84 International Organization from an investment over the life of the project.They can be used at the micro level to analyze any particular investment agreement,or at the macro level to analyze the direction of changes in North-South (center-periphery)economic relations.They have received some important initial testing,but much more is needed. The balance of bargaining power framework provides some important contrasts to,and amplifications of,the dependencia perspective.First,it ignores the capitalism-socialism dichotomy that is central to many dependency writers.It fo- cuses on the relative negotiating strength of the individual actors rather than on the way in which property is owned within the economic system.It suggests that scarce resources and optimizing behavior would produce the same outcome whether the "investor''were a private capitalist or a socialist state agency:that a Soviet co- production agreement for bauxite in Guinea (e.g.,Dembe)or a Roumanian pet- rochemical project in Latin America should not differ from an American co- production agreement for oil in Indonesia or a German petrochemical project in Africa unless the Soviet or Roumanian governments wanted to use the state agencies involved as aid-giving institutions (i.e.,direct them to refrain from optimizing behavior in the use of their scarce resources).Second,it suggests that the increasing spread of multinational corporations of diverse national origins (American,Euro- pean,Japanese)strengthens,rather than weakens,the position of Third World countries because it gives them more alternatives to choose from.Third,it offers hope to the periphery that the direction of change is in their favor:if one projects a world in which the competition among foreign investors is increasing,in which the growth of the Third World is continuing (irrespective of a rising or falling gap in relation to the developed countries),and in which the breadth of host country industrial capacity and the negotiating skills of host country bureaucrats are expand- ing,for example,one would predict that host countries would have to pay less for the services of foreign investors in the future than they have had to in the past.3 There would still be large numbers of cases where the distribution of the benefits was highly tilted toward the foreign company in accordance with the variables specified earlier (e.g.,in the poorest countries,and/or at the beginning of most investment agreements,and/or in those industries where technology was rapidly changing).But one would expect the benefits from foreign investment in the aggre- gate to be rising and the cost of securing those benefits to be falling,over time.14 There are two severe limitations to this balance of power framework,however, to which dependencia analysis should draw the attention of non-dependentistas: first,the possible separation between the potential capability of host goverments to exercise greater negotiating strength in their relations with foreign investors anc The transfer of bargaining power into Third World hands would be retarded,however,to the extent that host countries compete among themselves to get foreign investment. 14Some companies,of course,may try to balance their increasing vulnerability in certain product lines with more profitable operations in other lines where they have tighter control over technology.Peter Evans hypothesizes,for example,that American petrochemical companies in Brazil are trading "unat- tractive"arrangements for the production of chlorine for a profitable "'kicker''in polyurethane foam (TDI).Testing the New Alliance:The Brazilian State,the Multinationals and the Launching of the Polo Petroquimico at Camacari,Bahia,"Brown University,xerox,February 1976
84 International Organization from an investment over the life of the project. They can be used at the micro level to analyze any particular investment agreement, or at the macro level to analyze the direction of changes in North-South (center-periphery) economic relations. They have received some important initial testing, but much more is needed. The balance of bargaining power framework provides some important contrasts to, and amplifications of, the dependencia perspective. First, it ignores the capitalism-socialism dichotomy that is central to many dependency writers. It focuses on the relative negotiating strength of the individual actors rather than on the way in which property is owned within the economic system. It suggests that scarce resources and optimizing behavior would produce the same outcome whether the "investor" were a private capitalist or a socialist state agency: that a Soviet coproduction agreement for bauxite in Guinea (e.g., Dembe) or a Roumanian petrochemical project in Latin America should not differ from an American coproduction agreement for oil in Indonesia or a German petrochemical project in Africa unless the Soviet or Roumanian governments wanted to use the state agencies involved as aid-giving institutions (i.e., direct them to refrain from optimizing behavior in the use of their scarce resources). Second, it suggests that the increasing spread of multinational corporations of diverse national origins (American, European, Japanese) strengthens, rather than weakens, the position of Third World countries because it gives them more alternatives to choose from. Third, it offers hope to the periphery that the direction of change is in their favor: if one projects a world in which the competition among foreign investors is increasing, in which the growth of the Third World is continuing (irrespective of a rising or falling gap in relation to the developed countries), and in which the breadth of host country industrial capacity and the negotiating skills of host country bureaucrats are expanding, for example, one would predict that host countries would have to pay less for the services of foreign investors in the future than they have had to in the past.13 There would still be large numbers of cases where the distribution of the benefits was highly tilted toward the foreign company in accordance with the variables specified earlier (e.g., in the poorest countries, and/or at the beginning of most investment agreements, and/or in those industries where technology was rapidly changing). But one would expect the benefits from foreign investment in the aggregate to be rising and the cost of securing those benefits to be falling, over time.14 There are two severe limitations to this balance of power framework, however, to which dependencia analysis should draw the attention of non-dependentistas: first, the possible separation between the potential capability of host governments to exercise greater negotiating strength in their relations with foreign investors anc 13The transfer of bargaining power into Third World hands would be retarded, however, to the extent that host countries compete among themselves to get foreign investment. 14Some companies, of course, may try to balance their increasing vulnerability in certain product lines with more profitable operations in other lines where they have tighter control over technology. Peter Evans hypothesizes, for example, that American petrochemical companies in Brazil are trading "unattractive" arrangements for the production of chlorine for a profitable "kicker" in polyurethane foam (TDI). "Testing the New Alliance: The Brazilian State, the Multinationals and the Launching of the Polo Petroquimico at Camacari, Bahia," Brown University, xerox, February 1976
Multinational corporations and dependency 85 their political'will''to do so;second,the possible economic distortions produced by the investors in the local economy.Bargaining capability means little if it is not (or cannot be)exercised for political reasons.15 Greater"benefits''are of limited value if they lead to a"perversion''of the structure of the local economy.The final section of this essay will address the question of political constraints on the exercise of national will.The following section will address the question of economic distor- tions produced by foreign investment. I Dependencia proposition II Multinational corporations create distortions within the local economy.There are innumerable allegations of distortion,but four appear to occupy positions of preeminence in the dependencia literature:first,that multinational corporations "preempt''the development of an indigenous economic base by squeezing out local entrepreneurs in the most''dynamic''sectors of the host country economy;second, that multinational corporations employ "inappropriate capital-intensive tech- nologies when they move in,adding to host country unemployment;third,that multinational corporations worsen the distribution of income in the host country or even produce an absolute loss for the lower 40 percent;fourth,that multinational corporations alter consumer tastes and undermine the culture of the host country. Where might a dialogue between dependentistas and non-dependentistas begin with each of these propositions? 1.Multinational corporations "preempt''the development of an indigenous economic base by squeezing out local entrepreneurs in the most''dynamic''sectors of the host country economy. The analysis of direct investment as a response to foreign opportunities in imperfect markets,introduced in Section I,can be a powerful tool for analyzing the question of"preemption''or of foreign industrial domination.The work of Hymer, Kindleberger,Vernon,et al.,suggests that multinational corporations will concen- trate their activities in industries,or at stages within industries,where the greatest barriers to the entry of competition are located-that is,where there are the greatest scale factors,the highest amounts of R D,the largest advertising efforts,and so forth.16 By common definition,therefore,they are likely to be found in those sectors considered by the host country as most "'dynamic''or most glamorous.At the moment when the multinational first comes into the host country,however,domes- have tried to use this distinction to define "exploitation"and"complicity in exploitation'within the balance of power framework suggested here,and to measure the cost of such exploitation quantita- tively in'The Theory of Interational Exploitation in Large Natural Resource Investments,'in Stephen J.Rosen and James R.Kurth,eds.,Testing Theories of Economic Imperialism (Lexington:Lexington Books,1974). Certainly multinational corporations appear to locate their activities in industries that are highly concentrated.Fernando Fajnzylber and Trinidad Martinez Tarrago have found that in Mexico in 1970
Multinational corporations and dependency 85 their political "will" to do so; second, the possible economic distortions produced by the investors in the local economy. Bargaining capability means little if it is not (or cannot be) exercised for political reasons.15 Greater "benefits" are of limited value if they lead to a "perversion" of the structure of the local economy. The final section of this essay will address the question of political constraints on the exercise of national will. The following section will address the question of economic distortions produced by foreign investment. Dependencia proposition II Multinational corporations create distortions within the local economy. There are innumerable allegations of distortion, but four appear to occupy positions of preeminence in the dependencia literature: first, that multinational corporations "preempt" the development of an indigenous economic base by squeezing out local entrepreneurs in the most "dynamic" sectors of the host country economy; second, that multinational corporations employ "inappropriate" capital-intensive technologies when they move in, adding to host country unemployment; third, that multinational corporations worsen the distribution of income in the host country or even produce an absolute loss for the lower 40 percent; fourth, that multinational corporations alter consumer tastes and undermine the culture of the host country. Where might a dialogue between dependentistas and non-dependentistas begin with each of these propositions? 1. Multinational corporations "preempt" the development of an indigenous economic base by squeezing out local entrepreneurs in the most "dynamic" sectors of the host country economy. The analysis of direct investment as a response to foreign opportunities in imperfect markets, introduced in Section I, can be a powerful tool for analyzing the question of'' preemption'' or of foreign industrial domination. The work of Hymer, Kindleberger, Vernon, et al., suggests that multinational corporations will concentrate their activities in industries, or at stages within industries, where the greatest barriers to the entry of competition are located—that is, where there are the greatest scale factors, the highest amounts of R & D, the largest advertising efforts, and so forth.16 By common definition, therefore, they are likely to be found in those sectors considered by the host country as most "dynamic" or most glamorous. At the moment when the multinational first comes into the host country, however, domes- 15I have tried to use this distinction to define "exploitation" and "complicity in exploitation" within the balance of power framework suggested here, and to measure the cost of such exploitation quantitatively in "The Theory of International Exploitation in Large Natural Resource Investments," in Stephen ]. Rosen and James R. Kurth, eds., Testing Theories of Economic Imperialism (Lexington: Lexington Books, 1974). "Certainly multinational corporations appear to locate their activities in industries that are highly concentrated. Fernando Fajnzylber and Trinidad Martinez Tarrago have found that in Mexico in 1970
86 International Organization tic firms may possess special skills,special contacts,or special methods of exploit- ing local market imperfections that foreign investors need to become successfully established.Insofar as they do,as Franco and Wells have hypothesized,the for- eigners will have a strong inducement to penetrate the host country by means of some kind of merger with a domestic enterprise.17 But this may be a transitory stage for many multinationals.Once the parent corporation expands the horizon for the subsidiary's operations beyond sales in the domestic market,its primary concern focuses on how to integrate the subsidiary most effectively into its worldwide (or regional)operations.At this point,local partners become a hindrance (because their loyalty is to the profits and dividends of the domestic subsidiary,not to the profits and dividends of the multinational corporate system as a whole),and the parent company will tend to buy up or force out independent stockholders. This is a cycle that clearly fits the expectations of dependencia theorists:as they predict,import substitution policies are not dependable as a tool to build an independent business class (or a "national bourgeoisie").Rather such policies stimulate foreign firms to link up with indigenous economic groups,as the for. eigners move from exporting to production for the local market,and then to take over those groups as they incorporate local production into their large multinational organizations.In the process,local industrial development falls prey to foreign corporate domination.18 But the main variant of the idea of foreign investment as oligopolistic expansion-the product cycle model-lays the basis for expecting a strong,and largely successful,reaction to this "preemptive''cycle as well.19 Not only is the growth of multinational corporations a history of the attempts by the firms to exploit some barrier to entry,it is also the history of the diffusion of the techniques and of the technology that create the barrier to entry in the first place.20 As production processes introduced by the foreign investor into the host country are standardized markets probed,financial feasibility demonstrated,and uncertainty reduced,local entrepreneurs or local state agencies should be able to duplicate the activities of the for example,foreign investors sold 61 percent of their output in markets where the four largest plants accounted for at least 50 percent of sales.Similarly,in Brazil in 1972,Richard Newfarmer and Willard Mueller have discovered that 49 percent of a sample of 233 US subsidiaries enjoyed market shares of 25 percent or more,and 21 percent of the sample produced over half of the sales in the markets they served. Richard S.Newfarmer and Willard F.Mueller,"Multinational Corporations in Brazil and Mexico: Structural Sources of Economic and Non-economic Power,"Report to the Subcommittee on Multina- tional Corporations of the Committee on Foreign Relations,US Senate,p.133;and Fernando Fajnzylber and Trinidad Martinez Tarrago,"'Las empresas transnacionales en la industria mexicana,"(Mexico City: CONACYT/CIDE,1972),cited in Newfarmer and Mueller,p.60 et seq. Lawrence G.Franco,Joint Venture Survival in Multinational Corporations (New York:Praeger, 1971);Louis T.Wells,Jr.."The Multinational Business Enterprise:What Kind of International Organi- zation?",in Robert O.Keohane and Joseph S.Nye,Jr.,eds.,Transnational Relations and World Politics,special issue of International Organization Vol.25,No.3(Summer 1971). 1Franco and Wells argue,however,that multinational enterprises tend to evolve toward an organization based on"worldwide product divisions'which are"not accompanied by significant purchases of the interests of local partners or sales of the parent company's equity in joint ventures." For the"product cycle model.''see the citations of Vernon,Stobaugh,and Wells in footnote 3,and of Moran in footnote 4. 20N.B.In contrast to the rhetoric of"technology transfer"that is assumed to take place by the enthusiasts of foreign investment,the diffusion of technology occurs in the product cyele model despite the continual struggle of the corporations to keep it tightly held
86 International Organization tic firms may possess special skills, special contacts, or special methods of exploiting local market imperfections that foreign investors need to become successfully established. Insofar as they do, as Franco and Wells have hypothesized, the foreigners will have a strong inducement to penetrate the host country by means of some kind of merger with a domestic enterprise.17 But this may be a transitory stage for many multinationals. Once the parent corporation expands the horizon for the subsidiary's operations beyond sales in the domestic market, its primary concern focuses on how to integrate the subsidiary most effectively into its worldwide (or regional) operations. At this point, local partners become a hindrance (because their loyalty is to the profits and dividends of the domestic subsidiary, not to the profits and dividends of the multinational corporate system as a whole), and the parent company will tend to buy up or force out independent stockholders. This is a cycle that clearly fits the expectations of dependencia theorists: as they predict, import substitution policies are not dependable as a tool to build an independent business class (or a "national bourgeoisie"). Rather such policies stimulate foreign firms to link up with indigenous economic groups, as the foreigners move from exporting to production for the local market, and then to take over those groups as they incorporate local production into their large multinational organizations. In the process, local industrial development falls prey to foreign corporate domination.18 But the main variant of the idea of foreign investment as oligopolistic expansion—the product cycle model—lays the basis for expecting a strong, and largely successful, reaction to this "preemptive" cycle as well.19 Not only is the growth of multinational corporations a history of the attempts by the firms to exploit some barrier to entry, it is also the history of the diffusion of the techniques and of the technology that create the barrier to entry in the first place.20 As production processes introduced by the foreign investor into the host country are standardized, markets probed, financial feasibility demonstrated, and uncertainty reduced, local entrepreneurs or local state agencies should be able to duplicate the activities of the for example, foreign investors sold 61 percent of their output in markets where the four largest plants accounted for at least 50 percent of sales. Similarly, in Brazil in 1972, Richard Newfarmer and Willard Mueller have discovered that 49 percent of a sample of 233 US subsidiaries enjoyed market shares of 25 percent or more, and 21 percent of the sample produced over half of the sales in the markets they served. Richard S. Newfarmer and Willard F. Mueller, "Multinational Corporations in Brazil and Mexico: Structural Sources of Economic and Non-economic Power," Report to the Subcommittee on Multinational Corporations of the Committee on Foreign Relations, US Senate, p. 133; and Fernando Fajnzylber and Trinidad Martinez Tarrago, "Lasempresas transnacionales en la industria mexicana," (Mexico City: CONACYT/CIDE, 1972), cited in Newfarmer and Mueller, p. 60 et seq. "Lawrence G. Franco, Joint Venture Survival in Multinational Corporations (New York: Praeger, 1971); Louis T. Wells, Jr., "The Multinational Business Enterprise: What Kind of International Organization?", in Robert O. Keohane and Joseph S. Nye, Jr., eds., Transnational Relations and World Politics, special issue of International Organization Vol. 25, No. 3 (Summer 1971). 18Franco and Wells argue, however, that multinational enterprises tend to evolve toward an organization based on "worldwide product divisions" which are "not accompanied by significant purchases of the interests of local partners or sales of the parent company's equity in joint ventures." 19Forthe "product cycle model," see the citations of Vernon, Stobaugh, and Wells in footnote 3, and of Moran in footnote 4. 20N.B. In contrast to the rhetoric of "technology transfer" that is assumed to take place by the enthusiasts of foreign investment, the diffusion of technology occurs in the product cycle model despite the continual struggle of the corporations to keep it tightly held