I am grateful to the U.S. Departments of State and Defense for their support of this research. Harry Zubkoff, editor-in-chief, Department of Defense News Clipping and Analysis Service, gave generously of his time and assistance. I am also indebted to Robert Jervis, for his comments and suggestions on an earlier version of this article, to IO's reviewers, and to Christine Evans for her conscientious research assistance. The opinions expressed herein are those of the author and do not represent those of the U.S. government or her colleagues. This article is based on a data collection compiled by the author from a wide variety of open sources reporting on Third World production of major weapon systems between 1950 and 1980. Portions of this article are reprinted by permission of the publisher, from Arms Production in Developing Countries, by James E. Katz (Lexington, Mass.: Lexington Books, D.C. Heath and Company, Copyright 1984, D.C. Heath and Company).Google Scholar

1. The term Third World, used interchangeably with less developed countries (LDCs), refers to the regions of Africa, the Middle East, Latin America, and Asia (excluding Japan). This study uses the Overseas Development Council's definition of countries “with 1976 percapita incomes of less than $2000 and/or PQLI [Physical Quality of Life Index] Achievements of less than 90” (see ODC, Agenda 1977, p. 152). However, data on the defense industries of Greece, Spain, Turkey, and Yugoslavia, countries falling within that definition, were not included.

2. Some of the participants in the debate are Brown, Seyom, New Forces in World Politics (Washington, D.C: Brookings, 1974)Google Scholar; Hoffmann, Stanley, Primacy or World Order? (New York: McGraw-Hill, 1979)Google Scholar; Keohane, Robert and Nye, Joseph, Power and Interdependence (Boston: Little, Brown, 1977)Google Scholar; Petras, James F., “The Myth of the Decline of Capitalism,” Telos no. 28 (Summer 1976), pp. 181–87CrossRefGoogle Scholar; Ravenal, Earl, Never Again: Learning from America's Foreign Policy Failures (Philadelphia: Temple University Press, 1978)Google Scholar; Rosecrance, Richard, ed., America as an Ordinary Country: U.S. Foreign Policy and the Future (Ithaca: Cornell University Press, 1976)Google Scholar; Rosenau, James N., “Muddling, Meddling, and Modelling: Alternative Approaches to the Study of World Politics in an Era of Rapid Change,” Millenium 8 (Autumn 1979), pp. 130–44CrossRefGoogle Scholar; Russett, Bruce, “The American Retreat from World Power,” Political Science Quarterly 90, 1 (1975)CrossRefGoogle Scholar; Waltz, Kenneth, Theory of International Relations (Reading, Mass.: Addison-Wesley, 1979).Google Scholar

3. Gilpin, Robert, War and Change in World Politics (Cambridge: Cambridge University Press, 1981), p. 180.CrossRefGoogle Scholar

4. Miller, Steven E., “Arms and the Third World: Indigenous Weapons Production,” PSIS Occasional Paper no. 3 (University of Geneva Programme for Strategic and International Security Studies, December 1980), p. 25.Google Scholar Miller believes that although indigenous Third World arms production to date has had little impact on the global distribution of power, an expansion of these capabilities could bring about such a redistribution in the future.

5. The term “production” is used here in the generic sense, to include all phases of the maintenance and/or manufacture of defense items. This usage conforms to the U.S. Department of Defense definition of coproduction, which encompasses any program that “enables an eligible foreign government, international organization or designated commercial producer to acquire the ‘know-how’ to manufacture or assemble, repair, maintain and operate, in whole or in part, a specific weapon, communication or support system, or an individual military item” (Department of Defense Directive 2000.9 [ASD-I and L], International Co-production Projects and Agreements between the United States and Other Countries or International Organizations, 23 January 1974). I analyze the production and trade of major weapon systems only. In conformity with the Stockholm International Peace Research Institute (SIPRI) definition of “major weapons,” I include aircraft, missiles, armored vehicles, and ships. Small arms, artillery, and ammunition are not included.

6. According to one ACDA official, the large percentage increase in 1972 of Third World exports to 8.84% was due to an estimated $700 million delivery of arms from China to North Vietnam (interview, April 1980).

7. The term “third-country transfer” refers to the sale or gift of arms from Country A to Country B. Country B subsequently transfers those arms to a third country, Country C.

8. Because it is a member of NATO, Turkey is not included in the category “Third World” even though, according to most socioeconomic indicators, it is similar to many of the countries so categorized. Turkey is now becoming a more active arms producer. In 1981 it ratified an agreement with the United States that stipulated terms for the modernization of the moribund Turkish defense industry (U.S. House Subcommittee on Europe and the Middle East, United States-Turkey Defense and Economic Cooperation Agreement, 1980, 96th Cong., 2d sess., 7 May 1980. See also “Bilateral Agreement Bolsters U.S. Basing Plan,” Defense Electronics, April 1981, pp. 45–46.

9. China, devastated by revolution, began producing naval craft during this period but did not fully resurrect its arms industries until the late 1950s and early 1960s.

10. Peru, for example, which previously produced only ships, plans to establish an indigenous aircraft industry. With the assistance of Italy's Aermacchi, an assembly line will be set up for the MB-339 trainer and light strike aircraft (Milavnews, August 1981 and October 1981).

11. Generally, the newest defense items, incorporating the latest technological inventions, determine the relative level of sophistication of other military equipment. The meaning of “vintage,” or “intermediate,” or, for that matter, other qualitative descriptors such as “advanced,” “lead-edge,” or “critical,” is therefore constantly changing. However, to the consternation of methodological purists, the age of a weapons platform does not always reflect its relative technological sophistication. This is particularly true for “vintage” and “intermediate” defense items. Third World countries are constantly modernizing older weapon systems with new components, further smudging the already gray lines between categories. Israel's upgrading manufacturing capabilities, for example, are considerable. During a tour of Israel's tank depot, the author saw Centurion tanks refurbished with new assemblies and a forty-year old armored personnel carrier modified with half-tracks and rebuilt to modern specifications. Some effort has been made to define more exactly the terms “lead-edge” and “critical” (often used interchangeably). Maurice J. Mountain regards “critical technology” as those weapons in which the United States and its allies now have, or are likely to have, a margin of technological superiority over the USSR and the Warsaw Pact countries (“Technology Exports and National Security,” Foreign Policy no. 32 [Fall 1978], p. 96). Several U.S. agencies have cooperatively produced a list of military critical technologies in an attempt to change the way American exports are screened. Although the complete list of technologies is currently classified, it at least establishes a basis for policymakers' comparisons of defense items (Aviation Week and Space Technology, 13 October 1980, p. 23).

12. Herbert Wulf et al., Transnational Transfer of Arms Production Technology, A report written as a contribution to the United Nations study on the relationship between disarmament and development (University of Hamburg Institut fur Friedensforschung und Sicherheitspolitik, Study Group on Armaments and Underdevelopment, n.d. [c. 1980]), p. 10.

13. My data break down the arms industries operating in the Third World between 1950 and 1980 by category of weapon produced. With the exception of China, which maintained a ship industry in 1950, the oldest LDC producers (Argentina, Brazil, and India) began with aircraft industries producing simple trainers. More recently, because new states starting arms industries are often relatively small in size and industrial infrastructure, even trainers have been beyond their need and capability, and small naval craft have become the most common major military-industrial product in the Third World. The discrepancy in the number of LDC producers in 1980, 26, and the number in the period 1975–80, 27, reflects Bangladesh's termination of patrol craft production in 1977.

14. Production under license uses imported data packages as its source of design and technical information.

15. According to one source, the proportion of local to foreign components in the Merkava is 58% but is expected to rise to 80% (David Krivine, “Chariots of Steel,” Jerusalem Post Magazine, 11 November 1981, p. 4).

16. The capabilities of defense industries are rarely described in open sources. A recent survey of Egypt's military industries, designed to encourage Western investment, offers a unique view of the type of defense items produced in Third World countries. See Robinson, Clarence A. Jr, “Factories Tool for Alpha Jet Program,” Aviation Week and Space Technology, 18 January 1982, pp. 61–66.Google Scholar

17. The seven are Argentina, Brazil, China, India, Israel, South Korea, and South Africa. Prior to the revolution Iran rebuilt U.S. M-47 tanks, but is not known to have done so since. Reports of North Korean production of armored personnel carriers, believed to be Russian T-55s, began appearing in various sources after 1980 and so are not included here.

18. Milavnews, July 1981, p. 1.

19. SIPRI, World Armaments and Disarmament: SIPRI Yearbook, 1981 (New York: Crane, Russak, 1980), p. 77.Google Scholar

20. Ibid., and Tuomi, Helena and Vayrynen, Raimo, Transnational Corporations, Armaments and Development: A Study of Transnational Military Production, International Transfer of Military Technology and Their Impact on Development (Tampere, Finland: Tampere Peace Research Institute, 1980), p. 91.Google Scholar

21. Argentina's IA-33 Pulqui II aircraft program, which produced six prototypes, and the HF-24 manufactured by India illustrate the extent of foreign involvement in unlicensed Third World manufacturing efforts. Both were designed by a German aircraft engineer, Dr. Kurt Tank, and were supported by German engineers and at least one German test pilot. Dr. Tank's design very closely resembled an earlier Focke-Wulf design. The materials used in the Pulqui II airframe and engine were largely European (SIPRI, The Arms Trade with the Third World [New York: Humanities Press, 1971], pp. 763–64, 745)Google Scholar.

22. Most of the arms produced in China, for example, are based on Soviet models that were either license-produced previously or replicated without benefit of license. Examples are (aircraft) F-6 (MiG-19), BT-5 (Yak 18), AN-2, F-7 (MiG-21), B-5 (developed from IL-28); (missiles) CSA-1 (SA-2), AT-3, CSSN-1 (SSN-2). Israel's M-5 Nesher aircraft was converted from France's Mirage III, which was embargoed during the 1967 war, production ended in 1975. The Kfir C-2 fighter was developed from the Mirage V by Israel Aircraft Industries (IAT). North Korea's fast attack craft (“Iwon”-class and “Ku Song”-class, which are similar to the Soviet “P-2” and “D-3”) are generally classified as “indigenous” in the literature.

23. Interview with an IAI engineer in Israel, 14 August 1982. See also “Israel to Build Jet Fighters to Ease Reliance on U.S.,” New York Times, 8 February 1982; “Grumman Receives Lavi Primary Subcontract,” Aviation Week and Space Technology, 6 June 1983, p. 19.

24. Interview with an Israeli Ministry of Defense official in the United States, May 1983.

25. South Korea is not included here because the missile program for the Nike-Hercules surface-to-air missile was not confirmed in 1980 as in production.

26. There are some exceptions. North Korea also produces a frigate, small submarines, and an armored vehicle. The Philippines, Pakistan, Indonesia, and Nigeria also assemble helicopters but with relatively little local content.

27. Genuine air-to-air attack radar systems have been standard equipment on Soviet and U.S. fighters for a quarter of a century. Even Northrop's F-5E, developed from the F-5A during the 1960s and coproduced by the Taiwanese is reported to carry more effective air-to-air ordnance. See James B. Linder and A. James Gregor, “The Communist Air Force in the ‘Punitive’ War against Vietnam,” Air University Review, September-October 1981, pp. 69 and 75.

28. Milavnews, January 1981, p. 4.

29. Recent reports suggest that severe financial and technical constraints have prevented the Chinese from making more advanced weapon systems (Linder and Gregory, “Communist Air Force,” p. 69). See also Dreyer, June Teufel, China's Military Power (New York: China Council of the Asia Society, August 1982).Google Scholar

30. Harkavy, Robert E., The Arms Trade and International Systems (Cambridge, Mass: Ballinger, 1975), pp. 188–91.Google Scholar An index of military production capability (technical capabilities) was used, based in part on Harkavy's “Dependence-Autarky Continuum” (ibid., pp. 184–87).

31. Some countries, such as Sri Lanka, have remained at a rudimentary level of production (level 5), almost totally dependent on foreign suppliers for all but the simplest technologies.

32. South Korea's and Taiwan's arms industries are largely dependent on licensed production and the importation of U.S. components; Brazil's arms industry assembles some advanced components, but is essentially dependent upon imported sophisticated subsystems and foreign technical assistance. See the description of a Brazilian-Italian joint venture agreement to build a new light-weight strike fighter in Milavnews, December 1982, p. 7.

33. Israel illustrates the overriding importance of scale for military production. Although Israel does not lack technical skills, the most expensive and complicated manufacturing tasks are generally uneconomical to produce in the relatively small numbers required by Israel's armed forces. Even though the technological capability exists (or could easily be developed), factors of scale have created Israel's technological dependency.

34. Wulf, , Transnational Transfer of Arms, pp. 39–40Google Scholar; Harkavy, , Arms Trade and International Systems, Table 6–2, pp. 188–91Google Scholar; and Peleg, Ilan, “Military Production in Third World Countries: A Political Study,” in McGowan, Patrick J. and Kegley, Charles W., eds., Threats, Weapons and Foreign Policy, Sage International Yearbook of Foreign Policy Studies vol. 5 (Beverly Hills, Calif: Sage, 1980), pp. 209–30.Google Scholar Harkavy includes developed countries.

35. Kuznets, Simon, Six Lectures on Economic Growth (New York: Free Press, 1959)Google Scholar, suggests that the economic structure of a small country will of necessity be less diversified than that of larger ones, and production more concentrated in fewer industrial sectors. Its natural resources are likely to be fewer and it can sustain the minimum scale of a plant, particularly for modern industries (e.g., automobiles, airplanes, heavy electrical generators) only at an economic loss or on the basis of exports, often in a changing and volatile market. Moreover, smaller resources, even if extensive in one sector, such as oil, mean the country's limited capital and manpower will be concentrated in one or a few sectors. “A large nation can divide its greater volume of resources among a greater number of sectors with potential comparative advantage” (p. 92). Cf. Geoffrey Pattie, M.P. and under-secretary of state for Defence for the Royal Air Force, discussing the British defense dilemma in the Thatcher government's first Defence White Paper. He argued the need for the U.K. to create scale in production: “Overseas sales are, I believe, an important element in any defence industrial strategy. Extra orders beyond those from the U.K. mean more work, benefit to the balance of payments, economies of scale and thus reduction in the unit cost to ourselves” (Pattie, , “Needed: A Defence-Industrial Strategy for the U.K,” Defence and Foreign Affairs no. 9 [1980], p. 8Google Scholar).

36. Countries with fewer than one million inhabitants were not included.

37. Harkavy, , Arms Trade and International Systems, p. 204.Google Scholar

38. South Africa contains only 8% of sub-Saharan Africa's population and maintains an army comprising 9% of the region's forces, but produces 26% of its goods and services (GNP).

39. Jackman, Robert, “Dependence on Foreign Investment and Economic Growth in the Third World,” World Politics 34 (January 1982), p. 192.CrossRefGoogle Scholar Hollis Chenery and Moises Syrquin also find that size is particularly important for development in those LDCs above the lowest income levels. They hypothesize that countries with larger incomes have relatively advanced economic infrastructures and can, therefore, take advantage of economies of scale (Chenery, and Syrquin, , Patterns of Development, 1950–1970 [New York: Oxford University Press, 1975], chap. 4)Google Scholar.

40. After the signing of the Camp David accords (1979) Arab funding was withdrawn from the Arab Industrial Organization, which supported many of the military industries in Egypt

41. Peleg, , “Military Production,” p. 219.Google Scholar

42. Brazil and Argentina were subject to U.S. restrictions in 1977; Israel to restrictions from France in 1967 and the U.S. in 1973 and 1982–83; India was embargoed by the U.S. from 1965 to 1975, and China by the USSR in 1962 and the United States since 1948; South Africa has been subject to a voluntary UN embargo since 1963 and a mandatory one since 1977. Several of the second-rank LDC producers have also experienced political restrictions on their arms deliveries, e.g., Taiwan in 1978 from the United States and the Democratic Republic of Korea in 1974 from the USSR.

43. Pakistan was embargoed by the United States between 1965 and 1974; Chile by the United States since 1974; and Guatemala between 1978 and 1983. Restrictions on Iraq were imposed by the Soviet Union between June 1963 and May 1964.

44. The oldest and largest of the Third World arms producers, China, India, and Brazil, are all regional leaders, ranking first in size of population, military, and GNP. Argentina, also among the oldest of the producers, ranks second in size of military and third in size of population and GNP among Latin American states. Its production capabilities are also markedly less than its larger competitors.

45. See footnote 67 below for an explanation of “offsets.” The relationship between arms purchases and arms production is examined in Neuman, Stephanie, “Arms Transfers, Indigenous Defense Production and Dependency,” in Amirsadeghi, Hossein, ed., The Security of the Persian Gulf (London: Croom-Helm, 1981), pp. 131–50.Google Scholar

46. As discussed above, political factors can create, at least for a time, both “over-achievers” and “under-achievers” within the ranks of LDC arms producers. Over the long run this may prove equally true for European arms industries. One might argue that the relative inability of Western European governments— for fear of the political consequences of lay-offs— to permit any sector of their arms industries to reduce capacity despite inadequate demand is an exercise in “over-achievement” that factors of scale will eventually normalize.

47. Comparative advantage attributes the costs and benefits of international trade to differences among countries in the relative opportunity costs (costs in terms of goods given up) of producing the same commodities. For example, if Country A must give up three units of good x for every unit of good y produced, and Country B must give up two units of good x for every unit of good y produced, both countries would benefit if Country B specialized in the production of y and Country A specialized in the production of x.

48. For a discussion of the factors of scale and arms production in the Third World see Neuman, Stephanie, “Third World Arms Production and the Global Arms Transfer System,” in Katz, James E., ed., Arms Production in Developing Countries (Lexington, Mass.: D. C. Heath, Lexington Books, 1984).Google Scholar

49. “The Western Automobile Industry at the Crossroads,” Transatlantic Perspective no. 4 (January 1981), p. 13.Google Scholar

50. The “product life cycle” model, developed in the United States, describes the stages through which a manufactured item passes in terms of the investment behavior of U.S industries. See Vernon, Raymond, Sovereignty at Bay: The Multinational Spread of U.S Enterprises (New York: Basic Books, 1971).Google Scholar

51. For examples of the debate see Department of Defense, An Analysis of Export Control of U.S. Technology—A DOD Perspective, A report of the Defense Science Board Task Force on Export of U.S. Technology (February 1976); American Enterprise Institute, Legislative Analysis: Proposals for Reform of Export Controls for Advanced Technology (Washington D.C.: August 1979)Google Scholar; a symposium on “Trade, Technology, and Leverage,” in Foreign Policy no. 32 (Fall 1978), pp. 63–106Google Scholar; Burt, Richard, “U.S. Seeks to Guard Technological Edge,” New York Times, 11 December 1977Google Scholar; “Offsets and Technology Transfer,” Aviation Week and Space Technology, 16 July 1982.Google Scholar

52. “Brazilian Export Policy Sparks Criticism,” Aviation Week and Space Technology, 21 June 1982, p. 60.Google Scholar In 1982, Fairchild Aircraft Corp. lost its bid for import duties against Brazil's Embraer Bandeirante commuter aircraft before the U.S. International Trade Commission. Fair-child charged that its Metro commuter aircraft sales had been injured by the Brazilian financing subsidy (9% interest rate) to its customers. The American company claimed that “its share of the 19-passenger commuter transport market dropped from 60 percent in 1978 to a current 30% while Embraer's share rose proportionately” (Aviation Week and Space Technology, 27 September 1982, p. 24)Google Scholar.

53. “E-Systems Protests Foreign Arms Deals,” New York Times, 13 May 1982, p. D–2.Google Scholar Tadiran, the Israeli company that won the army contract, underbidding E-systems by $9 million, is 42% owned by General Telephone and Electronics Corporation and 42% owned by Koor Industries Ltd., one of Israel's largest industrial companies.

54. Jack I. Hope as quoted in Judge, John F., “Exporting Technology: An Exercise in Ignorance,” Government Executive, July 1977, n. p.Google Scholar

55. If France and Britain are added, the share of the industrialized countries increases to 90% (Thorsson, Inga, “Study on Disarmament and Development,” Bulletin of the Atomic Scientists, June/July 1982, pp. 41–44)Google Scholar.

56. See “The Coming Space War,” Foreign Report, 16 September 1982, pp. 1–3Google Scholar; and “U.S. to Increase Military Funds for Space Uses,” New York Times, 29 September 1982, p. 1.Google Scholar In the sphere of space technology, in general, the United States and USSR lead. Military applications, space explorations, manned orbital missions— all of these are beyond other countries' capabilities. Western Europe's combined expenditures on space projects are less than 20% of the NASA budget: NASA's budget in 1981 amounted to $5.54 billion, Europe's to a little more than $1 billion (Hoagland, John H., “Western Europe as a Space Power,” Europe/America Letter, June 1982, pp. 15–30Google Scholar).

57. See former Air Force Secretary Hans Mark's speech in the Albuquerque Journal, 16 January 1982, and the views of Patrick Friel, a former Department of Defense official, in Defense Daily, 29 June 1982, p. 326, and 30 June 1982 p. 332. Other specialists disagree, contending that laser technologies are not cost-effective and that various countermeasures could easily defeat them. The most detailed analysis appears in an M.I.T. study by Kosta Tsipis and Michael Callahan, reported in a Wall Street Journal article, “Laser Arms in Space Termed Impractical and a Potential Trigger to Nuclear War,” 22 December 1980, p. 10.

58. A large literature describing these developments is available. Some examples are Perry, William and Roberts, Cynthia A., “Winning through Sophistication: How to Meet the Russian Challenge,” Technology Review, July 1982, pp. 27–35Google Scholar; Russell, David M., “PGMs Achieve Stand-off Range with Sophisticated Guidance Systems,” Defense Electronics, November 1982, pp. 94–98Google Scholar; “Killer Electronic Weaponry: Tipping the Balance in Military Power,” Business Week, 20 September 1982, pp. 74–77, 80, 84.Google Scholar

59. See “Killer Electronic Weaponry,” p. 76; and “Tomorrow's Infantry: More Lethal and Much Swifter,” Business Week, 18 October 1982, pp. 189–90, 194.Google Scholar

60. Schemmer, Benjamin F., “‘Electronic Cameras’ with Instantaneous Ground Read-out Now Make Real-time, Precision Tactical Targeting Operationally Feasible,” Armed Forces Journal International, November 1982, p. 70.Google Scholar

61. Ibid., and “Tomorrow's Infantry…”

62. “Killer Electronic Weaponry,” p. 74.

63. Warwick, Graham, “Electronics on the Offensive,” Flight International, 23 October 1982, p. 1221.Google Scholar

64. For example, it takes an American company 3.5 to 5 years to build a new plant of any size and get it into production; finding sufficiently trained technicians has become equally time-consuming. Technology has become so advanced that unskilled labor cannot be quickly retrained as was the practice in the United States during World War II (Carlson, Robert J., “Strengthening the Defense Industrial Base: An Aircraft Manufacturer's View” [Paper delivered before a conference of the Scientists' Institute of Public Information, Boston, Mass., the Harvard Center for Science and International Affairs, and the American Academy of Arts and Sciences, 10 May 1982, p. 3)Google Scholar.

65. Computers, for example, in a variety of shapes, sizes, and capabilities are being installed (“embedded”) in almost every ship, plane, and weapon system—old and new—in the Pentagon's inventory (“Rule Dispute Snarls Military Computers,” Baltimore Sun, 21 November 1982, p. 1). The West German airforce is also extending the service lives of its Phantoms by installing new weapon system computers in over 160 of its F-4Fs (Milavnews, January 1983, p. 9).

66. The Northrop Corporation developed the F-5G without government funds and, without specific orders, put it into production in expectation of a market for more than 1,000 intermediate-level aircraft (Schemmer, Benjamin, “Pressures Build for DOD to Buy and Help Sell Northop F-5G as Its Business Prospects Look Bleaker than Advertised,” Armed Forces Journal International, September 1982, pp. 88–104Google Scholar). The Mirage F-3NG was designed with a similar market in mind. France's Dassault-Breguet has developed the Mirage 3NG (upgraded from the Mirage 50) as an “advanced technology export competitor to the Northrop F-5G and Israel's (IAI) Kfir” (Lenorovitz, Jefirey M., “Mirage 3NG Starts Final Ground Checks,” Aviation Week and Space Technology, 2 November 1982, pp. 52–55Google Scholar). The F-5G is now called the F-20.

67. When procurement of a complex and costly weapon system from a foreign supplier is pending, an LDC government is confronted by one or more of the following interrelated problems: 1) the outflow of foreign currencies; 2) an increase in technical and logistic dependence; 3) the underutilization of domestic defense industrial facilities and skilled labor (where they exist). In order to lessen or “offset” these negative features, the procuring country will generally demand either compensation for all or part of the worth of the purchase in reciprocal exports, or it will insist on coproduction of part of the weapon system in-country. The Brazilian government, for example, requires 50% offsets on all aviation imports. Some of these offsets have been achieved through coproduction agreements, but others have been realized through nonmilitary exports, e.g., coffee, sugar, soybeans, and other agricultural products (“Latin American Leaders Offer Challenge to U.S. Industry,” Aviation Week and Space Technology, 21 June 1982, pp. 60–61)Google Scholar.

68. For a description of increasing LDC exports to industrialized countries see Neuman, “Third World Military Production.”

69. See ibid.

70. Barter agreements between European arms suppliers and Third World oil producers, although seldom mentioned in public, are becoming more common as “offsets.” For example, between 1978 and 1980 Saudi Arabia's orders represented 20% of total French arms exports. Subsequently, one-half of France's oil imports came from Saudi Arabia. Similar arrangements were made with the Iraqis (Faligot, Roger, “Arms for the Middle East: Mitterrand Puts the Profit Motive First,” Middle East, October 1982, pp. 23–26)Google Scholar. See also fn. 74.

71. Thailand, for example, will assemble 47 of the 53 Fantrainer 60 aircraft it has purchased from West Germany as part of a coproduction agreement that also allows for some domestic manufacture. This will include wings, whose construction will be changed from the present composites to an all-metal structure more suited to the capabilities of Thailand's emergent aerospace industry (Milavnews, October 1982, p. 21).

72. For a description of rising intra-Third World trade, see Neuman, “Third World Arms Production.”

73. Examples include Iraq, which during 1982 received supplies from Spain and Switzerland, in addition to France and the United States. Spain has also been linked, unofficially, with the training of Iraqi pilots and ground crews. Switzerland is reported to have overhauled helicopters for the Iraqis. Argentina has purchased 27 Kurassier light tanks from Austria. The Benin government took delivery in 1982 of a transport plane from Holland. Bolivia plans to purchase all 52 of Belgium's surplus F-104 Starfighters (Milavnews, January 1983, pp. 1–24).

74. It is reported that in return for oil shipments and hard currency, North Korea was the source of about 40% of the $2 billion worth of weapons, ammunition, and equipment acquired by Iran in 1982 (Milavnews, January 1983, p. 20). Brazil, too, has been negotiating barter arms sales. The fact that many major weapons sales have been made to oil-exporting nations (Libya, Iraq, Qatar, Abu Dhabi, Angola, Nigeria, Indonesia) reveals Brazil's compelling need to export in order to offset its oil purchases. Brazil imports three-quarters of its oil (Williams, John Hoyt, “Brazil: Giant of the Southern Hemisphere,” National Defense, November 1982, p. 20)Google Scholar.

75. In addition to Iran, there are many examples of diversified sources of supply to Third World states in 1983. These include Iraq, which will continue to receive military equipment from France, the United States, Switzerland, and possibly West Germany; Argentina, committed to the purchase of over $1 billion in arms replacement from Israel, Peru, Brazil, France, Austria, West Germany, and possibly the Soviet Union; Egypt, which will receive defense items from China, France, the United States, and Britain; and India, which has contracted to purchase weapons from France, Britain, West Germany, and the United States (Milavnews, January 1983, pp. 1–24).

76. It is probable that China will one day join the ranks of the world's sophisticated arms producers, since its size provides the necessary factors of scale to make such production possible and profitable. Political factors and economic constraints, however, make such an eventuality a very long-term possibility.

77. Interview with U.S. State Department official, January 1983.

78. The Soviet Union, for instance, has offered to train personnel and set up fully operative plants for the production of the latest-model MiGs in India (“Soviet Union Says It's Eager to Help India Produce Arms,” New York Times, 20 March 1982). A French industrial group has been discussing with the Argentinian government the possibility of building a factory in Buenos Aires for the production of French armored vehicles under license (Milavnews, September 1982, p. 2). Within the Third World, more advanced producers are providing similar services to less advanced producers. The Chinese have already supplied Pakistan with a complete factory for overhauling F-6 aircraft (“Chinese Assisting Pakistani Industry,” Aviation Week and Space Technology, 30 March 1981).

79. The typical cost for a jet aircraft, for example, generally breaks down into one-third for the airframe, one-third for the engine, and one-third for the avionics. Most developing countries are dependent on major suppliers for the latter two items, or at least two-thirds of the cost of the plane. Brazil serves as a good illustration. Although Brazil has stopped buying fly-away military jet aircraft from the United States (because of domestic production), it continues to purchase components and transportation and commercial aircraft. U.S. aeronautical exports to Brazil totalled $192.8 million in 1981 (“Brazilian Export Policy Sparks Criticism,” pp. 60–61).