Congressional Research Service: Report for Congress, No. 94-82 E January 30, 1994 -ti- Japan's Keiretsu: Industrial Groups as Trade Barriers By Dick K. Nanto Specialist in Industry and Trade Economics Division JAPAN'S KEIRETSU: INDUSTRIAL GROUPS AS TRADE BARRIERS SUMMARY A prominent feature of Japan's capitalism consists of families of companies called keiretsu that are linked by crossholdings of stock shares, intra-group financing, and certain coordinating mechanisms. Two types of keiretsu exist: large horizontally organized industrial conglomerates, such as Mitsubishi, Mitsui, and Sumitomo, and vertically integrated manufacturers, such as Toyota, Nippon Steel, and Matsushita Electric. They have become a contentious issue in U.S. trade negotiations with Japan for several reasons. These keiretsu, or industrial groups, first, tend to buy from within their groups and may discriminate against American or other exporters to Japan. One survey of purchases by the Big Six conglomerate keiretsu (which account for a fifth of Japan's corporate capital and sales) found that they made 68 percent of their purchases from companies in which they had at least a 10 percent equity interest and bought only 5 percent from foreign, unrelated companies. The Japanese government claims that keiretsu firms make only 15 percent or less of their purchases from companies within their groups. The true figure, however, is more than four times that amount (according to Japan's Fair Trade Commission). The second reason that the keiretsu have become an issue in U.S.- Japan trade negotiations is that strong keiretsu ties between buyers and suppliers tend to keep outsiders out and insiders in, not only in Japan, but in Japanese assembly plants located in the United States. Japanese transplant automakers, in particular, have relied heavily on their traditional Japanese suppliers who have followed them to their U.S. plants. Third, keiretsu ties may provide an advantage to Japanese companies in developing new technology or in long-term planning. Fourth, keiretsu distribution systems may discriminate against foreign producers in reaching the retail Japanese consumer. And fifth, keiretsu stockholding patterns make the buying and selling of Japanese companies, let alone hostile takeovers, nearly impossible. In the current U.S.-Japan Framework talks, the keiretsu issue is being negotiated with Japan under the category of "economic harmonization." On the U.S. side, the Treasury Department is the lead negotiating agency. Progress in these talks is to be reviewed at a series of bilateral summit meetings with the first scheduled for February 11, 1994. Congressional interest in this issue stems partly from its oversight responsibilities, partly from proposals to change U.S. antitrust laws, and partly from complaints from constituents who are competing with keiretsu organizations. A separate issue is whether the keiretsu represent an alternative form of capitalism that could be acceptable in the United States. If they are efficient, are there other impediments that would preclude their being adopted by American companies? TABLE OF CONTENTS BACKGROUND TYPES OF INDUSTRIAL GROUPS CONGLOMERATE KEIRETSU Presidential Councils Cross Shareholding Main Bank Ties Personnel Ties Trading Companies Joint Investment Companies Intra-group Buying In-group Purchases by Horizontal keiretsu VERTICAL KEIRETSU POLICY CONSIDERATIONS JAPAN'S KEIRETSU: INDUSTRIAL GROUPS AS TRADE BARRIERS A prominent feature of Japan's capitalism consists of families of companies called keiretsu(1) that are linked by crossholdings of stock shares, intra-group financing, and certain coordinating mechanisms. They have become an issue in U.S. trade negotiations with Japan for several reasons. 1. The keiretsu (kay-ret-sue) also are referred to as zaibatsu (financial cliques). Zaibatsu, however, has a negative connotation and usually refers to Japan's prewar industrial combines characterized by holding companies. At the end of World War II, Japan's four largest zaibatsu controlled about a quarter of the paid-in capital of Japan's incorporated business. (See: Hadley, Eleanor M. Antitrust in Japan. Princeton, Princeton University Press, 1970.) These keiretsu, or industrial groups, first, tend to buy from within their groups and may discriminate against American and other exporters to Japan. Second, the strong keiretsu ties between buyers and suppliers tend to keep outsiders out and insiders in, not only in Japan, but in Japanese assembly plants located in the United States as well. Third, keiretsu ties may provide an advantage to Japanese companies in developing new technology or in long-term planning. Fourth, keiretsu distribution systems may discriminate against foreign producers in reaching the retail Japanese consumer. And fifth, keiretsu stockholding patterns make the buying and selling of Japanese companies, let alone hostile takeovers, nearly impossible. In the current U.S.-Japan Framework talks, the keiretsu issue is being negotiated with Japan under the category of "economic harmonization." On the U.S. side, the Treasury Department is the lead negotiating agency. Progress in these talks is to be reviewed at a series of bilateral summits held between the two nations; the first is scheduled for February 11, 1994. The keiretsu also were one of the targets under the Structural Impediments Initiative talks and follow- up actions between the United States and Japan from 1989-1992. Congressional interest in this issue stems partly from its oversight responsibilities, partly from proposals to change U.S. antitrust laws, and partly from complaints from constituents who are competing with keiretsu organizations. A separate public policy issue, however, is whether the keiretsu represent an alternative form of capitalism that could be acceptable in the United States in order for businesses to be more internationally competitive. If they are efficient, are there other legal impediments that would preclude their being adopted in the United States? BACKGROUND Japan's keiretsu developed partly in response to intense domestic competition among Japanese businesses but also from the hierarchal nature of Japanese society. These business groupings appear much like large industrial groups in other nations, but are distinctive in several ways. They consist of either vertical or conglomerate (horizontal) groupings of companies that are characterized by long-term association, cross-holdings of stock, exchanges of personnel, intra-group financing, extensive intra-group business dealings, and, sometimes, sharing of company name. The keiretsu reflect basic social patterns throughout Japanese society. Most of Japanese human interaction is based on hierarchical-vertical social structure combined with horizontal group cohesiveness. If America is seen as a society that operates on principles of egalitarianism and open opportunity, Japan is seen as a society that operates on principles of hierarchy and superior-inferior relationships. Japanese society resembles a mountain range of pyramids in which everyone has a niche. The "tyranny" of superior-inferior relationships, however, is tempered by strong horizontal ties among people who share similar ranks both within and without the hierarchies.(2) 2. Nakama, Ishiki. Nihonjin no Nakama Iishi (The collegial consciousness of Japanese). Tokyo, Kodansha, 1976. The keiretsu, per se, do not violate Japan's antitrust laws, but their activities can. Close buyer-supplier ties can lead to unfair trade practices such as participation in boycotts, refusals to trade, resale price maintenance, monopolization of markets, and other illegal activities. Horizontal collusive activities, however, cannot occur within a keiretsu that has but one firm in a particular market. Such collusive activities are more likely to occur in industrial associations or other horizontal organizations to which the keiretsu companies belong. The keiretsu were one of the targets of the Structural Impediments Initiative talks and follow-up actions between the United States and Japan in 1989-92. The United States claimed that the close links among Japanese corporations can "promote preferential group trade, negatively affect foreign direct investment in Japan, and give rise to anticompetitive business practices."(3) 3. Comments of the U.S. Delegation on the Interim Report by the Japanese Delegation. Appended to Japan-U.S. Structural Impediments Initiative, Interim Report by the Japanese Delegation. April 5, 1990. Released by the White House, Office of the Press Secretary, April 5, 1990. The United States also has claimed that the industrial groups can hinder access by U.S. firms to Japanese markets and can allow keiretsu companies to generate high profits in protected markets at home, thereby enabling them to shave profit margins in order to gain market share abroad. The long-term, buyer-supplier relationships also can lock out foreign suppliers, even those with competitive products, while the supplier-distributor links can prevent retailers from carrying competing products and can hinder price competition. The cross-holdings of shares also can impede foreign acquisitions of Japanese companies and make trading in stocks of certain companies thin. Many Japanese see the keiretsu as a natural outgrowth of their unique economic development and one of their strengths in international competition. Along with the elite government ministries, the core companies of the keiretsu are the first choice for employment among Japan's top college graduates. Japanese also point out that Germany has similar business organizations. Hence, they claim that the United States, not Japan, is out of step with the rest of the world.(4) 4 Russell, David. America's Hollow Victory. Business Tokyo, v. 4, June 1990. p. 34. TYPES OF INDUSTRIAL GROUPS More than two million non-financial corporations exist in Japan. As in other market-oriented, industrialized economies in the world, these corporations come in all types and sizes. Small businesses with just a handful of employees exist alongside of corporate giants whose names are familiar to consumers on all continents. On the surface, Japanese corporations fit the Western pattern. They issue stock, have boards of directors, generate profits, and pay dividends. Behind the outward form, however, are organizations that differ in ways that affect many of the operational decisions that frustrate foreign firms operating in or exporting to Japan. Japan's keiretsu come in two basic types. Japanese refer to them as vertical and horizontal. Vertical groups are similar to those found in other industrialized nations. They consist primarily of a dominant manufacturing firm, such as Toyota, with related firms in the supplier and distribution chain. The members of the firm come primarily from one sector, so the relationships are vertical. The vertically integrated groups include at least 39 blue chip manufacturers, such as Nippon Steel, Toyota, and Matsushita Electric, each with dozens of companies associated with it. These groups resemble the business empires found in all industrialized nations of the world. There is considerable debate in the literature over what to call the horizontal keiretsu. Even though the Japanese call them horizontal, in the antitrust definition of the word they are not. Horizontal ties usually refer to links among firms producing similar products, such as Chrysler's sales of cars made by Mitsubishi Motors. The Japanese horizontal keiretsu span the market and have firms in many different industries. In what Japanese call the one-set principle, each group tries to maintain one and only one company in each type of business. One of the unwritten rules of keiretsu behavior is that one firm in the group does not encroach upon the market of another. These horizontal keiretsu, therefore, have been referred to as intermarket keiretsu, as konzern, as diversified business groups, the financial keiretsu, or simply the Big Six keiretsu. The term "intermarket" emphasizes the fact that these groups span many sectors and markets. The term, conglomerate keiretsu, is perhaps the most intuitive and will be used interchangeably with horizontal in this report. The horizontal keiretsu are conglomerates because they comprise firms from many markets and are linked together primarily by stockholding and other financial arrangements. The conglomerate keiretsu often include member firms who are themselves vertical keiretsu. For analytical purposes, the Fair Trade Commission in Japan lists six large business groups as conglomerate keiretsu. These six are divided further into two groups. The first has origins in the prewar zaibatsu (industrial combines) --Mitsubishi, Mitsui, and Sumitomo--and the other consists of three bank centered groups--Fuyo (Fuji Bank), Sanwa (Sanwa Bank), and DKB (Dai-ichi Kangyo Bank). CONGLOMERATE KEIRETSU Each conglomerate keiretsu consists of core members (whose presidents meet in councils) and other members that may be smaller or more loosely connected to the group. Even though the core members represent only a fraction of all related keiretsu firms for each of the Big Six business groups, the extent of their market coverage is remarkable. Of the three groups descendant from the prewar zaibatsu, the 29 core firms in Mitsubishi operate in 19 of Japan's 23 major industrial sectors. The 24 Mitsui enterprises covered 15 industrial sectors, while the 20 Sumitomo firms operated in 13 sectors. For the three bank-centered groups, the 29 Fuyo core enterprises operated in 17 industrial sectors. The 44 Sanwa firms and the 47 DKB firms each operated in 18 industrial sectors. (See figure 2.) [PLEASE CONTACT GATEWAY JAPAN FOR THIS FIGURE] In particular industries, the keiretsu influence is remarkable. In chemical products, for example, Mitsubishi has 6 firms; Mitsui has 3, Sumitomo 2, Fuyo 3, Sanwa 8, and DKB also 8. In electrical machinery, Mitsubishi has Mitsubishi Electric; Mitsui has Toshiba, and Sumitomo has NEC. Each of the conglomerate keiretsu includes many more firms than the members of the core. Including subsidiary companies in which the core companies have an equity stake of 50 percent or more brings the number of firms in the Big Six to 763 for Mitsui (in 1989), 1,004 for Mitsubishi, 566 for Sumitomo, 873 for Fuyo, 1,604 for Sanwa, and 1,745 for DKB for a total of 5,980 firms.(5) Including the affiliated firms in which the core companies own equity shares of 10 to 49 percent, the number of firms in the Big Six doubles to more than 12,000.(6) 5. Japan Fair Trade Commission. Nihon no Roku Daikigyo Shudan. Tokyo, Toyo Keizai Shimposha, 1992. p. 139. (Hereafter referred to as Roku Daikigyo.) 6. JFTC, Kigyo Shudan (1989), p. 36. Even counting the subsidiaries and affiliated firms, the Big Six comprise only a small percentage of the more than 2 million enterprises in the whole Japanese economy. In terms of total capital and assets, however, the Big Six are large indeed. In fiscal year 1989, the 193 core companies accounted for 17 percent of the total corporate capital in Japan, 14 percent of total assets, and 16 percent of total sales. The group of 5,980 firms, which included subsidiaries with at least 50 percent equity investment by the core companies, accounted for 21 percent of total corporate capital, 18 percent of total assets, and 20 percent of total sales. In short, these six keiretsu represent more than a fifth of Japan's economy, and their share has been rising.(7) 7. JFTC, Roku Daikigyo (1992), p. 139. The reach of the Big Six worldwide should not be underestimated. Mitsubishi, for example, has 581 overseas offices, 140 overseas factories, 2,792 persons dispatched from Japan, and employs 74,417 local people in its overseas operations. Mitsubishi has favored other Asian nations (28,642 local employees) and North America (24,769 local employees) for investment, but its interest in Europe (7,814 local personnel) and South America (6,522 local personnel) is rising.(8) 8. Ohsono, Tomokazu. Hitome de Wakaru Kigyo Keiretsu to Gyokai Chizu (Charts to understand at a glance enterprise keiretsu and industrial circles). Tokyo, Nihon Jitsugyo Shuppansha, 1991. p. 19. Both the six conglomerate keiretsu and the many vertical keiretsu use a variety of methods to tie their enterprises together. These include: * presidential councils * crossholdings of shares * intra-group financing by a common bank * dispatching of executives and mutual appointments of officers * use of trading companies for marketing and organizing projects * joint investments in new industries. Presidential Councils The presidential councils comprise the presidents of the leading companies of the group, who meet periodically (usually monthly) to discuss matters of mutual interest. While the councils claim not to be policy making bodies for the group (as were the prewar holding companies), they do discuss such topics as economic and financial conditions, promising business activities, research and development, intra-group trademarks, and labor problems. They also can decide on joint investments in new industries, political contributions, public relations, rehabilitation of troubled member companies, and key personnel appointments.(9) 9. Dodwell Marketing Consultants. Industrial Groupings in Japan 9th Edition, 1990/91. Tokyo, Dodwell Marketing Consultants, 1990. p. 9. The importance of the presidential councils appears to be diminishing, since in the 1960s such councils met as often as weekly. The presidential councils do not violate antitrust laws prohibiting collusion because the companies involved represent different industries. Cross Shareholding Reciprocal holding of stock shares by firms is a major device used to cement relations across the keiretsu. This mutual shareholding stems from two major factors. First, as Japan liberalized its capital markets in the 1960s and 1970s, companies began to fear hostile takeovers from abroad. With government encouragement, they protected themselves by having friendly companies serve as stable stockholders. An amendment to Article 280 of Japan's Commerce Law allowed the board of a company to increase its capital by assigning new shares to other firms without formal approval from existing shareholders--in essence, companies simply exchanged new issues of stock. About 60 percent of Japan's outstanding company shares now are held by other companies. Most are considered to be friendly and stable. Foreign or other outsider companies, therefore, find it difficult to acquire enough shares of a company to mount a takeover, be it hostile or friendly. Second, Japan's antimonopoly law prohibits holding companies. The cross shareholding substitutes for vertical shareholding possible through holding company structures prevalent in other countries. The extent of stock crossholdings among the conglomerate keiretsu ranges from about 15 to 35 percent of total paid-up capital for the companies who form the core of the business groups. (See figure 3.) [PLEASE CONTACT GATEWAY JAPAN FOR THIS FIGURE] In FY 1989, for example, 35.4 percent of total shares for the 29 core members of the Mitsubishi Group were held by other members of the group. These percentages, however, represent totals for the group as a whole. Individual firms will hold a smaller percent of the stocks of the group or a specific member of the group. For the Mitsubishi Group, for example, one group member, on average, held only 1.8 percent of the stocks of the rest of the group. By law, a city bank may not hold more than 5 percent of a company's stock shares. Insurance companies are allowed to hold a higher 10 percent. The total shares of a specific keiretsu company held by all other group members varies from about 10 to 60 percent. Typically, about a quarter or third of a group company's shares will be held by other group members. The purposes of the mutual holdings of stock include cementing relationships and precluding hostile takeover attempts. They also provide a method of raising funds without diluting management control, since stable stockholders are expected to maintain their shares of the total when new stock is issued. The holdings of stock are rarely sold.(10) The direct financial impact of most of the cross shareholdings is subtle. Income from stocks sold to stable shareholders often is turned around and used to buy stocks in those same companies. Since some of the proceeds from stock issues go toward buying stocks of companies with which a company has cross shareholding arrangements, Japanese companies have to issue more new stock than is needed to raise a certain amount of funds. They also tend to rely more on banks and retained earnings for new capital. The mutual share holdings also reduce or even eliminate pressures on companies to produce short-term profits, since stable shareholders generally do not sell their holdings even when dividends are down. 10. Since most companies carry these stocks at their historical value, many Japanese companies have balance sheets in which net worth is considered to be understated. The decline in the stock market and recessionary economic conditions in the early 1990s, however, has been putting pressures on companies to sell some of their holdings of stock. U.S. critics of the keiretsu argue that the cross shareholdings affect the way that companies deal with each other, although many Japanese claim they have little real effect. Cross shareholdings among firms serve both as a symbol and financial incentive to do business with each other. If these cross-holdings had no meaning, why would such arrangements be maintained, particularly when Japanese firms are hungry for capital? Japanese authors, however, argue that the small percentages owned by one company in the stocks of another and the fact that they are reciprocal mean that the companies neither have the incentive nor the leverage to exert influence over each other.(11) 11. Odagiri, Hiroyuki. Growth Through Competition, Competition Through Growth. Oxford, Oxford University Press, 1992. p. 179-80. If the cross shareholdings of stock were the only tie among keiretsu companies, perhaps their significance would be minor. Cross shareholdings, however, are only one of several links. While no single firm with less than 5 percent interest in another is likely to be able to influence that company's purchasing or investment decisions, the mutual shareholdings combined with other ties make a combination that no member of a keiretsu is likely to ignore --neither would it be prudent to do so. Another way to look at the cross holdings is that the stable shareholders provide protection against outside intervention. Stable shareholders, under usual circumstances, can be counted on not to intervene directly in the management decisions of other member companies. Outside shareholder interests, however, tend to receive little attention (and sometimes are blatantly suppressed-- for example by hiring gangsters to quell dissent at stockholder meetings). Another effect of the interlocking shareholding is that the stable shareholders can help to prevent large price fluctuations in the stock, although they cannot reverse large market trends. When the market begins to move either up or down, stable shareholders can be counted on not to jump into the market and either acquire or dump shares. In some respects, however, this can also exacerbate fluctuations in share prices. Since less than half of all shares are available to be traded at any given time, markets for the shares of some companies can be quite thin, and small movements exaggerated. The cross shareholdings also may undervalue assets of a company. Since stocks usually are reported on balance sheets in terms of their acquisition value, large blocks of stock can be carried on balance sheets at the face value of 60 yen per share (at which they were originally issued). Calculating the net worth of a keiretsu company, therefore, can be difficult. Main Bank Ties Banks also play a central role in the conglomerate keiretsu. Of the six large groups, three are centered on banks while three are derived from former prewar zaibatsu and include a bank as a central member. The banks provide financing and hold blocks of shares of member companies. At one time, zaibatsu banks were referred to as "organ" banks or an integral part of the organization.(12) 12. Hadley, Eleanor. Antitrust in Japan. Princeton, Princeton University Press, 1970. p. 157. The companies in each group exchange information with the group bank and maintain large deposits there just to maintain satisfactory relationships. They also, however, make such deposits with other banks, just to ensure that ample credit will be available during periods of tight money and that no single bank will exert undue influence on their corporation.(13) Among the bank- centered industrial groups, nearly half of the member firms do not rely on the keiretsu bank for the largest share of their loans. 13. Abegglen, James C., and George Stalk, Jr. Kaisha, The Japanese Corporation. New York, Basic Books, 1985. p. 165-166. A company's main bank, however, will usually take the lead in arranging loan consortia among many banks and is responsible for monitoring the performance of those loans. If the borrowing company defaults on a loan, the main bank traditionally has last claim on any associated collateral. A company's main bank, therefore, will demand detailed information on the operations of a borrowing company and may step in to manage the company in particularly troubling circumstances. Main banks play an important role as a manager of last resort.(14) 14. See: Nakatani, Iwao. The Economic Role of Financial Corporate Grouping. In M. Aoki, ed.. The Economic Analysis of the Japanese Firm. Amsterdam, Elsevier, 1984. p. 241. In 1989, the Mitsui Bank served as the main bank for 109 corporations who on average relied on it for 25 percent of their loans. Likewise, Mitsubishi Bank was the primary bank for 156 firms and held an average of 28 percent of their loans.(15) 15. The averages for Sumitomo Bank were 28 percent for 138 firms, for Fuji Bank 25 percent for 160 firms, for Sanwa Bank 28 percent for 124 firms, for Dai-Ichi Kangyo Bank 27 percent for 176 firms. Trust and insurance companies, such as Sumitomo Trust Bank with 27 percent and 29 firms also acted as main banks. Institutions outside of the Big Six, such as the Industrial Bank of Japan with 32 percent and 160 firms also were important. Toyo Keizai Shimposha. Kigyo Keiretsu Soran (Overview of enterprise keiretsu) (quoted on p. 89 of Roku Daikigyo (1992).) The Japanese main bank system has been suggested as a solution to the monitoring problem pointed out by Adolf Berle and Gardiner Means in their 1933 study, The Modern Corporation and Private Enterprise. In American corporate governance, diverse and atomistic shareholders lack the ability, skill, and information to monitor the performance of specialized managers. Japan's main banks, however, hold substantial equity positions in corporations, speak with their own authority, and act as the delegated monitor for other banks lending to their client corporation. In essence, a main bank can perform some of the same sanctions of an outside board of directors or as a "takeover" market without some of the attendant costs.(16) 16. Gilson, Ronald J., and Mark J. Roe. Understanding the Japanese Keiretsu: Overlaps Between Corporate Governance and Industrial Organization. Yale Law Journal, v. 102, January 1993. p. 873-880. Does the main bank system provide and unfair advantage to keiretsu firms? A surprising result of research on keiretsu bank relationships during the 1960s was that keiretsu firms on average paid significantly higher rates of interest than other firms. The banks who are the leaders of the groups were able to collect additional interest of 0.2 to 0.3 percentage points on their loans to group members.(17) The higher interest rates, however, could have been compensation for the lower dividend payouts for the shares crossheld. During credit crunches, moreover, member firms probably receive preferential access to available funds. They also can receive favorable terms of repayment and extensions, if necessary. 17. Caves, Richard and Masu Uekusa. Industrial Organization. In Hugh Patrick and Henry Rosovsky. Asia's New Giant. Washington, Brookings, 1976. p.503. Another important role of the main bank is that it serves as a center of intelligence for the industrial group. In the process of making loans to member companies, banks amass considerable knowledge about the operations of companies, sectors, and national markets. It is no wonder that they take the lead in the keiretsu groups. American banks, on the other hand, are prohibited from holding stocks in and managing companies. Some question whether this places them and U.S. manufacturers at a disadvantage when competing with their Japanese (or European) counterparts. Personnel Ties While much emphasis is placed on the financial relationships among the keiretsu firms, personnel ties actually can have greater operational importance. It is in personnel ties that the hierarchy of relationships becomes apparent and harmonization of policies can be achieved. At the lower levels of personnel hierarchies, companies within industrial groups regularly exchange personnel. At higher levels, the leading companies in industrial groups, particularly the banks and trading companies, dispatch executives or retirees to other member companies. In 1989, about 60 percent of the companies in the Big Six had executives sent by other member companies of the same group. In the DKB group, for example, 48 out of 81 companies had members of their boards of directors from the Dai-ichi Kangyo Bank.(18) 18. Dodwell, Industrial Groupings, p. 432-454. In 1989, in the Big Six conglomerate keiretsu, over half of the executives in the Big Six industrial groups who had been dispatched to other member firms came from banks. Another 40 percent came from general trading companies.(19) 19. JFTC, Roku Daikigyo (1992), p. 81. Trading Companies Along with the banks, the general trading companies play a leadership role in most conglomerate keiretsu.(20) These huge companies operate diverse businesses on their own while providing many services to member firms. They procure raw materials, distribute products, finance some activities, organize diverse projects, and gather and disseminate intelligence. Since trading companies are involved in both importing and exporting, they can absorb considerable foreign exchange risk for the group. The trading company usually is considered to be the lead company or shares leadership with a bank or other major company in the group. 20. General trading companies are referred to as Sogo Shosha in Japanese. TABLE 1. General Trading Companies Associated with Japan's Conglomerate Keiretsu Industrial Group Trading Companies Mitsubishi Mitsubishi Corp. Mitsui Mitsui & Company Sumitomo Sumitomo Corp. Sanwa Nissho Iwai Corp. Nichimen Corp. Fuyo Marubeni Corp. Tokai Toyo Menka Kaisha DKB Itochu Co. Kanematsu Corp. Trading companies, moreover, engage in transactions not only with other Japanese firms but among buyers and sellers in third countries. Such transactions might include, for example, arranging for a sale of a U.S. chemical plant to Russia.(21) In the early 1980s, Japanese trading companies handled as much as 10 percent of all U.S. exports.(22) 21. Young, Alexander K. The Sogo Shosha: Japan's Multinational Trading Companies. Boulder, Colorado, Westview Press, 1979. p. 9-10. 22. Yoshino, M.Y., and Thomas B. Lifson. The Invisible Link, Japan's Sogo Shosha and the Organization of Trade. Cambridge, Mass., The MIT Press, 1986. p. 2. The role of trading companies in Japan's foreign trade, however, is diminishing rapidly. In 1985, the nine leading trading companies handled 44 percent of Japan's exports and 72 percent of its imports. By 1991, however, the shares had plummeted to only 13 percent of Japan's exports and 17 percent of its imports.(23) Some of this was due to a drop in the volume of transactions handled by trading companies, but most can be traced to more international transactions being handled directly by manufacturers and other distributors. 23. Keizai Koho Center. Japan 1990, An International Comparison. Tokyo, Keizai Koho Center, 1989. p. 46. And Japan 1993, An International Comparison. Tokyo, Keizai Koho Center, 1993. p. 45. Still, general trading companies wield considerable market power. Through control of key ports and shipping facilities in Japan, they can exert pressure on member companies in their buying and selling decisions and can hinder U.S. exports. Until the 19908 when such a high proportion of Japan's imports were concentrated in the hands of a few firms, moreover, the government was better able to exert "administrative guidance" to dampen imports of particular goods. This has occurred in steel and textiles.(24) 24. Lincoln, Edward J. Japan's Unequal Trade. Washington, Brookings Institution, 1990, p. 88. A trading company also can be a gatekeeper for sales to other firms of an industrial group. Having to go through the trading company adds to the cost of delivering the product but shifts some of the risks associated with the transaction from the foreign supplier to the trading company. Like the banks, trading companies form an important center for intelligence and information, particularly on markets. The large trading companies attract top graduates from universities each year who then are useful in interacting with government ministries and other top-level firms. Joint Investment Companies Another function of the industrial groups is to reduce the risk of investment in new business operations, in developing new technology, or in restructuring firms facing declining markets. These may be an agenda item at the presidential councils. Joint ventures, in particular, have become a favored method of developing new technology. The nature of high technology today, however, often requires expertise from outside the group. From April 1988 to December 1990, the Big Six groups established 761 joint investment companies. Only 6 percent, however, were of firms only within a single group. The vast majority, 77 percent, represented ventures with other Japanese companies, while 7 percent were with foreign companies.(25) 25. Japan Fair Trade Commission. The Outline of the Report on the Actual Conditions of the Six Major Corporate Groups. FTC/Japan Views, No. 13, June 1992, p. 34. Intra-group Buying U.S. businesses have charged that the conglomerate keiretsu discriminate against outsiders and prefer to buy from other member companies. Even competitive U.S. companies find it difficult to sell to keiretsu group companies that have ties to a competing supplier. Several economic studies have examined the effects of keiretsu on imports and exports. A widely quoted study by Robert Lawrence of Harvard University written while he was at the Brookings Institution concluded that both vertical and horizontal keiretsu relationships tend to reduce Japanese imports, although the horizontal keiretsu work primarily through inhibiting entry rather than by increasing efficiency--as is the case with vertical keiretsu. He also estimated that eliminating the keiretsu would increase Japan's manufactured imports by $30 billion from a 1985 base.(26) A study by Marcus Noland of the U.S. Council of Economic Advisers concluded that Japan's keiretsu were associated with higher than expected net exports and lower than expected imports. He noted, however, that it was unclear whether the lower imports were due to discrimination or enhanced efficiency.(27) 26. Lawrence, Robert Z. Efficient or Exclusionist? The Import Behavior of Japanese Corporate Groups. Brookings Papers on Economic Activity, No. 1, 1991. p. 311-341. This was critiqued in: Uryu, Fujio, et. al. Keiretsu wa Heisateki Ka: Ro-rensu Ronbun no Hihanteki Kento (Are keiretsu closed natured?: a critical examination of the Lawrence thesis). Japan, Research Institute of International Trade and Industry, Discussion Paper #92-DOJ-38, June 1992. 44 p. 27. Noland, Marcus. Public Policy, Private Preferences, and the Japanese Trade Pattern. Working paper, Council of Economic Advisers, Washington, D.C. August 1993. p. 22. These and other studies have been criticized by other economists. Paul Sheard of the Australian National University asserts that the econometric findings of such studies are inconclusive and their interpretation unclear. He emphasizes that imports could be low in keiretsu-dominated industries because of competitive corporate advantage by Japan's companies and not because of exclusionary or anticompetitive practices by keiretsu companies.(28) Masaru Yoshitomi, formerly of Japan's Economic Planning Agency, makes similar arguments.(29) David Weinstein of Harvard University, through an econometric analysis of keiretsu financial data concludes that the close ties between banks and keiretsu companies have induced the companies to expand production (through borrowing from banks) beyond what would be warranted by pure profit maximization. This, he concludes, may explain why entry into markets with strong keiretsu presence is often described as difficult.(30) 28. Sheard, Paul. Keiretsu and Closedness of the Japanese Market: An Economic Appraisal. Discussion Paper No. 273, The Institute of Social and Economic Research, Osaka University. June 1992. p. 19-21. 29. Yoshitomi, Masaru. Keiretsu: A Legitimate, Viable Capitalist Approach. The International Economy, vol. 3, May/June 1991. p. 68-75. 30. Weinstein, David E. and Yishay Yafeh. Japan's Corporate Groups: Collusive or Competitive? An Empirical Investigation of Keiretsu Behavior. Draft paper. Harvard University Department of Economics. November 2, 1992. 22 p. The econometric evidence, therefore, indicates that access to markets that have strong keiretsu presence is less than seems warranted by economic factors. Researchers do not agree, though, on whether that is caused by collusive or anticompetitive practices or simply by more competitive companies fostered by their keiretsu relationships. In-Group Purchases by Horizontal keiretsu The Japanese government response to this debate on the effect of the keiretsu on imports has been to emphasize that it does not matter which side is correct because the results of buying activities of the keiretsu firms are not discriminatory. Based on survey data, it asserts that on average conglomerate or horizontal keiretsu companies make only about 15 percent or less of their purchases with each other. If such intra-group purchasing is so low, how can it be discriminatory? The keiretsu purchasing survey was conducted by Japan's Fair Trade Commission (JFTC, Japan's antimonopoly agency) in 1992. The JFTC surveyed the buying and selling patterns of the Big Six conglomerate keiretsu to determine whether the results of their transactions show a pattern that favors group member firms over outsiders. This survey showed that the Big Six made only 16 percent of their purchases from other members of the industrial group. This is the basis for the assertion that the keiretsu are open to buying from the outside. Indeed, one can hardly argue that the Mitsui group, for example, tends to be closed to outsiders when it relies on other group members for only 8 percent of its purchases. This low percentage, however, can be quite misleading. It is low precisely because the number of firms being considered in the sample is low. In the Mitsui case, the 8-percent level for intra-group purchases applies only to the 24 core companies that belong to its presidents council. In addition to these, the Mitsui group held equity interests of at least 10 percent in 763 firms. It also had a number of overseas subsidiaries and joint ventures. When these firms are added to the core group being considered, the actual extent of intra-group purchasing becomes apparent. Table 2, shows the complete results on intra-group purchases reported by the JFTC. The JFTC asked each group company for its top 30 suppliers and amounts purchased from them. The purchases were then categorized as to whether they came from companies within the enterprise group, from related companies in which the group holds an equity interest, from other domestic firms, and from foreign firms.(31) 31. This method may understate purchases from foreign firms that would tend to be smaller suppliers and may not be included in the top 30 supplier firms. The astounding conclusion from this survey is that if you are a foreign firm without an equity relationship with a keiretsu, you have only one chance in twenty of selling to one. Conglomerate keiretsu firms on average made only 5 percent of their purchases from unrelated foreign companies. They made 68 percent of their purchases from companies within their group or with which they had an equity interest of at least 10 percent. The keiretsu also buy a considerable amount of imports or foreign products. On average, they bought 28 percent from foreign-related companies in which they had an equity interest. The problem is that they bought only 5 percent from foreign companies in which they had no equity interest. They also bought about 6 percent from competing keiretsu groups. In short, if a firm does not have an equity relationship with a Big Six company, chances are that it will not be doing much business with that company. It will be treated no worse, however, than a competing keiretsu company. The message of the data is clear. If a firm wants to deal with keiretsu companies, it must establish a long-term relationship with them. This long-term relationship may require the purchase or exchange of stock. Those firms without long-term relationships are likely to be relegated to the category of marginal suppliers. These data on actual transactions by members of enterprise groups seem to contradict what purchasers are reported to say they do. In a 1985 survey of Japan's machinery manufacturers by Japan's Ministry of International Trade and Industry, for example, 95.1 percent of the respondents said they would pick the superior good whether from within-group or imported, while 2.0 percent favored in-group goods even if imports were superior.(32) Companies claim that they do not discriminate against companies outside their keiretsu groups, yet their purchase patterns indicate that they do. A possible explanation for this apparent contradiction may be that stable relationships are considered to add to the quality of the product (through less risk and easier after market service) or that the close ties result in direct participation in product design which then is reflected in the final quality of the product. 32. Keizai Koho Center. Trading with Japan. Tokyo, Keizai Koho Center, 1985. p. 22. VERTICAL KEIRETSU In terms of the vertical keiretsu, in-group purchasing is even more pronounced. Major manufacturers in Japan maintain "families" of suppliers/buyers within their industries both upstream and downstream from the major keiretsu manufacturer. These firms also are connected by cross-shareholding of stock, personnel ties, and long-term buyer-supplier relationships. Many also control distribution chains. The supplier participates actively with the final manufacturer in designing products, upgrading technology and manufacturing processes, and implementing quality control. The buyer usually is allowed to examine the supplier's books, and cost savings generally are passed on to the final manufacturer to be incorporated into the retail price of the product. The supplier is an integral link in the competitive strategy of a Japanese manufacturer. This vertical keiretsu structure is a strategic choice of business organization. When forming a buyer-supplier relationship, a firm may internalize the supply source by forming a wholly owned subsidiary or it may rely on spot-market relationships in which a supply contract is awarded to the lowest bidder. It also may pursue something in between. One automaker, for example, may internalize the entire supply process from the iron mine to the smelter, to the mill, to the machining centers, to the fabricators, to the final assembly process, to the distributors, and then to the car dealers. Another might internalize only the distribution system and rely on independent suppliers for most inputs except for those with peculiar technology. A keiretsu structure is a hybrid of the two extremes. It is a matter of choice that is driven partly by market forces but also by management preferences. Toyota, for example, may buy most of its parts from its family of suppliers, while General Motors may buy most of its parts from its internal divisions. Keiretsu structures have worked well for Japanese industry partly because of historical development and partly because for certain industries, it can be a superior way to structure business ties. The industries in which a keiretsu structure seems to work the best are those in which technology or design are constantly changing and competition in the end product is intense. A keiretsu structure also is found in mature industries in which demand is stagnant and technical change small. The reason is that in such industries, firms seek ways to hold on to customers and to prevent debilitating price competition. A keiretsu structure can help stabilize an industry. The keiretsu type of organization has other advantages. The close links also substitute for legal work. In Japan, procurement contracts often do not contain the detailed contingency clauses common in American legal documents. If a problem arises, the relationship of mutual trust allows the companies to work out a satisfactory solution. The long-term nature of the relationship, moreover, means that if one side has to take a loss because of unforeseen difficulties, it may be favored the next time a problem arises. Hence, a kind of equity can be attained. Vertical keiretsu also play an important role in solving the employment problem of Japan's manufacturers. In Japan, subsidiaries and suppliers usually are required to accept retiring (voluntary or forced) employees from the lead manufacturer. This also can help the supplier, since the retiree usually turns around and deals with people in the parent company whom he formerly supervised. Such personnel transfers add to the difficulty of an outsider firm to break into a keiretsu buyer-supplier relationship, particularly if the firm does not have a substantial Japanese presence. Since suppliers are expected to accept retiring or surplus personnel from the buying company, a foreign firm without managerial positions in Japan has no such positions to offer. Japanese manufacturers also do not change suppliers without first consulting existing ones. If a competing supplier comes in with a lower price or new product, the existing supplier often is given a chance to match it. U.S. automotive parts suppliers also have complained that they cannot even get specifications for parts from Japanese automakers. They are sometimes told that they have to enter the process earlier. The existing suppliers have already long been involved in developing those specifications and manufacturing processes. Vertical keiretsu companies, however, also can get stuck with inefficient suppliers who cannot be circumvented easily. Just-in-time delivery also is usually required in keiretsu supplier contracts. Parts must be delivered as they are needed on the assembly line. This means that the supplier may have to make several small-lot deliveries at specific times each day. Also, manufacturers usually require their suppliers to insure that their products are 100 percent defect-free. Such parts can be used without being reinspected. In an October 1990 meeting called by Toyota Motors with its foreign suppliers, it indicated that it was finding one defect per 1,000 parts received from foreign auto partsmakers. This was considered to be too high compared with the performance of Japanese suppliers who averaged one defect per 100,000 parts.(33) 33. Sanger, David E. U.S. Suppliers Get a Toyota Lecture. New York Times, November 1, 1990. p. D1, D8. Such exacting requirements on the supplier mean that the buyer and supplier must have a relationship that goes beyond that typical of spot market purchasing or an open bidding process. There must be trust, loyalty, a mode of operation that allows for problems to be worked out in a mutually satisfactory manner, enough confidence in the relationship that the supplier is willing to invest in new technology, and a sharing of production and cost data that normally might be considered proprietary. Such relationships are difficult to cultivate without closer ties than those developed through arm's length transactions. Hence, in Japan vertical keiretsu have developed. In Western legal and economic literature, such transactions are called "relational contracting" as opposed to spot market transactions or internal sourcing. There is no a priori reason to expect that one type of buyer-supplier relationship is superior to another. Much depends on the type of product in question. If the product is standardized and requires minimal investment in specialized machinery or processes, a spot market obviously may work very well. Several suppliers may be capable of providing high quality nuts or bolts or rolled steel. If the product requires a fair amount of investment in design and machinery by the supplier, however, relational contracting or a keiretsu-type organization may be more efficient. The reason is that without a promise of long-term contracts and trust, the supplier may not be willing to make the investments needed. Also without such long-term relationships, the buyer may not be willing to disclose important future plans. The problem with long-term relationships, however, is that despite their efficiency, they may not be equitable. This may particularly apply to cases in which potential suppliers are international and the final product is traded across borders. The complaints of outsiders, not just foreigners but Japanese companies who are not members of the privileged group of suppliers, is that breaking into existing buyer-supplier relationships is nearly impossible in the short term. One American representative of a top U.S. telecommunications equipment and software supplier said that he often has the impression that potential Japanese buyers have already made up their minds not to buy the product even before it is shown. At times, his company cannot even get past the gate keepers.(34) In the long term, however, more competitive foreign firms may make it into the design-in process (i.e., participate in designing the products to be supplied) to become a stable supplier. 34. Statement by Glenn Fukushima in a seminar on keiretsu sponsored by the American Chamber of Commerce in Japan. October 12, 1992. Tokyo, Japan. Another problem is that Japanese automakers locating assembly plants in the United States have brought their own suppliers with them. Critics accuse these automakers of transplanting their keiretsu organization to American soil. Traditional American autoparts makers have not benefitted as much as anticipated because keiretsu parts makers with long ties to those Japanese automakers have followed them to the United States. POLICY CONSIDERATIONS No comprehensive U.S. policy toward the keiretsu has yet been developed. Rather, strategies and policies have been directed at specific issues or problems. In order to overcome the difficulties in breaking into the supply chain for Japanese automobile manufacturing keiretsu, under an agreement announced during President George Bush's visit to Tokyo in 1992, Japanese automakers indicated they would increase purchases of American autoparts by $9 billion. Such purchases would rise from $10 billion to $19 billion by 1994. This includes an increase in procurement by Japanese assembly plants located in the United States from local American suppliers (from about $7 billion to $15 billion) plus a doubling of U.S. exports of autoparts to Japan (from $2 billion to $4 billion). Also, under the North America Free Trade Agreement, automobiles assembled in the United States must meet a North American content level initially of 60 percent but rising to 62.5 percent (in order to qualify for the benefits of the Agreement). This encourages more local procurement of parts and labor. North American content, however, includes parts from Japanese suppliers incorporated in the United States, as long as they meet American country of origin requirements. Under the current U.S.-Japan Framework Talks, the keiretsu issue is being negotiated under the category of "economic harmonization." On the U.S. side, the Treasury Department is the lead negotiating agency. Progress in these talks is to be reviewed at bilateral summit meetings scheduled to begin on February 11, 1994. The keiretsu also were one of the targets under the Structural Impediments Initiative talks and follow-up actions between the United States and Japan from 1989-92. The Justice Department also has altered its policy toward pursuing antitrust violations by keiretsu in Japan by suing their U.S. subsidiaries. Under a reversal of policy announced in 1992, the Justice Department announced that U.S. antitrust laws can be applied to the activities of the keiretsu in Japan, if the conduct of such firms has a "direct, substantial, and reasonably foreseeable effect on U.S. commerce" whether or not there is direct harm to consumers. There still is no clear indication, however, of whether, and under what circumstances, a U.S. court would choose to assert jurisdiction over keiretsu activities in Japan.(35) 35. See ABA Antitrust Section, Antitrust Law Developments (2d ed. 1984), p. 530. And U.S. Library of Congress, Congressional Research Service, Extraterritorial Application of U.S. Antitrust Laws: Some History and Implications, by Janice E. Rubin. CRS Report No. 92-367 A. Washington, 199[sic]. 13 p. The U.S. Federal Trade Commission (FTC) has been investigating the activities of Japanese automobile companies and their parts suppliers operating in the United States. This investigation, begun in 1990, was to determine whether or not the automakers' propensity to buy components from suppliers in which they hold a financial interest illegally discriminates against competing parts makers.(36) The FTC investigation was pursued by the Bureau of Competition and the Bureau of Economics in a complementary fashion. They examined ownership interests, supply contracts, evidence of discriminatory pricing, and the possibility that such tactics may have produced exclusionary effects that limit the ability of U.S. firms to compete effectively.(37) As of the end of 1993, the case still is open, but no antitrust action has been initiated.(38) 36. Trade. Business Week, June 4, 1990. p. 71. 37. U.S. Federal Trade Commission. Prepared Statement of Janet Steiger, Chairman, Federal Trade Commission, Before the Committee on the Judiciary of the U.S. House of Representatives. May 3, 1990. pp. 3-4. 38. Telephone conversation with Michael McNeely, Bureau of Competition, U.S. Federal Trade Commission, Washington, D.C. February 1, 1994. The U.S. General Accounting Office also investigated Japan's keiretsu. Its study of corporate governance and the business environment in the United States, Japan, and Germany described and analyzed the keiretsu but did not make any policy recommendations.(39) 39. U.S. General Accounting Office. Competitiveness Issues, The Business Environment in the United States, Japan, and Germany. GAO/GGD-93-124. Washington, 1993. p. 81-91. In a report on Japan's distribution system, the U.S. International Trade Commission studied the problem of the keiretsu. It concluded that the keiretsu play distinct roles in the distribution of products in Japan and that in certain industries keiretsu manufacturers were able to exert substantial control over distribution channels. This may have made it harder for new firms to enter the Japanese market. The study found that enforcement of Japan's Antimonopoly Law was weak, but that changes in Japanese government policy as well as economic and social trends point toward increased openness in Japan's distribution system.(40) 40. United States International Trade Commission. Phase I: Japan's Distribution System and Options for Improving U.S. Access. Washington, U.S. ITC, 1990. p. vii-xi. In the 103d Congress, several bills would affect Japan's keiretsu. S. 777 (Senator Roth) would establish a U.S.-Japan Joint Antitrust Consultative Commission for intensive examination of antitrust activities in Japan and the United States. S. 99 (Senator Metzenbaum), S. 332 (Sen. Spector) and H.R. 2628 (Regula) would provide for greater latitude in enforcing U.S. laws against unfair foreign competition. A number of bills would extend the Super 301 provision (against priority unfair trade practices) or would monitor compliance with trade agreements.(41) Other bills rely on concepts of reciprocity or enforcing national treatment in order to gain greater access to Japanese markets. These may affect the keiretsu in general but probably not specifically. 41. H.R. 258 (Neal), H.R. 1573 (Levin), S. 268 (Baucus), S. 269 (Baucus), S. 301 (Daschle), S. 679 (Riegle). Japan's Keidanren (Federation of Economic Organizations), the voice of big business in Japan's politics, favors a review of Japan's competition policy and some increased enforcement of the antimonopoly law. It points out, however, that if the law were to be revised without also changing the statutory waivers from applications of the law for selected industries, inequities would develop. Keidanren favors establishing the rule of "free in principle, subject to regulation only in exceptional circumstances" and more transparency in administering the law and applying regulatory guidelines. It agrees with the United States that governmental administrative guidance should be given in writing, and not just orally.(42) 42. Keidanren (Japan Federation of Economic Organizations). Keidanren Position Paper on the Structural Impediments Initiative (SII) Talks. March 13, 1990. Tokyo, Keidanren. p. 5. Keidanren, however, comprises nearly all the keiretsu companies in Japan. While it favors a stronger JFTC, it still considers the keiretsu, in general, to be a strength of Japan's economy. Stronger antimonopoly enforcement, therefore, is not likely to lead to a demise of the keiretsu. Even if no particular action is taken to weaken the power of the keiretsu, market forces already are eroding the economic rationale for the keiretsu. The conglomerate keiretsu no doubt will continue to grow, but future growth will likely come at the expense of the traditional "family" ties within each group. Individual keiretsu companies are becoming more and more independent as they develop their own marketing mechanisms and establish links with firms in other countries and industries. In the three developed markets of the world, a recent trend is for consortia of firms in a specific industry to link together to market products simultaneously in all three markets. Corporations also are establishing networks by which they link with other firms to share technology and jointly develop products. The agreement between Sun Microsystems Computer Corporation and Fujitsu, Toshiba, Nihon Unisys, and Itochu companies to manufacture and market workstations in Japan is one such example.(43) Another is the Mitsubishi and Daimler-Benz agreement to cooperate over a wide range of business activities. This tie-in is expected to spawn joint projects in automobiles, electrical machinery, aerospace technology, and corporate telecommunications networks.(44) General Motors also heads a consortium that includes Isuzu and Suzuki in Japan and its subsidiaries in Europe. 43. Fishbein, Ed. Keiretsu Sayonara? World Trade Magazine, vol. 6, October 1993. p. 66, 68-69. 44. Smith, Charles. Two's Company. Far Eastern Economic Review, May 24, 1990. M'bishi, Daimler-Benz Mull 7 Joint Projects. Mainichi Daily News, May 24, 1990. As long as markets continue to expand, intra-group trading as a percent of total trade will likely diminish. During a severe recession, however, conglomerate keiretsu could retreat back into their industrial group and support each other as much as possible. The decline in imports into Japan during the 1992-93 recession seems to support this possibility. The cross-shareholdings of stock also could diminish. Given the heights reached by the Tokyo stock exchange in the late 19808, some companies questioned the value of keeping a portfolio with 80 many shares of other companies. In the early 1990s, as values on the Tokyo Stock Exchange have fallen, cross-shareholding has amplified the negative effects on keiretsu firms. Not only have the value of the member's own stocks fallen, but the stocks of affiliated companies in their financial portfolios also have fallen. Pressures on earnings, moreover, have made selling some cross-held stocks attractive. During 1993, the Japanese government's attempt to prop up its stock market by, for example, using funds from its Postal Savings system to bolster demand for stocks has reduced the financial pressures on keiretsu companies to re-examine and possibly adjust their cross-shareholding relationships. Japan's distribution system also is becoming less rigid. The high value of the yen has been encouraging direct imports, mail order sales, and discount stores that circumvent keiretsu distribution systems. U.S. pressures and stronger antitrust enforcement by the JFTC also are undermining control by manufacturers over the distribution chain. The United States has not argued that long-term, keiretsu-type relationships that make economic sense are wrong. Indeed, relationships based on trust that reduce the need for legal work enhance the efficiency of producers. The existence of keiretsu, per se, may not be the problem. The problem may be that the close coordination among group members facilitates violations of antitrust laws and dealings that can exclude U.S. exporters. In terms of reciprocity and equity, moreover, the ease with which Japanese companies can buy into U.S. firms compared with the difficulty of U.S. firms to do likewise in Japan offends the sense of fairness of many Americans. The fastest changes in the keiretsu system are likely to occur in increased networking and non-exclusive distribution systems. The economic rationale for the vertical buyer-supplier relationships is so strong that such keiretsu are unlikely to change much. The conglomerate keiretsu are likely to grow rather than to shrink, although coordination among member companies is likely to diminish as individual companies become more independent and networking outside the keiretsu system becomes more common. Some analysts consider Japan's keiretsu to be at the core of a different form of capitalism, one that some American corporations might adopt. A question for the twenty-first century is whether American forms of business organization, corporate governance, antitrust enforcement, and buyer-supplier contracting provide an adequate structure and incentives for companies to grow, provide quality employment opportunities, and compete internationally. From the vantage point of 1994, many would answer yes. Some scholars, however, have advocated consideration of a more keiretsu-like structure;(46) others point out that such structures in both nations evolved over a long period of time and in response to different economic and social forces and cannot be simply transplanted across national boundaries. 46. Ferguson, Charles H. Computers and the Coming of the U.S. Keiretsu. Harvard Business Review, v. 90, July-August 1990. p. 56. The immediate problem for American companies either exporting to or operating in Japan, however, is that the keiretsu companies do favor buying from related companies. A foreign firm whose prospective customer is a conglomerate keiretsu is facing a buyer who, on average, makes only 5 percent of its purchases from foreign, unrelated firms. If a foreign firm's prospective customer is a vertical keiretsu, it is facing a buyer who usually is looking for a supplier with whom it can or has established a long-term relationship based on trust, mutual risk sharing, and joint product development. Economic forces, more active enforcement of antitrust laws in Japan, and U.S. pressures, however, are presenting opportunities for some U.S. companies to break into these established keiretsu relationships. These American companies have discovered that doors in Japan are opening and that it is now up to them to step up with products, prices, and aftermarket service attractive to Japanese buyers.