《国际政治经济学》文献资料（Political Economy of International Development）Efficiency（Contradictions）of Multinational Corporations

THE EFFICIENCY (CONTRADICTIONS)OF MULTINATIONAL CORPORATIONS By STEPHEN HYMER Yale University Multinational corporations are a substitute for to reap the benefits of internal specialization aud the market as a method of organizing interna- exchange.Few studies have been made on the re- tional exchange.They are "..islands of con- lationship of foreign investment to a firm's overall scious power in an ocean of unconscious coopera- efficiency and as far as quantitative evidence is tion,"to use D.H.Robertson's phrase.1 This es- concerned,we must view this question as a com- say examines some of the contradictions of this pletely open one.With regard to the effect of latest stage in the development of private busi- size,the evidence is more plentiful but not con- ness enterprise. clusive.A number of studies on differences in per- At the outset,we should note that the multina- formance of large and small firms have in general tional corporation raises more questions than eco- concluded that firms experience economies of nomic theory can answer.Multinational corpora- scale up to a certain minimum size,after which tions are typically large firms operating in imper- there is little relationship between size and per- fect markets and the question of their efficiency is formance.Applying these results to the multina- a question of the efficiency of oligopolistic deci- tional corporation suggests that most parent firms sion making,an area where much of welfare eco- are large enough to have exhausted economies of nomics breaks down,especially the proposition scale without foreign investment,although many that competition allocates resources efficiently of their subsidiaries may be too small to stand on and that there is a harmony between private their own feet. proft maximization and the general interest. These tests,however,bave several inadequacies Moreover,multinational corporations bring into and may seriously underestimate the advantages high definition such social and political problems of size.The major difficulty is that large firms are as want creation,alienation,domination,and the seldom engaged in exactly the same activities as relationship or interface between corporations and medium-sized or smaller firms and their perfor- national states (including the question of imperi- mance is not really comparable.The fact that alism),which cannot be analyzed in purely "eco- very large firms do not seem to be significantly nomic”terms. more profitable than their smaller rivals or to grow significantly faster does not preclude the I.Division of Labor and the Extent of the Firm possibility that they are specializing in activities Our starting point is the fact that there are two where size is of great advantage and which would kinds of division of labor:the division of labor not be undertaken if the large firms did not exist. between firms coordinated by the markets;and The structure of output within a country could the division of labor within firms,coordinated by well be a function of the size distribution of its entrepreneurs.International trade theory has frms without there being observable differences been mainly concerned with the first of these and between large and small firms with regard to the has long stressed the desirability of widening in- more commonly studied characteristics. ternational markets to increase the division of la- The qualitative evidence on the structure of bor and exchange.Far less attention has been business enterprise and its evolution through time paid to the parallel proposition that the division suggests that both size and internationality have of labor within a firm is limited by the extent of important positive effects on a firm's strength the firm and the economic and social questions and ability.Since the beginning of the industrial this raises. revolution there has been a steady increase in the Unfortunately,the empirical evidence is not size of manufacturing firms,so persistent that it very helpful in deciding the degree to which large might almost be formulated as a general law of international firms should be encouraged in order capital accumulation.These increases in size were accompanied by important changes in organiza- D.H.Robertson quoted in R.H.Coase,"The tional structure involving both increased subdivi- Nature of the Firm,"Economica,New Series,1937, pp.386-405.Reprinted in G.S.Stigler and K.E. sion or differentiation of tasks and increased inte- Boulding,Readings in Price Theory (Richard D. gration through the creation of new organs of Irwin,Inc.,1932). control.Business administration became a highly 441

THE EFFICIENCY (CONTRADICTIONS) OF MULTINATIONAL CORPORATIONS By STEPHEN HYMER Yale University Multinational corporations are a substitute for the market as a method of organizing interna￾tional exchange. They are ".. . islands of con￾scious power in an ocean of unconscious coopera￾tion," to use D. H. Robertson's phrase.^ This es￾say examines some of the contradictions of this latest stage in the development of private busi￾ness enterprise. At the outset, we should note that the multina￾tional corporation raises more questions than eco￾nomic theory can answer. Multinational corpora￾tions are typically large firms operating in imper￾fect markets and the question of their efficiency is a question of the efficiency of oligopolistic deci￾sion making, an area where much of welfare eco￾nomics breaks down, especially the proposition that competition allocates resources efficiently and that there is a harmony between private profit maximization and the general interest. Moreover, multinational corporations bring into high definition such social and political problems as want creation, alienation, domination, and the relationship or interface between corporations and national states (including the question of imperi￾alism), which cannot be analyzed in purely "eco￾nomic" terms. I. Division of Lahor and the Extent of the Firm Our starting point is the fact that there are two kinds of division of labor: the division of labor between firms coordinated by the markets; and the division of labor within firms, coordinated by entrepreneurs. International trade theory has been mainly concerned with the first of these and has long stressed the desirability of widening in￾ternational markets to increase the division of la￾bor and exchange. Far less attention has been paid to the parallel proposition that the division of labor within a firm is limited by the extent of the firm and the economic and social questions this raises. Unfortunately, the empirical evidence is not very helpful in deciding the degree to which large international firms should be encouraged in order ' D. H. Robertson quoted in R. H. Coase, "The Nature of the Firm," Económica, New Series, 1937, pp. 386-405. Reprinted in G. S. Stigler and K. E. Boulding, Readings in Price Theory (Richard D. Irwin, Inc., 1932). to reap the benefits of internal specialization aud exchange. Few studies have been made on the re￾lationship of foreign investment to a firm's overall efficiency and as far as quantitative evidence is concerned, we must view this question as a com￾pletely open one. With regard to the effect of size, the evidence is more plentiful but not con￾clusive. A number of studies on differences in per￾formance of large and small firms have in general concluded that firms experience economies of scale up to a certain minimum size, after which there is little relationship between size and per￾formance. Applying these results to the multina￾tional corporation suggests that most parent firms are large enough to have exhausted economies of scale without foreign investment, although many of their subsidiaries may be too small to stand on their own feet. These tests, however, have several inadequacies and may seriously underestimate the advantages of size. The major difficulty is that large firms are seldom engaged in exactly the same activities as medium-sized or smaller firms and their perfor￾mance is not really comparable. The fact that very large firms do not seem to be significantly more profitable than their smaller rivals or to grow significantly faster does not preclude the possibility that they are specializing in activities where size is of great advantage and which would not be undertaken if the large firms did not exist. The structure of output within a country could well be a function of the size distribution of its firms without there being observable differences between large and small firms with regard to the more commonly studied characteristics. The qualitative evidence on the structure of business enterprise and its evolution through time suggests that both size and internationality have important positive effects on a firm's strength and ability. Since the beginning of the industrial revolution there has been a steady increase in the size of manufacturing firms, so persistent that it might almost be formulated as a general law of capital accumulation. These increases in size were accompanied by important changes in organiza￾tional structure involving both increased subdivi￾sion or differentiation of tasks and increased inte￾gration through the creation of new organs of control. Business administration became a highly 441

442 AMERICAN ECONOMIC ASSOCIATION specialized activity with its own elaborate division the top two levels are separated from the bottom of labor;and the corporation developed a brain level.In the multidivisional corporation,diff- to consciously coordinate the various specialties erentiation is far more complete;level three is and to plan for the survival of the organism as a completely split off from level two and is concen- whole. trated in the general office whose specific function Chandler2 distinguishes three major stages in is strategy,not tactics. the development of corporate capital.First,the In other words,the process of capital accumu- Marshallian firm,organized at the factory level, lation has become more and more specialized confined to a single function and a single indus- through time.As the corporation evolved,it de- try,and tightly controlled by one or a few men veloped an elaborate system of internal division who,as it were,see everything,and decide every- of labor,able to absorb and apply both the physi- thing.The second stage emerged in the United cal sciences and the social sciences to business ac- States at the end of the nineteenth century when tivity on a scale which could not be imagined in rapid growth and the merger movement led to earlier years.At the same time,it developed a high- large national corporations,and a new structure er brain to command its very large concentration of administration was developed to deal with the of wealth.This gave it the power to invest on a new strategy of continent-wide,vertically inte- much larger scale and with a much wider time-ho- grated production and marketing.The family firm rizon than the smaller,less developed firms that gave way to the modern corporation with a highly preceded it.The modern multidivisional corpora- elaborate administrative structure to organize the tion is thus a far cry from the marshallian firm in many disparate units of a giant enterprise.The both its vision and its strength.The Marshallian next stage,the multidivisional corporation,began capitalist ruled his factory from an office on the in the 1920's and gathered great momentum after second floor.At the turn of the century,the pres- the second World War.It too was a response to a ident of a large national corporation was lodged new marketing strategy.To meet the conditions in a higher building,say on the seventh floor, of continuous innovation,corporations were de- with wider perspectives and greater power.In the centralized into several divisions,each specializing giant corporation of today,managers rule from in one product line and organized as an almost the top of skyscrapers;on a clear day,they can autonomous unit similar in structure to the na- almost see the world. tional corporation.At the same time,an enlarged Each step in the evolution of business enter- corporate brain was created in the form of the prise had important implications for the structure general office to coordinate the various divisions of the international economy,just as each excur- and to plan overall growth and survival.This sion into the international economy provided new form is highly flexible and can operate in several challenges to the corporation and speeded its evo- industries and adjust quickly to rapidly changing lutionary development.In a world of Marshallian demands and technology. firms,commodity trade and portfolio capital were With each step in the development of business the main engines of international exchange. administration,capital obtained new power and Movement of enterprise between countries was new horizons.As Chandler and Redlicha point sharply limited because firms were small and out,there are three levels of business administra- lacked the appropriate administrative structure. tion.Level three,the lowest level,is concerned The diffusion of Marshall's vital fourth factor, with managing the day-to-day operations of the organization,from advanced to less advanced enterprise;i.e.,keeping it going within the estab- countries was therefore exceedingly slow.Move- lished framework.Level two is responsible for ments of portfolio capital were substantial,at coordinating the managers at level three.Level times,because the small Marshallian firms were one's function is goal determination and planning; associated with a highly developed banking and i.e.,setting the framework for the lower levels.In financial system.But the ability of less advanced the Marshallian firm all three levels are embodied countries to absorb capital (and technology)was in one entrepreneur.In the national corporation, limited to the rate at which they could build up their own organizations,a slow and difficult pro- 3 Alfred D.Chandler,Strategy and Structure (Dou- cess given the negative policies of most govern- bleday Co.,1961). ments in Africa,Asia,and Latin Ameri- Alfred D.Chandler and Fritz Redlich,"Recent ca,especially those in colonial dependencies.The Developments in American Business Administration and Their Conceptualization,"Bus.Hist.Rev., range of goods which could be produced was thus Spring,1961. restricted and the possibility for international

442 AMERICAN ECONOMIC ASSOCIATION specialized activity with its own elaborate division of labor; and the corporation developed a brain to consciously coordinate the various specialties and to plan for the survival of the organism as a whole. Chandler^ distinguishes three major stages in the development of corporate capital. First, the Marshallian firm, organized at the factory level, conñned to a single function and a single indus￾try, and tightly controlled by one or a few men who, as it were, see everything, and decide every￾thing. The second stage emerged in the United States at the end of the nineteenth century when rapid growth and the merger movement led to large national corporations, and a new structure of administration was developed to deal with the new strategy of continent-wide, vertically inte￾grated production and marketing. The family firm gave way to the modern corporation with a highly elaborate administrative structure to organize the many disparate units of a giant enterprise. The next stage, the multidivisional corporation, began in the 192O's and gathered great momentum after the second World War. It too was a response to a new marketing strategy. To meet the conditions of continuous innovation, corporations were de￾centralized into several divisions, each specializing in one product line and organized as an almost autonomous unit similar in structure to the na￾tional corporation. At the same time, an enlarged corporate brain was created in the form of the general office to coordinate the various divisions and to plan overall growth and survival. This form is highly flexible and can operate in several industries and adjust quickly to rapidly changing demands and technology. With each step in the development of business administration, capital obtained new power and new horizons. As Chandler and Redlich^ point out, there are three levels of business administra￾tion. Level three, the lowest level, is concerned with managing the day-to-day operations of the enterprise; i.e., keeping it going within the estab￾lished framework. Level two is responsible for coordinating the managers at level three. Level one's function is goal determination and planning; i.e., setting the framework for the lower levels. In the Marshallian firm all three levels are embodied in one entrepreneur. In the national corporation. ' Alfred D. Chandler, Strategy and Structure (Dou￾bleday & Co., 1961). •Alfred D. Chandler and Fritz Redlich, "Recent Developments in American Business Administration and Their Conceptualization," Bus. Hist. Rev., Spring, 1961. the top two levels are separated from the bottom level. In the multidivisional corporation, diff￾erentiation is far more complete; level three is completely split off from level two and is concen￾trated in the general office whose specific function is strategy, not tactics. In other words, the process of capital accumu￾lation has become more and more specialized through time. As the corporation evolved, it de￾veloped an elaborate system of internal division of labor, able to absorb and apply both the physi￾cal sciences and the social sciences to business ac￾tivity on a scale which could not be imagined in earlier years. At the same time, it developed a high￾er brain to command its very large concentration of wealth. This gave it the power to invest on a much larger scale and with a much wider time-ho￾rizon than the smaller, less developed firms that preceded it. The modern multidivisional corpora￾tion is thus a far cry from the Marshallian firm in both its vision and its strength. The Marshallian capitalist ruled his factory from an office on the second floor. At the turn of the century, the pres￾ident of a large national corporation was lodged in a higher building, say on the seventh floor, with wider perspectives and greater power. In the giant corporation of today, managers rule from the top of skyscrapers; on a clear day, they can almost see the world. Each step in the evolution of business enter￾prise had important implications for the structure of the international economy, just as each excur￾sion into the international economy provided new challenges to the corporation and speeded its evo￾lutionary development. In a world of Marshallian firms, commodity trade and portfolio capital were the main engines of international exchange. Movement of enterprise between countries was sharply limited because firms were small and lacked the appropriate administrative structure. The diffusion of Marshall's vital fourth factor, organization, from advanced to less advanced countries was therefore exceedingly slow. Move￾ments of portfolio capital were substantial, at times, because the small Marshallian firms were associated with a highly developed banking and financial system. But the ability of less advanced countries to absorb capital (and technology) was limited to the rate at which they could build up their own organizations, a slow and difficult pro￾cess given the negative policies of most govern￾ments in Africa, Asia, and Latin Ameri￾ca, especially those in colonial dependencies. The range of goods which could be produced was thus restricted and the possibility for intemational

THE INTERNATIONAL FIRM 443 trade to equalize factor prices was severely lim- certain highly intractable problems which greatly ited.* impede their efficiency.We turn to these consider- The national corporation opened new possibili- ations. ties of transferring organizational abilities inter- nationally.The new administrative structure and II.Bigness and Fewness financial power enabled firms to undertake direct Multinational corporations enlarge the domain foreign investments and organize large-scale pro- of centrally planned world production and de- duction in mining and manufacturing in foreign crease the domain of decentralized market-di- countries.However,this migration of business en- rected specialization and exchange.Bigness is thus terprise occurred only on a limited scale and was paid for,in part,by fewness,and a decline in usually restricted to a narrow activity;i.e.,to ac- competition since the size of the market is limited quiring raw materials used by the parent company by the size of the 'firm.The precise effect of the or to exploiting some technological advance or dif- present wave of direct investment on seller con- ferentiated product developed by the parent centration in world markets is not well estab- company.Moreover,to the extent that invest- lished.On the one hand,improved communica- ment strengthened the firm's market control,its tions are breaking down barriers to trade and wid- effect was considerably less beneficial and perhaps ening the market facing most buyers.On the even negative. other hand,direct foreign investment tends to re- The modern multidivisional or conglomerate duce the number of alternatives facing sellers and enterprise is a much more powerful organizational to stay the forces of international competition.A form than the national corporation and appears great deal of statistical work needs to be done to capable of integrating world production and ex- evaluate the net effect of these two tendencies change to a much larger extent.Larger size and a and establish the exact trend in the level of seller more advanced administrative structure give it a concentration,taking into account the growing in- much wider horizon leading in many cases to a ternational nature of the market.All that can be global outlook and a transformation to the stage said at present is that the world level of concen- of multinational enterprise.It seems that after a tration is much higher than it would be if foreign certain point,a corporation comes to think in investment and domestic mergers were restricted terms of its world market position rather than Since most countries are encouraging mergers at merely its United States or European market pos- home and foreign investment abroad,for better ition and to plan in terms of worldwide factor or worse,the opportunity to increase competition availabilities and demand patterns.Since the pro- by maintaining numbers is not being taken up. cess is just beginning,it is difficult to evaluate Direct foreign investment thus has a dual na- how strong this tendency will be.However,it is ture.It is an instrument which allows business clear that at present large corporations are con- firms to transfer capital,technology,and organi- sciously moving towards an international perspec- zational skill from one country to another.It is tive much faster than other institutions and espe- also an instrument for restraining competition be- cially much faster than governments,and are in tween firms of different nations.Analyzing any the vanguard of planners of the new international particular case is an exceedingly complex matter, economy created by the aeronautical and elec- as the antitrust literature shows.s For present tronic revolutions.Since multinational corpora- purposes,the important point is to note that the tions also have great financial and technical re- general presumption of international trade eco- sources,they will certainly have many successes nomists in favor of free trade and free factor and will be able to speed up the spread of tech- movements,on the grounds of allocative effic- nology and to organize activities until now impos- iency,does not apply to direct foreign investment sible.They are a large step forward but this is because of the anticompetitive effect inherently not,however,the same thing as saying that they associated with it.Just as in antitrust theory serve the general interest as well as their own, there are recognized reasons,within the frame- that they are the best way to exploit the possibili- ties of modern science,or that they do not create See also Stephen Hymer,"Direct Foreign Invest- ment and the National Economic Interest,"Peter Russel,ed.,Nationalism in Canada (Toronto:Mc- Stephen Hymer and Stephen Resnick,"Interna- Graw-Hill of Canada,1966);Yale Economic Growth tional Trade and Uneven Development,"in J.N. Center,Paper No.108;"L'Impact des Firmes Inter- Bhagwati,R.W.Jones,R.A.Mundell, Jaroslav nationals,"in M.Bye,ed.,La Politique Industrielle Vanek,eds.,Kindleberger Festschrift (M.I.T.Press, de L'Europe Integree (Paris:Presses Universitaires forthcoming). de France,1968)

THE INTERNATIONAL FIRM 443 trade to equalize factor prices was severely lim￾ited.* The national corporation opened new possibili￾ties of transferring organizational abilities inter￾nationally. The new administrative structure and financial power enabled firms to undertake direct foreign investments and organize large-scale pro￾duction in mining and manufacturing in foreign countries. However, this migration of business en￾terprise occurred only on a limited scale and was usually restricted to a narrow activity; i.e., to ac￾quiring raw materials used by the parent company or to exploiting some technological advance or dif￾ferentiated product developed by the parent company. Moreover, to the extent that invest￾ment strengthened the firm's market control, its effect was considerably less beneficial and perhaps even negative. The modern multidivisional or conglomerate enterprise is a much more powerful organizational form than the national corporation and appears capable of integrating world production and ex￾change to a much larger extent. Larger size and a more advanced administrative structure give it a much wider horizon leading in many cases to a global outlook and a transformation to the stage of multinational enterprise. It seems that after a certain point, a corporation comes to think in terms of its world market position rather than merely its United States or European market pos￾ition and to plan in terms of worldwide factor availabilities and demand patterns. Since the pro￾cess is just beginning, it is difficult to evaluate how strong this tendency will be. However, it is clear that at present large corporations are con￾sciously moving towards an international perspec￾tive much faster than other institutions and espe￾cially much faster than governments, and are in the vanguard of planners of the new international economy created by the aeronautical and elec￾tronic revolutions. Since multinational corpora￾tions also have great financial and technical re￾sources, they will certainly have many successes and will be able to speed up the spread of tech￾nology and to organize activities until now impos￾sible. They are a large step forward but this is not, however, the same thing as saying that they serve the general interest as well as their own, that they are the best way to exploit the possibili￾ties of modern science, or that they do not create ' Stephen Hymer and Stephen Resnick, "Interna￾tional Trade and Uneven Development," in J. N. Bhagwati, R. W. Jones, R. A. Mundell, Jaroslav Vanek, eds., Kindleberger Festschrift (M.I.T. Press, forthcoming). certain highly intractable problems which greatly impede their efficiency. We turn to these consider￾ations. II. Bigness and Fewness Multinational corporations enlarge the domain of centrally planned world production and de￾crease the domain of decentralized market-di￾rected specialization and exchange. Bigness is thus paid for, in part, by fewness, and a decline in competition since the size of the market is limited by the size of the firm. The precise effect of the present wave of direct investment on seller con￾centration in world markets is not well estab￾lished. On the one hand, improved communica￾tions are breaking down barriers to trade and wid￾ening the market facing most buyers. On the other hand, direct foreign investment tends to re￾duce the number of alternatives facing sellers and to stay the forces of international competition. A great deal of statistical work needs to be done to evaluate the net effect of these two tendencies and establish the exact trend in the level of seller concentration, taking into account the growing in￾ternational nature of the market. All that can be said at present is that the world level of concen￾tration is much higher than it would be if foreign investment and domestic mergers were restricted. Since most countries are encouraging mergers at home and foreign investment abroad, for better or worse, the opportunity to increase competition by maintaining numbers is not being taken up. Direct foreign investment thus has a dual na￾ture. It is an instrument which allows business firms to transfer capital, technology, and organi￾zational skill from one country to another. It is also an instrument for restraining competition be￾tween firms of different nations. Analyzing any particular case is an exceedingly complex matter, as the antitrust literature shows." For present purposes, the important point is to note that the general presumption of international trade eco￾nomists in favor of free trade and free factor movements, on the grounds of allocative effic￾iency, does not apply to direct foreign investment because of the anticompetitive effect inherently associated with it. Just as in antitrust theory there are recognized reasons, within the frame- ' See also Stephen Hymer, "Direct Foreign Invest￾ment and the National Economic Interest," Peter Rüssel, ed.. Nationalism in Canada (Toronto: Mc￾Graw-Hill of Canada, 1966) ; Yale Economic Growth Center, Paper No. 108; "L'Impact des Firmes Inter￾nationals," in M. Bye, ed., La Politique Industrielle de L'Europe Intégrée (Paris : Presses Universitaires de France, 1968)

444 AMERICAN ECONOMIC ASSOCIATION work of neoclassical economics,for preventing a creasing competitiveness,may improve general firm from merging with another firm or from in- welfare in the rich countries as well-although it creasing its share of the market by growth,there will harm those in the monopoly position. are also international.antitrust reasons for pre- venting a firm of one country from taking over a III.The International"Trickle Down" firm in another country or from acquiring or in- Many economists,in dealing with oligopoly, creasing its share of foreign production.Since this prefer to stress,as Schumpeter did,that the com- point can be easily misunderstood,it is important petition that counts lies in creative destruction to stress that this is not a second-best argument through the introduction of new technology and but a genuine argument on antimonopoly grounds new products.In that case,an oligopolistic for interfering in international markets.A re- market structure,even though it interferes with striction on direct investment or a policy to break static optimum allocation,may be a necessary or up a multinational corporation may be in some at least a contributing factor to dynamic opti- cases the only way of establishing a higher degree mum allocation in a private enterprise system,be- of competition in that industry.National anti- cause it allows innovators to capture some of the trust measures cannot substitute for international benefits of their discoveries and thus provides the antitrust when,for example,one of the major po- incentive for research and development.The rec- tential competitors to a domestic firm is its sister ord of the United States shows that one cer- or parent affiliate within the same multinational tainly cannot fault oligopoly on the grounds that group.In short,when we leave the conditions of it does not produce a very rapid rate of techno- perfect competition we lose the assumption of the logical change and product innovation.(Indeed it invisible hand. is easier to argue that the rate of change is too This argument,it should be noted,provides an high.)One can expect international oligopoly via important rationale for the infant entrepreneur multinational corporation to provide the same argument supporting protection.Temporary pro- kind of dynamic environment for the world econ- tection of a weak firm from a stronger firm can omy as a whole. improve the competitive structure of the industry The question of efficiency therefore hinges on in future periods by maintaining numbers.In the the direction of change rather than the rate of present context,the cost of this protection would change.An analysis of this problem involves an have to be borne by the country that offers it excursion into unexplored terrain since we do not while the benefits would accrue to the world as a now have an adequate theory on how corporations whole.Thus,in reverse of the usual arguments, choose between the available paths of innovation. myopic behavior will lead to too little protection We certainly cannot assume that market forces rather than too much.This presents a particularly compel firms to choose the optimum path.It is acute problem in the case of underdeveloped true that an innovation must,to some extent countries.These countries typically do not sell meet the market test for a corporation to survive. commodities or buy capital or technology in com- However,what is at stake here is not whether the petitive markets where there is an established consumer has some choice but rather whether an price at which they can trade whatever quantity oligopolistically competitive market structure they want.Instead,they frequently face only a provides him with the full range of choices possi- few potential buyers of their raw materials or ble.Oligopolists tend to copy each other,and their manufactured goods and a few potential sel- their predictions as to what the consumer wants lers of a particular technology.The price they re- are often self-fulfilling,since in fact this is all ceive or pay therefore depends on their skill and that the consumer is offered.If we had only large strength in bargaining and not on market condi- numbers of independent decision centers could we tions alone.The less developed the country,the assume that all avenues had been explored. greater its disadvantage in the bargaining process Since we cannot possibly treat this complex because it has fewer organizations that are in any topic in any detail in the present paper,let us way a match for the giant companies with which simply examine one theory of innovation closely it is dealing.Given the oligopolistic front main- associated with the multinational corporation and tained by the firms from developed countries,the the international demonstration effect.The mark- underdeveloped countries need to devote an impor- tant share of their scarce resources to building up Sean Gervasi,"Publicite et Croissance Econo- national enterprises which they can control and mique,"Economie et Humanisme,Nov.-Dec..1964 Opu- use in bargaining with foreign oligopolists.Ironi- Harry Johnson,"The Political Economy of lence,"in The Canadian Quandary (MeGraw-Hill, cally,their stronger bargaining position,by in- 1962)

444 AMERICAN ECONOMIC ASSOCIATION work of neoclassical economics, for preventing a firm from merging with another firm or from in￾creasing its share of the market by growth, there are also international, antitrust reasons for pre￾venting a firm of one country from taking over a firm in another country or from acquiring or in￾creasing its share of foreign production. Since this point can be easily misunderstood, it is important to stress that this is not a second-best argument but a genuine argument on antimonopoly grounds for interfering in international markets. A re￾striction on direct investment or a policy to break up a multinational corporation may be in some cases the only way of establishing a higher degree of competition in that industry. National anti￾trust measures cannot substitute for international antitrust when, for example, one of the major po￾tential competitors to a domestic firm is its sister or parent affiliate within the same multinational group. In short, when we leave the conditions of perfect competition we lose the assumption of the invisible hand. This argument, it should be noted, provides an important rationale for the infant entrepreneur argument supporting protection. Temporary pro￾tection of a weak firm from a stronger firm can improve the competitive structure of the industry in future periods by maintaining numbers. In the present context, the cost of this protection would have to be borne by the country that offers it while the benefits would accrue to the world as a whole. Thus, in reverse of the usual arguments, myopic behavior will lead to too little protection rather than too much. This presents a particularly acute problem in the case of underdeveloped countries. These countries typically do not sell commodities or huy capital or technology in com￾petitive markets where there is an established price at which they can trade whatever quantity they want. Instead, they frequently face only a few potential buyers of their raw materials or their manufactured goods and a few potential sel￾lers of a particular technology. The price they re￾ceive or pay therefore depends on their skill and strength in bargaining and not on market condi￾tions alone. The less developed the country, the greater its disadvantage in the bargaining process because it has fewer organizations that are in any way a match for the giant companies with which it is dealing. Given the oligopolistic front main￾tained by the firms from developed countries, the underdeveloped countries need to devote an impor￾tant share of their scarce resources to building up national enterprises which they can control and use in bargaining with foreign oligopolists. Ironi￾cally, their stronger bargaining position, by in￾creasing competitiveness, may improve general welfare in the rich countries as well—although it will harm those in the monopoly position. III. The International "Trickle Down" Many economists, in dealing with oligopoly, prefer to stress, as Schumpeter did, that the com￾petition that counts lies in creative destruction through the introduction of new technology and new products. In that case, an oligopolistic market structure, even though it interferes with static optimum allocation, may be a necessary or at least a contributing factor to dynamic opti￾mum allocation in a private enterprise system, be￾cause it allows innovators to capture some of the benefits of their discoveries and thus provides the incentive for research and development. The rec￾ord of the United States shows that one cer￾tainly cannot fault oligopoly on the grounds that it does not produce a very rapid rate of techno￾logical change and product innovation. (Indeed it is easier to argue that the rate of change is too high.) One can expect international oligopoly via multinational corporation to provide the same kind of dynamic environment for the world econ￾omy as a whole. The question of efficiency therefore hinges on the direction of change rather than the rate of change. An analysis of this problem involves an excursion into unexplored terrain since we do not now have an adequate theory on how corporations choose between the available paths of innovation. We certainly cannot assume that market forces compel firms to choose the optimum path. It is true that an innovation must, to some extent, meet the market test for a corporation to survive. However, what is at stake here is not whether the consumer has some choice hut rather whether an oligopolistically competitive market structure provides him with the full range of choices possi￾ble.* Oligopolists tend to copy each other, and their predictions as to what the consumer wants are often self-fulfilling, since in fact this is all that the consumer is offered. If we had only large numbers of independent decision centers could we assume that all avenues had been explored. Since we cannot possibly treat this complex topic in any detail in the present paper, let us simply examine one theory of innovation closely associated with the multinational corporation and the international demonstration effect. The mark- ' Sean Gervasi, "Publicité et Croissance Econo￾mique," Economie et Humanisme, Nov.-Déc, 1964. Harry Johnson, "The Political Economy of Opu￾lence," in The Canadian Quandary (McGraw-Hill, 1962)

THE INTERNATIONAL FIRM 445 eting literature suggests new products typically of the modern economic problem..."was the follow a cycle known as trickle-down or two-stage division of labor within the factory between those marketing.An innovation is first adopted by a who plan and organize economic activity and small group of individuals who act as opinion those who work for them.In the modern corpora- leaders and is then copied by others via the dem- tion the hierarchical structure of command and onstration effect.In this process,the rich get authority has been greatly elaborated from the more votes than everyone else,first of all because simple division between owners and workers in they have more money,second of all because they the Marshallian firm,but the tensions and have discretionary income and can afford to be conficts of autocracy remain.They take on par- experimental,and,third,because they have high ticular importance in the multinational corpora- status and are more likely to be copied.The prin- tion where problems of nationalism and problems ciple of consumer sovereignty cannot easily be of authoritarianism intertwine. applied to this process since,at most,only the Multinational corporations are torn in two di- special group in the first stage of the marketing rections.On the one hand,they must adapt to lo- process has something approaching a free choice. cal circumstances in each country.This calls for The rest have only the choice between conform- decentralized decision making.On the other hand, ing or being isolated. they must coordinate their activities in various In the international economy, trickle-down parts of the world and stimulate the flow of ideas marketing takes the form of the international from one part of their empire to another.This demonstration effect.Products are first intro- calls for centralized controls.They must therefore duced in the United States or Europe and then develop an organizational structure to balance the spread to other countries.Multinational corpora- need to coordinate and integrate operations with tions speed up this process by making it easier to the need to adapt to a patchwork quilt of lan- transfer new products and marketing methods to guages,laws,and customs.One solution is divi- less advanced countries.One of the key motives sion of labor based on nationality.Day-to-day for direct investment,cited by corporations,is to management in each country is left to nationals gain control over marketing facilities in order to of that country who are intimately familiar with facilitate the spread of their products.If firms local conditions and practices and best suited to were denied control over communication and mar- deal with local problems and local government. keting facilities in the foreign countries and we had These nationals remain rooted in one spot,but a regime of national firms (private or socialized) above them is a layer of people who move around rather than multinational firms,the pattern of from country to country,as bees among flowers, output would almost certainly be quite different transmitting information from one subsidiary to than the one that is now observed.There would another and from the lower levels to the general be more centers of innovation,and probably more office at the apex of the corporate structure.In variety of choices offered to the consumers,as the nature of things,these people,for the most each country developed products suited to its par- part,will be citizens of the country of the parent ticular characteristics.Products from one country corporation,just as we now find that the top ex- would spread to other countries either through ecutives of most of the major corporations in the trade or imitation but the movement would be United States are drawn from a relatively small coordinated by market competition rather than homogeneous cultural group quite distinct from the planning decisions of top management in a the population of the United States as a whole. few corporations whose interest it is to foreclose This creates two types of problems.In the first competition,to restrict the choices offered,and to place,there is the internal problem of creating in- insure the survival of their own organizations.It centives for foreigners whose access to the top is difficult to speak with professional certainty in corporate positions will be necessarily limited. this badly neglected field,but it does not appear The second problem is far more important and is to be socially efficient to allow corporations to in the nature of an external diseconomy.The sub- monopolize information on new possibilities cre- sidiaries of multinational corporations are fre- ated by science. quently amongst the largest corporations in their country of operations and their top executives play an influential role in the political,social,and IV.The International Hierarchy of Decision Making cultural life of the country.Yet these people, Marshall,like Marx,thought that the "chief 1 Alfred Marshall.Principles of Economics (Mac- fact in the form of modern civilization,the kernel millan,1961),pp.7475

THE INTERNATIONAL FIRM 445 eting literature suggests new products typically follow a cycle known as trickle-down or two-stage marketing. An innovation is first adopted by a small group of individuals who act as opinion leaders and is then copied by others via the dem￾onstration effect. In this process, the rich get more votes than everyone else, first of all because they have more money, second of all because they have discretionary income and can afford to be experimental, and, third, because they have high status and are more likely to be copied. The prin￾ciple of consumer sovereignty cannot easily be applied to this process since, at most, only the special group in the first stage of the marketing process has something approaching a free choice. The rest have only the choice between conform￾ing or being isolated. In the international economy, trickle-down marketing takes the form of the international demonstration effect. Products are first intro￾duced in the United States or Europe and then spread to other countries. Multinational corpora￾tions speed up this process by making it easier to transfer new products and marketing methods to less advanced countries. One of the key motives for direct investment, cited by corporations, is to gain control over marketing facilities in order to facilitate the spread of their products. If firms were denied control over communication and mar￾keting facilities in the foreign countries and we had a regime of national firms (private or socialized) rather than multinational firms, the pattern of output would almost certainly be quite different than the one that is now observed. There would be more centers of innovation, and probably more variety of choices offered to the consumers, as each country developed products suited to its par￾ticular characteristics. Products from one country would spread to other countries either through trade or imitation but the movement would be coordinated by market competition rather than the planning decisions of top management in a few corporations whose interest it is to foreclose competition, to restrict the choices offered, and to insure the survival of their own organizations. It is difficult to speak with professional certainty in this badly neglected field, but it does not appear to be socially efficient to allow corporations to monopolize information on new possibilities cre￾ated by science. IV. The International Hierarchy of Decision Making Marshall like Marx, thought that the "chief fact in the form of modern civilization, the kernel of the modem economic problem . . ."' was the division of labor within the factory between those who plan and organize economic activity and those who work for them. In the modern corpora￾tion the hierarchical structure of command and authority has been greatly elaborated from the simple division between owners and workers in the Marshallian firm, but the tensions and conflicts of autocracy remain. They take on par￾ticular importance in the multinational corpora￾tion where problems of nationalism and problems of authoritarianism intertwine. Multinational corporations are torn in two di￾rections. On the one hand, they must adapt to lo￾cal circumstances in each country. This calls for decentralized decision making. On the other hand, they must coordinate their activities in various parts of the world and stimulate the flow of ideas from one part of their empire to another. This calls for centralized controls. They must therefore develop an organizational structure to balance the need to coordinate and integrate operations with the need to adapt to a patchwork quilt of lan￾guages, laws, and customs. One solution is divi￾sion of labor based on nationality. Day-to-day management in each country is left to nationals of that country who are intimately familiar with local conditions and practices and best suited to deal with local problems and local government. These nationals remain rooted in one spot, but above them is a layer of people who move around from country to country, as bees among flowers, transmitting information from one subsidiary to another and from the lower levels to the general office at the apex of the corporate structure. In the nature of things, these people, for the most part, will be citizens of the country of the parent corporation, just as we now find that the top ex￾ecutives of most of the major corporations in the United States are drawn from a relatively small homogeneous cultural group quite distinct from the population of the United States as a whole. This creates two types of problems. In the first place, there is the internal problem of creating in￾centives for foreigners whose access to the top corporate positions will be necessarily limited. The second problem is far more important and is in the nature of an external diseconomy. The sub￾sidiaries of multinational corporations are fre￾quently amongst the largest corporations in their country of operations and their top executives play an influential role in the political, social, and cultural life of the country. Yet these people. ' Alfred Marshall. Principles of Economics (Mac￾millan, 1961), pp. 74-75