Multinational corporations and dependency 87 foreigner at a steadily decreasing cost.In contrast to dependencia theorists,product cycle analysts predict that the suppliers and auxiliary industries that begin by sup- porting the foreign multinational will eventually become a main source of competi- tion to it.21(The possible political implications of this hypothesis are discussed below.)In support of this process,host governments may well subsidize the growth of national alternatives to foreign multinationals on infant industry or national security grounds.From this perspective,the primary public policy problem for host country officials is not how to mobilize a massive revolution against foreign indus- trial domination,as some dependentista writers urge,but how to design and ad- minister an effective subsidy system that does not,as so often happens,degenerate into an endless drain of public funds needed for other projects in welfare or de- velopment.22 On balance,therefore,the analysis of multinational oligopolistic expansion suggests results that both support and limit dependencia predictions:foreign firms will continually be found in those "'dynamic''sectors where there are major market imperfections.They will continually attempt to integrate local production into their larger multinational organization so as to maximize the oligopoly returns to the system as a whole.But,in a dynamic sense,host country entrepreneurs (public and private)will meet ever-new opportunities to invade those areas that were originally the preserve of foreign multinationals,taking on more and more functions that initially only the foreigners could perform. This portends a pattern of challenge and response to foreign investors at odds with the rhetoric of"harmony of interests''that pervades the liberal characteriza- tion of the contribution of multinationals to Third World development.But it is a pattern of challenge and response that should not work cumulatively to host country disadvantage. The pace and extent of the economic counter-challenge to foreign corporate penetration in the Third World,however,has received surprisingly little rigorous study among either dependencia or non-dependencia analysts. 2.Multinational corporations employ "inappropriate''capital-intensive technologies that exacerbate rather than solve unemployment problems. This is an allegation that,at first glance,is difficult to reconcile with the idea of a profit-maximizing firm since an"appropriate"technology (one that takes advan- tage of local factor endowments,employing relatively more labor if labor is abun- dant)should minimize production costs.There are,however,three major hypoth- eses in non-dependencia literature that suggest why a multinational corporation might choose what would appear to be an"'inappropriate''technology. The first hypothesis proposes that foreign firms do not in fact compare the marginal costs of one production technique (e.g.,a capital-intensive technique)with the marginal costs of another production technique (e.g.,a labor-intensive tech- 2Evidence of this process has been discovered by Jorge Katz in the pharmaceutical,chemical,elec- tronics,automotive,and agricultural machinery industries.Programa Regional de Investigaciones en Ciencia y Tecnologia.CEPAL,Buenos Aires,1976. 2For an analysis of this point,see Harry G.Johnson,The Efficiency and Welfare Implications of the Intemational Corporation"in Charles P.Kindleberger,ed.,The International Corporation:A Sym- posium (Cambridge,Mass.:MIT Press,1970).Multinational corporations and dependency 87 foreigner at a steadily decreasing cost. In contrast to dependencia theorists, product cycle analysts predict that the suppliers and auxiliary industries that begin by sup￾porting the foreign multinational will eventually become a main source of competi￾tion to it.21 (The possible political implications of this hypothesis are discussed below.) In support of this process, host governments may well subsidize the growth of national alternatives to foreign multinationals on infant industry or national security grounds. From this perspective, the primary public policy problem for host country officials is not how to mobilize a massive revolution against foreign indus￾trial domination, as some dependentista writers urge, but how to design and ad￾minister an effective subsidy system that does not, as so often happens, degenerate into an endless drain of public funds needed for other projects in welfare or de￾velopment.22 On balance, therefore, the analysis of multinational oligopolistic expansion suggests results that both support and limit dependencia predictions: foreign firms will continually be found in those "dynamic" sectors where there are major market imperfections. They will continually attempt to integrate local production into their larger multinational organization so as to maximize the oligopoly returns to the system as a whole. But, in a dynamic sense, host country entrepreneurs (public and private) will meet ever-new opportunities to invade those areas that were originally the preserve of foreign multinationals, taking on more and more functions that initially only the foreigners could perform. This portends a pattern of challenge and response to foreign investors at odds with the rhetoric of "harmony of interests" that pervades the liberal characteriza￾tion of the contribution of multinationals to Third World development. But it is a pattern of challenge and response that should not work cumulatively to host country disadvantage. The pace and extent of the economic counter-challenge to foreign corporate penetration in the Third World, however, has received surprisingly little rigorous study among either dependencia or non-dependencia analysts. 2. Multinational corporations employ "inappropriate" capital-intensive technologies that exacerbate rather than solve unemployment problems. This is an allegation that, at first glance, is difficult to reconcile with the idea of a profit-maximizing firm since an "appropriate" technology (one that takes advan￾tage of local factor endowments, employing relatively more labor if labor is abun￾dant) should minimize production costs. There are, however, three major hypoth￾eses in non-dependencia literature that suggest why a multinational corporation might choose what would appear to be an "inappropriate" technology. The first hypothesis proposes that foreign firms do not in fact compare the marginal costs of one production technique (e.g., a capital-intensive technique) with the marginal costs of another production technique (e.g., a labor-intensive tech- 21Evidence of this process has been discovered by Jorge Katz in the pharmaceutical, chemical, elec￾tronics, automotive, and agricultural machinery industries. Programa Regional de Investigaciones en Ciencia y Tecnologia. CEPAL, Buenos Aires, 1976. 22For an analysis of this point, see Harry G. Johnson, "The Efficiency and Welfare Implications of the International Corporation" in Charles P. Kindleberger, ed., The International Corporation: A Sym￾posium (Cambridge, Mass.: MIT Press, 1970)