rates below prevailing market rates was FINAME. At that time the agency had three programs favoring capital goods industries which either financed domestic sales of local producers or provided credit to purchasers to buy locally produced items. Toward the close of the period, FINAME's operations underwent a massive increase, total commitments rising from about US$1.0 billion in 1975 to US$2.5 billion in 1976 and US$1.9 billion in / FINAME's financing appears to have 1/ The particularly large increase in 1976 reflected in part the fact that, for most of the year, there was a ceiling on the monetary correction (of 20%) resulting in highly negative real interest rates. See Annex III.

42 been concentrated mainly in heavy equipment and machinery which would be included among 10 of the branches analyzed in this report. The six excluded are agricultural equipment, office equipment and machinery, tractors, trucks and buses, motors and vehicle parts, and shipbuilding. For the specific product lines included, FINAME financed a substantial proportion of their local sales. However, in relation to the total value of capital goods produced for sale within the country (i.e., excluding exports), the agency's disbursements in 1976 and 1977 (US$1.0 billion and US$1.3 billion respectively) represented less than 10%. 1/ Its total commitments during those years (which would mainly be reflected in future production) were US$2.5 billion and US$1.9 billion or the equivalent of 10-15% of output. More detail on FINAME's operations are given in Annex III FINAME also played a role in implementing the policy adopted by the government after 1974 to increase the local content of capital goods production. From its inception in 1965, the agency had required a minimum of 50% local content in the equipment it financed. In 1974 this was raised to 67% and thereafter annually raised until the minimum limit reached 85% in In addition, it provided incentive to exceed those minima by offering interest rate differentials. F. Protection of Intermediate Inputs 2.34 A regime of overvalued exchange rates necessarily had to be accompanied by other forms of import restrictions. These restrictions on imports were also used for protecting certain industries. In Brazil, as in other countries where import restrictions have been employed to spur import substituting industrialization, nearly all industrial activities were affected by higher costs brought about by protection of industries supplying intermediate and capital goods inputs. The Brazilian capital goods sector was no exception. Protection afforded to Brazilian steel producers of intermediate inputs adversely affected the competitiveness of the capital goods sector. As the effective protection literature points out, 2/ the effects of protection on inputs must be considered in any assessment of net protection A recent study of effective protection in Brazil has employed realized tariffs to approximate protection. 3/ Their estimates, using 1975 tariff information and the 1970 IBGE Input-Output table, included separate 1/ Based on the data in Statistical Appendix Table 3, the current value of capital goods production in 1977 was estimated at US$14-16 billion. 2/ See Bela Balassa et al., The Structure of Protection in Developing Countries (Baltimore: Johns Hopkins University Press, 1971). 3/ Paulo Neuhaus and Helenamaria Lobato, "Protecao Efetiva a Industria no Brasil, ," Fundacao Centro de Estudos do Comercio Exterior, Rio de Janeiro, 1978, unpublished manuscript.

43 calculations for the protection on inputs. They are presented in Column 1 of Table 2.9. The Corden method estimates have been used to calculate the proportional effect on ex factory prices that protection on inputs can exercise for 16 capital goods industries. 1/ These estimates are presented as column 2 in Table 2.9. They range from a low of 2.4% (shipbuilding) to 10.3% (trucks and buses). For industrial equipment and machinery, the estimate of the price increase due to protection on inputs is a very low 3.4% It should be noted that the effect of protection on inputs appears to have been underestimated; the use of realized tariffs substantially understates the difference between domestic and international prices. An important example of this difficulty is the case of steel imports. Steel products constituted one of the most important intermediate inputs for capital goods production, accounting for 8-13% of the value of final products in the different capital goods industries. 2/ The 1975 realized tariff rate for steel imports was only 1%. Yet, the imports of steel have not been restricted simply through tariff measures. The government steel council CONSIDER established and administered a quota for steel imports. Consequently, the 1% realized tariff rate understates domestic and international price differences. In fact, some Brazilian capital goods producers complained that the domestic price of some internally produced specialty steel, e.g., stainless, were over twice the current international prices. 3/ 2.37 The underestimation of the effects from protection on inputs resulted in an overestimate of protection afforded to the production processes in question. If, on the other hand, all inputs could have been freely 1/ An explanation of the estimating procedure employed is provided in Annex II. 2/ For the Industrial Equipment and Machinery industry, the relevant technical coefficient for all steel inputs was IBGE, Matriz de Relacoes Interindustriais: Brasil (Rio de Janeiro: IBGE, 1976), p / While international prices reflect the disordered present situation in the world steel market, a recent study has presented some evidence indicating that for some basic steel items Brazilian prices do not compare unfavorably with the domestic market prices for the major steel exporting countries. See Instituto Brasileiro de Siderurgia, "O Preco dos Laminados de Aco no Brasil e no Exterior: Estudo Comparativo," Comissao Tecnica de Economia, unpublished report, June 1977.

44 Table 2.9: TARIt PRtECTICN ON INPUTS. AMD ZNACT ON MuCIS of CAPITAL GOODS Realized Tariff Protection ERtitated Dmestic Product tport Content of IBGE Cod on Inputs (Cordon Method) ftice Incrta, due to Final Products Code Category t!9 Al _te onm (X) b/ 121 Pumps & Motors Machine Parts Industrial Equipment & Machinery Agricultural Equipment & Machinery Office & Domestic Equipment Machinery Tractors Equipment for Electrical Energy Electric Conductors ' Electric Equipment Electrical Machinery Electronic Equipment Comiunications Equipment 1' Trucks & Buses Motors & Vehicle Parts Shipbuilding Railway & Aircraft Equipment Source: Columns I anm 2 from Paulo Neuhaus and Helon ria Lobato, ProteC;o Efotiva a tnduatria no Brasil , Cundagio Cen:ro de Estudos do Comercio Exterior, Rio de Janeiro, Column 3 from IBGE, Matrir de Relacoes Interindustriais: 3rasil Rio de Janeiro, 1977 a/ Based on 1975 data. b/ Import coefficients from the 1970 Input-Output table.

45 imported at world prices, the estimates of net protection would have been larger. In the absence of trade restrictions, the import content of Brazilian capital goods production would have been higher. The lowness of the import content for most capital goods industries is indicated in Column 3 of Table 2.9. For the Industrial Equipment and Machinery industry total imports used in production oy the industry accounted for less than 2% of the industry's output in 1970.

46 III. ESTIMATING NET PROTECTION A. Basic Estimates 3.01 After having discussed some of the individual policies relating to relative prices for Brazilian capital goods industry, the various elements can now be considered together and in doing so return to the central question posed by this study. To what extent was the Brazilian capital gqods sector protected or unprotected by the constellation of economic policies affecting it? Annex II presents a measurement technique for examining net protection. The measure of net protection for product j(pj) incorporates the effects of import restrictions, fiscal incentives, financial subsidies, the protection on inputs and exchange rate overvaluation. The rationale is to compare the possible prices to buyers for imports and domestically produced products after the effect of economic policies is taken into account. To undertake this comparison, possible prices are expressed in relation to unit import CIF prices and unit domestic ex factory prices. This means that, prior to any adjustments for economic policies, the ratio of the duty free import CIF price and the ex factory domestic product price is After adjustments for import restrictions, fiscal incentives, and financial subsidies, adjustments are then marlo' for other policies through a coefficient of distortion (0), reflecting th, effects of (1) exchange rate overvaluation and (2) protection on inputs. 1/ 3.02 Exonerations (partial or full) from import levies (including prior deposits) were generally granted to projects and, after a complex set of negotiations (described in the previous chapter), the specific pieces of imported equipment to receive these benefits were identified. Fiscal and financial subsidies were generally provided directly to locally manufactured pieces of equipment. Thus, effective protection for any item of equipment was not unique to the equipment and depended on a great deal of administrative discretion under the various policy schemes. As a consequence, it is only possible to estimate a range of net protection which would reflect the impact of the different policy elements involved Column 1 of Table 3.1 presents the price to the buyer of imported capital goods if all import taxes were charged. This would have been the case if the product in question were to be imported outside an official investment project scheme, i.e., without obtaining official approval. The Law of 1/ The measure of net protection (pj) can be written as (1) pj = P:/ 0 pj~~p Dj where PMj = the effective purchase price of imported product j, PDj = the effective purchase price of domestically produced product J, 0 = coefficient of distortion.

47 Table 3.1: EFFECTS OF ECONOMIC POLICIES ON PRICES OF IMPORTED AND DOMESTICALLY PRODUCED CAPITAL GOODS, 1978a/ Domestically Produced Product Prices to Import nrice to Buyer Buyers Coefficient IBGE Without Tariff Without D.L With D.L.1335 Incentives of Code C.vbe Category ~ Reductions PMO 0 Incentives Subsidies & Financial P 2D- O and Financial Subsidies PDJr. 1 Distortion 121 Pumps & Motors Machine Parts Industria Equipment & Machinery o Agricultural Equipment & Machinery Office & Domestic Equipment Machinery Tractors :31 Equipment for Electric Energy Electric Conductors Electric Equipment Electrical Machinery Electronic Equipment Communications Equipment Trucks & Buses Motors & Vehicle Parts Shipbuilding Railway & Aircraft Equipment a/ Possible prices are expressed in terms of unit CIF prices and unit domestic ex-factory prices. Source: Mission estimates; the details of the calculations are described in the text and Annex II.

48 National Similars was not applicable and full duties were charged. 1/ The combined effect of the taxes and import deposit was roughly double the CIF import price. On a unit CIF import price basis, the purchase price for an average product in the Industrial Equipment and Machinery category was 2.22 times the unit price if no import charges were levied. For Electrical Equipment, where there was a higher tariff, the comparable figure was If there were tariff and import tax reductions, the buyer's price of the imported product dropped accordingly. In the case of Industrial Equipment and Machinery, the prices for a 50% and 80% import tax reduction, as possibly granted by CDI, were 1.34 and 1.13 respectively. If there was a 100% reduction, while rare, the buyer's price for all capital goods categories was The purchase price of the domestically produced product, beginning from an equal unitary ex factory price, depended upon whether the fiscal incentives and financial subsidies were granted. In their absence, the buyer's prices of the domestic product for the 16 capital goods industries are given in Column 2 of Table 3.1. They range from 1.17 (Pumps & Motors) to 1.36 (Electronic Equipment), differing because of differences in the IPI over products. Column 3 of Table 3.1 lists the buyer's price if full allowable fiscal incentives and financial subsidies had been granted. As demonstrated, these Decree Law 1335 incentives and official financing arrangements made a very important difference in the possible buyer's price. The prices averaged around 0.65, the original ex factory price, with that for Industrial Equipment and Machinery being These subsidies were indeed powerful policy instruments The coefficient of distortion (0), as noted above, consists of two components: (1) any departure in practice from a shadow exchange rate and (2) the estimated domestic product price increase due to protection on inputs. Accordingly, 0 adjusts for distortions in the-prices facing producers and purchasers of capital goods which resulted from economic policies not dealing specifically with the industry in question. A producer will make the comparison between the actual buyer price in the market for imports and domestically produced goods, but the choice for society as a whole should more appropriately reflect distortions in the exchange rate and in the prices of inputs due to protection of those inputs. Whether or not a production process is protected or disprotected in a social sense depends on an entire constellation of economic policies. The use of the coefficient of distortion (0) adjusts for the effects of two major policy elements on the competitiveness of a certain industry Estimates of the coefficient of distortion (0) are listed in Column 4 of Table 3.1. As seen, 0 varies over industries (or products). This is because the estimated domestic product price increase due to protection on 1/ The Law of National Similars only applied to imports included under an official investment program, enjoying the benefit of government investment incentives. Imports could be undertaken without the CACEX examination of national similarity if there were no government incentives involved. In other words, despite all the attention given to the Law of National Similars, the latter had no protective teeth unless some government incentive scheme was involved.

49 inputs varies from industry to industry. Based on existing estimates of what would be an equilibrium exchange rate, 30% has been arbitrarily chosen as the degree of exchange rate overvaluation; this affects all industries uniformly ifferent combinations of policies actually implemented, of course, imply different effects on net protection that were in effect during Table 3.2 presents some alternatives under the administrative procedures. A net protection measure (pj) greater than one indicates the industry was protected, while a pj less than one denotes that the combined effect of economic policies serve to leave the industry negatively protected. A p. of 1.00 indicates there was a neutrality of all commercial policy, favoring neither imports nor domestic production Column 1 of Table 3.2 presents the measure of net protection with full import levies and without concessionary fiscal and financial subsidies. 1/ The P.'p ranges from a low 1.15 (Shipbuilding) to as much as 1.86 (Motors & Vehicles Parts). In no case was any capital goods industry disprotected. The net protection for the important Industrial Equipment and Machinery industry, under these circumstances, amounted to 34%. These estimates indicate that if all the legal tariffs and charges were levied on imports, the subsidization of domestic production would not have been required to protect the domestic capital goods industry. They also illustrate, however, that the elimination of the compulsory import deposit, without domestic production subsidies or compensating devaluation, would have left capital goods industries unprotected The relevance of the Column 1 estimates of possible net protection to actual protection is high. Many investment projects were not eligible for government investment incentives. In addition, a number of capital goods users preferred to import some capital goods in this manner, claiming that the uncertainty and delays of the official incentive schemes were at times too great to offset any potential private benefits to the investor. Consequently, if capital goods imports were to be undertaken, the prevailing tariffs and import deposit levels were applicable. Accordingly, the levels of protection afforded to the capital goods industries appear to be high, excessive in some cases. However, to the extent that the effects of protection on intermediate inputs have been understated, as is believed to be the case, the estimates of possible net protection in Column 1 are biased upward Combining the existing tariffs and import charges with the full complement of subsidies on domestic production raised substantially levels of net protection. This is demonstrated in Column 2 of Table 3.2. In every capital goods industry net protection was at least 200%. Such protection was clearly excessive. However, in practice this was not the normal case. If full domestic subsidies were allowed as a part of an official investment project, the equipment would be eligible for reduction of the import taxes actually charged. Thus, the estimates in Column 2 ceteris paribus are a maximum and an extreme. 1/ This measure equals the ratio of the estimates in Columns 1 and 2 of Table 3.1, divided by the coefficient of distortion in Column 4.

50 Table 3.2: NET PROTECTION (o.) UNDER ALTFRNATIVE PRa(sRAM SCHEMES FOR STXTEEN CAPITAL COODS INDUSTRIES _~~~~~ 9. _ Without Tariff and Border Tax Reduction via witi, 100% Tariff and Border Tax jed1ctilon via official Programs, i.e.. Full Tariffs Official lrograms, i.e., cu = 1.0 Suctor _without Concesslonary With Full Concessionary Without Concessionary With Full Concesstonary ( Id cs- Iucentives aud Subsidies Incentives and Subsidies Incentives and Subsidies Incentives and Suabsidies If lcation for vomestic Production, for tiomestic ProductJon, for Domestic Production, for DoMestic Productioln, Code 1ndustry 1.e. y - O i*e., y - I.. e.,e11 I 121 Ptupsalmotors Machiise Parts Industrial Equip. Machiinery Agricuiltural Equip. & Mach Office & Domestic Equip. & Mach Tractors Equip. for Electric Energy Etectrical Conductors Electrical Equip Electrical Mach w 115 Electronic EquIp ON 136 Cowunlcations Equip TruckB and UuBes Motors b Vehicle Parts shipbuildivig Railway & Aircraft Equip Source: Information in Table 3.1. For method of computation see text and Annex II.

51 An opposite, and equally unlikely, extreme is the case of a 100% reduction of import taxes via official programs without any allowable domestic production subsidies. These estimates are presented in Column 3 of Table 3.2. In this case all the domestic capital goods industries were disprotected by policies. It demonstrates that, in the absence of tariff protection and domestic production subsidies, some protection of the capital goods industries was necessary to overcome the disprotection afforded by exchange rate policy and the protection of intermediate inputs. These latter two policies discriminated against the capital goods sector, as well as against other manufacturing activities As is the case with all of the estimates presented in Table 3.2, Column 3 merely presents possible magnitudes of the net protection that occurred under different policy applications and mixes. However, the Column 3 estimates of net protection are unlikely to be those that actually took place, for two reasons. First, the 100% reduction in import taxes was an extreme, only to be granted under exceptional circumstances by presidential decree. The latest reductions were of either 50% or 80%, depending on the nature of the project, with the latter figure being the maximum allowable reduction under CDI procedures. Consequently, the disprotection evidenced in Column 3 of Table 3.2 is overstated. Nevertheless, for even a 50% reduction in import taxes, again without domestic production subsidies, the net protection (pj) for the Industrial Equipment and Machinery industry would have been negative. In other words, in this situation, without domestic production subsidies and with the waiving of the prior import deposit, the domestic industry was still disprotected A second reason why the possible levels of protection, i.e., disprotection, are unlikely to have occurred in practice concerns the use of the domestic production subsidies. In a project submitted to the CDI for the receipt of official government incentives, it is improbable for that project to have been granted tariff reductions for imported equipment but no domestic production subsidies for the items produced locally. In practice, the two types of incentive systems went together, although in reality they may have been applied to different capital goods components within the same project Column 4 of Table 3.2 presents the more likely case of both allowed tariff incentives (reductions) and domestic production subsidies for an investment project. As seen, even with the rare 100% reductions in import taxes, the existence and use of domestic production subsidies was sufficient to ensure positive net protection to every Brazilian capital goods industry. Ranging from 12% to 30% for all capital goods industries, the measure of net protection, under these conditions, for Industrial Equipment and Machinery would be 22%. Naturally, in the case of less than a 100% reduction in import taxes, i.e., a 50% or 80% reduction, net protection was even higher Comparing Column 1 and Column 4 in Table 3.2, it can be seen that the subsidies on domestic production were a more important form of protection for the Brazilian capital goods industry than the tariffs. These subsidies could have been substantial and could have resulted in very high levels of protection.

52 In concluding rather high levels of net protection for the Brazilian capital goods sector, three important qualifications must be borne in mind, the effect of each being to reduce the estimates of net protection. First, as noted, the effect of protection on inputs has probably been understated. This serves to bias downward the estimate of the coefficient of distortion, consequently resulting in an overestimate of net protection. Second, full fiscal and financial incentives were granted to only a limited number of capital goods. As a result, the estimates incorporating the production subsidies could be thought of as measures of net possible protection. If provided, these subsidies could boost the protection to extremely high levels. Finally, for those items which may have been recipients of financial subsidies, the net effect of those subsidies may have been overstated. Frequently, importers of capital goods will receive subsidized credit from their overseas suppliers, and the effect of this subsidy has not been taken into account The high observed levels of net protection for the Brazilian capital goods sector raises an important normative question. Should the capital goods industries have been protected? Based on previous Brazilian experience, the arguments are strong that there should have been some protection. Nevertheless, it appears that the observed levels of net protection, when full domestic production subsidies were applied, were excessive. The policy in force in 1978 appears to have been one of protectionary overkill, applied at the discretions of the authorities The most difficult element in the system of protection for capital goods was its discretionary nature. As has been demonstrated, the magnitude of protection actually provided could vary widely. And in actual practice it did so on a case-by-case basis. There was little, if any, evenness or automaticity in the system. Individual purchasers of capital goods were affected by separate government decisions, perhaps in a variety of agencies, related to the granting or non-granting of a multitude of investment incentives. Domestic producers of capital goods were faced by a continual uncertainty of not knowing whether the fiscal production subsidies would be granted from one sale to the next, or whether competing imports would be favored through tariff and prior deposit exonerations. Thus, as it was organized, the overall operation of the system was bound to be random and uneven A further question is raised by the use of subsidies for domestic capital goods production. Like the policy of exchange rate overvaluation combined with tariff exemptions, the effect of the use of subsidies for domestic capital goods production was to lower the relative cost to the investor of using capital in production. Thus, a further cost of subsidizing the development of the Brazilian capital goods industry may have been to continue the capital intensive industrial development in other sectors of the economy.

53 B. Indications of Competitiveness 3.21 In examining the effects of economic policies, the estimates of net protection have made the assumption that CIF import prices and domestic ex factory prices were the same. In reality, of course, they were not. There were other factors which affected economic costs. The extent to which domestic and international prices were comparable reflected the competitiveness of Brazilian capital goods. It is clear that Brazilian commercial policies affected the competitiveness of Brazilian products. Given those policies, it is possible to make some price comparisons to examine the actual competitiveness of Brazilian capital goods during the period under review. Given that the development of the capital goods sector was an objective of government policy, the question is whether, in view of prevailing price competitiveness, Brazilian producers actually needed the protection that government policy frequently provided Before making price comparisons of Brazilian capital goods with their internationally produced counterparts, two general observations can be made. First, Brazilian exports of capital goods were growing steadily and dramatically. Between 1973 and 1976 capital goods exports grew at an annual average rate of 49%, 1/ much more than for Brazilian exports in general, or for total manufactured exports. While those exports were still small relative to total domestic output, their growth suggests increased competitiveness. Moreover, the observed variety of Brazilian capital goods exports, ranging from simple machinery to heavy armaments, suggests that reasonable competitiveness was achieved along a broad front It is true that one of the reasons accounting for the increased international competitiveness and growth of Brazilian-manufactured exports was their package of export incentives. All indirect taxes were rebated for manufacturered exports, and there was also a pure fiscal subsidy element, as was shown in Column 4 of Table 2.5. In addition, there was a credit subsidy, provided principally through the provision of heavily subsidized working capital for the purpose of export production. The combined magnitude of the export subsidies could be substantial for capital goods, ranging from 20% (Electronic Equipment) of the exported product's value to 36.5% (Tractors). 2/ It should be noted, however, that (i) the export subsidies had been in place for ten years and there had been little appreciable change in their magnitude, especially for the fiscal subsidies, and (ii) the subsidies for manufactured exports for most industries were less than the amount of exchange rate overvaluation. Consequently, while the export subsidies may have been important in having promoted Brazil's exports of capital goods, they alone were an insufficient explanation. 1/ See Suely Barbosa Monnerat, "Comercio Externo na Area de Bens de Capital" - Periods: , EMBRAMEC, November 1977, unpublished report. 2/ Statistical Appendix Table 10 provides estimates of the export subsidies for the capital goods industries.

54 It is useful to compare the magnitude of the subsidies for the export of capital goods with the level of protection in the domestic market. The question is whether, given export promotional and import protectionist policies, profitability was greater in producing for the export market or the domestic market. For this comparison, use could be made of the estimates of net impqrt protection, (pj) assuming full domestic production subsidies and total tariff exemptions (Column 4 of Table 2.5). A measure of net export promotion could be constructed analogously, using the estimates of the fiscal and credit subsidies for export given in the Statistical Appendix, Table 10. In the case of every capital goods industry, it is seen that net import protection exceeded net export promotion. 1/ In other words, with the use of full domestic production subsidies, production for the domestic market was more profitable than production for export. The use of the full domestic production subsidies provided a net anti-export bias for the Brazilian capital goods sector A second general indication of improved competitiveness for the Brazilian capital goods sector can be seen in the movement of relative prices within Brazil. Over the prices of capital goods declined relative to prices for other industrial products (Table 3.3). While saying nothing about absolute international competitiveness, such a movement in relative prices domestically suggests that Brazil's competitiveness improved over time. The relative decline in capital goods prices was most likely related to the significant technological progress occuring in the sector. For example, a recent study has estimated that the annual proportional rate of Hicks neutral technological progress for both the electrical machinery and electrical and communications equipment industries was 18% for the period. 2/ 3.26 Table 3.4 presents some price comparisons for Brazilian and internationally produced capital goods, taken from a 1978 BNDE study. Prices were given in US dollars FOB and were direct firm price quotations. For the Brazilian products, the price effects of the incentives for manufactured exports were included. In other words, tax rebates and export subsidies were reflected in 1/ Alternatively, this can be seen by comparing Column 3 of Table 3.1 with 1 minus the estimates given in Column 3 of Statistical Appendix Table 10. 2/ Hicks neutral technological progress is that technological change which increases the marginal products of the factors of production proportionately so as to leave unchanged the marginal rate of technical substitution at a given factor price ratio. See William G. Tyler, "On Using Analysis of Covariance to Estimate a Cobb-Douglas Production Function: An Empirical Illustration with Data from the Brazilian Electrical Machinery Industry," University of Florida Working Papers in Economics, No , June A similar estimate was obtained for the machinery industry.

55 the Brazilian price quotations. Also, the Brazilian export prices were provided at the prevailing overvalued exchange rate. Out of the 17 product price comparisons listed in Table 3.4, in only 3 cases were Brazilian prices more than 30% greater than the price of the comparable foreign product. In most cases the prices were quite similar The BNDE study also examined in some detail prices for roadbuilding machinery and equipment. Again, direct FOB price comparisons in US dollars were made, based upon producer price quotations. In 30 separate price comparisons of comparable products, only 18 were found to disfavor Brazil. 1/ Of these, 8 comparisons related to heavy treaded tractors, where Brazilian prices seem to have been substantially higher than international price levels. The remaining price comparisons demonstrated great similarity in Brazilian and international export prices. Analysis of export price quotations, particularly under prevailing conditions in international markets, may not reflect actual cost-price relationships and competitiveness. Direct studies of these aspects would be required In connection with the question of export prospects in 1978, a consultant to the staff of the Brandt Commission undertook a study of the competitiveness vis-a-vis developed countries of the capital goods producing industries in more advanced LDCs (including Brazil, Argentina, India and Korea) for a sample of certain standardized pieces of equipment. The latter included diesel engines, electric motors, tractors, pumps, textile spinning equipment and trucks. Results of this study indicate that the prices ex factory quoted for Brazilian products (i.e. before the calculation of any export incentive) were competitive with comparable items produced in the industrialized countries as well as in other LDCs In considering the operation of the incentive-promotion policies in the period , it is concluded that the system was not overly protective and did not encourage the creation of capital goods industries unable to compete in international markets. In particular, the substantial number of very small "projects" which received exonerations in the period through 1974 suggest that the bureaucratic requirements for such applications were not a large burden. Thus, as pointed out above (para. 2.16), the relatively cheap import of capital goods does not appear to have been restricted. The consequence of this was that the local manufacturer of capital goods had to compete rather strongly with imports. Interviews with industrialists in the private sector (capital goods users) indicate that the domestically produced items were generally price competitive. Moreover, the decision to purchase local products was often based on factors such as convenience of servicing and ready supply of spare parts, elements which constituted "natural" protection. There is some evidence, however, that for large public sector investments (e.g., in the steel industry), official pressures were exerted to channel procurement to local sources with some resultant stimulus to high-cost production. 1/ EMBRAMEC, "Competitividade da Industria Brasileira de Bens de Capital," unpublished paper, 1978.

56 Table 3.3: BRAZILIAN WHOLESALE INDUSTRIAL PRICE INDEXES, (lst Semester 1969 = 100) (1) (2) (3) (4) (5) Relative Prices of Relative Electrical All Prices of Electrical Machinery Industrial Machinery Machinery & Equip. Year Products Machinery ( 2 )/(1)x10O & Equipment (4)/(1)x10O 1969 (1st Semester) (June) Source: Coniuntura Economica, various issues.

57 Table 3.4: SOME PRICE COMPARISONS BETWEEN BRAZIiIAN AND FOREIGN PRODUCED CAPITAL GOODS (1) (2) (3) (4) Price Quota- Price of Ratio of tions of Comparable Country of Brazilian Brazilian Foreign Origin for Price to Foreign Industry & Product Product Foreign Price Product (US$ FOB) (US$ FOB) Product (1)t(2) Machine Tools Bench Lathe a Taiwan 1.03 b. 1,550 1,380 USA 1.12 Parallel Lathe a. 16,811 11,806 Czechoslovakia 1.42 b. 9,422 9,925 Argentina 0.94 Bench Perforator a Rhodesia 1.06 b ,363 USA 0.60 Textile Machinery Spinning Machine a. 40,968 40,000 Europe 1.02 b. 41,735 40,000 Europe 1.04 Looms a. 3,661 3,000 South Korea 1.22 b. 7,674 9,500 Europe 0.81 Agricultural Machinery Automated Harvestor a. 29,411 23,000 Europe 1.27 b. 32,866 23,000 Europe 1.42 c. 29,330 23,000 Europe 1.27 d. 24,203 23,000 Europe 1.05 Tractors w/wheels a. 7,083 8,900 Europe 0.79 b. 9,608 10,404 USA 0.92 c. 9,383 7,658 Europe 1.22 Source: EMBRAMEC, "Competitividade da Industria Brasileira de Bens de Capital," unpublished paper, 1978.

58 IV. PROSPECTS A2ND PRINCIPAL POLICY RECOMMEDNDATIONS 4.01 The situation facing the capital goods sector at the beginning of 1979 bore great similarity to that of the early 1960s. Over-investment in capacity among the capital goods industries had resulted from the "euphoria" of the early 1970s. The recent reduction in public sector investment (as a result of measures to reduce inflation and the balance of payments deficit), along with stagnation in private sector capital outlays, had led to considerable excess capacity. The problem appeared to be particularly severe for some of the new facilities which produce items of more advanced technology or complexity, which have limited markets in Brazil The difficulties in the short term appeared formidable and, even under the best of circumstances, given the usual lags between investment decisions and actual implementation, were unlikely to be solved quickly. In the medium term, however, prospects appeared to be considerably brighter but there was need to focus on a number of critical problem areas in order to ensure expansion of the subsector A substantial proportion of existing plants were technically efficient by international standards. Given reasonable levels of capacity utilization, their price competitiveness would be clearly sensitive to the exchange rate-export incentive policy mix. Considerable progress had been made in developing export markets for standardized items (e.g., machine tools, diesel motors and trucks). However, only limited success had up to that point been achieved in the export of more complex pieces of equipment, which were generally sold as part of a total plant or facility package. The successful bidding to supply the generating equipment for the Itaipu dam was made possible primarily by the sales financing arrangements provided by FINAME. Further expansion of this type of export transaction would be closely linked to availability of finance In the past, the authorities appear to have been reluctant to undertake major programs for long-term export financing, particularly in view of the overall tight balance of payments situation. The position at the end of 1978, with both a high level of foreign indebtedness and gross foreign exchange reserves equivalent to about nine months' imports, while hardly completely satisfactory, nevertheless did suggest that there was room for a carefully designed export finance program At the same time, the re-assessment of priorities among economic sectors, which was being undertaken in connection with the changes in administration, needed to be extended to subsectoral problems. There is little doubt that efforts made to develop high technology enterprises in the late 1950s involved large costs, some of which were excessive. For example, in the case of commercial vehicles (even more so for automobiles), the requirements for virtually full domestic value by the end of the 1950s and early 1960s led to local production of some very complex parts at extremely high costs, which contributed to the high prices charged for those vehicles. It was not until the 1970s that levels of output were reached which supported efficient production of those items.

59 There was a need, therefore, to establish priorities among the subsectors which could efficiently upgrade technical processes. Market size would clearly be a major factor in such choices. For example, efforts were being made to obtain technology for production processes for a number of petrochemical products, but market conditions could be expected to limit the number of core plants (e.g., ethylene) in the foreseeable future Closely related to the efficiency of the subsector was the problem of excessive dispersion of production. Various BNDE studies indicated that, in several complex product lines, there were then a relatively large number of firms, e.g. from 2-8 for electrical equipment, whereas in more advanced countries there were only 1 or 2 local firms producing these items. It was felt that this fragmentation could interfere with efficient production and lead to higher unit costs. It must be noted that there were numerous instances in which the government policies to encourage foreign investment, pursued both during the 1950s and at the end of the 1960s and early 1970s, contributed to this development. Competing foreign firms, sensing the dynamism and potential large size of the Brazilian market, moved quickly to establish a position in that market Rationalization of output in these particular branches under these circumstances would be particularly difficult. Complete reliance on market forces might not lead necessarily to acceptable solutions, since operating decisions for the foreign-owned firms might depend on overall corporate strategies in which the current profitability of the Brazilian subsidiaries would be only one element. For this and other reasons, the Brazilian authorities considered administrative measures to promote the reduction in numbers of enterprises. For example, lending from official financial institutions (both BNDE and its subsidiary FINAME) would be restricted to encourage the rationalization process. As in the case of many of the incentive programs analyzed in Chapter III, this approach involved considerable discretionary authority and was not without its difficulties. 1/ 4.09 The government's stated objective of avoiding, on the one hand, monopoly power and, on the other, an uneconomic number of enterprises in one product line in the capital goods subsector 2/ would be substantially served in a policy framework encouraging a more open economy. As the history of these industries had shown, both supply and demand conditions for the bulk of these items were conducive to supporting efficient production in Brazil 1/ A heated controversy over a BNDE decision not to provide credit to a proposed new heavy equipment plant in Belo Horizonte illustrates the difficulties. Promoted by the Minas Gerais authorities, and sponsored by foreign interests, the plant would be producing a number of items already under production in the country, and BNDE officials believed that there was no justification for further competition. Under strong regional political pressures, the bank subsequently reversed its decision and is now providing some funding. 2/ From Resolution 9/77 of the Conselho de Desenvolvimento Economico; see para

60 which would meet external competition. Moreover, given the long-term needs for developing new exports, the successes already achieved indicate considerable scope for further expansion in external sales under an appropriate policy framework During the years , the measures which the government undertook for the capital goods industries introduced a substantial degree of protection in contrast to the more open economic environment for the subsector which had been provided during the prior periods of its impressive growth. Latest indications were that the efficiency of only a small number of those industries was adversely affected, essentially the newer lines of production among the more advanced heavy equipment producers. For already existing lines, the system adopted probably raised the levels of profit. While it is unlikely that, in the few years of high protection, major inefficiencies could have been introduced, continuation of these policies could have resulted in substantial distortions The principal policy recommendation emerging from the analysis of the system of import protection and export promotion which prevailed in the period 1974 through 1978 was that there was need for a complete revision of that system. Such a revision would focus on the exchange rate overvaluation, the tariffs on capital goods, prior import deposits, export subsidies and the fiscal subsidies for local production, in particular the discretionary nature of their application In January 1979, in an attempt to eliminate the distortions which had been introduced into the economy, the government announced a number of major changes in basic macroeconomic policies. These included: (a) the mini-devaluations would be accelerated so that by mid-1983 a real devaluation (relative to January 1979) of 25% would be achieved; (b) the prior import deposit system would be eliminated gradually by mid-1983; and (c) the subsidy element in export incentives, i.e., the tax credits in excess of the actual IPI applicable to an export item (see para. 2.28), would be eliminated gradually by mid In addition to these measures of a general nature, the government also announced the elimination of the fiscal subsidies for local capital goods production under D.L Applications for these incentives which were received prior to January 24, 1979, would be reviewed and granted under the prior administrative regulations governing their approval, but no new applications would be accepted. While there existed a decree law (1428) which authorized an automatic subsidy (up to 15%) for local capital goods production, the necessary regulations to implement this had not been issued and the government had indicated its intention not to use the mechanism, particularly in view of the budgetary implications during such a period of fiscal restraint.

61 This would mean that future protection for local capital goods industries would have to rest basically on the tariff structure and the application of the Law of National Similars in providing tariff exonerations There was a growing conviction, in both government circles and the private sector, that the Law of National Similars, as administered in most recent years within the context of the complex incentive structure which had existed, had not been an effective instrument for promoting the growth of efficient domestic capital goods industries. On the contrary, discussions with some industrialists suggested that the overall impact of the system was to encourage project designs which initially required high levels of imported equipment, so as to provide a strong bargaining position to the project authorities for the intricate negotiating procedures involved in the application of the law. Since the decisions regarding import exonerations or subsidies to domestic output were highly discretionary, producers of specific pieces of equipment had no assurance of their competitive position vis-a-vis imports; as pointed out in the earlier analysis, the same equipment item could in one case receive a subsidy for local production while in another it would receive treatment favoring its import While the elimination of production subsidies and the rationalization of commercial policy described above also eliminated most of the discretionary aspects of the system for protection, the Law of National Similars continued to be applicable to obtain tariff exonerations. As in the past, these exonerations would depend on the project and not on the individual pieces of equipment. To do away with this continued uncertainty for equipment producers, it was considered desirable to suspend application of the Law of National Similars The main element in the protection system would thus be the import tariff. In order for these measures to play their proper role, however, it would be necessary to review the current level of these duties. As noted in Chapter II, 1/ tariff rates have varied widely (from 7% to 205%) although most fall in the 35-55% range. The import-weighted average for capital goods categories is about 45%, but this measure understates the degree of protection since it does not take into account those items whose tariffs are so high as to amount to a virtual restriction on imports. There appears to be little basis for the current pattern of rates, many of which were set in the early 1950s as part of the import-substitution programs to encourage the establishment of a number of subsectors of relatively more advanced technologies, in particular to attract foreign investment Government officials have indicated that a general tariff reform is now under consideration, particularly in view of the measures which have already been taken to achieve a real devaluation and to cut back export subsidies. In the context of such a reform, it would be desirable to reduce substantially the tariff levels for most categories of capital goods, given the fact that generally efficient, competitive levels of production have been achieved in the subsector. Moderate levels could be retained for a carefully selected group of items which represent more advanced technologies where time required for the "learning" process would be critical. I/ See, in particular, table 2.5.