¹ For a detailed account of foreign investment, see Shapiro, Helen, Engines of Growth: The State and Transnational Auto Companies in Brazil (Cambridge and New YorkCrossRefGoogle Scholar, forthcoming).

² ANFAVEA, Indústria automobilística brasileira (Sāo PauloGoogle Scholar, various years); and IBGE, Anuário Estatístico do Brasil (various years).

³ For more on this early period, see Almeida, José, A implantaçāo da indústria automobilística no Brasil (Rio de Janeiro, 1972Google Scholar); and Gattás, Ramiz, A indústria automobilística e a segunda revoluçāo industrial no Brasil: origens e perspectivas (Sāo Paulo, 1981Google Scholar).

⁴ Guimarāes, Eduardo Augusto, “Industry, Market Structure, and the Growth of the Firm in the Brazilian Economy” (Ph.D. diss., University of London, 1980Google Scholar).

⁵ Maxcy, George and Silberston, Aubrey, The Motor Industry (London, 1959), 16–18Google Scholar. The aggregate import figures by country of origin mask firm nationality and hence exaggerate the inroads made by European-owned companies. Ford, unable to source its assembly plants with imports from the United States, shifted its export base to its European subsidiaries. The large increase in British imports in the late 1940s came in large part from the Ford subsidiary at Dagenham, which began exporting to Brazil in 1946. Later, French and Canadian subsidiaries were also used, depending on foreign exchange availability. Ford also engaged in barter deals, exporting Brazilian raw materials for imported vehicles (Manager's Monthly Letters, Ford Motor Company-Brazil, 1 March 1946, 1 March 1948, and 28 Jan. 1953, ID CF Storage B-38 119A, Wilkins Personal File). It should be noted that some European currencies became fully convertible only in 1959. For historical treatments of the industry, see also Bloomfield, Gerald, The World Automotive Industry (Newton Abbot, 1978Google Scholar); and Roos, Daniel and Altshuler, Alan, co-directors, The Future of the Automobile (Cambridge, Mass., 1984Google Scholar).

⁶ Maxcy and Silberston, The Motor Industry, 144n3.

⁷ Guimarāes, Eduardo Augusto, Acumulaçāo e crescimento da firma (Rio de Janeiro, 1987), 136Google Scholar.

⁸ Jenkins, Rhys Owen, Transnational Corporations and the Latin American Automobile Industry (Pittsburgh, Pa., 1987), 29CrossRefGoogle Scholar.

⁹ Tariffs on passenger cars had fallen slightly after the war, but still averaged over 30 percent in 1960 except in Germany, where they were 13–16 percent. By 1968, tariffs within the EEC were eliminated, and they had fallen to 17.6 percent in the United Kingdom. Vehicle trade among the major auto-producing countries in Europe was 6 percent of production in 1950, 7 percent in 1960, and 12.3 percent in 1970. Trade within Western Europe as a whole was 18.1, 18.8, and 25.8 percent of production for those same years (Roos and Altshuler, co-directors, The Future of the Automobile, 17 and 22). European companies had not penetrated each others' markets prior to this time, but the American firms—particularly Ford and GM—had done so through direct investment.

¹⁰ Ibid., 25. Europe's import share declined when the Big Three introduced their own compact models in the early 1960s, but it bounced back to 10 percent by 1970.

¹¹ Guimarāes, “Industry, Market Structure, and the Growth of the Firm in the Brazilian Economy,” 173. The next-largest countries—Argentina, Mexico, and Venezuela—each had only 200,000 vehicles in circulation.

¹² The domestic parts sector prospered during the war in response to the demand for replacement parts. Facing extinction after the war with the resumption of imports, makers lobbied vigorously for protection. For the industrial promoters, the fact that the sector's existence could be used as proof of Advisory 288's viability was more important than its lobbying efforts, particularly given its small size. They understood that the tiny sector would have short-run difficulties supplying even the items listed, but it nevertheless provided a springboard from which to begin. This characterization of the parts sector as still dependent on the state's tutelage and initiative, despite its growing organizational strength and participation in state decision making, concurs with that of Martins, Luciano, Pouvoir et developpement économique: formation et évolution des structures politiques au brésil (Paris, 1976Google Scholar). For an alternative view of the role of private capital, see Antonieta, M.Leopoldi, P., “Industrial Associations and Politics in Contemporary Brazil” (Ph.D. diss., Oxford University, 1984Google Scholar).

¹³ Willys-Overland had been a successful jeep and small truck producer in the United States and merged with Kaiser-Frazer in 1953 after establishing an assembly plant in Brazil.

¹⁴ Memo from G. K. Howard to Appropriations Committee, Ford Motor Company, 3 Feb. 1950, Ace. 106, box 26, Wilkins Personal File.

¹⁵ Mira Wilkins interview with Humberto Monteiro, 8 Nov. 1961, Wilkins Personal File. According to Branco Ribeiro, this phrase was often repeated by Walter McKee, Ford's vice-president for Latin America, as well. Ford-Brazil had to convince Dearborn that there was no alternative to manufacturing, which Dearborn thought was not financially viable. Interview with Ribeiro, 4 Aug. 1988, Sāo Paulo.

¹⁶ Executive Communication to Henry Ford II et al. from Arthur J. Wieland, vicepresident, Ford International, 23 Jan. 1953, Ace. AR-67-6, box 2, Ford Industrial Archives, Redford, Michigan.

¹⁷ Arthur J. Wieland, vice-president, Ford International, Executive Communication to Henry Ford II et al., 23 Jan. 1953; and Executive Communication to I. A. Duffy, 26 Jan. 1953, Ace. AR-67-6, box 2, Ford Industrial Archives. Ford International moved from New York to Dearborn in 1956.

¹⁸ Interview with Lúcio Meira, April 1985, Rio de Janeiro.

¹⁹ Steven Tolliday, “Rethinking the German Miracle: Volkswagen in Prosperity and Crisis, 1939–1992,” paper presented at the Business History Seminar, Harvard Business School, 4 Nov. 1991.

²⁰ VWD Frankfurt a. Main, nr. 171, “Director Nordhoff Concerning the VW Factory in Brazil,” 27 July 1953, Volkswagen Archives, Wolfsburg. As of July 1954, the Aufsichtsrat had approved DM5 million for only assembly and machining equipment in Brazil. (Letter from the head of the Aufsichtsrat to VW, 4 July 1954.) According to a report on the minutes of an Aufsichtsrat meeting on 27 Aug. 1954, the board clarified that the payment to VW do Brasil was for the sole objective of starting an assembly plant, not a manufacturing facility. Nordhoff informed Aranha of the supervisory board's decision in a letter dated 30 Aug. 1954, adding that building a complete factory would be put on hold to await further developments in Brazil. On convincing the Brazilians that the project had not been canceled, letter from Heinrich Nordhoff to F. W. Schultz-Wenk, 30 Oct. 1954, Volkswagen Archives.

²¹ Personal correspondence, F. W. Schultz-Wenk to Heinrich Nordhoff, 27 May 1954; official correspondence, Heinrich Nordhoff to F. W. Schultz-Wenk, 22 June 1954; letter from Schultz-Wenk to Nordhoff, 19 Nov. 1954, Volkswagen Archives.

²² Letter from O. E. de Souza Aranha to Heinrich Nordhoff 27 Sept. 1954; letter from F. W. Schultz-Wenk to Nordhoff, 29 Nov. 1954, relating Aranha's position; letter from Aranha to directors of VW-Wolfsburg, 26 April 1955, Volkswagen Archives.

²³ Letters from Nordhoff to Aranha, 12 Oct. and 13 Dec. 1954. In a 24 May 1955 letter to Aranha, Nordhoff explained that Wolfsburg did not share Aranha's optimism about Brazil, that they planned to wait until the next elections before moving ahead, and that future manufacturing plans would be contingent on how things went with the assembly plant. In a 26 May 1955 letter to Schultz-Wenk, Nordhoff repeated that, in contrast to Aranha, people outside Brazil did not judge the situation to be very good and that VW would begin with an assembly plant, which could later be transformed into a car body shop if necessary. Volkswagen Archives.

²⁴ Letter from Nordhoff to Schultz-Wenk, 20 April 1956, Volkswagen Archives.

²⁵ Ibid., 30 April 1956, Volkswagen Archives.

²⁶ Lúcio Meira, as quoted in Franco, Wellington Moreira, “A nacionalizaçāo de veículos no Brasil” (MA Thesis, University of Sāo PauloGoogle Scholar, n.d), Appendix No. 1, “Sinopse das Atas de Reuniāo da Subcomissāo de Jeeps, Tratores, Caminhōes e Automòveis da Comissāo de Desenvolvimento Industrial, May 14–15, 1952.” (Unless otherwise indicated, all translations from Portuguese are mine.) Despite some interest shown by Mercedes Benz and General Motors in the early 1950s in gradually beginning domestic manufacture, no major investments had come through. In an Executive Communication, Arthur J. Wieland suggested that GM retreated in response to Vargas's 1952 restrictions on profit remittances. Executive Communication to Henry Ford II et al. from Arthur J. Wieland, vice-president, Ford International, 23 Jan. 1953.

²⁷ Building an industry on the basis of a “national champion” was not considered a viable alternative to foreign direct investment. The Brazilian private sector was considered too weak to satisfy the capital and technological requirements of automobile manufacturing. Auto parts producers had made it clear to Vargas's subcommission that they were unable to support such an undertaking, even if the government were to provide the original financing. The lack of a strong capital market in Brazil made it impossible to tap financial sources outside the sector. Furthermore, capital goods would have to be imported and the technology licensed, if possible, both of which would necessitate large foreign exchange expenditures. State ownership was also ruled out as an alternative strategy, partly because the Brazilian state's fiscal capacity was extremely limited. More important, the role of the state under Kubitschek was limited to the coordination and subsidization of private investment. State enterprise was restricted to infrastructure and some basic industries. The National Motor Factory (FNM), established under the Lend-Lease program to produce airplane motors for the U.S. war effort, was controlled by the military and produced trucks under license after the war. Its plant was obsolete, and even when Vargas's subcommission was still discussing state involvement as an option, the firm was never considered a serious contender.

²⁸ Decree No. 39.412, 16 June 1956, establishing the overall plan and GEIA; Decree No. 39.568, 12 July 1956, for trucks; Decree No. 39.569, 12 July 1956, for jeeps; Decree No. 39.676, 30 July 1956, for utility vehicles; and Decree No. 41.018, 26 Feb. 1957, for passenger cars.

²⁹ In fact, a similar institution was approved by President Vargas in June 1954, only to be aborted after his suicide in August of that year.

³⁰ These included the minister of transportation and public works, who acted as president, the directors of the Carteira de Câmbio (“Foreign Exchange Bureau”) and Foreign Trade Bureau (CACEX) of the Bank of Brazil, the executive director of the Monetary Authority (SUMOC), and the director-superintendent of the National Bank for Economic Development (BNDE). Representatives of the following institutions were included subsequently: the War Ministry in mid-1957; the newly formed Council on Tariff Policy in August 1957; and the Agricultural Ministry in December 1959. It is noteworthy that representatives of the auto industry itself, though frequently consulted, were not included.

³¹ Although Instruction 113 was issued before Kubitschek took office, it came to be a widely used policy instrument only during his administration. Its benefits were not exclusive to the automotive firms, but the automotive industry as a whole (including the parts sector) was responsible for almost half of all investment entering Brazil under Instruction 113 during the Kubitschek administration: US$200.7 million of the US$419 million total. Instruction 113 discriminated against Brazilian investors, because they were unlikely to be able to take advantage of it. Foreign investors had access to foreign currency and could import equipment directly without relying on Brazil's foreign exchange holdings; Brazilians required foreign exchange cover to import equipment. The discriminatory treatment was justified by the scarcity of foreign exchange. Brazilian investors were compensated by the special exchange rates available for financing imports.

³² With the exception of separate exchange quotas for imported auto parts, none of these incentives was issued exclusively for or only available to the auto industry. The decrees made the industry eligible for them. For a fuller discussion of the financial subsidies, see Shapiro, Engines of Growth.

³³ Interview with Lúcio Meira, Minister of Transportation and Public Works and president of GEIA, April 1985, Rio de Janeiro. Sydney Latini, a secretary of GEIA, also used this term: “We have no time to lose. We must reach the target in the fixed time period. We must get beyond the ‘point of no returning.’ Afterwards, we will have the rest of our lives to discuss.” Latini, Sydney, SUMA Automobilística (Rio de Janeiro, 1984), 13Google Scholar.

³⁴ Brazilian policymakers wanted to take advantage of the country's being a Gerschenkronian “latecomer” and move immediately into capital-intensive, high technology sectors like auto. Rather than repeat the developmental stages of the more industrialized countries, they hoped to assimilate their technological legacy through foreign direct investment. The industry would then impel technological improvements in steel and other related sectors. Brazil's automotive program was therefore qualitatively different from those used in Europe as early as the 1920s and 1930s to protect and promote national industries in the face of U.S. competition. Although these programs all set up protectionist barriers to imports, Europe—which had given birth to automobile production—was starting from a very different base. Brazil was reversing the sequence of industrialization by using the auto sector to generate backward linkages, rather than to build on existing ones.

³⁵ Brazil was not unique in its concern with speed. Mexican policymakers also attempted to install the industry during the course of one presidential term, or the sexenio. See Bennett, Douglas C. and Sharpe, Kenneth E., Transnational Corporations versus the State: The Political Economy of the Mexican Auto Industry (Princeton, N.J., 1985), 128–29Google Scholar.

³⁶ For example, Ford's behavior was markedly different in Argentina and Mexico, where it took the lead in beginning domestic manufacture (Mira Wilkins interview with J. Sundelson, 11 May 1960, Wilkins Personal File). See also Bennett, Douglas C. and Sharpe, Kenneth E., “The World Automobile Industry and its Implications,” in Profits, Progress and Poverty, ed. Newfarmer, R. (Notre Dame, Ind., 1985), 214Google Scholar.

³⁷ Relevant estimates for optimum plant size vary. Bain estimated 300,000 (Bain, Joseph, Industrial Organization [New York, 1968], 284–87Google Scholar). Baranson showed production costs per unit leveling off at 120,000 for assembly, 240,000 for engines and other power train parts, and 600,000 for body stampings (Baranson, Jack, Automotive Industries in Developing Countries, World Bank Occasional Staff Papers, no. 8. [Washington, D.C., 1969]Google Scholar). Maxcy and Silberston estimated optimum capacity at 100,000 for assembly, 100,000 for casting, 400,000 for machining engines, and up to one million for body pressings, even though the rate of cost savings decreased as volumes grew. The economies of scale available to an individual firm differ from those for an individual plant. A firm can spread pre-production costs across plants and products, and a multiproduct mix allows some interchange of components. Measurement is difficult because of the nonhomogeneity of product and the variety of cost allocations used by firms. Maxcy and Silberston estimated that technical economies were exhausted at a million units, and that most of the substantive gains were made at the 400,000 mark (Maxcy and Silberston, The Motor Industry, 75–98). Optimum production for the industry as a whole would take into account the economies of scale available to suppliers as overall auto production increases. These estimates are all related to automobile production of the 1950s and 1960s. As a result of technological and organizational innovations, similar calculations for the industry in the 1990s would differ.

³⁸ For a similar attempt to identify the impact of firm-specific variables in host country-transnational firm bargaining at a later stage of the Brazilian auto industry, see Samuels, Barbara C. II, Managing Risk in Developing Countries (Princeton, N.J., 1990CrossRefGoogle Scholar).

³⁹ Minutes from GEIA meeting, 4 Dec. 1956, GEIA Archives, Conselho de Desenvolvimento, Ministério de Indústria e Comércio, Rio de Janeiro.

⁴⁰ Minutes from GEIA meeting, 26 Dec. 1956, GEIA Archives.

⁴¹ Internal memo from Eros Orosco, 8 Feb. 1957, attached to Ford truck project proposal, GEIA Resolution 16, file no. 217, GEIA Archives.

⁴² Minutes from GEIA meeting, 4 June 1957, GEIA Archives; and interview with Lúcio Meira, April 1985, Rio de Janeiro. Vemag was originally solely owned by Grupo Financeiro Nôvo Mundo, a banking and importing organization that had imported vehicles since 1945. In 1955, Vemag entere da licensing arrangement with Germany's Auto Union to produce their models. the next year, the company took in Auto Union, Frits Mueller Pressefabrik, and August Laepple as minority shareholders, but retained 87 percent of Vemag's equity capital; Dias, Vivianne Ventura, “The Motor Vehicle Industry in Brazil: A Case of Sectoral Planning” (MA Thesis, University of California, Berkeley, 1975), 43–44Google Scholar.

⁴³ Minutes from GEIA meeting, 6 Sept. 1957, GEIA Archives.

⁴⁴ Minutes from GEIA meeting, 28 Aug. 1957, GEIA Archives.

⁴⁵ Stefan Podgorski, Minutes from GEIA meeting, 4 Sept. 1957, GEIA Archives.

⁴⁶ Minutes from GEIA meeting, 28 Aug. 1957, GEIA Archives.

⁴⁷ Minutes from later GEIA meetings were not available.

⁴⁸ Minutes from GEIA meeting, 4 June 1957, GEIA Archives.

⁴⁹ Sydney Latini, Testimony before the Parliamentary Inquest Commission for the Verification of the Cost of the National Vehicle, 26 Oct. 1967, GEIA Archives.

⁵⁰ Japanese firms also stayed out of Brazil's passenger car market. At the time of Brazil's automotive decrees, Japan was producing fewer than 200,000 cars a year and exporting very few. See Cusumano, Michael, The Japanese Automobile Industry (Cambridge, Mass., 1985), 4CrossRefGoogle Scholar. Toyota submitted a small investment project to produce jeeps, which was accepted and implemented.

⁵¹ Memos attached to VW's auto project proposal, GEIA Resolution 63, file 39. GEIA Archives.

⁵² Minutes from GEIA meetings. The car plans of Mercedes and Borgward were ultimately accepted but never implemented. It is not clear whether Borgward's truck proposal was rejected or withdrawn.

⁵³ Interview with Lúcio Meira, April 1985, Rio de Janeiro.

⁵⁴ Minutes from GEIA meeting, 17 May 1957, GEIA Archives.

⁵⁵ Minutes from GEIA meeting, 4 June 1957, GEIA Archives.

⁵⁶ Interview with Sydney Latini, December 1984, Rio de Janeiro. Simca rented an old factory in Sāo Paulo to start assembly and to fulfill the domestic content guidelines. Although it did not meet the domestic content requirements of the program's second stage, Simca assumed that GEIA would still allow it to import spare parts. GEIA refused to release its foreign exchange allocation, and production was halted for six months. The firm ultimately had to scrounge for Brazilian-made parts and earned a reputation for poor quality. Simca sent a high-level executive from France to reactivate the plant. Despite its original promises, the firm never did relocate from Sāo Paulo to Minas Gerais. At one point, when it appeared that Simca would not be able to implement its project, Latini himself went to Turin to try to convince FIAT to substitute for Simca. Apparently, the families of the presidents of FIAT and Simca were related through marriage (Latini, Testimony before the Parliamentary Commission on Vehicle Cost, 26 Oct. 1967, GEIA Archives).

⁵⁷ Memo from D. B. Kitterman on Simca's Brazilian Car Proposal to Walter McKee, 2 May 1957, Ace. AR-67-6, box 2, Ford Industrial Archives.

⁵⁸ Project 997-A, introduced by Bilac Pinto, 58th session of Congress, 12 June 1956.

⁵⁹ Anais da Câmara dos Deputados (Congressional Record).

⁶⁰ In 1957, the average factory list price for a standard-equipped U.S. automobile was US$2,749. The average price for a four-door sedan was US$2,644, and for a two-door sedan, US$2,046. Ward's Automotive Report, 18 Feb. 1957.

⁶¹ Orosco, Eros, A indústria automobilística brasileira (Rio de Janeiro, 1961), 14Google Scholar.

⁶² Minutes from GEIA meeting, 14 June 1957, GEIA Archives.

⁶³ Minutes of Congress, 2 and 12 July 1957.

⁶⁴ The company was partially privatized in 1961, when 60 percent of its shares were sold to the public and the rest divided equally between state and federal governments. For more on VW's history, see Reich, Simon, The Fruits of Fascism (Ithaca, NY., 1990Google Scholar); for a more anecdotal account, see Nelson, Henry, Small Wonder (Boston, Mass., 1965Google Scholar).

⁶⁵ Letter from O. E. de Souza Aranha to O. W. Jensen, 21 April 1956, Volkswagen Archives.

⁶⁶ Internal communication from Heinrich Nordhoff to O. W. Jensen, 15 June 1956, Volkswagen Archives.

⁶⁷ Memorandum of the Planned Completion of VW Transporter in Brazil, submitted to the German Economic Ministry, August 1956, Volkswagen Archives.

⁶⁸ Letter from F. W. Schultz-Wenk to Heinrich Nordhoff, 8 Oct. 1956, Volkswagen Archives. (All translations from German courtesy of Laura Hastings.)

⁶⁹ Draft of letter to Kubitschek from Oeftering, attached to letter from Oeftering to Nordhoff, 28 Sept. 1956; telegram from Nordhoff to Oeftering, 10 Oct. 1956; letter from Schultz-Wenk to Nordhoff expressing surprise over his reaction to car production, 11 Oct. 1956; letter from Oeftering to Kubitschek, 13 Dec. 1956, all Volkswagen Archives.

⁷⁰ Letter from Nordhoff to Schultz-Wenk reiterating that cars were not under consideration, 22 Oct. 1956. In a letter to O. W. Jensen in Wolfsburg, Aranha wrote that VW do Brasil had requested BNDE financing for its car project, which in principle had been approved by management. He mentioned that he was told that Kubitschek personally advised the BNDE to help VW with its car financing. At the top of the letter is an annotation written by Nordhoff, in which he expresses shock that a financing project was being discussed on VW do Brasil letterhead for something (that is, cars) that had not yet been approved. Letter from O. E. de Souza Aranha to O. W. Jensen, 3 Dec. 1956, Volkswagen Archives.

⁷¹ Letter from Nordhoff to Schultz-Wenk, 12 April 1957.

⁷² VW had sent an earlier request for funding in its own name along with its van proposal in November 1956. Cruzeiro financing was provided for van construction; the car project was not pursued at that time.

⁷³ Bergsman, Joel, Brazil: Industrialization and Trade Policies (London, 1970), 45Google Scholar. The deutsche mark did not reach full convertibility until 1959, which probably fostered the ambiguity surrounding the exchange rate.

⁷⁴ O. E. de Souza Aranha no doubt helped raise expectations that the Brazilian government would be sympathetic to VW's requests. He wrote O. W. Jensen of Volkswagen-Wolfsburg on 16 April 1956 that one of his best friends, Lúcio Meira, has been appointed minister of transportation. Volkswagen Archives.

⁷⁵ Considerations external to the case's particular merits were not mentioned in the internal files to which I had access. GEDOC, Processos de Financiamento, BNDES Archives on Volkswagen Funding Requests, files R 641, F-171/56 and 172/56, Rio de Janeiro.

⁷⁶ Willis, Eliza, “The State as Banker: the Expansion of the Public Sector in Brazil” (Ph.D. diss., University of Texas at Austin, 1986), 256Google Scholar.

⁷⁷ Ibid., 289–93.

⁷⁸ Kubitschek dismissed Campos in June 1959 after breaking off negotiations with the IMF.

⁷⁹ John Goulden, general director of Ford, referred to Ford in this way in Testimony before the Parliamentary Commission on Vehicle Cost, 10 Oct. 1967, GEIA Archives.

⁸⁰ Mira Wilkins interview with Francisco Salles Cesar, 16 Nov. 1961, Wilkins Personal File.

⁸¹ In addition to profits earned locally in Brazil, from 1947 to 1955 the accounted profits earned by Ford-U.S. on exported units were estimated at US$18,541,000, and the economic profits at US$36,518,000; cited in background materials in the Presentation on the Brazil Truck Manufacturing Program made to the Executive Committee of Ford Motor Company, 30 Oct. 1956, Ace. AR-67-6, box 2, Ford Industrial Archives.

⁸² In a 3 Oct. 1956 letter to chairman Ernest R. Breech, A. J. Wieland, vice-president of Ford International wrote:

We wanted the (Executive) Committee to know the problem and the steps we were taking to try to keep our place in Brazil without following Government edicts which would have involved large sums of capital investment…. we must make some gestures to indicate to the Brazilian Government that we are interested in their problem and in their future growth. The plan that will be presented at the Executive Committee encompasses a probable investment of around $22,000,000, of which $14,000,000 would be in dollars and the balance in Brazilian currency. Obviously, we have no intention of recommending that we walk into Brazil with $14,000,000 as there are still a great many factors… that must be worked out.

Acc. AR-67-6, box 2, Ford Industrial Archives.

⁸³ Letter from J. S. Andrews, regional director for Latin America and the Orient, Ford International, to Tom Lilley, assistant general manager, Ford International, 30 March 1956. In anticipation of GEIA's automotive decrees, Andrews wrote, “In short, it would appear that the investment of dollar capital in Brazilian industry could well result in far more than an adequate return in cruzeiros, but that the outlook for remitting adequate return on the investment in dollars to the U.S. is limited and perhaps obscure.” Acc. AR-67-6, box 2, Ford Industrial Archives.

⁸⁴ In a 24 Oct. 1956 memo to T. O. Yntema, Tom Lilley discussed various ways to reduce Ford's direct investment, which included getting financing from the Export-Import Bank and the BNDE. “Our conversations in Brazil, however, have indicated that no such concessions will be granted and that to require them would be a way of rejecting the Brazilian plan.” Acc. AR-67-6, box 2, Ford Industrial Archives.

⁸⁵ Proposal for Truck Manufacturing Program, draft version, n.d., p. 25, Acc. AR-67-6, box 2, Ford Industrial Archives.

⁸⁶ Memo from Tom Lilley to T. O. Yntema, 24 Oct. 1956.

⁸⁷ Confidential letter to Humberto Monteiro, general manager, Ford Motor Company, Exports, Inc., Brazil, from J. S. Andrews, 31 Oct. 1956, Acc. AR-67-6, box 2, Ford Industrial Archives.

⁸⁸ Notes from 16 Jan. 1957 meeting at which Ford's International Division presented a revised truck proposal to Ford's Executive Committee; provided by J. C. Goulden to Mira Wilkins, Wilkins Personal File.

⁸⁹ Memo from Tom Lilley to J. S. Andrews on Goulart's visit, 14 May 1956, Acc. AR-67-6, box 2, Ford Industrial Archives.

⁹⁰ Letter from Humberto Monteiro, general manager, Ford Motor Company, Exports, Brazil, to J. S. Andrews, regional director, Latin America and the Orient, Ford International, 21 Dec. 1956, Acc. AR-67-6, box 2, Ford Industrial Archives.

⁹¹ Executive Communication from J. S. Bugas to Board of Directors, Subject: Brazil-Proposed Passenger Car Manufacturing and Facility Expansion Program, 14 Dec. 1960, Acc. AR-67-14, box 2, Ford Industrial Archives.

⁹² In 1950, Ford and GM had also decided not to compete with VW's Beetle in Germany but to compete with each other for the mid-size car market. Wilkins, Mira and Hill, Frank, American Business Abroad: Ford on Six Continents (Detroit, Mich., 1964), 391Google Scholar.

⁹³ Mira Wilkins interview with Walter McKee, 29 May 1962, Wilkins Personal File. According to McKee, the Chrysler board rejected the car program and fired all of the people involved.

⁹⁴ It is interesting to note that at this point, VW and other competitors assumed that Ford, and possibly GM, would enter the car market. In December 1958, VW's Schultz-Wenk reported to Wolfsburg that Mercedes Benz had given up on passenger cars for the moment, and that he thought this was because Mercedes feared competition from Ford and Chevrolet. He also wrote about the opening of Ford's factory and how Ford planned to use its V-8 engines for passenger cars as well as trucks. November Report on VW do Brasil from Schultz-Wenk to Nordhoff, 3 Dec. 1958, Volkswagen Archives.

⁹⁵ According to the proposal, reaching domestic content levels of 50 percent by March 1960 would be feasible without major new investments, because the car used many of the same components as Ford's small truck produced in Brazil. Reaching 95 percent domestic content by 1961 would be more complicated, because of the inability to use existing sheet metal dies. Executive Communication from Tom Lilley to members of the Executive Committee, Subject: Brazil—Proposed Manufacturing Program of 1959 Model Ford Car, 18 Feb. 1959, Acc. AR-75-63-430, box 29, Ford Industrial Archives.

⁹⁶ Ibid.

⁹⁷ Confidential telex from Monteiro to McKee, 26 March 1959, Acc. AR-75-63-430, box 29, Ford Industrial Archives.

⁹⁸ The economic stabilization program had been drafted by Finance Minister Lopes (who had served as Kubitschek's first director of the BNDE) and Roberto Campos, then director of the BNDE, in consultation with the IMF. Kubitschek broke off negotiations with the IMF in June 1959 and both Lopes and Campos were replaced soon after. For more on Brazil's break with the Fund, see Skidmore, Thomas E., Politics in Brazil, 1930–1964 (New York, 1967), 174–82Google Scholar; and Economic Commission on Latin America [Lessa, Carlos], “Fifteen Years of Economic Policy in Brazil,” Economic Bulletin for Latin America 9, 2 (1964Google Scholar).

⁹⁹ LT Interford Memo from Monteiro to McKee, 1 April 1959, Acc. AR-75-63-430, box 29, Ford Industrial Archives.

¹⁰⁰ The project was never implemented; Mercedes produced only trucks and buses in Brazil.

¹⁰¹ The proposal noted that 50 percent of the dollar investment could be utilized in production of a car other than the proposed model. After March 1960, Ford expected that imported equipment would become subject to import duties. Ford estimated that by using its US$4 million import license, it would save at least US$2 million on a future car program. On the initial production of 10,000 units, it was expected that Ford-U.S. would earn economic profits of US$1 million; profits to Ford of Brazil could be as high as US$10 million if the car had a retail price of US$8,500. Executive Communication from Tom Lilley to members of the Executive Committee, Subject: Brazil—Proposed Manufacturing Program of 1959 Model Ford Car, 1 April 1959.

¹⁰² Ford would use only about 60 percent and sell 40 percent to other ore users.

¹⁰³ Executive Communication from Tom Lilley to members of the Executive Committee, Subject: Brazil—Proposed Manufacturing Program of 1959 Model Ford Car, 1 April 1959.

¹⁰⁴ Memorandum from Walter McKee to J. S. Bugas et al., 1 May 1959, Acc. AR-65-71, box 27, PR archives, accession no. 742, Ford Industrial Archives.

¹⁰⁵ Telex from McKee to Monteiro, 7 May 1959, Acc. AR-75-63-430, box 29, Ford Industrial Archives.

¹⁰⁶ No record of later Ford projects was found among Brazilian archives and sources. Ford denied clearance to all archival materials pertaining to any aspect of its Brazilian operations after 1961. According to Ford archives until 1961, the Custom 300 was still being proposed in September 1959, but it had been replaced with the 1959 Ford Galaxie by August 1960. According to Brazilian records, the last proposal to be considered, and the most publicized and contentious, was for the 1959 Ford Fairlane.

¹⁰⁷ Possible alternatives included German class D cars, such as the 17M and its successor, the H or N XPD, with 1.5-liter engines, and the Ford Falcon, a Class E car with a 2.5-liter engine. Both were smaller than the Ford Custom 300, but neither would have competed directly with the smaller Class C cars produced by Volkswagen. Letter from Walter McKee to Humberto Monteiro, 15 July 1959, Acc. AR-75-63-430, box 29, Ford Industrial Archives.

¹⁰⁸ Letter from Andrew Masset, Ford Motor do Brasil, to Walter McKee, Ford International, 11 Sept. 1959, Acc. AR-75-63-430, box 29, Ford Industrial Archives.

¹⁰⁹ Ibid.

¹¹⁰ Executive Communication from J. S. Bugas to Board of Directors, Subject: Brazil—Proposed Passenger Car Manufacturing and Facility Expansion Program, 14 Dec. 1960, Acc. AR-67-14, box 2, Ford Industrial Archives.

¹¹¹ “There is no doubt that we will be facing a new crisis in our already difficult dealer situation as soon as competitors' cars appear in a volume on the market.” Letter from Andrew Masset to W. L. McKee, Ford International, 11 Sept. 1959.

¹¹² Memorandum on Brazil Car Manufacturing and Capacity Expansion Programs, 26 Oct. 1960, Ace. AR-67-14, box 2; and Executive Communication from J. S. Bugas to Board of Directors, Subject: Brazil—Proposed Passenger Car Manufacturing and Facility Expansion Program, 14 Dec. 1960.

¹¹³ Memorandum on Discussion of Factors, Problems, etc., in Connection with Brazil Car Manufacturing and Capacity Expansion Programs, 26 Oct. 1960, Acc. AR-6714, box 2, Ford Industrial Archives.

¹¹⁴ Letter from J. C. Goulden, finance manager, Ford Motor do Brasil, to W. L. McKee, regional director, Latin American Operations, Ford International, 17 Aug. 1960, discussing the 1959 Ford Galaxie proposal to be submitted to the Executive Committee. Ford assumed that it would be allowed to import US$3 million of equipment dutyfree and without exchange cover, so that the value of imported equipment would be carried on the books as capital investment. Ford predicted that the return on total added assets employed for car production would be 35 percent per year after taxes and that the additional fixed investment would be paid for within three years. In his 14 Dec. 1960 Executive Communication to the Board of Directors on Brazil—Proposed Passenger Car Manufacturing and Facility Expansion Program, J. S. Bugas provided somewhat higher investment figures for the Galaxie, but also assumed that all equipment would be imported duty-free and without exchange cover.

¹¹⁵ Interview with Lucas Lopes, May 1985, Rio de Janeiro.

¹¹⁶ Parecer do GEIA, 28 July 1961, published in its entirety in the Diário Oficial, 7 Aug. 1961.

¹¹⁷ Letter from Lúcio Meira to the Minister of Industry and Commerce, 31 Jan. 1963, GEIA Archives.

¹¹⁸ Ford and GM did not benefit from VW's loss of market share. Their respective shares of production in 1975 were 16 and 19 percent, and in 1980, 13 and 20 percent. FIAT entered the Brazilian car market in 1976, and by 1980 had 16 percent of total production, although a larger relative share went to export.

¹¹⁹ It could be argued that VW gained and maintained such a high market share because it had a more appropriate vehicle for Brazil. In countries such as Argentina and Mexico, however, no one firm dominated the domestic car market to the extent that VW did in Brazil.

¹²⁰ Guimarães suggests that a possible explanation for the late entry of Ford and GM was the shortage of managerial capacity in the 1950s, when the two companies were busy expanding in Europe. He also points out that the early and mid-1960s were profitable years for Ford and GM, and that the growth rate of the European market was slowing down; “Industry, Market Structure, and the Growth of the Firm in the Brazilian Economy,” 190.

¹²¹ According to Wilkins and Hill, an important reason for abandoning these markets was that local capital had to have a controlling interest in any project; Wilkins and Hill, American Business Abroad: Ford on Six Continents, 402.

¹²² Bennett and Sharpe also note that Ford was the first to indicate a willingness to begin domestic production in Argentina and Mexico. However, referring to the dynamics of defensive investment, which require that one firm take the lead while the others follow, they claim that Ford has traditionally made the first move in Latin America. In the case of Brazil, Ford was one of the most intransigent before the creation of GEIA; discussions had gone further with both GM and VW toward setting up domestic manufacture. With respect to cars in Brazil, Ford was not a leader but a follower. See Bennett and Sharpe, “The World Automobile Industry and its Implications,” 214.

¹²³ Baumgarten, Alfredo Luiz Jr, “Demanda de automóveis no Brasil,” Revista Brasileira de Economia 26 (1972): 203–97Google Scholar.

¹²⁴ Fishlow, Albert, “Some Reflections on Post-1964 Brazilian Economic Policy,” in Authoritarian Brazil, ed. Stepan, Alfred (New Haven, Conn., 1973), 104–5Google Scholar.

¹²⁵ See Guimarāes, “Industry, Market Structure, and the Growth of the Firm in the Brazilian Economy,” 195.

¹²⁶ Price rigidity in the face of falling demand is not unique to the Brazilian auto sector, but it has been demonstrated in other national auto industries and in oligopolized industries in general. Within Brazil, this pricing behavior was also characteristic of many other sectors, supporting the structuralists' arguments about the presence of market power and mark-up pricing.

¹²⁷ Vemag's president cited the firm's difficulty in raising capital as the primary reason for selling out. Vemag had tried selling stock but could not compete with other financial assets being offered. Another reason given was “that the authorities were not just stimulating but even pressuring the companies to accomplish mergers”; Ventura Dias, “The Motor Vehicle Industry in Brazil: A Case of Sectoral Planning,” 45.

¹²⁸ This process of concentration appeared to have the government's blessing. Secretary of planning Roberto Campos encouraged mergers. According to a Visāo interview, he viewed concentration as “an inevitable international tendency” that would reduce industrial costs and propagate competitive pricing. The denationalization of the motor vehicle industry occurred in other Latin American countries as well. Bennett and Sharpe claim that the speed of the Mexican program unintentionally may have been responsible. The TNCs were in a position to establish themselves more quickly since they had better access to resources and know-how. Moreover, they had ready suppliers for parts in the United States. This was particularly important in Mexico, which instituted only a 60 percent domestic content requirement. This was less critical under the 90 percent domestic content regime in Brazil, although foreign firms typically forced their major suppliers to accompany them; Bennett and Sharpe, Transnational Corporations versus the State, 128.

¹²⁹ The industry continued to expand throughout the 1970s, and by 1978 total vehicle production topped the “magic million” mark. An export promotion program was introduced in the early 1970s, and exports as a share of production increased, particularly in the 1980s when the domestic market stagnated.

¹³⁰ See Caves, Richard E., Multinational Enterprise and Economic Analysis (New York, 1982Google Scholar) for a general discussion, and Hymer, Stephen H., The International Operation of National Firms: A Study of Direct Foreign Investment (Cambridge, Mass., 1976Google Scholar) on the impetus behind direct foreign investment.

¹³¹ Knickerbocker, Frederick T., Oligopolistic Reaction and Multinational Enterprise (Boston, Mass., 1973Google Scholar).

¹³² The state is relatively absent in the work of Hymer and completely absent from Knickerbocker. Knickerbocker is less concerned with the reasons behind a first mover's entry than with why so many follow. However, in the Brazilian case, market closure and a brief window of investment opportunity affected all firms.

¹³³ In contrast to more market-oriented approaches to economic development, state-centric theorists argue that market forces alone will not lead to successful economic outcomes. Many have drawn their conclusions from the experience of Japan and East Asian newly industrialized countries such as South Korea. See Johnson, Chalmers, MITI and the Japanese Miracle (Palo Alto, Calif., 1982Google Scholar); Amsden, Alice, Asia's Next Giant: South Korea and Late Industrialization (New York, 1989Google Scholar); Haggard, Stephan, Pathways from the Periphery: The Politics of Growth in the Newly Industrializing Countries (Ithaca, N.Y., 1990Google Scholar); and Wade, Robert, Governing the Market: Economic Theory and the Role of Government in East Asian Industrialization (Princeton, N.J., 1990Google Scholar).

¹³⁴ See Geddes, Barbara, “Building State Autonomy in Brazil, 1930–1964,” Comparative Politics 22, 2 (Jan. 1990): 217–34CrossRefGoogle Scholar; and Lafer, Celso, “The Planning Process and the Political System in Brazil: A Study of Kubitschek's Target Plan, 1956–1961” (Ph.D. diss., Cornell University, 1970Google Scholar), on executive groups in general; see Moreira Franco, “A nacionalizaçāo de veículos no Brasil,” on GEIA on particular.

¹³⁵ Mexico's underlying economic conditions made it more difficult to require such high domestic-content levels. With a smaller population and per capita income level, Mexico did not possess the market potential of Brazil. the country's long border with the United States presented the country with different constraints. Unlike Brazil, Mexico's import-substitution process took place in the context of macroeconomic stability and of a fixed and unified exchange rate regime. Its auto program required that only 60 percent of a car's value (based on direct cost and including the engine) be manufactured in Mexico. By setting domestic content levels at 90–95 percent of a vehicle's weight, Brazil had ensured that body stamping would be done in the country. In general, the Mexicans were more concerned about relative costs and inflation, and there were large economies of scale in the stamping process. Also, with so much contact with the United States, Mexican consumers were exposed to the trends set by Detroit. A domestic content level of 60 percent allowed for body stampings to be imported and facilitated frequent model changes.

¹³⁶ See Evans, Peter, Rueschemeyer, Dietrich, and Skocpol, Theda, eds., Bringing the State Back In (New York, 1985CrossRefGoogle Scholar). For case studies on various international industries in Latin America, see Newfarmer, ed., Profits, Progress and Poverty.

¹³⁷ See Bennett and Sharpe, Transnational Corporations versus the State; Guimarāes, “Industry, Market Structure and the Growth of the Firm in the Brazilian Economy”; Jenkins, Transnational Corporations and the Latin American Automobile Industry; and Mericle, Kenneth S., “The Political Economy of the Brazilian Motor Vehicle Industry,” in The Political Economy of the Latin American Motor Vehicle Industry, ed. Kronish, Rich and Mericle, Kenneth S. (Cambridge, Mass., 1984Google Scholar).

¹³⁸ Bennett and Sharpe, “the World Automobile Industry and its Implications,” 222.

¹³⁹ Ibid., 208.

¹⁴⁰ Mericle, “The Political Economy of the Brazilian Motor Vehicle Industry,” 2 and 6.

¹⁴¹ See Gordon, Lincoln and Grommers, Engelbert L., U.S. Manufacturing Investment in Brazil: The Impact of Brazilian Government Policies, 1946–1960 (Cambridge, Mass., 1962Google Scholar); and John C. Goulden, general director of Ford do Brasil, Testimony before the Parliamentary Inquest Commission for the Verification of the Cost of the National Vehicle, 10 Oct. 1967, GEIA Archives.

¹⁴² Mira Wilkins interview with J. Sundelson, 11 May 1960, Wilkins Personal File.

¹⁴³ Many of the cited authors who have emphasized convergent interests were implicitly or explicitly responding to simpler bargaining models that assumed completely distinct interests and agendas on the part of states and foreign firms. These less sophisticated frameworks assumed that the state represented a vaguely defined national interest. To their credit, these scholars have identified competing interests in the peripheral state and society, some of which may not be distinct from those of the transnational firms. I would like to thank an anonymous reviewer for reminding me of this point.

¹⁴⁴ Guimarães, “Industry, Market Structure, and the Growth of the Firm in the Brazilian Economy,” 169–70. He cautions that an individual producer would have had difficulty getting an adequate supply of parts in the absence of a government policy for the sector as a whole.

¹⁴⁵ With the exception of BNDE financing, total subsidies on US$418.35 million of imported investment goods and complementary parts came to US$201.45 million, or 48 percent. See Shapiro, Engines of Growth.

¹⁴⁶ Orosco, A indústria automobilistica brasileira. 111.

¹⁴⁷ For example, GEIA withheld Simca's foreign exchange allotment when the firm failed to meet domestic content requirements, halting production for six months. Ford also saw its import quotas slashed; letter from Aranha to Volkswagen explaining the potential consequences of delaying or abandoning local engine production, 3 Feb. 1959, Volkswagen Archives.

¹⁴⁸ Since the focus of this paper has been the bargaining process in the implantation phase of the industry, other performance indicators have not been addressed. The evidence shows that the state's initial subsidies did not generate ongoing resource transfers to the sector. Rather, the state successfully taxed away an increasing share of the oligopolistic rents accruing to the industry. Brazil was also successful in generating the production externalities associated with the industry, which it would have sacrificed had it continued to import; the industry had relatively high linkage effects. Moreover, the industry's cost curve fell. In light truck production, Baranson found that the ex-factory costs net of taxes were 1.28 times those of the United States. In 1983, a World Bank study also concluded that Brazil's auto industry was a successful case of infant-industry development: World Bank, Brazil: Industrial Policies and Manufactured Exports. A World Bank Country Study (Washington, D. C., 1983), 116Google Scholar. These results were attributable to Brazil's sectoral policy, the nature of the market, the macroeconomic environment, and characteristics of the industry. They did not derive simply from Brazil's large-country advantage. Preliminary research indicates that neither Argentina nor Mexico achieved the same results with respect to rent redistribution or consolidation. For more on these issues, see Shapiro, Engines of Growth, and Shapiro, , “Rent-seeking or Rent-redistribution? Automobile Firms and the Brazilian State, 1956–1968”, in Developing Economies in Transition, ed. McCarthy, F. Desmond (Washington, D.C., 1990Google Scholar).