The Japanese main bank system: A transaction cost approach

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content THE JAPANESE MAIN BANK SYSTEM: A TRANSACTION COST APPROACH Copyright 2006 by Shinichi Suzuki A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (POLITICAL ECONOMY AND PUBLIC POLICY) May 2006 Shinichi Suzuki R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. UMI Number: 3233846 Copyright 2006 by Suzuki, Shinichi All rights reserved. INFORMATION TO USERS The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleed-through, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion. ® UMI UMI Microform 3233846 Copyright 2006 by ProQuest Information and Learning Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code. ProQuest Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml 48106-1346 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. T a b l e o f C o n t e n t s List of Tables.................................................................................................................................................v List of Figures...............................................................................................................................................x Abstract......................................................................................................................................................xiii Preface..........................................................................................................................................................xv C h a p t e r 1: T h e J a p a n e s e M a in B a n k S y s t e m : I t s C h a r a c t e r is t ic s a n d T h e o r i e s .............................................................................................................................................................. 1 1.1 In t r o d u c t io n .................... 1 1.2 F iv e S t y l iz e d F a c t s o f t h e M a in B a n k S y s t e m .................................................................................2 1.2.1 Bank Loans...............................................................................................................................6 1.2.2 Bond-Issue Related Services................................................................................................. 10 1.2.3 Stockholdings..........................................................................................................................14 1.2.4 Payment Settlement Account................................................................................................ 16 1.2.5 Information Services and Supply of Management Resources............................................17 1.3 S t a t e-C o n t in g e n t N a t u r e o f t h e M a in B a n k R e l a t io n s h ip................................................... 18 1.4 T h e T h e o r ie s o f t h e J a p a n e s e M a in B a n k S y s t e m .........................................................................19 1.4.1 The Agency Cost Theory.......................................................................................................19 1.4.2 The Contingent Governance of Teams.................................................................................24 1.4.3 The Implicit Contract Theory................................................................................................28 1.5 C o n c l u d in g R e m a r k s.....................................................................................................................................32 E n d n o t e s f o r C h a p t e r 1 ....................................................................................................................................... 35 C h a p t e r 2: T h e J a p a n e s e K e ir e t s u M o d e l ................................................................................................40 2.1 In t r o d u c t io n ...................................................................................................................................................... 40 2.2 T h e Ja p a n e s e K e ir e t s u M o d e l ..................................................................................................................41 2.2.1 Transaction Cost Economics: Asset Specificity and Market (Contractual Governance) versus Hierarchy (Organization).............................................................. 41 ii R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 2.2.2 T he F isher B o d y -G en eral M otors R elation: A n Illu stratio n ..................................................... 43 2.2.3 T he Japanese K eiretsu M odel: T he K eiretsu as a “M eta-F irm ” w ith the M ain B ank B eing a S u b s e t............................................................................................................... 45 2.3 A B r ie f O v e r v ie w o f t h e Z a ib a t s u H is t o r y : T h e E a r l y D e v e l o p m e n t of “C o m p l e m e n t a r it y ” a m o n g F i r m s ................................................................................................ 48 2.4 T h e P o s t w a r D e v e l o p m e n t o f K e ir e t s u G r o u p s W h e r e M a in B a n k s A r e S u b s e t s.......................................................................................................................................................... 54 2.4.1 F ro m Z aibatsu to K eiretsu: the P eriod o f “K eiretsu-ization” .................................................... 54 2.4.2 T he V arying Intra-G roup C ohesiveness am ong the B ig T hree: A T C E E x p lan atio n ............................................................................................................................................. 62 2.4.3 T he E ra o f the H eav y and C hem ical In d u stria liz a tio n .................................................................70 2.4.4 T he F orm ation o f th e “B ank G roups” and T h eir M otley N a tu r e .......................................... 103 2.5 T h e C o m p a r is o n a m o n g t h e S ix H o r iz o n t a l K e ir e t s u fr o m t h e T C E S t a n d p o in t : C r o s s-S h a r e h o l d in g s a n d F in a n c in g ...........................................................112 2.5.1 T he Intra-G roup S tockholdings...........................................................................................................113 2.5.2 T he M ain B ank and the N on-F inancial M e m b e rs......................................................................116 2.5.3 T he G T C and the N o n-F inancial M e m b e rs....................................................................................123 2.5.4 T he M ain B ank and the G T C ...............................................................................................................133 2.5.5 T he F inancial C onnection betw een th e M ain B an k and the N on-F inancial M e m b e rs................................................................................................................................................ 136 2.5.6 T he S tatus o f the Intra-K eiretsu T rades: T he F T C ’s R eports (1994, 1 9 9 8 )..................... 140 2.6 C o n c l u d in g R e m a r k s ...................................................................................................................................144 E n d n o t e s f o r C h a p t e r 2 ..................................................................................................................................... 146 C h a p t e r 3: T h e C o n t e m p o r a r y C r is is o f t h e J a p a n e s e M a in B a n k S y s t e m f r o m t h e In d u s t r ia l O r g a n iz a t io n a l P e r s p e c t iv e ..........................................................158 3.1 In t r o d u c t io n ..................................................................................................................................................... 158 3.2 T h e O il C r is e s: T h e T r a n s f o r m a t io n o f t h e In d u s t r ia l S t r u c t u r e a n d t h e R is e o f t h e V e r t ic a l K e ir e t s u ............................................................................................ 159 3.3 T h e St a g g e r in g O n e -S e t P r in c ip l e : T h e In t e r s e c t io n o f th e H o r iz o n t a l a n d V e r t ic a l K e ir e t s u a n d t h e E f f e c t o f D iv e r s if ic a t io n .......................................................................................................................................171 3.4 T h e B u b b l e E c o n o m y a n d Its A f t e r m a t h : E q u it y F in a n c in g , A s s e t- Sp e c if ic it y , a n d t h e M a in B a n k S y s t e m ................................................................................. 193 3.4.1 T he S tatus Q uo o f the E conom ics L iterature on th e B ubble E c o n o m y ............................... 193 3.4.2 T he P eriods o f F inancial L iberalization and the B ubble E c o n o m y .......................................195 3.4.3 T he P ost-B ubble E conom y P e rio d ....................................................................................................211 iii R eproduced with perm ission o f the copyright owner. Further reproduction prohibited without perm ission. 3.5 C o n c l u d in g R e m a r k s .................................................................................................................................. 228 E n d n o t e s f o r C h a p t e r 3 .....................................................................................................................................230 C h a p t e r 4: In s t it u t io n a l C o m p l e m e n t a r it y , A sse t S p e c if ic it y , a n d t h e E c o n o m ic P o l i c y ......................................................................................................................................238 4.1 In t r o d u c t io n ....................................................................................................................................................238 4.2 T h e C o n t e m p o r a r y P r o b l e m s o f t h e C o r p o r a t e G r o u p in g s in Ja p a n : T h e D is c r e p a n c y b e t w e e n t h e C h a n g in g B o u n d a r ie s o f t h e “M e t a - F ir m s” AND THE HlGH-GROWTH FRAMEWORKS............................................................................... 241 4.2.1 The Necessity to Reduce Cross-Shareholdings.................................................................243 4.2.2 The Necessity to Deregulate the Banking Sector: The Road for Investment Banking.............................................................................................................................245 4.3 B u il d in g t h e P o l ic y N o r m : In s t it u t io n a l C o m p l e m e n t a r it y a n d A sse t S p e c if ic it y ................................................................................................................................................249 4.4 A C a s e S t u d y : T h e B ig B a n g F in a n c ia l R e f o r m a n d Its P r o b l e m s.................................. 256 4.5 T h e P o l i c y P r o p o s a l s .................................................................................................................................264 4.5.1 The Pure Holding Company System..................................................................................265 4.5.2 Share Buyback..........................................................................................................................................270 4.5.3 Cross-Entry between Banks and Non-Financial Firms....................................................278 4.6 C o n c l u d in g R e m a r k s .................................................................................................................................. 287 E n d n o t e s f o r C h a p t e r 4 .....................................................................................................................................292 B ib l io g r a p h y ................................................................................................................................................................ 306 iv R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. L i s t o f T a b l e s Table 1. The Main Bank Representation of Main Bank Relations.................................................... 3 Table 2. Variations of Main Bank Relationships with Customer Firms (1963- 1983)....................................................................................................................................... 6 Table 3. External Sources of the Non-Financial Corporate Sector (1970-1992)............................ 7 Table 4. Changes in the Importance of Main Bank Borrowing........................................................ 9 Table 5. Bank Holdings of Domestic Corporate Bonds (Fiscal Year 1953-1991)....................... 11 Table 6. Financial Institutions’ Outstanding Lending (¥m).............................................................56 Table 7. The Rates of Cross-Shareholdings of the Mitsubishi, Sumitomo, and Mitsui Groups (1951-58)................................................................................................... 57 Table 8. Members of the Presidents’ Councils of the Mitsubishi, Sumitomo, and Mitsui Keiretsu, 1960.......................................................................................................... 59 Table 9. The Main Banks’ Lending to the Member Firms (1953-60).............................................64 Table 10. The Net Leakage Ratio, Defined as (C/A) - (B/A)............................................................66 Table 11. The Breakdown of Goods Traded in Sumitomo Corp. (the Second Half of 1952 to the Second Half of 1964)......................................................................................75 Table 12. Selected Examples of the Inter-Firm Relations of the Mitsubishi Keiretsu Established in the Heavy and Chemical Industrialization................................................76 Table 13. Selected Examples of the Inter-Firm Relations of the Sumitomo Keiretsu Established in the Heavy and Chemical Industrialization................................................77 Table 14. The Ranking of the Mitsui Firms in the Order of Mitsui Bank’s Lending (1960).................................................................................................................................... 78 v R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 15. The Ranking of the Mitsubishi Firms in the Order of Mitsubishi Bank’s Lending (1960).................................................................................................................... 81 Table 16. The Ranking of the Sumitomo Firms in the Order of Sumitomo Bank’s Lending (1960).................................................................................................................... 84 Table 17. The Results of the First Regressions (1960)..................................................................... 87 Table 18. The Results of the Second Regressions (1960)................................................................. 88 Table 19. Members of the Presidents’ Councils of the Mitsubishi, Sumitomo, and Mitsui Keiretsu, 1966......................................................................................................... 89 Table 20. The Ratios of the Intra-Keiretsu Stockholdings (1960)................................................... 91 Table 21. The Ratios of the Intra-Keiretsu Stockholdings (1966)....................................................92 Table 22. The Ranking of the Mitsui Firms in the Order of Mitsui Bank’s Lending (1966).................................................................................................................................... 93 Table 23. The Ranking of the Mitsubishi Firms in the Order of Mitsubishi Bank’s Lending (1966)....................................................................................................................96 Table 24. The Ranking of the Sumitomo Firms in the Order of Sumitomo Bank’s Lending (1966)....................................................................................................................99 Table 25. The Results of the First Regressions (1966).................................................................. 101 Table 26. The Results of the Second Regressions (1966).............................................................. 101 Table 27. Members of the Presidents’ Councils of the Bank Groups (the Fuyo, Sanwa, and DKB Groups), 1979...................................................................................... 104 Table 28. The Main Bank’s Stockholdings Relation Ratio............................................................. 118 Table 29. The Ratio of the Non-Financial Members’ Stockholdings Relation with the Main Bank.................................................................................................................... 121 Table 30. The Intra-Keiretsu Sales Ratio and the Intra-Keiretsu Purchases Ratio........................141 Table 31. Intra-Keiretsu Sales Ratio and Intra-Keiretsu Purchases Ratio (1992, 1996)................................................................................................................................... 142 vi R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 32. The Relations among the Three Types of Stockholding and the Three Types of Transaction (1992)........................................................................................... 143 Table 33. Investment, Financing, and Buying and Selling Relations among Sumitomo Bank, Sumitomo Corp., and Mazda Motors in FY 1998........................... 147 Table 34. The Energy Supply in Japan during 1955-1965.............................................................. 154 Table 35. Lending Rates by the Six Main Banks during 1988-1992............................................. 157 Table 36. Trends of Principal Economic Indicators, 1970 - 1983................................................. 160 Table 37. Intramural Expenditure on R&D as a Percentage of Sales, by Industry, FY 1976- 1981 ................................................................................................................ 162 Table 38. Change in Interindustry Gross Output Structure in the 1970s (Gross Output as 1975 Price = 100).............................................................................................163 Table 39. The Gross Profit Rates of Nine GTCs (1973-1982)....................................................... 165 Table 40. Correlations of Investment across Industries....................................................................166 Table 41. Change in the Proportion of Subcontracting Small and Medium Firm s........................169 Table 42. Diversification Matrix (Mitsui)......................................................................................... 178 Table 43. Diversification Matrix (Mitsubishi)..................................................................................180 Table 44. Diversification Matrix (Sumitomo)...................................................................................182 Table 45. Diversification Matrix (Fuyo)........................................................................................... 183 Table 46. Diversification Matrix (Sanwa)........................................................................................ 185 Table 47. Diversification Matrix (DKB)........................................................................................... 187 Table 48. Diversification of the Six Horizontal Keiretsu..................................................................190 Table 49. Transition of the Domestic Issuance Conditions and the Number of Firms Qualified............................................................................................................................. 197 vii R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 50. Funding Patterns of the Japanese Firms (% Distribution, 1975-95)............................ 199 Table 51. Security Financing by Listed Firms (1972-98)................................................................200 Table 52. Bond Trustee Administration (Answer to the First Query)............................................ 201 Table 53. Bond Trustee Administration (Answer to the Second Query)....................................... 202 Table 54. Cross-Shareholding in the Six Keiretsu (1980, 1985, 1990).......................................... 205 Table 55. The Loans to the Jusen Companies by the Six Main Banks (1995)...............................213 Table 56. The Bad Loans, the Bad-Loan Ratios, and the Capital-Asset Ratios of the Six Main Banks..................................................................................................................217 Table 57. The Latent Profits from Stockholdings of the Six Main Banks..................................... 218 Table 58. The Capital Consolidation by the Six Main Banks and the IBJ (October 1998 to March 1999)......................................................................................................... 220 Table 59. Moody’s Long-Term Credit Ratings on the Six Main Banks........................................ 221 Table 60. Mergers and Affiliations Made beyond the Traditional Keiretsu Boundaries by Non-financial Shacho-kai Firms (2000-02).......................................... 226 Table 61. Increasing Tie-Up Contracts with Outsiders in the Six Keiretsu (1975)....................... 231 Table 62. Dissolving Trends of Cross-Shareholdings during 1995-98.......................................... 236 Table 63. The Rates of Cross-Shareholdings Made among the Six Keiretsu Firms, the Non-Keiretsu Firms, and the Whole Market.............................................................239 Table 64. The Number of Subsidiaries Founded during 1982-91 in Japan....................................240 Table 65. Lending Shares in the Small and Medium-Sized Firms..................................................247 Table 66. Key Elements of the Big Bang...........................................................................................258 Table 67. The Japanese Firms Planning to Set Up Their Pure Holding Companies as of September 2000........................................................................................................ 268 viii R eproduced with perm ission o f the copyright owner. Further reproduction prohibited without perm ission. Table 68. The Reformations of the Share Buy-Back System..........................................................273 Table 69. Economic Positions of the Six Gigantic Horizontal Keiretsu in the Japanese Economy when Measured by the Indices (a)-(g)........................................... 292 Table 70. Values of Stockholdings of 9 Major Japanese Banks and Trusts in 2000 and Those of the Stocks They Planned to Sell Off in 2001 .......................................... 300 Table 71. Definition of Small and Medium-Sized Firms.................................................................300 Table 72. Number of M&As and Their Values (1998-2003).......................................................... 303 Table 73. Industry-wide Share of the GDP in Fiscal Years 1950,1970, and 1995.......................304 ix R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. L i s t o f F i g u r e s Figure 1. Intra-keiretsu stockholdings ratio.....................................................................................114 Figure 2. Intra-keiretsu stockholdings relation ratio.......................................................................115 Figure 3. Intra-keiretsu average stockholdings ratio.......................................................................116 Figure 4. Main bank’s stockholdings ratio........................................................................................ 117 Figure 5. Main bank’s average stockholdings ratio........................................................................ 119 Figure 6. Non-financial firms’ average stockholdings ratio............................................................120 Figure 7. Average ratios of the non-financial members’ stockholdings with the main bank............................................................................................................................122 Figure 8. GTC’s stockholdings ratio (non-unified version).............................................................124 Figure 9. GTC’s stockholdings ratio (unified version — unifying the data of three GTCs in Sanwa and three GTCs in DKB, respectively, and displaying them along with others)..................................................................................................... 125 Figure 10. GTC’s stockholdings relation ratio....................................................................................126 Figure 11. GTC’s average stockholdings ratio (non-unified version)...............................................127 Figure 12. GTC’s average stockholdings ratio (unified version).................................................... 127 Figure 13. Ratio of the members’ stockholdings with the GTC (non-unified version)................................................................................................................................128 Figure 14. Ratio of the members’ stockholdings with the GTC (unified version)........................... 129 Figure 15. Ratio of the members’ stockholdings with the GTC........................................................129 x R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Figure 16. Average ratio of the members’ stockholdings with the GTC......................................... 130 Figure 17. Average ratio of the members’ stockholdings with the GTC (unified version)................................................................................................................................131 Figure 18. The case when the main bank’s holding of the GTC’s stocks is eliminated (non-unified version).......................................................................................131 Figure 19. The case when the main bank’s holding of the GTC’s stocks is eliminated (unified version)...............................................................................................132 Figure 20. Ratio of the main bank’s stockholdings with the GTC....................................................134 Figure 21. Ratio of the GTC’s stockholdings with the main bank (non-unified version)................................................................................................................................135 Figure 22. Ratio of the GTC’s stockholdings with the main bank (unified version)......................135 Figure 23. Main bank’s financing relation ratio................................................................................. 137 Figure 24. Intra-keiretsu borrowing ratio............................................................................................ 138 Figure 25. Intra-keiretsu borrowing..................................................................................................... 139 Figure 26. Number of subsidiaries (Mitsui)........................................................................................ 172 Figure 27. Number of subsidiaries (Mitsubishi).................................................................................173 Figure 28. Number of subsidiaries (Sumitomo)..................................................................................173 Figure 29. Number of subsidiaries (Fuyo)...........................................................................................174 Figure 30. Number of subsidiaries (Sanwa)........................................................................................174 Figure 31. Number of subsidiaries (D K B ).......................................................................................... 175 Figure 32. Lending ratio in the construction sector........................................................................... 208 Figure 33. Lending ratio in the real estate sector............................................................................... 208 Figure 34. Lending ratio in the commerce sector...............................................................................209 xi R eproduced with perm ission o f the copyright owner. Further reproduction prohibited without perm ission. Figure 35. Lending ratios in the three non-production sectors......................................................... 209 Figure 36. Borrowing ratio (defined as borrowing/annual sales). Unit: %...................................... 262 Figure 37. Equity capital ratio (defined as equity capital/[equity capital + borrowing]). Unit: %..........................................................................................................262 xii R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. A b s t r a c t Contrary to their high performance in the past, Japanese banks suffered from bad loans and were deeply concerned with the prolonged depression of the Japanese economy during the 1990s. This raised questions of what changes occurred in the Japanese banking sector and what policies should be addressed to restore the sector. This dissertation gave answers to those questions from the standpoint of transaction cost economics, which was applied to the post-war major enterprise groups and their banks in Japan. The approach analyzed how the inter-firm transactions were molded in the groups and compared the groups qualitatively and quantitatively in light of transaction costs. The major findings were the following: (1) 1945 to the early 1970s—while the intra-group transaction supported by the group bank was common across the groups, there was divergence in how often such a transaction occurred. Typically, the former zaibatsu groups, especially the Mitsubishi group, outdid others in frequency. This was because their firms were mostly spin-offs in the heavy and chemical sectors during the zaibatsu era. (2) Early 1970s to the mid 1980s—all groups suffered from the two oil crises that damaged their intra-group transactions. Many group firms established their subsidiaries to explore new transaction networks. This tendency was less likely in the former zaibatsu groups, especially Mitsubishi. (3) Mid 1980s to the early 2000s—the group firms tapped into capital markets as financial deregulation progressed. The group banks then began speculative lending. This tendency occurred infrequently in Mitsubishi, where the group firms kept higher levels of transactions and, thus, were bound by higher levels of cross shareholdings. Consequently, Mitsubishi Bank faired well in the 1990s. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. The following recommendations were made so that financing can accommodate the new inter-firm transactions emerging after the oil crises: (1) introducing the holding company system, (2) initiating the share buyback system to reduce cross-shareholdings, (3) equipping firms with a banking function, and (4) allowing banks to invest in small and medium-sized firms. The laws concerning these policies accordingly need revision. Dissertation Comittee Robert Dekle, Ph.D., Committee Chair Department Chair, Professor, Department of Economics, University of Southern California, Los Angeles, California Saori N. Katada, Ph.D. Associate Professor, School of International Relations, University of Southern California, Los Angeles, California Najmedin Meshkati, Ph.D., CPE Associate Professor, Department of Civil/Environmental Engineering Department of Industrial and Systems Engineering Viterbi School of Engineering University of Southern California, Los Angeles, California R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. P r e f a c e This dissertation is a product of my long research of large Japanese banks and their affiliated firms, with particular attention to those belonging to the former six gigantic keiretsu (Mitsui, Mitsubishi, Sumitomo, Fuyo, Sanwa, DKB). The salient characteristic of the post-war Japanese financial sector is the dominance of the banking sector. From the 1950s to the early 1970s, those banks played a significant role in meeting enormous corporate financial needs, thereby contributing to the quantum leap of the Japanese economy during the period. Consequently, much research focused on the Japanese economy both in- and outside Japan, trying to unearth the secrets of the Japanese economy, including the peculiarity of its banking sector. When the recessive force weakened in the Japanese economy in the early 1990s, however, the evaluation of the Japanese banks suddenly reversed, and they were often identified as the reason for economic decline. Naturally, I was drawn to such queries as (a) What change(s) occurred in the Japanese banking sector? and (b) What policy prescription(s) should be addressed to remedy and restore the sector? While conducting the research, I constantly benefited from the literature listed in the bibliography. Among all, I found most beneficial those works by Professors Masahiko Aoki, Michael L. Gerlach, Ronald J. Gilson, Takeo Hoshi, Anil K. Kashyap, W. Carl Kester, Mark J. Roe, Paul Sheard, and Oliver E. Williamson for constructing my argument. As Professor Aoki (1989) exhibited, there are currently three main approaches to economic organization, and my own research principally falls into one of these three, “Transaction Cost Economics (TCE),” which Professor Williamson pioneered. Three chief reasons caused me to adopt this approach in this dissertation. First and foremost, a major characteristic of the Japanese economy lies in its teaming with the inter firm exchange networks, and the TCE analysis, when applied to such an economy, is expected to xv R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. prove both useful and powerful. Second, while a series of works by Professors Aoki and Sheard on the theorization of the Japanese main bank system neared completion, and the agency cost analyses were intensively presented in the forms of empirical analysis by Professors Hoshi and Kashyap in the first half of the 1990s, the TCE area remained little explored, except for a few masterpieces by Professors Gerlach (1992) and Kester (1991a). In particular, until today, the analysis on the inter- firm relations of the Japanese economy of the 1990s was virtually nil. Third, in the past, it has often been the case that Japanese banks were analyzed in their own right, being obscured from their relations with other industrial sectors of the economy. By adopting the TCE framework, which concerns primarily with a real rather than a financial transaction, such a bias is expected to be appeased, especially the part Japanese banks played in the real exchanges of the economy. Transactions are the alpha and omega of the economic activity. Issues arisng from the physical attributes of goods on transaction and the bounded-rationality of exchanging agents, however, have rarely been assigned the central position in economic thought. The Institutional School, from which TCE later branched off, waned after John R. Commons left the sign of its short life, if at all, in footnotes of economics textbooks. Waxing was Neoclassical Economics which, despite Professor Williamson’s advocacy of the New Institutional Economics in the 1970s, constantly reigned over the discipline as the paradigmatic thought during the postwar period. Even with its advanced, sophisticated analytical tools, however, Neoclassical Economics often fails to identify and tackle the difficulties lying in economic transactions. First, in the Neoclassical economics models, firms are equated with the production function, embodying the technological relation allegedly existing between the various combinations between capital and labor as inputs and the corresponding levels of output. Aside from whether such a technological relation exists for real, the issue implicit in such a setting is that the firm can smoothly acquire both inputs as much as it requires through market contracts. In 1937, this view encountered a serious challenge by Ronald H. Coase, who asked why the economy is populated with firms even when independent, self-employed people are allowed to mutually contract through markets. His answer was that the market xvi R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. transactions are not so smoothly carried out as economists envisaged, and that transactions are not only mediated by the market prices, but also unavoidably attended by information costs, bargaining costs, and other related costs to complete transactions, which were all to be comprehended by the concept of “transaction cost.” Establishing the firm, he argued, on the other hand, would contribute to economizing on these costs because a substantial part of the procurement activities is thereby shifted from markets to inside an organization (i.e., a firm). Second, even if the setting of the production function is justified, there remains the problem of the homogeneity of goods in the Neoclassical models which, like grains in the agricultural economy, are both inputted in and outputted from the production function, and ubiquitously exchanged among economic agents by the medium of the price mechanism. Such an assumption is often assumed for analytical convenience or defended by the concept of the Hicksian composite goods. In such economic models, however, there is no room for the issue of transactions to arise from the asset specificity of the goods under transaction. Standardization, if not homogenization, is actually prevalent in the contemporary U.S. economy. Among others, the recent “modularization” technique significantly contributed to standardizing interfaces of various production parts, or “modulars,” used in such industrial sectors as automobile, information technology, machine tool, and semi-conductor, thereby rejuvenating Silicon Valley and other parts of the U.S. economy in the 1990s. Through modularization, transaction costs that otherwise would have been exorbitant are considerably mitigated, and the market transactions of these sectors have, accordingly, become more invigorated than ever. But as any sort of modularization starts with the design process of standardizing various interfaces to be properly connected, it never is immune from the issue of asset specificity. Furthermore, a serious problem lies ahead; the more modularization progresses, the more the connection among interfaces goes out of control (Aoki and Ando 2002). This is exactly the point where the Japanese keiretsu, especially that of the vertical genealogy, shows its superiority, as it is often designed to overcome the ex-post problem of integrating parts among affiliated firms. In the face of modularization’s xvii R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. power that has started to carve out shares and niches, the Japanese entrepreneurs seem all too often to waver on their own production system. But they need to carefully discern its advantages and disadvantages alike from the long-haul perspective and thereby make sensible decisions. This would turn out to be even more crucial, if they clearly recognize the “institutional complementarity” of the goods, financial, and labor markets, and the way the systems differ between Japan and the U.S, as discussed in depth in Chapter 4. This dissertation includes four chapters. Chapter 1 identifies the main characteristics of the Japanese main bank system and reviews three notable theoretical works that explain them. In Japan, it is often the case that business firms, especially large ones, hold long-term relations with particular banks, which constantly meet a substantial portion of their borrowing needs. Japanese financial practitioners have traditionally called them “main banks,” implying that they are the “main” sources of the financial supply to their customer firms. Typically, there are five notable characteristics of the Japanese main bank system: (1) bank loans, (2) bond-issue-related service, (3) shareholdings, (4) offering the payment settlement account, and (5) supply of management information. The most salient, among all, is (5), which functions effectively often with (3) and (4). Explaining these functions are (a) the agency cost theory, (b) the implicit contract theory, and (c) Professor Aoki’s theory of contingent governance of team. For each, merits and demerits are equally examined. Currently, there remain mainly two problems to be overcome in these works. One is that in (b) the enforcement mechanism needs to be clarified to explain why the main bank mediates the keiretsu transaction and why it often incurs disproportionate costs involved in restructuring its financially ailing affiliated firms. The other is that in (c) the main bank is depicted as the sole monitor, and there is a necessity for other main bank monitors to be taken into account. Chapter 2 is titled “the Japanese Keiretsu Model,” which is the fourth hypothesis presented in this dissertation for explaining the Japanese main bank system phenomenon. The theoretical backbone of this hypothesis is TCE, which is applied to the Japanese (horizontal) keiretsu case. The advantage of adopting this hypothesis lies, first, in its being able to fill in the enforcement xviii R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. mechanism missing in the implicit contract theory and, second, in rectifying the view that the main bank is a monitoring paragon in Professor Aoki’s contingent governance of team. The essence of the Japanese keiretsu model is that keiretsu member firms positioned in various industrial sectors engage in the asset-specific, intra-keiretsu transactions with one another, and that the main bank, as a subset of the whole keiretsu constituency, supplies funds to those non-financial members to promote such intra-keiretsu transactions. During the period when the capital shortage was severe after WWII, the main banks of the Big Three (Mitsui, Mitsubishi, Sumitomo) keiretsu shared the advantageous position respectively within them, but it should not be dismissed that they were also subject to other non-financial members’ monitorihg whereby the non-financial members could secure the financial support from their main banks for their intra-keiretsu transactions. And it was the technique of cross-shareholdings across member firms that made such mutual monitoring possible. The technical conditions that enabled such intra-keiretsu, asset-specific transaction, on the other hand, were highly prepared in each of the Big Three zaibatsu groups (Mitsui, Mitsubishi, Sumitomo) by the end of WWII, through their high asset-specific-related spin-off investments across vital industries, which led to the unique industrial structure later dubbed the “one-set principle.” Such an industrial structure was then carried over to the postwar period along with the banks, which came to power during WWII. With this backdrop, after the Antimonopoly Law (April 1947) was administered and the holding company was consequently banned in Japan, the former zaibatsu subsidiaries naturally rallied to their affinities, through such alternative expediencies as cross-shareholdings and presidents’ councils, and formed the Big Three keiretsu groups. Their inter firm transactions were further accelerated by their intensive pursuit of heavy and chemical industrialization in the context of the postwar economic reconstruction. In addition to this general tendency across the Big Three keiretsu, Chapter 2 also discerns the dissimilarity that had subtly, yet persistently existed across them, thereby enhancing the credit of the TCE analysis. The dissimilarity among them arose primarily from the degree in the maturity of the inter-firm transaction relations molded during their zaibatsu period, in which the Mitsubishi and xix R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Sumitomo groups had generally outdone the Mitsui group because of their primarily engaging in the heavy and chemical industries. Such a dissimilarity was, then, directly translated into the frequencies of their intra-keiretsu transactions in the postwar period, once their parent companies (i.e., holding companies) were removed. Taking advantage of their highly developed inter-firm transaction networks, the Mitsubishi and Sumitomo groups could open their general trading companies and the presidents’ councils much more easily and, thus, earlier than the Mitsui group. The dissimilarity is also confirmed by two statistical measures: “Net Leakage Ratio,” which measures how much money flowed from one keiretsu group while receiving one yen from outside in each year from 1953-60, and a set of two regressions, which, on the other hand, measure the correlation between the main bank’s financing the member firms and the degree of cross shareholdings among the member firms and the bank in 1960 and 1966. In either case, Mitsubishi group’s highly developed intra-keiretsu exchange network and Mitsubishi Bank’s intensive intra- keiretsu financing are confirmed. The relevancy of the TCE approach proves also effective by the analyses on the three “Bank Groups” (Fuyo, Sanwa, and DKB), which, in the aftermath of heavy and chemical industrialization in the economy, rose in the late 1960s through the early 1970s. Similar to the case of the Big Three, these new groups were all characterized by the main bank, the intra-keiretsu transaction, cross-shareholdings, and the presidents’ council. Reflecting their motley nature of being composed of the former smaller zaibatsu groups, however, their intra-keiretsu transactions were, in general, severely constrained compared with the Big Three. As the TCE framework predicts, the rates of the intra-keiretsu cross-shareholdings in the Bank Groups were much lower than the Big Three, and, as the name of the “Bank Group” connotes, the groups were better characterized by the main banks’ financing their constituent firms, rather than promoting their intra-keiretsu transactions. Trying to find the source of the speculative bubble economy of the late 1980s in the real, rather than financial, economy, Chapter 3 analyzes how the inter-firm transactions inside each of the six keiretsu were encroached by the two oil crises of the 1970s. The difficulty brought about by the xx R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. oil crises against the keiretsu firms primarily lay in their losing the economic justification behind their intra-keiretsu transactions because they were mostly positioned in the heavy and chemical industries represented by oil, iron, and electricity. The disentangling of the keiretsu exchange networks was then furthered by the keiretsu member firms, especially those positioned in the assembly sectors, that started to spawn their subsidiaries and developed their new transaction networks known as the “vertical” keiretsu. By so doing, these firms diversified their businesses into the sectors more suitable for the post-oil-crises economic world, while benefiting simultaneously from the wage cut of the factory workers by moving them down to the subsidiaries and from the low transaction costs with the subsidiaries. In order to examine how much the one-set principle was encroached through such diversification, Chapter 3 introduces the “diversification matrix” for FY 1985, and examines how much earnings each keiretsu firm actually made from their mainstay and non-mainstay sectors. Consistent with the analysis of the previous chapter, the Mitsubishi group was the most faithful to the one-set principle, while the three Bank Groups were much less so. Under this industrial organizational condition financial deregulation was intensively pursued in Japan. With convertible bonds already in place, many keiretsu firms started to tap into the warrant-attached bonds allowed in the deregulation and to get around their traditional main bank borrowing. Correspondingly, the main banks started intensive financing in high-yield sectors like real estate, construction, and commerce, as they lost member firms from their customer bases. At this point, the prominence of the TCE analysis presents itself in its capability to predict the heterogeneous lending patterns across the six main banks from the institutional complementarity among the asset-specific investment, cross-shareholdings, and main bank lending. During the speculative bubble period, Mitsubishi Bank, which had kept the highest degree of cross shareholdings with other Mitsubishi firms and maintained the one-set principle in the most faithful form, significantly abstained from lending in the sectors. This was primarily because Mitsubishi Bank needed to fulfill the enormous borrowing needs from other Mitsubishi firms which, due to their high levels of cross-shareholdings, could not satisfactorily consolidate their capital. In trying to xxi R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. replenish their dissipating customer bases, on the other hand, the main banks of the three Bank Groups were forced to finance excessively in the three high-yielding, yet risky sectors. Once the euphoria of the bubble economy burst in the early 1990s, the main banks’ lending in the three sectors soured as bad loans. Parallel to the economic climate developing from bad to worse over time, bad loans accumulated to unprecedented levels across the main banks, severely hampering their operations. Having abstained from excessive lending in the three sectors, however, Mitsubishi Bank was least affected by bad loans. While Mitsubishi Bank was able to reserve its capacity to absorb Tokyo Bank (March 1995) and to refuse the government’s offer of taxpayer funds (March 1999), the other five main banks were eventually forced into mergers going beyond the traditional keiretsu lines, as a result of their submitting the restructuring plans to the government, in exchange for taxpayer funds. Chapter 4, the last chapter of this dissertation, explores the macroeconomic policy. Albeit not so conspicuous as in the case of the six keiretsu, the long-term trading, cross-shareholdings, and the development of the vertical keiretsu by large firms are also widely observed in other Japanese firms and banks that do not belong to the six keiretsu. The issues pointed out so far are, thus, not limited to the six keiretsu. First, the chapter establishes two policy norms. The first policy norm directly comes from the “institutional complementarity” concept developed by Professor Aoki. The concept particularly addresses the complementary functioning between the financial and labor markets. In the case of the Japanese economy, the banking sector and the firm-specific, interior labor market function complementarily, in that the former, in lieu of the securities market, integrates the entire monitoring process over the latter, thereby contributing to the firm-specific nature of the labor market, and in that the latter, under the former’s monitoring, struggles to improve its profit under the team-natured operation. The second norm is asset specificity consideration in the good market, which is missing in the institutional complementarity concept. The financial Big Bang, an extensive set of financial liberalization measures sequentially administered in Japan since 1997, is then shown to be an example of failing to take into consideration these two norms. Alternative to the xxii R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. financial Big Bang, three policies are proposed in this chapter, based on the above two policy norms. The first policy proposal is to allow the “unlisted” holding company, along with the allowances for its subsidiaries, to be listed in the securities market. As analyzed in-depth in Chapter 3, the inter-corporate transactions since the oil crises have, in general, moved from those among large firms to between large parent firms and their subsidiaries, with the consequence that those large firms have become “de-facto” holding companies of their subsidiaries. The holding company system, which was eventually introduced in June 1997 in Japan, accommodated this irreversible industrial trend. Also, from the point of view of corporate governance, the system contributed to recovering the effective governance over the large firms which, due to the governance lacuna, had often been induced to commit corporate scandals and other speculative actions during the bubble economy period. The current holding company system in Japan, however, leaves one remaining problem—how to effectively preserve the Japanese firm-specific, internal labor market. To resolve it, the system should further be advanced to allow the holding company to be delisted from the securities market. Otherwise, the firm-specific labor market would unavoidably intervene in the securities market, first at the holding company level, and then at its subsidiaries levels. However, delisting of the holding company raises another problem—the lack of governance over the holding company. In order to stay clear of this dilemma, it would be necessary to introduce outside directorship to the delisted holding company. Furthermore, because the delisted holding company would likely cause a problem of limiting the financial availability to its subsidiaries, it was necessary to allow the subsidiaries to be listed and, thereby, to tap into the securities market, even if the holding company is delisted. The second policy proposal is to introduce the share buyback system, which ideally should be administered simultaneously with the first proposal. This would lead to further improvement of corporate governance over the large firms. In order to expedite dissolving cross-shareholdings among large firms, the “Direct Mutual Share Buyback (DMSB)” is proposed, whereby the firm can xxiii R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. skip over the open security market and directly conduct the share buyback among themselves. This idea derives from the fact that in the past cross-shareholdings were often made by third party allocation (daisansha zoshi). Besides, for the short-term policy, the “Cross-shareholdings Purchase Corporation (CSPC)” is proposed, as an alternative to the “Banks’ Shareholdings Purchase Corporation (BSPC) (January 2002) currently adopted by the government. In contrast to the BSPC, which was assigned a limited role of purchasing the cross-held stocks between banks and their customers firms, the CSPC proposed here aims to reduce not only those stocks held between banks and their customers, but also those among large, non-financial firms. In implementing the CSPC, the share buyback should be carried out through acquisition prices rather than market prices because they were the prices for which cross-shareholdings were concluded in the past. By extending the function of the BSPC to that of the CSPC, there will be the direct benefit of saving the taxpayer fund disbursed to the BSPC. This is because re-exchanging the cross-held stocks among the firms concerned does not, in principle, require any expenditure anew. The third and final policy proposal is to equip large firms with a banking function, both commercial and investment, and to allow traditional banks to invest in small and medium-sized firms. Holding the banking function is particularly important for the firms whose holding companies are delisted from the securities market because they need to secure enough operational funds. Besides, the banking function would enable large firms to properly finance their subsidiaries, with much less constraint from their financial availability and borrowing from outside banks. These considerations, in turn, lead to the proposal of allowing the traditional banks to invest in small and medium-sized firms because not doing so would be unfair for the banks. This is even more relevant, especially when considering that large firms have been able to serve as “de-facto” banks for their subsidiaries, especially since the oil crises. Allowing traditional banks to serve as both commercial and investment banks in the area of small and medium-sized firms is actually very crucial today because they are substantially losing large firms as their customers. In particular, allowing investment in small and medium-sized firms to traditional banks would present them with the best xxiv R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. opportunity to acquire the essential skills of how to evaluate, underwrite, and monitor the small and medium-sized firms, with no recourse to raising collateral, which was often the case in the past. In order to arrange for the three proposals above, the overhaul of the laws concerned is indispensable: the Antimonopoly Law, the Securities and Exchange Law, the Commercial Law, and the Banking Law. In parallel, the policy coordination by the cabinet is required because the issues discussed above straddle multiple jurisdictions among the concerned ministries. This dissertation was submitted to the Graduate School, University of Southern California (USC), in the spring semester 2006, the final semester of the Political Economy and Public Policy (PEPP) program, USC. In completing this dissertation, I was often required to go through hard, yet stimulating times, as the topic involved not only the main TCE analysis, but a broad range of other disciplines as micro- and macroeconomics, industrial organization, law and economics, corporate financing and governance, labor economics, and political science. It was the PEPP program’s adoption of the inter-disciplinary approach that generously allowed me to select this topic for the present dissertation. For this, I herein express my deep thanks. Very regretfully, it was decided that the program be closed at the end of this semester. Among all the reasons that eventually led to the closure of the program, the most crucial one, I surmise, might have been the difficulty inherent in an inter-disciplinary approach. With no common methodology, language, and grammar bridging distinctive disciplines, the inter-disciplinary approach often imposes inordinate skills and ingenuity on the part of learners. The importance of the approach, however, is becoming even more eminent today, as the specialty and technicality are constantly intensified across disciplines. In closing this preface, I fear how much I could attain this grand goal and repay the program. I only wish that the present dissertation will provide a good discussion topic and be subject to various examinations. February 2006, Shinichi Suzuki R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Chapter 1 T h e J a p a n e s e M a in B a n k Sy s t e m : I ts C h a r a c t e r is t ic s a n d T h e o r ie s 1.1 Introduction The present dissertation explores the Japanese main bank system. What best characterizes the Japanese financial sector is its bank-dominant nature. Japanese banks, which are often linked with the “main bank” system, were the major force of the rapid economic growth of the Japanese economy during the mid-1950s through the 1970s. In the 1990s when the Japanese economy faced unprecedented depression and recurrent financial crises, however, a widely-held recognition ascribed these problems to a malfunction of the banking sector. Consequently, questions naturally arise as to what change(s) occurred in the Japanese banking sector and what policy prescription(s) should treat the section’s malfunction. Should the Japanese banks be completely abandoned and replaced by the Anglo-American securities market system? As the starting point to answer these questions, Chapter 1 overviews the main traits of the so-called Japanese “main bank system” and its theoretical works considered the most influential today. First, Section 1.2 confirms the representative roles played by the main bank as the five stylized facts, taking advantage of Aoki and Patrick (1994). To supplement Aoki and Patrick, historical accounts are also given when applicable. To better understand the corporate governance of the main bank, it is important to recognize that, quite often, the main bank displays its functions, critically depending on the financial states of its customer firm. The most salient characteristic is its intervention into the customer firm when it starts to suffer financial distress. Section 1.3 summarizes the state-contingent nature of the main bank’s functions, using Aoki and Patrick (1994). With those characteristics clarified, Section 1.4 reviews three notable theories of the Japanese main bank 1 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. system to determine which economical aspects the main bank considers the best of its functions: (1) the agency cost theory, (2) the contingent governance of teams, and (3) the implicit contract theory. Section 1.5 summarizes both the advantages and disadvantages of those theories for the purpose of model-building, which itself is delegated to Chapter 2. 1.2 Five Stylized Facts of the Main Bank System The term, “main bank,” has been continuously used in Japanese business circles since the postwar period. Quite recently, even researchers have started to use the term often. While the term has been used with almost no reference to its formal definition, there is certainly the core idea behind it. That is true especially when they say, “Bank A is the main bank of Company B,” or “Company B has Bank A as its main bank.” That is, the relationship between Bank A and Company B is characterized as particularly long, continuous, and harmonious, with the former being the largest lender to the latter (Aoki, Patrick, and Sheard 1994: p. 3; Teranishi 1993: p. 63). Almost all Japanese companies, both large and small, are said to have had their own main banks during the postwar period (Aoki, Patrick, and Sheard 1994: p. 5), so the main bank system is considered the fundamental element of Japanese-style financing. According to Miyashita and Russell (1994: p. 49), for instance, the main banks were the primary sources of borrowing for over two-thirds of Japan’s 873 biggest non-financial firms listed on the Tokyo Stock Exchange (TSE), and the secondary sources of funds for another 17 percent. In sum, the main banks are either the primary or secondary lenders to about 85 percent of Japan’s top companies (Miyashita and Russell 1994).1 Following the definition of the main bank given by the Economic Research Association (Keizai Chosa Kai) that the main bank offers the largest number of loans to its customer firms for the past three or more years consecutively, Horiuchi, Packer, and Fukuda (1988: pp. 162-4) also reported that nearly all the companies listed on the first section of TSE have main banks. Since main bank financing is so R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. widely accepted as the vital element of the Japanese corporate financing, this section reviews its characteristics in depth. Various works on the stylized facts of the Japanese main bank system are available today, but the most elaborate and comprehensive one is by Masahiko Aoki (see Aoki, Patrick, and Sheard 1994: pp. 5-19; Aoki and Okuno 1996: pp. 223-7). According to him, the following are the most conspicuous features of the Japanese main bank system, with each shifting its importance depending on the client firm’s four financial states: “Excellent,” “Favorable,” “Normal,” and Critical” (see Table 1): 1. Bank loans 2. Bond-issue related services (trustee administration in the case of domestic issues and co-lead management in the case of Euro issues) 3. Shareholding 4. Payment settlement accounts 5. The supply of management information and resources Table 1. The Main Bank Representation of Main Bank Relations Nature o f main bank relations Financial state o f the firm Excellent Favorable Normal Critical 1. Bank loans (•) • ® 2. Bond-issue-related services3 (•) • 3.. Shareholding • • • • 4. Payment settlement account • • • • 5. Supply o f management information and resources (•) (•) • Sources: Aoki, Patrick, and Sheard (1994: p. 7; Aoki and Okuno 1996: p. 226). Note. 3 Trustee administrator in the case o f domestic issues and co-lead management in the case o f Euro issues. • = indicates this role performed in this financial state. ® = indicates this role performed strongly in this financial state. (•) = indicates this role performed weakly in this financial state. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. These characteristics are mostly confirmed by other researchers such as Horiuchi and Sui (1992: pp. 1-2) and Miyashita and Russell (1994: pp. 49-53). Horiuchi and Sui (1992) pointed out the following four as stylized facts of the Japanese main bank system: 1. A main bank is the most important financial source to its customer firms, and the financial transactions between the bank and the firms are kept for a long period. 2. A main bank is not only the largest lender, but the largest shareholder to its customer firms. 3. It is often the case that a main bank dispatches its managers to its client firms. 4. When the client firm is in financial distress, its main bank intervenes in the firm and often devises a rescue plan. Miyashita and Russell (1994: pp. 49-53), on the other hand, observed the following four as the “primary functions of a main bank”: (1) Lender, (2) Stockholder, (3) Venture Capitalist, and (4) Company Doctor, meaning the “injection of new capital or debt forgiveness” (p. 52) in financially ailing firms, sending “several directors with managerial authority to help run the ailing company” (p. 52), and, in the worst case, making “a firm’s demise relatively quiet and dignified” (p. 53). Miyashita and Russell (pp. 50-1) might be a bit unique, compared with Aoki and Horiuchi and Sui, in pointing out “venture capitalist” as one of the primary functions of a main bank; they showed an example of venture capital supplied by Sumitomo Bank during the late 1970s when it supported NEC in outspending its two rivals for semiconductor plants. Considering, however, that they attributed the Sumitomo Bank’s financial support to NEC to its information management capacity, their injection of “venture capital” is considered equivalent to item 5 (the supply of management information and resources) of Aoki’s five stylized facts of the Japanese main bank system above (Aoki and Patrick 1994: pp. 5-19; Aoki and Okuno 1996: pp. 223-7)4 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Finally, Teranishi (1993: pp. 66-7) pointed out that a main bank’s lending often involves the formation of a loan syndicate or a loan consortium with other banks: Customarily, in Japan financing to firms is usually carried out by a loan syndicate, in which the main bank plays the managing role for the syndicate formed with other banks. Typically, finance given by a syndicate is for the short-term loan, most of which the main bank finances. When firms need long-term financing, three long-term credit banks play a supplemental role in providing them with long-term loans. Teranishi (1993) maintained that the loan syndication has an objective of spreading risk involved in financing across the banks. It is important to note here that, as pointed out by Horiuchi and Sui (1992: p. 1), the above main bank’s functions, especially those concerning items 3 and 5 of Aoki’s five stylized facts (Aoki, Patrick, and Sheard 1994: pp. 5-19; Aoki and Okuno 1996: pp. 223-7) are often made in the “implicit” form, in the sense that they are not often made in an explicit contract. In fact, it is uncertain client firms even know if their main banks purchase their shares, and, if they do, how much they purchase. Also, it is unclear how effective and cooperative the main bank can be in exchanging information with its customer firm, especially when it falls into financial crisis. All these actually contribute to rendering the term, “main bank,” ambiguous in nature. Accordingly, various hypotheses give economic rationale to explain all the stylized facts of the main bank system, as surveyed in the next section.4 While paying attention to the fact that the bank loan is often carried into practice by forming a loan syndicate, this chapter mainly relies on the five stylized facts pointed out by Aoki (Aoki, Patrick, and Sheard 1994: pp. 5-19; Aoki and Okuno 1996: pp. 223-7), because it is considered the most comprehensive and takes into consideration those stylized facts in the context of the client firm’s financial status. Each stylized fact is reviewed more closely in the following discussions. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 1.2.1 Bank Loans The primary factor that has constantly attracted the attention of researchers to the Japanese financial system is that the financial relationship between banks and non-financial firms in Japan is very stable. For instance, Horiuchi and Sui (1992: pp. 22-3) provide persuasive data. They selected 658 firms which operated during 1963-1983 out of the first and second sections of TSE, examining how many of them kept borrowing from their main banks between 1963-1973 and 1974-1983. They then discovered that 564 firms in 1973 and 521 firms in 1983 had borrowed from their main banks beginning in 1963. They also discovered that the firms that had changed their man banks between 1963-1973 were likely to have changed their main banks in the 1974-1983 period; 94 firms changed their main banks between 1963-1973, whereas 47 firms maintained the relationship with the same banks by 1983 (see Table 2). In a similar fashion, Aoki, Patrick, and Sheard (1994: pp. 22-3) selected 200 listed firms from TSE which had no need to borrow from banks between 1965-91 and determined how many returned to their main bank relationships in the period. Sixty-six out of 75 firms resumed bank borrowing after they terminated it for one period or two, and that only 10 out of those 66 firms changed their main banks.5 Table 2. Variations of Main Bank Relationships with Customer Firms (1963-1983) Period A: 1963-1973 Period B: 1974-1983 Number o f banks that are the same as in 1963 Number o f banks that are different than in 1963 Indeterminate cases Sum Period A = Period B 521 47 X 568 Period A ^Period B 26 12 X 38 Indeterminate (Period B) 17 9 16 42 Revivals (Period B) X X 10 10 Sum 564 68 26 658 Source: Horiuchi and Sui (1992: p. 23). R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 3 shows the high dependency of non-financial firms on the private financial institutions in Japan: Table 3. External Sources of the Non-Financial Corporate Sector (1970-1992) (In Billion Yen) Bonds Borrowing Year Total Equity Domestic Overseas Private Institution Public Institutions Commercial Paper Foreign Debt 1970 11191 980 346 10 8617 805 4310 1971 16068 842 642 -4 12485 908 1189 11972 17998 1206 354 -27 15193 845 427 1973 16914 1185 779 -37 13519 1447 -52 1974 15891 831 541 29 11558 1482 1147 1975 16855 1161 1287 325 12827 1923 -670 1976 17163 934 661 20X 13248 1639 380 1977 13259 1079 647 144 10256 1646 -514 1978 12933 1148 747 142 8724 1340 830 1979 13696 1368 901 396 8338 1962 628 1980 18313 1440 563 214 13359 1939 7602 1981 21159 21 12 1050 209 16289 2578 -1175 1982 23122 2045 632 S02 17279 2280 7 1983 22604 1564 20.s 1286 17903 1585 27 1984 25411 2229 857 1420 19953 1329 -379 1985 29772 2017 681 2346 22997 1010 739 1986 33415 2205 1579 2587 26285 553 202 1987 42675 4124 2364 3760 25772 1955 2999 1988 54193 5225 1607 4282 30112 3863 7578 1513 1989 72971 10124 1463 9513 37797 6515 780 3777 1990 66740 4438 3130 3768 50()31 0730 2696 6934 1991 38319 1277 3633 6516 23938 6365 3362 -95 1992 Mo. 1-3 2765 207 911 588 -24 558 -2690 1745 Mo. 4-6 3722 53 1277 66 -312 66 -238 -168 Mo. 7-9 10614 154 1157 -667 6637 -667 359 17 Source: Aoki, Patrick, and Sheard (1994: p. 9). Throughout the 1970-86 period, the aggregate borrowing from the private financial institutions accounted for more than 70 percent of the total external sources of funds available to non-financial firms. However, in most cases, the main bank focused on supplying short-term loans. Its long-term R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. loans accounted, in general, for only 15% or so of its total lending. Also, the degree of non-financial firms’ dependency on their main banks is relatively small; Horiuchi, Packer, and Fukuda (1988: p. 163) assessed on the basis of their definition of the main bank6 that the borrowing by Japan’s major companies listed on the first section of TSE accounted for 30-40 percent of all their borrowing from the major city banks and three long-term credit banks between 1962-1983. Similarly, defining the main bank as the one providing the largest share of loan to its customer firm, Horiuchi and Sui (1992: p. 10) showed that the loans made by the main banks to their customers listed on the first section of TSE accounted for about 20 percent in the early 1960s, about 15 percent in the late 1960s to the early 1970s, and about 10 percent after the late 1970s to 1988 (these results are replicated in Table 4). Thus, although it is true that the main bank is an indispensable lender, the client firms often need to collect funds from “the main bank’s competitors” (Horiuchi, Packer, and Fukuda 1998), some of which are the participants of their loan syndicates (Teranishi 1993). Besides, in order to supplement the short-term loans by the main banks, the firms sometimes need to rely on the long-term credit banks for their long-term loans. See Table 4 for IBJ’s (Industrial Bank of Japan) high ratios in both (A) and (B) throughout the time. The generally low ratios in both (A) and (B) across all banks in Table 4 make the main bank relationship less significant, however. This is because having the largest share of short-term loans, rather than the size of the overall loan share, is thought to be the hallmark of the main bank, as this is what defines its pivotal oversight role in the ongoing financing of the firm and at times its ability to obtain and exercise leverage over the firm’s operations and management. (Aoki, Patrick, and Sheard 1994: p. 8) 8 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 4. Changes in the Importance of Main Bank Borrowing Bank Ratio 1962 1967 1972 1977 (1982) 1983 (1986) (1988) Dai-Ichi Kangyo (A) 36.5 (N/A) 44.7 (N/A) 29.9 (12.7) 27.1 (12.3) (12.9) 26.1 (14.4) (14.3) (B) 38.6 29.7 26.5 36.8 33.6 Mitsui (A) 28.4 (16.0) 25.6 (13.6) 24.2 (10.9) 21.9 (9.4) (8.3) 19.7 (8.4) (8.4) (B) 60 4 64.8 66.9 63.8 59.2 Mitsubishi (A) 41.4 (21.2) 32.0 (14.7) 31.0 (12.9) 29.4 (12.4) (11.9) 28.0 (12.1) (11.7) (B) 54.2 52.2 60.2 57.0 51.0 Sanwa (A) 40.0 (23.4) 31.7 (15.8) 32.0 (13.6) 31.9 (13.7) (12.8) 28.1 (13.5) (12.8) (B) 41.3 46.6 39.4 42.3 41.2 Sumitomo (A) 40.8 (21.6) 35.1 (17.2) 30.3 (13.7) 28.3 (11.9) (11.6) 26.5 (12.7) (12.2) (B) 63.9 59.6 58.4 48.3 45.3 Fuji (A) 36.8 (21.7) 34.8 (17.9) 32.9 (15.1) 27.4 (12.7) (11.6) 24.5 (13.0) (12.2) (B) 35.3 43.5 41.0 43.1 38.3 Industrial Bank of (A) 39.7 42.9 38.2 40.3 37.5 Japan (B) 56.4 41.2 34.1 24.8 27.6 Tokai (A) 41.5 40.7 37.4 40.5 31.8 (B) 16.5 12.7 10.1 16.4 13.4 Daiwa (A) 55.5 56.9 40.4 40.2 34.4 (B) 28.0 18.4 15.3 14.6 13.5 Kyowa (A) 39.4 55.7 62.3 49.6 33.6 (B) 4.3 2.6 6.5 1.6 2.2 Hokkaido (A) 47.4 53.7 40.2 18.3 22.6 Development Bank (B) 2.6 2.9 3.1 1.6 3.4 Sources: Horiuchi, Packer, and Fukuda (1988: p. 163), Horiuchi and Sui (1992: p. 10). Note. Unit: %. The numbers in parentheses are from Horiuchi and Sui (1992: p. 10). Definition o f Main Bank: (a) Horiuchi, Packer, and Fukuda: “the bank that provides a particular firm with the largest loan for three or more years consecutively (in such a case, the firm is regarded as belonging to its main bank’s ‘financial group’ [‘kinyu keiretsu'])”; (b) Horiuchi and Sui: “the bank that provides a particular firm with the largest loan in a particular year.” (A) refers to the ratio o f the borrowing by the companies listed on the first section o f TSE from their main banks to their total borrowing. (B) refers to the ratio o f the main banks’ lending to their customer companies listed on TSE to their total lending. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. The main bank’s most vital role factor is to monitor its clients quite intensively within relatively short lending periods, on behalf of itself and others in the same loan consortium, to determine the viability of its clients’ projects. The monitoring activity of this sort by the main bank, in turn, contributed to decreasing the probabilities of moral hazard on the side of the client firms, and equally to avoiding costly duplication involved in monitoring costs (Aoki, Patrick, and Sheard 1994: p. 8; Hoshi, Kashyap, and Sharfstein 1991). 1.2.2 Bond-Issue Related Services The second stylized fact of the main bank system is “bond-issue related services.” Although banks and trust banks were both prohibited from underwriting bonds and equities from n Section 65 of the Securities and Exchange Law of 1948, the main bank has played a significant role in the domestic and international bond issuance as a “trustee bank (jutaku ginko).” As reported in Table 5, Japanese city banks, as trustee banks, have traditionally garnered enormous amounts of corporate bonds, although their bond holdings have declined steadily since the 1970s. Trustee banks granted Japanese city banks privileges like (a) gaining access to collateral assessment submitted by issuing firms, (b) discretion to convene creditors’ meeting as chief managers, and (c) tapping into lucrative opportunities to earn fees from assessment and custodian services. Also, those banks were often expected to play a large role in customer firms’ bankruptcies by purchasing their outstanding bonds at face value (Aoki, Patrick, and Sheard 1994: p. 12). 10 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 5. Bank Holdings of Domestic Corporate Bonds (Fiscal Year 1953-1991) Fiscal Year Absolute Value (trillion yen) Share (%) Fiscal Year Absolute Value (trillion yen) Share (%) 1953 0.1 74.7 1973 1.8 36.9 1954 0.2 75.7 1974 1.9 33.9 1955 0.2 73.8 1975 2.1 30.9 1956 0.2 68.0 1976 2.0 26.5 1957 0.2 69.0 1977 1.9 23.0 1958 0.3 68.9 1978 1.8 20.7 1959 0.4 67.0 1979 1.9 19.6 1960 0.5 53.8 1980 2.0 19.4 1961 0.6 52.7 1981 2.1 18.6 1962 0.7 52.6 1982 2.0 16.5 1963 0.7 52.5 1983 1.8 14.4 1964 0.8 51.5 1984 1.7 12.6 1965 1.0 56.5 1985 2.0 11.0 1966 1.1 56.4 1986 2.1 10.3 1967 1.2 55.7 1987 5.0 14.7 1968 1.4 55.7 1988 5.3 14.2 1969 1.5 54.6 1989 5.4 13.3 1970 1.6 51.2 1990 5.8 13.6 1971 1.7 46.0 1991 6.7 14.3 1972 1.7 41.0 Source: Aoki, Patrick, and Sheard (1994: p. 13). These kinds of bond-issue related services are mostly peculiar to Japanese financing. In Europe and the U.S., there is, in principle, no such collateral requirement on the issuing firms. Instead, they are subject to assessment by various rating agencies, and the general public makes the purchasing decision in reference to the ratings made by those agencies (Kurosawa 1999: pp. 28-56, Nihon Kakuzuke Toshi Joho Senta 1998: pp. 15-46). “Collateral principle” in Japan, according to Rosenbluth (1989: pp. 140-141), dates back to the financial panic of the late 1920s; in its aftermath, the MOF (Ministry of Finance), in pursuit of broadening and strengthening its ability to oversight the Japanese financial circle, drafted a new Banking Act in 1927,9 forming a Bond Issue Arrangement Committee (Kisai-kai) comprised of about thirty banks.1 0 The Committee’s motto was 11 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. “Corporate bonds shall not be issued without sufficient collateral” (Rosenbluth 1989: p. 140)." This tendency of choking up the bond market in the encouragement of indirect financing by banks was further strengthened by the war efforts starting from the 1930s and then carried over to the postwar Japanese economy; in exchange for the introduction of Section 65 of the Securities Exchange Law, which prohibited banks from the underwriting business, the Bond Committee and its collateral principle, once abolished in 1945, were reinstated. The Temporary Interest Rates Adjustment Law (Rinji Kinri Chosei-Ho) of January 1947 placed the Bond Committee inside the Bank of Japan along with other representatives from the MOF, the Bank of Japan (BOJ), and the Economic Stabilization Board (Rosenbluth 1989: p. 143; Ueda 1999: p. 57).1 2 With the auspices of the Law, the Federation of Bankers’ Association and the Regional Banks Association could hold down interest rates so that bonds issued by the government and the IBJ for the purpose of economic 13 recovery would be efficiently absorbed by other city banks. When the Bank of Japan retreated from the credit allocation work in 1955, the Bond Committee exclusively began to screen corporate bond issuance. Its role became more pronounced with a tight monetary policy in 1957 (Nihon Koshasai Kenkyujo 1995: p. 151; Ueda 1994: p. 98). The main purpose of the Bond Committee around that time was to not crowd out the bond issuance from the heavy and chemical industries such as electric utility and marine transportation (Aoki 1994c: pp. 113-7; Rosenbluth 1989: p. 143; Ueda 1994: p. 98). Showing that, in 1955, the proportion of funding by the government financial institutions and long-term credit banks (notably, the IBJ) represented about a half of new funding sources, Aoki argued that city banks’ ex-ante monitoring on new investment projects was complemented in a critical manner by those banks (pp. 113-7). After the first oil crisis of 1973 hit the Japanese economy strong enough to lower its real GNP growth rates from 10-12 percent to 3-6 percent, and after firms left for the Eurobond market to avoid onerous interest payments on their bank debts, the main banks began to follow their customers in Europe to act as “co-managers” of bond issuance through their subsidiaries in Europe. They could 12 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. do so because there was no restriction comparable to Section 65 of the Securities and Exchange Law in Europe, although such a trial was often circumvented by the so-called “Three Bureau Agreement” (sankyoku shido, 1975) among the MOF’s Banking, Securities, and International Finance Bureaus (Aoki, Patrick, and Sheard 1994: p. 12; Rosenbluth 1989: pp. 147-163).1 4 In the early 1970s, the Euromarket accounted for 1.7 percent of Japanese corporate financing, but the figure rose to 19.6 percent in the second half of the 1970s (Rosenbluth 1989: p. 149). After the “Yen-Dollar Committee” (November 1983) ended in May 1984, the Euroyen bond issuance skyrocketed; the total outstanding amount of Euroyen bond issues surged from 70 billion yen in 1983 to 1.6 trillion yen in 1986.1 5 All this impacted on the Bond Committee, and Japanese banks were forced to ease the conditions for domestic bond issuance. By the end of 1984, 20 companies became eligible to issue straight bonds, and 100 qualified for convertible bonds. O f the latter, nearly half were permitted to issue convertible bonds without collateral. In March 1998, the conditions were further relaxed, and 180 firms were allowed to issue unsecured domestic bonds. Banks assented to further gradual relaxation of domestic bond issuance rules by introducing a new rating system to be used along with the Bond Committee’s issuance conditions (Nihon Koshasai Kenkyujo 1995: pp. 158- 161; Rosenbluth 1989: pp. 159-163).1 6 In sum, banks’ privileges in the bond-issue related services have been on the constant decline, which, in Table 5, is clearly exhibited in their steadily decreasing bond holdings since the early 1970s. Banks, however, did not stand by idly and struck back against the securities industry, trying to regain their cut by winning the long-prohibited underwriting businesses. After seven years of the intense debate at the MOF, they eventually came to a compromise with the securities sector, with the consequence of the June 1992 enactment of “the Financial System Reform Law,” which allowed them to underwrite both bonds and equities through their securities subsidiaries. Banks’ revival in this area is yet to be seen, as they are now competing with securities firms on virtually the 17 same ground the law provided with them. 13 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 1.2.3 Stockholdings The Japanese Anti-Monopoly Law allows banks to hold up to 5 percent of corporate stocks, and, typically, the main bank holds its customer firm’s shares large enough to be ranked as the top shareholder, and, if not, at least, among the top five shareholders (Aoki and Okuno 1996: p. 223; Aoki, Patrick, and Sheard 1994: pp. 12-4). Besides, the main bank seldom sells off the customer’s shares unless it abandons its main bank position. Although the upper limit of a banks’ stockholdings is set at 5 percent, the main bank often mobilizes other affiliated firms to hold more. Such mobilization is pronounced, especially when the customer firm is put under the pressure of hostile takeover (Aoki, Patrick, and Sheard 1994: p. 14). During the postwar period, banks’ holding of corporate stocks was often carried out in the context of “cross-” or “mutual” shareholdings with their customer firms, which, as those terms connote, is attained by their mutual purchases and subsequent holdings of their equities, and such an action itself expedited the counterparts among non-financial firms sharing the same main bank. All these mutual shareholdings then contributed to the creation of various corporate groups or 18 “keiretsu,” a phenomenal feature of the postwar Japanese economy. Cross-shareholding and “keiretsufication” had been intensively pursued in the Japanese economy during the 1950s throughout the 1960s. The first round of the movement was set in motion by the U.S. Occupation Force’s policy shift to strengthen the Japanese economy to serve as a bulwark against communism; first, while starting to disseminate the stocks owned by the zaibatsu corporate groups through the Holding Company Liquidation Commission (mochikabu gaisha seiri iinkai, HCLC), the Occupation 19 Force made a decision not to dismantle the zaibatsu banks and kept them virtually intact. No sooner had the Peace Treaty been concluded between Japan and the U.S. in September 1951 than the former zaibatsu groups began to re-assemble; shortly after the conclusion of the Treaty, the measure that had forbidden the former zaibatsu firms and family members to hold the shares issued by their 14 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. former affiliates was officially lifted (Teranishi 1993: p. 85). The human networks also started to revive in the form of Presidents’ Councils (shacho-kai), in which the executives of the former zaibatsu firms gathered in line with the same zaibatsu lineage, and held meetings on a regular basis. These councils gradually became consolidated as they were given the economic rationale by cross shareholdings, which themselves were pursued intensively when those firms tried to circumvent takeover attempts gradually becoming rampant through the liquid securities market.2 0 For its part, the Ministry of International Trade and Industry (MITI) played a significant role in promoting cross shareholdings by amending the Antimonopoly Law in 1953, whereby banks’ owning corporate 21 stocks were allowed to increase from 5% to 10%. The second round began in the mid-1960s, when the securities market was haunted by serious depression, and when the Japanese government was preparing to join the OECD in April 1964: in face of the securities market depression, two institutions were founded, respectively, in 1964 and 1965, for the major purpose of absorbing the excess supply of securities. One was the Japan Cooperation Securities (nihon kyodo shoken), founded under the collaboration among major city banks, securities firms, the IBJ, and the Bank of Japan, and the other, Japan Securities Holding Institute (nihon shoken hoyu kiko), founded among major securities firms and the Bank of Japan. They both collectively absorbed 5.1% of the total stocks in circulation, but to no avail. In May 1965, 22 Yamauchi Securities was eventually forced into bankruptcy. The grave experience in the securities market prompted large listed firms to further intensify their mutual stockholdings. Japan’s joining the OECD also stirred the fears among major Japanese firms that they might be taken over by foreign firms, thereby spurring them toward further cross-shareholdings. The lax financial policy prior to the “Nixon Shock” of 1971, and the subsequent inundation of foreign currency also contributed to strengthening the crossholdings (Nakajima 1991: p. 46). Consequently, by the early 1970s, the corporate sector had come to possess about sixty percent of all the traded stocks (Miyamoto 1999; pp. 141-142). 15 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. The third round was in the late 1980s, the period of a swelling bubble economy. Major corporations started to issue an enormous number of bonds in the form of “equity financing,” expecting that the price hike of those bonds would soon lead to their threshold values with which to transform them into equities. Once transformed into equities, they were exchanged in the context of cross-shareholdings within each keiretsu. Tsuru (1994: p. 30) reported that, as a consequence of the frantic equity financing during the bubble economy period, in 1993 the corporate sector had come to hold about 70 percent of all stocks, in which holdings by the non-financial and financial firms, respectively, accounted for 23.9 and 40.8 percent, respectively, whereas individuals, only 23.7 percent. As will be reviewed and discussed in depth in Chapters 3 and 4, in reaction to the frenzy speculations during the bubble economy period, banks and non-financial firms started to intensively decrease their cross-shareholdings from the late 1990s, and now the level of their cross shareholdings is even less than that of the pre-bubble economy period. Together with the Big Bang financial market reform of 1996, which espoused the active, liquid securities market, cross shareholdings now seem to be constantly waning. But the recent revision of the Investment Trust Law (December 2004), which allowed banks to engage in the trust business, opened another path for banks to access the securities market. Besides, there is currently a movement toward the abolition of the five percent stockholding ceiling of the Antimonopoly Law, which, in the event of implementation, will provide banks with a much higher degree of freedom to hold stocks than ever. It is still important to watch how the stockholding function of banks will develop as these events unfold. 1.2.4 Payment Settlement Account As pointed out in Section 1.2.2, when the main bank provides a loan, it often involves the formation of a loan consortium with other non-main banks. Once the loan is in the consortium, 16 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. however, the main bank in the loan consortium makes a significant difference compared to other lenders, because all the financial transactions of the borrowing firm go through its payment settlement account set up at the main bank, and not others (Aoki, Patrick, and Sheard 1994: p. 14; Aoki and Okuno 1996). Another trait that distinguishes the main bank from other lenders is its offering such services to the borrowing firm as issuing promissory notes (normally three months in duration) and dealing with foreign exchange businesses (Aoki, Patrick, and Sheard 1994: p. 15). All these functions provide the main bank with useful information that other lenders in the consortium cannot access; by keeping track of all the monetary transactions of the borrowing firm reflected in its payment settlement account and other service subscriptions, the main bank can effectively monitor if the firm is fulfilling its investment project as originally planned, and if it is in a sound financial state. 1.2.5 Information Services and Supply o f Management Resources. Aoki and Okuno (1996: p. 224) and Aoki, Patrick, and Sheard (1994: pp. 15-6) point out that the main bank often participates in the management of its client firms by sharing information regarding the acquisition and liquidation of their business assets and their business partners by having its executive sit as a director or as auditors on the board of the client firms. The main sources of bank directors and auditors are those those forming the core of the Japanese main bank system such as the large city banks and the IBJ. For instance, Sheard (1985: pp. 16-8) reported that, in 1982, 1773 listed firms had a total of 1676 directors on their boards who were formerly or concurrently bank executives, and that in 1980 about half of the firms listed on the first section of the Tokyo Stock Exchange had on their boards at least one executive from their main bank. The close managerial relationship between the main bank and its client firm is often accompanied by long-run financing. Before the mid-1970s, this practice was prevalent even in large firms when they were more dependent on bank loans. But as many firms started to stay away from their main banks 17 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. after that, the bargaining power tipped the balance toward the customer firms, and the practice has is now applied to those firms whith less than excellent financial status such as small and medium-sized firms. 1.3 State-Contingent Nature of the Main Bank Relationship The above main-bank-relation items 1, 2, and 5 in Table 1 (bank loans, bond-issue-related, and supply of management information and resources, respectively) should be understood in combination with four financial states (excellent, favorable, normal, and critical) because they critically depend upon the financial states of the client firm (Aoki and Okuno 1996: pp. 224-7; Aoki, Patrick, and Sheard 1994: pp. 16-9). Firms with excellent financial status (E firms) are equipped with ample accumulated reserve, so that they do not have to rely on bank loans very much. They would rather limit their use of banks to payment settlement and occasional bond issuance. Obviously, in this case, the bargaining power is for the firms’ side, and the competition among banks over gaining the main bank position with E firms is likely very harsh, as it would provide more lucrative opportunities for the bank.2 3 Firms not so good as E firms, but somewhat favorable in financial position (F firms), occasionally need a loan from their main banks, but they also have the ability to raise funds by issuing bonds in both domestic and overseas securities markets. As explained in Section 1.2.2, the main bank usually serves as a trustee bank for the F firm when the firm needs to issue bonds in the domestic market, and as co-manager in the case of overseas markets the bank earns management fees. The majority of firms belong to the N-firms category: N firms, by definition, do not have enough internal reserves for investment and do not have the ability to issue bonds on their own, so 18 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. they necessarily need to rely on the loans supplied by their main banks. Besides, these firms are often eager to accept directors from the main banks on their boards. The main banks’ unique function is even more pronounced in the case when their client firms are put into a critical financial status and require an additional injection of funds for survival (C firms). If its client firm is identified as a C firm, the main bank often dispatches its mangers as the residual claimant (stockholders) and begins to screen whether the firm deserves a rescue plan (additional financial supply, waiver, or rollover of the interest and principal payments partly or entirely)2 4 or descent into bankruptcy. When a rescue plan is granted, it almost always entails a replacement of the firm’s executive by the main bank. Some C firms are encouraged to merge with another firm under the lead of the main bank.2 5 1.4 The Theories of the Japanese Main Bank System This section reviews three important theories that explain the salient features of the main bank system, as examined in the previous section. This task is indispensable in discerning the achievements and problems involved in the present theories on the Japanese main bank system. The theories that will be subject to examination are (a) the agency cost theory, (b) the “contingency governance of teams” (1994), and (c) the implicit contract theory. They are all considered good trials of capturing the saliencies of the main bank system, and have had great impact on the subsequent works.2 6 The review of these theories is solely to identify the problems involved in them and, thereafter, build another model on the Japanese main bank system. 1.4.1 The Agency Cost Theory The salience of the main bank system is often made clear by comparing the two types of firms: ones which maintain close ties with their main banks, and the others, which keep a very weak relationship with, or even completely independent of, banks. It has often been pointed out that the 19 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. firms with close ties with their main banks are likely to avoid bankruptcy when they face financial distress, and show faster recovery than those without a main bank relationship. For example, Suzuki and Wright (1985) showcased the close financial relationship maintained between Mazda Motors and Sumitomo Bank; the first oil crisis of 1973 hurt the profitability of Mazda, which saw a huge decline in the demand for their gas-guzzling, rotary-engine cars. In response to Mazda’s plight, Sumitomo Bank implemented a rescue plan, which involved a new loan at a favorable rate, sending several top executives and the advice to sell some of its shares to Sumitomo Bank. In parallel, Sumitomo Bank urged other Sumitomo firms, notably Sumitomo Trust Bank and Sumitomo Corporation, to take part in this rescue plan. In response, Sumitomo Trust Bank dispatched its management team to Mazda; and Sumitomo Corporation, the group trading company, took charge of the issue of distributing the automobiles produced by Mazda. Furthermore, these three firms supported the boost of Mazda’s sales by promoting its automobiles among their customer firms, suppliers, and employees. Due to all these endeavors by Sumitomo firms, Mazda eventually turned around into a profitable company (Sheard 1985; Sheard 1994c: pp. 220-1; Hoshi, Kashyap, and Sharfstein 1990: pp. 73-4; Hoshi, Kashyap, and Sharfstein 1991: p. 39; Hoshi 1994b: pp. 295-6). Showcasing the four largest bankruptcy cases (Kojin, Eidai Sangyo, Sawa Shokai, and Rikkar) in postwar Japan, on the other hand, Sheard (1985) pointed out that the weak bank-firm relationship was the common factor that significantly contributed to these failures. According to his research, no bank was willing to step in and organize a rescue process for any of these four cases. The above observations necessarily lead one to consider what incentive(s) would induce banks to break into their financially troubled client firms and what mechanism would work in the main bank’s bailing out process. One theoretical framework addressed in the earlier period to deal with these questions is the “agency cost theory” by Jensen and Meckling (1976). According to Jensen and Meckling, the agency cost necessarily arises under the “agency relationship,” which is defined as “a contract under which one or more persons (principals]) engage another person (agent) to perform some service on behalf which involves delegating some decision making authority to the 20 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. agent” (p. 308). Such an agency relationship, when translated into modem corporate organization, turns into the one between owners and managers of a firm, and, accordingly, the agency cost of a modem corporation primarily originates from the separation between its ownership and control (management). More plainly put, “outside equity ownership” dilutes the owners’ stakes, thereby giving the manager the incentive to acquire perquisite in the form of corporate payment. Adding to this is the “corporate debt” that engenders the possibilities of (a) the cost of a highly leveraged firm to commit moral-hazardous actions under the limited liability of the firm, (b) the outside monitoring cost to prevent such actions, and (c) the bankruptcy cost, which involves difficulty for the firm to negotiate simultaneously with all of its creditors and the problem of holdout creditors who try to free-ride on the other creditors’ negotiations for e improved future cash flows. In light of the theory of the agency cost as explained above, the Japanese main bank system seems, in many respects, to have contributed to considerably reducing the agency cost associated with corporate financing. First of all, as pointed by Gilson and Roe (1992: p. 10), the main bank is a solution to the problem inherent in the separation of ownership and control of a firm posed by Berle and Mean in 1932:2 7 Serving as the major shareholders to its customer firms, the main bank could effectively avoid the problem of dilution of the outside equity ownership. Such an ability of the main bank is even more pronounced when it is taken into account that other shareholders of the client firms are under the strong influence of the main bank. Second, by dispatching the top executive to its client firms, the main bank has successfully filled in the information gap with them, thereby reducing such problems as the perquisite acquisition and hazardous moral actions, on the part of the client firms. Third, by being simultaneously both the major shareholders of, and the major lenders to, its client firms, the main bank has effectively appeased the problem of the conflict of interests that could have otherwise arisen between shareholders and lenders, especially when the firm concerned is under the bankrupt state.2 8 Taking all these together, the Japanese main banks are largely considered to have functioned as the mechanism substitute for the missing takeover market in Japan (Gilson and Roe 1992: p. 10). 21 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Probably, the most notable examples of applying the agency cost theory to the area of the Japanese main bank system are Hoshi, Kashyap, and Scharfstein (1990, 1991), who conducted empirical analyses of the relationship between the corporate investment activity and the main bank’s financing in order to delineate such an aspect as characterized by the agency cost in the main bank system. Their basic findings were that (1) the firms with main banks are likely to recover faster from the state of financial distress than otherwise, and that (2) such firms can often undertake long-term investment activity without suffering from the restriction of their internal funds. From these findings, Hoshi, Kashyap, and Scharfstein (1990; 1991)concluded that their initial assumption was firmly supported that the firms with strong ties with their main banks have easier access to external financing, that is, the lower agency costs associated with the main bank financing, which, in turn, mitigate the problems of financial distress and the shortage of internal funds. In the first work, Hoshi, Kashyap, and Scharfstein (1990) collected a data set of 125 manufacturing firms which experienced financial distress, which itself is defined as the coverage ratio less than one, for two consecutive years during 1978-1985, and sorted them out into 45 “group firms” and 80 “non-group 29 firms,” following Naktani’s (1984) classification of the group. What they first found was that the average annual investment rate of the group firms in the year immediately following the financial distress was much higher than that of the non-group firms (Hoshi, Kashyap, and Scharfstein 1990:: p. 80). Next, they regressed the investment rates of the 125 firms over the three years immediately after their financial distressed period on the following three variables: (1) the group dummy (GROUP), (2) the rate of bank borrowing that comes from the firm’s largest lender (TOPLEND), and (3) the rate of the firm’s outstanding shares held by the largest lender (SHARE). The results were quite supportive of the agency cost theory, in that the coefficients were both positive and statistically significant for GROUP and TOPLEND (p. 81).3 0 In the second work, Hoshi, Kashyap, and Scharfstein (1991) turned their attention to another question of how liquidity affects investment activities more for the non-group firms than for the group firms. The problem involved in this 22 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. approach is that liquidity serves as a proxy for other determinants, notably the profitability of investment. One way to get around this problem is to incorporate into the regression equation both the variables of liquidity and Tobin’s q, a proxy for the expected profitability. If liquidity constraints are unimportant, Tobin’s q will turn out to be the sole determinant of investment. Otherwise, the theory of agency cost predicts that the liquidity coefficient of the regression equation for the non group firms will be higher than that for the group firms. Again, following the Nakatani’s (1984) classification, Hoshi, Kashyap, and Scharfstein (1991) compiled the panel data sets of 121 group firms and of 83 non-group firms from the manufacturing industry in the period of 1965-1986, and regressed their annual investment rates on (1) cash flow, (2) short-term securities, (3) Tobin’s q, (4) two yearly dummies (the firm dummy and the yearly dummy), and (5) lagged production.3 1 The result was that, as the theory of agency cost predicted, the coefficients of both liquidity variables, that is, (1) and (2), are much larger for the non-group firms than for the group firms and statistically significant, whereas the coefficients of Tobin’s q are almost the same for both the group and non group firms. From these findings, Hoshi, Kashyap, and Scharfstein (1991) concluded the Japanese main bank system contributed to mitigating the capital market imperfection. Also, in concluding, Hoshi, Kashap, and Scharfstein (1991: p. 57) re-emphasized the importance of the financial institution’s holding the majority of both equities and credits of its clients in reducing the agency cost. Unless this condition is met, the shareholders—and subsequently the managers, too—might exploit their status of limited liability to run a risky project, and try to induce the bondholders by raising the interest rate to counteract their moral hazardous action. Thus, ultimately, the coordination between the shareholders and the bondholders (creditors) becomes important in reducing the agency cost. In this regard, the authors argued (p. 57) that the U.S. Glass-Steagall Act is a costly statue in prohibiting the U.S. banks from owning equities issued by firms: under the Act, U.S. banks would find it inefficient to exercise control over the client firms since their main objective (i.e., ensuring the repayment of loans plus interest) is significantly 32 different from that of shareholders. In contrast, the Japanese main banks are the paragon of 23 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. appeasing the conflict of interests that might arise between these two parties, and can focus on the monitoring over their client firms all the more for that. 1.4.2 The Contingent Governance o f Teams While the agency cost literature by Hoshi, Kashyap, and Scharfstein (1990, 19991) successfully showed that the close main bank relationship often contributes to allowing the financially distressed firms to tap into a cheaper money supply from their main banks for recovery, it comes to a standstill at the question of why some firms, even with their maintaining close main bank relationship, still fall into bankruptcy. For that question to be answered, the “theory of the contingent governance of teams” presented by Masahiko Aoki (1994a) seems the most prospective as of today, as it explicitly takes into consideration the main bank’s incentive scheme to be incorporated in the customer firm’s surplus maximization problem, in which one option is to allow the firm to bankrupt. In Aoki’s model, there are three agents: a shareholder, a monitor (bank), and a team (firm). A team consists of homogeneous workers and solely engages in production. What distinguishes Aoki’s model from that of the agency cost theory is his incorporation of the “team” as the production agent. Aoki set up the “team” in his model because the mode of the team production is prevalent across Japanese firms, where the job demarcation is often not clearly defined among workers (Aoki 1989a, 1994d, 1995a).3 3 Under such a condition, there is always a possibility of shirking on the side of workers in the team. In order to minimize the level of shirking inside the team, the monitor (i.e., the bank) necessarily has to devise the incentive scheme that properly reflects the wage contract. Such a wage contract, however, needs to be linked to the total amount of product that the team produces, because that is the only way the monitor can assess its performance. In order for the team production to be maximized, two other agents (the shareholder and monitor) 24 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. are also subject to their own reward schemes. The optimal wage contract schemes depending on the three possible states of the team’s performance (“poor,” “critical,” and “good”) obtain as follows:3 4 1. The “poor” state of the production: a. The team is dismissed out of the firm after receiving the minimum wage (the bankruptcy of the firm). The workers of the team suffer from this dismissal because their contextual skills acquired in the firm cannot be easily applied to other firms. b. The shareholder receives the minimum residual, which is smaller than those he receives in the “critical” and “good” states of the production. c. The monitor incurs cost in this state, because, in Aoki’s scheme, the amount of the product is short of payments for the minimum wage and residual for the team and the shareholder. In this state of the production, the monitor declares bankruptcy of the firm. Along with the minimum wage payment, this serves as the “stick” for the team. 2. The “critical” state of the production: a. The ream receives only the minimum wage. b. The shareholder receives the “normal” amount of residual, which is larger than that he receives in the “poor” state of the production. c. There are two regions for the monitor in this state: the one allows the monitor to receive the positive residual after payments for the team and the shareholder. The other incurs cost to the monitor because the amount of the product is short of money for payments for the team and the shareholder. 25 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 3. The “good” state of the production: a. The team receives all the residual less the normal residual for the shareholder. This serves as the “carrot” to the team. b. The shareholder receives the “normal” residual. c. The monitor receives nothing. By not receiving any surplus, the monitor grants the excess residual as a reward to the team. Aoki (1994a: pp. 671-3) argued that this “state-contingent nature” of the product distribution well captures the essential aspect of the Japanese main bank system, and that the main bank system functions in a complementary fashion to the team nature of production widely accepted in Japanese firms.3 5 In relation to the model of the contingent governance of teams, Aoki pointed out two vital factors, both of which also have ramifications for the better understanding of the contemporary Japanese economy. One is the “firm-specific nature” of the Japanese workers, and the other, the role of the government regulation in creating the “main bank rent.” First, it is important to recall that the firm depicted in Aoki’s model as a team composed of homogeneous workers is actually a mirror image of the typical Japanese firm, where the workers are expected to develop the firm-specific skills in the context of their lifetime tenures. On the team nature of the Japanese firms, Aoki (1994a) explained: [T]he Japanese firm has developed a type of internal organization which facilitates lateral coordination among different task units on the basis of information sharing, joint responsibilities, and help. These cross-functional interactions permit firms to gain comparative advantage in certain manufacturing industries by quickly adapting to continually changing technological and market environments, if not under stationary or radically changing environments. One possible consequence of the development of such an internal organization has been the manifestation of strong team nature, (p. 672) 26 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. The consequence of the workers’ learning and acquisition of the firm-specific skills is that such skills cannot be easily transferable to other firms (Aoki 1989a; 1994a: p. 672; 1995a: pp. 16, 72; Aoki and Okuno 1996: pp. 209-10). Losing a job in the economy where the firm-specific skills are the essential ingredient of employment is virtually “punishment” to the worker.3 6 Under such an employment structure, it is just a matter of course that most workers choose to stay in the same firms during their lifetime employment tenures, while always trying to attain the highest level possible in their performance. Thus, concluded Aoki, “the main bank system,” “the team production” and “the lifetime employment system” (shiishin koyo seido) are “complemental” to each other (Aoki 1989a; 1994a; 1994c; 1995a; Aoki, Patrick, and Sheard 1994; Aoki and Okuno 1996). Second, Aoki tried to find the effect of various government regulations in their sustaining the Japanese main bank system. More specifically, he considereds how the government regulations have catapulted the main banks’ incentive to rescue their financially ailing firms even if they need to incur cost in doing so (recall that the main bank has to assume cost in the bad and critical states). Aoki (1995a: pp. 121-152; and Aoki and Okuno 1996: pp. 213-220, pp. 237-244) pointed out that the following regulations have contributed to generating the “main bank rent,” whereby to enhance the banks’ incentive to rescue their financially distressed firms: (1) holding down the deposit interest rate, while keeping its real rate positive by controlling the inflation rate (see the “Temporary Interest Rates Adjustment Law” [Rinji Kinri Chosei-Ho] of January 1947 in Section 1.2.2); (2) restriction of the bond issuance market (see Section 1.2.2); (3) restriction of the entry into the banking sector, while excluding banks out of the dealing and underwriting businesses; and (4) adopting the carrot-and-stick policy for improving the performance of banks, which often took the forms of authorization of opening a new branch and of sending the former the MOF officers as executives. Considering that banks took over the reign of the Japanese financial market at the expense of the development of the securities market for almost all of the postwar period, it is quite natural to conjecture that these regulations have created some sort of “rent” in the banking industry, 37 thereby providing banks with the leeway to sustain the firms at the brink of bankruptcy. Besides, 27 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. the concept of the main bank rent also wins relevance in explaining the weakening governance of the main banks after a high growth period. As seen in Section 1.2.2, the financial deregulation that progressed, especially in the form of relaxation of the bond issuance conditions, contributed to significantly reducing the main bank rent, and, thereby, considerably sapping the main banks of their capability to govern their customer firms. Such weakening governance of the main banks, according to Aoki, culminated in the bubble economy period of the late 1980s, when many large Japanese firms went through the lacuna of their corporate governance by their main banks (Aoki 1995a: pp. 145-8; Aoki and Okuno 1996: pp. 243-244). Providing a solid theoretical foundation for how a series of the financial deregulations eventually led to the emergence of the bubble economy in Japan, the theory of the contingent governance of teams and the concept of the main bank rent have both gained the credibility across the wide range of economists today. 1.4.3 The Implicit Contract Theory One topic often discussed in the area of Japanese economy study is the uniqueness of the Japanese style contract. This uniqueness is distinguished largely by its being a little more than a legal boilerplate that only expresses both parties’ intentions for engaging in the mutual, beneficial exchange relationship. This forms a sharp contrast, especially with the Anglo-American style contract, which adopts the legally-binding, formal written style and specifies every term according 38 to all the contingencies possible in the future. Due to the lack of such contractual specificity as found in the Anglo-American contractual system, the Japanese contract is often pointed out to leave some vagueness and interpretability between the parties concerned. The contract between a bank and a firm is not exceptional. While the contract usually clearly states the terms of loans such as the interest rate and the due date, it still opens to the vagueness for the contract renewal and the rescue plan, as discussed in depth in the previous section. With all these observations, it is quite natural for 28 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. many researchers to direct their attention to the “implicit contract theory” in trying to elicit the economic motives of banks and business firms that might be tacit behind the contract. The first work that paid attention to this area in the connection with the Japanese main bank system is presented by Nakatani (1984), who went beyond the relationship between banks and firms, and placed much more emphasis on the “industrial group” or “keiretsu” where all sorts of risk-sharing is assumed across all the members belonging in the same group.3 9 For the purpose of clarifing the difference between the financial keiretsu firms (those belonging to one of the following: Mitsui, Mitsubishi, Sumitomo, Fuji, Dai-ichi Kangin, and Sanwa) and non-keiretsu affiliated firms between the period of 1971-82, Nakatani, through his regression models, found not only that the group affiliated firms have less chance of exposure to the interest rate variations (Nakatani 1984: p. 242), but also that (1) the group-affiliated firms’ objectives are placed not on the maximization of their profits and sales growth, but on the improvement of compensation for their employees, (2) the improvement of compensation for the employees in the group affiliated firms often entails their low-dividend policy, and (3) the group-affiliated firms have higher debt-equity ratios than non-affiliated firms. Putting (l)-(3) together, Nakatani concluded that there exists an “implicit mutual insurance mechanism” in the keiretsu groups by which some idiosyncratic shock is not only borne by the firm itself, but absorbed, in varying degree, by the other members, including the main bank (Nakatani 1984: pp. 244-5). Furthermore he asserted that “‘(the) main bank’ usually acts as an insurer of group-affiliated companies, and when the latter are faced with financial and/or managerial difficulties, the former renders necessary assistance even beyond the level that normal business reciprocity requires” (Nakatani 1984: p. 241). The main banks of the industrial groups can do so because the group-affiliated firms are less likely to go bankrupt with their stabilized profit- and sales-growth rates and interest rate variations. Later on, Nakatani’s work was succeeded and further explored by Sheard (1985, 1991a, 1994d). In the perspective of implicit, mutual risk-sharing of the industrial groups, he argued that the “lifetime employment system” (shiism koyo seido) and the “seniority wage system” (nenko chingin seido), both of which are the two prominent features of 29 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. the Japanese labor market, are the outcomes of the industrial groups’ pooling member firms’ risks through the internalization of the capital market through cross-shareholdings. In parallel, he proceeded, the main banks of the industrial groups play a significant role in disciplining the group firms’ managers in lieu of the capital market that has ceased to function for that role as a result of cross-shareholdings. Sheard (1991a) then traced those Junctional roles played by the group firms and the main bank in the process of scrapping the aluminum production firm in each group during the late 1970s to the late 1980s (Sumitomo Aluminum Smelting in the Sumitomo group [Sumitomo Group], Mitsubishi Light Metal [Mitsubishi group], Showa Aluminum Industries [Fuyo group], Sumikei Aluminum [Sumitomo group], Mitsui Aluminum [Mitsui group], and Nippon Light Metal [Dai-Ichi Kangin group]). His major finding was that the main banks of the industrial groups offered sizable help to their group aluminum producing firms in their downsizing processes, particularly through bearing disproportionately high costs such as interest reduction and exemption (Sheard 1991a: pp. 23-4, 27). Although starting from the motivation to develop Nakatani’s idea of implicit, mutual risk-sharing in each of the industrial groups, Sheard’s (1985; 1991a; 1994d) works have, thus, come to largely share with Aoki’s concept of “institutional complementarity,” the core idea of which is that various economic institutions, including markets, function complementarily, not independently, of each other.4 0 It is much to the credit of Nakatani and Sheard that they paid attention to the industrial groups or keiretsu in Japan and identified the most essential role of the main bank in each of these groups in the framework of implicit, mutual contract. There is, however, a fundamental flaw involved in such a framework; in and of itself, the implicit contract theory is a black box, and not able to give the mechanism on how the risk-sharing is enforced among affiliate firms; Sheard, himself, seemed to realize this issue when he pointed out, “There is the issue of enforcement of the risk-sharing arrangements in adverse states of nature, given that to varying degrees they are implicit in nature or based on incomplete contracts the precise details of which get filled over time” (Sheard 1991a: p. 28). He then conjectured that such enforcement comes from the equilibrium behaviors by 30 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. the participants in the context of a long term repeated game in which firms are establishing and maintaining reputations for trustworthy behavior (Sheard 1991a: p. 28; 1994d: p. 21). Contrary to Sheard, Ramsayer (1994) asserted that there is no such thing as an implicit contract between the main bank and its client firm. He, instead, presented the raison d’etre of the main bank’s rescuing its financially ailing firm from a legal viewpoint. First, he considered whether the firm would want to purchase insurance in the form of the implicit contract with its main bank, and concluded that it would not. This is because, if the bank is to offer insurance against the firm’s failure, it will continuously raise its premium until the safest firm disappears from the insurance market (adverse selection). He then analyzed that the main bank would not want to sell such insurance, either; if the bank sells such insurance to the firm committing a morally hazardous action, that will benefit not only the manager of the firm through the channel of shareholders’ paying supra- market salary to them, but also eventually those shareholders. Adding to these two considerations, Ramsayer continued, is the fact that Japanese banks seldom rescue the small and medium-sized firms, which generally run less diversified operations than large firms: if banks and large firms can negotiate mutually advantageous implicit insurance, so should banks and small and medium-sized firms. Having analyzed all these cases, Ramsayer argued that the lack of “the doctrine of subordination” in Japan often encourages Japanese banks to intervene in their financially troubled firms, even though there was no rescue contract concluded at the outset. The doctrine of equitable subordination, which is firmly established in the U.S. under the buttress by courts, often prevents American banks from intervening in the financially troubled firms against the possibility that the banks restructure the debtors to their advantage. To the contrary in Japan, the lack of such a doctrine makes the banks’ ex ante threat to the borrowing firms less credible, and gives even the most ruthless bank an ex post incentive to rescue the troubled firms, especially when the latter owes huge debts (Ramsayer 1994: pp. 247-52).4 1 Ramsayer’s argument on the doctrine is, in fact, quite relevant, especially in regard to how the bad loans were disposed of during the post bubble economy period of the late 1990s through the early 2000s: when large banks started to dispose of their bad 31 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. loans as the issue became imminent in the late 1990s, those bad loans were mostly the kinds that were financed by either only one bank or a quite small number of banks joining in the same loan syndicates. Once those kinds of bad loans were largely disposed of, then the pace of the disposition slowed down again, as a larger number of banks in the same syndicates were often held up with each other in which one of these should make a first move in the disposition. This issue prolonged awhile with no recourse to the doctrine of equitable subordination (see, for example, Wada 2003:: pp. 74-5, 180). 1.5 Concluding Remarks This chapter first confirmed the chief roles of the Japanese main bank and their state- contingent nature. Those roles, among all the main bank’s rescuing the financially ailing customer firm, are incorporated in the following three notable theories: (1) the agency cost theory, (2) the contractual governance of teams, and (3) the implicit contract theory. The first two are both concerned with micro-analysis of the relationship between the main bank and its client firm, whereas the third one places the main bank in the industry-wide, keiretsu transactions. The main contentions and problems involved in each of these theories are summarized in the following: 1. The agency cost theory: the close tie between the main bank and its client firm often allows the latter to easily access a money supply with a cheap agency cost (represented in the form of interest rate) from the former. Besides, the problem arising from the liquidity availability is much less severe for the keiretsu firms than the non-keiretsu firms. One vital factor that enables these is the main bank’s stockholdings of its client firm, whereby the main bank disciplines the executives of the client firms and reduces the agency cost involved (the “Berle-Means” problem). Also, concentrating stockholdings and lending in the hands of the main bank often contributes to solve the 32 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. issue of conflict of interest that might otherwise be serious between stockholders and bondholders. 2. The contractual governance o f teams: The main bank plays an essential role in monitoring its client firm, where the performance of individual workers is not discemable (team production). This is actually the main characteristics that most of the Japanese firms share. For each of three possible states of the firm’s performance (poor, critical, and good), the payment schemes are designed for the three agents (team, shareholder, main bank) as an incentive device. The main bank’s decision on whether it should rescue its client firm at the brink of bankruptcy hinges upon the firm’s future prospect and the “main bank rent,” which is influenced by various government financial regulations. Despite its successful incorporation of the salient characteristic of the Japanese firm (a team production) and the main bank’s monitoring through its efficient wage contract scheme, the model underestimates the role played by cross shareholdings between the bank and its client firm. Also, limiting the analysis on the bank-firm relation, the cross-shareholdings among non-financial firms are grossly overlooked. 3. The implicit contract theory: Paying attention to the industry-wide economic transactions by the industrial group or keiretsu, the theory identifies the main bank as an “insurer” for the affiliate firms of the same group. The ties among the affiliate firms are consolidated by cross-shareholdings, which, in turn, entail the demise of the effective capital market. The main bank assumes the financing role in lieu of capital market. Nakatani’s empirical research suggests that the industrial groups are more concerned with compensation for their employees and preserving the lifetime employment system rather than the group firms’ profit- or sales-maximization. Despite 33 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. its plausibility, the theory is incapable of showing the enforcement mechanism by which the main bank functions as the insurer to other affiliate group firms. With all these advantages and disadvantages alike, the next chapter develops another model of the Japanese main bank system that is not well-captured in those three theories above and examines its relevancy. 34 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Endnotes for Chapter 1 1. Unfortunately, Miyashita and Russell (1994: p. 49) did not mention the year o f the data from which they derived these statistical results. In referring to smaller firms, they argued that the figures o f the main bank lending would be even higher, and that their main banks would not be city banks, but much smaller regional banks (p. 49). Statistically examining the impact of the size o f firms and main banks on their relationships, Horiuchi (1994) derived the conclusion that the relationship between small firms and small main banks is much more fragile than that between large firms and large main banks. In the analysis concerned, I am going to delimit the main bank relationships between large firms and large banks. 2. Horiuchi and Sui (1992; pp. 2-5) also admitted the importance o f information management capacity of bank, although they did not list it as one o f the stylized facts. 3. Teranishi (1993, 1994) explained that the loan syndication in Japan started in 1937 when the Sino- Japanese war broke out, and the Japanese government began to strengthen its oversight over the banking industry to encourage financing for the munitions industry. Under such an economic circumstance, “the banks were naturally reluctant to make single large loans” (Teranishi 1993: p. 76), and, there occurred “the spontaneous formation o f large numbers o f loan consortia during this period” (p. 75). Teranishi’s interesting finding on this issue is that the Industrial Bank o f Japan had often helped form the loan syndications during the war. He also clarified how the loan consortiums built up during the wartime was succeeded into the post-war main bank system (p. 75). 4. The formation o f a loan consortium might be also part o f such ambiguity, because how financial burden should be allocated among the participants when the client firm is in financial trouble is not usually made into an explicit contract when they form the loan consortium. Therefore, there seems to have been an implicit contract among the participants that the main bank should incur the disproportionate financial burden among the participants when the client firm is in financial distress (Sheard 1992: p. 13). 5. Although skeptical o f the close relationship between banks and non-financial firms, as represented by the main bank relationship, Miwa (1990: pp. 144-6; 1996: pp. 101-4) also found that 66.8 percent of 743 firms picked up from the first section o f the Tokyo Stock Exchange had maintained the main bank relationship during 1973-83. 6. Their definition o f the main bank was adopted from Economic Research Association (Keizai Chosa- kai, various issues), which regards a company as belonging to the “financial group” (kin ’ yu keiretsu) o f its lending bank, when it was given the largest loan from the bank for three or more years consecutively. See Horiuchi, Packer and Fukuda (1988: p. 162). 7. The only exception to this is underwriting debentures issued by three long-credit banks. 8. Note that the issuance o f corporate bonds was severely limited to a small number o f large companies such as electric power companies. The flip side o f the constant decline in the share o f banks’ bond holdings since the early 1970s was their increase by insurance companies and the Trust Department of the MOF. See Aoki, Patrick, and Sheard (1994: p. 12). 35 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 9. The new Banking Act o f 1927 implemented the MOF with the bank licensing system (Clause 2), the minimum capital requirement on banks (Clause 3-41), and the discretionary supervising authority over the financial institutions (Clause 20-3). Taking advantage o f these new clauses, the MOF could successfully squelch the financial panic in the 1930s. See Ueda (1999: p 39) and Teranishi (1993: pp. 69-71) for details. 10. See also Ramsyer (1994: pp. 237-9) for how the Japanese bond market evolved after the 1930s, and what function it has played in relation with non-financial corporations and the MOF. 11. Rosenbluth (1989: p. 140) proceeded: At the core o f the Bond Committee were eight banks, with the Industrial Bank of Japan central among them. Security firms were given a cut o f the market as underwriting members o f the Bond Committee, but only banks could earn the collateral fee. When Yamaichi Securities attempted in the late 1920s to take over a trustee bank to capture some o f the collateral-management business, the MOF blocked the move. One o f the enduring effects o f the Bond Committee’s rule was the shift it engendered in Japanese corporate finance to reliance on bank loans. For the comprehensive survey o f the history o f the Bond Committee and its function, see Nihon Koshasai Kenkyujo (1995: pp. 148-161, pp. 358-69). 12. One big difference between the prewar and postwar Bond Committees is that the latter incorporated four large securities firms (Nomura Securities, Daiwa Securities, Nikko Securities, and Yamaichi Securities). Such incorporation was done smoothly because banks let the securities industry tap into their monopoly rent in the form o f the underwriting fees while strategically pricing their collateral management service fees (Ramsayer 1994: pp. 238-9; Rosenbluth 1989: p. 145-7). The underwriting fee structure determined in this period had lasted until the Big Bang financial reform. 13. In 1946, GHQ banned financial institutions from issuing bonds on the ground that banks would crowd out other corporate bonds in the market. The IBJ overturned this measure so that it would be able to issue the bank debentures (kin’ yii-sai) to other city banks (Rosenbluth 1989: p. 144). During this period, banks absorbed over half o f all corporate bonds and financial debentures, and agreed not to resell them on the secondary market. In exchange for this, they started to claim that they had the right to determine the amount o f the monthly subscription o f those bonds (Rosenbluth 1989: p. 145). 14. Rosenbluth (1989: p. 163) explained that the Three Bureaus Agreement (sankyoku shido) was abolished by the MOF in March 1988 on account o f banks’ assent on further relaxation on domestic bond issuance rules. It is probably true that the role o f the Agreement was virtually over at the time, but the Agreement had actually remained even in the nominal form by 1998 when the Big Bang enactment o f mutual entry between banking and securities sectors was administered in Japan. A survey data by Ramsayer (1994: p. 238) showed that 85.7 percent o f Japanese firms had paid commissions to Japanese banks for their service o f selling bonds abroad. Their commission fee totaled 3.5 million yen. 15. According to Nihon Koshasai Kenkyujo (1995: p. 158), in 1985, the value o f the Eurobond issuance exceeded that o f domestic issuance. 36 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 16. In November 1990, the rating system o f the Bond Committee was finally abolished, and the issuance o f straight and convertible bonds, both secured and unsecured, was completely liberalized. Private rating companies were to take a full grip on the rating job thereafter. These agencies are “Nihon Koshasai Kenkyujo (JBRI, founded as a subsidiary o f Nikkei Shimbunsha in April 1985),” “Nihon Kakuzuke Kenkyujo (JCR, founded by Bank o f Tokyo, Japan Long-Term Credit Bank, and other institutional investors in April 1985),” and “Nihon Investors Service (JIS, founded by the IBJ, some city and regional banks and some securities firms in March 1985).” The Bond Committee was officially abolished in 1996. For the detailed history o f the rating system in Japan, see Kurosawa (1999), Nihon Kakuzuke Toshi Joho Senta (1998) and Nihon Koshasai Kenkyujo (1995). 17. Vogel (1996: pp. 181-195) reported in detail the battle between the banking and securities sectors and the role played by the MOF from 1985-92. 18. Actually, cross-shareholdings can be witnessed in the prewar period. For details, see Nakajima (1991: pp. 35-7), Shimotani (1996: p. 84, 91), and Teranishi (1993: p. 71, 77). Important to note, however, is that cross-shareholding o f the postwar period is inseparably related with the indirect financing by the main banks. This issue will be discussed in depth in the next chapter. 19. On this matter, Nakajima (1991: p. 64) noted this: “There was no financial firm that was forced to dissolve. This was to soon exert a far-reaching impact on the creation o f “keiretsu” or “corporate groups” centering around the banks. 20. During this period, there occasionally occurred hostile takeover trials. Some examples are Fuji Electric, Mitsui Mining, Nippon Electric (NEC), Nakayama Steel, Taisho Kijo Insurance, Yowa Real Estate, and Kanto Real Estate. For details o f these incidents, see Aoki (1995a: p. 102), Kikkawa (1992: pp. 261-3), Miyajima (1992: p. 242), Miyashita and Russell (1994: p. 67) and Teranishi (1993: p. 85). 21. Two other important developments that contributed to reviving the former zaibatsu into the postwar keiretsu groups during the period were (a) the lifting ban on using old zaibatsu names in 1952, and (b) the abolition o f the Law Prohibiting Excess Concentration o f Economic Power (kado keizai shuchu shuchti haijoho haijoho) in 1955, which was actually done by the Japanese government. 22. Yamauchi Securities was finally salvaged by the operation o f Section 25 o f the Bank o f Japan Law. 23. For example, Toyota Motor Co., despite its guise o f belonging to the Mitsui group, has long maintained the main bank relationship, not only with Sakura Bank, but also with Tokyo Bank and Sanwa Bank. Also, Nakajima (1991: pp. 57-8) pointed out that the large shareholders o f Hitachi Factory (Hitachi Seisakujo), as o f March 1975, were the IBJ, Sanwa Bank, Fuji Bank and Dai-Ichi Kangyo Bank, all o f which were competing with each other for the main bank position with Hitachi Factory. 24. The rescue plan is usually administered by the main bank’s consultation with other banks that joined in the same loan syndicate. In almost all cases o f the rescue plan, the main bank is supposed to assume a disproportionate share o f the burden compared with other banks in the syndicate. See, for instance, Sheard (1994c: p. 195). 25. Two sets o f case studies for those belonging to C firms were offered by Sheard (1992b: pp. 30-60) and Sheard (1994c: pp. 213-226). Aoki (1994a: p. 671) stated that the main banks barely make benevolent rescue operations and that to do so could have invited serious morally hazard hazardous behavior on the part o f their client firms. 37 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 26. The aspect o f the formation o f a loan syndication involved in the main bank system, albeit important, will be dismissed in this section, as it was not listed as a part o f the stylized fact in the previous section. Interested readers are advised to see Sheard (1994a 1994b, 1994d) and Teranishi (1994). 27. According to Jensen and Meckling (1976: p. 327), the cost associated with the “separation of ownership and control” o f a firm was first pointed out by Adam Smith’s The Wealth o f Nations (1776), and became widely known by Adolf Berle and Gardiner Means’s The Modem Corporation and Private Property (1932). 28. Also note that, acting as the manger o f a loan consortium, the main bank also contributes to lowering the bankruptcy cost. It often takes the form o f the main bank’s disproportionately assuming the bad loan compared with other lenders in the same consortium. 29. Nakatani’s (1984: p. 233) classification was based on the three factors o f (a) the strength o f a firm’s relationship to the financial institutions in the group, (b) the propensity to borrow from group banks and insurance companies, and (c) the percentages o f shares held by other group firms. 30. For the concrete regression results, see Hoshi, Kashyap, and Scharfstein (1990: pp. 81-5). Somehow, the coefficient o f SHARE was negative and not statistically significant. They carried out two more interesting regressions (Model 2 and Model 3): Model 2 is to incorporate the interaction effects, “GROUP*TOPLEND,” “GROUP*SHARE,” and “TOPLEND*SHARE,” and all these three variables turn out to be negative. For the first two terms, the results make sense because the “GROUP” term is considered to have already taken some o f their effects. The negative sign o f the “TOPLEND*SHARE” roughly means that lending and equity shares are a substitute, and not complementary. This seems to contradict Fukao’s (1999: p. 166) argument that “by lending and holding equities at a time, Japanese banks could enjoy their client firms’ success through the appreciated o f their equities.” Model 3 is to consider sales growth instead o f investment rate. The result o f this model is essentially the same as Model 2. 31. Both the dependent and independent variables are normalized by capital stock in order to eliminate the scale effects. The two dummy variables were incorporated in order to remove the firm-specific effect and the macroeconomic shock. The precise definitions o f those variables are in Hoshi, Kashyap, and Scharfstein (1991: pp. 44-5). 32. Jensen (1986) argued that, under the restriction imposed by the Glass-Steagall Act, it is advisable that U.S. firms issue more debt in exchange for stock since debt creation has a bonding effect on the managers to pay out future cash flow to the bondholders, thus reducing the agency cost o f free cash flow that otherwise could be spent as perquisites for managers. 33. The condition that best characterizes the team, vis-a-vis the principal-agent relation, is that the principal and its agent share the same objective. See Aoki (1989b: p. 14). 34. Aoki’s original model (Aoki 1994a) is very complicated. For its truncated version, see Aoki and Okuno (1996: p. 210). Note that the optimal solution o f the wage contracts is the “second best” in that, as long as the monitor cannot oversee each worker’s performance, there is a possibility that the production capacity is not fully utilized (p. 209). 35. Aoki recommended this state-contingent governance scheme to the transitional economies, where the economic system is transforming from the socialist to market-based economies, on the ground that the state-contingent governance best serves to break down the entrenchment o f workers and managers that took place en masse in the democratization process. 38 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 36. In fact, Aoki (1994a) derived that, by increasing the cost o f such punishment in the model, the second- best solution approaches the first best. 37. Excessive main bank rent would give the main banks the incentive to rescue firms in the bad or critical state quite often, and that would ruin the benefit o f the contingent governance system. Thus, in order for the system to properly work, the main bank rent should not be too low or too high (Aoki and Okuno 1996: p. 215). 38. Kester (1991a: p. 19) cited two clauses widely used in Japanese contracts. One often used as the final clause reads, “Concerning matters not stipulated herein or any doubt about the stipulation both parties shall settle them upon deliberation.” A similar one reads, “Should a disagreement arise, the parties will settle it amicably by consultation.” While suspecting that the Japanese contract typically represented by these two clauses undermines the very purpose o f entering into a contract in the first place, Kester assesses that such a contract allows the business entities to adapt very quickly to fluid market conditions (Kester 1991a: pp. 62-7; 1992: p. 28). For the Japanese contract adopted in the automobile and electric appliance industries, see Asanuma 1989: pp. 3-9). 39. The “so-called” implicit contract theory was originally presented to explain the wage rigidity arising from the asymmetric information regarding the quality o f workers in the labor market: according to the theory, the wage rigidity is the result o f the manager’s bidding higher-than-competitive wages for inducing workers o f good quality into his firm from among all the available workers whose qualities are unknown to him. The idea was then inherited by Fried and Howit (1980), who demonstrated that the interest rate and credit rationing occur in the financial market because o f the (risk-neutral) bank’s transferring the interest-variation risks from the (risk-avert) customers o f good quality. In any case, it is important to note that, despite his occasional reference to Fried and Howit (1980), Nakatani’s (1984) work is disparate from their work. Horiuchi and Sui (1992) noticed this when they said, “The interest rate variations are neither only one risk nor the most significant risk to both banks and firms” (Horiuchi and Sui 1992: p. 8). They pointed out that the insurance mechanism explained by the implicit contract theory and that which is used to explain the main bank’s rescue activity are essentially different (p. 17). The “implicit contract” in the sense of Fried and Howit (1980) was later succeeded by Horiuchi (1988) and Hirota (1991), who conducted empirical research to find out whether there was a correlation between the operating profit and the financial cost in the firms maintaining the main bank relationship. 40. This suggests that Nakatani’s main results— items (1) to (3) above— can be reinterpreted as institutional complementarity among the financial and labor markets in each o f the industrial groups. 41. Ramsayer (1994: pp. 244-5) pointed out that the main bank scholars envision the bank-firm relationship, typically in the following fashion: The Mitsubishi Bank, for example, may explicitly agree to lend Iroha Sushi 30 per cent o f the bank loans Iroha needs. At the times it does so, it may also implicitly agree to bear 80 per cent o f the costs o f monitoring Iroha and to absorb 80 per cent o f any losses should Iroha fail. Ramsayer argued that this kind o f implicit contract is hard to imagine because it can be easily made explicit and much simpler. The straightforward way to reach essentially the same result is just to have Mitsubishi lend 80 per cent o f the money to Iroha Sushi. By making such an explicit, simpler contract, Mitsubishi will have a greater incentive to monitor Iroha, and be willing to absorb 80 percent o f any resulting loss. Because the total loan outstanding will not change, the explicit scheme will not raise the bank’s capital requirements. Besides, because Mitsubishi bears 80 percent o f Iroha’s default risk under either the explicit or implicit scheme, neither does it reduce Mitsubishi’ s diversification: in short, there is no reason for Mitsubishi to issue insurance to Iroha. 39 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Chapter 2 T h e J a p a n e s e K e ir e t s u M o d e l 2.1 Introduction Three theories of the Japanese main bank were reviewed in the previous chapter and found to have both advantages and disadvantages: while respectively paying attention to two dimensions in which the main bank deemed necessary to play a significant economic role, the agency cost theory and the contingent governance of teams have the common ground of being micro-theoretic and explicitly incorporating the bank’s decision-making in regard to financing and equities holdings. Due to the very nature of their micro-economic settings, however, these theories suffer from their inability to derive macroeconomic implications. The implicit contract theory, on the other hand, is rather macro-theoretic, taking into consideration a set of firms arranged across industries in their formation of an enterprise group called keiretsu. The theory, however, suffers from another problem of failing to articulate the enforcement mechanism of the mutual shareholdings among firms that leads to the keiretsu formation and the subsequent improvement of their employees’ compensation. In light of these problems, the most promising way to make a theoretical breakthrough in research on the Japanese main bank is to reconstruct a model based on the Japanese keiretsu, which incorporates a main bank as its constituent, yet, unlike Nakatani and Sheard, in such a way as to take into consideration some aspects of the main bank’s decision-making and, when necessary, other affiliate firms. Indeed, many representative Japanese main banks have traditionally been the vital constituents of their own keiretsu groups and often involved in joint projects with their affiliates. In order to grasp such an aspect of division of labor among the firms inside a keiretsu group, it is beneficial to take advantage of the fruits of the transaction cost economics (TCE) developed in a 40 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. series of works by Williamson (1985, 1987a, 1987b, 1988). Section 2.2 reviews the core idea of TCE and then considers how it can be reconfigured for model building of the Japanese keiretsu of which the main bank is a part. In order to confirm the relevancy of the model thus built, the sections overview how the zaibatsu and keiretsu groups were formed, and conduct a comparative statistical analysis over the six keiretsu that emerged in postwar Japan. The key variables for the analysis are those pertaining to cross-shareholdings and affiliate firms’ borrowing from their main banks. 2.2 The Japanese Keiretsu Model 2.2.1 Transaction Cost Economics: Asset Specificity and Market (Contractual Governance) versus Hierarchy (Organization) A model of the Japanese keiretsu needs to incorporate the concept of “asset specificity” or “relation specific investment,” developed in “transaction cost economics (TCE),” which refers to the highly idiosyncratic nature of investment made among the economic agents under transaction. The most common features of the modem economy are its division of labor in production and the exchange of products systems. Production is mostly taken up by economic agents, who have developed and embodied the skill-specific knowledge of how to produce their products cost- effectively and lucratively. When the products thus produced are of intermediate kinds, they need to be further exchanged with other production agents, instead of being delivered to consumers. This is exactly where the modem marketplace places strain on both sides of the transaction agents in regard to whether their transaction will go through as they originally planned: the agents, at this juncture, are both exposed to the hazard of the self-interested, opportunistic behavior on the part of the other agent, who may take advantage of the other’s having made a “nontrivial, relation specific” investment.1 For instance, when the selling agent is already committed to an asset-specific investment, the purchasing agent may threaten him by suggesting the possibility of trade termination, and start to haggle in the negotiation. In the case, on the other hand, when the 41 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. purchasing agent is committed to an asset-specific investment, the selling agent may threaten the purchasing agent by suggesting termination of trade and demand more reward. Consequently, both buyers and sellers with asset-specific investment are urged to reconsider partaking in transactions, which may lead to a sheer exchange failure. To circumvent such opportunism in the marketplace, two instruments have been designed and developed: “the pure market contract” and “the pure organization” or “vertical integration.” The pure contract intends to prevent opportunism by agents’ concluding an ex-ante market contract that specifies all the terms of trade contingent on all future outcomes conceivable at the time of the conclusion. Under this pure market contract scheme, the damage caused by opportunism is recouped through judicial enforcement. The pure organization, on the other hand, makes a sharp contrast with the pure market contract in its complete absence of any term of trade between agents. This is possible through one trading agent buying up the assets of the other and thereby replacing a centralized, hierarchical organization for the market transaction. The pure market contract and the pure organization are, according to TCE, the two extreme institutional alternatives used for facilitating transactions between different agents (Williamson 1985: pp. 41-2). In determining whether to take the pure market contract or the pure organization, the factor that matters the most is the degree of asset specificity of goods under transaction. In order to approach this issue, one must consider two types of goods: “general purpose” and “special puipose”. Generally, the former is best characterized by the nature of standardization and the latter, to the contrary, by the nature of non-triviality and idiosyncrasy. Due to their nature, the general purpose goods are normally available in the open market and pose no such risk as maladjustment between the transaction agents. Even if there occurs an accident in the course of transaction, new trade relations and contracts are usually rearranged in a less costly manner. The special purpose goods, on the other hand, are not often available in the open market and are suitable only for those agents who share the interest in trading such types of goods. While the agents can benefit from the transaction of this type of goods, as they cannot meet their demand in the open market, they are more 42 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. vulnerable to the serious problems posed by such a transaction in the forms of opportunism and ex- post maladjustment as previously explained. From the analogy above, the proposition derives that the more asset-specific goods are, the more transaction becomes susceptible to the risk of the ex post maladjustment. Accordingly, as the degree of asset specificity develops, the introduction of vertical integration (organization) will become more imperative. When the degree enters some threshold along the continuum spanned over the two ends of the general purpose and special purpose goods, the pure market contract is eventually supplanted by the outright ownership of vertical organization (the “Fundamental Transformation”) (Williamson 1985: p. 54, 61, 88). 2.2.2 The Fisher Body-General Motors Relation: An Illustration Gilson and Roe (1992: pp. 23-6) and Klein (1988) demonstrated a case study from U.S. economic history, which is considered a good example of how the “Fundamental Transformation” occurs in the real economy. The story is about the trade relationship between General Motors (GM) and Fisher Body, both of which entered into a long-term exclusive contract in 1919: GM decided to introduce a closed metal automobile body, which, at that time, was mostly wooden, open, and individually made, and began to look for a firm with the capacity to build it. Fisher Body had such capacity and was looking for a new customer. Seemingly, their interests perfectly matched, but while they sat at the negotiation table, they realized a serious problem looming ahead. To build the metal door GM needed, Fisher had to make a highly specific investment in stamping machines and ties, which, to Fisher Body, appeared as a significant source of leverage; the technological specificity, once adopted, would allow GM to appropriate Fisher’s quasi-rent by threatening to reduce GM’s demand for Fisher-produced bodies, or to terminate GM’s purchase when GM could not meet its request for price adjustment with Fisher Body. Fisher Body was, thus, unwilling to commit to the relation- specific investments without assured purchases from GM. 43 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. They overcame this problem by devising an extensive dealing clause in the negotiation, by which GM agreed to purchase all their closed-metal bodies from Fisher over a ten-year period. In its turn, however, this clause created the opportunity for Fisher Body to threaten GM, since Fisher could now exploit GM by increasing the price or decreasing the quality, taking advantage of the clause that required GM to purchase exclusively from Fisher Body. For GM’s protection, Fisher agreed to a formula for setting the price equal to Fisher’s “variable costs” (the wages and transportation costs) plus 17.6 percent of the costs representing the profit or of the value of the specified assets Fisher had employed. With the inclusion of the procurement clause and price formula, both GM and Fisher finally entered into a long-term contract. After the conclusion of the contract, however, GM recognized that the demand for closed- metal bodies grew rapidly. Compared with the traditional wooden, open bodies dominant in the market at that time, the closed metal bodies were essentially a novelty. By 1924, the sales of closed- metal bodies accounted for more than 65 percent of GM’s automobile production. This shift in demand began to work very favorably to Fisher under the original 1919 contract, which failed to foresee such an extreme run-up in demand. Despite GM’s requests, Fisher adopted a relatively inefficient, highly labor-intensive technology and refused to locate the body-producing plants adjacent to GM’s assembly plants, thereby effectively holding up GM.3 By so doing, Fisher could take advantage by increasing its profits because 17.6 percent of profit upcharge specified in the original contract was more pronounced under Fisher’s labor-intensive technology. Obviously, all this was detrimental to GM which could have saved money if Fisher were closer to its own plants and by Fisher’s adoption of more cost-saving, capital-intensive technology. Eventually, in 1926 GM acquired Fisher Body stock at terms highly favorable for Fisher, and integrated the company into GM’s subsidiaries. William Fisher, President of Fisher Body, became general manager of “GM’s” new Fisher Body division. Through this vertical integration of Fisher into GM, the original contract of 1919 was brought to naught, and Fisher’s asset-specific technology was carried over in the hands of GM for the continuation of producing closed-metal 44 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. automobile bodies. The technology that GM acquired from Fisher was actually not limited to stamping machines and dies, but included the human capital of how to produce closed-metal bodies: if the hold-up problem were only a matter of specific physical investments made by Fisher, GM could have solved the problem by stamping machines and dies and drawn another contractual agreement with Fisher. Through this vertical integration, GM converted all the Fisher employees into GM ones and moved from “buying” automobile bodies to “producing” them. 2.2.3 The Japanese Keiretsu Model: The Keiretsu as a “ Meta-Firm” with the Main Bank Being a Subset Asset-specific investment was very prevalent in the postwar Japanese economy. It, however, rarely took the “Fundamental Transformation” form (complete integration) as argued by Williamson, and was rather characterized by partial integration through partial ownership among a set of affiliated firms forming the industrial group or “keiretsu.”4 As witnessed in the historical account of Section 1.2.3, the stockholdings between the main bank and its client firms were often part of the whole group formation process that also involved cross-shareholdings among non- financial firms. In fact, the former zaibatsu firms and banks (Mitsui, Mitsubishi, and Sumitomo) and those belonging to newly-established bank groups (Dai-Ichi Kangyo, Fuyo, Sanwa) both utilized cross-shareholdings on various occasions in the postwar period, thus consolidating the industrial groups known as the “horizontal-” or “intermarket” keiretsu. Unlike zaibatsu firms, those belonging to a keiretsu group are not under the command of a holding company. Rather, they are a set of firms and (main) bank each of which aims to solidify the intra-group ties through cross- or mutual- shareholdings, while, at the same time, securing their corporate independence. Their main reason to do so is to expedite their “asset-specific” or “relation-specific” investment among their group firms widely dispersed across industries;5 while the stocks held by each constituent firm may be small in number, and, thus, the outright ownership of one firm over the other is eschewed inside the group, the aggregate stocks inside the group often become substantive enough to secure the governance 45 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. over the constituents, so that any malfunctioning in the intra-group transactions can be relatively easily corrected. To clarify this point, let us suppose a case where one member firm impinges on its original contract with other affiliates to engage in the intra-group transaction, and to opportunistically take advantage of others’ asset-specific investments.6 If the market contract is the only binding means for these firms, and the contract fails to specify how such an opportunistic action is handled, the outcome of such a market contract will become disastrous; their transaction will not only come to halt, but turn inordinately costly because of the very nature of their idiosyncratic investments. If cross-shareholdings are made among those firms at the outset, however, the situation takes a totally different course; in such a case, there is still a remedy for other members to rally around and exercise their collected residual right against the opportunistic firm and to force out the evil executive. This logic holds for other cases, too, such as the one involving the executive showing bad, if not opportunistic, performance in the intra-group transaction.7 Exchanging stocks in the group is thus understood as the submission of pledges or hostages among the group members, or, equivalently, the confirmation of their intent for the credible commitment to accomplish the intra-group transaction, rather than obtaining the means to earn dividends. The transaction relationship, thus, established, on the other hand, is best characterized by the “mixed modes of firms and market organization” (Williamson 1985: pp. 41-2). As Kester noted, such a relationship is “an attempt to secure the best of two worlds,” since [b]y tying themselves to one another in groups, yet eschewing outright ownership and control, Japanese corporations have been able to exploit some of the high-powered incentives of the market that derive from independent ownership of assets, while relying on selective intervention by key equity owners to adapt contracts to new circumstances as needed. (Kester 1991a: p. 80; 1991b: p. 41) The traditional framework of the main bank monitoring can now be reconfigured into the keiretsu production system discussed above. That is, the primary task of the main bank is reinterpreted as providing group firms with money as part of the investment goods, just as one 46 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. subcontractor delivers its products to its group affiliate firm. For this reason, the main bank administers intensive monitoring on the member firms and sees whether they are closely following their original investment and production projects. This is the place where Aoki’s “contingent governance of teams” comes into play in the context of the keiretsu production scheme. The main bank’s holding of their members’ settlement accounts (toza yokin) (see Section 1.2.4) makes its monitoring much more efficient and effective, as it can see every financial transaction made by the members. The main bank’s holding of its members’ equities furthermore contributes to disciplining the firms because of its possible future intervention.8 When one member firm is about to falter, the main bank intervenes and starts reconstruction procedures for the firm, such as replacing the problem executives and managers and providing a fresh money supply to continue the firm’s operation. The agency cost theory of the Berle-Means tradition tries to explain this rescue action mainly from the residual loss that the bank incurs. Indeed, the residual loss is something that the main bank wishes to avoid in any event, ft should be noted, however, that the effect of such residual loss will be much more pronounced when the main bank holds the multiple equities issued by the firms which are closely connected through the relational investment activity; if such an incident occurs, the failure of one member firm will spillover to other members through a chain effect of production failures. The main bank’s credible commitment to the group production activity shakes off such a vicious chain effect: through timely intervention, the main bank can recover not only the faltering firm, but also other related firms engaging in part production with that firm.9 In sum, in the framework of relational investment, the main bank’s primary objective lies in smoothing out a chain of the intra-group production across the member firms, rather than maximizing the residual gains from the stocks it owns: the bank is always interested in maximizing residual gains, but such an interest is often subjugated by its primal goal of accomplishing the group production as a whole. Otherwise, its residual gains may become even worse.1 0 47 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 2.3 A Brief Overview of the Zaibatsu History: The Early Development of “Complementarity” among Firms The concept recurrently discussed in development economics is the “coordination” or “complementarity” among various industrial sectors, which ensures the simultaneous growth among industrial sectors or, by which, according to G. B. Richardson, “their combined profitability when undertaken simultaneously, exceeds the sum of the profits to be obtained from each of them, if undertaken by itself (Gerlach 1992: p. 206, citing Richardson, Information and Investment,I960: p. 72). In such a case, “investments may be complementary either because the costs of one are reduced when the other is undertaken, or because the demand for the output of the other” (Gerlach 1992). One good exposition of the industrial coordination is the relationship between the steel and shipbuilding industries; the growth of the shipbuilding industry, on the one hand, promotes that of the steel industry through the former’s consumption of steel supplied by the latter as the primary raw material. Through the economics of scale, on the other hand, the growth of the steel industry contributes to the shipbuilding industry by reducing the steel price. Once such mutual benefits are realized, the coordination is viewed as existing between these two sectors and keeps benefiting them in the established circle.1 1 No doubt, the importance of the inter-industrial coordination is most pronounced in economically backward countries, where “a venture in any single big enterprise may be attractive only if the simultaneous development of other industries is undertaken” (Gerlach 1992: p. 206, citing William W. Lockwood, The Economic Development in Japan: Growth and Structural Change, 1968: p. 227). In Japan, where modernization started with a relatively primitive base in the late nineteenth century, the zaibatsu groups, notably Mitsui, Mitsubishi, and Sumitomo, played significant roles in diversifying their investments in various industrial areas in an asset-specific, coordinating fashion. In fact, perusal of the zaibatsu histories discloses that many zaibatsu firms maintained close, asset-specific relations with other affiliates, due primarily to their originally 12 sprouting up from the well-established parent firms in the groups. First of all, in the early days of 48 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. the Mitsubishi zaibatsu, Nippon Yusen Shipment Service and Takashima Coal-Mining (Takashima Taonko) were sprung off from Mitsubishi Shokai, the origin of the Mitsubishi zaibatsu. At this very beginning, there was a direct, asset specific relation between these two subsidiaries. Coal produced by Takashima Coal-Mining was shipped by Nippon Yusen, and Nippon Yusen consumed the coal as its fuel. In order to meet the demands related with shipment service, Mitsubishi Shokai set up non life insurance and warehouses during the 1870s through the 1890s. Shipping service, on the other hand, was soon to be added with the advanced shipbuilding technology. In 1877, the government- owned Nagasaki Shipbuilding Factory was sold to Mitsubishi Shokai under the decision by Hirofumi Ito, then Industry Minister, who recognized the importance of “complementarity” between shipment service and shipbuilding. The electric rigging technology was, then, developed in Nagasaki and Kobe Shipbuilding Factories. In 1921, the rigging plants in these two shipbuilding factories merged and became independent as Mitsubishi Electric {Mitsubishi Denki). Takashima Coal-Mining, on the other hand, was a successively set up the sales division since 1881, and those sales divisions started to deal with outside suppliers. In 1918, Takashima Coal-Mining made a decision to lump them together and gave independence to the newly established firm under the name _ 13 of Mitsubishi Corp. {Mitsubishi Shoji). Other zaibatsus, notably, the Sumitomo and Mitsui zaibatsu, also developed through the inter-industrial coordination. In the case of the Sumitomo zaibatsu, many firms sprang up from Besshi Copper Mining Co., Sumitomo’s patriarchal company that started business in the late seventeenth century. First, warehouse and financial businesses were hived off from Besshi in the early Meiji period: the warehouse inside Besshi became indispensable when Sumitomo closed the wholesaling copper business and shifted to the retailing. Also, part of the warehouse was allotted for the crops submitted as collateral, and financial business expanded in the same period. The warehouse and the financial business grew respectively into Sumitomo Warehousing (1893) and Sumitomo Bank (1895). Besshi Mining Co. went on to bring forth other metal-related firms. It set up the machine-production department to save the cost of imported technology. The department 49 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. came to produce other types of machines for other Sumitomo firms and gained independence as Niihama Factory (Niihama Seisaku-jo, 1928), which was, in 1934, renamed Sumitomo Machine Co. (Sumitomo Kikai Kogyo). Besshi also bought out Nihon Copper Production Co. (Nihon Seido) as its processing department in 1897. After acquiring other advanced techniques as copper-rolling and steel-tubing, the department won independence as Sumitomo Copper-Rolling & Steel-Tubing in 1926. In trying to enter the steel industry, Sumitomo also purchased Nihon Steel Production (Nihon Chukd-jo) and renamed it Sumitomo Steel Production (Sumitomo Chuko-jo) in 1910. Sumitomo’s entry into the chemistry industry, on the other hand, was due primarily to an ironical reason of resolving the pollution problem exacerbated by Besshi Copper Mining. Implemented with the technology that could transform sulfuric acid gas discharged from Besshi into fertilizer, Sumitomo Fertilizer (Sumitomo Hiryo Seizd-jo) was established in 1913.1 4 Quite contrary to the development paths of the Mitsubishi and Sumitomo zaibatsu, where the production sector took precedence over the financial and distribution sectors, the Mitsui zaibatsu initiated its modernization of the financial sector, which, at the time, was Mitsui’s leading sector. First, the distribution department (kokusan- gata) of the Ryogae-sho financial sector was spun off as Mitsui Corp. (Mitsui Bussan) in 1876. In the meantime, Mitsui, like Mitsubishi and Sumitomo, tried to strengthen the production sector, but that was often achieved by takeovers of outside firms that Mitsui Bank was financing, not by the interior branching out. The first example of this was Kamioka Mining of Nakanishi-gumi, which was taken over by Mitsui Bank in 1886 as a result of the Nakagami bankruptcy. Together with Miike Mining, which was originally sold off by the government to Mitsui in 1888, Mitsui’s headquarters unified the two mining companies into Mitsui Mining Co. (Mitsui Kozan, 1892). Mitsui Corp., then, started to take advantage of shipping abroad coal exploited by Mitsui Mining. Apart from this close relation between Mitsui Mining Co. and Mitsui Corp., however, it is hardly possible to find interior, industry-wide coordination in Mitsui’s diversification activities, most of which were rather characterized by a series of ad hoc takeovers. Shibaura Factory (Shibaura Seisaku-jo, 1893) and Onoda Cement (1894) were other examples of Mitsui’s acquisition of outside 50 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. firms through the bad-loan disposal. Mitsui’s entry into the steel industry was also through the takeover of Hokkaido Coal Mine and Shipping {Hokkaido Tanko Kiseri) in 1913, which had owned Nihon Steel {Nihon Tekkd-jo) as its subsidiary since 1907. As the diversification progressed, these three zaibatsus gradually recognized that their directorship and monitoring over their subsidiaries were limited, due primarily due to their undifferentiated ownership and directorship. Their solution to this issue was to set up holding companies {mochikabu gaisha) to separate ownership and directorship in their organization: by separating these two distinctive functions, they hoped that the division of labor would function in such an ideal manner that, while the owner assumes the monitoring role for the performance of the directors and managers, the directors and managers take the responsibility of efficiently running the firms with their managerial expertise. Shortly after the management crisis occurred inside Mitsui Bank, in 1891, the Mitsui families held the “Temporary Council” (Kari Hyogi-kai), consisting of the Mitsui clan and their directors, with the main agenda of consolidating management over the subsidiaries. To accomplish such a goal, the council made a decision to convert all four major Mitsui companies (Mitsui Bank, Mitsui Corp., Mitsui Mining Co., and Mitsui Kimono Department) into an “unlimited partnership” {gomei gaisha) in 1893. The Council also drew up a series of rules regarding personnel affairs and corporate disbursement in the form of the Family Constitution {Kaken), which was to be accomplished by 1900. With the background of all these efforts, Mitsui’s headquarters developed into Mitsui Gomei, Mitsui’s own holding company, in 1909. During the same period as Mitsui started to open its council, Mitsubishi Co. {Mitsubishi-sha) was also setting about its corporate reform. After making the regulatory arrangement of capital, personnel allocation, and self-financing system {dokuritsu saisan-sei) for the subsidiaries, Mitsubishi & Co., Ltd, itself metamorphosed into “Mitsubishi Limited” {Mitsubishi Goshi Gaisha) in 1893, drawing a line between Iwasaki families’ asset and Iwasaki’s ownership of the Mitsubishi firms.1 5 Eventually, “Mitsubishi Goshi” was set up as Mitsubishi’s own holding company in 1917. Having lagged behind Mitsui and Mitsubishi mainly due to the extremely centralized governance by Kichizaemon 51 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Sumitomo, who invested an enormous sum in all Sumitomo firms, the Sumitomo zaibatsu finally shifted to the holding company regime in 1921 through the reorganization of Kichizaemon’s private firm into “Sumitomo & Co. Ltd.” (Sumitomo Goshi).1 6 The period from the early 1920s to the early 1930s is best characterized by the mass sprouting of new, smaller zaibatsu, and the Big Three’s (Mitsui, Mitsubishi, Sumitomo) expansion of their territory into the heavy and chemical industries. In no time, World War I breaking out in Europe in 1917 brought about a huge demand for shipment service in Japan, which, then, spilled over into other sectors such as shipbuilding and steel.1 7 Taking advantage of this favorable economic climate, many new faces entered these sectors through diversification, and some of them set up their own holding companies. They were Furukawa & Co. (1917), Okura-gumi (1917), Fujita-gumi (1917), Asano (1918), Kawasaki (1920), Okawa & Co. (1920), and Kuhara (1920). Already shifted to unlimited partnership in 1912, Suzuki, like keiretsu of the post World War II period, promoted spinning off its subsidiaries during this period (Hashimoto 1992: p. 107; Miyashita 18 and Russell 1994: pp. 27-28). Once the War ended and the prolonged depression set in the Japanese economy in the early 1920s, however, the majority of these small zaibatsu found themselves having to pay dearly for their reckless expansion during the war boom.1 9 Standing in sharp contrast with those small zaibatsu, the Big three somehow muddled through the dismal 1920s by their prudent investment policies. Probably, the most conservative was Sumitomo, where, in the late 1910s, Masaya Suzuki, then general director of Sumitomo Co. & Ltd., turned down the proposal for setting up Sumitomo’s own trading company, insisting that Sumitomo’s business line be restricted to its traditional copper mining and related areas (Hashimoto 1992: pp. 94-5). Trying to stay away from speculative expansion, on the other hand, Mitsubishi Shipbuilding Co. (Mitsubishi Zosen, 1912) only accepted orders from regular customers such as the Japanese Navy and Nippon Yusen Shipment Service (Hashimoto 1992: pp. 97-8). In Mitsui, Mitsui Corp. had long ago 20 subjugated its vessel department, and did not grant the department expansive shipbuilding. In 52 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. response to the expansion of trade, Mitsui Corp. spun off Tokyo Raw Cotton {Toyo Menka) in April 1920, and tried to consolidate the risk management of the cotton dealings. While the representative areas of the Big Three, such as mining, banking, and commerce, were all matured, new economic conditions were prepared in the Japanese economy once the 1930s ushered in, and Japan’s comparative disadvantage of heavy and chemical industries was significantly mitigated (Hashimoto 1992: p. 130; Okazaki 1999: p. 65).2 1 Trying to prepare for the fuel substitution from coal to petroleum, Mitsubishi Corp. bought out the patent from Associated Co. of the U.S. in order to deal with refined petroleum in Japan. As the next step, Mitsubishi Corp. set up Mitsubishi Petroleum (Mitsubishi Sekiyu) in 1931 for the ultimate purpose of developing domestic refined technology. In Mitsubishi, Mitsubishi Heavy Industry {Mitsubishi Jukogyo) was founded in 1934 by the merger between Mitsubishi Airplane {Mitsubishi Kokiiki) and Mitsubishi Shipbuilding. In Sumitomo, Sumitomo Metal {Sumitomo Kinzoku) was established in 1935 by the merger between Sumitomo Steel and Sumitomo Copper-Rolling Steel Tubing. Also, in Sumitomo, Sumitomo Fertilizer was reorganized into Sumitomo Chemical {Sumitomo Kagaku) in 1934 after adding more advanced technology capable of ammonia synthesis. In Mitsui, Toyo Koatsu was founded in 1933 for ammonia synthesis with the Dupon method. When the Sino-Japan War broke out in July 1937, the heavy and chemical industrialization was further explored as part of the war effort. From 1936 to 1941, the paid-in capital of the heavy and chemical industries dramatically increased in the Big Three. In particular, conglomeration was promoted inside the Big Three’s mining companies during this period, through the investment in the various related fields, such as chemical, metal, and petroleum (Sawai 1992: p. 161-62).2 2 With the wartime economy settling in the Japanese economy, however, the inter-industrial development of the Big Three eventually came to an end. After the Pacific War broke out in December 1941, the Big Three’s heavy and chemical industrialization became more biased toward “mechanization” under the military’s seizure of resources. The Big Three also underwent other anomalies, such as (1) the restrictions on shareholders’ power in corporations and (2) the active 53 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. participation of the Big Three’s banks in corporate financing: throughout 1939-43, the government successively legislated laws regarding corporate reform with the aim of effectively running the wartime economy by restricting the profit motives of shareholders and giving more incentives to both managers and workers to actively participate in the production (Okazaki 1999: pp. 110-120; 23 Teranishi 1993: p. 74, 78, 79). This wartime corporate reform served as a centrifugal force to the Big Three’s governance over their subsidiaries, thereby bringing about their decentralization in more progressive degrees (Sawai 1992: pp. 189-195).2 4 To supplement capital market that was very much choked up through the various restrictions, the formation of the loan consortia started in 1939 among city banks, the IBJ, and Wartime Finance Bank (Senji Kin’ yu Kinko). In lieu of the capital market, the lending consortia took on a monitoring task, where the main lender often took advantage of the lower monitoring cost by lending to the firms belonging to the same group firms. Along with the support from the BOJ and the MOF, the consortium lending had, in fact, served as a substitute for a capital market and played a significant role for corporate financing in the rest of the wartime period (Okazaki 1999: pp. 115-117; Teranishi 1993: p. 78-79; Ueda 1999: p. 44-46).2 5 Obviously, this is the prototype of the main bank system in postwar Japan. Anomalous as it was in light of the whole zaibatsu history, the wartime economy prepared the basis for the postwar Japanese keiretsu through the institutionalization of both the new corporate governance whereby managers are detached from capital market, and the lending through the associate banks whereby the demise of capital market is countervailed. 2.4 The Postwar Development of Keiretsu Groups Where Main Banks Are Subsets 2.4.1 From Zaibatsu to Keiretsu: the Period o f “Keiretsu-ization ” The period which researchers such as Aoki considered as “the heyday of the Japanese main bank system” spans roughly over 1951-1975 (Aoki and Patrick 1994: p. 46), which substantially 54 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. corresponds to Japan’s high economic growth. This section witnesses that, in the earlier, trial-and- error period of 1945-55, the basic framework of keiretsu groupings was consolidated on both legal and functional bases. In the earlier Occupation regime, General Headquarters (GHQ) significantly impacted the Japanese economy, especially through (1) the dissolution of the zaibatsu groups, (2) the purge of corporate directors from public service (January 1947), and (3) the administration of the “Law Prohibiting Excess Concentrations of Economic Power” (LPECEP, Kado Keizai Shuchu Haijoho) (December 1947) to both Mitsui Corp. and Mitsubishi Corp. (July 1947). As a result of (l)-(3) above, the following three events respectively ensued: (1) the subsidiaries of the holding companies became independent; (2) virtually all managing directors employed during the wartime were forced out of their firms and replaced by new ones promoted from the factory level; and (3) Mitsui Corp. and Mitsubishi Corp. were divided into 223 and 139 petite companies, respectively, and brought into dysfunction.2 6 But the GHQ’s decision of September 1947 not to administer LPECEP to the 27 Japanese banks and the legislation of the Corporate Reconstruction Preparation Law (Kigyd Saiken Seibiho) of 19482 8 worked so favorably for the Japanese banks that they had less difficulty resuming their business: their lending balance noticeably increased after the suspension of the Reconstruction Finance Bank (RFB, Fukko K in’ yu Koko, 1946) due to the.Dodge Line of 1949 (see Table 6). Noteworthy of these city banks regaining strength is that their finance focused mainly on former customers (i.e., the former, fellow zaibatsu subsidiaries) in trying to collect the unpaid part of bad wartime bad loans to their new accounts (October 1948); city banks began to support reconstruction of their group firms through refinancing in the early 1950s (Miyajima 1992: pp. 226-29). 55 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 6. Financial Institutions’ Outstanding Lending (¥m) Month/ Year Wartime Finance Bank (1) Industrial Bank o f Japan (2) Reconstruction Finance Bank (3) Ordinary banks (4) [(l)+(2)+ (3)]/(4) (%) 3/1942 2,599 — 15,662 — 16.3 3/1943 9,766 306 19,034 — 52.9 3/1944 6,002 999 25,054 — 27.9 3/1945 12,106 2,903 40,354 — 37.2 9/1945 14,649 3,074 56,429 . . . 31.4 12/1947 — — 135,711 44,210 32.6 12/1948 — — 332,006 111,159 33.5 3/1949 17,970 . . . 131,965 357,096 42.0 12/1949 40,997 . . . 108,410 588,593 25.4 12/1950 69,158 — 89,895 845,510 18.8 12/1951 102,964 . . . 79,247 1,286,780 14.2 Source: Teranishi (1993), p. 84. In addition to the resurgence of the banking sector, noticeable progress occurred in cross shareholdings in the early 1950s, another essential factor in sustaining the keiretsu transaction. The vital events that contributed to this were the Peace Treaty of April 1952, which lifted the remaining 29 regulation on stockholdings, and the revision of the Antimonopoly Law in September 1953, which allowed banks to increase their stockholdings from 5% to 10% and non-fmancial firms to hold stocks with no restrictions. Not to miss this golden opportunity, city banks began to promote the mutual shareholdings with their group firms. There were two major reasons for the banks to do so at the time. First, by promoting mutual shareholdings, banks could avert the holdup actions on the part of their customer firms, most of which were still debt-ridden at the time (Miyajima 1992: pp. 245- 7). Second, banks, with the mutual shareholdings, could protect some of their customer firms suffering from the stagnant stock prices and the possibility of hostile takeovers (Miyajima 1992). Even more important, from the keiretsu perspective, however, is the development, after 1953, of cross-shareholdings that gradually progressed among non-financial firms. Table 7 exhibits this 56 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. trend, on the assumption that the member firms are those of the three Presidents’ Councils (Shaco- kai) in 1959.3 0 Table 7. The Rates of Cross-Shareholdings of the Mitsubishi, Sumitomo, and Mitsui Groups (1951-58) Group 1951 1952 1953 1954 1955 1956 1957 1958 Mitsubishi Total (2.7) 9.8 10.6 11.5 11.1 11.6 13.2 14.1 Ordinary Bank (0.6) 1.5 1.9 1.8 1.6 2.1 2.3 2.5 Trust Bank (1.0) 2.7 2.6 3.0 2.1 1.0 2.8 3.7 Life Insurance (0.3) 2.3 2.2 2.5 2.8 3.4 3.0 3.0 Casualty (0.4) 2.8 3.0 3.4 3.5 3.5 3.4 3.2 Non-financials (0.3) 0.6 0.9 0.9 1.1 1.5 1.6 1.8 Sumitomo Total 0.3 9.5 11.2 14.0 14.0 14.7 15.5 17.1 Ordinary Bank 0.0 2.0 1.8 2.7 2.8 2.9 3.2 3.7 Trust Bank 0.2 3.3 3.6 4.0 3.8 2.1 2.5 3.1 Life Insurance 0.0 1.6 1.5 2.0 2.4 2.7 2.8 3.1 Casualty 0.0 1.0 1.0 1.3 1.3 1.4 1.1 1.4 Non-financials 0.1 1.6 3.3 4.1 3.7 5.6 6.0 5.8 Mitsui Total 1.9 4.0 5.2 5.8 5.2 6.2 6.1 6.7 Ordinary Bank 0.2 0.5 0.9 1.0 1.1 1.1 1.3 1.3 Trust Bank 0.3 0.7 0.6 0.5 0.6 0.1 0.4 0.8 Life Insurance 0.0 0.2 0.0 0.0 0.1 1.0 1.1 1.3 Casualty 0.5 0.9 1.5 1.5 1.3 1.6 1.1 1.0 Non-financials 0.9 1.7 2.3 2.8 2.1 2.3 2.2 2.3 Source: Kikkawa (1992: p. 264) using Jojo Kaisha Soran, various issues. Note. The numbers were calculated by the member firms’ stocks which were held by each institution (Bank, Trust Bank, Casualty, and Non-financials) s- the total number o f the stocks issued by the member firms x 100. The member firms are those listed in Table 8. Meiji Life Insurance, Sumitomo Life Insurance, and Mitsui Life Insurance were out o f the calculations because they were not stock companies. Mitsubishi Oil, Mitsubishi Cement, Sumitomo Real Estate, Mitsubishi Real Estate (1951-52), Mitsubishi Corp (1951-54), and Mitsui Corp. (1951-59) were not considered in the calculations. For the Mitsubishi group, the members o f the informal gathering were also considered in the calculations. For the Mitsui group, some Getsuyo-kai members were eliminated from the calculations if they were not the members o f Itsuka-kai, the sub-meeting o f Getsuyo-kai. The numbers o f the parentheses o f the Mitsubishi group are the results after eliminating Mitsubishi Heavy Ind. The company was not considered in the calculation because they were divided into three smaller companies by the Exclusion o f Excessive Economic Concentration Law, and because the stocks o f these three companies were not listed on the stock market yet as o f 1951. 57 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Cross-shareholdings that go beyond the mutual shareholdings between a bank and its customers are the key step toward the keiretsu formation because, as shown in the Japanese Keiretsu Model, they serve to facilitate transactions among member firms through their securing voting rights against each other. In fact, as the Japanese Keiretsu Model predicts, parallel to the increasing cross shareholdings among non-financial firms of each former zaibatsu lineage were the foundations of the general trading companies (GTC) for the purpose of expediting their respective intra-group transactions: during 1948-49 when free trade resumed both domestically and internationally, many former zaibatsu firms, now in orphanhood, found themselves totally at a loss regarding how they could restart trading with no representative agent for them. An executive of Mitsubishi Electric once lamented, “Except the government purchases, we have been consistently working with Mitsubishi Corp. since 1924, and its dissolution is such fatal damage to us” (Miyajima 1992: p. 223). The breaking up of the general trading companies into fragments (in the case of Mitsui and Mitsubishi), or sheer absence of such a corporate entity (in the case of Sumitomo), only contributed to increasing transaction costs among members and to the contraction of the intra-group trade in both scale and scope. After the second revision of the Antimonopoly Law, the zaibatsu orphans began to rally for building up their own GTCs. In June 1952, under the widely-held auspices of the Sumitomo firms, Nippon Construction {Nippon Kensetsu) was transformed into “Sumitomo Corp. {Sumitomo Shoji)." The foundation of Sumitomo Corp. was actually the reflection of the aspiration across the Sumitomo firms “to reduce transaction costs, and to expand trades among the Sumitomo firms” (Kikkawa 1992: p. 270). In response to this, Sumitomo Corp. soon began to play a central role in selling raw materials to and purchasing the final products from, Sumitomo Metal Ind., Sumitomo Electric, Sumitomo Machines, and Sumitomo Chemical (Kikkawa 1992: p. 271).3 1 Similarly, Mitsubishi Corp. was reconstructed in April 1954: for the purpose of reducing transaction costs among the members, the Mitsubishi firms were all in tandem for the unification of four trading companies (Kowa Jitsugyo, Fuji Shoji, Tozai Boeki, and Tokyo Boeki). In particular, “thrice-partitioned” Mitsubishi Ind’s and Mitsubishi Bank played a central role in the unification (Kikkawa 1992: pp. 58 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 272-73). The reconstruction of Mitsui Corp., on the other hand, was delayed to 1959, although the Mitsui firms started the discussion in 1952 (Miyajima 1992: p. 222-24).3 2 Having enlisted the general trading company in addition to a bank in the corporate constellation composed of production firms from various sectors, the firms of the former zaibatsu lineage were growing out of the status of disbanded, zaibatsu orphans, and were securing the golden condition for the concerted group actions. The most telling about this is the formations of the three “Presidents’ Councils” or “Shacho-kaf', the symbolic events for the emergence of the three notable “keiretsu” groups (Mitsui keiretsu, Mitsubishi keiretsu, and Sumitomo keiretsu) in the postwar era.3 3 They were “Hakusui-kai” of the Sumitomo keiretsu (April 1951), “K in’ yo-kai” of the Mitsubishi keiretsu (about 1954), and “Nimoku-kaf ’ of the Mitsui keiretsu (1959). Typically, a Presidents’ Council is a monthly meeting, where the members are selected from Presidents (shacho) of the representative firms (see Table 8). Table 8. Members of the Presidents’ Councils of the Mitsubishi, Sumitomo, and Mitsui Keiretsu, 1960 Sector/Presidents’ Councils Kin’yo-kai (Mitsubishi) Hakusui-kai (Sumitomo) Itsuka-kai (Mitsui) Bank Mitsubishi Bank Mitsubishi Trust Sumitomo Bank Sumitomo Trust Mitsui Bank Mitsui Trust Insurance Tokio F&Ms Sumitomo Life Sumitomo F&M Mitsui Life Taisho F&M Trade & commerce Mitsui Corp. Sumitomo Corp. Mitsui Corp. (Toyo Cotton) (Tokyo Foods) (General Corp.) Agriculture and forestry (Mitsui Norin) Coal Mitsubishi Mining Sumitomo Coal - Mining Mitsui Mining (Hokkaido Tanko Kisen) Construction (Mitsui Construction) (Sanki Kogyo) Foods (Kirin Beer) Table continues next page 59 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 8 (continued). Members of the Presidents’ Councils of the Mitsubishi, Sumitomo, and Mitsui Keiretsu, 1960 Sector/Presidents’ Councils Kin’yo-kai (Mitsubishi) Hakusui-kai (Sumitomo) Itsuka-kai (Musui) Textiles Mitsubishi Rayon Toyo Rayon Paper and pulp Mitsubishi Paper Chemicals Mitsubishi Kasei Mitsubishi Oil Sumitomo Chemical Mitsui Chemical Toyo Koatsu (Mike Gosei) (Mitsui Petrochemical) Petroleum Mitsubishi Petroleum Glass and cement (Ceramics) Asahi Glass Mitsubishi Cement Nippon Sheet Glass Steel Mitsubishi Steel Mitsubishi Kozai Sumitomo Kinzoku Kogyo (Nihon Seikojo) Non-ferrous metals Mitsubishi Metal Sumitomo Kinzoku Kozan Sumitomo Electric Mitsui Metal Electrical machinery Mitsubishi Electric NEC (Nihon Denki) Transportation machinery Mitsubishi Shipbuilding Mitsubishi Nihon Heavy Ind. Shin-Mitsubishi Heavy Ind. Mitsui Shipbuilding (Showa Airplanes) General & precision machinery Sumitomo Machinery (Mitsui Seiki) Real estate Mitsubishi Realty Sumitomo Realty Mitsui Realty Shipping Mitsubishi Marine Transportation (Nippon Yusen) Mitsui Marine Transportation Warehousing Mitsubishi Warehouse Sumitomo Warehouse Mitsui Warehouse Source: Keizai ChOsa Kyokai (1961), p. 10. Note. The parenthesized firms in the Mitsubishi keiretsu belong to other gatherings of the presidents of the Mitsubishi keiretsu. The parenthesized firms in the Mitsui keiretsu belong to Getsuyo-kai, and those not parenthesized, Itsuka-kai, the antecedent of Nimoku-kai (1961). As the zaibatsu descendents, these member firms were chosen from a wide range of core industries, yet had little overlap between the product lines with each other. This embodies one of the significant features of the councils, which Miyazaki (1966) appropriately dubbed the “one-set principle,” referring to a keiretsu group’s intent to place one and only one firm in an industry to forestall internecine competition among member firms. In these councils, any content discussed in 60 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. the meetings is usually kept confidential, and their PR offices explain that the primary purpose of holding them is for the members to exchange information and to promote mutual friendship (Toyo Keizai Shinposha [1994a: pp. 13-14]). The Councils, however, have arguably played much more important roles, especially, in making the group strategy (Kikkawa [1992: pp. 258-59; 1994: p. 147]; Okumura [1976: pp. 96-98, pp. 109-11; 1992: pp. 21-26; 1998: pp. 110-26];and Shimotani [1996: p. 144]). Such an argument is actually convincing because the councils are considered grand shareholders’ meetings, as a result of cross-shareholdings that contributed to pooling more of the total outstanding stocks issued by the members within the group. Thus, as Okumura (1976: p. 95) noted, “The foundation of a Presidents’ Council is the cross-shareholding among the members by which the council can wield its authoritative power.” Taking an example of Mitsubishi’s Kin ’ yo-kai, he argued that Kin ’ yo-kai is the de-facto stockholders’ meeting among the Mitsubishi firms for the following reason: Mitsubishi firms are, in fact, large shareholders to each other in the following fashion; in the first place, Meiji Life Insurance, Tokyo Casualty Insurance, Mitsubishi Heavy Industry, and Mitsubishi Corp. are the large shareholders of Mitsubishi Bank. In the second place, Mitsubishi Bank, Meiji Life Insurance, Tokyo Casualty Insurance, and Mitsubishi Corp. are the large shareholders of Mitsubishi Heavy Industry. In the third place, and finally, Mitsubishi Bank, Tokyo Casualty Insurance, Meiji Life Insurance, and Mitsubishi Heavy Industry are the large shareholders of Mitsubishi Corp. In this way, the member firms of K in’ yo-kai own, in total, 27%, 26%, and 41% of the outstanding stocks issued respectively by Mitsubishi Bank, Mitsubishi Heavy Industry, and Mitsubishi Corp. Therefore, Kin ’ yo-kai is actually considered as a general meeting of the stockholders. (Okumura 1976: p. 76) That K in’ yo-kai is a general stockholders’ meeting implies that K in’ yo-kai is a group strategy meeting for Kin ’ yo-kai’s members through which their revealed interests are filtered into the whole group policy. From this line of reasoning, Okumura (1976: pp. 102-3) conjectured that one of the major reasons behind cross-shareholding is to “govern” the member firms with each other. On the same ground, Kikkawa (1992: p. 252; 1996: p. 147) argued that “a Presidents’ Council is actually 61 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. considered as a grand stockholders’ meeting for the corporate members,” and that “the formation of a Presidents’ Council is a good indication for the formation of an enterprise group.” As a grand stockholders’ meeting based on cross-shareholdings among the member firms, the Presidents’ Council laid the foundation of its decision-making on the mutual consultation among members. This is the major difference from the prewar zaibatsu, where the decision-making was solely delegated to their holding companies.3 4 Consequent to the shift into the pluralism in decision making was the conflict of interest inevitably occurring among the group members in the council. As described well by Gerlach (1992: p. 109), the postwar Presidents’ Council was poised on a subtle power balance between the general trading company and, among all, the bank, playing the centripetal role of maintaining significant linkages with the rest of the members, and the industrial firms playing the centrifugal role of trying to secure their independence. Under such a delicate power balance, the chronic shortage of capital in the earlier postwar period often worked favorably 35 to the former. Once a conflict arose among the members and turned out to be unresolved through the discussion, the settlement was often resorted to the ultimate technique of the members’ votes, so that the conflict would be eventually resolved and that the policies of the members and the whole group would be brought into harmony (Miyashita and Russell 1994: p. 62; Okumura 1976: pp. 89- 90).3 6 2.4.2 The Varying Intra-Group Cohesiveness among the Big Three: A TCE Explanation In the previous section, the analytic focus was placed upon the common feature among the Big Three Keiretsu in their early postwar period, where the disbanded, former zaibatsu firms had come to restore their inter-corporate transaction relations through cross-shareholding, first between bank and firms, and then among non-financial firms, and thereby formed the GTCs and the Presidents’ Councils. It should be noted, however, that there also existed some differential factors among the Big Three in the development of their keiretsu formation as characterized above. For 62 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. instance, Table 7 showed not only the general trend of increasing cross-shareholdings across the Big Three over time, but also the divergence in the pace of their processing them: throughout the period, Sumitomo ranked at the top, followed by Mitsubishi by narrow margins, and Mitsui, a far cry from those two. Another variation among the Big Three was the period when they could form their GTCs and Presidents’ Councils: Sumitomo and Mitsubishi concurrently formed these two organizations in quite earlier periods after Occupation ended (Sumitomo Corp. in June 1952, Hakusui-kai in April 1951, Mitsubishi Corp. in April 1954, K in’ yo-kai in about 1954). Lagging behind these two was, again, Mitsui that could start up Mitsui Corp. and Nimoku-kai in 1959. These observations seem to inform of a positive correlation between the rate of cross-shareholdings and the readiness to open the GTC and the Presidents’ Council. This section tries to provide an answer to the question of why such divergence persisted among the Big Three, and of where it came from. First, in order to see the importance of the keiretsu financing by each group bank, one must consider Table 9, which shows an annual data set of the group lending and borrowing activities between 1953-60 sorted by five categories: A: the main bank’s lending to the member firms (the keiretsu financing), B: the total borrowing of the member firms, C: the main bank’s total lending, D = A/B, and E = A/C. By comparing the Big Three by the above five categories, it is possible to draw the following characteristics of the Big Three’s financial activity during the period: (1) Sumitomo overwhelmed Mitsubishi and Mitsui in D in nearly all the periods (the exceptions in 1956 and of 1960 only). (2) While its D’s were generally a bit lower than those of Sumitomo, Mitsubishi overwhelmed Mitsui and Sumitomo in A, B, C and E over the entire period. This means that, although Mitsubishi Bank had excelled in the keiretsu financing A, it could not satisfy the huge financial needs of the Mitsubishi firms which needed to borrow from outside as well. (3) Mitsui had been almost consistently dominated by Sumitomo and Mitsubishi in A, C and D (the exceptions are A of 1959 and of 1960, and D of 1957 which is bigger than that of Mitsubishi by the very slight margin of 0.1%). 63 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 9. The Main Banks’ Lending to the Member Firms (1953-60) Main bank 1953 1954 1955 1956 1957 1958 1959 1960 Mitsubishi A 10,566 21,517 20,723 22,488 37,108 44,216 42,932 54,318 B 50,118 108,554 113,253 121,209 163,065 199,001 208,093 267,748 C 143,359 169,330 187,352 254,393 334,322 394,011 447,424 525,737 D 21.1 19.8 18.3 18.6 22.8 22.2 20.6 20.3 E 7.4 12.7 11.1 8.8 11.1 11.2 9.6 10.3 Sumitomo A 6,139 9,000 7,436 8,050 14,572 16,644 16,997 21,367 B 23,460 30,875 29,703 43,814 53,395 66,140 81,168 108,255 C 142,417 166,536 184,844 241,574 312,881 394,011 447,424 525,737 D 26.2 29.1 25.0 18.4 27.3 25.2 20.9 19.7 E 4.3 5.4 4.0 3.3 4.7 4.2 3.8 4.1 Mitsui A 5,679 7,626 3,717 5,250 7,348 9,725 18,537 28,668 B 33,860 37,178 22,202 30,239 32,078 50,751 98,091 160,894 C 115,230 130,984 131,138 167,537 208,675 257,018 292,150 354,610 D 16.8 20.5 16.7 17.4 22.9 19.2 18.9 17.8 E 4.9 5.8 2.8 3.1 3.5 3.8 6.3 8.1 Source: Keizai Chosa Kyokai (1961), pp. 22-30. Note. A = the main bank’s lending to the member firms; B = the total borrowing of the member firms; C = the main bank’s total lending; D = A/B (%); E = A/C (%). The members are those listed in Table 8 except Mitsubishi Petroleum, Mitsubishi Cement, and Sumitomo Realty. Mitsubishi Shipping and Nihon Yusen are excluded from both A and B of 1953. Mitsubishi Shipping is excluded from 1955-59. Mitsubishi Corp. of 1953 and Mitsui Corp. of 1953-59 are excluded. This means that, due to its lack of lending ability, Mitsui Bank had been consistently poor in satisfying the financial needs of Mitsui firms. Interestingly, it is readily discemable from (l)-(3) above that D’s are ranked roughly in the order of Sumitomo, Mitsubishi, and Mitsui throughout the periods, which corresponded to their opening GTCs and the Presidents’ Councils. This observation is consonant with the widely accepted view that the keiretsu financing—the term “A” in the present analysis—played significant roles in the keiretsu formations (Okumura 1976: p. 115): offering the keiretsu finance that could bring up D to the highest level, and, thereby, blocking the way for other banks to exercise their influence over the member firms, Sumitomo Bank served as the strong impetus for the formation of the Sumitomo Keiretsu. As reflected in the group’s lower record of D, R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Mitsubishi Bank might have been less salient in solidifying the group ties. Mitsubishi’s lower D was mainly ascribed to the huge investment in the heavy and chemical industry by the Mitsubishi firms during the period, which led them to borrow considerably from outside (Kikkawa 1992: p. 268). In the midst of such a high borrowing demand of the group firms, Mitsubishi Bank responded very favorably to them by offering the generous keiretsu financing, as indicated in the highest A and E throughout the period, and, thereby, made a considerable success in rallying them around. Albeit in a paradoxical way, the importance of the keiretsu financing is also shown by the case of Mitsui, which had a very bad reputation of its weak financial position and group ties (e.g., see Miyashita and Russell 1994: p. 85, 86; Okumura 1976: pp. 184-85): as reflected in the lowest values in both relative (D and E) and absolute terms (C and A), Mitsui Bank’s poor showing in the keiretsu financing is considered the main reason for its delay in the formation of Mitsui Corp. and Nimoku- kai. In fact, there is the record that the poor financing by Mitsui Bank hampered clearing up the bad debt of Dai-Ichi Tsusho, thereby becoming a hard obstacle against the reorganization of Mitsui Corp (Kikkawa 1992: p. 273). Finance is only half of the story. An equally important factor for the keiretsu formation is the asset-specific transaction relations among non-financial firms. Such relations are reflected, albeit indirectly, in the lending-borrowing relations of Table 9. First, by comparing these two keiretsu, it can be readily seen that three absolute numbers, A, B, and C, and two relative numbers, D and E, are 37 both lower in Mitsui than in Mitsubishi. This implies that, although the Mitsui firms were not so much engaged in money-guzzling, heavy and chemical industrialization as those of Mitsubishi (as reflected in Mitsui’s lower B), the keiretsu financing (A) by Mitsui Bank was disproportionately shrunk (as reflected in Mitsui’s lower D and E). This disproportionate reduction is, then, considered to measure its unwillingness to engage in the keiretsu financing for its group firms. This analysis can be developed in a more precise form by introducing simple math manipulation: in Table 9, the relation D > E holds for all the Big Three every year, and it is equivalent to (1/D) < (1/E), or (B/A) < (C/A). From here, it is possible to derive (C/A) - (B/A) (> 0). Considering that “C = the main 65 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. bank’s total lending = the main bank’s lending to the member firms + the main bank’s lending to the non-member firms” and “B = the total borrowing of the member firms = the borrowing from the main bank + the borrowing from the non-main bank,” and that “the main bank’s lending to the member firms = the borrowing from the main bank,” (C/A) - (B/A) turns out to be equivalent to (1/A) • [(the main bank’s lending to the non-member firms) - (the member firms’ borrowing from the non-main banks)] (> 0). As the numerator can be interpreted as “the net financial leakage to the non-member firms,” (C/A) - (B/A) is considered the “net leakage ratios,” which is defined as the ratio of the net financial leakage to the non-member firms relative to keiretsu financing. In a more intuitive term, the net leakage ratio means how much the main bank lends to the non-member firms while lending one yen to the member firms at the same time.3 8 Table 10 compiles the net leakage ratio of the Big Three between 1953-60, the calculation based on the data reported in Table 9. Table 10. The Net Leakage Ratio, Defined as (C/A) - (B/A) Main bank 1953 1954 1955 1956 1957 1958 1959 1960 Aver./Stand. Div. Mitsubishi 00 OO 2.8 3.6 6.0 4.6 4.4 5.6 4.8 5.1/1.7 Sumitomo 19.4 15.1 21.0 24.9 17.6 19.8 21.5 19.3 19.8/2.7 Mitsui 14.5 12.4 29.7 26.5 24.2 21.1 10.6 6.7 18.2/1.8 Note. The unit is yen. The calculations are based on the data o f Table 9. In the table, it is quite obvious that Mitsubishi Bank ran its fund most efficiently among the Big Three in the sense that its financial leakage ratio, ranging from 2.8 to 8.87 (the arithmetic average = 5.1), took the least values among the Big Three throughout the whole period. Being poles apart is Mitsui Bank whose net leakage ratio ranged from 6.7 to 29.7 (the arithmetic average = 18.2), much higher than the records of Mitsubishi. Considering that Mitsui consistently kept the lowest values in A, C, D, and E,3 9 this result apparently reflects the inefficiency of Mitsui Bank’s financing, or again, its unwillingness to engage in keiretsu financing. Why, then, did such 66 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. inefficiency arise in Mitsui Bank? To this question, the asset-specificity perspective poses the hypothesis that, in Mitsubishi, the member firms were densely connected through their asset- specific transactions, while in Mitsui they were not. In other words, the Mitsui firms were generally more independent from one another than the Mitsubishi firms. This is because the more (less) closely the group firms are connected through transactions, the more likely (unlikely) the main bank can retrieve the funds it injected into a group of firms, and, thus, the more the main bank is given the incentive to lend to its group members. Therefore, the asset-specific transaction activity of a group likely leads the main bank to bring down its net leakage ratio. This effect was considered pronounced particularly in the 1950s, all the more for the scarcity of capital.4 0 There is, of course, a possibility that the net leakage ratio substantially reflects how much investment was carried out in the heavy and chemical industry during the period. This is especially true of Mitsubishi, whose lower net leakage ratio is most likely the reflection of the need of Mitsubishi Bank to efficiently run its funds in meeting the high financial demand from its member firms to pursue the heavy and chemical industrialization. Likewise, the higher net leakage ratio of Mitsui most likely represented the overall delay in the heavy and chemical industrialization in the group, which, in fact, coincided with its generally lower B, and also with the widespread view that “Mitsui is generally poor in the heavy and chemical industry” (Okumura 1976: p. 185). But this does not vitiate the TCE interpretation because the heavy and chemical industrialization itself was the main factor of promoting asset-specific transactions among the member firms. Introducing the concept of “internalization of the external economies,” Miyazaki (1955: p. 54, 180: pp. 72-3) argued that the heavy and chemical industrialization necessarily brings about the benefits accruing from the inter-sectoral exchanges within a keiretsu group. He illustrated the concept by taking up an example of the automobile industry:4 1 As the automobile industry is established and developed in the Japanese economy, the manufacturers of thin steel plate, tires, piston rings, ball-bearings, electrical fittings and other automobile parts will naturally benefit from it. . . . By its essential nature this kind of 67 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. “external economy” produces effects on the generality of industries through the intermediacy of the element of complementary production, through the relations existing among the good produced. (1966: pp. 54- 55; 1980: p. 72) Miyazaki’s “internalization of the external economies” suggested the dynamic nature of the synergy effect, where the close (distant) inter-firm relations brought about by the group investment activities in the past encourages (discourages) the group bank to finance more of the group firms, and that, in turn, feedback to further consolidate (weaken) the inter-firm relations. In order for such a synergistic effect to be brought into full play in the postwar period, the basic inter-firm relations must have been already prepared, to a large extent, during the prewar, zaibatsu period. In this regard, Gerlach (1992: pp. 22-3) argued that “[cjontemporary patterns of intercorporate relations in Japan reflect distinctive path-dependent trajectories that represent a considerable carryover from the prewar period.” As witnessed in Section 2.3, the ways the inter-firm relations were molded differed significantly between the Mitsubishi and Mitsui zaibatsu. In the Mitsubishi zaibatsu, the firms were, in many cases, founded endogenously in the process of branching out activities and thus could keep their technological continuity even after positioned independently in various industrial sectors. That the Mitsubishi zaibatsu developed this way was mainly ascribed to its initiating the production and conglomeration activities, mostly in the heavy and chemical industrial sectors, as represented by Takashima Coal-Mining, Nagasaki Shipbuilding, and Mitsubishi Electric. To the Mitsubishi zaibatsu, doing so was virtually the only way for its future prosperity because, unlike the Mitsui and Sumitomo zaibatsu, Mitsubishi had no industrial establishment prior to the Meiji era. By contrast, in the Mitsui zaibatsu, the inter-firm relations were hardly developed in the context of the heavy and chemical industrialization. This is mainly because the Mitsui firms were mostly bought out randomly from outside by Mitsui Bank, and, except the mining business, were biased somewhat toward the light industry (camphor, paper, sugar, synthetic dyes and so on). Above all, Mitsui’s forte was primarily in the trading activity of Mitsui Corp., which was of no use in the development of asset-specific, inter-firm relations. All those 68 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. characteristics of the Mitsui zaibatsu derived, in turn, from its long history of financial business, which prompted Mitsui to take the development path based on its bountiful commercial capital in the Meiji era.4 2 The inter-firm relations thus nurtured in varying degrees across the zaibatsu groups had seemingly disappeared in the process of the zaibatsu dissolution. But they were suddenly put to the test when disbanded, former zaibatsu firms were given the chance to reunite in the form of keiretsu in the 1950s. With the tight inter-firm transaction ties still remaining among the disbanded members, Mitsubishi Corp. was reorganized smoothly after Occupation ended. In trying to reduce transaction costs, the Mitsubishi firms were all in tandem with supporting the foundation of Kowa Jitsugyo (April 1950), the chief successor of the former Mitsubishi Corp. When Kowa Jitsugyo tried to inherit the Mitsubishi Corp. name, it could easily gain the quadripartite agreement with the other three trading companies (Fuji Shoji, Tozai Boeki, and Tokyo Boeki). In July 1954, unification of these four small GTCs was completed under the auspices of other Mitsubishi firms, Mitsubishi Bank and “three-partitioned” Mitsubishi Heavy Ind. (Kikkawa 1992: pp. 279-82). On the other hand, despite its long, prestigious career in prewar time, the reorganization of Mitsui Corp. was far from easy, with a series of confrontations among the Mitsui firms and fourteen smaller GTCs that used to be the parts of the former Mitsui Corp. First, contrary to the case of Kowa Jitsugyo of the Mitsubishi keiretsu, there occurred a dispute among Mitsui’s GTCs in June 1951 when Nitto Soko Tatemono, the one deemed as succeeding the former Mitsui Corp., tried to adopt the name of Mitsui Corp. In April 1952, these trading companies held a meeting to settle the problem of the reunification of Mitsui Corp. and of the usage of the group symbol, but to no avail. Adding to this was virtually no cooperation for the reunification on the side of the Mitsui firms, at least, before 1954. All these facts are congruent with the previous observations of how Mitsui’s development path failed to nurture its intra-firm transaction relations. Mitsui Corp., itself, admitted that, pointing out the following as the primary reason for its delay in the reunification: “Mitsubishi Corp. was reorganized quite earlier, and that is mainly because of its historical background, being quite different from that of Mitsui 69 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Corp per se. First of all, in Mitsubishi, there existed gigantic makers. First and subsequently, Mitsubishi Corp. was set up for satisfying the needs of these makers through the supply of the raw materials and the export of their products. To the contrary, in Mitsui, there existed Mitsui Corp. first, and the gigantic makers, such as Toyo Rayon and Mitsui Shipbuilding, were subsequently set up as its subsidiaries. In Mitsubishi where the strong leadership of the various industrial firms was manifested, they had a high demand for owning their general trading company, thus facilitating the reorganization of Mitsubishi Corp., particularly, during heavy and chemical industrialization underway (Kikkawa 1992: p. 274, citing The Challenge and Creation: One Hundred Years o f Mitsubishi Corp., p. 185).” The case of the Sumitomo keiretsu, the analysis of which has been postponed by now, is not so clear-cut as in the Mitsui and Mitsubishi keiretsu, because it is impossible to tell whether its highest D was brought about by higher A (the numerator of D), which correlates with high asset specificity among firms, or lower B (the denominator of D), which correlates with lower asset specificity among them. But Sumitomo’s inter-industrial expansion, originated by Besshi Copper and Mining Co., suggests its steadfast inter-corporate relations. Besides, there is a record that Sumitomo Corp. was transformed from Nippon Construction, which itself was developed from Sumitomo Land & Construction in 1945 for the encouragement of the group’s interior transactions and that, at its inception, Sumitomo firms provided assistance by dispatching their directors and workers (Okumura 1976: p. 201). Thus, like Mitsubishi firms, Sumitomo Corp. was generally ruled by other group firms, which were in close contact with one another through the past development process, and that, in turn, facilitated the establishment of Sumitomo Corp. in the earlier, postwar period of 1952.43’4 4 2.4.3 The Era o f the Heavy and Chemical Industrialization The first wave of the heavy and chemical industrialization in postwar Japan took place in the first half of the 1950s under the policy of the “rationalization investment” (gorika toshi), which 70 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. was originally kindled by the special procurement demand of the Korean War (June 1950 - July 1953), and which, thereby, invigorated the Japanese economy dampened since the closure of the RFB and Japan’s shift into a single exchange rate system in April 1949.4 5 The focal point of the rationalization investment was to overcome the technological backwardness in the three areas of heavy industry—steel, marine transportation, and coal—through transplanting more advanced, foreign technologies (Hashimoto 1995: pp.89-92; Kato 1975: p. 117; Kikkawa 1992: p. 257): with the outcomes of the wartime effort, these three industries embodied the most advanced technological levels at the time in Japan, yet were retarded by the Western standard. It was under such a technological condition that the former Big Three zaibatsu firms rallied around and undertook the rationalization investment during the period. With the resurgence of city banks and the support from the Japan Development Bank (JDB, nihon kaihatsu ginko, founded in 1952), the Big Three’s keiretsu activities gradually became salient by their main banks’ supplying funds to member firms, among all, those belonging to these three industries, and by the members holding equities issued from those positioned in the three industries (Kikkawa 1992: p. 265, pp. 269-70. See also Kato [1975: Table 4 on p. 116] for the lists of the companies which borrowed more than 1 billion yen from their main banks in 195 5).46 Also noteworthy of the rationalization investment, especially, from the keiretsu perspective, is that these three industries were chosen as targets not only because of their technological importance of those days, but also because of the significance in their inter industry relations: in the early 1950s, the Japanese economy was caught by a “coordination failure” when expensive coal induced the production of expensive steel and, consequently, expensive marine vessels which, through a vicious circle, resulted in expensive coal imports. The Japanese government aimed to break away from this vicious circle by upgrading the technological standard of the three industries (Hashimoto 1995: p. 148; Okazaki 1996)4 7 The rationalization investment, however, came to an end around 1955, after the armistice of the Korean War in 1953, and the BOJ exacted a tight money policy due to its lack of foreign currencies. 71 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. The rationalization investment being short-lived notwithstanding, the private investment quickly regained its strength in the form of a second wave of heavy and chemical industrialization which, through its strong sustainability, brought about high economic growth in Japan with an annual average rate of more than 10 percent until 1970. During this period various inter-industry relations and, accordingly, asset-specific relations among firms were remarkably improved in Japan: underway during the period were the unprecedented level of ventures by young corporate executives, who were abruptly promoted when their supervisors were all purged in the Occupation regime, and the “energy revolution” where petroleum and its derivatives gained a significant footing in the supply of energy and raw and medium materials. Their consequences were sweeping, inter industry relations emerging among the new types of industry, notably the steel-related industry (automobiles, home electric appliances, factory machinery, and shipbuilding) and the petroleum- related industry (petroleum refining, petro-chemistry, synthetic resins, and synthetic fibers) (Kikkawa 1992: pp. 276-77). Because the products of these two industries were often raw or medium inputs of other industries, it became compelling that, in order to secure the premise of their long-term growth, they established a dense transaction network with no regard to in which production sector they were positioned. This is the very reason why enterprise groupings became necessitated on the part of firms during the high growth era: by joining in an enterprise group or “keiretsu,” firms could expect the general trading company (GTC), on the side of real transaction, to facilitate their purchasing and selling their inputs and outputs among themselves in an asset-specific fashion, and the bank, on the side of financial transaction, to support disbursing their extravagant production plans in the heavy and chemical industries. Such technological connections among the firms were also found favorable on the part of banks because, in the midst of financial difficulty of the early postwar period, they were considered to improve the monetary flow-back to banks through transactions among affiliate firms (Okumura 1976: p. 110; Suzuki 1993: p. 79). Okumura (1976) well captured this aspect of the keiretsu financial transaction in the following example: 72 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. When Sumitomo Steel, for example, proposes to improve its working capital, it is always the case that Sumitomo Bank lends to Sumitomo Corp. instead of Sumitomo Steel. If the fund is actually allotted for the improvement of working capital of Sumitomo Steel, then the fund will, sooner or later, be returned to Sumitomo Bank through Sumitomo Corp. That is, the fund only runs on the following circles where Sumitomo Bank is at both the starting and ending points: Sumitomo Bank — > Sumitomo Corp. -» Sumitomo Steel (-» Sumitomo Corp. -») — > Sumitomo Bank. In this case, the fund never flows out of Sumitomo Bank, because the fund transfers among these agents are carried out only by the name transfers within Sumitomo Bank. In addition, Sumitomo Bank can benefit through this transaction because it can increase both the lending and deposit at a time.4 8 Thus, as witnessed in the previous sections, it was quite natural for both banks and firms to reunite on the basis of their former zaibatsu belongings, and to precipitate their groupings in the mode of the “one-set principle,” where the members were positioned across the wide variety of industries, yet in such a manner to eschew their internecine competition by the duplication in industries (see again Table 8 for how firms of the Presidents’ Councils were positioned in the representative industries).The inter-corporate relations for encouraging asset-specific transactions, as illustrated in the Japanese Keiretsu Model, were thus being forged during this period. First, the nature of the GTC’s was noticeably changed in the course of the heavy and chemical industrialization in all the keiretsu groups: their role as wholesale dealers based on foreign trades, which itself was molded during the prewar period, gradually receded and, instead, that of intermediating the extensive range of domestic, inter-corporate transactions— from soup noodles to a rocket—had come to the fore (Suzuki 1998: p. 191).4 9 Among all, their offering credit became crucial in intermediating corporate transactions: based on their credit checking in advance, the GTCs usually granted credit to buyers and assumed all the default risks involved if the transaction was carried out directly between buyers and sellers (Dore 1986: p. 68; Miyashita and Russell 1994: pp. 56-8).5 0 Furthermore, by offering credit, the GTCs facilitated inter-corporate transactions more than the value of cash they received from their main banks: for example, in 1970, the total credits that the GTCs created were about twice as much as their borrowings from their main banks, and about ten 73 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. times as much as their own capital (Yamanaka 1989: pp. 77-8). While doing so, the GTCs usually let their purchasing firms issue short-term (usually more than 5 months) promissory notes addressed to them (Yamanaka 1989). The GTCs had, thus, significantly contributed to smoothing out transactions even when the buyers were not prepared for cash to pay (Uchida 1976: p. 347).5 1 Deploying their constituent GTCs’ superb skill in intermediating corporate transactions, all keiretsu groups increased their intra-group, asset-specific transactions, which were becoming more and more significant as the heavy and chemical industrialization progressed. Uchida (1976: pp. 345- 46), for example, reported that, in 1966, although the GTCs’ aggregate transactions totaled only 12% of the total wholesaling in Japan, their shares in chemical and steel products were respectively 52 at quite high ratios of 38% and 21%. Backed up by the group’s heavy and chemical industrialization policy, Mitsubishi Corp. came to deal with 30% of all the intra-group transactions of the Mitsubishi keiretsu (Uchida 1976: 343). Despite its earlier stalemate of the intra-group, steel transactions due to the breakup of Mitsubishi Heavy Ind. into three smaller companies by LPECEP,5 3 Mitsubishi Corp. gradually improved the situation through its effort to re-establish the relations with them. For example, throughout the period of the first to seventeenth government shipbuilding plans (1947-1960), Mitsubishi Corp. secured the orders of shipbuilding from Nippon Yusen, Mitsubishi Shipping, and Osaka Shosen,5 4 expanding its supply of steel to the three- partitioned Heavy Inds. Besides, Mitsubishi Corp. served as the group’s chief organizer by opening new trade channels between group firms and the three-partitioned Heavy Inds. when the latter decided to embark on the “landing operations” in the face of the overall depression in the shipbuilding industry in the late 1950s. In 1954, Mitsubishi Corp. eventually achieved the unification of the three-partitioned Heavy Inds. into Mitsubishi Ind. and made an exclusive contract with the united Mitsubishi Ind. to appoint Mitsubishi Corp. as its sole trade agent (Kikkawa 1992: p. 280; Suzuki 1993: p. 126). After the contract, Heavy Ind.’s steel purchase from the Corp. doubled (from 40% to 80%) of its total steel purchase (Kikkawa 1992: p. 281; Suzuki 1993: p. 107). 74 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Mitsubishi Corp. also played a significant role in the resolution of the intra-group conflict, when some Mitsubishi firms rushed into the oil industry (Suzuki 1993: pp. 107-8, 109-10). In a similar fashion, Sumitomo Corp. started to play a central role in improving intra-group transactions: in 1950-55, Sumitomo Corp. increased its purchase of chemical products from Sumitomo Kasei up to 35% (Suzuki 1993: p. 111). In the second half of 1951, Sumitomo Corp.’s purchase from Sumitomo Metal, Sumitomo Kasei, and Sumitomo Kinzoku Kozan accounted for 60% of their total output, and its sales to Sumitomo Metal, Sumitomo Electric, and Toyo Aluminum reached 41% of the Corp.’s total (Suzuki 1993). As displayed in Table 11, the state of high intra-group transactions, especially in the areas of metal, chemical, and machinery, was consistently maintained throughout the whole high-growth period.5 5 Some major inter-corporate relations inside the Mitsubishi and Sumitomo keiretsu that had developed during the high-growth era are summarized in Tables 12 and 13.5 6 Table 11. The Breakdown of Goods Traded in Sumitomo Corp. (the Second Half of 1952 to the Second Half of 1964) Goods The second half o f 1952 (%) The second half o f 1954 (%) The second half o f 1959 (%) The second half o f 1964 (%) Steel 44 43 41 37 Nonferrous Metals 16 20 18 15 Electric Machinery 8 7 6 6 Machinery 8 7 13 9 Fertilizer, Foods 14 15 10 12 Chemicals 2 4 5 7 Fibers 8 4 2 3 Fuel 0 1 5 11 Total 100 100 100 100 Source: Sumitomo Shoji Kabushiki Gaisha (1992: pp. 337, 383, 401). 75 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 12. Selected Examples of the Inter-Firm Relations of the Mitsubishi Keiretsu Established in the Heavy and Chemical Industrialization Period Products or mediatory roles The flow o f goods 1947-1960 (the 1st to 17th government shipbuilding plans) steel materials 3 Mitsubishi Heavy Inds. -» (MC) — > Nippon Yusen, Mitsubishi Marine Transportation, Osaka Shosen 1954 freight cars, and their electric ntartc Mitsubishi Electric — > (MC: electric parts) -» New Mitsubishi Heavy Ind. — » (MC: freight cars) -» Mitsubishi Kogyo 1955 (the year when the Mitsubishi Atomic Power Committee and the Mitsubishi Missile Committee were set up) various electric products (the aimiation o f the two firms) MC— Mitsubishi Electric and other member firms 1957 steel materials and machinery (mutual trade via MC) Yawata Steel, Nihon Kokan, Kawasaki Steel -» (MC: steel materials) -» 3 Mitsubishi Heavy Inds. (MC: machinery) — > Yawata Steel, Nihon Kokan, Kawasaki Steel 1958-1959 (the landing operations by 3 Mitsubishi Heavy Inds.) assorted heavy and chemical products 3 Mitsubishi Heavy Inds.— (MC)— other Mitsubishi firms About 1960 lubricant Mitsubishi Petroleum -» (MC) -» Mitsubishi Kasei About 1960 the establishment o f Mitsubishi LPG, the group’s representative sales agent o f LPG Mitsubishi Petroleum, Mitsubishi Kasei, Mitsubishi Yuka — » (MC) — » Mitsubishi LPG About 1960 bauxite, aluminum (MC: imported bauxite) — > Mitsubishi Kasei — > (MC: refined aluminum materials) -» Mitsubishi Raynolds -» (MC: processed aluminum) *Mitsubishi Raynolds was established in 1962 mainly by Mitsubishi Kinzoku Kogyo and Mitsubishi Kasei under the intermediation by Mitsubishi Corp. Early 1960 petroleum derivatives: MC mediated some Mitsubishi firms in conflict when they tried to enter the petrochemical industry Mitsubishi Yuka— (MC)— Mitsubishi Kasei, Mitsubishi Kasei— (MC)— Mitsubishi Rayon, etc. 1964 various heavy and chemical products: support o f the unification among 3 Heavy Inds. MC— Mitsubishi Heavy Ind. Source: Compiled from Kikkawa (1992: pp. 295-96) and Suzuki (1993: pp. 104-10). Note. The arrows in the table signify the direction o f the goods under transaction. “MC” stands for Mitsubishi Corp. As the transaction mediator, it is parenthesized in the transaction flows. When the inter-firm transaction becomes rather complicated, the goods involved are respectively displayed in the right hand side within the parentheses. Suzuki (1993: pp. 74-101) and Aki and Okita (Showa Dojinkai 1960) also abound with inter corporate relations established during the period, but do not explicate whether they were intermediated by GTCs (general trading companies). 76 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 13. Selected Examples of the Inter-Firm Relations of the Sumitomo Keiretsu Established in the Heavy and Chemical Industrialization Period Products or mediatory roles The flow of goods The late 1940s - the early 1950s steel pipes for the chemical industries Sumitomo Kinzoku Kogyo — > (SC) — » Sumitomo Chemical - steel materials Sumitomo Kinzoku Kogyo -» (SC) -» Sumitomo Machinery -- dregs o f sulfuric acid, aluminum Sumitomo Chemical — > (SC) -> Sumitomo Kinzoku Kogyo - production machinery, plant for steel production Sumitomo Machinery — » ■ (SC) — > Sumitomo Kinzoku Kogyo - special electric wires Sumitomo Electric -» SC) -» Sumitomo Kinzoku Kogyo - copper, nickel, electricity- charged copper Sumitomo Kinzoku Kozan -> (SC) — » Sumitomo Electric - unknown Sumitomo Chemical - (SC) - Sumitomo Electric - unknown Sumitomo Coal Mining - (SC) - Sumitomo Kinzoku Kogyo - aluminum, electricity-charged copper Sumitomo Kinzoku Kogyo, Sumitomo Chemical -» (SC) — > Sumitomo Light Metal 1954 affiliation through the reconstruction Sumitomo Machinery - SC 1959 (the year when the Sumitomo Atomic Power was set up) various electric products: the affiliation o f the two firms SC - Sumitomo Electric and other member firms About 1960 electric products: affiliation Meidensha - SC 1961 cement: affiliation lwaki Cement - SC Uncertain acrylic materials Sumitomo Chemical-* (SC) -» Nihon Exlan Uncertain synthetic fibers Sumitomo Chemical -» (SC) -> Toyobo, Asahi Kasei Source: Compiled from Suzuki (1993: pp. 112, 128). Note: “SC” stands for Sumitomo Corp. As the transaction mediator, it is encircled by a parenthesis in the transaction flows. The arrows in the table signify the direction o f the goods under transaction. “SC” stands for Sumitomo Corp. As the transaction mediator, it is parenthesized in the transaction flows. When the inter-firm transaction becomes rather complicated, the goods involved are respectively displayed in the right hand side within the parentheses. The inter-corporate connections thus developed were then to be coupled with the lending support from the main bank: in general, during the high-growth era, the main banks of the Big Three keiretsu intensively financed their member GTCs and others positioned in the heavy and chemical 77 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. industries, thereby encouraging their intra-group transactions, both directly and indirectly. This is portrayed in Tables 14-16, which respectively exhibit the Presidents’ Councils members ranked in the order of lending they received from their main banks in 1960. Table 14. The Ranking of the Mitsui Firms in the Order of Mitsui Bank’s Lending (1960) A B C 1. Mitsui Corp. Mitsui Norin (0.069) N/A (14,704:23.19%) Showa Airplane (0.04) N/A Mitsui Realty (0.025) 15/17(0.0003) General Corp. (0.023) 4/6 (0.0005) Mitsui Bank (0.02) 3/17 (0.072) 2. Toyo Rayon Mitsui Petro-chemical (0.1) N/A (3,700: 24.03%) Mitsui Bank (0.008) N one/17 Mitsui Realty (0.008) 13/17(0.002) Mitsui Chemical (0.006) None/7 Mitsui Corp. (0.001) None/17 3. Mitsui Mining Mitsui Construction (0.52) N/A (3,446: 14.46%) Miike Gosei (0.119) 5/9 (0.004) Mitsui Petro-chemical (0.1) N/A Mitsui Norin (0.033) N/A Mitsui Bank (0.011) 13/17(0.017) 4. (Toyo Cotton) Mitsui Warehouse (0.015) 3/11 (0.009) (2,831: 19.28%) Mitsui Bank (0.002) 9/17 (0.026) Mitsui Shipping (0.002) 5/9 (0.005) 5. General Corp. Mitsui Warehouse (0.01) 4/11 (0.006) (2,272: 24.65%) Shipping/Realty (0.003) 3/9 (0.013)/N o n e/17 Mitsui Bank (0.002) 6/17 (0.033) Mitsui Corp. (0.0005) 5/17 (0.023) 6. Nihon Seikojo) Mitsui Bank (0.01) 4/17(0.064) (1,985: 25.82%) Mitsui Realty (0.01) 5/17(0.026) M'Nui Mining (0.003) None/8 M tsui Shipping (0.001) None/9 Mitsui Corp. (0.001) 15/17(0.001) Table continues next page R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 14 (continued). The Ranking of the Mitsui Firms in the Order of Mitsui Bank’s Lending (1960) A B C 7. Mitsui Chemical Mitsui Petro-chemical (0.1) N/A (1,930: 40.45%) Miike Gosei (0.03) None/9 Mitsui Bank (0.008) 12/17(0.019) Shipping /Realty (0.004) None/9 / 10/17(0.007) Mitsui Warehouse (0.001) None/11 8. Mitsui Realty Mitsui Construction (0.399) N/A (1,872: 20.85%) Mitsui Norin (0.127) N/A Mitsui Petro-chemical (0.1) N/A Mitsui Warehouse (0.052) 6/11 (0.001) Nihon Seikojo (0.026) 1/6 (0.01) 9. (Tokyo Shokuhin) Mitsui Norin (0.02) N/A (1,604: 24.48%) Mitsui Bank (0.003) 5/17(0.05) Mitsui Warehouse (0.003) 2/11 (0.014) Mitsui Shipping (0.001) 2/9(0.014) 10. ToyoKoatsu Miike Gosei (0.125) 3/9 (0.003) (1,000: 8.1%) Mitsui Petro-chemical (0.1) N/A Mitsui Bank (0.014) 8/17 (0.018) Mitsui Realty (0.005) 14/17 (0.0005) Mitsui Corp. (0.004) 8/17(0.016) 11. Mitsui Shipping Tokyo Shokuhin (0.014) 3/5 (0.001) (794: 2.92%) General Corp. (0.013) 2/6 (0.003) Mitsui Bank (0.01) 6/17 (0.033) Corp./Toyo Cotton (0.005) 5/17 (0.023) / 4/6 (0.002) Mitsui Zosen (0.001) 2/10(0.034) 12. (Hokkaido Tanko Risen) Mitsui Norin (0.12) N/A (780: 8.71%) Mitsui Bank (0.011) None/17 Mitsui Realty (0.008) None/17 13. Mitsui Metal i Mitsui Petro-chemical (0.1) \ V (773:20.27%) MiikcGosci (0.021) None 0 Mitsui Realty (0.01) None 17 Mitsui Shipping (0.007) Nune/9 Mitsui Bank (0.006) ! 11/17(0.02) 14. Mitsui Warehouse i Mitsui Norin (0.018) i N/A (288:41.68%) Tokyo Shokuhin (0.014) ! 2/5 (0.003) Toyo Cottons (0.009) i 1/6(0.015) General Corp. (0.006) i 1/6(0.01) ____________ ___ _________________ I Mitsu Bank (0.003) ________ | 7/17(0.03) Table continues next page 79 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 14 (continued). The Ranking of the Mitsui Firms in the Order of Mitsui Bank’s Lending (1960) A B C 15. Mitsui Shipbuilding Showa Airplane (0.185) None (161: 10.94%) Mitsui Shipping (0.034) 6/9 (0.001) Mitsui Petro-chemical (0.025) N/A Mitsui Bank (0.008) None/17 Miike Gosei (0.004) None/9 16. (Miike Gosei) Mitsui Pctro-chemical (0.1) N/A (104: 33.55%) Mitsui Construction (0.005) N/A Koatsu/Realty (0.003) 1/6 (0.125)/8/17 (0.01) Mitsui Mining (0.004) 2/8 (0.119) Mitsui Corp. (0.001) 10/17(0.011) 17. (Sanki Kogyo) Mitsui Realty (0.005) N one/17 (100: 11.43%) Mitsui Bank (0.003) 7/17 (0.03) Mitsui Corp. (0.003) 5/17 (0.023) Mitsui Warehouse (0.003) 6/11 (0.001) Source: The above data are calculated from Keizai Chosa Kyokai (1964). Note. The firms considered in this table are limited to those belonging to the Presidents’ Council (in this case, thus, Getsuyo-kai and Itsuka-kai). Columns A, B, and C, respectively, contain the following information: Column A: Mitsui firms ranked in the order o f Mitsui Bank’s lending. The two numbers in the parenthesis under each firm’s name are Mitsui Bank’s lending to the firm, that is, the amount o f borrowing by the firm from Mitsui Bank (the left-hand side in the parenthesis. The unit is billion yen.) and the borrowing ratio, defined as “the borrowing from Mitsui Bank divided by the total borrowing” o f the firm (the right-hand side of the parenthesis). The parenthesized firms are the members o f Getsuyo-kai, and those not parenthesized are the members o f Itsuka-kai, both being the antecedents o f Nimoku-kai. Column B: the top five companies ranked in the order o f the ratios o f the stocks held by the firm o f Column A. The number in the parenthesis denotes the ratio. Column C: the numerator o f the fraction denotes the ranking o f the firm o f Column A when measured by the ratio o f its stock held by the firm o f Column B. The denominator denotes the total number o f group firms whose stocks are held by the firm o f Column B. N/A in Column C means that the firm o f Column B does not hold any stock issued by the firm of Column A. The number in the parenthesis is the ratio o f the number o f the stock o f the firm A held by the firm o f Column B, relative to the total number o f the stocks issued by the firm of Column A. 80 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 15. The Ranking of the Mitsubishi Firms in the Order of Mitsubishi Bank’s Lending (1960) A B C 1. Mitsubishi Corp. Mitsubishi Oil (0.06) N/A (12,830: 24.79%) Mitsubishi Warehouse (0.021) 2/16 (0.008) Mitsubishi Shipping (0.016) ! 1/13(0.01) Mitsubishi Mining (0.014) 5/11 (0.008) Mitsubishi Kozai(0.013) 5/8 (0.0003) 2. Shin-Milsubishi Mitsubishi Seiko (0.025) 4/6(0.0004) Heavy Ind. Mitsubishi Corp. (0.02) 11/21 (0.004) (6,488: 18.19%) Mitsubishi Bank (0.018) 9/20(0.027) Mitsubishi Shipping (0.014) 1 9/13 (0.0004) Mitsubishi Realty (0.011) 11/17 (0.001) 3. Mitsubishi Electric Mitsubishi Corp. (0.016) 9/21 (0.06) (5,078: 19.64%) Mitsubishi Realty (0.012) 10/17(0.002) Mitsubishi Bank (0.011) 16/20 (0.013) Mitsubishi Shipping (0.005) 8/13(0.001) Mitsubishi Kasei (0.004) 11/15 (0.002) 4. Mitsubishi Rayon Mitsubishi Oil (0.12) N/A (4,440: 48.94%) Mitsubishi Shipbuilding (0.025) - ! 4/10 (0.01) Mitsubishi Bank (0.01) 1 5/20(0.044) Kasei/Corp. (0.005) I 6/15 (0.01)/7/21 (0.01) Mitsubishi Warehouse (0.004) 5/16(0.001) 5. Mitsubishi Kasei Mitsubishi Oil (0.121) 1 N/A (3,184: 19.98%) Mitsubishi Cement (0.083) None/4 Mitsubishi Mining (0.048) 2/11 (0.02) Mitsubishi Shipping (0.032) 6/13 (0.003) Mitsubishi Bank (0.014) 8/20(0.034) 6. Asahi Glass Mitsubishi Oil (0.122) i N/A (3,100: 32.89%) Mitsubishi Cement (0.083) None/4 Mitsubishi Realty (0.028) 10/17(0.002) Mitsubishi Warehouse (0.023) i 5/16(0.001) Shin-Mitsubishi Heavy (0.009) 10/14(0.005) 7. Mitsubishi Mitsubishi Corp. (0.041) 11/21 (0.004) Shipbuilding Mitsubishi Seiko (0.025) 2/10(0.001) (2,843: 16.65%) Mitsubishi Shipping (0.019) None/13 Mitsubishi Rayon (0.01) 2/15 (0.025) Mitsubishi Bank (0.009) 9/20 (0.027) 8. Mitsubishi Nihon Mitsubishi Seiko (0.025) 3/6(0.0005) Heavy Ind. Mitsubishi Shipping (0.021) None/13 (2,777: 20.82%) Mitsubishi Corp. (0.016) 10/21 (0.005) Mitsubishi Kozai(0.013) None/8 Mitsubishi Realty (0.009) : 11/17 (0.001) Table continues next page 81 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 15 (continued). The Ranking of the Mitsubishi Firms in the Order of Mitsubishi Bank’s Lending (1960) A B C 9. (Nippon Yusen) Mitsubishi Bank (0.004) None/20 (2,230: 7.08%) Mitsubishi Corp. (0.004) None/13 Shin-Mitsubishi Heavy (0.0005) 11/14 (0.004) 10. Mitsubishi Metal Mitsubishi Oil (0.062) N/A (1,814:48.62%) Mitsubishi Cement (0.017) None/4 Mitsubishi Mining (0.009) 6/11 (0.005) Seiko/Bank/Realty (0.005) None/6/14/20(.018)/10/17(.002) Mitsubishi Corp. (0.004) 11/21 (0.004) 11. Mitsubishi Realty Mitsubishi Cement (0.017) None/4 (1,671:9.92%) Mitsubishi Warehouse (0.016) 4/16(0.003) Mitsubishi Mining (0.014) 4/13 (0.005) Mitsubishi Shipping (0.009) 4/13 (0.005) Mitsubishi Corp. (0.008) 9/21 (0.006) 12. (Kirin Beer) N/A N/A (1,530: 34.13%) ............. “ ........ ........ ‘ ' ......... 13. Mitsubishi Kozai Mitsubishi Bank (0.004) 10/20 (0.025) (1,420: 42.24%) Shipping/Cement (0.002) 3/13 (0.008)/N one/4 Mitsubishi Realty (0.001) None/17 Shin-Mitsubishi/Shipbuilding (0.0004) None/14 / None/10 Mitsubishi Corp. (0.001) 6/21 (0.013) 14. Mitsubishi Petro. Mitsubishi Shipping (0.014) 5/13(0.001) (1,310:21.28%) Mitsubishi Bank (0.009) None/20 Mitsubishi Corp. (0.006) 8/21 (0.008) Mitsubishi Cement (0.003) None/4 Mitsubishi Realty (0.001) 10/17(0.002) 15. Mitsubishi Paper Mitsubishi Shipping (0.07) None/13 (955: 28.30%) Mitsubishi Rayon (0.06) None/13 Mitsubishi Bank (0.009) 11/20 (0.022) Mitsubishi Corp. (0.004) 11/21 (0.004) Mitsubishi Nihon (0.001) None/11 Table continues next page 82 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 15 (continued). The Ranking of the Mitsubishi Firms in the Order of Mitsubishi Bank’s Lending (1960) A B C 16. Mitsubishi Mining Mitsubishi Cement (0.501) None/4 (939: 9.93%) Mitsubishi Kasei (0.026) 3/15 (0.048) Mitsubishi Oil (0.023) N/A Mitsubishi Realty (0.013) 4/17(0.014) Mitsubishi Corp. (0.008) 5/21 (0.014) 17. Mitsubishi Seiko Mitsubishi Bank (0.001) 10/20 (0.025) (612: 17.19%) Mitsubishi Shipbuilding (0.001) 2/10 (0.025) Mitsubishi Nihon (0.0005) 1/11 (0.025) Shin-Mitsubishi (0.0004) 1/14 (0.025) Mitsubishi Mining (0.0003) 6/11 (0.005) 18. Mitsubishi Mitsubishi Shipping (0.01) 3/13 (0.008) Warehouse Mitsubishi Corp. (0.008) 3/21 (0.021) (598: 50.46%) Mitsubishi Cement (0.004) y4 (o.oo3) Realty / Bank (0.003) 2/17 (0.016)/5/20 (0.044) Denki /Asahi /Rayon (0.001) 3/10(.011 )/4/l5(.023)/6/l5 (.004) 19. Mitsubishi Shipping Mitsubishi Corp. (0.021) 4/21 (0.016) (504: 6.84%) Mining/Asahi (0.01) N o n e /ll/ 11/15 (0.004) Kozai/Warehouse (0.008) 2/8 (0.002)/l/16 (0.01) Bank/Realty (0.004) 3/20 (0.049) /5/17 (0.009) Mitsubishi Petro. (0.005) 1/16(0.014) Source: The above data are calculated from Keizai Chosa Kyokai (1964). Note. The firms which are not parenthesized in this table are the members o f Kin’yo-kai, the Mitsubishi keiretsu’s Presidents’ Council. Those parenthesized are the members o f the other council o f the Mitsubishi keiretsu. Columns A, B, and C, respectively, contain the following information: Column A: Mitsubishi firms ranked in the order o f Mitsubishi Bank’s lending. The two numbers in the parenthesis under each firm’s name are Mitsubishi Bank’s lending to the firm, that is, the amount o f borrowing by the firm from Mitsubishi Bank (the left-hand side in the parenthesis. The unit is billion yen.) and the borrowing ratio, defined as “the borrowing from Mitsubishi Bank divided by the total borrowing” o f the firm (the right-hand side o f the parenthesis). The firms which are not parenthesized in this table are the members of Kin’yo-kai, the Mitsubishi keiretsu’s Presidents’ Council. Those parenthesized are the members o f the other council o f the Mitsubishi keiretsu. Column B: the top five companies ranked in the order o f the ratios o f the stocks held by the firm o f Column A. The number in the parenthesis denotes the ratio. Column C: the numerator o f the fraction denotes the ranking o f the firm o f Column A when measured by the ratio o f its stock held by the firm o f Column B. The denominator denotes the total number o f group firms whose stocks are held by the firm o f Colum B. N/A in Column C means that the firm o f Column B does not hold any stock issued by the firm o f Column A. The number in the parenthesis is the ratio o f the number o f the stock o f the firm A held by the firm o f Column B, relative to the total number o f the stocks issued by the firm of Column A. 83 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 16. The Ranking of the Sumitomo Firms in the Order of Sumitomo Bank’s Lending (1960) A B C 1. Sumitomo Corp. Sumitomo Realty (0.058) N/A (7,097: 39.94%) Kinzoku Kozan (0.043) 2/12 (0.034) Kinzoku Kogyo (0.031) 1/9 (0.089) Sumitomo Machinery (0.028) 4/12 (0.006) Sumitomo Coal Mining (0.023) 1/11 (0.02) 2. Nihon Denki Sumitomo Coal Mining (0.059) 1/11 (0.02) (2,782: 19.62%) Sumitomo Electric (0.043) 1/12 (0.06) Kinzoku Kozan (0.024) 3/12 (0.028) Sumitomo Corp. (0.023) 7/13 (0.016) Sumitomo Machinery (0.018) 8/12 (0.001) 3. Sumitomo Electric Nihon Denki (0.06) 2/12 (0.043) (2,360: 17.20%) Sumitomo Corp. (0.044) 10/13 (0.01) Sumitomo Realty (0.035) N/A Kinzoku Kozan (0.027) 7/12(0.013) Sumitomo Machinery (0.026) 8/12 (0.001) 4. Sumitomo Kinzoku Kozan Sumitomo Realty (0.052) N/A (1,899: 30.18%) Sumitomo Corp. (0.034) 2/13 (0.043) Nihon Denki (0.028) 4/12 (0.024) Sumitomo Machinery (0.026) 2/12 (0.008) Sumitomo Coal Mining (0.017) 5/11 (0.01) 5. Sumitomo Kinzoku Kogyo Sumitomo Corp. (0.089) 3/13 (0.031) (1,899: 5.59%) Sumitomo Machinery (0.026) 8/12 (0.001) Sumitomo Bank (0.025) 4/11 (0.051) Sumitomo Coal Mining (0.018) 8/11 (0.003) Nihon Denki (0.017) 8/12 (0.009) 6. Nippon Sheet Glass Sumitomo Realty (0.031) None/12 (1,830: 31.52%) Sumitomo Corp. (0.028) 7/13(0.016) Sumitomo Machinery (0.026) N one/12 Sumitomo Bank (0.02) 8/11 (0.04) Sumitomo Warehouse (0.008) None/12 7. Sumitomo Coal Mining Nihon Denki (0.02) 1/12 (0.059) (1,715: 19.15%) Sumitomo Machinery (0.013) 3/12 (0.007) Sumitomo Bank (0.012) 2/11 (0.058) Kinzoku Kozan (0.01) 6/12 (0.017) Sumitomo Warehouse (0.006) 6/12(0.002) Table continues next page 84 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 16 (continued). The Ranking of the Sumitomo Firms in the Order of Sumitomo Bank’s Lending (1960) A B C 8. Sumitomo Chemical Sumitomo Corp. (0.075) 11/13 (0.008) (1,340: 11.24%) Sumitomo Machinery (0.026) 8/12(0.001) Sumitomo Bank (0.025) 11/11 (0.025) Sumitomo Coal Mining (0.017) 7/11 (0.005) Kinzoku Kozan (0.012) 11/12 (0.003) 9. Sumitomo Warehouse Sumitomo Realty (0.038) N/A (385: 77.15%) Sumitomo Machinery (0.006) 1/12(0.01) Sumitomo Corp. (0.004) 11/13 (0.008) Nihon Denki (0.003) 9/12 (0.008) Electric/Coal Mining (0.002) 8/12 (0.011)/6/11 (0.006) 10. Sumitomo Machinery Sumitomo Warehouse (0.01) 3/12 (0.006) (70: 17.95%) Sumitomo Realty (0.008) N/A Sumitomo Coal Mining (0.007) 3/11 (0.013) Sumitomo Corp. (0.006) 5/13 (0.028) Sumitomo Bank (0.002) 6/11 (0.044) Source: The data are calculated from Keizai Chosa Kyokai (1964). Note. The firms in this table are the members o f Hakusui-kai, the Sumitomo keiretsu’s Presidents’ Council. Columns A, B, and C, respectively, contain the following information: Column A: Sumitomo firms ranked in the order o f Sumitomo Bank’s lending. The two numbers in the parenthesis under each firm’s name are Sumitomo Bank’s lending to the firm, that is, the amount o f borrowing by the firm from Sumitomo Bank (the left-hand side in the parenthesis. The unit is billion yen.) and the borrowing ratio, defined as “the borrowing from Sumitomo Bank divided by the total borrowing” o f the firm (the right-hand side o f the parenthesis). Column B: the top five companies ranked in the order o f the ratios o f the stocks held by the firm o f Column A. The number in the parenthesis denotes the ratio. Column C: the numerator o f the fraction denotes the ranking o f the firm o f Column A when measured by the ratio o f its stock held by the firm o f Column B. The denominator denotes the total number o f group firms whose stocks are held by the firm o f Column B. N/A in Column C means that the firm o f Column B does not hold any stock issued by the firm o f Column A. The number in the parenthesis is the ratio o f the number o f the stock o f the firm A held by the firm o f Column B, relative to the total number o f the stocks issued by the firm of Column A. These tables show that the three main banks (Mitsui Bank, Mitsubishi Bank, and Sumitomo Bank) all placed the highest lending to their group GTCs (Mitsui Corp., Mitsubishi Corp. and Sumitomo Corp.) in the year, with huge differences with their runners-up (Toyo Rayon, Shin- Mitsubishi Heavy Ind., and Nihon Denki). This fact indicates that the main banks of the Big Three strongly supported the intra-group transactions underway in their GTCs during the period. 85 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Furthermore, following the GTCs in the ranking are typically those firms engaged in the heavy and chemical industries, such as Mitsui Chemical and Toyo Koatsu (Mitsui keiretsu), 3 Heavy Inds., Mitsubishi Electric and Mitsubishi Rayon (Mitsubishi keiretsu), and Nihon Denki and Sumitomo Electric (Sumitomo keiretsu).5 7 The other factor equally important to the main banks’ lending in solidifying the intra-group transactions is cross-shareholdings among member firms, and their data are summarized in Columns B and C of Tables 14, 15, and 16. For each breakdown, Column B lists the top five companies ranked in order of the stockholdings ratio by the firm of Column A (The numbers in the parentheses denote ratios). In Column C, the state of cross-shareholdings between the firms of Columns B and C are captured by fractions. In those fractions, the numerator denotes the ranking of the Column A firm, measured by the ratio of the stockholdings by the firm of Column B. The denominator, on the other hand, denotes the total number of member firms whose stocks are held by the firm of Column B (the numbers in the parentheses are the ratios of what percentage of stock issued by the firm of Column A is owned by the firm of Column B in the same breakdown). Thus, Columns B and C roughly outline the state of cross-shareholdings carried out between the firms of Columns A and B in the same breakdown. A couple of points are to be addressed by comparing these two Columns. First, quite often, cross-shareholdings were made between Column A and B firms, although in the case of Mitsui keiretsu, they were less salient. Second, Column A firms made cross-shareholdings with their main banks and GTCs, and, again, this feature was less salient in the Mitsui keiretsu. This, in turn, suggests that the firms of Column A were often blessed with the chance to tap into loans supplied by their main banks and to utilize their GTCs to fulfill their group transaction needs. In order to see how effective each main bank assisted asset-specific transactions in 1960, a set of two regressions is conducted. The first is to regress the members’ borrowing ratios (= the borrowing from the main bank divided by the total borrowing) on the ratios of the main bank’s holding of the stock issued by the borrowing firms (= the number of the borrowing firm’s stock held by the main bank divided by the total number of stock issued by the borrowing firm). This 86 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. regression is done to verify whether the main banks likely secured more (less) voting rights through the purchase of their customer’s stocks as the customer borrowed more (less) from them. The second regression, on the other hand, is to regress the borrowing ratios, used in the first regression, on the ratios of the stockholdings by those other members for which the members in the first regression were the primal shareholders. This regression is to check the correlation between the borrowing ratios of the member firms and the likeliness of their transactions with their partners in the same Presidents’ Councils by taking into consideration the effect of cross-shareholdings whereby such mutual transactions are considered to occur more often as their cross-shareholdings are more progressed. Put together, these two regressions serve to analyze whether the Big Three’s main banks financed their member firms in the way that mediates the transactions between the cross-shared firms, while securing their voting rights in the borrowing members. To focus on the main banks’ involvement in the tangible goods transaction between their members, other financial institutions (trust and insurance companies), and other non-financial firms considered not engaging 58 in the tangible goods transactions are eliminated out of the regression. The regression results are summarized in Tables 17 (for the first regression) and 18 (for the second regression). Table 17. The Results of the First Regressions (1960) Keiretsu Constant Coefficient /?-square Mitsui 0.019 0.019 0.007 (0.702) (0.170) Mitsubishi 0.027 0.011 0.012 (2.691) (0.354) Sumitomo 0.037 0.016 0.035 (3.729) (0.424) Note. The parenthesized numbers are t values. 87 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 18. The Results of the Second Regressions (1960) Keiretsu Constant Coefficient /{-square Mitsui 0.002 -0.006 0.310 (1.680) (-1.341) Mitsubishi 0.005 0.0004 0.00006 (0.760) (0.025) Sumitomo 0.018 0.026 0.060 (1.478) (0.565) Note. The parenthesized numbers are t values. In the cases of the Mitsubishi and Sumitomo keiretsu, the positive coefficients obtained in both the regression, as expected from the TCE viewpoint. They are, however, not statistically significant. In the case of Mitsui, on the other hand, the positive coefficient obtained in the first regression, but the negative coefficient in the second regression, which is against what is expected from the TCE perspective. Besides, they are not statistically significant. For these coefficients to be statistically significant, especially for Mitsubishi and Sumitomo, the economic conditions at this earlier phase of the high economic growth might have not matured enough for the main banks to be able to fully display its function of promoting asset-specific, intra-group transactions. In this regard, Kikkawa (1992: pp. 275-7), Shibagaki (1975: p. 71-2), and Shimokawa (1975: pp. 71-2) actually pointed out that the heavy and chemical industrialization was to be further advanced by the “energy revolution” in the early 1960s. Among all, this energy revolution owed much to the liberalization of the petroleum imports that the Japanese government eventually allowed in 1961.5 9 The main banks’ financing function of supporting the intra-group, asset-specific transactions is yet to be seen until the effect of heavy and chemical industrialization is widely diffused in the economy. The Big Three’s effort to accommodate the energy revolution and to promote further the heavy and chemical industrialization was readily confirmed in the changes in the members of their Presidents’ Councils. Tablel9 shows the member firms of the Big Three in 1966. 88 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 19. Members of the Presidents’ Councils of the Mitsubishi, Sumitomo, and Mitsui Keiretsu, 1966 Sector Kin’yo-kai (Mitsubishi) Hakusui-kai (Sumitomo) Nimoku-kai (Mitsui) Bank •Mitsubishi Bank •Mitsubishi Trust •Sumitomo Bank •Sumitomo Trust •Mitsui Bank •Mitsui Trust Insurance •Meiji Life •Tokyo Marine •Sumitomo Life •Sumitomo Marine •Mitsui Life •Taisho Marine Trade •Mitsui Corp. •Sumitomo Corp. •Mitsui Corp. Agriculture and forestry (•Mitsui Norin) Coal •Mitsubishi Mining •Sumitomo Coal - Mining •Mitsui Mining • Hokkaido Tanko Kisen Construction •Sanki Kogyo (•Mitsui Construction) Foods •Kirin Beer (•Nihon Flour) Textiles •Mitsubishi Rayon •Toyo Rayon Paper and Pulp •Mitsubishi Paper Chemicals •Mitsubishi Kasei •Mitsubishi Oil •Mitsubishi Resin •Mitsubishi Monsanto Chemical •Mitsubishi Edogawa Chemical •Sumitomo Chemical •Mitsui Chemical •Toyo Koatsu •Mitsui Petrochemical Petroleum •Mitsubishi Petroleum (•General Petroleum) Glass and cement (ceramics) •Asahi Glass •Mitsubishi Cement •Nippon Sheet Glass •Sumitomo Cement Steel •Mitsubishi Steel •Sumitomo Kinzoku Kozan Kogyo •Nihon Seikojo Non-ferrous metals •Mitsubishi Metal •Sumitomo Kinzoku Kozan •Sumitomo Electric •Sumitomo Light Metal •Mitsui Metal Electrical machinery •Mitsubishi Electric •NEC (Nihon Denki) Transportation machinery •Mitsubishi Eleavy Ind. •Mitsui Shipbuilding (•Showa Airplanes) Machines •Mitsubishi Kakoki •Sumitomo Machinery (•Mitsui Seiki) (•Mitsui-Miike Seisakujo) Real estate •Mitsubishi Realty •Sumitomo Realty •Mitsui Realty Shipping •Nihon Yusen (•Osaka Shosen-Mitsui Shipping) Warehousing •Mitsubishi Warehouse •Sumitomo Warehouse •Mitsui Warehouse Source: Keizai Chosa Kyokai (1967: p. 11). Note. The parenthesized firms in the Mitsui keiretsu belong to Getsuyo-kai. 89 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. The change in the m em ber firms betw een 1960 and 1966 can be obtained by comparing Tables 8 and 19, as in the follow ing. (1) New entries and mergers occurred in the heavy and chemical industries, which indicate that the heavy and chemical industrialization progressed more between the two periods. They were made by the following firms: Mitsubishi Resin (developed from Nagahama Resin in June 1962), Mitsubishi Monsanto Chemical (developed from Monsanto Kasei in August 1958), Mitsubishi Edogawa Chemical (developed from Edogawa Chemical in June 1962), Mitsui Petrochemical (founded by the group investment in 1955), Nihon Seikojo (founded already in 1907, but placed in a peripheral position in the Mitsui keiretsu), Mitsubishi Kakoki (founded already in 1935, but placed in a peripheral position in the Mitsubishi keiretsu), and Sumitomo Light Metal (branched out from Sumitomo Metal in September 1959), “new” Mitsubishi Heavy Ind. (founded by the merger among the three Heavy Inds.), and “new” Mitsubishi Metal (founded by the merger between the “old” Mitsubishi Metal and Mitsubishi Kozai).6 0 (2) As a result of the energy revolution, scrapping coal-mining facilities became unavoidable. The cement sector then came to the fore in taking over the resources from the coal mining sector. Sumitomo Cement’s joining Hakusui-kai represents this.6 1 Also, as a result of the rationalization of the shipping industry in the early 1960s, Mitsubishi Shipping and Mitsui Shipping disappeared from the Kin’yo-kai and Nimoku-kai, respectively. Instead, Nihon Yusen joined Kin’yo-kai after its absorption of Mitsubishi Shipping. The change in the industrial landscape brought about by the energy revolution was also reflected in the change, between the two periods, of the “intra-keiretsu stockholdings ratio,” which is defined as “the number of stocks held by the Presidents’ Council members divided by the total number of stocks issued by the members.” The ratios are reported in Tables 20 (1960) and 21 (1966) for the Big Three. Comparing these two tables gives three observations: (1) The new entrants in the 90 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Presidents’ Councils ranked quite higher in terms of the ratio (Mitsui Petrochemical, Mitsubishi Resin, Mitsubishi Kakoki, Edogawa Chemical, Sumitomo Light Metal, and Sumitomo Cement). (2) Even the lowest ranked Edogawa Chemical in the Mitsubishi keiretsu swelled to 10.1% of the intra- keiretsu stockholdings. (3) The GTCs and other firms of the heavy and chemical industry are still ranked higher in terms of the ratio, in roughly the same order as in I960.6 2 Table 20. The Ratios of the Intra-Keiretsu Stockholdings (1960) Mitsui Mitsubishi Sumitomo Mitsui Realty: 16.7 Mitsubishi Cement: 77.5 Sumitomo Corp.: 37.8 (6.6) (75.8) (32.3) Mitsui Shipping: 9.2 Mitsubishi Oil: 58.2 Sumitomo Machinery: 24.0 (7.0) (50.7) (19.6) Mitsui Corp.: 8.3 Mitsubishi Corp.: 19.2 Nihon Denki: 23.5 (1.1) (14.0) (15.8) Toyo Koatsu: 4.6 Mitsubishi Shipping: 17.7 Sumitomo Realty: 22.0 (1.8) (17.7) (22.0) Mitsui Warehouse: 4.6 Mitsubishi Realty: 16.0 Sumitomo Coal-Mining: 20.3 (1.6) (11.5) (14.5) Mitsui Chemical: 3.2 Mitsubishi Mining: 13.6 Sumitomo Kinzoku Kozan: 15.6 (1.3) (11.5) (12.5) Mitsui Mining: 2.9 Mitsubishi Warehouse 12.9 Sumitomo Warehouse: 13.2 (2.8) (9.1) (8.2) Mitsubishi Steel: 11.8 Sumitomo Electric: 12.1 (9.3) (8.3) Mitsubishi Rayon: 10.5 Sumitomo Kinzoku Kogyo: 11.1 (6.1) (6) Mitsubishi Kasei: 9.0 (5.6) Source: The ratios are calculated from the data in Keizai Chosa Kyokai (1961). Note. The ratio o f the intra-group keiretsu stockholdings is defined as “the number o f stocks held within the keiretsu concerned divided by the total number o f stocks issued.” The unit is percent. The numbers in parentheses are the ratios when the main banks’ partaking in the stockholdings is excluded. R eproduced with perm ission o f the copyright owner. Further reproduction prohibited without perm ission. Table 21. The Ratios of the Intra-Keiretsu Stockholdings (1966) Mitsui Mitsubishi Sumitomo Mitsui Petrochemical 55.0 Mitsubishi Cement: 78.2 Sumitomo Light Metal: 45.9 (46.2) (76.5) (40.1) Mitsui Realty: 15.2 Mitsubishi Oil: 48.8 Sumitomo Corp.: 34.1 (6.1) (40.7) (26.1) Mitsui Corp.: 8.7 Mitsubishi Resin: 45.8 Sumitomo Realty: 30.0 (1.6) (44.1) (21.7) Nihon Seikojo: 7.6 Mitsubishi Corp.: 17.8 Sumitomo Cement: 28.7 (1.2) (12.6) (21.7) Mitsui Chemical: 5.1 Mitsubishi Kakoki: 15.6 Sumitomo Coal-Mining: (3.2) (11.4) 25.5 Toyo Koatsu: 4.8 Mitsubishi Mining: (9.6) (19.7) (1.9) Mitsubishi Rayon: (6.8) Sumitomo Machinery: 23.2 Mitsui Mining: 4.5 Mitsubishi Steel: 11 (18.9) (2.8) (6.5) Nihon Denki: 21.9 Sanki Kogyo: 4.4 Mitsubishi Realty: 10.9 (13.2) (1.1) (6.9) Sumitomo Kinzoku Kozan: Mitsui Warehouse: 3.9 Edogawa Chemical: 10.1 16.5 (6.1) (4) (13.2) Sumitomo Electric: 12.1 (8.4) Sumitomo Warehouse: 11.1 (6.1) Sumitomo Kinzoku Kogyo: 10.0 (4.3) Source: The ratios are calculated from the data in Keizai Chosa Kyokai (1961). Note. The ratio o f the intra-group keiretsu stockholdings is defined as “the number o f stocks held within the keiretsu concerned divided by the total number o f stocks issued.” The unit is %. The numbers in parentheses are the ratios when the main banks’ partaking in the stockholdings is excluded. The main banks’ lending patterns of 1966 are summarized in Tables 22, 23, and 24 in exactly the same fashion as in Tables 14, 15, and 16. 92 R eproduced with perm ission o f the copyright owner. Further reproduction prohibited without perm ission. Table 22. The Ranking of the Mitsui Firms in the Order of Mitsui Bank’s Lending (1966) A B C 1. Mitsui Corp. Mitsui Bank (0.206) 4/20 (0.071) (31,747: 17.4%) Mitsui Seiki (0.15) N/A Mitsui Petrochemical (0.075) None/2 Mitsui Norin (0.069) N/A Miike Sesakujo/Construction N /A /N one/7 (0.025) 2. (Toyo Cotton) Mitsui Bank (0.098) 9/20 (0.071) (9,336: 12.6%) Nihon Flour (0.011) 4/5 (0.007) Mitsui Warehouse (0.009) 5/12 (0.007) 3. Mitsui Petrochemical Mitsui-Miike Seisakujo (0.01) N/A (7,152: 15.4%) 4. Mitsui Realty Mitsui Construction (0.267) 2/7 (0.008) (5,684: 17.2%) Mitsui Seiki (0.133) N/A Mits Bank (0.132) 2/20 (0.097) Mitsu Norin (0.127) N/A N Ian Seikojo (0.012) 5/5 (0.002) 5. Mitsui Chemical Mitsui Bank (0.145) 14/20 (0.019) (5,130: 16.1%) Mitsui-Miike Seisakujo (0.015) N/A Mitsui Petrochemical (0.012) N/A Mitsu Realty (0.005) 10/18 (0.007) Mitsui Mining (0.004) None/3 6. Nihon Seikojo Mitsui Bank (0.1) 5/20 (0.064) (4,576: 22.9%) Mitsui Seiki (0.05) N/A Mitsui-Miike Seisakujo (0.01) N/A Mitsui Realty (0.002) 7/18(0.012) 7. Mitsui Mining Mitsui-Miike Seisakujo (0.739) N/A (4,498: 8.6%) Mitsui Construction (0.202) 4/7 (0.003) Mitsui Petrochemical (0.004) None/2 Table continues next page 93 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 22 (continued). The Ranking of the Mitsui Firms in the Order of Mitsui Bank’s Lending (1966) A B C 8. Mitsui Shipbuilding Mitsui Bank (0.104) None/20 (4,301: 10.6%) Mitsui Seiki (0.05) N/A Mitsui Petrochemical (0.022) None/2 Mitsui Realty (0.007) 13/18 (0.002) Mitsui Corp. (0.002) 10/17(0.017) 9. Toyo Rayon Mitsui Petrochemical (0.23) None/2 (4,283: 9.3%) Mitsui Bank (0.079) None/20 Mitsui Chemical (0.009) 11/13 (0.0004) Mitsui Corp. (0.005) 16/17 (0.002) 10. Toyo Koatsu Mitsui Bank (0.145) 12/20 (0.028) (3,196: 8.3%) Mitsui-Miike Seisakujo (0.015) N/A Mitsui Petrochemical (0.012) None/2 Mitsui Realty (0.005) 13/18(0.002) Mitsui Mining (0.004) None/3 11. Mitsui Construction Mitsui-Miike Seisakujo (0.01) N/A (2,445: 30.6%) Mitsui Realty (0.008) 1/18 (0.267) Mitsui Bank (0.006) 18/20 (0.008) Mitsui Mining (0.003) 2/3 (0.202) Kinzoku Kogyo/Koatsu (0.001) 5/10 (0.014)/4/8 (0.013) 12. Hokkaido Tanko Kisen Mitsui Bank (0.111) 17/20 (0.01) (1,324: 6.1%) Mitsui Norin (0.04) N/A Mitsui Realty (0.007) None/18 13. Mitsui Metal Mitsui Bank (0.065) 13/20 (0.02) (1,088: 7.1%) Mitsui Petrochemical (0.022) None/2 Mitsui-Miike Seisakujo (0.015) N/A Mitsui Construction (0.014) 5/7 (0.001) Mitsui Mining (0.013) None/3 14. (Nihon Flour) Mitsui Bank (0.111) 7/20(0.051) (1,040:21.8%) Toyo Cotton (0.007) 3/5(0.011) Mitsui Corp. (0.003) 8/17(0.022) ....... ..................... Table continues next page 94 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 22 (continued). The Ranking of the Mitsui Firms in the Order of Mitsui Bank’s Lending (1966) A B C 15. Mitsui Warehouse Mitsui Bank (0.034) 11/20 (0.03) (963: 36.0%) Mitsui Norin (0.018) Toshoku (0.014) Toyo Cotton (0.007) General Sekiyu (0.006) N/A 5/5 (0.002) 4/5 (0.009) 3/8 (0.01) 16. Osaka Shosen-Mitsui Shipping (615: 1.0%) N/A 17. Sanki Kogyo (378: 14.3%) Mitsui Bank (0.029) Construction/Realty (0.005) Mitsui Corp. (0.003) Mitsui Warehouse (0.002) 9/20 (0.033) None/7 / None/18 12/17 (0.01) 9/12 (0.001) Source: Calculated and compiled from Keizai ChOsa Kyokai (1967). Note. The parenthesized firms belong to Getsuyo-kai. The rest belong to the Nimoku-kai. Columns A, B, and C, respectively, contain the following information: Column A: Mitsui firms ranked in the order o f Mitsui Bank’s lending. The two numbers in the parenthesis under each firm’s name are Mitsui Bank’s lending to the firm, that is, the amount o f borrowing by the firm from Mitsui Bank (the left-hand side in the parenthesis. The unit is billion yen.) and the borrowing ratio, defined as “the borrowing from Mitsui Bank divided by the total borrowing,” o f the firm (the right-hand side of the parenthesis). The parenthesized firms are the members o f Getsuyo-kai, and those not parenthesized are the members o f Itsuka-kai, both being the antecedents o f Nimoku-kai. Column B: the top five companies ranked in the order o f the ratios o f the stocks held by the firm o f Column A. The number in the parenthesis denotes the ratio. Column C: the numerator o f the fraction denotes the ranking o f the firm o f Column A when measured by the ratio o f its stock held by the firm o f Column B. The denominator denotes the total number o f group firms whose stocks are held by the firm o f Colomn B. N/A in Column C means that the firm o f Column B does not hold any stock issued by the firm o f Column A. The number in the parenthesis is the ratio o f the number o f the stock o f the firm A held by the firm o f Column B, relative to the total number o f the stocks issued by the firm of Column A. 95 R eproduced with perm ission o f the copyright owner. Further reproduction prohibited without perm ission. Table 23. The Ranking of the Mitsubishi Firms in the Order of Mitsubishi Bank’s Lending (1966) A B C 1. Mitsubishi Heavy Ind. Mitsubishi Steel (0.051) 3/6 (0.001) (33,475: 11.3%) Mitsubishi Kakoki (0.038) 5/7(0.0001) Mitsubishi Corp./Bank (0.034) 11/20 (0.006)/12/21 (0.031) Mitsubishi Cement (0.024) N/A Nihon Yusen (0.021) 7/8 (0.0004) 2. Mitsubishi Corp. Mitsubishi Oil (0.054) N/A (32,652: 24.8%) Mitsubishi Kasei (0.05) 9/15 (0.008) Mitsubishi Kakoki (0.042) 2/7 (0.001) Mitsubishi Paper (0.027) 3/7(0.004) Mitsubishi Warehouse (0.02) 1/13 (0.008) 3. Mitsubishi Electric Mitsubishi Corp. (0.016) 12/20 (0.005) (12,880: 15.7%) Mitsubishi Bank/Mining (0.001) 15/21 (0.017)/9/12 (0.001) Mitsubishi Kakoki (0.013) None/7 Mitsubishi Realty (0.006) 6/10 (0.003) Mitsubishi Kasei (0.003) 12/15 (0.002) 4. Mitsubishi Realty Mitsubishi Cement (0.016) N/A (6,563: 12.5%) Mitsubishi Mining (0.014) 6/11 (0.002) Mitsubishi Corp. (0.007) 11/20 (0.006) Mitsubishi Bank/Kasei (0.005) 8/21 (0.04) 19115 (0.008) Asahi Glass (0.003) 4/12(0.028) 5. Mitsubishi Kasei Mitsubishi Resin (0.414) 0 (5,958: 9.7%) Mitsubishi Oil (0.109) N/A Mitsubishi Cement (0.081) N/A Mitsubishi Mining (0.048) 2/12 (0.022) Mitsubishi Kakoki (0.022) 3/7 (0.0005) 6. Mitsubishi Rayon Mitsubishi Oil (0.108) N/A (4,380: 12.9%) Mitsubishi Bank/Cement (0.007) 6/21 (0.044)/N/A Mitsubishi Kasei/Corp. (0.005) 6/15 (0.021)/8/20 (0.014) Kinzoku Mining/Heavy ind. (0.002) N o n e/1 2 /7/11 (0.017) Asahi Glass (0.001) 10/12 (0.005) 7. Mitsubishi Steel Mitsubishi Bank (0.002) 5/21 (0.045) (3,857: 28.8%) Mitsubishi Heavy ind. (0.001) 1/11 (0.051) Nihon Yusen (0.0005) None/8 Mitsubishi Corp. /Realty (0.0003) 8/20 (0.014)/None/10 Table continues next page 96 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 23 (continued). The Ranking of the Mitsubishi Firms in the Order of Mitsubishi Bank’s Lending (1966) A B C 8. Mitsubishi Petroleum (3,230: 12.0%) Mitsubishi Corp. (0.006) Mitsubishi Cement (0.002) Mitsubishi Banlc/Yusen (0.001) Mitsubishi Realty (0.0008) 8/20 (0.014) N/A 17/21 (0.013)/4/8 (0.005) None/10 9. Kirin Beer (3,200: 22.8%) Mitsubishi Bank (0.006) Mitsubishi Corp. (0.003) None/21 14/20 (0.002) 10. Asahi Glass (2,818: 16.3%) Mitsubishi Oil (0.11) Mitsubishi Cement (0.081) Mitsubishi Warehouse (0.037) Mitsubishi Realty (0.028) Edogawa Chemical (0.017) N/A N/A..................................................... 3/13 (0.003) 6/10 (0.003) 3/10 (0.002) 11. Mitsubishi Metal (2,666: 12.6%) Mitsubishi Oil (0.056) Mitsubishi Cement (0.016) Mitsubishi Mining (0.009) N/A N/A 7/12(0.005) Mitsubishi Bank (0.005) 14/21 (0.019) 12. Nippon Yusen (2,506: 3.3%) 13. Mitsubishi Edogawa Rayon/Corp. (0.003) Mitsubishi Corp. (0.015) Mitsubishi Mining (0.01) Mitsubishi Bank (0.007) Mitsubishi Sekiyu (0.005) Mitsubishi Realty (0.003) Mitsubishi Bank (0.003) 5/10(0.002)/ 10/20 (0.008) 6/20 (0.02) 8/12 (0.002) 16/21 (0.015) 4/6 (0.001) None/10 2/21 (0.062) Chemical (2,038: 23.8%) Kasei/Resin/Glass/Paper (0.002) 7/15(.017)/l/4(.002)/5/12(.017)/ 3/7(.004) Mitsubishi Electric (0.001) None/8 Mitsubishi Corp. /Realty (0.0003) N one/20/N one 10 14. Mitsubishi Mining (2,003: 7.4%) Mitsubishi Cement (0.52) Mitsubishi Kasei (0.022) Mitsubishi Oil (0.02) Mitsubishi Corp. (0.007) Mitsubishi Realty (0.006) N/A 4/15 (0.048) N/A 8/20 (0.014) 2/10 (0.014) Table continues next page R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 23 (continued). The Ranking of the Mitsubishi Firms in the Order of Mitsubishi Bank’s Lending (1966) B 15. Mitsubishi Paper (1,435: 10.0%) Mitsubishi Bank/Rayon (0.006) Edogawa Chemical/Corp. (0.004) Mitsubishi Heavy Ind. (0.0008) 2/21 (0.062)/5/10 (0.002) 3/10 (0.002)/5/20 (0.027) N one/11 16. Mitsubishi Warehouse (1,265: 34.1%) 17. Mitsubishi Resin (1,206: 14.6%) Mitsubishi Corp. (0.008) Mitsubishi Cement (0.004) Bank/Glass/Realty (0.003) Nihon Yusen (0.002) Electric/Kasei (0.001) Edogawa Chemical (0.002) Mitsubishi Bank (0.001) Mitsubishi Heavy ind. (0.0001) 6/20 (0 .02) N/A 9/21(.038)/3/12(.037)/None/10 None/8 None/8 / None/15 3/10 (0.002) 15721 (0.017) 8/11 (0 .01 ) 18. Mitsubishi Kakoki (673: 35.9%) Bank/Corp. /Mining (0.001) Mitsubishi Kasei (0.0005) Kinzoku Mining (0.0004) Mitsubishi Heavy ind. (0.0001) 7/21(,042)/3/20(.042)/None/11 5/15 (0.022) None/11 2/11 (0.038) Source: Calculated and compiled from Keizai Chosa Kyokai (1967). Note. Columns A, B, and C, respectively, contain the following information: Column A: Mitubishi firms ranked in the order o f Mitsubishi Bank’s lending. The two numbers in the parenthesis under each firm’s name are Mitsubishi Bank’s lending to the firm, that is, the amount o f borrowing by the firm from Mitsubishi Bank (the left-hand side in the parenthesis. The unit is billion yen.) and the borrowing ratio, defined as “the borrowing from Mitsubishi Bank divided by the total borrowing,” o f the firm (the right-hand side o f the parenthesis). Column B: the top five companies ranked in the order o f the ratios o f the stocks held by the firm o f Column A. The number in the parenthesis denotes the ratio. Column C: the numerator o f the fraction denotes the ranking o f the firm o f Column A when measured by the ratio o f its stock held by the firm o f Column B. The denominator denotes the total number o f group firms whose stocks are held by the firm o f Column B. N/A in Column C means that the firm o f Column B does not hold any stock issued by the firm o f Column A. The number in the parenthesis is the ratio o f the number o f the stock o f the firm A held by the firm o f Column B, relative to the total number o f the stocks issued by the firm of Column A. 98 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 24. The Ranking of the Sumitomo Firms in the Order of Sumitomo Bank’s Lending (1966) A B C 1. Sumitomo Corp. Realty/Kinzoku Kozan (0.046) N/A /2/14 (0.034) (17,961:25.3%) Sumitomo Coal-Mining (0.032) 2/14(0.02) Sumitomo Cement (0.03) None/4 Sumitomo Machinery (0.028) 6/14 (0.007) Sumitomo Bank (0.018) 2/14(0.081) 2. Nihon Denki Sumitomo Coal-mining (0.059) i 5/14(0.012) (10,775: 1.9%) Sumitomo Electric (0.043) 1/11 (0.052) Kinzoku Kozan (0.032) ... 1 ..5/14.(0.02)............ Sumitomo Realty (0.025) ... ! ..N/A Sumitomo Bank (0.023) | 1/14(0.087) 3. Sumitomo Chemical Sumitomo Corp. (0.072) 10/15 (0.009) (9,609: 14.6%) Sumitomo Bank (0.03) | 13/14 (0.037) Sumitomo Machinery (0.028) 8/14(0.004) Sumitomo Coal-mining (0.026) 8/14(0.005) Sumitomo Cement (0.015) | None/4 4. Sumitomo Electric Nihon Denki (0.052) 2/13 (0.043) (6,799: 26.8%) Sumitomo Corp. (0.032) 1 11/15 (0.008) Sumitomo Realty (0.027) N/A Sumitomo Machinery (0.021) 10/14(0.002) Kinzoku Kozan (0.017) 6/14(0.013) 5. Sumitomo Kinzoku Kogyo Sumitomo Light Metal (0.369) | None/5 (6,421: 8.3%) Sumitomo Corp. (0.055) 10/15 (0.009) Sumitomo Coal-mining (0.042) 9/14(0.004) Sumitomo Machinery (0.025) 9/14(0.003) Sumitomo Bank (0.023) 1 7/14 (0.057) 6. Nihon Sheet Glass Sumitomo Corp. (0.028) 6/15 (0.016) (5,155: 32.4%) Machinery/Realty (0.025) 11/14 (0.001)/N /A Bank/Kinzoku Kogyo (0.02) 11/14 (0.04)/None/10 Sumitomo Cement (0.009) None/4 Sumitomo Warehouse (0.008) None/14 7. Sumitomo Cement Sumitomo Realty (0.006) N/A (3,590: 17.5%) Sumitomo Bank (0.005) j 4/14 (0.07) Kinzoku Kogyo (0.0003) 9/10(0.015) Table continues next page 99 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 24 (continued). The Ranking of the Sumitomo Firms in the Order of Sumitomo Bank’s Lending (1966) A B C 8. Sumitomo Kinzoku Kozan Sumitomo Realty (0.041) N/A (2,710: 22.0%) Sumitomo Corp. (0.034) 1/15 (0.046) Sumitomo Coal-mining (0.021) 8/15 (0.033) Machinery/Nihon Denki (0.02) 5/14 (0.008)/4/13 (0.032) Sumitomo Electric (0.013) 6/11 (0.017) 9. Sumitomo Machinery Sumitomo Cement (0.039) None/4 (2,630: 26.7%) Sumitomo Coal-mining (0.016) 3/14(0.019) Sumitomo Warehouse (0.014) 3/14(0.006) Kinzoku Kozan (0.008) 5/14 (0.02) Sumitomo Corp. (0.007) 4/15 (0.028) 10. Sumitomo Light Metal Sumitomo Corp. (0.01) 9/15(0.01) (1,790: 11.9%) Bank/Kinzoku Kozan (0.003) 5/14 (0.059)/9/14 (0.004) Sumitomo Chemical (0.0005) 9/12 (0.007) 11. Sumitomo Coal-Mining Sumitomo Cement (0.099) None/4 (1,645: 6.2%) Sumitomo Corp. (0.02) 2/15 (0.032) Sumitomo Machinery (0.019) 2/14(0.016) Sumitomo Realty (0.017) N/A Bank/Kozan/Denki (0.012) 6/!4(.058)/4/14(.021)/l/13(.059) 12. Sumitomo Warehouse Sumitomo Realty (0.03) N/A (985: 34.5%) Bank/Kinzoku Kogyo (0.02) 9/14 (0.05)/7/10 (0.017) Sumitomo Corp. (0.004) 11/15 (0.008) Nihon Denki (0.003) None/13 Coal-mining/Electric (0.002) 7/14(0.006) /N one/ll Source: Calculated and compiled from Keizai Chosa Kyokai (1967). Note. Columns A, B, and C, respectively, contain the following information: Column A: Sumitomo firms ranked in the order o f Sumitomo Bank’s lending. The two numbers in the parenthesis under each firm’s name are Sumitomo Bank’s lending to the firm, that is, the amount o f borrowing by the firm from Sumitomo Bank (the left-hand side in the parenthesis. The unit is billion yen.) and the borrowing ratio, defined as “the borrowing from Sumitomo Bank divided by the total borrowing” o f the firm (the right-hand side o f the parenthesis). Column B: the top five companies ranked in the order o f the ratios o f the stocks held by the firm o f Column A. The number in the parenthesis denotes the ratio. Column C: the numerator o f the fraction denotes the ranking o f the firm o f Column A when measured by the ratio o f its stock held by the firm o f Column B. The denominator denotes the total number o f group firms whose stocks are held by the firm o f Column B. N/A in Column C means that the firm o f Column B does not hold any stock issued by the firm of Column A. The number in the parenthesis is the ratio o f the number o f the stock o f the firm A held by the firm o f Column B, relative to the total number o f the stocks issued by the firm o f Column A. 100 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. According to these tables, (1) Mitsui Corp. and Sumitomo Corp. still maintained the highest priority in their main banks’ lending, while Mitsubishi Corp. was replaced by “new” Mitsubishi Heavy Ind. by a slight margin in the highest priority position chosen by Mitsubishi Bank. (2) There came newly established firms in the higher rankings, such as Mitsui Petrochemical, Mitsubishi Resin, Sumitomo Cement, and Sumitomo Light Metal. The concern here is to see, once again, how effective each main bank of the Big Three financed their members in a way that asset- specific transactions were encouraged with their group affiliates. In order to examine this, a set of two regressions was conducted under the same methodology as previously.6 3 The results are reported in Tables 25 and 26. Table 25. The Results of the First Regressions (1966) Keiretsu Constant Coefficient /{-square Mitsui -0.019 0.357 0.546 (-0.970) (2.685) Mitsubishi 0.021 0.089 0.197 (2.213) (1.716) Sumitomo 0.045 0.008 0.0008 (1.495) (0.066) Note. The parenthesized numbers are t values. Table 26. The Results of the Second Regressions (1966) Keiretsu Constant Coefficient /{-square Mitsui 0.007 -0.013 0.015 (1.043) (-0.299) Mitsubishi -0.012 0.146 0.624 (-2.016) (4.458) Sumitomo 0.011 0.025 0.012 (0.435) (0.247) Note. The parenthesized numbers are t values. 101 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. As in 1960, the effect of Mitsui Bank’s financing in directing the intra-Mitsui transactions seems dubious because of the negative coefficient in the second regression. Its coefficients in both the first and second regressions are not statistically significant, either. As for Sumitomo Bank, the effect is somewhat vague. The signs of the coefficients in the first and second regressions are both positive, as is expected from the TCE point of view, but the coefficients are not statistically significant. No doubt, the most impressive is the case of the Mitsubishi keiretsu, where the expected signs obtained for the coefficients of both the first and second regressions, with both being statistically significant. Together with the result of 1960, the conclusion can be drawn for Mitsubishi as follows: as the heavy and chemical industrialization further progressed with the energy revolution, Mitsubishi firms, most of them being positioned across the wide range of the heavy and chemical sectors, increased their intra-group transactions and, in parallel, their cross-shareholdings. Their intra-group transactions were executed effectively by the finance provided by Mitsubishi Bank, which itself was motivated to do so all the more for its increased holdings of the stocks issued from those non-financial members in the course of the energy revolution. Because asset-specific transactions are more likely to occur when the effect of the heavy and chemical industrialization is widely diffused among affiliate firms, this conclusion also conforms with the well-accepted account that the Mitsubishi keiretsu is best characterized by its superiority in the heavy and chemical industry (e.g., see Miyashita and Russell [1994: p. 90] and Okumura [1976: p. 176]).6 4 Severed entirely as they seemed in the zaibatsu dissolution process, the inter-corporate ties of the Big Three (Mitsui, Mitsubishi, Sumitomo) revived, and even consolidated, through their technological and transaction relations in the course of the postwar heavy and chemical industrialization during the early 1950s through the late 1960s. Prohibited from rallying around under a holding company by the Antimonopoly Law, a group of member firms was associated through cross-shareholdings, with each holding fractional amounts of affiliates’ stocks, yet with a whole group securing portions large enough to exercise voting rights to the members. To do so was considered beneficial to the entire group, as it served to circumvent hostile-takeovers and to smooth 102 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. out intra-group transactions. Not directly engaging in production, yet playing the pivotal roles of adjusting, and even improving, the intra-group transactions were the group’s main bank and GTC. The main bank actively supported the group’s heavy and chemical industrialization through financing those members positioned in the areas. Besides, in its highest priority of lending, the main bank enormously financed the group GTC to facilitate the intra-group transactions. While receiving abundant financing from the main bank, the GTC took the role of intermediating transactions among members, in a way that minimized the transaction costs involved. The GTC furthermore extended credit to member firms to lubricate their transactions even more. Characterized as the intermeshed transaction networks and cross-shareholdings as thus described, a set of group firms, dubbed “keiretsu” vis-a-vis zaibatsu, is regarded as a meta-firm, in which the production of goods, varying from soup noodles to a rocket, are expedited in a seamless, transaction-cost minimizing fashion. While sharing this feature, the keiretsu grouping across the notable Big Three keiretsu differed in maturity, which, in turn, can be traced back to their industrial development nurtured during the zaibatsu era. Having focused more on the international trade, commerce, and finance during prewar period, the Mitsui keiretsu was significantly delayed in the keiretsu formation in the postwar period. The most successful was the Mitsubishi keiretsu, where the transaction networks were already highly developed by the group firms, especially in the area of the heavy and chemical industries, in the zaibatsu era. 2.4.4 The Formation o f the “Bank Groups” and Their Motley Nature After 1966, three additional new keiretsu groups emerged on the Japanese economy scene: the Fuyo Group, the Sanwa Group, and the DKB Group. They respectively announced to launch their Presidents’ Councils: Fuyo-kai in January 1966, Sansui-kai in February 1967, and Sankin-kai in January 1978. Crowned with the title “keiretsu,” these three groups shared the main traits with their forerunners, the Big Three: (1) retaining member firms in the Presidents’ Council, which were 103 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. selected from a wide range of industrial sectors, including the main bank and the GTC and (2) holding council meetings on a regular basis. The formations of these three new groups were largely owed to their main banks’ efforts (Fuji Bank, Sanwa Bank, Dai-Ichi Bank, and Nippon Kangyo Bank) to consolidate the financing relations with their customer firms, which, as in the case of the Big Three, were mostly positioned across various sectors of the heavy and chemical industry.6 5 Due to this bank-led group formation, these three groups were often referred to as the “Bank Groups,” as opposed to the “Former Zaibatsu Groups” (the Big Three). In selecting customer firms, the three main banks emphasized on their former zaibatsu lineages that dated back to World War I when smaller zaibatsu sprouted en masse, as overviewed in Section 2.3. More specifically, the Bank Groups’ major lineups were made of the firms that formally belonged to the following small zaibatsu groups: (1) The Fuyo Group: Yasuda, Asano, Nezu, Okura, Mori, and Nissan. (2) The Sanwa Group: Suzuki Shoten, Iwai Sangyo, and Hitachi. (3) The DKB Group: Furukawa and Kawasaki. The details of the member firms, their industrial positions, and their zaibatsu origins are summarized in Table 27.6 6 ’6 7 Table 27. Members of the Presidents’ Councils of the Bank Groups (the Fuyo, Sanwa, and DKB Groups), 1979 Category Fuyo-kai (Fuyo) Sansui-kai (Sanwa) Sankin-kai (DKB) Bank (l)Fuji Bank (l)Yasuda Tmst Bank Sanwa Bank Toyo Trust Bank (9) (*)Dai-Ichi Kangyo Bank Insurance (l)Yasuda Life (l)Yasuda F&M Nippon Life (9) Asahi Life (*) Fukoku Life (6) (*) Nissan F&M Taisei F&M Table continues next page 104 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 27 (continued). Members of the Presidents’ Councils of the Bank Groups (the Fuyo, Sanwa, and DKB Groups), 1979 Category Fuyo-kai (Fuyo) Sansui-kai (Sanwa) Sankin-kai (DKB) Trade & commerce Marubeni Nichimen (7) (8) Nissho-Iwai Iwatani Takashimaya Itochu (*) Kanematu-Gosho (7) (8) Nissho-Iwai (10) Kawatetsu Corp. (*) Seibu Department Store Agriculture and forestry Coal Construction (4) Taisei Construction Ohbayashi (1) Toyo Construction Sekisui House Shimizu Construction Foods (3) Nisshin Flour Milling (3) Sapporo Beer (6) Nichirei Ito Ham Textiles (3) Nisshin Spinning (l)Toho Rayon Unitica (7) Teijin Asahi Chemical Paper and pulp Sanyo Kokusaku Pulp (*) Honshu Paper Chemicals (5) Showa Denko (6) Nippon Oil & Fat Kureha Chemical (8) Tokuyama Soda Sekisui Chemical Ube Industries (6) Hitachi Chemical Tanabe Pharmacy Fujisawa Pharmacy (8) Kansai Paint (*) Denki Kagaku (9) Nihon Zeon (9) Asahi Denka (*) Sankyo (*) Shiseido Lion Petroleum Toa Nenryo Maruzen Oil Showa Shell Oil Glass and cement (Ceramics) (2) Nihon Cement Osaka Cement Chichibu Cement Steel (2) Nihon Kokan (7) Kobe Steel Nisshin Steel Nakayama Steel (6) Hitachi Metal (10) Kawasaki Steel Kobe Steel Japan Metals & Chemicals Table continues next page 105 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 27 (continued). Members of the Presidents’ Councils of the Bank Groups (the Fuyo, Sanwa, and DKB Groups), 1979 Category Fuyo-kai (Fuyo) Sansui-kai (Sanwa) Sankin-kai (DKB) Non-ferrous metals (6) Hitachi Cable (9) Nippon Light Metal (9) Furukawa Mining (9) Furukawa Electric Electrical machinery (6) Hitachi Co. (1) Oki Electric Yokogawa Electric (6) Hitachi Co. Iwatsu Electric Sharp (6) Hitachi Co. (6) Fuji Electric Yasukawa Electric (9) Fujitsu (*) Nippon Columbia Transportation machinery (6) Nissan Motors (6) Hitachi Shipbuilding (6) Shin Meiwa Daihatsu (10) Kawasaki Heavy Ind. Ishikawajima-Harima Heavy Ind. Isuzu Motors General & precision machinery Real estate Cannon Kubota Ironwork (1) (3) Nihon Seiko (1) Tokyo Tatemono NTN Toyo Bearings Asahi Optical (*) Niigata Ironworks Ebara Factory Shipping Warehousing (3) Tobu Railway Keihin Kyuko Railway (1) Showa Marine Transportation Hankyu Railway Nippon Express Yamashita Shin Steamship (*) Nippon Express (10) Kawasaki Steamship (9) Shibusawa Warehouse Other industries Toyo Rubber Orient Lease (9) Yokohama Rubber (*) Korakuen Stadium (*) Nihon Kangyo Kakumaru Securities Sources: T5yo Keizai Shinposha, Kigyd Keiretsu Soran (1980: p. 38), Okumura (1976: pp. 207-43), T0y5 Keizai Shinposha (1994c: pp. 64-81). Note. The numbered firms are o f the former zaibatsu lineage. The numbers indicate which one or more o f the following zaibatsu these firms used to belong to: (l)Yasuda, (2) Asano, (3) Nezu, (4) Okura, (5) Mori, (6) Hitachi-Nissan (Kuhara, Ayukawa), (7) Suzuki Shoten, (8) Iwai, (9) Furukawa, (10) Kawasaki. Those belonging to “Kangin Jugosha-kai” are marked (*). As briefly pointed out above, the emergence of the Bank Groups mainly ascribed to the heavy and chemical industrialization of the Japanese economy, which, albeit in somewhat 106 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. decelerating degrees in the late 1960s, consistently gave these banks the incentive to lump their customer firms into whole, integrated groups for better business opportunities and to restrain fund outflow from their group affiliations (Dai-Ichi Kangyo Ginko 1992: p. 119; Okumura 1976: p. 110; Suzuta 1976: p. 136). Their move toward group formations was clearly expressed in their adaptation of policy slogans as the “mainstream economic transaction” (keizai shuryu torihiki) (Fuji Bank) and the “structural transaction” (kozd torihiki) (Sanwa Bank) (Suzuki 1993: p. 136). Probably, the best example for showcasing the grouping process is the case of Sanwa Bank. Sanwa Bank, traditionally known for its financing in the textile-related industries, came to its turning point when it arranged the affiliation between two of its customer firms, Ube Industries and Nippon Rayon, when the latter was searching for a transaction partner with the capability of supplying caprolactam for its nylon production. The arrangement turned out to be successful, as these two firms came to terms on affiliation in May 1954 (Sanwa Bank 1974: pp. 407-9).6 8 This successful merger directed Sanwa to more seriously grapple with the then impending economic issue that “grouping of the customer firms cannot be avoidable if a bank promotes the heavy and chemical industrialization among them” (Sanwa Bank 1974: p. 409). Sanwa, subsequently, embarked on a joint project on the petrochemical products with its customer, Maruzen Oil (Okumura 1976: p. 235; Sanwa Bank 1974: pp. 417-21).6 9 The purpose was not only to secure more customers in the area, but also to “rearrange its financing pattern in the most suitable manner with the heavy and chemical industrialization underway” (Sanwa Bank 1974: p. 417). Sanwa’s plan for the corporate grouping or “keiretsu-ization,” through the financing of Maruzen Oil, was expressed concisely by President Watanabe’s speech at the 25th anniversary of Sanwa Bank (December 1958): “Our major customers are being connected mutually, especially under the significant influence of Maruzen Oil. Our bank is currently paying every effort in promoting this” (Sanwa Bank 1974: p. 417). With more solid ground, the head of the business department, Hara, made the following speech in the same commemoration, stressing the importance of Sanwa’s grouping policy: 107 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. The petrochemical industry generates innumerable derivatives, which, in turn, commands the proportionate demands from a wide range of industrial sectors. Also, the industry often needs enormous funds for plant investment. It is virtually impossible for a single firm to accomplish all these on its own. It is more appropriate that a number of firms in the industry cooperate with each other for achieving the goal. In fact, the former zaibatsu groups are promoting the petrochemical industrialization by various affiliation policies. Our bank should adopt a similar policy, too, by mobilizing Maruzen Oil and others in order to promote the petrochemical industrialization. Such a policy will give us a chance to enter the chemical industry, thereby fortifying our earning ability enough to beat other mega-banks. If we fail to do this, our group ties will eventually be unfastened, and our customers will, consequently, be controlled by other groups in the supply of raw materials. (Sanwa Bank 1974: pp. 417-8)7 0 While consolidating group transactions centripetal to Maruzen Oil, Sanwa sequentially set up the “councilors’ office” (san’ yo-shitsu, July 1960), the ‘unified’ “screening section” (shinsa-bu, May 1961), and the “enterprise section” (kikaku-bu, April 1966). Ultimately, in February 1967, “Sansui- kai” was formed, with 22 member firms and Sanwa Bank. Concurrently, Sanwa lent impetus for the merger of its two affiliate GTCs, “Nissho” and “Iwai Sangyo,” which led to the establishment of “Nissho-Iwai” in October 1968. The major reason behind the merger was in “the response to the increasing needs of the group transactions” (Uchida 1976: p. 344). The development of the Fuyo and DKB Groups took similar paths to the Sanwa Group. For example, Fuji Bank noticed that “after 1955, our customers began to wish more than ever to affiliate with one another in order to better cope with the exploitation of the atomic power and the rise of the petrochemical industry” (Fuji Ginko 1982: p. 925). In replying to such growing request, Fuji Bank proposed “Marubeni,” its customer trading company, to absorb “Takashimaya-Iida,” which gained fame in its high volume steel trading, but was suffering from its failure in the soybeans dealings at the time. Despite continuing the main bank relation with Sumitomo Bank, Marubeni consented to the proposal, in the hope of growing out of the status of a textile trading company and to become a full-fledged GTC in the surge of the heavy and chemical industrialization yet to come. In September 1955, “Marubeni-Iida” was eventually founded after the merger between the two trading companies, 108 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. and started to function as the main GTC for Fuji’s customer firms (Suzuta 1976: p. 50).7 1 Concurrently, Fuji Bank set up “Section One” in its Screening Department to make close contacts with customers which had started group-like transactions. Since 1960, Section One recurrently held liaison conferences with those customers and came to the agreement with them to set up “Fuyo-kai” (Fuji Ginko 1982: pp. 926-27). In the case of the DKB Group, there was a grouping process underway, in the 1960s, between the Furukawa and Kawasaki Groups, which both belonged to the Dai-Ichi Bank Group. This was a natural development of their past transactions, where Kawasaki Heavy Ind. and Kawasaki Airplanes routinely relied on the supply of copper and rubber tires from Furukawa Electric and Yokohama Rubber (Keizai Chosa Kyokai 1967: pp. 12-13). Accepting the proposal from Dai-Ichi Bank, these two groups decided to hold a joint Presidents’ Council every few months in 1966 (Keizai Chosa Kyokai 1967: p. 13). As for the GTC, Itochu served as the main GTC for the Kawasaki Group. Itochu long supplied such raw materials as coal and steel to Kawatetsu, and, as a major shareholder, processed the merger between Kawasaki Heavy Ind. and 72 Yokoyama Kogyo in November 1966 (Suzuta 1976: p. 50). The corporate grouping or “keiretsu-ization” no longer became the privilege of the Big Three. The technological spillovers across industrial sectors, characteristic to the heavy and chemical industrialization, was enhancing the possibility for other non-Big Three firms alike to be intimately connected through asset-specific transactions. Yet, there exists a serious issue pertaining to transaction cost that critically differentiates the Bank Groups from the Big Three. That is, the Bank Groups are significantly differed from the Big Three in how well their member firms are attuned to the intra-group transactions so that the entire group can function ideally as a meta-firm. Those belonging to the Bank Groups are not well-prepared to act in tandem with other members for the intra-group transactions because group formations are based on the motley collections of former minor zaibatsu and other independents. For example, in the case of the Fuyo Group, there were no cross-shareholdings among Nihon Kokan (independent), Nissan Motors (the former Ayukawa or 109 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Hitachi-Nissan zaibatsu), and Marubeni (independent) before Fuyo-kai was opened (Okumura 1976: p. 210, 271). In the Sanwa Group, on the other hand, its three major constituents, Hitachi Shipbuilding (the former Ayukawa or Hitachi-Nissan zaibatsu), Ube Industries (independent), and Teijin (the former Suzuki Shoten), constantly kept their independent positions from the Group (Okumura 1976: p. 238). Besides, in the Sanwa Group, two other quasi-Presidents’ Councils (“Colover-kai” and “Midori-kai”) were composed of those which did not join in Sansui-kai. Most explicit in showing the motley nature of the Bank Groups is the DKB, which consisted of “Furukawa Sanksui-kai” (1955-) of the Furukawa Group, “Mutsumi-kai” (1955-) of the Kawasaki Group, “Kangin Jugosha-kai” (1970-) of the former Nippon Kangyo Bank, and those led by Dai- Ichi Bank. The DKB Group was too immature for being a keiretsu group because its interior transactions were negligible other than those within and between the Furukawa and Kawasaki 73 Groups (Okumura 1976: p. 221; Suzuta 1976: p. 15). Such immaturity seems to be reflected in the formation of their own Presidents’ Council (Sankin-kai, January 1978) and their less frequent holding of the council (every three months) than other five keiretsu groups (every month). Consequent to such an ad hoc way of forming keiretsu groups were doubled, and, occasionally, even tripled investments in the same industrial sectors by the member firms, which necessarily spurred their internecine competition. Obviously, such competitions run counter to the spirit of the one-set principle. The problem often arose, as shown in Table 27 in such sectors as trade and commerce, chemicals, steel, electrical machinery, and transportation machinery. For example, in 1971, an incident occurred in which the MITI resorted to administrative guidance to urge affiliation between Hitachi Co. and Fujitsu, the two members of the DKB Group, in order to remove their excessive competition before the trade liberalization of computers was implemented in April 1976 (Johnson 1982: p. 302; Suzuta 1976: p. 32). For another example, a failed merger occurred among Asahi Chemical, Yokohama Rubber, and Nihon Zeon in the DKB Group shortly after the DKB Group was formed: at the last moment when the agreement was about to be reached, the Furukawa Group suddenly changed its position and thwarted it (Suzuta 1976: p. 40).7 4 The one- 110 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. set principle was also breached by some constituents traversing multiple keiretsu, such as Itochu, Kobe Steel, Hitachi Co., and Nippon Express. Their straddling multiple keiretsu indicated their limited involvement in the membership of each. As shown in Table 27, Hitachi Co. was the best example of taking a noncommittal position, crossing over all three Bank Groups at a time. Reportedly, Hitachi capitalized on its prestige in the electric industry, inducing fierce lending competition among Fuji, Sanwa, and Dai-Ichi Banks until the cease-fire agreement was reached that they lend Hitachi in the way that Hitachi’s borrowing rates become equalized (Suzuki 1993: p. 87). The GTCs were not immune from the above problem. In the Sanwa and DKB Groups, their GTC’s were involved in multiple investment issues (Nichimen, Nissho-Iwai, and Iwatani in the Sanwa Group. Itochu, Kanematsu-Gosho, Nissho-Iwai, and Kawatetsu Corp. in the DKB Group) and multiple memberships (Nissho-Iwai joining in both Sanwa and DKB). DKB, among all, provided the most confusing case where its four sub-groups and four GTCs produced extremely intricate combinations: (1) the Furukawa Group—Nissho, (2) Mutsumi-kai—Nissho, (3) Kangin Jugosha-kai -Kanematsu, and (4) the Kawasaki Group—Itochu and Kawatetsu Corp. (Suzuta 1976: pp. 50-2). The melange of those sub-groups, led by Dai-Ichi and Kangyo Banks, was followed by a series of accidents, while these banks tried to integrate them the hardest way. First, after Nissho turned to the Sanwa Group when Sansui-kai was formed, Itochu gradually came to power, and the Furukawa Group and Mutsumi-kai were both advised by Dai-Ichi Kangyo Bank to switch their GTC to Itochu, with which they had previously made no transaction.7 5 Following this was the mis- coordination among Itochu, Furukawa Aluminum, and Nippon Light Metal, in April 1973, when the latter two were involved in an internecine feud in the Furukawa Group over aluminum production. Not much interested in making the entire group transactions coherent, Itochu financed Furukawa Aluminum and let it enter the aluminum industry, with the consequence that there occurred a double-investment in the industry (Suzuta 1976: p. 42, 212). Also, in the same period, when Itochu was proposing the production of the high-temperature gas reactors in the “Dai-Ichi Atomic Power Group,” other members declined Itochu’s proposal, and, instead, supported Fuji Electric’s plan on 111 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. the production of the hydro-pressured high reactors (Suzuta 1976: pp. 210-11). Even after the formation of the DKB Group in 1978, Itochu seemed far from the “group organizer.” For instance, even after officially joining the DKB Group, Ishikawajima-Harima Heavy Ind. never utilized Itochu, but kept relying on other non-group GTCs such as Mitsui Corp., Nissho-Iwai, Mitsubishi Corp., and Marubeni (Suzuta 1976: p. 150).7 6 Although not expressed clearly in the numeric form, the high transaction costs involved in the intra-group transactions are almost evident in the three Bank Groups. That was particularly true of the DKB Group. They frequently doubled, even tripled investments in the identical industries, and the internal feuds and transaction stalemate all indicate this. All these disadvantages, in turn, ascribe to their motley character of collecting the smaller former zaibatsu firms, which, unlike those belonging to the Big Three, was never arranged in the mode of the one-set principle in the prewar period. A paradoxical consequence ensued that the three main banks (Fuji, Sanwa, and Dai-Ichi Kangyo Bank), boasting of their superb financial capabilities to pull out firms from across industries, later suffered from the financial leakage out of their groups. Suzuta (1976) described the situation in the following: In the case of Dai-Ichi Kangyo Bank, there often occurred a problem that the money the Bank financed to its group firms easily flew away to other banks because the Bank does not hold so many keiretsu firms closely related, (p. 89) 2.5 The Comparison among the Six Horizontal Keiretsu from the TCE Standpoint: Cross-Shareholdings and Financing Now that the historical overview and the related analyses were discussed for both the Former Zaibatsu Groups (the Big Three) and the Bank Groups, this section puts all the six keiretsu together and carries out additional comparative data analyses. To confirm the relevancy of the TCE hypothesis adopted so far, this section mainly focuses on variables pertaining to stockholdings of each keiretsu. This is because they are considered proxies for the intra-keiretsu transactions. As the 112 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. stockholdings relations are often intricate across firms in each keiretsu, the analyses proceed along the subsections, each of which focuses on the dimension of (1) the whole keiretsu group, (2) the main bank, (3) the GTC, and (4) the main bank and the GTC. Equally as important as the stockholdings variables is financing made by the main bank and its positions in the keiretsu firms. Therefore, the comparison of the borrowing ratios of the six keiretsu ensued. These works will then be supplemented by the actual trade data given by the Fair Trade Commission (FTC). The analysis starts from 1973 when the stockholdings data of the six keiretsu were made available by Toyo Keizai Shinposha, Kigyd Keiretsu Soran (various issues. The analysis of the DKB Group starts from 1978.). 2.5.1 The Intra-Group Stockholdings As the first step for confirming the TCE hypothesis validity, this subsection examines the overall characteristics of the inter-corporate stockholdings of the six keiretsu through three variables: “the intra-keiretsu stockholdings ratio,” “the intra-keiretsu stockholdings relation ratio,” and “the intra- keiretsu average stockholdings ratio.” They prove useful for both the cross-sectorial and time-series comparisons of the six keiretsu in terms of the inter-corporate stockholdings. First, “the intra-keiretsu stockholding ratio” is the percentage of the number of stocks issued by members relative to the total balance of the stocks issued by members in a keiretsu group. As displayed in Figure 1, the ratios of Mitsubishi and Sumitomo keiretsu are impressively higher than those of four other keiretsu throughout the whole period, whereas that of DKB stays consistently the lowest. Between these two extremes are the Mitsui, Sanwa, and Fuyo keiretsu, with Mitsui somewhat exceeding Sanwa and Fuyo for almost all the period. This result is congruent with the states of the inter-corporate transaction relations of the six keiretsu analyzed so far: piling up more (less) members’ stocks within a keiretsu implies that members are more (less) likely concerned with 113 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. the corporate governance of other members for the purpose of securing the stable transactions with them. The higher intra-keiretsu stockholdings ratios of the Big Three, in particular Mitsubishi and Sumitomo, are the reflections of their pressing need for frequent intra-keiretsu transactions centering in the heavy and chemical industries. Diametrically opposite to this case is DKB, where the members are the least inclined to cross-shareholdings because of their very sparse intra-keiretsu transactions. Intra-Keiretsu Stockholdings Ratio 35 Mitsui Mitsubishi — * — Fuyo - • * • - Sanw a — o — DKB S. 15 a ^ a* aQ dp dd oh eft op o fo 6S o> .0,' .op oP < sP oP ,oP ,oP .oP .oP .dp .dP .dp .dp .dp .dp Y ear F igure 1. Intra-keiretsu stockholdings ratio. The above point is confirmed again by looking further into two other variables: “the intra- keiretsu stockholdings relation ratio” and “the intra-keiretsu average stockholdings ratio.” Defined as “the number of firms whose stocks are held by other members devided by the number of the possible intra-keiretsu stockholdings relations,” “the intra-keiretsu stockholdings relation ratio” measures how much cross-shareholdings are diffused in a keiretsu group. As displayed in Figure 2, throughout the time, the ratios are ranked in the order of Sumitomo, Mitsubishi, Mitsui, Fuyo, DKB, 114 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. and Sanwa, validating the contention that cross-shareholdings are more prevalent among the members of the Big Three, particularly Sumitomo and Mitsubishi, than those of the Bank Groups. Intra-Keiretsu Stockholdings Relation Ratio — • — Mitsui — E— Mitsubishi — A— Sumitomo — X— Fuyo ; — * — Sanw a • - o - -DKB 1973 1977 1981 1985 1987 1989 1992 1996 Y ear F igure 2. Intra-keiretsu stockholdings relation ratio. The other variable, “the intra-keiretsu average stockholding ratio,” is defined as “the sum of each member’s stockholdings ratios divided by the number of the actual stockholdings relations,” and, as drawn in Figure 3, is higher in the Big Three, roughly in the order of Mitsubishi, Sumitomo, and Mitsui, than the Bank Groups, roughly in the order of Sanwa, Fuyo, and DKB, over time. This is, 77 again, the expected result from the TCE perspective. 115 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Intra-Keiretsu Average Stockholdings Ratio 4.5 3.5 — *— Mitsui - • • - - Mitsubishi — a — Sumitomo .2 2.5 ■*— Sanwa O — DKB -O 0.5 1973 1977 1981 1987 1989 1992 1996 Y ear F igure 3. Intra-keiretsu average stockholdings ratio. 2.5.2 The Main Bank and the Non-Financial Members The Japanese Keiretsu Model showed that the main bank’s holding of stocks issued by its non-financial members plays a crucial role in securing hostages for the main bank, thereby allowing it to raise its voice against those non-financial members. In order to check how much the main banks of the six keiretsu raked in their members’ stocks in the past, this section explores three main banks’ stockholdings variables in the similar way to the previous section: “the main bank’s stockholdings ratio,” “the main bank’s stockholdings relation ratio,” and “the main bank’s average 78 stockholdings ratio.” Figure 4 plots “the main bank’s stockholdings ratio,” defined as “the number of the non-financial members’ stocks held by the main bank divided by the total balance of the stocks issued by the non-financial members.” 116 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Main Bank's Stockholdings Ratio 4.5 3.5 — Mitsui — □— Mitsubishi — A— Sumitomo — H— Fuyo - - - X - ■ ■ Sanwa — • — DKB o 1 5 0.5 ^ ^ ^ N # > ,c £ > N c^ ^ s c £ > ^ Year F igure 4. M ain b a n k ’ s stockholdings ratio. As shown in the graph, the ratios of the six main banks are constantly concentrated and show the general declining trend from about 4.5% in the early 1970s to about 2.5% in the mid- 1980s. Under such concentration, however, the ratios of the Bank Groups are often slightly higher than the Big Three: they are almost constantly higher than Sumitomo and Mitsui. It was not until 1989 when Mitsubishi, which, by the time, stayed somewhere between Fuyo and Sanwa and DKB, could catch up with them. This actually matches with the previous observation that the Bank Groups are characterized more by their tight relations between the main banks and the member firms. The concentration of the ratios, along with their general decline from the early 1970 to the early 1980s, has actually to do with the legal issue of Section 11 of the Antimonopoly Law, which prohibited city banks from owning more than 10% of corporate stocks between 1953-1976, and, which, in 1977, strengthened its restriction by bringing down the rate to 5%. Table 28, on the other hand, plots “the main bank’s stockholdings relation ratio,” defined as “the number of non-financial firms whose stocks are possessed by the main bank divided by the total number of the non-financial member 117 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. firms,” and displays how extensively the six main banks, under the constraint by the Antimonopoly Law, collected the stocks from their group constituency. Table 28. The Main Bank’s Stockholdings Relation Ratio Main Bank 1973 1974 1978 1979 1980 1983 1984 1985 1986 Mitsui 21/21 21/21 22/22 22/22 22/22 21/21 21/21 21/21 21/21 Mitsubishi 22/22 22/22 22/22 22/22 24/24 22/22 22/22 22/22 24/24 Sumitomo 14/14 14/14 18/18 18/18 19/19 18/18 18/18 19/19 18/18 Fuyo 27/27 27/27 27/27 27/27 27/27 27/27 27/27 28/28 27/27 Sanwa 28/28 34/34 37/37 37/37 37/37 39/39 39/39 41/41 34/34 i DKB 40/40 40/40 41/41 41/41 42/42 42/42 34/34 Main Bank 1987 1988 1989 1990 1991 1992 1993 1996 1998 Mitsui 21/21 21/21 i 21/21 21/21 23/23 23/23 24/24 24/24 23/23 Mitsubishi 24/24 25/25 1 25/25 25/25 25/25 25/25 26/26 25/25 24/24 Sumitomo 18/18 18/19 18/18 18/18 18/18 18/18 18/18 18/18 18/18 Fuyo 27/27 27/27 27/27 27/27 27/27 26/26 27/27 27/27 26/26 Sanwa 34/34 41/41 41/41 41/41 41/41 41/41 41/41 41/41 39/39 DKB 34/34 43/43 43/43 43/43 43/43 44/44 44/44 44/44 44/44 Note. The numerator denotes “the number o f non-financial firms whose stocks are possessed by the main bank” and the denomninator, “the total number o f the non-financial member firms.” In contrast to “the stockholdings relation ratio” witnessed in the previous section, the main bank’s stockholdings relations with the non-financial members are constantly 100% with no regard to which keiretsu the main bank belongs to. The six main banks were, thus, constantly in the best position to oversee their client members through holding their residual rights, in addition to lending. No doubt, this is one of the most important aspects of the Japanese main bank system. Finally, taking into consideration the divergence in the number of member firms across the six keiretsu, “the average ratio of the main bank’s stockholding of the non-financial members” is derived for the six main banks, denoting “the sum of the main bank’s stockholding ratio for each 118 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. non-financial member divided by the number of the main bank’s actual stockholding relations with the non-financial members.” As exhibited in Figure 5, both the movement and ranking of the ratios show the similar pattern to “the main bank’s share in the aggregate non-financial members’ stocks,” indicating the effect of the Antimonopoly Law. Main Bank's Average Stockholdings Ratio 5 — ♦— Mitsui — B— M itsubishi — A— Sum itom o — X— Fuyo - - * - - S anw a — O— DKB 4 .2 3 1974 1977 1980 1985 1987 1989 1992 1996 Y ear F igure 5. M ain b a n k ’ s average stockholdings ratio. There is yet another element that buttressed the Japanese main bank system. That is, how much stock issued by their main bank are the non-financial members willing to possess: as demonstrated in the Japanese Keiretsu Model, in order for the non-financial members to be able to exercise the countervailing influence toward their main bank, and thereby to expedite financing into the production process, they also have to secure part of the stocks issued by the main bank as their hostages. To examine this in detail, the following three variables are introduced in the similar fashion as previously: “the ratio of the non-financial members’ stockholdings with the main bank,” “the ratio of the non-financial members’ stockholdings relation with the main bank,” and “the 119 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. average ratio of the non-financial members’ stockholdings with the main bank.” Figure 6 displays the “the ratio of the non-financial stockholdings with the main bank,” defined as “the number of the main bank’s stocks owned by the non-financial members divided by the total balance of the main bank’s stocks.” Non-Financial Firms' Stockholdings Ratio o £ 30 25 20 ■0-^0--^ 10 5 0 -♦— Mitsui - a — Mitsubishi - it— Sumitomo —X— Fuyo ■ * - - Sanw a - O - -DKB Year F igure 6. N on-financial firm s ’ a vera g e stockholdings ratio. Reflecting that there is no legal restriction imposed on the side of the non-financial members in holding the stocks issued by their main bank, the rates show the divergence across the six keiretsu, with the Bank Groups generally positioned higher than the Big Three. Among all, Fuyo and Sanwa are consistently ranked remarkably high throughout the time. While dominated by the Bank Groups, by 1989 the Big Three are ranked in the order of Mitsui, Mitsubishi, and Sumitomo and thereafter, the first two had reversed their rankings because of Mitsui’s sharp drop in 1990. Defined as “the number of the non-financial members holding the stocks issued by their main bank divided by the total number of the non-financial members,” “the ratio of the non-financial members’ 120 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. stockholdings relation with the main bank,” on the other hand, is to check how widely the main bank’s stocks are possessed among the non-financial members. As displayed in Table 29, the ratios of the six keiretsu are almost 100%, if not completely, throughout the time: in a mutual fashion, the 79 non-financial firms secured their stakes in their main banks throughout the time. Table 29. The Ratio of the Non-Financial Members’ Stockholdings Relation with the Main Bank Keiretsu 1973 1974 1978 1979 1980 1983 1984 1985 1986 1987 Mitsui 16/16 13/13 18/18 18/18 17/17 18/18 18/18 18/18 18/18 18/18 Mitsubishi 15/16 13/13 17/18 17/18 14/17 16/18 16/18 18/18 18/18 18/18 Sumitomo 12/12 12/12 15/15 15/15 15/15 14/16 13/16 13/16 13/16 13/16 Fuyo 9/16 9/13 18/18 18/18 17/17 18/18 18/18 18/18 18/18 17/18 Sanwa 12/16 13/14 17/17 17/17 17/17 18/18 18/18 18/18 17/18 17/18 DKB 17/17 17/17 16/16 18/18 18/18 18/18 17/18 17/18 Note. The numerator o f the fraction denotes the number o f the non-financial firms owning the stocks issued by the main bank, and the denominator, the total number o f the non-financial members. Finally, considering these relation rates, “the average ratio of the non-financial firms’ stockholdings with the main bank” is derived by the definition of “the number of the main bank’s stocks owned by the non-financial members divided by the number of the non-financial members’ stockholdings relations with the main banks” (see Figure 7). Somewhat concentrated, the movement and ranking of the ratios display the similar pattern to the first variable. The only exception is DKB, which is now ranked the lowest almost all the time. This indicates that the fairly high ranking of DKB in the first variable (“the ratio of the non-financial members’ stockholdings with the main bank”) was primarily ascribed to the huge number of Sankin-kai’s member firms. Taking also into consideration that the stockholdings relations among the non-financial members of DKB are extremely scarce, the non-financial members of the DKB 121 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Group are the least likely to take the joint action for exercising the voting right against Dai-Ichi Kangyo Bank in order to direct the Bank to properly finance the intra-keiretsu production. Average Ratio of the Non-Financial Members' Stockholdings with the Main Bank 2.5 o 0.5 z /b ^ a Q qQ ofb o j 1 ( & flfil qN ^yV ryb rSb oS .<v .<v .c S ® .< $ > .Cp .oP .CP .C^ .OP .< X ® .< X ® .(X ® .c s® .< x ® .e x ® - Mitsui -M itsubishi; — A— Sumitomo — K— Fuyo Sanwa — 0— DKB Year Figure 7. Average ratios o f the non-financial members ’ stockholdings with the main bank. Combining all the above results helps to make clear the dissimilarity among the six keiretsu, particularly that between the Big Three and the Bank Groups. First of all, it should be pointed out that the Bank Groups, among all, Fuyo and Sanwa are characterized by their higher cross-shareholdings between the main banks (Fuji Bank, Sanwa Bank) and the non-financial members than the Big Three. The Bank Groups’ bias toward the inter-corporate relations between the main banks and the non-financial members is evident all the more for their poor records in the three variables examined in the previous section. Such a characteristic of the Bank Groups is obviously ascribed to their origins where the main banks played the pivotal role in mobilizing their member firms. Diametrically opposite to Fuyo and Sanwa in this regard is Sumitomo, which was characterized by the tenuous cross-shareholdings between Sumitomo Bank and the non-financial members of Hakusui-kai. This, in turn, is the reflection of Sumitomo’s development path in which the firms positioned in the mining sectors often took precedence and, thus, were induced to make 122 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 80 more cross-shareholdings among them, rather than with Sumitomo Bank. In between the two extremes is Mitsui, which, despite its status as one of the Big Three, seems to take the closer pattern to the Bank Group in its keeping the mediocre positions in all the variables examined in the present and previous subsections. This seems to convey Mitsui’s state of the maturity in the intra-keiretsu transaction, which is mediocre in comparison with Sumitomo and Mitsubishi, yet is more advanced than the Bank Groups because of its being one of the earlier starters. In closing this subsection, it should be noted that all the six keiretsu had, more or less, experienced the declines in the first and third variables. This, of course, indicates that their non- financial members had gradually thrown away their stakes in their main banks, which is obviously erosive to their intra-keiretsu production activities. The next chapter discusses this problem in detail, especially in connection with the recent crisis in the Japanese main bank system. 2.5.3 The GTC and the Non-Financial Members From the historical overviews and the related analyses showing the states of the intra- keiretsu transactions divergent across the six keiretsu, it is predicted from the TCE viewpoint that such divergence is properly reflected in the stockholdings variables related with the GTCs. In order to validate the prediction, this section examines the following six variables: “the GTC’s stockholdings ratio,” “the GTC’s stockholdings relation ratio,” “the GTC’s average stockholdings ratio,” “the rate of the members’ stockholdings with the GTC,” “the rate of the members’ stockholdings relation with the GTC,” and “the average rate of the members’ stockholdings with the GTC.” To derive these six variables, the following ten GTCs are selected from the six keiretsu: (Mitsui Corp., Mitsubishi Corp., Sumitomo Corp., Marubeni (Fuyo), Nichimen (Sanwa), Nissho- Iwai (Sanwa), Iwatani Corp. (Sanwa), Itochu (DKB), Kanematsu-Gosho (DKB), and Nissho-Iwai (DKB).8 1 123 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. “The GTC’s stockholdings ratio” is defined as “the number of the members’ stocks held by the GTC divided by the total balance of the stocks issued by the members,” and plotted for the ten GTCs in Figure 8. GTCs Stockholdings Ratio 2 .00 % 1.80% 1.60% 1.40 % 1.20 % 0 1 1.00% a 0 .80% 0 .60 % 0 .40 % 0 .20% 0.00% . 1 - • • i.. -1—J.—X —i.. -1- .12:: D rrS xo A A oU (JJ ip ip ,« Q .- V < < i x ' ex' .ex' .ex' . a y . cxv cxP .a y aSP txP .a P cxP .op .aP op S f K °> ' K °> ' A X<SP kQ ? tp o lp d d _ 0 ? 5 0 ° > Year — 0— Mitsui — ■— Mitsubishi — A— Sumitomo — X— Fuyo — * — Nichimen — • — Nissho — I — Iwatani —■ ■ ■ —— Itochu - - — ■ ■ K anem atsu --■ ❖ --N issh o - - O - - Kawatetsu Figure 8. GTC’ s stockholdings ratio (non-unified version). As expected, the rate is higher in the Big Three, particularly in Sumitomo and Mitsubishi throughout the period. While dominated by the Big Three, in the Bank Groups Fuyo dominates the rest belonging to either Sanwa or DKB. For a more simplified view, Figure 9 unifies the data of three GTCs in Sanwa and three GTCs in DKB, respectively, and displays them along with others as in Figure 8. As a result of the summations, Sanwa and DKB both became comparable to Fuyo and 82 Mitsui, but the preeminence of Sumitomo and Mitsubishi was not affected. For the purpose of examining how widely such high (low) stakes of the Big Three (the Bank Groups) are distributed among the members, “the GTC’s stockholdings relation ratio” is in order, with the definition being 124 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. “the number of the members whose stocks are possessed by the GTC divided by the total number of the member firms.” GTCs Stockholdings Ratio 2 .00% 1.80% 1.60% 1.40% x- X 1.20% o 1.00% w 0.80% 0.60% 0.40% 0.20% 0 .00% Jv5 -S?3 ..C _C y> „<#> ...Op — ♦— Mitsui -■ — Mitsubishi — Sumitomo - ■ - X ■ ■ Fuyo — * — Sanwa — O — DKB Year Figure 9. GTC’ s stockholdings ratio (unified version — unifying the data o f three GTCs in Sanwa and three GTCs in DKB, respectively, and displaying them along with others). As shown in Figure 10, the ratio is strikingly higher in the Big Three, which is then followed by Fuyo and, then, those belonging to either Sanwa or DKB. While Sumitomo’s achievement of 100% level for nearly all the time is partly attributable to its small number of the member firms, the ranking of the groups in Figure 10 is considered highly correlated with the frequency of the intra-keiretsu transactions within each keiretsu relative to others. 125 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. GTC's Stockholdings Relation Ratio 100.00% r tr 90.00% 80.00% 70.00% 60.00% o 50.00% 1 5 40.00% 30.00% 20 .00% 10.00% 0 .00% op op < > 3 o P o P d P o S d V d P d P d P dp .dp .w > .dp .dp .dp .dp .dp .dp .dp .dp & .<** .d J ? 1 .dJ? .<# .<& Mitsui Mitsubishi Sum itom o Fuyo * - - Nichimen O— N issho H — Iwatani —— Itochu • — - -K anem a - □ — N issho Year Figure 10. GTC’ s stockholdings relation ratio. Finally, “the GTC’s average stockholdings ratio” is derived by “the GTC’s stockholdings ratio divided by the number of the GTC’s stockholdings relations” (see Figures 11 [non-unified version] and 12 [unified version]). The ranking of the GTC’s exhibited in these figures is the same as those of the other two variables examined above, except the minor alteration between Sumitomo and Mitsubishi. Putting all these results together, the conclusion is drawn that the high (low) stockholdings ratios attained by the GTC’s of the Big Three (the Bank Groups) are the outcome of adding up their high (low) average stockholdings ratios through their high (sparse) stockholdings relation ratios. 126 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. G TC s Average Stockholdings Ratio 5.00% 4 .50% 4 .00% 3.50% 3.00% '1 2.50% O ' 2 .00 % 1.50% 1.00% 0- 0 . 0.50% 0 .00 % Vb ,4b gfb gb1 n b n \ gSb o P > rjP ot^ cO' d P d P O ?4 .O,' .0 0 .OO .OP .OP OP .OP ,o P ,o P .OP OP Op .OP .Cip o p Op Op - o — Mitsui - ■ — M itsubihsi -A — S um itom o -X — Fuyo - * — N ichim en - o — N issh o — I — Iwatani - Itochu - - - - - K an em a - - -O - - N issh o Y e ar Figure 11. GTC’ s average stockholdings ratio (non-unified version). GTC's Average Stockholdings Ratio (Unified Version) 4 .50% 4 .00% 3 .50% 3.00% 2 .50% “ 2 .00 % 1.50% 1.00% 0 .50% 0 .00% Vb A 4 * A * ^ A * 4 dP o f* Q i4 O ? 4 9^ 9^ dp < $ > dP dp & < & v .oo .oO .oO .dp .op .dp .dp .dp .dp .dp .dP .op .dp .dp .dp .dp .oP - Mitsui -M itsubishi - S u m ito m o -F u y o - S a n w a -DKB Y e ar Figure 12. GTC’ s average stockholdings ratio (unified version). 127 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Now in consideration of the possible cross-shareholdings between the GTC(s) and other members, stockholdings relations are analyzed from the opposite direction. “The ratio of the members’ stockholdings with the GTC,” defined as “the number of the GTC’s stocks owned by the members divided by the total balance of the stocks issued by the GTC,” and “the ratio of the members’ stockholdings relation with the main bank,” defined as “the number of members owning the stocks issued by the GTC divided by the total number of the members,” are respectively shown in Figures 13 (non-unified version) and 14 (unified version), and Figure 15. Ratio of the Members' Stockholdings with the GTC 35 30 25 15 10 5 0 V b otb o b * q Id aS q S > o S 4 cIP o f* ctt O p dP C > P V .<V .O,' .O,' .O p .O p .O p .O p .O p ,<sp .O p .O p .C S4 .K 4 .CS4 .OP .CS4 .tx4 - ♦ — Mitsui - a — M itsubishi -A — Sum itom o -X — Fuyo - * — N ichim en - o — N issh o — Iwatani — • - Itochu K a n em a - ■ -0- - - N issh o Y e ar Figure 13. Ratio o f the members ’ stockholdings with the GTC (non-unified version). 128 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Ratio of the Members' Stcockholdings with the GTC O f S 35 30 25 20 15 10 5 0 & | Mitsui ! — 0 — Mitsubishi — A—Sum itom o! K Fuyo Sanwa — O— DKB Y ear Figure 14. Ratio o f the members ’ stockholdings with the GTC (unified version). Ratio of the Members' Stockholdings with the GTC r a 100.00% 90.00% 80.00% 70 .00% 60.00% 50 .00 % 40.00% 30 .00% 20.00% 10.00 % 0 .00% V b * p& pfb otp pS > oS p9? pP) p& qN pfb pvb rib p& Y e ar ♦ - Mitsui - ■ — M itsubishi - h . — S u m ito m o j -X — Fuyo - * — N ichim en - o — N issh o —I — Iw atani — Itochu - - o - - K an em a ■ - -0- - - N issho Figure 15. Ratio o f the members ’ stockholdings with the GTC. 129 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. These graphs indicate that, except Mitsui’s decline in the first variable, the rankings parallel those of “the GTC’s stockholdings ratio” and “the GTC’s stockholdings relation ratio.” The third variable, “the average ratio of the members’ stockholdings with the GTC,” is derived by “the number of the GTC’s stocks owned by the members divided by the number of members’ stockholdings relations with the GTC.” As shown in Figure 16 (non-unified version), the rate is prominently higher in Iwatani (Sanwa) and Kanematsu-Gosho (DKB), and constantly the lowest in Mitsui. The higher rankings of Iwatani and Kanematsu-Gosho are also reflected in Figure 17 (unified version), where Sanwa and DKB are positioned as the second for nearly all the time. This result is considered due both to the extremely low records of Iwatani and Kanematsu-Gosho in the second variable (see Figure 15, and to their main banks’ (Sanwa Bank and Dai-Ichi Kangyo Bank) holding of the large blocks of their stocks. Figure 18 and 19 (unified version) both report the case when the main banks’ holding of the GTCs’ stocks is eliminated. Average Ratio of the Members' Stockholdings with the GTC 7 6 5 4 3 2 1 0 , , , , ; , , , , , --------- VS /V VS ^ oS o \ pS b (d V s S ot' (A oS (5) (& & ^ ^ ^ ^ ^ ^ ^ -4 — Mitsui - s — Mitsubishi -is— Sumitomo -X— Fuyo - * — Nichimen -O— Nissho 1 — Iwatani — - - - Itochu — ± — Kanema - - O - - Nissho Year Figure 16. Average ratio o f the members ’ stockholdings with the GTC. 130 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Average Ratio of the Members' Stockholdings with the GTC (Unified Version) 6 5 4 3 2 1 0 & .<# ,d£ .c# .<# .c# .<# .< & .<# .<# V b /ik / i< 6 /iQ i o O < A a,' o > ' o > ' o,' dp dP - Mitsui -M itsubishi -S um itom o - Fuyo -S an w a -DKB Y ear Figure 17. Average ratio o f the members ’ stockholdings with the GTC (unified version). The Case When the Main Bank's Holding of the GTCs Stocks is Eliminated 3.5 2.5 o r a 0.5 {b pSh gfb pb< o(D ofo < * A pSb p Q > qN gif pvb p& pfe > a ' o,' o ' dP dP dP dP dP dP dP dP dP dP d 5 .dP o fi dP - ♦ — Mitsui I H i— M itsubishi; -A — Sum itom o -X — Fuyo - * — Nichimen - o — N issho H — Iwatani j Itochu — A— K anem a N issho Y ear Figure 18. The case when the main bank’ s holding o f the GTC’ s stocks is eliminated (non-unified version). 131 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. The Case When the Main Bank's Holding of the G TC s Stocks is Eliminated 3.5 — Mitsui —n — M itsubishi —is— S um itom o —* — Fuyo — Sanwa —o — DKB _ 2.5 0.5 -V P /it* Vb ofb o P * pfo q \ o S b p& dp oN rtt' op oP dP v .<v . a ' ,o>' .oP .d p .o p .dP .oP .oP .o p .oP .cs4 .cs4 .0? .o 4 .o 4 .cs4 Y e ar Figure 19. The case when the main bank’ s holding o f the GTC’ s stocks is eliminated (unified version). As shown in these figures, after the elimination of the main banks’ stockholdings, Iwatani and Kanematsu-Gosho recede to modest positions, and their orders are replaced by Sumitomo and Mitsubishi. This result, in fact, agrees with the observation in the previous section that the Bank Groups are best characterized by their main banks’ partaking in the intra-keiretsu stockholdings as the major shareholders. The extremely low profile of Mitsui, on the other hand, was not essentially altered after the elimination. The lower positions of the Bank Groups and Mitsui in their non- financial members’ stockholdings of their GTCs again indicate their fragile states of the intra- keiretsu transactions. To sum up, in proportion to their phenomenally high frequencies of the intra-keiretsu transactions, the GTCs and the member firms of the Mitsubishi and Sumitomo keiretsu are positively engaged in cross-shareholdings with each other. Standing in sharp contrast with them are 132 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. the Bank Groups, among all, Sanwa and DKB, where cross-shareholdings between the GTCs and the rest of the group members are poor in both the aggregate and average terms. As the final remark of this subsection, it is important to note that, except for Mitsubishi, all the rest of the keiretsu have experienced, as the general trend, the declines in both “the GTC’s stockholdings ratio” and “the ratio of the members’ stockholdings with the GTC” over time. Such a trend is, in turn, ascribed to either the simultaneous declines in the GTC’s stockholding relations with the members, and the members’ stockholdings relations with the GTCs, or the drops in the amounts of the stockholdings on both parts of the GTCs and the members (or the synergetic effect of both). No doubt, either of these cases would be detrimental to the intra-keiretsu transactions. This problem is taken up in the next chapter and argued in connection with the recent crisis of the Japanese main bank system. 2.5.4 The Main Bank and the GTC As discussed in subsection 2.4.3, during the high economic growth of the early 1960s, the GTCs of the Big Three were all provided with ample financing from their main banks, whereby the buying member firms were given credits, and the intra-keiretsu transactions were consequently promoted. Under such circumstances, it is inferred that these GTCs had strong incentives to possess more of their main banks’ stocks to win favorable lending terms.8 3 It is also inferred that the main banks, on their part, had the urge to secure more of the stocks issued by the GTCs because they could pressure the GTCs to observe their original repayment plans by hinting at possible future intervention. How much the main bank shared in the total balance of each GTC’s stocks is displayed in Figure 20 (“the ratio of the main bank’s stockholdings with the GTC”). 133 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Ratio of the Main Bank's Stockholdings with the GTC n > 02 12 10 8 6 4 2 0 J v 3 -Q ? 3 -9 v > - A * * A * -A ^ .9? .< £ > ^ -CV5 -C?3 -C?3 — ♦— Mitsui — □ — M itsubishi — A— S um itom o — K— Fuyo — * — N ichim en — O— N issh o — I — Iwatani ■ - O ■ • Itochu — — K a n em a N issh o Year Figure 20. Ratio o f the main bank’ s stockholdings with the GTC. As predicted, Figure 20 shows that all six main banks partook of the stockholdings of their member GTCs over time. The ratios were concentrated in the narrow ranges of 6-8% by the mid 1980s, and of 4-6% thereafter. This was primarily because of the Antimonopoly Law, as seen in subsection 84 2.5.2. Yet, still discemable is that, except the case of Itochu, the three main banks of the Bank Groups kept higher ratios in the mid-1980s. This result reconfirms the bank-dominated nature of the Bank Groups. In consideration of possible cross-shareholdings, “the ratio of the GTC’s stockholdings with the main bank” is in order here to examine how much stocks the GTCs obtained from those issued by the main banks (see Figures 21 [non-unified version] and 22 [unified version]). 134 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Ratio of the G TC s Stockholdings with the Main Bank 3 1.5 /{b C tS C k ' O v ' O v ' ^ ^ <$r n cT n cT ^ .& .<& $ > & K ° > ' ST SP k Q P ,SP k < 3 P k Q P ,SF . < & ■ _ d p - 5 3 ° ^ J ♦ Mitsui -H — M itsubishi -A — S u m ito m o -H — F uyo - * — N ichim en • ■ ■ N issh o — I — Iw atani - - A ■ ■ Itochu — A— K a n em a — 0— N issh o Y e a r Figure 21. Ratio o f the GTC’ s stockholdings with the main bank (non-unified version). Ratio of the GTC's Stockholdings with the Main Bank 3.5 2.5 o 0.5 A * " * * A ^* A *P A^ P$^ Pv^ p^* d o P^ P^ P?* Pv* dS> ot^ £$• o p dp dP o s .o s .o s .OP .oP .OP .oP .oP .oP .oP .OP .cs5 .OP .op .o p .o p .o p -M itsui - M itsubishi -S u m ito m o - Fuyo - S a n w a -D K B Y e ar Figure 22. Ratio o f the GTC’ s stockholdings with the main bank (unified version). 135 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. From these figures, it is ascertained that, except for the minor case of Iwatani and Kanematsu- Gosho, the GTCs always possessed the main banks’ stocks, and that the ranking was, for most of the time, in the order of Mitsubishi, Sumitomo, Mitsui, Fuyo, and those of either Sanwa or DKB. Obviously, this ranking corresponds to the degree of the intra-keiretsu transactions analyzed so far. This result also reconfirms the observations in subsections 2.5.2 and 2.5.3 that the Bank Groups are characterized more by financial relations between their main banks and the non-financial members, except the GTC. Finally, it should be noted that Figures 21 and 22 indicate that, except Sumitomo and Fuyo, the GTC’s holding of the stocks issued by the main banks were under general decline over the course of time. From the TCE viewpoint, this is considered to have been caused by either GTC’s not wishing to be financed from their main banks or their declining role in processing the asset-specific transactions among members. 2.5.5 The Financial Connection between the Main Bank and the Non-Financial Members Equally as important as the various stockholdings relations examined above are those bound by financial obligations arising from the lending and borrowing between the main bank and the non-financial members in each keiretsu. This section assesses this aspect of the inter-firm relations for the six keiretsu by focusing on the following two variables: “the main bank’s financing relation ratio” and “the intra-keiretsu borrowing ratio.” Defined as “the relation rate of the intra-keiretsu lending by the main bank” denoting “the number of the non-financial member firms which accept loans from the main bank divided by the total number of the non-financial member firms,” “the main bank’s financing relation ratio” measures how extensively the main bank supplies financing to its group non-financial member 136 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. firms. As exhibited in Figure 23, all six keiretsu had high ratios throughout the periods, which indicates that the six main banks, in general, extensively financed their non-financial member firms. Main Bank's Financing Relation Ratio 120.00 100.00 80.00 .2 60.00 1 3 tt. 40.00 20.00 0.00 f t —♦ — M itsui - a — M itsubishi —is — S u m ito m o —X— Fuyo — Sanwa —o — DKB 1974 1978 1980 1985 1987 1988 1989 1992 1993 1996 1998 Y e a r Figure 23. Main bank's financing relation ratio. It should also be noted, however, that Mitsui, Sanwa and DKB were often positioned at somewhat lower ratios than the rest. In particular, Sanwa, which once was positioned at 100% in 1974, shows the generally declining trend in the ratio since then. Contrary to Sanwa, DKB was much less likely to vary the ratio, staying constantly in the very narrow margin between 92% and 95%, but never reaching 100% in the entire period. Most consistently at the 100% level were Sumitomo and Fuyo, which dropped off from that level only once (Sumitomo in 1980) or twice (Fuyo in 1987 and 1992).8 5 “The intra-keiretsu borrowing ratio,” on the other hand, is defined as “the sum of the members’ borrowing from their main bank divided by the sum of the total borrowing of the 137 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. members,” measuring the degrees of financial stake that the main banks hold in their own keiretsu (see Figure 24). Intra-Keiretsu Borrowing Ratio o r e 0 £ 18 16 14 12 10 8 6 4 2 0 .VI) ^ ^ oS i gift gfc* n 6 gfo q\ gSS q d 4 1 (ft* 4 1 ( f t ( f t ( f t os .o,' o,' .dP ,c>P ,cS° .dP .dP .cSP .dp .dp .op .Op .Op .op .op .op -M itsui -M itsubishi - Sum itom o - Fuyo - S an w a -D K B Y ear Figure 24. Intra-keiretsu borrowing ratio. While it is hard to clearly delineate from the figure the trend of the ranking among the six keiretsu, other than Sumitomo’s predominance for almost the entire period, there exists the bipolarization between the top three (Sumitomo, Mitsubishi, and Sanwa) and the bottom three (Mitsui, Fuyo, and DKB), in which changes in the ranking occur only in each of these two groups. Breaking beyond the boundary between the two groups, the remarkable changes in the ranking occurred in 1985-88 and in the 1990s when Mitsui and DKB, respectively, went up to the second highest position. In light of the fact that the speculative bubbles and the subsequent financial mess occurred in these periods, however, Mitsui’s and DKB’s huge leap upward in these periods was most likely anomalous. Mitsui’s unusual amplifying the intra-keiretsu borrowing ratio during 1985-86 is mostly ascribed to 138 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. the Mitsui firms’ rapidly increased borrowing from Mitsui Bank (see Figure 25), which itself was brought about by the BOJ’s incessant reduction in its official discount rate and to their not relying so much on such other means as equity fiancing.8 6 Intra-Keiretsu Borrowing 1,400,000 1,200,000 a 1,000,000 800,000 600,000 o 400,000 200,000 £ .oP .<# .<# .<# , < £ > N .<# .<# .<# .<# Mitsui Mitsubishi Sum itom o Fuyo S anw a DKB Y ear Figure 25. Intra-keiretsu borrowing. Mitsui’s huge leap in the ratio is also highlighted by the declines of Mitsubishi, Sumitomo, Sanwa, and DKB during the same period, but their declines were primarily due to their disproportionate reliance on equity financing relative to borrowing from their main banks, as the 87 firms of those keiretsu also increased their borrowing from their main banks (see Figure 25 again). DKB’s sudden increase in the ratio in the 1990s, on the other hand, arose mainly from the other non-main banks’ credit withdrawal from Sankin-kai members after the collapse of the bubble economy. This is because the balance of the loans made by Dai-Ichi Kangyo Bank to Sankin-kai members were nearly constant between 1991-95 (Figure 25). Thus, DKB’s unusual ascension in the 139 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. ratio during the 1990s was not by Dai-Ichi Kangyo Bank’s fortifying a lending grip on Sansui-kai members; it was rather the reflection of the members’ suffering from the credit crunch under their high exposure to other non-main banks. To avoid these anomalies, it would be safer to look at the data prior to the bubble economy period, when the bipolarization likely persisted between the top three (Sumitomo, Mitsubishi, and Sanwa) and the bottom three (Mitsui, Fuyo, and DKB). Together with the states of cross-shareholdings examined in the previous subsections, the conclusion can be drawn that the Mitsubishi and Sumitomo keiretsu bear the closest resemblance to the Japanese Keiretsu Model, as they are both characterized by the high rates of cross-shareholdings intermeshed among the main bank, the GTC, and other non-financial firms, and the high leveraged states of the non-financial firms promoted by their main banks. The other four keiretsu also keep such resemblance, but in lesser degrees. Among all, DKB is judged as the least to keep such resemblance because of its constantly weak cross-shareholdings rates and lower leveraged states, across the Sankin-kai members.8 8 2.5.6 The Status o f the Intra-Keiretsu Trades: The FTC’ s Reports (1994,1998) The hard obstacle in analyzing the Japanese keiretsu in terms of the transaction cost approach is that, unlike the stockholdings and financing, it is practically impossible to gain access to the information of the actual trades carried out among member firms. This is because “each company holds the records of such trades [the group internal trades], but never publishes them” (Okumura 1976: p. 130, the brackets added by the present author). Fortunately, today, the reports based on the survey research conducted by the Fair Trade Commission (FTC) in FY 1992 and 1996 are available (Toy5 Keizai Shinposha 1994b; Kosei Torihiki Iinkai 1998a, b). Although the survey research started in 1982, the Commission successfully collected actual transaction data from the keiretsu member firms and translated them in the forms of “the intra-keiretsu sales ratios” and “the intra-keiretsu purchases ratios.”8 9 140 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. “The intra-keiretsu sales ratio” and “the intra-keiretsu purchases ratio” are respectively defined as “the member production firms’ sales to other member firms divided by the total sales by the member production firms” and “the member production firms’ purchasing from other member firms divided by the total purchasing by the member production firms,” and reported for FY 1981, 1989, 1992, and 1996 in Table 30. Table 30. The Intra-Keiretsu Sales Ratio and the Intra-Keiretsu Purchases Ratio Ratio Mitsui Mitsubishi Sumitomo Fuyo Sanwa DKB Intra- keiretsu sales ratio 1981 25.4 29.6 31.0 16.5 7.3 13.3 1989 8.97 25.66 38.11 12.65 6.29 11.86 1992 8.41 22.62 31.67 9.6 5.15 10.19 1996 6.75 18.65 18.75 9.59 3.59 9.37 Intra- keiretsu purchases ratio 1981 16.2 21.9 14.0 5.1 6.2 9.5 1989 4.67 14.97 16.44 3.96 5.24 7.79 1992 3.89 15.87 9.46 3.67 4.67 6.5 1996 3.57 11.47 6.39 2.49 4.27 4.72 Source: The data are from Toyo Keizai Shinposha (1994b: p. 139) and Kosei Torihiki Iinkai Jimukyoku (1998b: p. 172). Note. The unit is %. In agreement with their high magnitudes in various stockholdings relations, cross shareholdings, and financing, both ratios of the Mitsubishi and Sumitomo keiretsu are shown to be much higher than those of the Bank Groups’ in all the four years. Despite its longstanding history as one of the Big Three, Mitsui’s records in the two ratios are somewhat mediocre, plummeted in 1989, and have remained in doldrums since then. The divergence in the intra-keiretsu sales and purchasing across the six keiretsu is also reflected in how much their intra-keiretsu transactions are intermediated by their GTCs. The items (a)-(d) of Table 31 show the intra-keiretsu sales and purchases ratios mediated by the GTCs for each keiretsu in FY 1992 and 1996. Considerable 141 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. amounts of the intra-keiretsu sales and purchases in the Mitsubishi and Sumitomo keiretsu are via Mitsubishi Corp. and Sumitomo Corp. Quite contrary, in the Sanwa and DKB keiretsu, the sales and purchases that were not mediated by the GTCs are almost comparable to those intermediated by the GTCs. In particular, in Sanwa, the former was outstripped by the latter in 1996. These facts actually conform with the different states of cross-shareholding between the Big Three and the Bank Groups. Table 31. Intra-Keiretsu Sales Ratio and Intra-Keiretsu Purchases Ratio (1992,1996) 1 itr -Keiretsu Ratio (a) Intra-Keiretsu Sales Ratio (1992) Mitsui Mitsubishi Sumitomo Fuyo Sanwa DKB From members to GTC From members to non-GTC members 6.64 1.77 18.76 3.86 31.06 0.61 8.55 1.05 3.65 1.49 6.02 4.17 (b) Intra-Keiretsu Sales Ratio (1996) Mitsui Mitsubishi Sumitomo Fuyo Sanwa DKB From members to GTC From members to non-GTC members 5.8 0.95 15.43 3.22 18.29 0.46 8.87 0.72 2.63 0.96 5.84 3.53 (c) Intra-Keiretsu Purchases Ratio (1992) Mitsui Mitsubishi Sumitomo Fuyo Sanwa DKB From members to GTC From members to non-GTC members 2.61 1.28 10.47 5.4 8.5 0.96 2.44 1.23 2.73 1.94 3.79 2.71 (d) Intra-Keiretsu Purchases Ratio (1996) i Mitsui Mitsubishi Sumitomo Fuyo Sanwa DKB From members to GTC From members to non-GTC members 2.66 0.92 7.58 3.89 5.97 0.42 1.5 0.99 1.81 2.46 2.63 2.09 Sources: Toyo Keizai Shinposha (1994b) and Kosei Torihki Iinkai Jimukyoku (1998a, b). Note. The unit is %. The above survey reports, when conjoined by the states of cross-shareholdings among the six keiretsu analyzed by the previous subsections, suggest that the rates of the intra-keiretsu transactions (both sales and purchases) kept pace with those of cross-shareholdings among member 142 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. firms, including the GTCs. Although based on the survey from 1992, only Toyo Keizai Shinposha (1994b: pp. 48, 138) validates this viewpoint. Table 32 reports more detailed data showing the frequencies of three types of transactions (bilateral, unilateral, and none) under each of three possible forms of stockholdings among members (bilateral, unilateral, and none). The financial institutions, such as the main bank, are excluded from the data. Table 32. The Relations among the Three Types of Stockholding and the Three Types of Transaction (1992) Stockholding type Mitsui Mitsubishi Sumitomo Fuyo Sanwa DKB Cross-shareholding 32.90 56.33 87.50 25.00 11.74 15.24 bilateral trade 16.02 25.33 40.00 8.33 5.26 7.20 unilateral trade 12.12 20.00 29.17 9.33 3.51 5.24 no trade 4.76 11.00 18.33 7.33 2.97 2.80 Bilateral-stockholding 20.35 25.00 10.00 15.00 15.79 16.59 bilateral trade 2.60 5.00 2.50 3.33 4.72 5.12 unilateral trade 13.85 13.00 5.83 6.33 5.94 6.10 no trade 3.90 7.00 1.67 5.33 5.13 5.37 No-stockholding 46.75 18.67 2.50 60.00 72.47 68.17 bilateral trade 3.90 1.00 0.83 3.00 5.40 7.44 unilateral trade 17.75 7.67 0.83 13.67 16.33 15.85 no trade 25.11 10.00 0.83 43.33 50.74 44.88 Source: The data are from Toyo Keizai Shinposha (1994b: p. 138). Note. The financial institutions are excluded from the consideration. The value in each item is “the number of the inter-firm relations that is actually applicable for the case concerned divided by the number o f the possible inter-firm relations.” The unit is %. From Table 32, it is possible to make out two important points in light of the transaction cost approach, one, the common character across the six keiretsu, and the other, the way the six keiretsu start to deviate from each other, in regard to which combination of stockholdings or trade pattern is most preferred. First, it is readily discemable from the table that, no matter which keiretsu 143 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. is concerned, the following three combinations between the type of stockholdings and the trade pattern are the most favorably selected under each type of stockholdings: (cross-shareholding— bilateral trade), (unilateral stockholding—unilateral trade), and (no stockholding—no trade). This finding actually accords with the TCE hypothesis that, as transactions between two parties progress, their stockholdings, in parallel, become denser, and, ultimately, mutual. In other words, stockholdings and transaction are prone to develop side by side, from the null stage (no stockholding—no trade), via the middle stage (unilateral stockholding—unilateral trade), to the full- fledged stage (cross-shareholdings—bilateral trade). Second, such patterns detectable across all the six keiretsu notwithstanding, Table 32 also shows a stark difference, particularly between the Big Three and the Bank Groups, in regard to which one of these combinations is most often selected. As seen in Table 32, the Big Three, among all, Mitsubishi and Sumitomo were best characterized by the combinations of (cross-shareholdings—bilateral trade) and (unilateral stockholdings—unilateral trade), whereas the Bank Groups, the combination of (no stockholdings—no trade). Finally, the FTC’s survey reports showing the declines of both the intra-keiretsu sales and purchases ratios during 1981-96 are also congruent with the previous analysis of declining tendency of cross-shareholdings in the six keiretsu. The next chapter considers this issue in depth. 2.6 Concluding Remarks The aspect of the Japanese main bank system that has often been missed in the past literature is its role to facilitate the intra-keiretsu transactions. Based on this recognition, this chapter first set up the Japanese Keiretsu Model and examined its relevance by reviewing the history of how the keiretsu groups were formed in postwar Japan. Along with this historical review, the statistical comparisons were made for the six notable keiretsu groups, showing that cross-shareholdings, the proxies of the intra-keiretsu transactions, are, in general, much more progressed in the Big Three (the former Zaibatsu Groups) than in the Bank Groups. This result is chiefly ascribed to the fact that 144 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. the firms belonging to the Big Three were developed in tandem with each other through the diversification process during the prewar zaibatsu era (“the one-set principle”), and that the Bank Groups were, on the other hand, made in a somewhat ad hoc manner because the former smaller zaibatsu had had no historical background of trading with one another. Reflecting more intermeshed cross-shareholdings among the member firms, the role of the main banks of the Big Three is likely more pronounced than that of the Bank Groups. Among all, the role played by Mitsubishi Bank in the Mitsubishi keiretsu turned out to be the best fit with the Japanese Keiretsu Model through the analyses of the net-leakage ratio of the early 1960s and a set of two regressions analyses of the late 1960s. 145 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Endnotes for Chapter 2 1. This is often referred to as the “hold-up problem” in the TCE literature. The problem became even more serious when the sunk cost involved in such an investment was large. 2. In addition to the idyosincrasy o f goods under transaction, “bounded rationality” and “opportunism” are assumed as the essential trait o f human nature affecting transactions in the transaction cost econmics. 3. Klein (1988: pp. 201-2) pointed out that even if the contract is incomplete, there is the “self-enforcing range” o f the contractual relationship that limits the economic feasibility o f hold-up threats. Such range, Klein explains, mainly derives out o f the transacting parties’ taking care o f their reputation: by committing the hold-up action while utilizing the incomplete contract, the party might suffer from bad reputation in the future which functions against its trying to establish a new relationship with other parties. But there is some probability that market conditions drastically change so that it pays for one party to hold up the other regardless o f its loss o f good reputation— as Fisher held up GM under the original contract. 4. The term “keiretsu,” if used without any notice, is synonymous with the “horizontal keiretsu” and the “intermarket keiretsu.” Thus, these three terms are used interchangeably hereafter. The adjective “horizontal” is used in these terms because, in keiretsu, asset-specific investment and cross shareholdings are made between large firms with roughly the same size, as in the cases o f the former zaibatsu groups and the bank groups. The concept that forms a contrast with the horizontal keiretsu is the “vertical keiretsu,” which, as is discussed in depth in Chapter 3, is characterized by asset-specific investment between a large (parent) firm and a smaller or subsidiary firm, and by the exclusive ownership (stockholdings) o f the former over the latter. When the vertical keiretsu is referred in the present thesis, the adjective “vertical” is never be abbreviated. For the concepts o f the horizontal and vertical keiretsu, see Gerlach (1992) and Ito (1993: p. 178). 5. On the issue o f cross-shareholdings, the majority o f researchers have so far taken the stance that their primary objective is to avoid hostile takeovers. While it is certainly true that the cross-shareholdings were done occasionally for such a purpose, the present thesis does not place an importance to such an aspect o f the cross-shareholding practice. In this regard, the following statement by Gilson and Roe (1992: p. 34) is right to the point: (T)he motivation for acquiring cross holdings may differ from the function the holdings came to perform. The contractual benefits o f cross-holdings might not be obvious or easy to construct. A catalyst— fear o f takeovers— might have made them happen; thereafter the positive functions were seen, or survived. 6. Quite often, the intra-keiretsu trades are often done by the mediation o f the “general trading company (sogo shosha, GTC).” For the matter o f simplicity, the present discussion omits the GTC. 7. Another kind o f thereat is to sell off the stocks issued by a problematic firm on the open securities market. Gilson and Roe (1992: p. 21) pointed out that dumping stock seems to be the implicit threat in Japan. As reviewed closely in Chapter 3, dumping the stocks among keiretsu firms, especially in the mutual or retaliatory fashion between the bank and other non-financial firms, intensified in the late 1990s. From the TCE standpoint, this symbolizes the demise o f the keiretsu groups. 146 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 8. Note that the main bank is also held up by other members through the stockholdings and, thus, is subject to their monitoring. So far, there has been no incident reported where the main bank was intervened by non-financial member firms. But, in the late 1990 when the banks’ performance generally deteriorated, non-financial firms made the first move in dumping stocks issued by the main banks. 9. The following Table 33 shows the investment, as well as financing, buying, and selling relations among Sumitomo Bank, Sumitomo Corp., and Mazda Motors in FY 1998 (read Tables 33a and b from left to right to see these relations). The numbers in parentheses are the rankings. Although the data are more than 20 years later after Mazda experienced financial distress, they show that these three companies still maintained close transaction relations through the channels o f investment, financing, and buying and selling in the year. It can be readily understood from the tables that if Mazda went bankrupt, under such dense transaction relations, that would trigger huge residual losses to Sumitomo Bank and Sumitomo Corp. through both direct and indirect effects. Table 33. Investment, Financing, and Buying and Selling Relations among Sumitomo Bank, Sumitomo Corp., and Mazda Motors in FY 1998 Table 33a: Investment and Financing Relations Sumitomo Bank Sumitomo Corp. Mazda Sumitomo Bank - Stockholdings (13) Financing (2) Stockholdings (20) Financing (8) Sumitomo Corp. Stockholdings (6) - Mazda - Table 33b: Buying and Selling Relations Sumitomo Corp. Mazda Sumitomo Corp. - Selling (1) Buying (3) Mazda Selling (2) Buying (3) - Note. The data are from Toyo Keizai Shinposha, Kigyo Keiretsu Soran (2000). The numbers in parentheses are the rankings. 10. Although the theoretical ground differed from TCE, Aoki (1984b) showed that the amount o f money supply that banks provide to their client firms often deviates from that corresponding to the maximum stock values. When building up the Japanese Keiretsu Model, the present author gained valuable insight from Gerlach (1992), Kester (1991a, 1991b, 1992), and especially Gilson and Roe (1992). 11. The similar point is made by Okazaki (1999: pp. 67-9). 12. Takeda (1992: p. 56) pointed out that the inter-industrial economic development gained strength in the Japanese economy in the late 1890s. For the details o f the inter-industrial development o f the Mitsui, Mitsubishi, Sumitomo, and other smaller zaibatsu (Yasuda, Asano, and Nissan), see Okazaki (1999: pp. 58-98) and Hashimoto and Takeda (1992: Chapters 2, 3, and 4). 147 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 13. O f course, not all Mitsubishi firms stemmed from the shipment service and the coal-mining sector. For instance, Mitsubishi Estate (Mitsubishi Jiho,1937) developed from the real estate division (1906) o f Mitsubishi Shokai whose objective was mainly in managing the real estate o f the Marunouchi area, Tokyo, which was disposed o f by the government in 1890. Asahi Glass (1907) grew out o f the company set up from the personal interest o f Toshiya Iwasai. Mitsubishi Bank (1919) originated from the merger o f Dai-29 and Dai-119 banks in 1885. For all these details, see Okazaki (1999: pp. 74-7). Takeda (1992: pp. 57-68) showed that the Mitsubishi zaibatsu thrived on the investment in the coal mining and copper producing companies in the early 1920s. 14. In the Sumitomo zaibatsu, too, there were firms that had nothing to do with Besshi’s diversification. These firms were Nihon Plate Glass (Nihon Ita Garasu. Absorbed in the Sumitomo zaibatsu shortly after World War I), Osaka Kitaminato Co. (1919), Shikoku Chuo Electric Power (1934), Sumitomo Trust (1929), Sumitomo Life Insurance (1924), and Nihon Electric (NEC. somewhere in the 1930s). See Okazaki (1999: pp. 82-85) for the details. 15. The amendment o f the Commercial Code o f 1893 legalized the joint-stock system in Japan, thereby making these corporate reforms possible. See Okazaki (1999: p. 104 109) and Takeda (1992: p. 75) on this issue. 16. Before their holding companies became stock companies, the zaibatsu firms were all dependent upon the capital collected from the founding families. Once collected, the capital was not allowed for free disposal at their will (Takeda 1992: p. 78). This system is known as “Soyu-sei” or the collective contribution system. For example, article 28 o f the Mitsui Family Constitution states “regardless o f the amounts originally invested, the shares among our families are fixed at the following ratios: (1) 23/100 for the SoryO-ke (the eldest son’s family), (2) 11.5fl00 11.5/100 for the head family, and (3) 3.9/100 for branch families (Miyazaki 1966: p. 223).” According to Takeda (1992: p. 77), the collected capital was managed either by the holding companies (in the case o f Mitsui and Sumitomo) or by each’s subsidiaries (in the case o f Mitsubishi). 17. From 1915 through 1918, the average profit rates o f the shipment and shipbuilding industry and o f the steel industry were 72.3% and 29.5%, respectively. They were 2nd and 9th ranked in the top 16 highest-profit-rate industries during the period in Japan. See Hashimoto (1992: p. 92). 18. As is discussed later, these smaller zaibatsu were to become important constituents o f the three, bank- oriented keiretsu groups (Fuyo, Sanwa, and DKB) in the late 1960s and the early 1970s: Asano and Okura were to be incorporated in the Fuyo group; Suzuki and Iwai, in the Sanwa group; and Furukawa and Kawasaki, in the DKB group. Kuhara went bankrupt in 1926, and its assets were bought out by Gisuke Ayukawa, the founder o f the Nissan zaibatsu. The two major constituents of the Nissan zaibatsu, Hitachi Factory and Nissan Motors, were to be incorporated in the Sanwa and FuyO groups, respectively. 19. In 1921, the operation o f Furukawa Corp. was suspended by its parent firm, Furukawa & Co. (Hashimoto 1992: pp. 112-3). Suzuki Shoten went bankrupt together with its major lender, Taiwan Bank, when a huge financial crisis occurred in Japan in 1927 (Hashimoto 1992: p. 129; Yoshino and Lifson 1986: pp. 17-9). Asano started to suffer from the chronic loss in the late 1920s, which, itself was often caused by the easy-financing policy by Yasuda Bank. After disposing o f huge bad loans, Yasuda withdrew from its vested sectors such as electric machines, electric power, railways, and paper, and limited itself to financial business (Hashimoto 1992: p. 129). Hashimoto (1992: p. 128) pointed out that the major reason for the failure o f the new zaibatsu groups was their lack o f a decentralized decision-making system between the headquarters (holding company) and the factories (subsidiaries). 148 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 20. Again, in sharp contrast with the new zaibatsu groups, the Big Three could also thrive in the banking sector by carefully selecting their customers (Hashimoto 1992: p. 117). 21. Through the study o f the profit rates o f a broad range o f the Japanese industries over the period of 1932-36, Okazaki (1999: pp. 165-66) classified the following four types o f industries the zaibatsu firms had been operating: (1) leading industry: ceramics, (2) growing industries: metals, machines, and chemicals, (3) stagnant industries: mining, and real estate and warehouses, and (4) declining industries: finance, and commerce. 22. It is important to note, however, that Mitsui’s heavy and chemical industrialization was not so progressed as Mitsubishi’s and Sumitomo’s during the period. For instance, in the heavy and chemical industries during the period, Mitsui had only one subsidiary (Mitsui Chemical), whereas Mitsubishi and Sumitomo had four and six, respectively. Besides, among the top ten companies measured by paid-in capital for the period o f 1937-41, only two were from Mitsui (Mitsui Mining and Mitsui Corp.), whereas five companies were respectively from Mitsubishi and Sumitomo. Mitsui was severely damaged when Tokyo-Shibaura Electric gained independence from Mitsui and when Tokyo Electric merged with Shibaura Factory in July 1939. See Sawai (1992: pp. 153-55, 159,161-62,175). 23. “The Corporate Profits, Dividends, and Capital Accommodation Directive” (Kaisha Rieki Haito Oyobi Shikin Yuzo Rei, April 1939) authorized that dividends from firms capitalized at more than ¥200,000 be subject to the MOF’s regulation (Okazaki 1999: p. 111). The Directive was replaced by “The Company Accounts Control Directive” (Kaisha Keiri Tosei Rei, October 1940), which further tightened the MOF’s control by placing the requirement o f ministerial approval for the dividends that exceeded 8 percent o f capital or the rate o f previous year (Okazaki 1999: p. 114). “The Munitions Corporations Law” (Gunju Gaisha Ho, October 1943), on the other hand, authorized that the government or the association in control appoint a corporate president with no regard to shareholders (Okazaki 1999: p. 118-19). After the enactment o f the Company Accounts Control Directive, both the zaibatsu and non-zaibatsu firms alike started to reduce their dividend propensity, and the dividend ratios ceased to correlate with their rates o f profit (Okazaki 1999: p. 114). 24. Another centrifugal force was their holding companies’ transformation into stock companies (Mitsubishi & Co. Ltd. and Sumitomo & Co. Ltd. in 1937. Mitsui & Co. in 1940). Sawai (1992:p. 188) pointed out that their motivation was mainly in raising funds for inheritance tax by releasing part o f their stocks to the public. 25. Since 1940, Mitsui Bank consistently increased lending to its group firms, and, as a result, the lending balance came to exceed its deposits. Mitsui Bank and Sumitomo Bank also increased lending to their group firms. In the case o f Mitsubishi, group financing by the Mitsubishi’s financial institutions (bank, trust, insurance [marine, fire, and life]) exceeded more than 40% in 1944 (Sawai 1992: pp. 182-83). After the Munitions Corporations Law (October 1943) was enacted, however, the loan consortium system was replaced by the system o f designating specific financial institutions by which financing was exclusively taken on by the lead bank. Under this system, banks were forced to lend at the bidding o f corporate managers because the Munitions Corporations Law granted them full authority o f the designation. Consequently, banks’ monitoring became virtually ineffective. See Hoshi et al. (1991) and Teranishi (1993: p. 78) for the details. 26. Mitsubishi Corp. was divided into over 170 petite companies (Miyashita and Russell 1994: p. 37). 27. That the Finance Division o f the Economic and Scientific Section o f the GHQ seized the jurisdiction over the Japanese banks greatly contributed to keeping the banks almost intact because the Division had no interest in trustbusting (Miyashita and Russell 1994: p. 38). Besides, the Finance Division was often at enmity with the Antitrust Division set up inside the same Section (Johnson 1982: p. 174), which, according to Hashimoto (1995: p. 106), insisted on severing Japanese banks. 149 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 28. The Corporate Reconstruction Preparation Law was arranged in 1948 for the disposition o f bad loans arising from wartime finances. Bad loans were processed first by writing-off corporate clients’ capital up to 90%, and then by writing-off banks’ capital for the exceeding amount. 29. The first revision o f the Antimonopoly Law did not contribute to cross-shareholdings because cross shareholdings among the former zaibatsu firms were forbidden at the time o f the Peace Treaty. See Miyajima (1992: p. 245) and Teranshi (1999: p. 85). The Japanese government declared in July 1951 that the government had completed the whole zaibatsu dissolution process and dissolved the Holding Company Liquidation Committee (HCLC, mochikabu gaisha seiri iinkai) in the following day. 30. The member firms o f the Presidential Presidents’ Councils o f the Mitsui, Mitsubishi, and Sumitomo groups were not known for the 1951-58 period. To the best o f the present author’s knowledge, it was Keiretsu no Kenkyu kenkyu (1960) which first publicized the members o f these councils from 1959. 31. In the Sumitomo group, there was a rapid growth in cross-shareholdings, not only between Sumitomo Bank and the group firms, but among other group firms (Kikkawa 1992: pp. 264-5; Teranishi 1993: p. 85). Nippon Construction was developed from its predecessor, Sumitomo Land & Construction (Sumitomo Tochi KOmu, 1944) in 1945, and functioned as the de-facto general trading company for the Sumitomo firms (Kikkawa 1992: p. 270; Okumura 1976: p. 201; Toyo Keizai Shinposha 1994a: pp. 58-9). By the time Sumitomo Corp. was formed, the Sumitomo firms often relied on two Osaka traders, Itochu Shoji and Marubeni, both o f which sprouted out o f Daiken Construction (Daiken Kensetsu) under the administration o f LPECPE in 1949 (Miyashita and Russell 1994: p. 97). 32. Behind the reconstruction o f these general trading companies was the MITI which “helped rebuild the trading companies by issuing laws that authorized tax write-offs for the costs o f opening foreign branches and for contingency funds against bad debt trade contracts” (Johnson 1982: p. 206). Johnson (1982) pointed out that the MITI’s Industrial Rationalization Council (Sangyo GOrika Shingikai) called for “keiretsu-ization” o f trading companies by employing its licensing authority and ability to supply preferential financing, and that the Council winnowed about 2800 trading companies after the occupation down to around 20 big ones, each serving a bank keiretsu or a cartel o f smaller products. See Miyashita and Russell (1994: pp. 37-38) for a similar account. 33. Dore and Whittaker Imai and Komiya (1994: p. 10) pointed out that the term “keiretsu” used to have a limited meaning in the pyramidal structure o f subsidiaries and highly dependent suppliers and subcontractors grouped around a large manufacturing firm. Admittedly, the three corporate groups discussed in this section used to be called “enterprise groups” (kigyO shudan) in their earlier ages in Japan. But in the judgment o f the present author since the term “keiretsu” is fully mature matured to describe these groups, it is constantly adopted hereafter (In addition, the term was used in that way by Keiretsu no Kenkyu kenkyu published in 1960). 34. When Makoto Usami, a former President o f Mitsubishi Bank, visited foreign countries, he was occasionally asked an embarrassing question as to which firm is representative o f the Mitsubishi Group (Miyazaki 1966: pp. 220-5). This episode reports well on the mutual, horizontal nature o f the postwar Presidents’ Councils. 35. Miyazaki (1966: p. 25) reported that, despite the banks’ holding tight reins over the Presdents’ Councils, the balance o f power occasionally tilted toward the industrial firms when new technological development became the major concern o f the councils. 150 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 36. Top secret as they were, the contents o f the Presidents’ Councils were occasionally scooped by mass media, or even publicized by the councils themselves. They were mostly factional disputes regarding group policy or personnel affairs. Some examples are (a) the dispute over the election o f the chief executive o f Mitsubishi Petrochemical (1972. See Miyashita and Russell (1994: p. 62), (b) Mitsubishi Bank’s attempt to abandon the “three diamond logo” (in the 1970s. See Miyashita and Russell 1994: p. 65), (c) the battle between Sumitomo Light Metal and Sumitomo Chemical over their jurisdiction o f the aluminum industry (in the early 1970s), and (d) the Mitsukoshi scandal, in which President Shigeru Okada was forced to resign after piling up a huge inventory purchased from Michi Takehisa, the proprietress o f foreign-made products distributors (1982. See Gerlach 1992: p. 111-13). 37. The only one exception is, again, Mitsui’s D o f 1957, which was bigger by the negligible margin of 0.1% than that o f Mitsubishi. 38 Let us note that “the net leakage ratio” is always positive because both (C/A)— (B/A) and A are always positive. This also means that the keiretsu financing exceeded the amount o f investment in each o f the Big Three during the period. 39. There is one minor exception in this relation, as pointed out in (37). 40. This hypothesis is examined in the context o f the heavy and chemical industrialization in the next section. 41. This explanation is actually a part o f his higher-rank concept o f the “set control,” which refers to the “investment behavior aiming at control by keiretsu o f a complete set o f all industries related to one another” (Miyazaki 1980: p. 65), and is occasionally referred to as “keiretsu one-set principle,” the term coined by Miyazaki himself. This kind o f investment pattern had actually a close bearing on the members o f the Presidents’ Councils being selected from diverse industrial sectors with no duplication (see again Table 8). The issue o f the one-set principle is reviewed more in detail in the next section. 42. In a similar fashion Similarly, Okumura (1976: pp. 174-5) drew a line between the Mitsubishi and Mitsui zaibatsu, arguing that the former is best characterized by its inter-industrial development, while the latter by its jack-of-all trades. 43. Okumura (1976: p. 200) pointed out that Sumitomo’s top three companies had traditionally been Sumitomo Bank, Sumitomo Metal, and Sumitomo Chemicals and that Sumitomo Corp. was put in a weaker position in the Sumitomo group. In a the similar vein, Kikkawa (1992: p. 271) pointed out that Sumitomo Corp. was developed under the support o f the group firms. He explained that the main purpose o f the Sumitomo firms for opening Sumitomo Corp. was to reduce the transaction costs involved in the procurement o f raw materials and the sales o f their products. 44. With the observation that the heavy and chemical industrialization had been well advanced in Sumitomo, how can one explain its higher net leakage ratio (ranging from 15.1 to 24.9, the arithmetic average = 19.8) in Table 10? Okumura (1976: p. 198) gave a clue to this question: according to him, Sumitomo was traditionally characterized by Sumitomo Bank’s precedence over other Sumitomo firms. That propelled Sumitomo Bank into the intensive “Sumitomo-ization” activity through financing the non-Sumitomo firms. 151 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 45 In April 1949, the foreign exchange rate was determined at the fixed rate o f SI = ¥360. Before April 1949, differing exchange rates were applied across industries: for instance, compared with the above single exchange rate, many fiber-related industries adopted strong yen exchange rates, while many machine-related industries, weak yen exchange rates. Thus, once the singe exchange rate was introduced, some kinds o f machines (for example, sewing machines, watches, automobiles, radios, cameras, and marine vessels) suffered export difficulty (see Okazaki 1996: pp. 320-2). 46. The JDB played a significant role in financing the marine transportation and coal industries, as both industries suffered severe damages during the war and also required long-term investment (Kikkawa 1992: pp. 269-70). Kikkawa’s Table 6-7 (1992: p. 265) shows that the keiretsu firms belonging to these industries were supported by other members through the high rate o f stockholdings. 47. The central body o f planning the inter-industry coordination during the time was the MITI’s “Industry Rationalization Council” (sangyo gorika shingi-kai, founded in December 1949). See Okazaki (1996: pp. 326-36) for the details o f its activities. 48. Teranishi (1993: p. 83, pp. 87-8) pointed out that the MOF’s policy o f controlling the branch networks o f banks and o f introducing the criterion based on their deposit earnings together contributed to corporate groupings because such a policy restricted banks’ ability to rake in household savings, and, thereby, directed banks to focus more on securing corporate deposits. 49. Symbolic to the change in the function o f the GTCs is that it was not until the postwar period that the term “general trading company” (Sogo Shosha) became a part o f the Japanese business parlance. In other words, such a term was not common in prewar Japan (Okumura 1976: p. 136)). 50. The advantage o f this seemingly complex system is that by intermediating the transaction, the shosha eliminates the need for the seller to extend credit to some tiny firm it has never heard of; instead, it extends credit to a giant trading company. Similarly, the buyer is in debt not to the supplier but to a large, well-connected trading company which can manage the risk much better than the seller. (Miyashita and Russell 1994: p. 57) For this reason, the GTCs are often called “a kind o f quasi-insurance agency” (Miyashita and Russell (1994: p. 58). Because o f this quasi-insurance agency function o f the GTCs, there occasionally happens a case where “the plants o f supplier and purchaser are located adjacent to one another in the same industrial complex with direct feedthrough o f materials, and yet the trading company earns commission income as a financial intermediary in the transaction” (Miyashita and Russell 1994: p. 58). 51. The GTC’s credit creation function was even more valuable among both banks and firms, as they often hesitated to invest in exploring the distribution channels under the general lack o f liquidity during the high-growth era (Yamanaka 1989: p. 10, 87). The BOJ’s discount policy often supplemented the GTCs’ credit creation; it often discounted the promissory notes issued by both the GTCs and their owing customers, thereby supplying high-powered money twice for identical transactions (Yamanaka 1989: p. 84). 52. Although the year selected probably no longer falls into the typical high growth era, Yamanaka (1989: p. 13) reported that in 1973 six gigantic GTCs (Mitsui, Mitsubishi, Sumitomo, Marubeni, Nissho-Iwai, and Itochu) accounted for about 70% o f the total steel wholesaling, and 50% o f the total synthetic fibers wholesaling in Japan. 152 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 53. Under the LPECEP administration in 1948, Mitsubishi Heavy Ind. was severed into three smaller companies o f (new) Mitsubishi Heavy Ind., Mitsubishi Nihon Heavy Ind., and Mitsubishi Shipbuilding. After the severance, (new) Mitsubishi Heavy Ind. and Mitsubishi Shipbuilding, respectively, established their trade relations with Fuji Steel-Kinoshita Corp. and Yawata Steel-Iwai Corp. With these two new trade relations established, Mitsubishi Corp. could not restore the status of sharing the highest steel wholesaling to these two firms until they, along with Mitsubishi Nihon Heavy Ind., came into merger in 1964 (Suzuki 1993: p. 105). 54. Note that Osaka Shosen was actually an affiliate o f the Sumitomo keiretsu. Osaka Shosen placed the orders o f shipbuilding to Mitsubishi Shipbuilding because there was no shipbuilding firm in the Sumitomo keiretsu. This led the Sumitomo keiretsu to buy out the non-group machinery firms later on. 55. The heavy and chemical industrialization also lent impetus for the Sumitomo keiretsu to extend their trade channels to non-Sumitomo firms. Consequently, the primal materials produced by the group’s main constituents, such as Sumitomo Metal, Sumitomo Chemical, and Sumitomo Electric, came to be exported, via Sumitomo Corp., to other non-Sumitomo production firms, and Sumitomo Corp., accordingly, became less and less dependent on Sumitomo’s traditional products such as metals, nonferrous metals, electric machinery, and electric wires. See Kikkawa (1992: pp. 297-8), Okumura (1976: pp. 204-6), and Suzuki (1993: pp. 112-4). 56. The trade bias toward the products characteristic to the heavy and chemical industrialization was also true o f five trading companies in the Kansai area (Tomen, Nichimen, Gosho, Itochu, and Marubeni), which, as often referred to as “kansai gomen” in the Japanese language, traditionally engaged in the dealings o f the cotton-related products. In the process o f the heavy and chemical industrialization, they gradually reduced the sales volume o f their cotton-related products, while increasing those of machinery, steel, and chemical products. For details, see Hashimoto (1995: p. 107), Yamanaka (1995: p. 7), and Uchida (1976: pp. 343-4). During this period, Mitsui’s intra-group transactions were developed, but still delayed compared with Mitsubishi and Sumitomo. Suzuki (1993: p. 88) explained that that was the primary reason for Toshiba to stay out o f the official Mitsui membership. 57. Probably, as the reflection o f delay in the heavy and chemical industrialization, in the Mitsui keiretsu, there are some firms ranked in the table, which are appropriately categorized into the old type industry (Mitsui Mining) and the light industry (Toyo Cotton). 58. The firms categorized in one o f the following types are eliminated from the samples: (a) those engaging in the warehouse industry; (b) those engaging in the realty industry; (c) those engaging in the industries under scrapping, such as shipping and mining industries; (d) those engaging in the cement industry (Its primary role at the time was to inherit the resources from the mining industry.); (e) those spawned from other member firms, (I) those set up by the group investment (They were, in many cases, engaged in the booming industries such as oil and petroleum.); (g) those which constantly received the orders from the government (In 1960, only Nihon Denki (NEC) was categorized under.); (g) specifically, the following firms are eliminated from the samples: Mitsui Realty, Mitsui Warehouse, Mitsui Mining, Mitsui Shipping, Mitsubishi Mining, Mitsubishi Cement, Mitsubishi Petroleum, Mitsubishi Oil, Mitsubishi Realty, Mitsubishi Shipping, Sumitomo Realty, and Nihon Denki. 153 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 59. After World War II, a large number o f oil fields were developed in the Middle-East, contributing to bringing down the price o f petroleum. In 1958, the cost o f petroleum per calorie fell below that o f coal (Kikkawa 1992: pp. 275-6) and, after the liberalization o f the petroleum import, the energy switch from coal to petroleum immediately ensued in Japan (see the Table 34 below for the energy supply in Japan during 1955-1965). Table 34. The Energy Supply in Japan during 1955-1965 Year Petroleum Hydro-Powered Coal Others Sum Import Ratio 1955 20.2 21.2 49.2 9.4 100.0 (56,016) 24 0 1960 37.7 15.3 41.5 5.5 100.0 (93,749) 44.2 1965 58.4 11.3 27.3 3.0 100.0(165,614) 66.2 Source: Shibagaki (1975: p. 72), originally from Tsusansho. 1973. Nihon no enerugi mondai (the MITI. The energy problem o f Japan). Note. Unit: %, 1,000 kl. 60. It is also reported that the foundations o f Sumitomo Light Metal and “new” Mitsubishi Heavy Ind. were partly due to the group members’ preparation for the capital liberalization o f 1965. See Kikkawa (1992: p. 280). 61. In contrast to Sumitomo Cement, Mitsubishi Cement was not joined in Nimoku-kai. This is considered primarily due to its heavy technological reliance on Onoda Cement, a non-affiliate with the Mitsui keiretsu. 62. The firms characterized by extremely high intra-keiretsu stockholdings ratios in 1966— the firms such as Mitsui Petrochemical (55.0%), Mitsubishi Cement (78.2%), Sumitomo Light Metal (45.9%), and Sumitomo Cement (28.7%)— were built largely from investment by other member firms: most o f the stocks issued by these new firms were accepted first by other, long-established members until they were acknowledged to be grown enough to be independent. On their sides, the new firms took part in cross-shareholdings only in limited ranges. See Keizai Chosa Kyokai (1967). 63. This time, the newly set-up firms must be eliminated from the samples besides those specified in endnote 58 because they are judged as very unlikely to take part in the intra-keiretsu transactions on a regular basis. More specifically, the following firms are eliminated from the samples this time: (a) Mitsui keiretsu: Hokkaido Tanko Kisen, Mitsui Mining, Mitsui-Miike Seisakujo, Mitsui Norin, Mitsui Petrochemical, Mitsui Realty, Mitsui Seiki, and Mitsui Warehouse; (b) Mitsubishi keiretsu: Mitsubishi Oil, Mitsubishi Cement, Mitsubishi Mining, Kirin Beer, Mitsubishi Realty, Nihon Yusen, Mitsubishi Warehouse; and (c) Sumitomo keiretsu: Sumitomo Coal-Mining, Nihon Electric, Sumitomo Cement, Sumitomo Warehouse, and Sumitomo Realty. 64. In 1965, the Mitsubishi keiretsu further strengthened the Presidents’ Council’s (Kin’yo-kai)’s function by equipping itself with the authority to appoint presidents o f its member firms (Miyazaki 1976: p. 255; Okazaki 1992: pp. 312-20). 65. Dai-ichi Bank and Nippon Kangyo Bank merged into Dai-ichi Kangyo Bank in 1971. 154 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 66. It has been often pointed out that the Fuyo Group maintained stronger interior cohesiveness than other two Bank Groups because Fuyo’s firms were mostly carry-overs from the prewar Yasuda zaibatsu. During the time when Zenjiro Yasuda (1938-1921), President o f Yasuda Bank, took the reins over the Yasuda zaibatsu, Yasuda zaibatsu had already established long-term financial relations with the Asano, Nezu, and Okura zaibatsu. Also, its financial relations with the Nissan and Mori zaibatsu started in the wartime period (Toyo Keizai Shinposha 1994b: p. 65). 67. Caves and Uekusa (1976: p. 63) described the formations o f the Bank Groups as follows: “The smaller and less complete o f the prewar groups did not reappear in their previous form. However, three major groups have emerged centered on giant banks...especially, Fuji (successors o f the Yasuda zaibatsu), Dai-ichi, and Sanwa.” 68. Later in 1969, Nippon Rayon and Nichibo were merged into Unitika under strong support from Sanwa Bank. 69. Maruzen was one o f the oil companies permitted to enter the oil industry in the First Domestic Oil Production Plan by the MITI in the late 1950s. Other companies permitted were Nihon Petrochemical, Mitsui Petrochemical, Sumitomo Petrochemical, and Mitsubishi Oil. 70. Later on in the early 1960s, Maruzen Oil fell into a financial crisis (Okumura 1965: p. 235; Sanwa Bank 1974: pp. 424-30). After the crisis, Sanwa Bank set up a new policy o f accelerating the grouping (Okumura 1965: p. 235). 71. In May 1964, Marubeni-Iida simplified its name to “Marubeni.” 72. Although Itochu was the main GTC for Kawatetsu, its main bank was Fuji Bank at the time. 73. It should be noted, however, that the rates o f cross-shareholdings within the Furukawa and Kawasaki Groups were respectively high in the earlier postwar period (Okumura 1976: p. 221). The cross shareholdings between these two groups were, on the other hand, not so progressed during 1959-64 (Suzuki 1993: p. 98). That the Furukawa Group often relied on the financing from Mitsubishi Bank also contributed to weaken the group ties within the DKB Group (Okumura 1976: p. 229). 74. Note that Yokohama Rubber and Nihon Zeon were the members o f the Furukawa Group. Dai-ichi Kangyo Bank considered the merger among the three firms as impending and, thus, tried to promote cross-shareholdings among them. 75. Also, Itochu’s successfully intermediating the merger between Nissan Motors and General Motors in 1973 contributed to establishing its position in the DKB Group (Okumura 1976: p. 229; Suzuta 1976: p. 146). 76. It is well known that Ishikawajima-Harima had long been affiliated with Toshiba o f the Mitsui Group. Since 1991, Ishikawajima-Harima has been an official member o f both Nimoku-kai and Sankin-kai. 77. The sharp decline o f Sanwa’s average stockholdings ratio is primarily due to the increase o f its member firms: from 1970 to 1981, the number o f Sansui-kai increased from 22 to 40. 78. Note that trust banks and insurance companies are excluded from the analysis. 155 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 79. Kigyo Keiretsu Soran's (various issues) reports are flawed in one respect: they only list the twenty highest member stockholders for each keiretsu, no matter how many members participate in it. Now note that “the ratio o f the non-fmancial members’ stockholdings with the main bank” and “the average ratio o f the non-financial firms’ stockholdings with the main bank” in this subsection are derived under such a restriction. The two variables introduced in the next subsection (“the ratio o f the non-fmancial firms’ stockholdings relation with the GTC” and “the average ratio o f the non-fmancial firms’ stockholdings with the GTC) were also derived under this restriction. 80. Such tenuous cross-shareholdings between Sumitomo Bank and its affiliates are considered one o f the main factors that made Sumitomo Bank quite vulnerable to the bubble economy that occurred in the late 1980s. This issue is taken up in the next chapter. 81. The term, “the members,” in this subsection refers to the main bank and the non-financial member firms. Trust banks and insurance companies are excluded from the members. Also, “Kawasaki Corp.” o f the DKB Group is eliminated from the analysis, as it specializes in the steel trade with three other firms o f the Kawasaki Group (Kawasaki Steel, Kawasaki Heavy Ind., Kawasaki Steamship) and is hardly qualified as a GTC. 82. Note that the unified data in Figure 9 are introduced just for eliminating the complexity involved in the comparison among the ten GTC’s, and that they should be viewed with the reservation that those GTC’s whose data are unified in Figure 9 are not actually merged. This argument holds true for other unified data in the following discussion. 83. While the interest rates were strictly regulated by the MOF during the high growth period, there was still room for banks to manipulate them via imposing the “compensating deposit” to their customer firms. Typically, the compensating deposit is made o f the two types, the one, “butsumi yokin,” requiring the customers to place a deposit for discounted bills, and the other, “ryodate yokin," requiring them to place a deposit for the loan made in exchange for the bills (see, for example, Tsuda 1993: pp. 160-2). The compensating deposit was a recurrent headache to the business firms since 1963 (Miyazaki 1966: pp. 135-6), but gradually faded away as the power balance between banks and their customers turned unfavorable to the former in the late 1970s (Rosenbluth 1989: p. 158, 166). 84. The ratios o f the main bank’s stockholdings with the GTC are directly replicated from Kigyo keiretsu soan. As Figure 20 shows, all six main banks continued possessing member GTCs’ stocks more than 5% by the mid-1980s, which apparently runs counter to the Antimonopoly Law. Candidly, the present author has no idea as to why this occurred. 85. In the Big Three keiretsu, the firms which had been less reliant on the finance from the main banks’ are those gaining relatively independent positions from the group. Notably, they are Mitsukoshi (the Mitsui group), Toyota Motors (the Mitsui group), and Asahi Glass (the Mitsubishi group). 86. The increased corporate borrowing was mostly the response to the BOJ’s incessant reduction o f its official discount rate after 1986. Also note that the balances o f equity financing issued by Mitsui and Fuyo’s firms were smaller than others. For instance, during 1992-95, the balance o f equity financing (the sum o f the convertible and warrant-attached bonds that failed to transform into equities) in each keiretsu was the following: 2.1843 trillions yen (Mitsubishi), 0.8117 trillion yen (Mitsui), 1.7155 trilion yen (Sumitomo), and 1.1176 trillion yen (Fuyo). See Miyazaki (1992: pp. 218-9). 87. See endnote 86 above for the amounts o f equity financing issued by Mitsubishi, Sumitomo, and Sanwa. 156 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 88. Sankin-kai members’ borrowing from non-main banks was heavier than the non-financial members o f other five keiretsu, and they were, thus, very susceptible to sudden credit withdrawal from other non-main banks (the ratios o f borrowing from non-main banks in 1992 were the following for the six keiretsu. See Toyo Keizai Shinposha (1994b: p. 88, 153) and Kosei Torihiki Iinkai Jimukyoku (1998b: p. 68, 162) for more details: the Mitsui firms: 51.52%, the Mitsubishi firms: 50.85%, the Sumitomo firms: 57.82%, the Fuyo firms: 51.54%, the Sanwa firms: 51.93%, the DKB firms: 55.48%). Considering that Dai-ichi Kangyo Bank’s high lending ratio (i.e., lending/total assets) to its member firms, Sankin-kai members’ heavy reliance on the non-main banks was mainly ascribed to Dai-Ichi’s incapacity to sustain the huge number o f Sankin-kai members. See the following table for the lending rates by the six main banks during 1988-92. The rates are calculated by using the data from Kigyo keiretsu soran. Table 35. Lending Rates by the Six Main Banks during 1988-1992 Main Bank 1988 1989 1990 1991 1992 Mitsui 0.0207 0.0177 0.0156 0.0109 0.0103 Mitsubishi 0.0148 0.0127 0.0113 0.0126 0.0109 Sumitomo 0.0120 0.0077 0.0069 0.0075 0.0079 Fuyo 0.0120 0.0100 0.0086 O.U09(> 0.009' Sanwa 0.0183 0.0169 0.0148 0.0171 0.0170 DKB 0.0196 0.0174 0.0150 0.0190 0.0191 89. Unfortunately, the FTC suspended this survey research after 1994. 157 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Chapter 3 T h e C o n t e m p o r a r y C r isis o f t h e J a p a n e s e M a in B a n k S y s t e m f r o m t h e In d u s t r ia l O r g a n iz a t io n a l P e r s p e c t iv e 3.1 Introduction The main purpose of this thesis is to explain the contemporary plight of the Japanese main bank system. In order to do this, however, this chapter focuses more on the industrial side, rather than the financial side. This paradox stems mainly from the present thesis’ adoption of the Japanese Keiretsu Model, which treats the main bank as only one constituent of a keiretsu group. As confirmed in the previous chapter, one equally crucial factor for enabling the intra-keiretsu transaction is the one-set principle adopted in each keiretsu, which, albeit in varying degrees across the six keiretsu, was molded in the process of the heavy and chemical industrialization. From today’s vantage point, it is relatively easy to infer that such a co-prosperous relation between the main bank and other member production firms through the one-set principle was thwarted significantly by the oil crises of the 1970s. This chapter, thus, explores the issues of the oil crises, first reviewing their macroeconomic impact on the Japanese economy and then analyzing specifically how they affected the intra-keiretsu transaction of each of the six keiretsu. Finally, based on the analytical results thus obtained, the analysis is directed to the main banks of the six keiretsu to explain their varying responses to the financial liberalization that started in the early 1990s. 158 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 3.2 The Oil Crises: The Transformation of the Industrial Structure and the Rise of the Vertical Keiretsu As previously witnessed, the Big Three keiretsu (the Mitsui, Mitsubishi, and Sumitomo keiretsu) significantly contributed to Japan’s high economic growth since the early 1950s, deploying their member firms positioned across wide-ranges of industrial sectors in the mode of the “one-set principle,” while systematically consolidating their frameworks (the main bank system, the asset- specific investment via the GTC, and cross-shareholdings among members) under the process of economic growth. In particular, after the “energy revolution” of the early 1960s, which replace petroleum for coal as the primary domestic energy source, their intra-keiretsu exchanges were considerably accelerated through the extensive technological spillover effects over the heavy and chemical industries, thereby furthering their “keiretsu-ization” toward the most ideal form as demonstrated in the Japanese Keiretsu Model. Recognizing clearly the gains accruing from such increasing inter-corporate exchanges, Sanwa and Fuji Banks, on the other hand, followed suit of the Big Three by formally setting up their own keiretsu (the Fuyo Group (1966), the Sanwa Group (1967)) by remobilizing their former affiliates and independents. Such concurrence of the high economic growth and the development of the keiretsu groups, however, gradually came under pressure, as the fierce market force intervened in the Japanese economy through the Kenney Round (1964-67) and the dissolution of the Bretton Woods fixed-exchange rate system (1973). The coup de grace was the first oil crisis of 1973, which brought the rapport into a complete stalemate: being heavily dependent on imported oil for some two-thirds of its primary energy needs, the Japanese economy was extremely vulnerable to such an energy crisis. A fourfold increase in the oil price brought about by the crisis exacerbated Japan’s international balance of payments and economic growth which, in 1974, declined for the first time since the postwar period started (see Table 36 for the principal economic indicators during this period). 159 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 36. Trends of Principal Economic Indicators, 1970 - 1983 Annual growth rate of Year Real GNP Private final consumption expenditure (%) Private investment in new plants and equipment (real) (%) GNP deflator (%) Wholesale prices (%) Consumer prices (%) Unemploy ment rate (%) Current balance of payments ($U.S. million) 1970 1 9.9 6.9 17.7 7.3 3.6 7.7 1.1 1.970 1971 . 4.7 5.9 -0.4 5.2 -0.8 6.1 1.2 5,797 1972 9.0 9.5 7.4 5.2 0.8 4.5 1.4 6,624 1973 O O : 0 0 ■ 9.3 16.0 11.9 15.7 11.7 1.2 -136 1974 -1.2 -0.7 -7.4 20.6 31.6 24.5 1.4 -4,693 1975 i 2.4 4.1 -3.2 7.8 3.0 11.8 1.9 -682 1976 : 5.3 3.4 3.1 6.4 5.0 9.3 2.0 3,680 1977 ! 5.3 3.8 2.2 5.7 1.9 8.1 2.0 10,918 1978 5.1 4.7 6.6 4.6 -2.6 3.8 2.2 16,534 1979 i 5.2 5.9 7.7 2.6 7.3 3.6 2.1 -8,754 1980 : 4.8 1.3 3.0 2.8 17.8 8.0 2.0 -i'i.'4h 1981 ‘ 3.8 0.8 3.5 2.7 1.4 4.9 2.2 4,770 1982 1 3.3 4.2 2.5 1.7 1.8 2.7 2.4 6,850 1983 ! 3.0 1 .3 0.5 -2.2 1.0 2.7 20,799 Sources: Economic Planning Agency (Japan), Annual Report on National Accounts, 1984a, March (Tokyo: Government Planning Office); and Economic Planning Agency (Japan), Japanese Economic Indicators, 1984b, various issues, monthly. Despite unprecedented economic turbulence in the postwar period, the solution was principally delegated to the private sectors (Imai 1982: p. 58; Uekusa 1987: p. 481, 486). In particular, at the time, the MITI’s influence over the Japanese industries, through its ability to enforce cartel-formation and administrative guidance (fixed investment adjustment), was no longer dominant as before because of its paving the way for the import liberalization and shifting the policy priority away from maximum economic growth to the environmental issues (Dore 1986: p. 14, 128; Johnson 1982: pp. 275-304; Uekusa 1987: pp. 475-9, pp. 482-3; Uekusa and Caves 1976: p. 6).1 The MITI did intervene in the economy under “the Law on Temporary Measures for the Stabilization of Specified Structurally Depressed Industries” (1978), but in a limited way: the law designated only a handful of industries as “structurally depressed” which were allowed to take R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 2 collective actions for capacity reduction and to be exempt from the antimonopoly legislation. Besides, it did not take any of its measures such as import restrictions and government subsidies, -5 from concern about foreign criticism (Uekusa 1987: p. 492). With virtually no governmental support under the skyrocketing energy costs and the militant labor groups crying out for salary increases, Japanese firms were compelled to convert their technology-related strategy from purchasing from abroad to inventing their own for the energy conservation and labor-saving purposes in line with “streamlined management” (genryo keiei). In attempting to make a breakthrough in the technology-related areas, their research and development (R&D) expenditures increased from 1.91 percent of GNE (Gross National Expenditure) in 1965 to 2.45 percent in 1975 to 2.52 percent in 1979 (Kosei Torihiki Iinkai Jimukyoku 1992), while their average annual growth rate of payments for imported technology declined from 21.0 percent in the 1960s to 6.0 percent in the late 1970 (Uekusa 1987: p. 490).4 Among all, as shown in Table 37, the ratios of R&D expenditures over sales were higher in chemicals, electrical machinery, transport machinery, and precision machinery. Apparently, the last three of those industries are the primary sectors of the so-called “knowledge-intensive industry” which, in agreeing with the advent of the low growth era, refrained from excessive use of basic materials in the production process and, instead, promoted the realization of the high-added values through highly developed technologies invented by R&D. With successful R&D efforts in the areas of microelectronics, mechatronics, and biotechnology, such technologies as high-quality integrated circuits, microcomputers, numerical control machines, industrial robots, new industrial materials and medicines, these industries rapidly increased their market shares, finally swinging the balance against the heavy and chemical industries (mining, chemicals, petroleum and coal products, nonmetalic mineral products, and metal products) by 1980 (see Table 38). 161 R eproduced with perm ission o f the copyright owner. Further reproduction prohibited without perm ission. Table 37. Intramural Expenditure on R&D as a Percentage of Sales, by Industry, FY 1976 -1981 Industry 1976 1977 1978 1979 1980 1981 1982 All industries 1.42% 1.48% 1.57% 1.49% 1.48% 1.62% 1.78% Agriculture, forestry, and fisheries 0.24 0.31 0.60 0.45 0.17 0.26 0.27 Mining 0.57 0.50 0.54 0.48 0.52 0.46 0.64 Construction 0.48 0.53 0.42 (1.40 0.46 0.37 0.43 Manufacturing 1.64 1.70 1.82 1.71 1.73 1.91 2.15 Food 0.49 0.50 0.51 0.51 0.58 0.55 0.63 Textile mill 0.66 0.56 0.77 0.82 0.77 1.09 1.13 Pulp and paper 0.47 0.46 0.49 0.42 0.41 0.43 0.52 Publishing and printing 0.46 0.41 0.36 0.27 0.26 0.21 0.39 Chemical products 2.39 2.62 2.71 2.54 2.55 2.37 3.05 Industrial chemicals 1.69 1.87 1.92 1.71 1.85 2.01 2.17 Oils and paints 2.40 2.71 2.73 2.17 2.48 2.56 2.6d Drugs and medicines 5.05 4.84 5.00 5.53 5.45 5.85 5.56 Other chemical products 2.38 3.12 3.03 2.38 2.19 3.03 3.43 Petroleum and coal products 0.18 0.23 0.27 0.18 0.30 0.18 0.20 Rubber products 2.25 1.96 2.(in 2.44 2.10 2.33 2.47 Clay and stones 1.40 1.22 1.29 1.27 1.30 1.39 1.64 Iron and steel 1.02 1.11 1.08 1.04 1.14 1.30 1.50 Nonferrous metals 0.96 1.01 1.00 0.87 1.03 1.36 1.57 Fabricated metals 1.00 1.18 1.08 1.28 1.15 1.22 1.43 General machinery 1.79 2.01 1.93 1.85 1.90 2.10 2.34 Electrical machinery 3.66 3.61 3.74 3.55 3.71 4.06 4.52 Electrical appliances 3.49 3.49 3.59 3.19 3.35 3.80 4.17 Electronic equipment 3.80 3.71 3.89 3.91 3.94 4 21 4.72 Transport machinery 2.08 2.27 2.44 2.37 2.34 2.62 2.69 Motor vehicles 2.20 2.32 2.60 2.51 2.38 2.82 3.02 Others 1.76 2.12 1.90 1.85 2.15 1.94 1.67 Precision machinery 2.37 2.91 3.15 2.96 3.02 3.47 3.07 Other manufacturing 1.24 1.15 1.16 0.91 1.16 1.11 1.30 Transportation, communication, and public utilities 0.27 0.33 0.35 0.40 0.32 0.36 0.32 Source: Japan, Office of the Prime Minister, Bureau of Statistics, Report on the Survey o f Research and Development, 1982 (Tokyo: Nihon Tokei Kyokai, 1983), p. 59, as cited in Uekusa (1976: p. 491). 162 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 38. Change in Interindustry Gross Output Structure in the 1970s (Gross Output as 1975 Price = 100) Sector/industry Proportion of gross output Growth of index of gross output (1970= 100) 1970 1975 1980 1975 1980 Basic material sector Agriculture 3.( 2.9% 2.3% 102 99 Forestry and fishery 1.3 1.0 1.0 93 117 Mining 0.7 0.5 0.6 82 137 Textile 3.5 2.8 2.4 102 109 Lumber and wood products 1.9 1.6 1.4 106 111 Pulp and paper products 1.0 0.9 0.9 110 134 Chemicals 3.5 3.2 3.6 116 158 Petroleum and coal products 2.7 2.9 2.4 133 138 Nonmetalic mineral products 1.8 1.5 1.6 101 137 Basic metals 7.3 6.5 6.5 111 138 Metal products 2.2 1.9 2.0 109 143 SUBTOTAL 29.6% 25.8% 24.7% 109 130 Processing and assembly sector Food 5.7% 5.7% 5.2% 126 141 Leather and rubber goods 0.6 0.5 0.5 104 136 General machinery 3.8 3.9 4.3 127 175 Electric Machinery 1.8 1.4 1.9 96 162 Transport equipment 4.5 4.5 5.1 127 179 Other manufacturing products 1.5 1.3 1.4 112 152 Construction 10.3 10.3 9.5 125 143 SUBTOTAL 28.2% 27.6% 28.1% 123 154 Information sector Information goods 1.7% 1.2% 1.8% 86 165 Information equipment 2.3 2.3 3.9 123 256 Information media 3.5 3.0 2.8 108 123 Information services 4.3 4.1 4.7 120 172 Social services 4.1 5.3 4.8 162 183 SUBTOTAL 15.9% 15.9% 18.1% 125 175 Service sector Public utilities 1.9% 2.0% 2.1% 133 170 Wholesale and retail 8.8 9.0 9.0 128 159 Real estate 3.3 4.8 4.2 181 196 Transportation 4.1 5.9 5.8 182 222 Other services 5.9 6.8 6.4 145 167 Miscellaneous 2.3 2.2 1.6 118 108 SUBTOTAL 26.3 30.7 29.1 146 172 TOTAL 100.0% 100.0% 100.0% 125 172 Source: Uekusa (1987: p. 474). 163 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. The quick industrial transformation thus achieved was one of the main factors that enabled Japan to resolve its economic disequilibrium (the domestic unemployment and the deficit in the international balance of payments) within three years and to effectively cope with the second oil crisis of January, 1979.5 Despite such a buoyant macroeconomic performance, the industrial transformation was greeted with mixed feelings among industrial circles, then involved in the painful readaptation process toward a new low growth regime. The most troubled in such a re-adapting process were, obviously, those severely hit by the two oil crises, that is, those heavily reliant on imported oil and the related sectors of the heavy and chemical industry, the majority of which were typically members of various keiretsu groups. Severely damaged by the oil crises as oil importers, their GTCs too had become embroiled in hardship of the so-called “winter years of the GTCs,” trying to locate their new market niche in the post-oil crisis economy.6 The lesson they learned in the years of struggle was that they had been losing profits, not only because of the sudden rise in the oil price, but because of the changing economic conditions that no longer allowed for their traditional way of earning profits with basic materials such as steel and chemicals (Yamanaka 1989: p. 15). For the data on the gross profit rates of nine GTC during 1973-82, see Table 39. 164 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 39. The Gross Profit Rates of Nine GTCs (1973-1982) General Trading Company 1973 1974 1976 1978 1980 1982 Mitsubishi Corp. 2.08% 2.10% 1.97% 2.00% 1.62% 1.43% Mitsui Corp. 2.15 2.06 1.72 1.79 1.63 1.72 Itochu 2.63 1.96 1.93 1.91 1.64 1.29 Marubeni 2.83 2.36 2.22 2.07 1.61 1.44 Sumitomo Corp. 2.16 1.97 1.92 1.88 1.59 1.62 Nissho-Iwai 2.45 2.33 2.15 2.15 1.77 1.57 Tomen 3.04 2.80 2.37 2.21 1.97 1.76 Kanematsu-Gosho 2.26 2.04 2.07 2.11 1.99 1.68 Nichimen 2.82 2.28 2.55 2.63 2.19 1.93 Average 2.39 2.21 2.01 2.00 1.68 1.54 Source: Yamanaka (1989: p. 26). Table 40 substantiates this, showing how much investment in one sector affected those in other sectors (the correlation coefficients of investment across industries) for the two sets of period demarcated by 1973, the one, 1966, and the other, 1973-85. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 40. Correlations of Investment across Industries Industry Food Textiles Pulp & paper Chemic. Iron & steel Non- ferrous m etals Fabric’ d m etals G eneral machin. Electr. machin. Transport. Equipm. Electric. Service 1966 to 1972 Food 1.00 Textiles -0.14 1.00 Pulp& paper 0.06 0.50 1.00 Chem icals 0.47 0.49 0.48 1.00 Iron&steel -0.14 0.15 0.10 0.22 1.00 Non-ferrous m etals 0.46 -0.15 0.27 0.46 0.30 1.00 F abricated m etals 0.32 0.40 0.22 0.39 0.29 0.15 1.00 G eneral m achinery 0.31 0.51 0.37 0.66 0.46 0.30 0.75 1.00 Electrical machinery 0.23 0.51 0.31 0.59 0.61 0.37 0.67 0.84 1.00 Transportation equipm ent 0.45 0.46 0.53 0.46 0.40 0.30 0.56 0.60 0.66 1.00 Electricity 0.16 0.40 0.33 0.42 0.02 0.12 0.02 0.12 0.38 0.27 1.00 Service -0.30 0.24 0.06 0.10 -0.07 -0.20 -0.28 -0.09 0.19 -0.30 0.45 1.00 A v e r a g e 0.17 0.31 0.29 0.43 0.21 0.22 0.32 0.47 0.49 0.40 0.25 -0.02 1973 to 1985 I Food 1.00 Textiles 0.12 1.00 Pulp&paper 0.21 0.13 1.00 Chem icals 0.03 -0.11 0.33 1.00 Iron&steel -0.12 0.08 -0.29 -0.04 1.00 Non-ferrous m etals 0.06 0.05 0.15 0.20 -0.14 1.00 Fabricated m etals 0.06 0.13 0.18 0.02 -0.28 0.18 1.00 G eneral m achinery 0.20 0.28 0.10 0.12 -0.31 0.07 0.32 1.00 Electrical machinery 0.16 0.30 -0.04 -0.16 -0.25 0.03 0.56 0.52 1.00 Transportation equipm ent 0.20 -0.17 -0.01 -0.03 -0.21 0.20 0.24 0.36 0.51 1.00 Electricity -0.12 -0.05 0.02 -0.37 -0.05 -0.50 0.14 0.04 0.16 0.01 1.00 Service 0.18 0.38 0.02 -0.10 -0.06 -0.31 0.26 0 34 0.53 0 22 -0.03 1.00 A v e r a g e 0.09 0.10 0.07 -0.01 -0.15 0.00 0.17 0.19 0.21 0.12 -0.07 0.14 Source: Y oshikaw a (1995: p. 207). R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. By comparing the average coefficients values of each industry, it is readily observed that (1) the average coefficients of all but the service sector (+0.16) decreased in the second period, and that (2) the decreases in the value were drastic, in particular the sectors of chemicals (-0.44), iron (- *7 0.37), and electricity (-0.28). These findings suggest that, after the first oil crisis, technological spillovers across industries dramatically decreased, not only in the chemicals sector (including the petrochemical sector), but also in the other two sectors representated in the heavy industry, iron and electricity. The widespread stagnation of the heavy and chemical industries, brought about by the industrial reorganization after the first oil crisis, thus became the major hindrance against the GTCs to earn the traditional way of intermediating the heavy and chemical products across industries. Other members of the keiretsu groups were not immune from this problem, either, as they are typically positioned in the heavy and chemical industries. Their solid inter-corporate ties frayed as their interior joint actions lost economic justification.8 “During the rapid growth period,” noted Imai (1982: p. 57), “it was possible for these informal organizations [the Presidents’ Councils] to coordinate growth plans of member firms because the vast majority of potential conflicts were resolved by demand growth itself.” However, after the experience of the two oil crises, already established firms that had created new-products departments and those that set up joint ventures with other firms entered fields that were experiencing new growth. In this way, the groups of traditional enterprise complexes lost their economic justification and reshuffuled to adapt to changes in the business environment. Member firms of each group are now pursuing their own objectives, choosing between internal and interfirm organization based on economic cost comparison with alternate situation. (Imai 1982: p. 59) Rising streamlined management, prevailing across industries, also contributed to further changing the inter-corporate relations of the Japanese firms in a distinctive way. That is, large manufacturing firms, whether those of the six gigantic keiretsu or independents, started to intensively spin off some of their departments or sections as small and medium-sized subsidiaries, and to accelerate subcontracting transactions with them (Aoki 1987: p. 282; Imai 1992: p. 223; 167 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Uekusa 1987: p. 503; Watanabe 1999: pp. 214-5). Between 1966-1976, for example, the number of subcontracting small and medium-sized firms in the manufacturing sectors increased from 300,000 to 370,000 (Yaginuma 1993: p. 8).9 It was often the case that, by setting up the subsidiaries, large firms aimed to keep employees as homogeneous as possible, in terms of career prospects, salary payments, and labor conditions, at each firm level, thereby exercising efficient labor management and securing access to the low-cost labor in their subsidiaries (Aoki 1987: p. 284; Caves and Uekusa 1976: p. 70; Shimotani 1996: pp. 123-4.1 0 The subcontracting relation, thus established, is often referred to as the “vertical keiretsu,” vis-a-vis the “horizontal keiretsu,” since the former is best characterized by transactions principally initiated by the parent’s direction to its subsidiaries, which themselves are smaller in size, whereas the latter, in contrast, the reciprocal trade association between large firms, almost equal in size, as in the case of the six gigantic keiretsu (see, for example, Gerlach [1992] for the usage of the terminology).1 1 Spinning-off subsidiaries (bunsha-ka) or “vertical keiretsu-ization” brought about a great impact on firms’ shifting their inter-firm transactions more toward those between parent firms (large firms or headquarters or subcontractees) 12 and their subsidiaries from those among large firms. Consequently, transactions became much more fine-tuned and asset-specific than those among large firms, as such transactions were mostly made for satisfying the specific orders placed by parent firms to their subsidiaries. This trend was further accelerated, as parents spawned more subsidiaries, and those subsidiaries, in turn, brought forth their own subsidiaries in a pyramidal fashion. Through such a process of multiplying subsidiaries, the parent firms came to acquire the status of “holding companies” (mochikabu gaisha), which, through their ownership, partook in the governance over their subsidiaries in such a coherent manner as to form enterprise groups.1 3 Quite often, the vertical keiretsu were well developed in the aforementioned knowledge-intensive industries, such as general machinery, electric machinery, transport machinery, and precision machinery (see Table 41). 168 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 41. Change in the Proportion of Subcontracting Small and Medium Firms Sector 1966 1976 1986 Change 1966-86 Manufacturing total 53% 61% 56% 3% Food 17 15 8 -9 Textile manufacturing 80 85 80 0 Clothing and other textile 74 84 79 5 Wood and wood products 35 43 22 -13 Furniture and furnishings 46 41 39 -7 Pulp and paper 51 45 41 -10 Publishing and printing 46 51 42 -4 Chemical manufacturing 40 37 22 -18 Petroleum and coal products 30 27 18 -12 Plastics - -- 69 - Rubber products 62 61 65 3 Leather and leather products 60 63 65 5 Ceramic, stone, and clay products 34 29 35 1 Steel 66 70 52 -14 Nonferrous metal products 67 69 62 -5 Metal products 66 75 71 5 General machinery 71 83 75 4 Electric machinery 81 82 79 -2 Transport machinery 67 86 80 13 Precision machinery 72 72 70 -2 Other manufacturing -- 56 43 - Source: Chusho Kigyocho (Various issues), Kigyo Jittai Kihon Chosa [Small and Medium Enterprise Agency] (Tokyo: Chusho Kigyocho). This was mainly due to their requirements as assemblers for large numbers of parts, components and related goods for the production of final goods (for example, an automobile consists of about 20,000 items). It was in this context of expanding their production networks in a pyramidal fashion that such notable companies of the knowledge intensive industry as Toyota and Matsushita Electric Industry came to power in the Japanese economy in the 1970s. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. The major characteristic of the vertical keiretsu lies in its “half-market and half organization” nature (Imai 1982: p. 61, 1992: p. 215, 217), as in the case of the horizontal keiretsu. However, as for the vertical keiretsu, such a character is brought in substantially by the parent firm’s partial ownership of its subsidiaries.1 4 As Shimotani (1996: p. 127) pointed out, standing in sharp contrast with the firms of the U.S. and European soil, which completely (100%) own their subsidiaries, the parent firms in Japan hold various degrees of equity (20% to 100%) issued by their subsidiaries. For instance, in the case of Matsushita Electric Inc., in 1991, only 32 out of 84 subsidiaries were owned by Matsushita Electric: as for the rest, 8 subsidiaries were owned 80% to 90%, 11 subsidiaries, 60% to 70%, 17 subsidiaries, 50% to 60%, and 16 subsidiaries, less than 50% (Shimotani 1996).1 5 This kind of partial ownership by a parent firm is what Yoshimura, Ueno, and Kagano (1999: p. 29) called the “dual governance structure,” where “both the parent and the market govern these subsidiaries” (Yashimura, Ueno, and Kagano 1999)”1 6 It is through such a dual governance structure that a parent can simultaneously secure the premise for both the organizational (hierarchical) and market aspects at a time, because it allows the parent to selectively intervene in its 17 subsidiaries via its partial ownership, while delegating the governance to the public at other times. Such a double-faceted nature of the vertical keiretsu comes into full play in the trade dimension: as if they were originally in the same organization, the parent and the subsidiaries both worked over the specification of parts and then, the knowledge-sharing of technology (Aoki 1987: p. 283; Imai 1992: p. 215; Shimokawa 1993: p. 55). The relation thus established between them is likely self binding and thus strengthens the organizational aspect of the vertical keiretsu further, because the suspension of trade would incur a huge loss on both of them arising from the sunk-cost paid for their asset-specific investment (Yaginuma 1993: pp. 35-6). Simultaneously, such self-binding is constantly under market pressure. In the case of the automobile and electric machinery industries, it is now known that those subsidiaries which failed to meet the standards given by the parents (in terms of technological performance, quality, sales price, etc.) were often prodded out of the keiretsu 170 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. membership and replaced by more competent ones (Asanuma 1989: p. 2, 1992: p. 115; Yaginuma 1993: p. 29). The vertical keiretsu progressively gained force, as the low growth economy settled in the domestic economy. In fact, the prevalence of the vertical keiretsu eventually led to a change in the accounting system. In March 1978, “the consolidated accounting system” was put into effect, requiring parent firms to consolidate their financial statements with those of their 50% or more- owned subsidiaries. Also, in March 1984, the equity method (mochibun-ho) came into force, calling for the consolidation of the parents’ income statement with those of their 20% to 50% owned- 18 subsidiaries. The primary purpose for these new systems was to protect investors through disclosure of consolidated data, which are deemed reflecting more accurate financial state of each enterprise group (kigyo gurupu) where a parent and its subsidiaries are associated, virtually, as “a single operative body” (Shimotani 1996: pp. 131-3). The irreversible trend of the low growth economy now became obvious. The Japanese economy was inevitably coming into the new age when the traditional way of conducting business by the horizontal keiretsu was required reconsideration. 3.3 The Staggering One-Set Principle: The Intersection of the Horizontal and Vertical Keiretsu and the Effect of Diversification The firms of the six horizontal keiretsu were not immune from the above trend. This section examines the development of the vertical keiretsu and its impact on the keiretsu firms through the increase in the number of their subsidiaries and their diversification. The latter, in particular, intends to measure how much the “one-set principle” was eroded by diversification through the newly set-up subsidiaries. In the past literature, probably due to the organizational disparity between the horizontal and vertical keiretsu, they were studied separately or treated dichotomously with the consequence 171 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. that their relations were ignored or left obscured. Perusal of member firms of the six horizontal keiretsu showed, however, that these two types of keiretsu are not exclusive to each other. Rather, they form a multi-layered inter-corporate complex in the way that, while joining in their horizontal keiretsu, they respectively serve as parent firms standing at the apexes of their subsidiaries. The average numbers of subsidiaries in the production sectors are reported in Figures 26 to 31 for the six keiretsu. They show phenomenal increases in the subsidiaries of the two notable industries that were significantly developed after the oil crises—electric machinery and transportation machinery.1 9 , 2 0 Number of Subsidaries (Mitsui) 350.0 300.0 250.0 fe 200.0 .a E Z 150.0 100.0 50.0 =1 0.0 1995 1983 1985 1990 - Paper&Pulp -Chem ical -Textile - Ceram ics - Steel - Non Ferrous - Electronics -Transportation Year F igure 26. N um ber o f subsidiaries (M itsui). 172 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Number of Subsidiaries (Mitsubishi) 120 100 80 60 40 20 0 1995 1985 1990 1983 -♦ — Paper&Pulp -■ — Chemical -A— Textile -X— Petroleum - * — Ceramics -©— Steel — l— Non-ferrous General Machinery T ransportation - - Q - - Precision Year Figure 27. Number o f subsidiaries (Mitsubishi). Number of Subsidiaries (Sumitomo) 140 120 100 80 60 40 20 0 1990 1995 1983 1985 — 0— Chemical — ■— Ceram ics — A— Steel - - -X - - - Non-ferrous — X— General Machinery — • — Electronics Year Figure 28. Number o f subsidiaries (Sumitomo). R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Number of Subsidiaries (Fuyo) 180 160 140 120 o 100 20 1983 1985 1990 1995 — ♦— Textile — ■ — Paper& Pulp — i t— Chem ical - - -X - ■ ■ Petroleum Ceram ics Steel G eneral M achinery Electronics Transportation Precision Year Figure 29. Number o f subsidiaries (Fuyo). Number of Subsidiaries (Sanwa) B 60 50 40 20 10 0 1983 1985 1990 1995 -Textile -Chemical - Petroleum -Gum - Ceram ics I — • — Steel — I — Non-ferrous General Machinery Electronics - - O - - Precision Year Figure 30. Number o f subsidiaries (Sanwa). R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Number of Subsidiaries (DKB) Paper& Pulp E — Chem ical * — Petroleum * — Gum * — Steel O - ■ Non-ferrous H — G eneral Machinery — Electronics — T ransportation ■ 0 — Precision Figure 31. Number o f subsidiaries (DKB). The increment of subsidiaries, or, equivalently, the development of the vertical keiretsu across sectors, indicates that the parent firms had, more or less, shifted their trade patterns from the ones with other large firms belonging in the same Presidents’ Council to other ones with their own subsidiaries. The development of the vertical keiretsu also often entailed business diversification with parent firms directing their subsidiaries into the sectors that were not traditional to their mainstay business (Dore 1986: pp. 61-3; Kosei Torihiki Iinkai Jimukyoku 1992: pp. 86-98). The motivations behind the parents’ diversification, or “ta k a k u -k a were basically threefold. First, many parent firms perceived that, as the growth of the economy decelerated, their mainstay businesses matured or even became obsolete (Kosei Torihiki Iinkai Jimukyoku 1992: p. 87). In face of the poor market prospects of their matured mainstay businesses, a large number of Japanese firms were forced into diversification (Dore 1986: p. 63; Nakamura 1992: pp. 26-8). For instance, the survey conducted by the Japan Committee for Economic Development (keizai doyu-kai) in 1985 showed that 78.3% of 175 180 160 140 120 o .G £ 3 Z 100 1983 1985 1990 1995 Y ear R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. the sample firms replied that their mainstay business became fully matured (Kosei Torihiki Iinkai Jimukyoku 1992). Among those firms, the survey continued, more than half replied that they were willing to enter other industrial areas. (Kosei Torihiki Iinkai Jimukyoku 1992: p. 88). Second, technological innovations created many new industrial fields and encouraged entry from various sectors where their raw materials and production methods could be easily replicated to them (Kosei Torihiki Iinkai Jimukyoku 1992: pp. 86-7). Some notable examples of such new fields were the IC industry, where the entry occurred from electric machinery, computers, and communication equipment: the optical fiber industry, where the entry occurred from electric wires, glass makers; and chemical makers, and the biotechnology industry, where the entry occurred from medical, foods, chemical, and electric machinery (Kosei Torihiki Iinkai Jimukyoku 1992: p. 87). Third, as a result of the commensurability of production methods between adjacent fields, firms were given more opportunity to reduce the aggregate costs by producing multiple products at a time (“economies of scope”) (Kosei Torihiki Iinkai Jimukyoku 1992: pp. 97-8; Nakamura 1992: pp. 29- 30). Some examples of this are shipbuilding firms where their technologies were redeployed for plant engineering, prefabricated housing, and large-scale environmental modification schemes, and agricultural firms, which, utilizing their technologies, started to increase the production of electric motors and fire pumps (Dore 1986: p. 63). The member firms of the six keiretsu were, in general, subject to these industrial trends. In order to see how much diversification progressed in each of the six keiretsu, “the diversification matrix” is hereupon introduced for each (see Tables 42-47). The main idea of the diversification matrix is to determine in which industrial sectors the member firms and their subsidiaries made 21 turnovers in a given time. The sectors considered in the matrix are in total 16, which were all chosen from production sectors (foods, textile, lumber, pulp, printing, chemical, petroleum, gum, ceramics, steel, non-ferrous metal, machinery, electric machinery, transportation machinery, and precision machinery). In the matrix, the sectors vertically arranged represent the parent firms’ mainstay business (hongyo), whereas those horizontally arranged represent the sectors in which 176 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. those parent firms diversified by setting up their subsidiaries. Thus, the matrix becomes 16x16, and reading it horizontally gives the idea of how much the parents in the 16 industrial sectors earned through diversification, and reading it vertically gives the idea from which sector the entries were actually made for each of the 16 sectors. Because parents’ earnings from their mainstay business appear only on the diagonal line of the matrix, let us define “the mainstay ratio” as the ratio of each of those values on the diagonal relative to the total earnings made by the parent and its subsidiaries. Let us also define “the non-mainstay ratio” as the ratio of the earnings made by the non-mainstay business relative to the total earnings made by the non-mainstay and mainstay businesses, in each of the 16 sectors. The diversification matrix is created for each of the six keiretsu in the following way: first of all, the analysis is targeted to FY 1985. This is because the ultimate goal of the present chapter is to determine the effect of the financial liberalization that occurred thereafter, which is fully discussed in the next section. The shacho-kai member firms of the 16 production sectors were selected from Kigyo Keiretsu Soran (Toyo Keizai Shinposha, published in 1986). The sales of those firms and their parents were then summed up, respectively, in the most appropriate industries in the matrix, in light of the company’s description in Nihon no Kigyo Gurupu (Toyo Keizai Shinposha, published in 1986) and the Japan Standard Industrial Classification (Nihon hyojun sangyo bunrui) released by the Statistics Bureau of the Ministry of Public Management, Home Affairs, and Posts and Telecommunication (Somucho Tokeikyoku 2002). Tables 42 to 47 are the diversification matrices obtained by the procedures above. “H” and “V” under each of the sales values, respectively, stand for the mainstay ratio and the non-mainstay ratio (the unit is percent). 177 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 42. Diversification Matrix (Mitsui) M ain business and subsidiaries: Earnings (million yen) and ratios (%) for “H” and “V” SUM 188991 667540 2301136 IT) O b L O C N J S r - ~ - C O O C O O b r-- co C O o> L O C M r- co Indis- cem. Non- Production g r - '- C N I o n (\j 5b X > 321774 1-1:14, V:13.7 x > 7918 H:4.4, V:0.3 5b w o g t o C M X > g o d 3 S i > LU 39378 H:1.7, V:22.0 18085 H:4.2, V:10.1 49503 H:10.1, V:27.7 71670 H:1.6, V:40.1 Precise 13437 1 H:0.6, V:10.2 s S s x > Transp. 0 0 ^ ■ 0 0 eo X > Electric. 2665 H:0.1, V:0.1 40081 H:9.4, V:1.1 15311 H:8.4, V:0.4 to O) ^ X > 3575843 ! H:81.8. : V:98.3 Machines 18411 H:2.8, V:3.6 < £ > g o o C b X > 23879 H:13.1, V:4.7 C M c o * C O s g ° 120663 H:2.8, V:23.8 Metals S ^ ® ¥ i > Non- Steel 14691 H:0.6, V:3.3 392092 H:80.4, V:87.0 35725 H:0.8, V:7.9 Steel o> ^ co m if t- 4941 H:1, V:1.5 12881 H:0.3, V:4.0 Ceramics I 300197 I H:70.5. I V:81.1 3689 H:0.8, V:1.0 3 ^ 2 Gum Petrol. 7346 H:1.5, V :0.4 i O J " M " . I T b O s — ^ X > Chemical 1767670 i H:76.8, V:97.4 20664 H:4.9, V:1.1 Printing Pulp 504457 H:75.6, V:100 Lumber 67611 H:10.1, V:100 Textile 134658 H:5.9, V:1 Foods 188991 H:100, V:86.4 25969 H:3.9, V:11.9 3837 I H:0.2, V:1.8 Parent firm’ s main business Foods Textile Lumber | Pulp Printing Chemical Petroleum Gum Ceramics Steel Non-Steel Metals Machines Electric. 178 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table continues next p age Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Table 42 (continued). Diversification Matrix (Mitsui) Parent Main b u sin ess and subsidiaries: Earnings (million yen) and ratios (%) for “H" and “ V" firm’ s main business Foods Textile Lumber Pulp Printing Chemical Petrol. Gum Ceramics Steel Non- Steel Metals Machines Electric. Transp. Precise Etc.(P) Non- Production Indis- cern. SUM Transp. 171190 H:1.5, V:52.9 8128 H:0.1, V:1.8 340232 H:3, V:67.0 9335428 H:81.9, V:100 1580 H:0.0, V:1.2 1544187 H:13.5, V:66.0 11400745 Precise Etc.(P) Non-Prod. Indiscem SUM 218797 134658 67611 504457 1815200 370331 323608 450636 4550 507553 3638246 9339168 132273 178636 2340850 20026574 Sources: T he shacho-kai m em ber firms of the 16 production secto rs w ere selected from Toyo Keizai S hinposha, Kigyo Keiretsu Soran (1986). The sa le s of th o se firms and their parents w ere then sum m ed up, respectively, in th e m ost appropriate industries in the matrix, in light of the com pany’s description in Toyo Keizai S hinposha, Nihon no kigyo gumpu (1986) and the Japan Standard Industrial Classification released by the Statistics B ureau of the Ministry of Public M anagem ent, H om e Affairs, and P osts and Telecom m unication (S om uchb Tokeikyoku 2002). Notes. H = S ales value m ainstay ratio (%). V = S ales value non-m ainstay ratio (%). Petrol. = Petroleum . M achines = G eneral m achinery. Electric. = Electric m achinery. Transp. = Transportation m achinery. P recise = Precision m achinery. Etc.(P) = O ther production sector. N on-P. = N on-production sector. Indiscem . = Indiscernible sector. The 16 secto rs in the matrix w ere chosen from production sectors (foods, textile, lumber, pulp, printing, chem ical, petroleum , gum , ceram ics, steel, non-ferrous m etals, general m achinery, electric m achinery, transportation m achinery, and precision machinery). T he secto rs vertically arranged represent the parent firms' m ainstay b u sin ess (hongyd); those horizontally arranged rep resen t the sectors in which those parent firms diversified by setting up their subsidiaries. Reading th e matrix horizontally gives the idea of how m uch the paren ts in the 16 industrial secto rs earn through diversification; reading it vertically gives the idea from which sector the entries are m ade for ea ch of the 16 sectors. The m ainstay ratio is the ratio of ea ch of the values on the diagonal line relative to the total earnings m ade by the parent and its subsidiaries. The non-m ainstay ratio is the ratio of th e earnings m ade by the non-m ainstay b u sin ess relative to the total earnings m ade by the non-m ainstay and m ainstay b u sin esses in each of th e 16 sectors. The analysis is targeted to FY 1985. <i V O Table 43. Diversification Matrix (M itsubishi) M ain business and subsidiaries: Earnings (million yen) and ratios (%) for “H” and “V" SUM 1484128 170463 4194148 1453945 h- c m C M 8 C O co o Indis cem . 37792 H.0.9, V:80.2 co T ~ m co § Non-P. 51182 H:3.4, V:7.3 140195 H:3.3, V:19.9 53215 H:3.7, V:7.5 32981 H:2.9, V:4.7 7457 H:7.2, V:1.1 Etc. (P) Precise 1 1 6060 H:0.5, V:2.7 Transp. Electric. 21620 H:0.5, V:0.9 1396 H:1.4, V:0.1 M achines 45460 H:1.1, V:12.0 9099 H:0.8, V:2.4 8096 H:7.8, V:2.1 M etals Non- Steel 150274 H:3.6, V:13.2 Steel o ® co oo co cb S * co X > C eram ics 29185 H.0.7, V:2.7 1058078 H:92.6, V:97.3 Gum 4571 H.0.1, V:100 Petroleum 619183 H:14.8, V:32.0 1307770 H:89.9, V:67.7 5942 H:0.5, V:0.3 Chem ical 3125435 H.74.5, V:96.4 91654 H:6.3, V:2.8 24776 H:2.2, V:0.8 Printing Q. 3 C L 170463 H:100, V:100 Lum-| ber I Textile 9262 H.0.2, V:100 F oods 1432946 H:96.6, V:98.9 rZ co ' " d o — X > 5281 H:0.5, V:0.4 P arent firm’ s main business F oods Textile Lum ber Pulp Printing Chem ical Petrol. Gum C eram ics Steel 180 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table continues next p age Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Table 43 (continued). Diversification Matrix (Mitsubishi) P arent firm's Main bu sin ess and subsidiaries: E arnings (million yen) and ratios (%) for “H” and ‘V main business Foods Textile Lum ber Pulp Printing Chem ical Petroleum Gum C eram ics S teel Non- Steel M etals M achines Electric. Transp. Precise Etc. (P) Non-P. Indis cem . SUM Non-Steel 951990 H:89.9, V:83.5 20767 H:2.0, V:0.6 7621 H:0.7, V:32.8 78936 H:7.5, V:11.2 1059314 Metals M achines 34271 H:84.2, V:9.0 6452 H:15.8, V:0.9 40723 Electric. 37332 H:1.5, V:3.3 2701 H:0.1, V:100 2534 H:0.1, V:0.7 2243270 H:91.6, V:95.2 160428 H:6.6, V:22.7 2063 H:0.1, V:4.4 2448328 Transp. 22318 H:0.6, V:100 157 H:0.0, V:0.0 3055 H:0.1, V:3.4 279581 H:6.9, V.73.8 90873 H:2.2, V:3.9 3465412 H:85.4, V:99.4 15616 H:0.4, V:67.2 173895 H:4.3, V:24.7 5989 H:0.1, V:12.7 4056896 Precise 216264 H:99.7, V.97.3 572 H:0.3, V:0.1 216836 Etc.(P) Non-P. Indiscem. SUM 1449398 9262 170463 22318 3242022 1932895 4571 1087263 89435 1139596 2701 379041 2357159 3486179 222324 23237 705313 47150 16370327 Source: The shacho-kai m em ber firms of the 16 production sectors w ere selected from Toyo Keizai S hinposha, Kigyo Keiretsu Soran (1986). The sa le s of th o se firms and their parents w ere then sum m ed up, respectively, in the m ost appropriate industries in the matrix, in light of the com pany’ s description in Toyo Keizai S hinposha, Nihon no kigyo gurupu (1986) and the Japan Standard Industrial Classification released by the Statistics B ureau of th e Ministry of Public M anagem ent, Home Affairs, and P o sts and Telecom m unication (S om ucho Tokeikyoku 2002). Note. The notes for Table 42 apply to this table also. H = S ale s value m ainstay ratio (%). V = S a le s value non-m ainstay ratio (%). M ainstay ratio = the ratio (in %) of e a ch of the earnings m ade by the p arent firm divided by the total earnings m ade by the p aren t and its subsidiaries. N on-m ainstay ratio = the ratio (in %) of the earnings m a d e by the non-m ainstay b u sin ess divided by th e total earnings m ade by the non-m ainstay and m ainstay b u s in e sse s in e a ch of the 16 sectors. Petrol. = Petroleum . M achines = G eneral m achinery. Electric. = Electric m achinery. Transp. = Transportation m achinery. P recise = Precision machinery. Etc.(P) = O ther production sector. Non-P. = Non-production sector. Indiscem . = Indiscernible sector. Table 44. Diversification Matrix (Sumitomo) P arent firm’ s main b u sin ess Main b u sin ess and subsidiaries: Earnings (million yen) and ratios (%) for “H” and “V” Foods Textile Lum ber Pulp Printing Chem ical Petroleum Gum Foods Textile Lumber Pulp Printing Chem ical Petroleum Gum Ceram ics Steel 1757591 H:86.8, V:96.7 30514 H:6.6, V:1.7 27168 H:0.9, V:1.5 203306 H:6.9, V:100 N on-Steel 2988 H:0.1, V:0.2 280182 H:13.1, V:100 Metals M achines 178 H:0.0, V:0.0 Electric. Transp. Precise Etc (P) Non-P. Indiscem. SUM 1818439 203306 280182 Sources: The shacho-kai m em ber firms of the 16 production secto rs w ere selected from Toyo Keizai S hinposha, Kigyo Keiretsu Soran (1986). The sale s of th o se firms and their parents w ere then sum m ed up, respectively, in the m ost appropriate industries in the matrix, in light of the com pany’s description in Toyo Keizai S hinposha, Nihon no kigyo gurnpu (1986) and the Japan Standard Industrial Classification released by the Statistics Bureau of the Ministry of Public M anagem ent, Home Affairs, and P o sts and Telecom m unication (SOmucho Tokeikyoku 2002). Notes. H = S ale s value m ainstay ratio (%). V = S ale s value non-m ainstay ratio (%). Petrol. = Petroleum . M achines = G eneral m achinery. Electric. = Electric m achinery. T ransp. = Transportation m achinery. P recise = Precision m achinery. Etc.(P) = O ther production sector. Non-P. = N on-production sector. Indiscem . = Indiscernible sector. The 16 secto rs in the matrix w ere ch o sen from production secto rs (foods, textile, lumber, pulp, printing, chem ical, petroleum , gum, ceram ics, steel, non-ferrous m etals, general m achinery, electric m achinery, transportation m achinery, and precision machinery). The sectors vertically arranged rep resen t the p arent firms’ m ainstay b u sin ess (hongyo); th o se horizontally arranged rep resen t the secto rs in which those p arent firms diversified by setting up their subsidiaries. Reading the matrix horizontally gives the idea of how m uch the parents in the 16 industrial sectors earn through diversification; reading it vertically gives the idea from which secto r the entries are m ade for e a ch of the 16 sectors. T he m ainstay ratio is the ratio of each of the values on the diagonal line relative to the total earnings m ade by the p arent and its subsidiaries. T he non-m ainstay ratio is the ratio of the earnings m ade by the non-m ainstay b u sin ess relative to the total earnings m ade by the non-m ainstay and m ainstay b u sin esses in each of the 16 sectors. The analysis is targeted to FY 1985. 182 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 45. Diversification M atrix (Fuyo) M ain business and subsidiaries: Earnings (million yen) and ratios (%) for “H” and “V ” SUM 1397822 6 2 3 0 5 6 h- C O O C M C O C O 1366647 1832334 C O C M C M 2 0 8 3 9 4 2 Indis- cern. 1706 H:0.3, V:100 Non-P. 18574 H:1.3, V:1.2 I s m co ^ o n i > ^ LO C O & co X > CN C M C D to c m S? X > 30908 H :1.7, V:2.0 260029 j H :12.5, V:16.5 I Q -r- L U 317 8 H:0.2, V :86.0 519 H:0.0, V :14.0 Precise Transp. 5 < * > n n c i c m X > g C M X > Electric. 122458 H :19.7, V:2.4 Machines C M co" co C T > o 2 X > c m " V - C M X > 28817 H:1.4, V :1.6 Metals 1171 H:0.1, V:0.3 3 1 119 H :1.5. V:8.3 Non- | Steel j C O M o t o ^ x > 6155 H:0.3, V:35.1 Steel 1711970 H .82.2, V :99.9 Ceramics 2304 H:0.3, V:0.5 2 2 3044 H :100, V :48.3 Gum Petroleum D - ? t n c o g 5 o l ? x > 3459 H:0.2, V:0.2 Chemical 58 864 H:4.2, V:4.0 174262 H :28.0, V :11.8 S tOlO o co CD C O C O 8 x > 3 0 8229 H :16.8, V :20.9 18566 H:0.9, V:1.3 Printing Pulp 3 026 H:0.5, V:0.5 621317 H:93.8, V :99.5 Lum ber Textile 2 8 3 5 1 4 H :45.5, V:100 Foods 1320384 H :94.5, V:100 Parent firm's main business Foods i Textile I Lumber Pulp Printing j Chemical | Petroleum | Gum j Ceramics Steel Non-Steel Metals R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table continues next p ag e Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Table 45 (continued). Diversification Matrix (Fuyo) Parent firm's Main b u sin ess and subsidiaries: Earnings (million yen) and ratios (%) for “H” and “V” main business Foods Textile Lum ber Pulp Printing Chemical Petroleum Gum Ceramics Steel Non- Steel Metals Machines Electric. Transp. Precise Etc. (P) Non-P. Indis- cern. SUM Machines 8250 H:0.7, V:0.6 5403 H:0.5, V:1.2 975513 H:85.5, V:55.4 1 2 H:0.0, V: 0 .0 49020 H:4.3, V:0.7 102326 H:9.0, V:6.5 1140524 Electric. 2868 H:0.0, V:100 500 H:0.0, V:0.0 231454 H:3.2, V:50.1 1481 H:0.0, V:0.1 283339 H:3.9, V:75.4 731677 H:10.0, V:41.6 4983979 H:68.0, V:96.2 241122 H:3.3, V:3.6 227134 H:3.1, V:18.4 627107 H:8.6, V:29.8 7330661 Transp. 57996 H-.0.9, V:100 60180 H:0.9, V:16.0 75681 H-.1.1, V:1.5 6337300 H:94.9,.V: 95.0 98189 H.1.5, V:8.0 51656 H.0.8, V:3.3 6681002 Precise 11571 H:1.3, V:0.7 908346 H:98.7, V:73.6 919917 Etc.(P) Non-P Indiscem. SUM 1320384 283514 624343 2868 1477735 1496656 57996 462205 1713451 17553 375809 1760810 5182130 6672494 1233669 3697 1573996 1706 24261016 Sources: The shacho-kai m em ber firms of the 16 production secto rs w ere selected from Toyo Keizai S hinposha, Kigyo Keiretsu Soran (1986). The sa le s of th o se firms and their parents w ere then sum m ed up, respectively, in the m ost appropriate industries in the matrix, in light of the com pany's description in Toyo Keizai S hinposha, Nihon no kigyo gurupu (1986) and the Japan Standard Industrial Classification released by th e Statistics B ureau of the Ministry of Public M anagem ent, H om e Affaire, and P o sts and Telecom m unication (Som ucho Tokeikyoku 2002). Notes. H = S ales value m ainstay ratio (%). V = S ale s value non-m ainstay ratio (%). Petrol. = Petroleum . M achines = G eneral m achinery. Electric. = Electric m achinery. T ransp. = Transportation m achinery. P recise = Precision machinery. Etc.(P) = O ther production sector. Non-P. = Non-production sector. Indiscem . = Indiscernible sector. The 16 secto rs in the matrix w ere ch o sen from production secto rs (foods, textile, lumber, pulp, printing, chem ical, petroleum , gum , ceram ics, steel, non-ferrous m etals, g eneral m achinery, electric m achinery, transportation m achinery, and precision m achinery). T he secto rs vertically arranged represent the p aren t firms' m ainstay b u sin ess (hongyo)-, th o se horizontally arranged rep resen t the sectors in which those parent firms diversified by setting up their subsidiaries. Reading the matrix horizontally gives th e idea of how m uch the parents in the 16 industrial secto rs ea rn through diversification; reading it vertically gives the idea from which secto r the entries are m ade for each of the 16 sectors. The m ainstay ratio is the ratio of ea ch of the values on the diagonal line relative to the total earnings m ade by the p arent and its subsidiaries. The non-m ainstay ratio is the ratio of the earnings m ade by th e non-m ainstay b u sin ess relative to th e total earnings m a d e by th e non-m ainstay and m ainstay ^ b u sin esses in e a ch of th e 16 sectors. The analysis is targeted to FY 1985. oo Table 46. Diversification Matrix (Sanwa) P arent firm’ s main business Main busin ess and subsidiaries: Earnings (million yen) and ratios (%) for “H" and “V" F oods Textile Lum ber Pulp Printing Chem ical Petroleum Gum Foods 381386 H:100, V:97.4 Textile Lumber 4499 H:8.4, V:1.1 385174 H:72.0, V:97.1 136 H:0.0, V:5.5 107203 H:20.0, V:3.6 Pulp Printing Chem ical Petroleum Gum C eram ics 5780 H:0.2, V:1.5 11329 H:0.3, V:2.9 2691033 H:72.4, V:90.1 150855 H:7.2, V:5.0 8692 H:0.2, V:0.4 1817344 H:86.8,V:92.7 2873 H:0.1, V:1.1 237125 H:91.5, V:92.6 22101 H:8.5, V:0.7 868 H:0.3, V:0.0 Steel 2344 H:0.1, V:94.5 7162 H:0.2, V:0.2 134910 H:3.5, V:6.9 Non-Steel M etals 4098 H:1.0, V:0.1 10084 H:2.5, V:3.9 M achines Electric. Transp. 2630 H:0.1, V:100 4759 H 0 1 V:0.2 190 H:0.0, V:0.0 5865 H:0.3, V:2.3 T able continues next p age 185 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 46 (continued). Diversification Matrix (Sanwa) P arent firm’ s main b usiness Main b u sin ess and subsidiaries: Earnings (million yen) and ratios (%) for “H" and “V” F oods Textile Lumber Pulp Printing Chem ical Petroleum Gum Precise Etc.(P) Non-P. Indiscem. SUM 391665 396503 2480 2630 2988269 1960946 255947 Sources: T he shacho-kai m em ber firms of the 16 production secto rs w ere selected from T o y s Keizai S hinposha, Kigyo Keiretsu Soran (1986). The sa le s of th o se firms and their parents w ere then sum m ed up, respectively, in the m ost appropriate industries in the matrix, in light of the com pany’ s description in T o y s Keizai S hinposha, Nihon no kigyo gurupu (1986) and the Japan Standard Industrial Classification released by the Statistics B ureau of the Ministry of Public M anagem ent, Home Affairs, and P o sts and Telecom m unication (Som ucho Tokeikyoku 2002). Notes. H = S ale s value m ainstay ratio (%). V = S ales value non-m ainstay ratio (%). Petrol. = Petroleum . M achines = G eneral machinery. Electric. = Electric m achinery. Transp. = Transportation m achinery. P recise = Precision m achinery. Etc.(P) = O ther production sector. Non-P. = N on-production sector. Indiscem . = Indiscernible sector. The 16 secto rs in the matrix w ere ch o sen from production secto rs (foods, textile, lumber, pulp, printing, chem ical, petroleum , gum, ceram ics, steel, non-ferrous m etals, general m achinery, electric m achinery, transportation m achinery, and precision machinery). The sectors vertically arranged rep resen t the p arent firms’ m ainstay b u sin ess (hongyo); th o se horizontally arranged rep resen t the secto rs in which th o se p arent firms diversified by setting up their subsidiaries. Reading the matrix horizontally gives the idea of how m uch the parents in the 16 industrial secto rs earn through diversification; reading it vertically gives the idea from which secto r the entries are m ade for each of the 16 sectors. T he m ainstay ratio is the ratio of ea ch of the values on the diagonal line relative to th e total earnings m ade by the p arent and its subsidiaries. T he non-m ainstay ratio is the ratio of the earnings m ade by the non-m ainstay b u sin ess relative to the total earnings m ade by the non-m ainstay and m ainstay b u sin esses in e a ch of the 16 sectors. The analysis is targeted to FY 1985. 186 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 47. Diversification Matrix (DKB) M ain business and subsidiaries: Earnings (million yen) and ratios (%) for “H” and “V* SUM 679665 3543662 2377992 273268 6537575 1601509 724248 Indis- cern. 48703 H:1.4, V:55.3 C O s 0 ^ ^ X > Non-P. 65002 H:9.6, V:3.5 14917 H:0.4, V:0.8 66172 H:2.8, V:3.6 4439 H:1.6, V:0.2 291688 H:4.5, V:15.9 159823 H',10.0, V:8.7 126267 H:17.4, V:6.9 Etc. (P) 2280 H:0.1, V:23.7 Precise 13000 H:0.4, V:7.0 C M ^ § 5 5 1 co X > Transp. L O C O C M S P O h- X > 1015 H:0.4, V:0.0 C M n o t -: ^ x > 41815 H:2.6, V:1.1 1713 H:0.2, V.0.0 6 1 U J 83247 H:1.3. V:1.0 29377 H:1.8, V:0.4 C M C O ' C O C M X > Machines 17514 H:0.5, V:1.0 1103090 I H:1.6, ! V:6.1 ^ C O C M tT X > 563451 H:77.8, V.33.6 Metals 84541 H:1.3, V:15.6 436143 H.27.2, V:80.7 T — ^ to 0 0 C O X > Non-Steel 56283 H:0.9, V:4.8 910757 H-,56.9, V:77.0 Steel 5675642 H:86.8, V:95.3 Ceramics 202929 H:5.7, V:41.7 36217 H:0.6, V:7.4 1468 H-.0.1, V:0.3 £ 3 o O C N J < D X > 267814 H:98.0, V:99.8 1 Petroleum j i 2254537 H:94.8, V:94.4 134910 H:0.2, V:5.6 Chemical 1044 H:0.2, V:0.0 3049514 H:86.1, V:96.8 49658 H:2.1, V:1.6 13528 H:0.2, V:0.4 i 18091 : H:1.1, V:0.6 5943 H:0.8, V:0.2 Printing Pulp hi r-T co o s O ) ifi X > 4093 H:0.1, V:0.7 Lum ber t o 9 tri £> cm o o <2 x > 2344 H:0.0, V:14.9 Textile T- < £ ) “ OO c r> o in ^ X > 68518 H:1.9, V:94.2 Foods 121558 H:3.4, V:97.2 OO O O i C O X > ; Parent fiirm's main busi ness Foods Textile Lumber Pulp Printing Chem. Petrol. Gum Ceram. Steel Non- Steel Metals Mach. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table continues next page Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Table 47 (continued). Diversification Matrix (DKB) Parent fiirm’ s Main b u sin ess and subsidiaries: Earnings (million yen) and ratios (%) for “H" and “V” main busi ness Foods Textile Lum ber Pulp Printing Chemical Petroleum Gum Ceramics Steel Non-Steel Metals Machines Electric. Transp. Precise Etc. (P) Non-P. Indis cem. SUM Electric. 12009 H:0.1, V:0.4 229769 H:2.1, V:47.2 282580 H:2.6, V:4.7 215636 H:2.0, V:18.2 815123 H:7.6, V:48.6 7847406 H:73.2, V:97.8 305457 H:2.8, V:8.1 56113 H:0.5, V:30.4 925022 H:8.6, V 50.3 33318 H:0.3. V:37.8 10722433 Transp. 5029 H:0.1, V:100 16670 H:0.4, V:3.4 71 H.0.0, V:0.0 16689 H:0.4, V:3.1 170834 H:4.5, V:10.2 37707 H:1.0, V:0.5 3376972 H:88.3, V:89.4 7038 H:0.2, V:3.8 7339 H 0 2 V./6.3 185719 H 4.9, V 10.1 2068 H:0.1, V:2.3 3826136 Precise 2228 H:2.0, V:0.1 2017 H:1.9, V:0.0 104587 H:96.1, V:56.7 108832 Etc.(P) Non-P. Indis cem. SUM 125066 72709 15701 600164 5029 3149787 2389447 268450 487053 5958222 1182747 540524 1676275 8023477 3779351 184540 9619 1839049 88110 30395320 Sources: The shacho-kai m em ber firms of the 16 production secto rs w ere selected from Toyo Keizai S hinposha, Kigyo Keiretsu S oran (1986). The sa te s of th o se firms and their p aren ts w ere then sum m ed up, respectively, in the m ost appropriate industries in the matrix, in light of the com pany's description in Toyo Keizai S hinposha, Nihon no kigyo gurupu (1986) and the Japan Standard Industrial Classification released by the Statistics B ureau of the Ministry of Public M anagem ent, Home Affairs, and P o sts and Telecom m unication (Som ucho Tokeikyoku 2002). Notes. H = S ales value m ainstay ratio (%). V = S ale s value non-m ainstay ratio (%). Petrol. = Petroleum . M achines = G eneral m achinery. Electric. = Electric m achinery. Transp. = Transportation m achinery. P recise = Precision m achinery. Etc.(P) = O ther production sector. Non-P. = Non-production sector. Indiscem . = Indiscernible sector. T he 16 secto rs in the matrix w ere ch o sen from production secto rs (foods, textile, lumber, pulp, printing, chem ical, petroleum , gum , ceram ics, steel, non-ferrous m etals, g eneral m achinery, electric m achinery, transportation m achinery, and precision machinery). T he secto rs vertically arranged represent the p arent firms’ m ainstay business (hongyo); th o se horizontally arranged rep resen t the sectors in which th o se p aren t firms diversified by setting up their subsidiaries. R eading the matrix horizontally gives th e id e a of how m uch the parents in th e 16 industrial secto rs earn through diversification; reading it vertically gives the idea from which sector the entries are m ade for e a ch of the 16 sectors. The m ainstay ratio is th e ratio of each of the values on the diagonal line relative to the total earnings m ade by the parent and its subsidiaries. T he non-m ainstay ratio is the ratio of the earnings m ade by the non-m ainstay b u sin ess relative to the total earnings m ade by the non-m ainstay and m ainstay b u sin esses in e a ch of the 16 sectors. The analysis is targeted to FY 1985. ‘The information for "Chicchibu Cem ent" is not available. oo oo As expected, the diversification activities are observed off the diagonals of the matrices throughout all six keiretsu. The following three points should be made about the matrices. First, the firms whose mainstay business lay in the chemical (Mitsui, Mitsubishi, Fuyo, Sanwa) and steel (Mitsui, Sumitomo) sectors, both of which were representatives of the high-growth economy, started to enter other sectors, such as service and petroleum in the case of the chemical sector, and machines and electric machinry in the case of the steel sector. Most impressive is the case of Sumitomo, where the mainstay ratio of its steel sector (Sumitomo Metal Ind. or Sumitomo Kinzoku Kogyo) was scaled down to 65.9%, despite its enormous involvement in the steel production during the high-growth era. Second, simultaneous to the restructuring of the above two industries, there occurred entries into the machinery and precision machinery sectors, the two leading sectors of the low growth economy, mainly from the transportation (Mitsubishi), steel (Sumitomo), and electric machinry (Fuyo, Sanwa, DKB) sectors, and from the electric machinry sector (Fuyo, Sanwa, DKB), respectively. Third, there is a somewhat large variance in the sales share of the service sector across the six keiretsu. The sales share of the service sector is relatively higher in the Bank Groups (Fuyo: 6.5%, Sanwa: 8.3%, DKB: 6.1%), and particularly in the Mitsui Group (11.7%), and lower in the Mitsubishi (4.3%) and Sumitomo Groups (2.7%). This result indicates that the tight inter-firm transaction relations of the Mitsubishi and Sumitomo Groups were less affected by the incentive to diversify into the service sector even in the middle of the low growth trend. Now the problem is, after the diversification observed as above, how much of the industrial configuration set up in line with the one-set principle was affected in each keiretsus. Table 48 serves to examine this problem by allowing one to check, in each keiretsu, how many sectors were engaged in their mainstay business, while better protected from the entry from the outside sectors, through the comparison of “the high mainstay ratio” and “the low non-mainstay ratio.” 189 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 48. Diversification of the Six Horizontal Keiretsu Mitsui Mitsubishi Sumitomo Fuyo S anw a DKB M ainstay Ratio: High (80.4% or above) foods (100), non-ferrous m etals (80.4), electric (81.8), transportation (81.9) foods (96.6), pulp (100), petroleum (89.9), ceram ics (92.6), non- ferrous m etals (89.9), electric (91.6), precision (99.7), steel (83.6), m achinery (84.2), transportation (85.4) chem ical (86.8), ceram ics (93.3), electric (97.5), m achinery (84.5) foods (94.5), pulp (93.8), petroleum (81.5), ceram ics (100), m achinery (85.5), transportation (94.9), precision (99.7), steel (82.2) foods (100), petroleum (86.8), gum (91.5), ceram ics (95.7), non- ferrous m etals (93.1), m achinery (100), steel (83.3), transportation (85.0), precision (82.8) pulp (87.7), chem ical (86.1), petroleum (94.8), gum (98.0), steel (86.8), transportation (88.3), precision (96.1) M ainstay Ratio: Low (below 80.4%) pulp (75.6), chem ical (78.8), ceram ics (70.5), steel (74.1) chem ical (74.5) steel (65.9), non-ferrous m etals (67.4) textile (45.5), chem ical (66.5), electric (68.0) textile (72.0), chem ical (72.4), electric (69.0) non-ferrous m etals (56.9), m achinery (77.8), electric (73.2) Ave./Std. 80.4/9.1 89.8/7.3 82.5/12.0 82.9/16.2 86.0/10.2 8 4 .6 /1 1 .8 Non- M ainstay Ratio: High (below 77.4%) steel (41.8) petroleum (67.7), m achinery (9.0) m achinery (57.0) chem ical (61.5), ceram ics (48,3), m achinery (55.4), precision (73.6) ceram ics (49.1), non-ferrous m etals (56.4), m achinery (22.8), precision (60.9) m achinery (33.6), precision (56.7), non- ferrous m etals (77.0) Non- M ainstay Ratio: Low (77.4% or above) foods (98.4), pulp (100), chem ical (97.4), ceram ics (91.1), non-ferrous m etals (97.0), electric (98.3), transportation (100) foods (99.9), pulp (100), chem ical (96.4), ceram ics (97.3), steel (96.6), electric (95.2), transportation (99.4), precision (97.3), non- ferrous m etals (83.5) chem ical (98.7), ceram ics (90.3), steel (100), non- ferrous m etals (88.0), electric (90.1) foods (100), textile (100), pulp (98.5), petroleum (99.8), steel (99.9), electric (96.2), transportation (95.0) foods (97.4), textile (97.1), chem ical (90.1), petroleum (92.7), gum (92.8), steel (91.0), electric (95.3), transportation (83.4) pulp (99.3), chem ical (96.8), petroleum (94.4), gum (99.8), steel (95.3), electric (97.8), transportation (89.4) Ave./Std. 90.5/18.6 85.7/25.9 87.4/13.4 84.4/19.3 77.4/23.2 84.0/21.1 Note. M ainstay ratio = the ratio (in %) of e a ch of the earnings m ade by the p arent firm divided by the total earnings m ade by the parent and its subsidiaries. N on-m ainstay ratio = the ratio (in %) of the earnings m ade by the non-m ainstay b u sin ess divided by the total earnings m ade by the non-m ainstay and m ainstay b u sin esses in each of the 16 sectors. T he num bers in p aren th e se s a re either the m ainstay ratio or the non-m ainstay ratio. Machinery = G eneral m achinery. Electric. = Electric m achinery. Transportation = Transportation machinery. Precision = Precision m achinery. Ave. = average. Std. = standard deviation. For com parison, the following are the results of the FTC conducted in 1986: High mainstay: foods, pulp & paper, printing, petroleum & coal, non-ferrous m etals, electric machinery. Low mainstay: textile, non-ferrous, precision m achinery, lumber. High non-mainstay: lumber, chem ical, non-ferrous m etals, general machinery, pulp & paper, ceram ics, non-ferrous m etals, transportation m achinery. Low non mainstay: foods, textile, printing, petroleum & coai, gum , steel, electric m achinery, precision machinery. 190 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. More specifically, Table 48 was created by classifying the industries of the 6 diversification matrices by the following four categories: high mainstay ratio, low mainstay ratio, high non mainstay ratio, and low non-mainstay ratio. The criterion for the high (low) mainstay ratio is that the industries’ mainstay ratios be equal to or more (less) than 80.4%, which is actually the average mainstay ratio of the Mitsui Group, the lowest average among the six keiretsu. Likewise, the criterion for the high (low) non-mainstay ratio is that the industries’ non-mainstay ratios be equal to or less (more) than 77.4%, which is the average non-mainstay ratio of the Sanwa Group, the lowest average among the six keiretsu. The idea behind setting these criteria is to list as many industries as possible under both categories of high mainstay ratio and low non-mainstay ratio, and, thereby, to offer the maximum chance to find the industries overlapping in these two categories for each keiretsu. This process will prove helpful when checking how the one-set principle is robust within each keiretsu. Under such a generous setting, the overlapping sectors are 4 in Mitsui, 7 in Mitsubishi, 3 in Sumitomo, 5 in Fuyo, 5 in Sanwa, and 6 in DKB. Their ratios to the sectoral coverage are then, in order of the high ranking, 64% (7 out of 11 sectors) in Mitsubishi, 60% (6 out of 10 sectors) in DKB, 50% (4 out of 8 sectors) in Mitsui, 50% (3 out of 6 sectors) in Sumitomo, 45% (5 out of 11 sectors) in Fuyo, and 42% (5 out of 12 sectors) in Sanwa. Thus, in light of the criterion adopted above, Mitsubishi seems to be the least affected by various diversifications and to be the most faithful to the one-set principle. This way of measuring the robustness of the one-set principle, of course, involves some flaws. For instance, DKB’s second ranking should be evaluated with caution because its standard deviations are quite higher in both the mainstay and non-mainstay ratios (11.8 in the mainstay ratio, which is bigger than Mitsui [9.1] and Sanwa [10.2], 21.1 in the non-mainstay ratio, which is bigger than Mitsui [18.6], Sumitomo [13.4], and Fuyo [19.3]). This implies that, while DKB seemingly boasts of six sectors respectively listed in the high mainstay and low-mainstay categories, it has other sectors where diversification from the non-mainstay business intensively occurs. In fact, as 191 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. shown in Table 47, the entry from the electric machinry sector to the machinery sector is remarkable in the group: the electric machinry sector earned 7.6% of its total sales from the machinery sector, and those sales account for 48.6% of the aggregated sales in the machinery sector. DKB’s second ranking is largely owed to the methodology of the present analysis that tends to boost both the mainstay and non-mainstay ratios through the mere addition of the sales of multiple firms in the same sectors. Top-ranked Mitsubishi, too, has the highest standard deviation in the non-mainstay ratio (25.9%), but that owes much to the lower non-mainstay ratio of the petroleum sector (67.7%), which, in turn, was caused largely by the entries of other Mitsubishi firms from the chemical sector (Mitsubishi Kasei, Mitsubishi Gas, Mitsubishi Oil, Mitsubishi Plastics, Mitsubishi Rayon). For those Mitsubishi firms, it was quite natural to rely on the petroleum supply from Mitsubishi Petroleum, the one and only firm in the group licensed by the government to engage in the petroleum refining business. Considering also that of those five Mitsubishi firms, Mitsubishi Kasei, Mitsubishi Rayon, and Mitsubishi Oil which maintained affiliated relations, it is more relevant to view the competitive aspect of their relations as fairly reconciled by the sense of co-prosperity.2 2 Introducing different criteria might generate other interesting results. The prospective criteria are (1) the number of sectors listed in the high mainstay ratio relative to all the sectors shacho-kai members are engaged in, (2) the average value of the mainstay ratio, (3) the number of sectors listed in the low non-mainstay ratio relative to all the sectors shacho-kai members are engaged in, and (4) the average value of the non-mainstay ratio. The corresponding rankings by those criteria include the following: (1) Mitsubishi (90.9%) > Sanwa (75.0%) > Fuyo (72.7%) > DKB (70.0%) > Sumitomo (66.7%) > Mitsui (50.0%), (2) Mitsubishi (89.8%) > Sanwa (86.0%) > DKB (84.6%) > Fuyo (82.9%) > Sumitomo (82.5%) > Mitsui (80.4%), (3) Mitsui (87.5%) > Sumitomo (83.3%) > Mitsubishi (81.8%) > DKB (70.0%) > Sanwa (66.7%) > Fuyo (63.6%), and (4) Mitsui (90.5%) > Sumitomo (87.4%) > Mitsubishi (85.7%) > Fuyo (84.4%) > DKB (84.0%) > Sanwa (77.4%). 192 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Obviously, those rankings drastically differed from the first one, but there is one commonality: no matter which criterion is chosen, Mitsubishi constantly ranked higher than the three Bank Groups. Together with the outcomes derived in the previous chapters, the Mitsubishi Group seems to bear the closest resemblance to the Japanese Keiretsu Model. Mitsubishi’s distinctively solid inter-corporate ties are to be proved again in the course of financial liberalization that started in the 1980s. 3.4 The Bubble Economy and Its Aftermath: Equity Financing, Asset- Specificity, and the Main Bank System 3.4.1 The Status Quo o f the Economics Literature on the Bubble Economy The previous two sections showed that the two oil crises and the subsequent arrival of the low growth economy in Japan significantly transformed its industrial structure. The impact was, however, not limited to the industrial arena: ensuing to the oil crises was a series of financial liberalizations, which started as a response to the request on the part of industry to reduce various financial burdens on the widely-held recognition that there would no longer be lucrative business as in the high growth period. By focusing on the six keiretsu again, this section reinterprets the development of financial liberalization from the TCE perspective, and, thereby validates its contention under such a context. As of today, a large number of economists have ascribed the contemporary plight of the Japanese economy to the financial liberalization intensively pursued in the 1980s, coupled with the BOJ’s loose monetary policy adopted after the Plaza Accord of September 1985, identifying these factors as having encouraged the speculative behaviors of large firms and financial institutions. For example, Miyazaki (1992) and Noguchi (1992), the two notable protagonists of this view, argued that big enterprises’ tapping into equity financing—a form of direct financing by issuance of the convertible and warrant-attached bonds, which were made more attractive in the process of the 193 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. financial liberalization—and the banks’ reaction in the form of intensive financing in the small-and- medium sized firms and the real estate sector, together stirred up the speculative bubbles in both the financial and real estate markets, under the extraordinarily lenient monetary policy adopted by the BOJ. In a similar vein, Hoshi and Kashyap (2001) argued that the speculative bubble and the subsequent crisis of banks were chiefly ascribed to financial deregulation in the 1980s, yet with more emphasis on its slow and uneven nature of the financial deregulation, and with some doubt on the effect of the BOJ’s low interest rate policy on triggering the increase in real estate prices.2 3 In a much more theoretical approach, on the other hand, Aoki (1994c, 1995a) and Aoki and Okuno (1996) too singled out financial liberalization as the factor that deprived banks of their regulatory privileges and thereby narrowed their “main-bank rent,” and argued that, while trying to replenish their disappearing rent by consolidating the tie with their borrowing firms, the (main) banks failed to dissuade them from speculation. The present thesis does not intend to overturn their arguments, but points out that they missed one essential factor regarding how equity financing was processed in the keiretsu groups during the period. As Okumura (1993: pp. 12-4, 1996: pp. 78-80, 1998a: pp. 63-4, 1998b: pp. 52-62) recurrently pointed out, and Miyazaki (1992: pp. 164-9, pp. 214- 20) noticed in passing in his argument, equity financing in the latter half of the 1980s was put into practice often with cross-shareholdings among associated firms or within each keiretsu group.2 4 This was primarily because, as they explained, the issuing firms were nervously concerned with securing their stockholders, with the possibility in mind that their convertible and warrant-attached bonds would be eventually switched into stocks through the exercise of the attached options. Taking also into consideration the varying degrees of cross-shareholdings processed among the six keiretsu groups, the present thesis provides a consistent explanation of differing behaviors among these keiretsu groups and, in particular, their main banks, in the bubble economy and its aftermath. In doing so, it heavily relies on the central ideas of TCE applied to the Japanese keiretsu that (1) the more group firms are engaged in asset-specific, intra-keiretsu trades, the more willingly they partake 194 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. in cross-shareholdings with one another, (2) due to the nature of cross-shareholdings, which offset capital among the issuing firms, and, thus, does not serve as an effective means for raising capital, the issuing firms necessitate financing from their main bank, and (3) consequently, the intra-keiretsu transactions, cross-shareholdings, and the main bank financing keep pace with each other in a complementary fashion. In this regard, the case of the Mitsubishi Group deserves particular attention because it is the immaculate example of showing the positive interaction among the above three factors in the development of the bubble economy. Mitsubishi Bank’s abstaining from excessive financing in the real estate sector and its encouraging financing to the member firms in the late 1980s should be explained as a collorary of such an interaction. The other five main banks, which—in sharp contrast with Mitsubishi Bank—enormously financed the real estate sector and suffered subsequently from the huge bad loans, should also be understood in the same framework, but as a case in which the three factors failed to interact positively. To sum up, by placing equity financing into the organizational context of the keiretsu internal trade, the TCE perspective presents an emulating view on how the speculative bubble economy occurred and also explains why the main banks responded to it in varying ways. 3.4.2 The Periods o f Financial Liberalization and the Bubble Economy It has been well documented that corporate financing of Japanese firms gradually shifted from bank borrowing to the direct issuance of bonds, particularly in the Eurobond market, as they began to perceive the bank debts as quite onerous in the perpetuation of the low growing domestic economy, and that such a shift in corporate financing significantly contributed to creating the financial dis-itermediation between the financial and industrial sectors in Japan. First of all, as Rosenbluth (1989: pp. 137-66) showed in detail, such a shift in corporate financing of Japanese firms occurred mainly because the Eurobond market could offer the Japanese firms with good ratings a set of attractive flexible rate instruments and swaps that reduced interest 195 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. rates and hedged against exchange rate risks, and also because it did not impose any further conditions that were often required by the “Bond Committee” (kisai-kai) in the Japanese bond market, such as minimum mandatory capital, collateral, prospectus, and collateral assessment and custody fees. As a result, “beginning in the mid-1970s, the number of Japanese corporate bond issues in the Euromarket began to skyrocket” (Rosenbluth 1989: p. 149). In the stampede of Japanese firms to the Euromarket, Japanese authorities, including the IBJ, were forced to reconsider domestic issuance conditions and gradually relaxed them: in 1975, the Bond Committee adopted a policy of honoring the amount of bond issues as requested by each firm, although they still had to meet the domestic bond issuance criteria (Hoshi and Kashyap 2000: p. 229). In 1984, the MOF commissioned a Bond Study Group (Shasai mondai kenkyukai) to examine the domestic bond market, and the standard was successively relaxed in October 1985 and in March 1988 (Rosenbluth 1989: 159-63). Table 49 exhibits a comprehensive survey of the deregulation process of the domestic issuance conditions, along with the numbers of firms qualified under each condition. Over the course of bond market deregulation, the “warrant-attached bond” was introduced in 1981. It is a type of equity-linked bond whereby its subscribers, through the exercise of the option or “warrant” in a stipulated period, are entitled to purchase a new stock from the issuing firm at the 25 price determined when the bond is purchased (Miyazaki 1992: p. 134, pp. 155-6). Together with the convertible bond, which had been in place since 193 8,2 6 it prepared the way for Japanese firms to utilize “equity financing,”—the bond-issuing the issuance of new stocks in the future—which the late 1980s.2 7 196 corporate financing accompanying a possibility of was to culminate in the bubble economic period of R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 49. Transition of the Domestic Issuance Conditions and the Number of Firms Qualified Years Straight bond Convertible bond Secured Unsecured Secured Unsecured 1950’s Apr. 1959: Bond Committee started. 1960’s May 1963: the first revision of the restrictions. May, 1965: the second revision of the restrictions. May 1966: the third revision of the restrictions. 1970’s Oct. 1972: the fourth revision of the restrictions. Mar. 1979: the restrictions drawn. 15 billion yen or above. (2) Apr. 1970: the voluntary restrictions drawn. Mar. 1979: the restrictions drawn in parallel with the secured straight bonds. 150 billion yen or above. (2) Sep. 1976: the restrictions relaxed. 1980’s Jun. 1987: the fifth revision of the restrictions. 3 billion yen or above. (1060) Apr. 1984: the qualitative restrictions relaxed. (16) Jun. 1987: the restrictions relaxed. 3 billion yen or above. (1050) Jan. 1983: a new restrictions drawn. 110 billion yen or above. (25) Nov. 1988: the rating system introduced and co-existing with the traditional restrictions. BBB or above. (1300) Oct. 1985: the restrictions relaxed. 110 billion yen or above. (57) Nov. 1988: the rating system introduced. BB or above. (1400) Apr. 1984: the restrictions relaxed (97). Jul. 1987: the rating system introduced. AA. Or A or above and 55 billion yen or above. (180) Jul. 1985: the restriction relaxed. 33 billion yen or above. (175) Nov. 1988: the restrictions relaxed. 55 billion yen or above. Or AA. Or 33 billion yen or above and A or above. (300) Feb. 1987: the restrictions relaxed. 20 billion yen or above. Or A. Or 550 billion yen or above and BBB or above. (240) 1980’s Jul. 1987: the rating system introduced. (300) Nov. 1988: the restrictions relaxed. 20 billion yen or above. Or A. Or 33 billion yen or above and BBB or above. (500) 1990’s Nov. 1990: convergence to the rating system. BBB or above. Nov. 1990: convergence to the rating system. Nov. 1990: convergence to the rating system. BB or above. Nov. 1990: convergence to the rating system. BBB or above. the restrictions abolished. Source: Nihon Koshasai Kenkyujo (1995: p. 159). Note. The numbers in parentheses denote those companies qualified under respective restrictions/the rating system. The rating of this table belongs to the Japan Bond Research Institute (Nihon Koshasai KenkyQjo), which assessed the rating by nine stages (AAA, AA, A, BBB, BB, B, CCC, CC, C). 197 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. In addition to the domestic needs for deregulation, the “foreign pressure,” especially from the U.S., accelerated the deregulation process in the areas of issuance in the Euroyen bond market and foreign exchanges: based on the measures adopted by the “Yen-Dollar Committee” of 1984, the MOF relaxed the restrictions imposed on the issuance in the Euroyen bond market in 1984 and 1985 (Enkyo 1993: pp. 106-8; Miyazaki 1992: pp. 111-2; Rosenbluth 1989: p. 81, 159). Furthermore, in 1984, the MOF eliminated both the real demand principle and the restrictions on conversion between the yen and foreign currencies, which arranged for the options to utilize foreign currencies to the Japanese firms (Hoshi and Kashyap 2000: pp. 232-6; Miyazaki 1992: pp. 112-5; Rosenbluth 1989: p. 57). Particularly noteworthy of this deregulation is that it paved the way for Japanese firms to raise foreign currencies for their corporate financing purposes, via the issuance of warrant- attached bonds. All these deregulation processes, no doubt, lent momentum to Japanese firms switching from the traditional bank borrowing to the bond issuance in their corporate financing (see Tables 50 and 51). And this is exactly what the majority of economists pinpointed to be the source of the bubble economy of the late 1980s in Japan. 198 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 50. Funding Patterns of the Japanese Firms (% Distribution, 1975-95) Year- end Securities market Borrowed funds Total Equity Domestic bonds Foreign bonds Commercial Paper Total Private lender Public lender 1975 13.6 4.4 0.3 9.0 0.0 86.4 78.4 8.0 1976 13.4 4.3 0.5 8.5 0.0 86.6 78.6 8.0 1977 13.3 4.4 0.6 8.3 0.0 86.7 78.4 8.3 1978 13.5 4.5 0.7 8.3 0.0 86.5 77.9 8.6 1979 13.6 4.6 0.9 8.1 0.0 86.4 77.5 8.9 1980 13.3 4.6 0.9 7.9 0.0 86.7 77.6 9.1 1981 13.0 4.6 0.7 7.7 0.0 87.0 77.6 9.4 1982 12.9 4.4 1.0 7.5 0.0 87.1 77.6 9.5 1983 15.5 4.1 1.2 10.2 0.0 84.5 75.5 9.0 1984 15.6 4.0 1.4 10.1 0.0 84.4 75.7 8.7 1985 17.0 5.1 2.0 9.9 0.0 83.0 74.7 8.3 1986 17.7 5.1 2.3 10.3 0.0 82.3 74.6 7.7 1987 20.0 7.2 2.6 9.8 0.4 80.0 69.9 10.2 1988 21.3 6.8 2.9 9.8 1.8 78.7 68.8 9.9 1989 23.0 6.3 4.0 10.4 2.3 77.0 67.2 9.9 1990 23.5 6.1 4.8 10.1 2.5 76.5 66.6 9.9 1991 23.1 6.3 5.3 9.7 1.8 76.9 66.5 10.3 1992 22.8 6.- 4.9 ‘>.4 1.7 77.2 65.9 11.3 1993 22.0 7.2 4.0 9.3 1.5 78.0 65.6 12.3 1994 22.0 7.7 3.6 9.4 1.3 78.0 64.9 13.1 1995 22.2 8.2 3.1 9.5 1.4 77.8 64.6 13.2 Source: Hoshi and Kashyap (2000: p. 245), which derived the data from the Bank o f Japan, Economic Statistics Quarterly, various issues. Note. Entries are measured in the book value. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 51. Security Financing by Listed Firms (1972-98) Year Total securities (million yen) Distribution (%) Public stock offering as % total Foreign bonds Stock Bonds as % total as % total 1972 1,784,689 58.4 41.6 37.3 0.0 0.0 1973 2,240,766 41.9 58.1 25.2 0.1 0.2 1974 1,741,396 31.2 68.8 15.9 3.4 4.9 1975 3,187,449 31.4 68.6 6.9 11.7 17.0 1976 2,302,001 29.9 70.1 21.2 18.1 25.8 1977 2,543,740 36.3 63.7 23.7 14.7 23.1 1978 2,972,270 30.2 69.8 19.0 16.4 23.5 1979 3,298,028 28.9 71.1 19.1 24.0 33.8 1980 2,883,285 36.5 63.5 30.6 24.1 37.9 1981 4,400,028 43.8 56.2 31.7 19.8 35.2 1982 4,084,502 33.0 67.0 26.3 27.7 41.3 1983 4,048,420 19.8 80.2 11.6 44.8 55.9 1984 5,409,408 19.3 80.7 15.1 44.0 54.5 1985 6,890,503 12.5 87.5 7.3 51.0 58.2 1986 8,395,196 10.4 89.6 4.8 48.2 53.8 1987 14,455,291 20.8 79.2 9.6 39.9 46.6 1988 17,636,098 27.1 72.9 14.6 30.5 41.9 1989 28,410,407 31.1 68.9 20.5 41.4 60.0 1990 14,441,448 26.3 73.7 13.7 35.8 48.5 1991 12,500,454 6.5 93.5 1.0 63.5 67.9 1992 9,619,910 4.4 95.6 0.0 60.2 62.9 1993 11,143,567 7.4 92.6 0.1 45.0 48.6 1994 8,499,604 11.0 89.0 1.6 22.6 25.4 1995 8,094,650 7.9 92.1 0.4 23.2 25.1 1996 13,616,878 15.2 84.8 2.2 18.3 21.6 1997 10,162,545 11.4 88.6 1.3 23.4 26.4 1998 15,906,750 9.7 90.3 1.8 10.0 11.1 Source: Hoshi and Kashyap (2000: p. 235), which derived the data from the Tokyo Stock Exchange, Annual Securities Statistics, 1998. Note. This excludes the proceeds from warrant exercise, offerings to current shareholders, private placements, and preferred issues. 200 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. The firms and banks of the six horizontal keiretsu were, of course, not free from those deregulation processes. While there still remained collateral requirements in the declining forms, the main banks of the six keiretsu came to play a significant role as “bond guarantors” and/or “trustee banks” (Jutaku ginko) for their member firms now eagerly issuing convertible and/or warrant- attached bonds in place of bank borrowings. Tables 52 and 53, which are both the replications of Campbell and Hamao (1994: pp. 347-8), reported the states of bond trustee administration by the six main banks for five chosen periods (1970-75, 1976-79, 1980-83, 1984-87, and 1988-91. In the case of warrant-attached bonds, two periods of 1984-87 and 1988-91 only) provide answers to the following two questions: (1) Out of the bond issues trusteed by bank A, what proportion is for firms that have bank A as their main bank? And (2), out of the trusteed bonds issued by a firm with bank A as its main bank, what proportion is trusteed by bank A? Table 52. Bond Trustee Administration (Answer to the First Query) Straight bond Convertible bond Warrant 1970 -75 1976- 79 1980- 83 1984- 87 1988 -91 1970- 75 1976- 79 1980- 83 1984- 87 1988 -91 1984- 87 1988- 91 DKB 70 83 100 95 0 53 92 60 50 64 80 100 Sakura Bank 95 97 98 100 100 97 50 49 75 67 100 100 Mitsubishi Bank 72 80 73 100 100 98 100 99 93 75 100 100 Fuji Bank 60 79 100 94 100 63 92 74 56 61 0 100 Sumitomo Bank 96 97 98 95 0 91 76 83 85 94 0 76 Sanwa Bank 53 74 66 69 0 60 94 80 37 53 100 100 Source: Campbell and Hamao (1994: p. 347). Note. This table gives the answer to the question: Out of the trustee bonds issued by a firm that has bank A as its main bank, what percentage (of the total amount) is trusteed (as the first trustee) by bank A? 201 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 53. Bond Trustee Administration (Answer to the Second Query) Straight bond Convertible bond Warrant 1970 -75 1976- 79 1980- 83 1984- 87 1988- 91 1970 -75 1976- 79 1980- 83 1984- 87 1988 -91 1984- 87 1988 -91 DKB 98 91 98 6 0 88 100 87 73 83 100 100 Sakura Bank 96 97 98 100 100 72 49 42 50 47 100 100 Mitsubishi Bank 75 77 82 51 100 66 63 71 57 63 58 25 Fuji Bank 40 26 45 100 100 61 73 47 65 60 0 75 Sumitomo Bank 73 79 71 100 0 77 69 56 62 53 0 100 Sanwa Bank 92 94 98 97 0 81 96 70 49 76 100 61 Source: Campbell and Hamao (1994: p. 348). Note. This table gives the answer to the question: Out of the bond issues trusteed by bank A (as the first trustee), what percentage (of the total amount) is for firms that have bank A as their main bank? Although the firms incorporated in Campbell and Hamao (1994) are those in which the largest quantity of loans is made by one of 19 major banks—“the main bank firms” (Campbell and Hamao 1994: p. 325) in their terminology—and, thus, are much broader in number than the shacho- kai member firms considered in the present thesis, the data are still useful for gaining an insight into the inter-firm and bank-firm relations of the six keiretsu. The first to be noticed in Tables 52 and 53, as pointed out by Campbell and Hamao (1994: p. 345), is that the trustee role of the main bank 28 alliance was not disturbed very much by the course of the bond market deregulation. From this observation, they concluded that, in general, banks remain likely to trustee the bonds issued by their main bank clients, and that if a bond is trusteed, it is likely to be done by the issuer’s main bank, rather than other banks (Campbell and Hamao 1994: p. 345). But even more important, particularly from the TCE perspective, is to see through the differences in the degree of the inter-firm connection across the six keiretsu, hidden behind the data on main banks’ trustee administration of convertible and warrant-attached bonds, which, within certain periods, were to be transformed into 202 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. stocks and distributed in varying degrees, among affiliated firms in the form of cross-shareholdings. Granted, in Tables 52 and 53, there is no information of the addressees of the convertible and warrant-attached bonds issued whereby to grasp precisely the effect of issuing them on cross shareholdings among members. Yet, by reading Tables 52 and 53 simultaneously, it is possible to see through a few distinctive issuing patterns among the six main banks, especially in the case of the convertible bond, which turned out to be concordant with the analyses in the previous chapters: (1) In any period, Mitsubishi and Sumitomo Banks trustee almost all convertible bonds issued from their main bank firms, and still could afford to trustee those from non-main bank firms. This result is interpreted as the reflection of not only the strong ties between these two banks and their own main bank firms and their enormous financial capability to guarantee the trusteed bonds in the case of default, but also the solid inter-firm exchange networks of the Mitsubishi and Sumitomo Groups. This is because, given the dense intra-keiretsu exchanges, which these two banks can easily confirm through their members’ checking accounts (toza yokiri), the debt incurred by one member is relatively easily replaced by its credit(s) to other member(s), and, thus, the probability of losses due to defaults is significantly lowered. In such an exchange circumstance, the bank can, in relatively easy terms, let their members issue the bonds to other members in the same keiretsu group. Besides, the dense intra-keiretsu trades have the advantage of minimizing the outflow of funds from these banks. All these conditions allow the banks more financial leeway and, thus, provide them with the incentive to trustee more of the bonds issued from their members. In the case when an addressee is engaged in transaction with the issuer and expects that the transaction will perpetuate, its convertible bonds are switched into stocks on the due dates, and thereby constitute part of the intra-keiretsu 29 cross-shareholdings. (2) Sakura (Mitsui), Fuji, and Sanwa Banks do not trustee convertible bonds from their main bank firms so much as Mitsubishi and Sumitomo Banks. But that does not mean that these three banks could not afford to trustee more of those from their main bank firms: actually, they considerably trustee those issued from non-main bank firms. This implies that the degrees of the inter-firm connections between these three banks and their main bank firms are generally weak. 203 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. The reason is opposite to (1) above: in the case of the Mitsui, Fujo, and Sanwa Groups, where the intra-keiretsu exchanges are not so intense and where the members are often engaged in transaction with non-member firms, their main banks cannot easily see that the debt incurred by the members will be replaced by its credit(s) to other member(s), through a dense interior exchange network. Also, these banks may feel hesitant to trustee bonds when the addressee is a non-member because of their concerns about the leakage of their funds to outsider(s) in the future. The main bank, if put in such an exchange circumstance, likely makes its decision rather on a case-by-case basis, just caring for the repayment ability of an issuing firm and the likeliness of the inflow of funds to the bank in the future. (3) Dai-Ichi Kangyo Bank does not trustee convertible bonds so much as Mitsubishi and Sumitomo Banks, although it trusteed most of the convertible bonds issued from its main bank firms. This situation most likely reflects that due to the extremely large number of member firms, Dai-lchi Kangyo Bank cannot satisfy all trustee requests, even if the bank restricts the trustee business within the group members.3 0 The new way of raising capital, which enabled firms to get around the traditional main bank borrowing for better borrowing terms notwithstanding, the keiretsu firms soon faced the problem recurrently besetting them since the early days of their initial public offerings: who is going to accept those convertible and warrant-attached bonds, which, sooner or later, will be matured into stocks? (Miyazaki 1992: p. 164) To this problem, the majority responded in the conventional way by allocating the stocks within the same keiretsu firms in line with cross-shareholdings (Miyazaki 1992: pp. 165-9; Okumura 1993: p. 12-3, 15, 1996: p. 74, 86, 1998: pp. 57-8). This was, indeed, an aspect unduly neglected in the past literature of how the speculative bubble evolved expansively in the late 1980s in Japan. Yet, still missing is the fact that, despite their common inclination to gamer the members’ stocks inside their groups, their pace in doing so significantly differed from each other. This is clearly shown in Table 54, which presents the data of cross-shareholdings processed in each of the six keiretsu in the three chosen years of 1980, 1985, and 1990. 204 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 54. Cross-Shareholding in the Six Keiretsu (1980,1985,1990) Year Mitsui (Nimoku-kai) Mitsubishi (Kin’yo-kai) Total stocks issued Rate of cross shareholding in the group Number of cross shared stocks in the group Total stocks issued Cross shareholding rate in the group Number of cross shared stocks in the group 1980 13,655.8 17.62 2,406.1 16,752.6 29.26 4,901.8 1985 16,686.8 17.62 2,940.2 19,272.8 25.18 4,852.9 1990 21,294.8 16.54 3,522.2 24,457.6 26.89 6,576.6 1985/1980 (%) 22.2 22.2 15.0 -1.0 1990/1985 (%) 27.6 19.8 26.9 35.5 Year Sumitomo (Hakusui-kai) Fuyo (Fuyo-kai) Total stocks issued Rate of cross shareholding in the group Number of cross shared stocks in the group Total stocks issued Cross shareholding rate in the group Number of cross shared stocks in the group 1980 11,161.8 26.74 2,984.7 18,012.4 16.26 2,928.8 1985 13,409.5 25.01 3,353.7 21,200.5 15.79 3,347.5 1990 16,247.3 24.06 3,909.1 26,165.4 15.44 4,039.9 1985/1980 (%) 20.1 12.4 17.7 14.3 1990/1985 (%) 21.2 16.6 23.4 20.7 Year Sanwa (Sansui-kai) DK.B (Sankin-kai) Total stocks issued Rate of cross shareholding in the group Number of cross shared stocks in the group Total stocks issued Cross shareholding rate in the group Number of cross shared stocks in the group 1980 18,168.1 16.78 3,048.6 22,465.7 14.2 3,172.2 1985 21,173.5 16.84 3,565.6 26,542.6 13.33 3,538.1 1990 26,104.0 16.40 4,281.1 32,156.3 12.06 3,878.0 1985/1980 (%) 16.5 17.0 18.1 11.5 1990/1985 (%) 23.3 20.1 21.1 9.6 Source: Miyazaki (1992: p.167). Note. The original sources were TO yO Keizai Shinposha, Kigyd Keiretsu Soran (1987, 1992). R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. While it shows that six keiretsu groups overall experienced an increase in the rate of the total stocks issued in 1985-90, Mitsui and DKB experienced declines in the rate of cross shareholdings in the group in the period. This implies their weaker inter-firm connections than the other four keiretsu groups. Furthermore, among those four keiretsu, Sumitomo, Fuyo, and Sanwa respectively saw that their number of cross-shared stocks in the group shrink compared to their rate of increase in the total stocks issued. In this regard, they can be viewed as having experienced declines in the rate of intra-keiretsu stockholdings. The exception to the above tendencies is Mitsubishi from 1985-90, in which the entire group absorbed more stocks issued than those newly issued, by Kin’yo-kai members. This is apparently the manifestation of the most solid inter-firm associations within the Mitsubishi Group. The data of Table 54 thus indicate that the inter-firm relations that were analyzed previously were, in general, carried over to each keiretsu in the 1980s. Those differing states of cross-shareholdings across the six keiretsu then came to impact the way their main banks finance in a crucial fashion. In the case, for example, where members are generally under close contact through transactions, and, thus, bound firmly by cross-shareholdings, as in the Mitsubishi Group, the main bank is more likely poised to promote financing toward its members in order to encourage joint-production among them. This is primarily because cross shareholdings, while solidifying the intra-keiretsu trade relations, do not, in essence, increase the paid-in capital among member firms, and because there emerges a situation where those members starve for capital injection from their main bank. Okumura (1993: p. 15), for example, explained the negating impact of cross-shareholdings on capital consolidation: When company A increases its paid-in capital by 100 million yen and company B pays up this amount, and company B, in turn, increases its paid-in capital by the equal amount that is to be paid up by company A, the paid-in capital of each of these companies will be increased by 100 million yen, without their making 1 yen of real investment. If this is done repeatedly, there is no limit to the extent to which the paid-in capital can be puffed up, but, in reality, no investment will be made at all.3 1 206 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. This is exactly the situation where the main bank is called upon to increase fresh funds into the member firms so that they can maintain the intra-keiretsu transactions. Sheard (1994b: p. 323) summarized such complementarity between cross-shareholdings and the main bank financing in the following: [M]aking financial and their associated control instruments illiquid may enhance their commitment value. Interlocking shareholdings based on stable shareholdings arrangements are shareholdings, which, in a sense, have been made more illiquid. Bank loans, particularly when there exists no secondary market for their trading, are another example of an illiquid instrument. A bank loan can be viewed as an instrument that allows the supplier of funds credibly to commit itself to certain actions, e.g., to not withdraw its funds or (in the case of non-variable interest rates) to not alter the cost of funds, even should new information arrive. In the opposite case where trade association among members is sparse, as in the Mitsui and the Bank Groups, the main bank is most likely forced out of the traditional lending business with its member firms. The member firms, which are not so much engaged in transactions with other large group firms and are rather interested in developing the trade relations with their subsidiaries, are more motivated to take the action to utilize equity financing for better borrowing terms. Facing the loss of customers even from its own shacho-kai, on the other hand, the main bank will be pressed to seek out new ones from the non-traditional keiretsu lines or may be tempted to venture into the financing of the new sectors such as real estate and construction. The above predictions prove quite accuare. As is well known today, during the late 1980s, Japanese banks made an unprecedented level of lending in the three non-production sectors of real estate, construction, and commerce (Miyazaki 1992: pp. 139-41, p. 170; Noguchi 1992: pp. 125-31; Okumura 1999: pp. 133-6). Obviously, such anomalous bank lending, alongside equity financing on the part of non-financial firms, was one of the main factors for the creation of the bubble economy. Admittedly, such anomalous bank lending, along with equity financing on the part of non-financial firms, added fuel to the exploding real estate prices, contributing significantly to the formation of the bubble economy. A close look into the lending records of each of the six main banks, however, 207 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. reveals that their lending attitudes towards these three sectors during the period were not uniform. Figures 32 to 34* respectively, display the lending ratios of the six main banks in each of the three sectors (construction, real estate, commerce), and Figure 35, their aggregated lending ratios in the three sectors, over the period of 1985-92.3 2 Lending Ratio in the Construction Sector 16 14 12 10 8 6 4 2 0 1984 1985 1986 1987 1988 1989 1990 1991 1992 Year ■ ♦ — Mitsui - - B - - Mitsubishi -A— Sumitomo Figure 32. Lending ratio in the construction sector. Lending Ratio in the Real Estate Sector 8 7 6 5 4 3 2 1 0 1984 1985 1986 1987 1988 1989 1990 1991 1992 ■ • Mitsui Mitsubishi Sumitomo Figure 33. Lending ratio in the real estate sector. 208 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Lending Ratio in the Commerce Sector 35 3 0 2 5 20 1 5 10 5 0 1 9 8 4 1 9 8 5 1 9 8 6 1 9 8 7 1 9 8 8 1 9 8 9 1 9 9 0 1 9 9 1 1 9 9 2 — 0— Mitsui - ■ — Mitsubishi ■ -A - - Sumitomo — X— Fuyo - * — Sanw a - • — DKB Y ear Figure 34. Lending ratio in the commerce sector. Lending Ratios in the Three Non-Production Sectors 5 0 4 5 4 0 - o - 3 5 3 0 2 5 20 1 5 10 5 0 1 9 8 4 1 9 8 5 1 9 8 6 1 9 8 7 1 9 8 8 1 9 8 9 1 9 9 0 1 9 9 1 1 9 9 2 — ♦— Mitsui — ■— Mitsubishi — A— Sumitomo — K— Fuyo Sanwa — o— DKB Year Figure 35. Lending ratios in the three non-production sectors. 209 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. It is readily verified from Figures 32-34 that Mitsubishi Bank’s lending ratios in the construction and commerce sectors are much smaller than other banks during the whole period. Most of all, its lending ratio in the three sectors (Figure 35) recorded the lowest value in five years and only minutely higher values (in the range of 0.05% to 0.48%) than Sanwa Bank in the rest of the chosen years. Mitsubishi Bank’s abstaining from financing too much in the three sectors is actually correlated with its constantly increasing lending to Kin’yo-kai members during 1985-91 (Figure 25).3 3 Sanwa Bank and Dai-Ichi Kangyo Bank are also shown to have refrained from financing extravagantly in the three sectors. But, probably as a result of their rather incoherent inter-firm transaction networks, Sanwa’s financing in the construction sector and Dai-Ichi Kangyo’s financing in the commerce sector were somewhat higher. Consequently, their lending ratios in the three sectors were often higher than Mitsubishi Bank’s. Diametrically opposite to Mitsubishi Bank were Mitsui, Sumitomo, and Fuji Banks, which all decreased lending to their members, while increasing financing in the three sectors during the period. As they saw their lending to their shacho-kai members gradually diminishing over the course of the late 1980s, they targeted their lending toward the construction sector (in the case of Sumitomo Bank), the real estate sector (in the case of Mitsui Bank), and the commerce sector (in the case of Fuji Bank), eventually arriving at quite high lending ratios in all three. It is a bit unexpected, however, that Sumitomo Bank’s lending ratios were, in general, quite higher in all three sectors, and that, in particular, its aggregate lending ratio in these sectors constantly hit the highest value among all the six main banks, despite the Sumitomo Group’s remarkably high levels of interior cross-shareholdings. This puzzle can be best solved by paying attention to non-financial Hakusui-kai members’ extremely low level of holding Sumitomo Bank’s stocks (see Figures 6 and 7). Because of this, Sumitomo Bank was not properly governed by other Hakusui-kai members, and, instead, let President Ichiro Isoda sway the Bank for a long period of 14 years and go on speculative businesses, such as absorption of Heiwa Sogo Bank in 1986, intensive 210 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. land speculation, and unscrupulous financing to Itoman, a medium-sized trading house, in the late 1980s. In the end, the bank infringed Sumitomo’s long-tradition of not pursuing bubbles, which was first decreed by Sumitomo’s family constitution drafted in the early days of the zaibatsu era.3 4 3.4.3 The Post-Bubble Economy Period The over-swollen bubbles were doomed to crash in any event whether they were of the ever-soaring stock market or of banks’ reckless financing in the three non-production sectors. The decelerating impact was first given by the Bank of Japan (BOJ) which, leery of the domestic inflationary pressures and the weakening of the yen, decidedly raised its official discount rate four consecutive times from May 31, 1989 to March 20, 1990. The rate consequently jumped from 2.5% to 5.25%, and the stock prices dropped in response. On April 2, 1990, the Nikkei Stock Average suddenly plummeted to 28,002.07 yen,3 5 and, after a short lull, on October 1, 1990, further plunged to 20,221.86 yen, a 49 % decline of the previous year’s record. Nose-diving stock prices exerted a widespread disastrous impact on the economy. First, many firms began to suffer the bulk of the remaining convertible and warrant-attached bonds that they had issued during the bubble period, because the actual stock prices went far below the option-exercising stock prices of these bonds. The firms were now in the position to redeem those bonds, utilizing their own reserved funds or, in some cases, even issuing straight bonds at higher interest rates. As a matter of course, this redemption process significantly sapped firms of their investment potentials. The doldrums of the stock market also affected other stockholders such as banks and individuals, with the former shrinking their lending in observing the BIS capital ratio regulation, and with the latter tightening their purse strings while disposing of their assets accumulated during the bubble period.3 6 Another decelerating impact was given by the MOF, which, in April 1990, restricted the growth of real estate lending by each bank to no more than the growth in its overall loan portfolio (soryo kisei). The new measure 37 squelched swelling land prices, but with an unexpected side-effect of their incessant dropping, 211 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. followed by a series of bankruptcies in the real estate industry and the deterioration of the real estate collateral submitted to the banks. The euphoria of the frenzy bubble became thwarted, and the economy took a reactionary path, entailing unprecedented events, such as occasional financial panic and prolonged depression during the 1990s. Among all, the growing piles of soured loans, concentrated in the three non-production sectors, became pervasive and aggravating over time across Japanese banks, debilitating their profit-earning abilities. Their sluggish continuation of lending to troubled firms, which were reportedly en masse in the three non-production sectors, often resulted in augmenting their bad loans. Their liquidation of troubled firms, on the other hand, contributed to creating deflationary pressure on the economy, and, at the same time, seriously harmed their capital. The incessant deflationary process further ate into their capital through precipitation of the value of stocks they held for cross-shareholdings. Consequently, under the BIS capital-ratio requirement, banks were often compelled to refrain from financing small and medium sized firms, which further accelerated the deflationary pressure on the economy and their accumulation of bad loans. Japanese banks are, in general, still struggling in this vicious circle. Interestingly, however, the seriousness of the bad loan problem actually differed significantly among the six keiretsu main banks. Notably, the bad loan problem was less serious in Mitsubishi Bank: for instance, in the early 1992 period, when the majority of the Japanese banks evaded the problem and chose a forbearance policy, Mitsubishi Bank, confident in settling its bad loan problem quickly, proposed boldly to the MOF that, in order to relieve the public distrust, all Japanese banks should disclose their bad loan records (Nihon Keizai Shimbunsha 2001: p. 290).3 8 In fact, in the rest of the 1990s, Mitsubishi Bank kept its bad loan ratio lower, and its capital-asset ratio higher, than others with the least support from taxpayer money and with no reliance on a merger. In sharp contrast with Mitsubishi, the other five banks which long suffered from the increasing bad loans took more taxpayer money and were finally forced into mergers. As the bad loan problem substantially mirrors banks’ reckless lending in 212 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. the three non-production sectors during the bubble economy period, the examination of the problem is considered a credit to the TCE perspective. The first event that brought the seriousness of the bad loan problem of Japanese banks to the public eyes was the so-called “jusen” problem, where the bulk of the loans made by banks, the agricultural cooperatives, and other financial institutions to seven home mortgage loan companies {jutaku senmon kin ’ yu gaisha. See the first column of Table 55 for those companies.), often referred to as “jusen” in abridgement, turned out to be unrecoverable in the mid-1990s. Table 55. The Loans to the Jusen Companies by the Six Main Banks (1995) Jusen Companies Sakura Bank (Mitsui) Mitsubishi Bank Sumitomo Bank Fuji Bank Sanwa Bank DKB Total bad loans/ total lending Nippon Flousing Loan *255.2 0 0 0.1 *258.0 0 1,674.3/ (6/1971) (5%) (0%) (0%) (0%) (5%) (0%) 2,257.4 Housing Loan Service *50.1 *50.1 *50.1 *50.1 0 *50.1 1,083.3/ (9/1971) (5%) (5%) 5%) (5%) (0%) (5%) 1,419.6 Juso 9.0 0.5 5.8 0.3 3.0 0.7 1,290.7/ (10/1971) (0%) (0%) (0%) (0%) (0%) (0%) 1,609.4 Sogo Jukin 0 1.0 4.1 0 0 0 960.6/ (7/1972) (0%) (0%) (0%) (0%) (0%) (0%) 1,118.3 Dai-Ichi Housing Loan 1.0 1.2 4.8 5.0 5.0 7.0 991.4/ (12/1975) (0%) (0%) (0%) (0%) (0%) (0%) 1,505.8 Chigin- Seiho Jutaku 0.2 16.4 24.8 0 0 0.2 695.1/ Loan (6/1976) (0%) (0%) (0%) (0%) (0%) (0%) 877.9 Japan Housing Loan 0 4.0 0 11.2 47.0 0 1,436.7/ (6/1976) (0%) (0%) (0%) (0%) (0%) (0%) 1,931.2 Sum 315.5 73.2 89.6 66.7 313.0 58.0 8,132.1/ 10,719.6 Sources: Mihaupt and Miller (1997: p. 25), Nihon Keizai Shimbunsha (1995: p. 207, 2000: p. 25, 37, 51) Note. The numbers in the parenthesis under the name of a jusen company represents the month and year of its establishment. The unit is billion yen. The percentage in the parentheses represents the rate of shareholding. The lending amounts are as of March 1995. The total bad loans are as of August 1995, when the second on-site inspection by the MOF was completed. The total bad loan is the summation of the following three types, categorized by the “Prompt Corrective Action” introduced in June 1996: (a) Category II: the loan which fails to satisfy some of the conditions for recoverability so that it is uncertain to be recoverable in the future, (b) Category III: the loan which is considered as hard to recover in the future, (c) Category IV: the loan which is no longer retrievable. An asterisk (*) indicates the founding institution of the jusen company concerned. For comprehensive data on the founding and non-founding institutions and their lending amounts to the Jusen companies, see Nihon Keizai Shimbunsha (1995: p. 207). 213 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. The jusen companies were established successively during 1971-76 by banks (ordinary banks, trust banks, and the three long-term credit banks) and other financial institutions, under the guidance and encouragement of the MOF, for the chief purpose of meeting the individual housing loan demand, which Japanese banks, at the time, could not fully satisfy in their preoccupation with corporate financing. Despite the MOF’s formal intent, however, the foundation of some jusen companies seems to have more connection with the keiretsu inter-corporate relations of the founding banks. One case in point was Sanwa Bank’s taking the lead in establishing the first jusen company, “Nippon Housing Loan,” in June 1971 (Milhaupt and Miller 1997: p. 24), and Mitsui Bank’s involvement in founding two jusen companies, the aforementioned Nippon Housing Loan and “Housing Loan Service” in September 1971: hardly able to find the opportunity for corporate financing within their own keiretsu, Sanwa Bank and Mitsui Bank were eager to seek other financial outlets, like jusen companies, to make full use of their funds. Mitsui’s involvement in founding two jusen companies was actually peculiar, as all other banks joined in establishing only one jusen company. These two banks and the two related jusen companies were later to encounter the direst financial consequence when the bubble economy came to an end, and the jusen problem subsequently surfaced. The origin of the jusen problem is primarily ascribed to the regulatory blank over the companies: not being deposit-taking institutions, they fell outside the MOF’s jurisdiction and thus were freed from any obligation to report their lending records to the MOF. Correspondingly, their founding banks evaded the regulation of lending limits imposed on jusen companies and wrote off their nonperforming assets out of the books. Joining in this regulatory lacuna was the ever-growing financial supply from the agricultural cooperatives, which were granted the approval from the Ministry of Agriculture, Forest, and Fisheries (MAFF) to finance the jusen companies in 1980. As the low-growth paradigm set in the domestic economy, and the founding banks started to reinforce their home mortgage loan business in compensating for shrinking corporate lending, the jusen 214 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. companies, to their disadvantage, targeted small and medium-sized firms in the real estate industry. Under the favorable climate of souring land prices in the late 1980s, jusen’s new lending policy was seemingly successful until the MOF introduced the aforementioned lending regulation on the real estate: its squelching souring land prices, while not interfering with ongoing lending by the agricultural cooperatives, only contributed to generating the massive nonperforming loans in the jusen companies. This financial fiasco triggered fierce battles between the founding banks and the agricultural cooperatives over the loss allocation of the jusen companies in late 1995. In December 1995, a compromise was finally reached by introducing the “modified founder liability” as the loss allocating scheme, dictating all founding institutions to write off their bad loans and all other lenders to bear losses in proportion to their loans. A taxpayer fund of 680 billion yen was used for making 39 up the shortage resulting from the scheme. Although the loss-covering contribution that each bank had to bear was not officially publicized, it is conjectured from the data of lending made by the six keiretsu main banks by 1995 (see Table 55) and the modified-founders-liability principle that Mitsui Bank and Sanwa Bank must have incurred disproportionate contributions compared with others. This result indicates that the problem of losing corporate customers was particularly serious in Mitsui and Sanwa, as their loans to jusen companies were continued until the MOF introduced the lending regulation in April 1990 (see, again, Figure 24 for the intra-keiretsu borrowing ratio. Mitsui Group’s sudden trough in the ratio was especially phenomenal). The jusen incident was only the first-round of the bad loan problem: already in 1994, bad loans of some credit cooperatives and regional banks cornered them to bankruptcy.4 0 In March of the following year, due to its disposing of bulky bad loans, Sumitomo Bank was compelled to report an annual loss of 820 billion yen, the first annual loss recorded by a city bank in postwar banking history. As previously noted, the bad loan problem arose primarily from banks’ speculative financing in the three non-production sectors during the bubble period, of which financing by the jusen companies was only part. Besides, the problem turned out to be omnipresent throughout the 215 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Japanese banking sector. Caution, however, is in order. Despite the general outlook of the banking sector being exacerbated by the growing bad loans, the seriousness of the issue was never uniform across individual banks. As for the six keiretsu main banks, the bipolarity between Mitsubishi Bank and others existed in the burden of the bad loans from the early 1990s, and such bipolarity became much more pronounced in the second half of the 1990s. That is clearly shown in Table 56 reporting three key variables of measuring the sound banking operation (the amount of bad loans, the bad loan ratio, and the capital-asset ratio) for the six keiretsu main banks over all the seven years starting from 1995 or 1996 4 1 Agreeing with the TCE perspective, according to Table 56, in 1995 and 1997, Mitsubishi Bank outperformed others in both bad loan and capital-asset ratios. Obviously, this result obtains from the fact that Mitsubishi Bank restrained itself from the heavy financing toward the three non production sectors throughout the bubble period. Furthermore, the enormous latent profits (fukum ekt) accruing from Mitsubishi Bank’s stockholdings (see Table 57 for Mitsubishi’s latent profits in 1994 and 1996) prepared the ample reserves for the bank to deal with the bad loans 4 2 216 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 56. The Bad Loans, the Bad-Loan Ratios, and the Capital-Asset Ratios of the Six Main Banks Main Bank 1995 1997 1998 1999 2000 2001 2002 Sakura Bank (Mitsui) Bad Loan 1,932.8 1,775.5 1,475.4 N/A 1,672.4 1,215.1 *5,169.5 B.L. Ratio 10.4% 4.8% 3.98% N/A 5.24% 4.0% *7.51% C.A Ratio 8.37% 8.92% 9.12% 12.53% 12.53% 11.31% *10.42% Mitsubishi Bank Bad Loan 1,156.1 1,156.5 2,250.2 5.10% N/A 4.52% 1,723.3 3,194.1 1,928.8 B.L. Ratio 6.1% 2.6% 4.86% 9.27% 5.5% C.A Rano 9.93 % 9.28% 8.53% 11.46% 11.46% 9.69% 10.43% Sumitomo Bank Bad Loan 1,487.6 1,074.2 1,469.1 N/A 1,672.4 1,561.6 See(*) above. B.L. Ratio 8.4% 2.9% 3.91% 5.80% 6.0% 4.9% C.A. Ratio 9.2 % 8.75% 9.23% 11.60% 12.53% 10.94% Fuji Bank Bad Loan 1,712.5 1,839.6 1,692.7 N/A 1,295.9 1,172.0 **2,127.6 B.L. Ratio 13.2% 5.4% 4.93% N/A 4.14% 3.70% **5.61% C.A. Ratio 8.36% 9.22% 9.41% 11.00% 11.00% 10.81% **9.47% Sanwa Bank Bad Loan 1,157.0 1,245.4 1,287.6 N/A 1,251.7 1,215.4 3,400.3 B.L. Ratio 5.7% 3.5% 3.50% N/A 4.16% 4.09% 6.98% C.A. Ratio 9.12% 9.10% 9.60% 12.25% 12.25% 10.51% 12.00% DKB Bad Loan 1,378.7 1,434.6 1,471.4 N/A 1,726.1 1,641.0 See(**) above. B.L. Ratio 6.4% 3.9% 4.03% 5.9% 5.32% 5.2% C.A. Ratio 9.40% 8.75% I 9.08% 12.11% 12.11% 11.58% Sources: Mainichi Shimbunsha, Shukan Ekonomisuto (6/26/2001), Ekonomisuto Special Issue (October 13, 2003: p. 99-100); Kyuno (2000: pp. 52-3); Nakamura (1997: p. .87, 91, 97); TO yO Keizai ShinpOsha, .Shukan Toyo Keizai (October 26, 1996: p. 33). Note. The unit of the bad loan is billion yen. “B.L. Ratio” and “C.A. Ratio” stands respectively for “the bad loan ratio” and “the capital-asset ratio.” The calculation of the bad loan and the bad loan ratio are based on the “risk management bad loan,” defined as the summation of the following four types of bad loan: (a) the bad loan to the failed firms, (b) the bad loan whose interest payment was deferred for more than 6 months, (c) the bad loan whose interest payment was revised to be lower than the BOJ’s official bank rate, and (d) the bad loan to the firm the bank decided to support by giving up part of its loan. Note that, by February 1996, the risk management bad loan was defined as the summation of (a) and (b) only. The bad loans in the table do not include those made to the jusen companies. The records of Mitsubishi Bank for the years after 1997 are of the Bank of Tokyo-Mitsubishi Ltd. (*) and (**), respectively, indicate the records of Sumitomo-Mitsui Bank and Mizuho Financial Group. 217 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 57. The Latent Profits from Stockholdings of the Six Main Banks Main Bank March 1994 March 1997 Sakura Bank (Mitsui) 763.4 (-59.3) 764.7 Mitsubishi Bank (The Bank o f Tokyo-Mitsubishi in March 1997) 878.8 (-47.3) 1,601.7 Sumitomo Bank 406.5 (-70.0) 688.8 Fuji Bank 343.5 (-74.4) 385.0 Sanwa Bank 795.4 (-48.7) 719.0 DKB 627.2 (-56.7) 770.5 Sources: Nihon Keizai Shimbunsha (1995: p. 89), Nakamura (1997: p. 93). Note. The unit is billion yen. The negative value in the parenthesis demonstrates the rate o f decrease in the latent profits compared with the previous year. Mitsubishi Banks’ highest latent profits from its stockholdings imply that the stocks the bank held were very unlikely sold off, which further indicates the tight cross-shareholdings among the Mitsubishi Group. Taking advantage of all those favorable conditions, in March 1995, Mitsubishi Bank embarked on a “surprising” merger with Tokyo Bank, reorganizing itself into “the Bank of Tokyo-Mitsubishi Ltd.” Mitsubishi’s intent for the merger primarily lay in reinforcing its foreign exchange network by adding the one developed by Tokyo Bank since 1947 under the status of Japan’s special bank licensed to engage in the foreign exchange business, and thus was opposite in nature from Sumitomo’s merger with Heiwa Sogo Bank (Nihon Keizai Shimbunsha 1995: p. 12, 22-3, 28; Kishi et al., and Gendai Kin’yu Mondai Ken’kyu Gurupu 1995: p. 46).4 3 Taking advantage of the rock-bottom bad loan ratio and exceedingly ample latent profits of the stockholdings, in March 1997, the Bank of Tokyo-Mitsubishi made a decision on a bulk sale of its bad loans in the book value of 5 billion yen, in collaboration with Cargill Financial Service, the first case of the bulk sale of the bad loans by a Japanese bank (Tasaku 2002: p. 104, 143; Wada 2003: pp. 74-5, 180). After the success of the first bulk sale, the bank kept selling off its bad loans to Goldman-Sachs, in the value of 10 billion yen, and to Lone Star, in the value of 20 billion yen, respectively, throughout the rest of 1997 (Wada 2003). 218 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Between 1998 and 2001, Mitsubishi’s records in the bad loan and capital-asset ratios suddenly soured (see Table 56). Among all, Mitsubishi’s achievement in the two ratios in 1998, and in the two ratios and the bad loans in 2001, was the worst across all six banks. Those results are, however, only spurious. The sudden reversal between Mitsubishi and others in the operation performance in 1998 and thereafter is mostly ascribed to the MOF’s introduction of the “acquisition-cost accounting (genka-ho)4 4 and to the injection of taxpayer funds disproportionately assinged to other banks than Mitsubishi. For example, analyzing the financial statements issued by major 17 banks in March 1998, Watanabe (2001: p. 152, 176) found that, if the latent losses of the stockholdings were subtracted from the tier 2 component of their capital, the average value of the capital-asset ratios of those banks suddenly scaled down to only 6.63%, and that those that could go beyond the 8% hurdle of the BIS regulation were only the Bank of Tokyo-Mitsubishi and Sumitomo Bank. In fact, in accordance with Watanabe’s estimation, it was soon made public that, after the government put into practice the first-round of the taxpayer fund injection under “the Law on the Financial Function Stabilization (kin’ yu kind anteika sochi-ho) in March 1998, the Bank of Tokyo-Mitsubishi did not originally wish to apply for the injection of the fund (Watanabe 2001: p. 176).4 5 The superiority in the operation performance by Mitsubishi, along with its sustaining a tight relationship with other Mitsubishi firms, was even more striking in the second round of taxpayer fund injections in March 1999, made possible by the Diet’s enacting “the Law on Emergency Measures to Promptly Restore the Function of the Financial System” (kin ’ yu kind soki kenzenka kinkyu sochi-ho) in October, 1998, in place of the Law on the Financial Function Stabilization.4 6 At this juncture of the government’s arranging for taxpayer funds on a much lager scale, the Bank of Tokyo-Mitsubishi stood steadfast and declined the government’s offer, while others could not help accepting the fund in the form of either preferred stock or subordinated debts. Table 58 exhibits the amounts of the taxpayer funds injected into the six main banks and the IBJ in March 1999, their 219 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. demarcations between preferred stock and subordinated debts, and the amounts of the funds raised by these banks from the market or the group members.4 7 Table 58. The Capital Consolidation by the Six Main Banks and the IBJ (October 1998 to March 1999) Tax-payer fund (March 1999) Market- or group-financing (October 1998 - March 1999) Main bank Grand sum Sum Preferred stock Subordinated debt Sum Common and preferred stocks Preferred financing securities Sakura 1,145.0 800.0 800.0 (800.0) - 345.0 86.2 258.8 Tokyo- Mitsubishi 765.8 - - - 765.8 244.2 521.6 Sumitomo 801.0 501.0 501.0 (501.0) - 300.0 - 300.0 Fuji 1,217.0 1,000.0 800.0 (500.0) 200.0 217.0 217.0 - Sanwa 880.0 700.0 600.0 (600.0) 100.0 180.0 - 180.0 DKB 900.0 900.0 700.0 (400.0) 200.0 - - - IBJ 817.0 600.0 350.0 (350.0) 250.0 217.0 67.0 150.0 Sum 6,525.8 4,501.1 3,751.0 (3,151.0) 750.0 2,024.8 614.4 1,410.1 Source: Ohara (2000: p. 59). Note. The unit is billion yen. The number in the parentheses under “Preferred stock” o f “Tax-payer fund” is the preferred stock convertible to subordinated debts in the future. Preferred financing securities (yusen shusshi shoken) were issued by special purpose companies o f the banks concerned. The 450.1 billion yen o f the public fund injected to the six keiretsu main banks and the IBJ shares about 62% o f the total tax-payer fund injected (725.93 billion yen) in March 1999. While declining the government’s offer for public funds, the Bank of Tokyo-Mitsubishi, as Table 58 shows, could raise the massive funds through the issuance of common or preferred stocks (244.2 billion yen), most of which were accepted by Kin’yo-kai members 4 8 Those factors, often obscured in the sheer values of the bad loans and capital asset ratio, seem to have been properly taken into consideration in the long-term credit ratings of the six main banks by Moody’s: as shown 220 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. in Table 59, Tokyo-Mitsubishi had been rated the highest between 1998-2000, as well as in other periods, despite its seemingly lagging behind some or all other banks in the bad-loan and capital- asset ratios in the same period. Table 59. Moody’s Long-Term Credit Ratings on the Six Main Banks Main Bank 1995 1996 1997 1998 1999 2000 Sakura Bank a : \ - A3 Baal Baa2 Baa2 Mitsubishi Bank Aa3 Aa2 Aa2 A1 A2 A2 Sumitomo Bank A1 A1 A1 A2 A3 A3 Fuji Bank A1 A1 A1 Baal Baa2 Baa2 Sanwa Bank Aa3 Aa3 Aa3 A1 A3 A3 DKB A1 A1 A1 A3 Baa2 Baa2 Source: M oody's Japan at www.modys.co.jp Note. Moody’s rating evolves, basing on the following scheme: Aaa, Aal, Aa2 Aa3, Al, A2, A3, Baal, Baa2, Baa3 Bal, Ba2, Ba3, B l, B2, B3, Caa, Ca, C (eligible) w (speculative) Yet another set of the worst records of Tokyo-Mitsubishi in March 2001, on the other hand, had the close bearing on the so-called the “Tokyo-Mitsubishi shock,” the bank’s accumulating more bad loans by introducing of a much stricter standard. The purpose for doing that, according to the bank, was “to put an end to the bad-loan problem in an earlier stage” (Mainichi Shimbunsha 2001, June 26 issue, pp. 26-7). As a result of the bank’s incorporating more risk factors in estimating bad loans originally classified as “the special attention,” which along with “the normal loans,” was not required to be incorporated into the bad-loan calculations, those classified under “the special management” drastically increased, pushing up the total bad loans to more than 3 trillion yen (Mainichi Shimbunsha 2001, June 26 issue, p. 26). In witnessing Tokyo-Mitsubishi’s confidence in eliminating the suddenly puffed-up bad loans on its own, some analysts gauged that others would significantly suffer, if they followed Tokyo-Mitsubishi’s way of estimation (Mainichi 221 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Shimbunsha 2001, June 26 issue: p. 27). Behind the more than 80% increase of the bad loan accumulation, from 1,723.3 billion (March 2000) to 3,194.1 billion yen (March 2001), was implied Tokyo-Mitsubishi’s readiness to pay back 100 billion yen of the taxpayer funds of March 1998. In fact, the bank did it at the same time the Tokyo-Mitsubishi shock was reported.4 9 In 2002, as Table 56 indicates, Tokyo-Mitsubishi significantly recovered from its bad loan issue: its bad-loan ratio declined to the lowest value, and its capital-asset ratio ranked second to Sanwa, whose capital account was then yet leveled up by 600 billion yen of the preferred stock the government purchased. Put under the generally inferior status to Tokyo-Mitsubishi, on the other hand, the other five main banks built their restructuring plans. First, in reacting to their pledges in the “business improving plans” (keiei kenzenka keikaku) submitted to the government in exchange for taxpayer funds in March 1999 that they redeem the funds in the future via the improvement of their profitability, they seriously began to deliberate on affiliation or merger with other banks in a bid to generate synergy (Nihon Keizai Shimbunsha 2000: p. 211). In August 1999, Fuji Bank and DKB, joined by Industrial Bank of Japan (IBJ), announced their plan to establish “Mizuho Holdings” by fall 2000, taking advantage of the government’s revision of Section 9 of the Antimonopoly Law to make possible establishing the holding company in December 1997. Established officially in September 2000 by designating the three banks as subsidiaries, Mizuho Holdings boasted of 141 trillion yen of assets, 95 trillion yen of deposits, and its large securities subsidiaries’ market share (16.3%) in the underwriting business second to Nomura Securities (24.3%).5 0 In order to further promote reorganization of the three banks into “Mizuho Bank,” which was to undertake the retail and wholesale financing, and “Mizuho Corporate Bank,” which was to inherit the investment business, Mizuho Holdings, together with Mizuho Trust and Mizuho Asset Trust, opened their supreme holding company, “Mizuho Financial Group,” in October 2001. By the end of 2001, the Group completed the reorganization by exchanging the stocks of the three disappearing banks for those issued by Mizuho Bank and Mizuho Corporate Bank. Aiming to consolidate Sakura’s profit 222 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. earning ability and to expand Sumitomo’s business opportunities in the Metropolitan area (Maihichi Shimbunsha, 2001, April 17 issue, p. 28; Ohara 2000: pp. 45-6), on the other hand, Sakura Bank and Sumitomo Bank announced in October 1999 their plans of business affiliation on the premise that they complete the merger by April 2002. Saving one year from the original schedule, these two banks merged in April 2001, establishing “Sumitomo Mitsui Bank Corp. (SMBC).”5 1 Remaining in limbo in the wave of mergers, Sanwa Bank finally made a decision to join in the merger plan announced by Tokai Bank and Asahi Bank in September 1998. Hoping to consolidate the retail financing basis in the Metropolitan, Midsection, and Kansai Area through the merger, the three banks announced in March 2000 that they would come under the management of a common holding company by April 2001. Asahi, however, suddenly broke away from the deal in June 2000 in fear of Sanwa’s dominance in the event of the merger (Nihon Keizai Shimbunsha 2000: pp. 234-5), and Toyo Trust, in place, joined the merger plan the following month. The reorganization was, thus, started among Tokai, Sanwa, and Toyo Trust, with Tokai-Sanwa’s transfer of their trust business to Toyo, and Toyo’s commercial banking business to Tokai-Sanwa. In parallel, the stocks of those three banks were exchanged for those issued by UFJ (United Financial of Japan) Holdings Inc., officially established in April 2001. In January 2002, the reorganization was completed, and the new commercial banking and trust banking sectors, which were renamed respectively UFJ and UFJ Trust, were placed under the umbrella of UFJ Holdings Inc. Originally planned to generate synergy, however, those three cases of merger necessarily blurred the boundaries of, and even the dissolution of, the five keiretsu groups, where the five keiretsu main banks, prior to the mergers, had traditionally been assigned the centripetal role of financing their members arranged in line with the one-set principle in their groups: since those banks merged with each other, despite their fierce competition based upon the one-set principle formula in the past, they were all pressed to reconsider the relations with those from their own shacho-kai, as well as those from outside. Engulfed with those issues, Fuji Bank and DKB, when setting forth their plan to establish Mizuho Holdings to the Fair Trade Commission in September 223 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 2000, were called upon to pledge to the Commission to reconsider their continuous participation in Fuyo-kai and Sankin-kai. Immediately thereafter, Fuji Bank and DKB transferred the administrative offices of Fuyo-kai and Sankin-kai, traditionally placed in these two banks, to Marubeni and Itochu, respectively, (www.yomiuri.co.jp/atmoney/special/f23/mizuho01.htm, www.jftc.go.jp/pressrelease/ 00.june/0006l.html both retrieved on May 26, 2004). In a similar move, in January 2001, Sanwa Bank informed the Fair Trade Commission of its intent not to force the customer firms of Tokai Bank to join in Sansui-kai and pledged that UFJ would not elaborate any merger plan for its customer firms (www.jftc.go.jp/pressrelease/01.january/0101102 retrieved on May 26, 2004). Mitsui Bank and Sumitomo Bank, on the other hand, informed the Commission of their refusal to stay away from Nimoku-kai and Hakusui-kai, while insisting that they were not going to commit any form of manipulation of the member firms belonging to these two shacho-kai (www.jftc.go.jp/pressrelease/00.december/00125 retrieved on May 26, 2004). In this case, the keiretsu boundary was affected through the mergers ventured across the two groups, not by the Bank’s turning away from shacho-kai: as for the first move, in November 2000 Mitsui Chemical and Sumitomo Chemical announced their merger plan completion by March 2003.5 2 Also, in October 2001 and November 2002, “Mitsui Sumitomo Insurance Co.” and “Sumitomo Mitsui Construction Co. Ltd.” were established, respectively, by the mergers between Mitsui Marine Insurance and Sumitomo Marine Insurance, and between Mitsui Construction and Sumitomo Construction. These two newly established firms joined Nimoku-kai and Hakusui-kai, encroaching on the traditional keiretsu boundary between the two groups. As of today, however, the successful mergers between these two Groups are limited to these two cases above, and Sumitomo-Mitsui’s arching over the two different shacho-kai is apparently deviating from the gist of the one-set principle. The mergers among the banks were not the only factor that contributed to cracking the traditional keiretsu borders. In the second half of the 1990s, banks and their client firms intensively dissolved their cross-shareholdings and eventually sold off the amount equal to, or even more than, 224 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. what they had purchased during the bubble econom y period. A ccording to Inoue (1998, 1999, c -I 2000), the clearing of cross-shareholdings was first started by non-financial firms selling off the stocks issued by banks in the mid-1990s and then, accelerated by banks selling off those issued by non-financial firms in the late 1990s, with a twist of banks’ frequently engaging in the “cross trading,” whereby banks bought back some of those they sold, after realizing their latent profits into cash (Watanabe 2001: pp. 131-2; Yamamoto 2003: pp. 69-73).5 4 Targeting 2,472 banks and business firms, Inoue (2000) also found that substantial part of the stocks issued during the bubble economy period were sold off to the market in such a short period of 1996-98, showing concern that there would soon emerge the problem of securing stable shareholders among firms, had it not been for the discontinuation of such a movement.5 5 Despite such a concern as revealed by Inoue, the government established “the Law Restricting the Possession of Shares, etc. by Banks, etc.” (ginko nado no kabushiki nado no hoyii wo seigen suru horitsu) in September 2001, to encourage banks to shrink their stockholdings and thereby to mitigate their exposure to the vicissitude of the stock prices. Furthermore, in January 2002, in supplementing the above law, the government started to operate “Banks’ Shareholdings Purchase Corporation” (BSPC, ginko nado hoyii kabushiki shutoku kiko) in collaboration with banks for purchasing and reselling the stocks accumulated in the banks. The “market valuation accounting” (Jika kaikei) of September 2001 is also another factor promoting this trend, as it requires both financial and non-financial firms to subtract what used to be treated as the “latent losses” of their stockholdings out of their tier 2 capital. In closing, it is important to understand that, while all the events reviewed in this subsection specifically concern the issues arising from either bad loans or the stock market crash or both, they are actually the manifestation of the one-set principle that had become dysfunctional due to the oil crises in the 1970s. That was already confirmed by various stockholdings’ indices taken up in Chapter 2. Even the least damaged Mitsubishi Bank and other Mitsubishi firms will never be immune from such an impact, as their stockholdings variables were also shown to be on a general 225 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. declining trend. The state of the dysfunctional one-set principle is reflected in the recent wave of the affiliations made beyond the traditional keiretsu boundaries by some shacho-kai members (see Table 60); getting unleashed from cross-shareholdings and their distinctive roles assigned from the standpoint of the one-set principle, the keiretsu members have started to increasingly pursue alliance with independents or those with different keiretsu backgrounds, just for the sake of their own economic motives and/or technological needs.5 6 Table 60. Mergers and Affiliations Made beyond the Traditional Keiretsu Boundaries by Non-financial Shacho-kai Firms (2000-02) Date Firms Purpose January 2000 Nippon Steel (IBJ), Sumitomo Kinzoku Kogyo (Sumitomo) The affiliation for the production o f H- type and stainless steel April 2000 IHI Heavy Ind. (Mitsui, DKB), Nissan Auto (Fuyo) IHI Heavy Ind.’s buying out o f the aerospace and self-defense technology sections from Nissan May 2000 Mitsubishi Heavy Ind. (Mitsubishi), Hitachi Co. (Fuyo, Sanwa, DKB) The unification o f the steel production technologies October 2000 Mitsubishi Motors (Mitsubishi), Nissan Motors (Fuyo) The affiliation for the production o f forklift tracks October 2000 NEC (Sumitomo), Hitachi Co. (Fuji, Sanwa, DKB) The affiliation for the production o f optical communication machies October 2000 C. Ito (DKB), Marubeni (Fuyo) The unification o f the steel sections November 2000 Mitsui Chemicals (Mitsui), Sumitomo Chemical (Sumitomo) The merger to be completed by October 2003. (but, finally cancelled in March 2003) November 2000 NEC (Sumitomo), Hitachi Co. (Fuyo, Sanwa, DKB) The unification o f the semiconductors DRAM sections January 2001 Mitsubishi Corp. (Mitsubishi), Nissho-Iwai (Sanwa, DKB) The unification o f the steel sections April 2001 NKK (Fuyo), Kawasaki Steel (DKB) The plan to establish “JFE Holdings” by October 2002. May 2001 Matsushita Electric (Independent), Hitachi Co. (Fuyo, Sanwa, DKB) The affiliation for the improvement o f the home electric machines Table continues next page 226 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 60 (continued). Mergers and Affiliations Made beyond the Traditional Keiretsu Boundaries by Non-financial Shacho-kai Firms (2000-02) Date Firms Purpose May 2001 Mitsui Corp. (Mitsui), Sumitomo Corp. (Sumitomo) The comprehensive affiliation. June 2001 Sumitomo Corp. (Sumitomo), Nissho-Iwai (Sanwa, DKB) The affiliation for developing the LNG sections. September 2001 Tanabe (Sanwa), Taisho (independent) The plan to establish a holding company October 2001 Mitsubishi Motors (Mitsubishi), Nissan Motors (Fuyo) The unification o f the autotransmission sections October 2001 Toshiba (Mitsui), Matsushita Electric (independent) The unification o f the liquid crystal display sections. November 2001 Nippon Steel (IBJ), Sumitomo Kinzoku Kogyo (Sumitomo) The unification o f the welding sections by July 2002 December 2001 Nippon Steel (IBJ), Kobe Steel (Sanwa) The affiliation for the supply o f raw materials and semi-finished goods December 2001 Nippon Steel (IBJ), Sumitomo Metal Ind. (Sumitomo), Kobe Steel (Sanwa) The comprehensive affiliation January 2002 Mitsui Construction (Mitsui), Sumitomo Construction (Sumitomo) The plan for the unification within a few years March 2002 Mitsubishi Electric (Mitsubishi), Hitachi Co. (Fuyo, Sanwa, DKB) The unification o f the system LSI project June 2002 Nippon Seiko K.K. (Fuyo), NTN (Sanwa) The unification o f the shaft bearings sections July 2002 Asahi Chemical (DKB), Mitsubishi Chemical (Mitsubishi), Idemitsu Petrochemical (independent) The unification o f the domestic polystyrene production sections within the year July 2002 Isuzu Motors (DKB), Yanmar Co. (independent) The affiliation for the production o f diesel engines for the industrial use September 2002 Toyota Motors (Mitsui), Nissan Motors (Fuyo) The affiliation for the production o f less pollutive, hybrid autos December 2002 Nissho-Iwai (Sanwa, DKB), Nichimen (DKB) The plan to establish a holding company by April 2003 Source: Kaisha Shikiho (http://job.toyokeizai.co.jp/reorga/saihen.html), Megacompany Research (2003), and Takarajima (2003). 227 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 3.5 Concluding Remarks The analysis of the six main banks for the period of the mid-1980s through the early 1990s proved the robustness of the TCE perspective applied to the Japanese keiretsu. In the case of Mitsubishi Group, equity financing among the Kin’yo-kai members, in the wake of financial liberalization, were treated in line with their traditional practicing of cross-shareholdings which, in turn, were based upon their frequent intra-keiretsu transactions. Mitsubishi Bank was then assigned the crucial role of financing those members in filling up their capital deficiency resulted from cross shareholdings. This complementary mechanism significantly helped divert the bank from financing excessively in the three non-production sectors of real estate, construction, and commerce and thereby from the subsequent bad loans in these sectors. Taking advantage of suffering the least from bad loans, the bank could make a great advance by merging with Tokyo Bank and by settling the bad loan problem, including the reimbursement of taxpayer money to the government. In the case of other keiretsu groups, on the other hand, their somewhat shaky asset-specific exchange networks failed to induce member firms to take on convertible and warrant-attached bonds in the context of cross-shareholdings. As a result, the member firms had come to lean toward equity financing in lieu of the traditional main bank financing, and the main banks started to heavily finance the aforementioned three non-production sectors in order to offset their declining profits, with the dire consequence of their suffering huge bad loans throughout the 1990s. Successfully deriving the bipolarity between Mitsubishi Bank and others in their responses to financial deregulation, the TCE perspective applied to the Japanese keiretsu presents one emulating view against that upheld by the majority of economists. Typically, economists’ view on how the bubble economy was brought about is exclusively drawn on the financial deregulation of the 1980s, whether the empirical approach in explaining the rising equity financing on the part of the non-financial corporations and the speculative financing in the three non-production sectors on the part of banks (Hoshi and Kashyap, Miyazaki, Noguchi), or of the theoretical approach in 228 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. explaining the shrinking “main bank rent” (Aoki, Okuno). The TCE perspective, on the other hand, emphasises loosening inter-corporate exchange networks in the keiretsu groups, which had already been conspicuous in the three Bank Groups at the outset, and which had come to be more serious across all the groups on account of the downfall of the heavy and chemical industries brought about by the oil crises. The weakening of the inter-corporate exchange networks were then translated in the group constituents’ lesser commitment to cross-shareholdings, especially with the main bank and the GTC(s) which, in turn, prepared the way for the main bank to lose its customers even from its own keiretsu group. Also, choosing an enterprise group as a basic unit of analysis, the TCE perspective takes into consideration the non-unanimity among the keiretsu groups in the pace of their losing asset-specific transactions among their member firms. As Japan and the U.S. experienced the bubble phenomena in common in the wake of financial deregulation, it might be natural for researchers to cast their eyes on its impacts on the financial market. It is, yet, essential to take notice that the way the financial market is organized significantly differs in these two economies. In particular, in Japan, financial institutions and production firms are more intimately tied through cross-shareholdings, and consequently the bank financing is often inseparably related with providing the group firms with financing for completing their joint production. Accordingly, their inter-corporate transaction relations should properly be taken into analysis, even if the main concern is on the financial market. Its origin being of the tradition of the institutional economics, TCE offers a useful analytical tool for such a case. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Endnotes for Chapter 3 1. The MITI’s drastic policy change was first announced to the public through the Industrial Structure Council’s declaration (1971), which contained the following agenda: (1) a transformation o f policy goals from the pursuit o f maximum economic growth to the utilization o f economic growth to improve social and economic performance. (2) a change from the vigorous protection and promotion o f promising industries to limited intervention into firm behavior and “maximum use o f the market mechanism.” (3) promotion of a shift from a capital-intensive, energy-intensive and environmentally destructive industrial structure to a knowledge-intensive, energy-conservative, and environmentally sound one. And (4) promotion o f international cooperation (Dore 1986: p. 13, 14; Uekusa 1987: pp. 475-6). 2. These industries were aluminum refining and synthetic fibers, (hurt by high energy costs), shipbuilding (low world demand), electric furnace steel making, ferrosilicon, and linerboard (low domestic demand), and cotton spinning, combed-wool spinning, and chemical fertilizers (hit by the increasing comparative advantages o f newly industrializing countries). 3. In the absence o f the MITI’s positive involvement in the adjustment process, the keiretsu groups played significant roles, especially in such aspects as (1) financial assistance, including interest reductions, from the main banks, (2) spreading o f financial losses among member firms, and (3) transferring displaced workers to other member firms. Their roles were particularly pronounced in the shipbuilding and aluminum refining industries. See Sheard (1991) and Uekusa (1987: pp. 494-9). 4. Parallel to the MITI’s shying away from economic affairs during the 1970s, the government’s involvement in the R&D activity was negligible during the same period: it accounted for only 2% of all R&D costs o f the private sector, which stood in sharp contrast with Germany (16%), Britain and France (25-30%), and the U.S. (35%) (Dore 1986: p. 135). 5. In fact, the elasticity o f petroleum consumption with respect to GNP in Japan decreased from 1.0-1.2 in the high growth period to 0.25 in 1974-79 (Shibagaki 1981: pp. 65-6, 73; Yamanaka 1989: pp. 16- 7). Another factor that helped Japan to quickly recover from the first oil crisis was its expansion of exports. Interestingly, it is possible to detect Japan’s industrial transformation in its changing the composition o f the export goods: while the composition ratios o f textiles and foods decreased, respectively, from 16.1% (1965-69) to 9.4% (1970-74) to 5.5% (1975-79) and from 3.6% (1965-69) to 2.2% (1970-74) to 1.2% (1975-79), the ratio o f machinery, represented by automobiles, electric machinery, precision machinery, and optical instruments, drastically increased from 41.5% (1965-69) to 51.4% (1970-74) to 60.7% (1975-79) (Shibagaki 1981: pp. 64-5). 6. The GTCs were the first agents to bear the direct brunt o f the first oil crisis. In December 1976, due to its irretrievable credit in NRC (Newfoundland Refining Company), Ataka Corp. was forced into failure and inherited by Itochu Corp. under the guidance o f Sumitomo Bank (Itochu Shoji Chosabu, Ed. 1992: p. 90). Later on, Itochu itself was forced to abandon its affiliation with Toa Sekiyu because o f the sharp fluctuations in oil prices (Itochu Shoji Chosabu, Ed., 1992: p. 90; Yamanaka 1989: pp. 27-8). Mitsui Corp. was also forced into a fix because o f its joint venture with the Iranian national petrochemical company (Itochu Shoji Chosabu, Ed., 1992: p. 90). 230 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 7. Note that “chemicals” in Table 40 include the petrochemicals sector. Two other points to be made from the table are that (3) the decreases in the coefficients were relatively minor in the sectors o f food (-0.08), textiles (-0.20), and pulp and paper (-0.22), and that (4) the decreases in the sectors o f general machinery (-0.28), electrical machinery (-0.28), and transportation equipment (-0.28) were somewhere between the industries grouped in (2) and (3) above. These results imply that the technological relations o f the sectors o f food, textiles and pulp and paper with others were constantly weak, and that the sectors o f three newly rising industries (general machinery, electrical machinery, transportation equipment) were not so severely affected by the first oil crisis as in chemicals, iron, and steel. 8. The following data in Table 61 (Imai, 1982: p. 59) o f 1975 show that the six keiretsu were increasing their tie-up contracts with other keiretsu firms or non-keiretsu firms, indicating their breaking interior-ties. Table 61. Increasing Tie-Up Contracts with Outsiders in the Six Keiretsu (1975) Mein Bank Within Group With Firms in Other Groups With Other Outside Firms Mitsui group 7.2% 11.8°/ 52.7% Mitsubishi group 9.1 7.1 36.9 Sumitomo group 9.4 17.3 64.7 Fuyo group 2.7 6.3 27.7 Sanwa group 1.9 6.2 29.0 DKB group 2.0 7.5 24.3 9. In contrast, during 1976-1986, the number only increased from 370,000 to 380,000 (Yaginuma 1993). 10. There seem to be two more purposes for large firms to spin off subsidiaries, as some economists point out. They are (1) realization o f the economies o f scale for the assembly maker and parts producers in each hierarchical level (Uekusa 1987: p. 503; Yaginuma 1993: p.21, 28) which, as will be discussed shortly, is relevant to the manufacturing industries, and (2) diversification through setting up subsidiaries (Dore 1986: p. 61, 63, 72, 75, 85-6, 149, 215-6; Kosei Torihiki Iinkai Jimukyoku 1992: pp. 86-96 pp. 266-79) which were very popular among the members o f the six gigantic keiretsu (horizontal keiretsu). Number 2 will be confirmed in the next section. 11. In some other literature, “horizontal keiretsu” is alternatively referred to as “intermarket keiretsu” or “kigyo shudan (the enterprise conglomerate).” “Vertical keiretsu,” on the other hand, is alternatively referred to as “kigyo guriipu ” (the enterprise group). 12. Gerlach (1992: p. 146) showed an interesting case in which one electric company formed both a horizontal and vertical keiretsu at the same time, placing the transaction weight more on the latter. A major electric company, disguised as “Takeshita Kogyo,” relies on shacho-kai firms for only 1.6% o f its total purchases, most o f which are from the group trading company and several raw material and heavy equipment producers. Takeshita’s sales to the group’s firms, most o f which are the group’s main bank and steel company, on the other hand, account for 1.9% o f its total sales. In contrast, about 40% o f its total purchases came from affiliated supplier firms, and fully 90% o f its semiconductor sales passed through two captive distributors in its vertical keiretsu. 231 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 13. Note that while engaged in the real business activity, the parent firms avoided impinging on Section 9 o f the Antimonopoly Law. This is because the section prohibits only the setting up o f “pure holding companies” (junsui mochikabu gaisha), which are exclusively concerned with the governance o f their subsidiaries through their stockholdings. 14. Actually, cross-shareholdings were occasionally observed between parent firms and their subsidiaries before the amendment o f the Commercial Code in 1981, which prohibited companies owning more than 50% o f their stocks from owning their owners’ stocks. See Nakajima (1991) and Okumura (1993: p. 16) for detail. 15. Parallel to the varying degree o f the parent firm’s ownership, the degree o f independence o f the subsidiaries varies, especially in terms o f personnel exchange. Yoshimura, Ueno, and Kagno (1990: p. 25) pointed out that the largely held production subsidiaries received a large number o f managers from Matsushita Electric, while most o f the sales companies employed managers and sales personnel by themselves from the local labor market. 16. The dual governance nature o f subsidiaries is probably the main reason why Aoki (1987) preferred to use the term “the quasi-disintegration” instead o f “spinning-off’ when he explained the phenomenon o f increasing subsidiaries after the oil crises. 17. Yoshimura, Ueno, and Kagano’s (1999: p. 30) major finding was that when subsidiaries are owned mainly by parent firms, their performance, measured by ROE and ROA, improves very much. 18. Since the introduction o f the equity method, it has become customary to call the 50% or more- owned subsidiaries, “kogaisha ” (subsidiaries) and the 20-50% owned-subsidiaries, “kanren gaisha (associate companies).” But, unless the distinction between these two terms is crucial, the present thesis uses the term “subsidiaries.” 19. The primary reason that the “average” number is introduced for counting the subsidiaries is that typically in the Bank Groups there are often multiple firms positioned in identical sectors. Toshiba and Hitachi Co. are not included in Figures 26 and 29-31 because its data are based on the SEC standard. Besides, the following firms were not incorporated for the data calculations because o f their missing data or adoption o f the SEC standard: Kureha Kogyo in Figure 29, Tanabe, Kansai Paint, Nakayama Steel in Figure 30, Chichibu Cement, Kawasaki Steel in Figure 31. 20. The relatively stagnant pace o f the increment o f subsidiaries in such sectors as paper and pulp, ceramics, chemical and petroleum is well explained by Yaginuma (1993: pp. 13-5): their production process did not depend so much on assembly-work, but on the economies o f scale. 21. The diversification matrix was first introduced in 1979 by the Fair Trade Commission, which intended to measure diversification by the ten largest capitalized firms in 17 production sectors, which were listed in the TSE first section (KOsei Torihiki Iinkai Jimukyoku 1992: pp. 88-95). A drawback involved in the FTC’s approach is that it failed to consider the subsidiaries o f these firms. (Kosei Torihiki Iinkai Jimukyoku 1992: pp. 265-82). 22. Mitsubishi Kasei and Mitsubishi Rayon were both branched off from the same company, the former Mitsubishi Kasei, in 1950. Mitsubishi Oil was founded in 1956 under the strong support from Mitsubishi Kasei and Mitsubishi Rayon, both o f which invested 22% o f its paid-in capital. See Kikkawa (1992: p. 301). 232 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 23. The reasons why they were skeptical about the BOJ’s monetary policy in raising the asset-prices are twofold: First, a loose monetary policy should lead to a general increase in prices, but inflation in goods prices never occurred in Japan in the late 1980s. Second, there was the timing inconsistency between the BOJ’s first interest cutting (January 1986) and the remarkable rise o f 12.7% in land prices in six major cities (March 1985). See Hoshi and Kashyap (2001: p. 226, 228, 312). 24. While highly evaluating Okumura’s work in turning attention to the interaction between equity financing and cross-shareholdings, the present thesis diverges from his work in taking into explicit consideration the differing degrees o f the intra-keiretsu transactions and cross-shareholdings across the six keiretsu groups. With no TCE view employed, Okumura interprets the high correlation between equity financing and cross-shareholdings as the keiretsu firms’ desire to maintain high stock prices and to circumvent hostile takeovers. By the same reason, he fails to explain differing responses among the main banks on how to cope with the bubble economy and its aftermath. 25. In 1986, the detached warrant components were also allowed in Japan. 26. The issuance o f the convertible bonds actually became popular in Japan after Nippon Express’ (Nippon Tuuri) issuance in 1966. 27. Note that financing through the issuance o f the convertible and/or warrant-attached bonds is advantageous to the issuing firms when the stock prices are on the upward trend because the issuing firms can restrain the interest payment to the bondholders in exchange for capital gains o f the newly issued stocks. Furthermore, in the case o f the “Eurodollar” warrant-attached bond, it was often the case that by making a long-term contract o f foreign exchange, the issuing firms could enjoy “negative” interests. See Miyazaki (1992: pp. 134-5, 155-9). 28. That DKB’s accepting straight bonds from its main bank firms amounts to only 6% o f all its trusteed bonds between 1984-87 (Table 53) seems most likely a typographic error or a calculation mistake. 29. Very impressive o f Mitsubishi Bank is that before the pre-deregulation period covered by the three periods in the Table 52 (1970-75, 1976-79, 1980-83) the bank’s acceptance o f the trustee requests from its main bank firms was extremely high, compared with other five banks (Table 52). This seems correlated with the extremely high rate o f cross-shareholdings in the Mitsubishi Group during those periods. 30. In the case o f the warrant-attached bond, Sakura (Mitsui) and Sanwa Banks’ records seem in disagreement with the TCE perspective: despite the tenuous intra-keiretsu exchanges in the Mitsui and Sanwa Groups, Sakura and Sanwa Banks both trustee all (100%) members’ trustee requests. Other than this, the three observations hold true in the case o f the warrant-attached bond. 31. Essentially the same argument can be found in Sheard (2001: pp. 69-70) and Tanaka (2003: pp. 66- 9). They furthermore point out that Japanese firms could often increase the book value of their capital ad infinitum by simply carrying out cross-shareholdings. This issue is discussed in depth with reference to “share buyback” in the next chapter. 32. The lending ratio is defined as the “lending made in the sector in question/the total lending.” The lending “ratio” was chosen for the comparison purpose because the six main banks were heterogenous in the compostition o f their capital and asset. The ratios were calculated from TQyo Keizai Shinposha, Kigyo keiretsu soran (1987-1994 issues). 233 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 33. Confident o f its sound financing, Mitsubishi Bank took on the initial public offering at the New York Stock Exchange in September 1989, where the stringent SEC standard was adopted as the listing condition, such as releasing the profit reports four times a year. Japanese banks have often been cited among the top-ten ranking o f the worldwide excellent banks in many journals, but, as of 2003, Mitsubishi Bank was the only Japanese bank listed on the New York Stock Exchange. Probably due to the stringent requirements o f the Stock Exchange, Japanese firms, including Mitsubishi Bank, account for only eight in total, such as Sony, Honda, Hitachi, and TDK. 34. Nihon Keizai Shimbunsha (2001: pp. 73-6) and Yanai (2002: pp. 58-65) pointed out that President Isoda’s reorganization plan o f Sumitomo Bank o f July 1979, which subjugated the monitoring section (shinsa bumon) under the business section (eigyo bumon), greatly spurred the bank’s speculative lending in the late 1980s. For details o f other scandalous incidents o f Sumitomo Bank, see Nihon Keizai Shimbunsha (2000), Nogi (2001), Sataka (1994), and Yanai (2002). Also note that the second lowest record o f Sumitomo’s mainstay ratio in 1986 symbolizes its weakening centripetal force. 35. Very perplexingly, the bond prices were also sharply declining in April 1990. The most convincing explanation for this is that there occurred a capital flight from Tokyo to Frankfurt, seeking the investment opportunities in unified Germany (Miyazaki 1992: pp. 199-204). The successive decline in the stock prices by April 1990, on the other hand, is deemed ascribable to the BOJ’s raising the official discount rate because by that time the Deutsche mark was declining against the yen (see Figure 111-23 o f Miyazaki 1992: p. 203). 36. In 1990-94, the average annual growth rate o f real GDP was 1.5%, compared to 5.5% for the previous four years. To counter this, the government abandoned its policy goal o f a balanced budget, which was achieved in 1991 after 16 years o f deficits. 37. The residential land price in Tokyo declined at an annual rate of 1.0 % in 1991, 12.7% in 1992, and 2.3% in 1993. The commercial area, on the other hand, declined in value at an annual rate o f 0.3% in 1991, 12.5% in 1992, and 20.5% in 1993. 38. Mitsubishi Bank’s proposal was eventually turned off by the Banking Bureau o f the MOF, which expected that many regional banks would be forced into troubles by such a disclosure (Nihon Keizai Shimbunsha 2001: pp. 290-1). 39. The bad loans, which were expected to be at least partly recoverable in the future (Categories II and III) were sent to the “Housing Loan Administration Corporation” (jutaku kin 'yu kanri kiko, HLAC). The corporation was established in June 1996 and planned to close in 15 years. For more details, see Milhaupt and Miller (1997). In April 1999, the HLAC merged with the Resolution and Collection Bank (seiri kaishu ginko), formerly known as Tokyo Kyodo Bank (a bridge bank for failed Tokyo Kyowa Bank and Anzen Credit Bank in the late 1994), and developed into the “Resolution and Collection Corporation” (RCC). 40. The following are the failed credit cooperatives in 1994-96: Tokyo Kyowa Credit Association (December 1995), Anzen Credit Bank (December 1995), Cosmo Credit Coop (July 1995), and Kizu Shin’yo Kumiai (August 1995). The failed regional banks are Hyogo Bank (August 1995) and Hanwa Bank (September 1996), which were both second-tier banks. The bankruptcy o f Hyogo Bank is particularly remarkable because it was the first case in the postwar era in which the MOF issued an order o f the business suspension to an ordinary bank. The MOF’s decision to close Hyogo Bank was based on its policy change from the convoy-like protection approach to the market-based approach after the jusen problem was settled down in June 1996. 234 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 41. As o f today, there are three definitions o f a bad loan in Japan. They are “the risk management loans,” “the self-classified loans,” and “the classified loans per Financial Reconstruction Act.” The bad loans reported in Table 56 are all calculated under “the risk management loans.” By March 1996, they were only composed o f (a) loans to the failed firms and (b) loans past due more than six months. But, in March 1996, (c) loans whose interest were reduced or exempted and (d) loans to the firms under support, were added. In March 1998, (e) loans past due more than three months was also added, and (f) loans with relaxed conditions was defined anew so as to replace (c) and (d) above. As a result o f those revisions, the risk management loans became virtually equivalent to the U.S. SEC standard. For the details o f the definitions o f bad loans, see Hoshi and Kashyap (2001: pp. 281-3), Yabushita and Bushimata (2002: p. 82), Yoshida (2001: pp. 62-71), and Watanabe (2001: pp. 30-59). 42. No matter which way it is to be disposed of, accumulating reserves or write-offs from the book, bad loans must be redeemed from the capital account o f the lending institution, which is made o f the following two components: (a) the fundamental component (Tier 1): the sum o f “capital (common and preferred stocks),” “legal reserves,” and “surplus fund.” (b) the supplemental component (Tier 2): the sum o f “the latent profits o f the securities,” “the latent profits o f land,” “the bad loan reserve,” and “the long-term subordinate debt. ” 43. On its part, Tokyo Bank was also attracted to the merger with Mitsubishi Bank. This was because Tokyo Bank had constantly eroded its privileged status as the foreign exchange bank since the revision o f Foreign Exchange Law in 1980s. 44. The acquisition-cost accounting was introduced by the MOF as a temporal measure in August 1992 and March 1998. The purpose was for corporate stockholders to evade the realization o f the latent loss. In closing the books in March 1998, city banks, long-term credit banks, 16 trust banks, and 5 insurance companies adopted this accounting method. See Inoue (1998). 45. “The Law on the Financial Function Stabilization” was passed on February 16, 1998, paving the way for taxpayer money injection into financially ailing banks other than credit cooperatives and shinking banks. Under the law, “the screening committee of the financial crisis management” (kin ’ yu kiki kanri shinsa iinkai) chaired by Yoko Sasanami made a decision o f injecting the total 1,800 billion yen o f taxpayer money into 21 banks (city and trust banks) and 3 long-term credit banks. The decision by the committee later incurred the rancor among city banks, as it distributed uniformly 100 billion yen to each o f them with no close attention to their varied bad loan ratios. 46. During the same period, the eight bills, which were packaged as “the Financial Reconstruction Law” (kin ’ yu saisei-ho) were enacted under the strong support o f the Democratic Party. The main purpose of the law was to set up the bridge bank for failed banks. For the purchase o f the bad loans inherited from failed banks, “the Financial Revitalization Account” was set up, with 18 trillion yen of taxpayer funds in the Deposit Insurance Corporation o f Japan. These laws and the account set up at the corporation were abolished when the payoff plan was put into action in April 2005. 47. In August 1998, Sakura Bank announced the plan for raising capital, to the amount o f 300 billion yen, mainly from Nimoku-kai members, such as Mitsui Corp., Mitsui Fudosan, and Toyota Motors. In anticipation o f the Bank’s soon receiving the taxpayer funds, Toyota declined the request. By the end o f November o f the year, the bank could collect 345 billion yen from its 21 associate firms. See Asahi Shimbunsha (1999: p. 241) and Dore (2000: p. 160, 205). 235 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 48. Interestingly, especially from the TCE perspective, Table 58 reports that the main banks o f the three Bank Groups could not raise so much capital as the Big Three, despite their overwhelming numbers o f member firms. Also, considering the fact that they accepted taxpayer funds mostly in the form of subordinate debt and thereby allowed the government’s direct intervention, their problems o f capital shortage are deemed to have been extremely serious as o f March 1999. 49. By March 2004, only four banks (Tokyo-Mitsubishi, Mitsubishi Trust Bank, Kansai Sawayaka Bank, and Sumitomo Trust Bank) could complete the redemption o f the taxpayer money. In March 2004, Mizuho Financial Group, Mitsui Trust Bank and Ashikaga Bank respectively paid back, respectively, 575 billion yen, 50 billion yen, and 30 billion yen, all o f which were injected in the form o f subordinated bonds in February 1999 (March 23, 2004. Nikkei Shimbun). 50. Mizuho Holdings’ securities subsidiaries were established through the merger among Fuji Securities, DKB Securities, and IBJ Securities. 51. Allegedly, Sakura Bank was, at the beginning, considering a merger with Sanwa Bank (Nihon Keizai Shimbunsha 2000: p. 227). Note that the merger between Sakura Bank and Sumitomo Bank was done without setting up a holding company. 52. The merger plan o f these two firms, however, was cancelled in March 2003. 53. With Kigyo keiretsu soran discontinued in December 1999, which itself symbolizes the end o f the keiretsu era, the present thesis exclusively relies upon Inoue (1998, 1999, 2000) and Kuroda (2001, 2002) for the data o f cross-shareholdings after 1999. 54. Targeting cross-shareholdings carried out nationwide, Inoue (1999) found the dissolving trends of cross-shareholdings during 1995-98, as summarized in the following table. Note that the dissolution of cross-shareholdings was carried out among non-financial firms on a large scale during the period (0.59% decrease during 1995-98). Table 62. Dissolving Trends of Cross-Shareholdings during 1995-98 Year Rate of cross shareholdings (aggregated) (%) Stockholdings by business enterprises of: Stockholdings by banks of: Those issued by business enterprises (%) Those issued by banks (%) Those issued by business enterprises (%) Those issued by banks (%) 1995 20.31 4.85 5.36 6.98 0.33 1996 19.52 4.97 4.17 7.69 0.22 1997 18.19 4.71 1 3.51 7.53 0.11 1998 16.02 4.26 3.19 6.51 0.04 55. Inoue (2000: p. 10) reported that, out o f 12.66 trillion stocks issued by banks, and 16.46 trillion stocks issued by non-financial firms during the bubble economy period, 8.36 trillion stocks and 6.89 trillion stocks were respectively sold off to the market during 1996-1998. 236 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 56. Kosei Torihiki Iinkai Jimukyoku (2001b) conducted a survey research on the firms’ motives behind selling off the partners’ stocks in the late 1990s. Among the top 100 firms measured by non-consolidated assets in March 2000, 61.5% o f them replied that they sold off their partners stocks out o f concern o f the latent losses from the stockholdings, and 23.1% replied that they did so because o f the ineffectiveness o f cross-shareholdings with the firms they no longer had any trade relation with. 237 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Chapter 4 In s t it u t io n a l C o m p l e m e n t a r it y , A sse t S p e c if ic it y , a n d t h e E c o n o m ic P o l ic y 4.1 Introduction Now that the Japanese main bank system was analyzed fully over the period starting from its incipiency through its heyday in the high-growth era to its recent crisis, the policy proposals for the system are in order in this chapter. Consistent with the TCE framework adopted, the focal point is how to reconstruct the inter-corporate relation between banks and non-financial firms. This consideration is, indeed, unavoidable because the contemporary crisis of the Japanese main bank system, as analyzed in the previous chapter, principally arose from banks losing transaction ties with their group firms. Unlike the tenet upheld by the majority of economics professions, it is virtually impossible to single out such an issue as if it is purely a financial phenomenon.1 Although the policy proposals of this chapter are principally based on the results obtained in the previous chapter focusing on the six gigantic horizontal keiretsu, they have close relevance to the whole Japanese economy.2 This is actually possible because as recurrently emphasized by Gerlach (1992: p. 18, 170, 172, 178, 292) and Kester (1991a, 1991b, 1992), the salient characteristics of the six horizontal keiretsu, such as stable stock cross-holdings, long-term reciprocal trading, and financial relationship, are not limited to those keiretsu, but widely held 3 . . . . . throughout Japanese firms and banks. Their contention is, in fact, validated by various statistics. Targeting all the listed Japanese firms throughout the periods of 1987-2000, Kuroki (2001: p. 18) 238 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. reported that cross-shareholdings by non-horizontal keiretsu firms are similar to those of horizontal keiretsu, concluding that cross-shareholdings practice is widespread in the Japanese economy (Table 63). Table 63. The Rates of Cross-Shareholdings Made among the Six Keiretsu Firms, the Non- Keiretsu Firms, and the Whole Market. Year Six Keiretsu Non-Keiretsu firms The whole market 1987 28.01 12.10 18.39 1990 26.21 11.13 18.01 1993 26.36 11.14 17.51 1996 23.78 10.17 16.26 1999 20.36 9.39 10.58 2000 16.71 7.47 10.10 Source: Kuroki (2001: p. 18). Note. The unit is percent. Kosei Torihiki Iinkai Jimukyoku (2001a: pp. 6-7), on the other hand, reported that, among the 100 largest Japanese firms measured in terms of non-consolidated asset, 42 were shacho-kai member firms, and 44 would not decrease or increase cross-shareholdings in the future. The primary reason (77.3%) is to maintain or consolidate trade relations with their partners.4 Furthermore, in the same survey research, 76.3% of 97 firms were shown to have their own main banks, which supplied 56.0% of their corporate financing (Kosei Torihiki Iinkai Jimukyoku 2001a: pp. 9-10). The “vertical keiretsu-ization” has also been a common characteristic of the Japanese economy, especially after the oil crises. As shown in Table 64 below, the number of subsidiaries newly established in each fiscal year steadily increased until the late 1980s, although some of the increase, especially in the late 1980s, might be the reflection of the euphoria of the bubble economy. 239 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 64. The Number of Subsidiaries Founded during 1982-91 in Japan Subsidiaries 1982 : 1983 : 1984 ! 1985 1986 1987 1988 1989 1990 1991 Production Sector 183 165 214 204 257 242 215 227 164 138 Non-Production Sector 477 596 591 712 823 886 804 806 638 521 Sum 660 761 805 916 1,080 1,128 1,019 1,033 802 659 Source: TO yO Keizai ShinpOsha (1993: p. 30). The aforementioned survey research by Kosei Torihiki Iinkai Jimukyoku (2001a: p. 19) indicated that, during the aftermath of the bubble economy in the 1990s, the vertical keiretsu-ization had not decelerated among large firms: from FY 1993 to FY 1999, the ratio of the consolidated to non-consolidated asset, which averaged over 73 respondent firms, increased from 1.34 to 1.43, with the average values of non-consolidated asset virtually unchanged between these two years. The factor which turned out to be no less important than financial liberalization in contributing to the bubble economy was that the cross-shareholdings were maintained in the horizontal keiretsu, despite their actual intra-group transactions’ debilitation since the oil crises. If the government could properly map out the policies based on such recognition, it could have at least appeased the disastrous impacts of the speculative bubbles. Until quite recently, however, the government’s major concern was not in adjusting the stockholding relations among large firms in light of the current industrial organization, but rather in mobilizing fiscal and monetary policies in pumping up the depressed economy, which soon became inefficacious.5 Some important policies on the subject have been introduced over time. A good example is the June 27 lifting of the ban on holding company. But they were generally sporadic, inconsistent, and far from completion. What is required of the government today is to make the legal and infrastructural frameworks consistent and comprehensive so that the ever-lasting cross-shareholdings in the horizontal keiretsu will be corrected to reflect the contemporary industrial organization, which is best characterized by the development of the vertical keiretsu.6 Only thereafter will the effects of fiscal and monetary policies, if any, be unfolded. 240 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. The present chapter proceeds in the following order: first, Section 4.2 extracts from the previous chapter the contemporary financial and industrial organizational problems the Japanese economy faces today. In preparation for the policy making, Section 4.3 sets out the policy norm, with which to sieve out possible policy plans. Particular concern here is the concept of the “institutional complementarity,” elaborated by Aoki (1995a ) and Aoki and Okuno (1996), which refers to the complemental relation across institutions that serves to reinforce their mutual existence. The consideration of this concept is indispensable for policy making because, as reviewed in the section on the “state-contingent nature of the main bank relation” in Chapter 1, a bilateral, functional correlation exists in the Japanese financial and labor markets, and because a disordering impact on one of these markets necessarily spills over to the other. Besides the institutional complementarity, the TCE consideration is made in the policy norm creation. In Section 4.4., the so- called “Big Bang” financial reform (1996) is examined and evaluated in light of the policy norm thus established. In Section 4.5, policy proposals are based on analyses of the previous chapter and the policy norm established in Section 4.3. More specifically, (a) introduction of the holding company, (b) introduction of the share buy-back system, and (c) the cross-entry by banks and non- financial firms are proposed. Section 4.6 concludes this chapter. 4.2 The Contemporary Problems of the Corporate Groupings in Japan: The Discrepancy between the Changing Boundaries of the “Meta-Firms” and the High-Growth Frameworks As analyzed in-depth in the previous chapter, the contemporary crisis of the Japanese main bank system is closely associated with the debilitation of the intra-keiretsu transactions, which became an irretrievable trend after the two oil crises of the 1970s. The intra-keiretsu transactions, or the joint productions among keiretsu members, since then have stagnated. Accelerating this tendency was the members’ shifting their transactions with their own subsidiaries. With a series of the financial liberalizations that followed, large-scale keiretsu firms tapped directly into the financial 241 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. market in place of the traditional main bank borrowing, which climaxed in the bubble economy of the late 1980s: during the period, the main banks, except probably Mitsubishi Bank, were forced into desperate positions in their keiretsu groups, where they could not but accept the disproportionate numbers of convertible and warrant-attached bonds from their escaping keiretsu associates, and, simultaneously, finance excessively the three non-production sectors of real estate, construction, and commerce. Once the euphoria of the bubble subsided, the “lost decade” of the 1990s ushered in. Banks were left with huge bad loans amassed in the three sectors, and the inordinate number of equities converted from the bonds whose prices started to precipitate. The non- financial firms, on the other hand, were left with excessive numbers of convertible and warrant- attached bonds which failed to convert into equities because of continuously falling stock prices. Later on, in the late 1990s small and medium-sized firms suffered, as banks, desperately disposing of their bad loans, gradually withdrew their loans and refused to issue new ones. It is not correct that politicians and government officials failed to take any remedial measure at the sight of the adverse impacts on the heavy and chemical industries brought about by the oil crises. The government, in fact, mobilized special measures for promoting capacity reduction and employment adjustment of the industries severely hit by the oil crises from 1978 to 1988. They failed to correctly recognize that the various frameworks which once were compatible with the main bank system and the heavy and chemical industrialization were losing their causes, and, thus, needed corrections to fit in with the low growth regime. The problems that arose from the uncorrected high growth frameworks are basically twofold and inter-dependent, as explained more in-depth below. 242 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 4.2.1 The Necessity to Reduce Cross-Shareholdings With transactions among large-firms less and less frequent after the oil crises, cross shareholdings in each horizontal keiretsu should have shrunk to proportionate degrees of their decreased transactions. There are three reasons for that. First, when the trading partner became more and more estranged as it shifted the transaction weight to its subsidiaries, holding the stock issued by such a partner became more onerous. In more modem terminology, that likely increases the agency cost of the holding firm: when in the high-growth period the issuing company used to be a close transaction partner, monitoring the company was an easy job of quickly examining the products regularly delivered if it was a supplier, or of simply accepting the orders from the company if it was a buyer. But now, whether a buyer or seller, with the transaction volume steadily decreasing with the company, monitoring the company becomes more difficult. It requires in-depth knowledge of its subsidiaries, to which the company shifted its transaction channel in the past years, and of the products, which the company produces in collaboration with those subsidiaries. This is exactly the reasoning behind the statement by a Japanese executive that “there is no sense in holding shares in a company with which business ties are slim” (Gerlach 1992: p. 78). Second, with the rising monitoring or agency cost, another negative effect of cross shareholdings, which offsets capital raised between/among the firms concerned, became more likely pronounced. In the simple term of mutual stock exchanges, cross-shareholdings entail canceling-out capital for those firms involved, and does not contribute to consolidating their capital. During the high-growth era, such a negative effect was overwhelmed by its benefit of smoothing out transactions among involved firms. Now, however, the positions of benefits and cost are reversed, as those firms are less involved in transactions and thus less interested in the mutual corporate governance through cross-shareholdings. 243 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Third, it is extremely hazardous that the keiretsu firms issue their stock, despite their infrequent intra-keiretsu transactions, and that the main banks, as a result, become the reservoirs of those stocks. Such a situation would most likely lead to the destabilization of the banks’ operations, especially when the economy trends downward. This situation actually captures well what the main banks of the three Bank Groups (Fuji Bank, Sanwa Bank, and Dai-Ichi Kangyo Bank) and Mitsui Bank went through in the late 1980s and 1990s: according to Figures 5 and 6 (Chapter 2, Section 2.5.2) the stockholdings are concentrated mostly in the hands of the main banks of the three Bank Groups and the Mitsui Group during the period. When the economy trended upward, those banks benefited from holding excessive stocks issued by the members, as the BIS (Bank for International Settlements) regulation allowed the Japanese banks to incorporate 45% of capital gains into their capital, and thereby encouraged their lending even further. Once the economy faltered, however, those banks significantly suffered from the sharp decline of stock prices which ate up their capital. According to the keiretsu or TCE perspective, the skewed accumulation of stocks in the hands of the main banks is well explained by the weakened asset-specific transactions among member firms: once the asset-specific transactions are weakened among non-financial members, the role of the main bank in filling out their capital void arising from cross-shareholdings would accordingly become modest, because non-financial members are now less willing to be involved in such a practice. In response to such a change, the main banks would be motivated to hold more of the members’ stocks, in order to reinforce the transaction relations with the fleeing customers. In the long-run, even the Mitsubishi Group, the epitome of the Japanese Keiretsu Model, will by no means be safeguarded by such a tendency, as the low growth pattern that settled in since the oil crises will hardly reverse its course in the future.8 244 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 4.2.2 The Necessity to Deregulate the Banking Sector: The Road for Investment Banking Aoki (1995a: pp. 138-48) and Aoki and Okuno (1996: pp. 237-43) argued that the four pillars of postwar financial regulation—the Temporary Interest Rate Adjustment Law (rinji kinri chosei-ho) of 1947, the MOF’s repression of the bond market by setting the high standards for collateral requirement, the entry restriction of the banking and securities sectors respectively set by the Banking Law and the Securities and Exchange Law, and the MOF’s carrot-and-stick policies of granting banks their branches and dispatching their ex-officers to those only performing well—once catapulted the banking sector through creating the so-called “main bank rent9, allowing banks the leeway to support ailing firms but, in the process of financial liberalization progressing, became less effective and even started to be the shackles on the banking sector. While agreeing with some of these factors pointed out by Aoki,1 0 the TCE assesses the issue from a different perspective, and also raises some other issues that Aoki (1995a) and Aoki and Okuno (1996) failed to point out. First, the previous argument of the changing boundaries of the keiretsu “meta-firms” suggests that the ban on banks to open pure holding companies, which is substantiated in Section 9 of the Antimonopoly Law, has become obsolete and even harmful on the banking sector. The primary reason for this is that, even after the main banks lost substantial numbers of non-financial keiretsu firms as their customers as they shifted their transaction base to their own subsidiaries,1 1 Section 9 did not allow them to approach the small and medium-sized firms in the same way as the non-financial keiretsu members—acquiring, with virtually no restriction, the stocks issued from small and medium-sized firms as stipulated in Section 10 of the Antimonopoly Law. As Shimotani (1996: pp. 20-9) pointed out, the revision of Section 10 in 1949 had a significant impact on the Japanese (non-financial) firms to develop into “operating holding companies” (Jigyd mochikabu gaisha) through the acquisition of shares issued from other associate firms on the same keiretsu line. 245 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Along with Section 11 of the Law allowing banks to hold up to 5% of shares issued by 12 other keiretsu associate firms, Section 10 thereby contributed to the formation of horizontal keiretsu during the high-growth period (Shimotani 1996: p. 29).1 3 By the time the oil crises assaulted the Japanese economy, the co-existence of the horizontal keiretsu, a part of which is the main bank system, Sections 9 and 10 had not raised a serious issue, as the non-financial keiretsu members were still bound firmly via asset-specific transactions in the area of heavy and chemical industry, and as there was room for the main banks to play a role in their horizontal keiretsu by financially facilitating the transactions, and filling the capital void arising from cross-shareholdings. But, once such transactions stagnated, the problem came to the surface. Unlike their non-financial keiretsu associates, which were now eager to develop their own vertical keiretsu under the auspices of Section 10, the main banks were completely blocked from owning their subsidiaries out of a flock of small and medium-sized firms. Furthermore, in the course of that process, they were substantially deprived of the opportunities to develop their ability to evaluate the potentiality of new technologies developed by small and medium-sized firms. As Aoki (1995a: p. 220) pointed out, such evaluation is now best done by non-financial firms.1 4 Those issues became even more serious when major non-financial firms started lending business to their subsidiaries through their “non banks” or “financial subsidiary companies” (kin’ yu kogaisha),1 5 thereby enormously stripping the banks of financing jobs: according to Sakamoto and Shimotani (pp. 195-225: 1987) and Ubukata (1986), many major Japanese firms set up their own financial subsidiary companies in the mid- 1980s, with their vantage points of abundant internal reserves and privileges to raise capital through the issuance of CPs in the foreign financial markets, and started to finance their own subsidiaries, often at better terms than their main banks could offer. At this point, the seclusion of the traditional banks from both the horizontal and vertical business lines was culminated.1 6 Desperate, banks tried to secure the niche in the small and medium-sized firms financing, but they did so very poorly from a monitoring viewpoint: as shown in Table 65, the city banks 246 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. played a major role in the area of the small and medium-sized firms financing by continuously increasing their lending since 1975. Correspondingly, Yanai (2002: p.53) estimated that the share of the city banks’ lending to those firms in their total lending steadily increased from 34.5% (1975), through 46.7% (1980) and 51.8% (1985), to 71.3% (1990). Table 65. Lending Shares in the Small and Medium-Sized Firms Type Fiscal Year 1955 1975 1980 1985 1988 Dec. 1989 Private Bank account City banks 27.3 20.9 21.5 26.1 30.1 30.3 banks Regional banks 28.5 20.3 19.8 20.5 20.2 20.5 Regional banks 9.3 (Tire II) Trust banks 2.4 0.7 0.7 2.6 3.3 3.1 Long-term credit 1.9 2.3 3.8 4.3 5.4 5.6 banks Subtotal 60.1 44.2 45.8 53.6 59.0 68.8 Trust account 0.7 2.8 3.2 3.5 3.9 4.2 Sum 60.8 47.0 48.9 57.1 62.9 73.0 Specialized Private Sogo banks 19.4 16.1 15.0 12.7 10.9 banks institutions Shinkin banks 11.0 20.9 17.2 14.7 13.2 13.6 Credit co-ops 1.7 5.2 5.9 5.1 4.8 5.0 Sum 32.0 42.2 38.1 32.5 28.9 18.6 Governmental Shoko chukin 3.4 4.3 4.5 4.3 3.5 3.4 institutions Chusho koko 1.8 3.0 3.8 2.7 2.1 2.3 Kokumin koko 2.1 2.5 3.4 2.7 2.2 2.3 Subtotal 7.2 9.8 11.7 9.7 7.8 8.0 Sum 39.2 52.0 49.8 42.2 36.7 26.6 Grand sum 100.0 100.0 100.0 100.0 100.0 100.0 Grand sum (million yen) 158 5,832 10,983 17,936 24,493 26,582 Source: Nakamura (1992: p. 159). Note. The unit is percent. “Grand sum” includes the lending from the institutions not listed in the table. Note that, in 1989, Sogo banks were transformed into Regional banks (Tire II). Consequently, the financial difficulties characteristic to the small and medium-sized firms were ostensibly overcome (Nakamura pp. 162-4). The city banks’ enormous lending to those firms, 247 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. however, was often associated with their securing the real estate collateral from the borrowing firms, 17 and not firmly based on their close monitoring of those firms: : under the continually soaring land prices throughout the postwar period, banks speculated that they could fully compensate the higher default risks involved in financing the small and medium-sized firms by simply securing their real- 18 estate collateral (Yabushita and Bushimata 2002: p. 73-4). The remaining rigidity in the lending interest rates, mostly attributable to the regulated borrowing interest rates, also contributed to preventing banks from differentiating lending interest rates by carefully monitoring the financial states of borrowing firms.1 9 Thus implemented, the collateral-cum-financing started to automate itself, paving the way for the bubble economy (Okumura 1999: pp. 122-5, 135; Yabushita and Bushimata 2002: pp. 75-6). Once the bubble economy collapsed, however, the values of the collateral plummeted, and the banks began to fasten their lending outlets, focusing on consolidating their capital. The DI index of the small and medium-sized firms, which climbed to unprecedented levels of around 30 during the bubble economy period, declined to virtually zero in 1992 (Okumura 20 1999: p. 157). Although the index became slightly restored thereafter, the firms, suffering debt- overhang, no longer wished to borrow (Okumura 1999: p. 165). After the Law on Emergency Measures to Promptly Restore the Function of the Financial System (kin’ yu soki zesei-ho) was introduced in October 1998, the declining status of the city banks as the main banks for the small 21 and medium-sized firms became much more pronounced. In order for banks to acquire the proper monitoring capability, it is inevitable to deregulate the area of investment by banks so that they can invest in small and medium-sized firms on an equal footing with non-financial business firms. With this deregulation, they will be exposed to opportunities to nurture their monitoring capability so as to effectively compete with large, non- financial business firms in the area of investment in small and medium-sized firms. Besides, they will be urged to reconsider their collateral raising, and, instead, to factor in various risks involved in financing as well. Such a deregulation directly involves revising Section 9 of the Antimonopoly 248 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Law and the Banking Law, both of which have long prohibited banks from becoming investment banks. Simultaneously, the revision of these two laws requires that of, or even the abolition of, Section 11 of the Antimonopoly Law because it sets a 5% investment ceiling against banks, a far too small ratio for banks to compete with other non-financial firms and to exercise governance effectively over small and medium-sized firms. Capping all this revision process is the abolition of Section 65 of the Securities and Exchange Law: even if banks are allowed to hold as many equities issued by small and medium-sized firms as they wish, they will still need to purchase them indirectly via securities firms because Section 65 bans any securities-related business (dealing, brokerage, and underwriting) by banks. Quite justifiably, banks will feel uneasy paying extra fees for securities firms when they have already achieved the ability to carry out the whole investment process, including helping small and medium-sized firms complete issuing their equities. Besides traditional financing, achieving the three functions of dealing, brokerage, and underwriting, with no restriction on equities holdings, will fully equip banks with the capabilities of governing small and medium-sized firms, and, if they so wish, allow them to build their own corporate groups, as have large, non-financial firms since 1947.2 2 4.3 Building the Policy Norm: Institutional Complementarity and Asset Specificity Before hammering out specific proposals, it is necessary to establish the policy norm with which to sieve out possible policy plans. In so doing, this section explicitly considers the Japanese labor market, which has been left out of the analyses so far due to the TCE approach adopted in this thesis. The main reason for considering the labor market is not out of the typical macroeconomic concerns, but rather from the consideration that Japanese goods, financiing, and labor markets maintain the complemental relations with one another, and that, because of this, the economic policy targeting one of these three markets will eventually spill over and trigger systemic influences to the 249 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. other two markets. In order to achieve a better understanding of such an issue for creating the policy norm, this section reviews Aoki (1995a) and Aoki and Okuno (1996), who first pointed out the above issue and elaborated the concept of “institutional complementarity.” Next, the section revisits the TCE issue in order to examine its compatibility with the concept. According to Aoki, an “institution,” including the market system is best comprehended by the concept of “Nash equilibrium,” where the choice by a player becomes optimal only when his or her choice follows those chosen by the majority of the market participants (“strategic complementarity”). The equilibrium thus reached is stable, as any one of the players in the game hardly receives an incentive to change his or her choice (the “self-enforcing” nature of the Nash equilibrium). At this stage, “institutional complementarity” plays a vital role in the way that institutions, thus established as stable Nash equilibria, function in an complemental manner, thereby reinforcing them as a whole system by giving more incentives to the players in each institution to choose the most prevalent modes (Aoki 1995a: p. 19, pp. 91-2; Aoki and Okuno 1996: p. 2, 35, 325; Okuno-Fujiwara 1999). Aoki, then, argued that there exists institutional complementarity in the Japanese main bank system, where two factors function in a comlemental fashion make effective the “contingent governance” by the main bank over its customer firm: the one is the “incomplete corporate governance market” (securities market), which is substantially inactive due to the widespread cross-shareholdings, and the other, the “incomplete internal labor market,” which is composed mostly of homogeneous, as opposed to skill-specific, workers so as to be incorporated into the “team” production of the firm. These two institutions are complemental in the sense that, while the production by the team, composed of homogeneous workers, precludes the main bank from discerning the performance of each worker, therefore requiring such an incentive-inducing scheme as the wage system contingent upon their total output as reflected in the main bank’s “contingent governance” scheme, the main bank’s contingent governance per se increases the chances for the workers to share operating processes with one another, as if they work in a “team,” because of the incomplete governance market (securities market) failing to specify the contractual 250 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. agents in detail (Aoki 1989a, 1994a, 1994b, 1994c, 1994d). Because of this complementarity, Japanese firms, especially since the late 1950s, have been able to develop a seamless system that encouraged their workers to cultivate shared learning across various work units through frequent job rotations and OJT (on-the-job training) inside their factories or offices so that they could effectively respond to changing industrial climates and on-site technical exigencies, with no recourse to managerial interventions (Aoki 1989a, 1994a, 1994b, 1994d; Aoki and Patrick 1994; Koike 1997; 23 24 Miyamoto 1999). ’ At the same time, under the contingent governance by the main bank, the Japanese firms were required to develop an internal institution that could effectively forestall the problem arising from the less clear-cut job demarcation, such as difficulty in associating the salaries with specific jobs, and the potentiality of the development of local or sectional goals inconsistent with the whole organizational one. To this end, the rank system (shokuno seido) was devised and extensively adopted by Japanese firms, as the best device for the human resource management, whereby the workers were paid by their rankings, regardless of what jobs they engage in, and whereby the centralized personnel department can effectively administer the internal recruitment of managers through competition among the workers over the higher rankings (Aoki 1989a: p. 349, 350, 351, 354, 1994d: p. 40, 51, 55; Koike 1997: pp. 100-8; Miyamoto 1999: pp. 74-84, pp. 94- 25 113). Thus, giving equally the two differing incentives of cooperation in the team and of competition for promotion, Japanese firms, in general, succeed in securing a stable labor- management relation, where the workers acquire the contextual skills specific to the firms they belong to and efficiently cope with the firm-specific issues over the long-periods of their tenures (Miyamoto 1999: pp. 94-113).2 6 What is remarkable of the concept of institutional complementarity is that once it is applied to another economy where the market conditions significantly differ from Japan’s, another consequence will obtain. The U.S. economy is a case in point. In sharp contrast with Japan’s counterpart, the U.S. financial market clearly draws a line between credit (bank loan, bond, 251 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. debenture) and the stock market, and the stockholders exclusively hold the reigns over the top management through the market-oriented incentive contracts solely for the purpose of maximization 27 of the residual values of their firms (Aoki 1989a: p. 354). The high-powered, market-oriented incentive contracts are first made with the top managers, who are assigned the entire production and marketing tasks based on their expertise. The production process is divided into a series of special functions, with each standardized, and where each work unit is required to exclusively perform the specific job dictated by the original plan (1994d: p. 43). With the original plan once drawn at the higher office and hardly changed, those managers, unlike those of the Japanese firms, are not motivated to change the production scheme even in the face of unexpected technical difficulty and changing market conditions in the course of production. They choose rather to cope with those issues by such measures as preparing buffer inventories and troubleshooting specialists (Aoki 1994d: p. 43). Other employees positioned under the top management in the corporate hierarchy, too, are employed from each segment of the labor market that pertains to each specific job required by each work unit. The employment pattern thus established is referred to as the “job system” (shokumu seido), which is premised on workers’ standardized skills in each type of job that were acquired through their formal training or past job experiences, and on the facilitation of employment, from the top manger to shop-floor operators, through their competitions in the labor 28 29 market appropriately segmented for each existing job (Miyamoto 1999: p. 79, 99). ’ Obviously, the job system is the reflection of the institutional comlementarity existing between the stock and labor markets in the U.S. economy. Essential to note, at this stage, is that while the Japanese and the U.S. economies both maintain institutional complementarity between the financial and labor markets, the way it is maintained is diametrically opposite in these two economies: in Japan, the inactive stock market is linked, via the contingent governance of the main bank, with the development of the firm-specific, internal labor market based on the rank system, where the workers are offered the opportunities to learn and acquire the wide range of firm-specific, contextual skills 252 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. during their lifetime tenures. The pay raise is considered for those displaying higher performances, yet often in tandem with their promotion. In exchange for their lifetime tenures, however, the workers, once unemployed, often find it hard to find jobs outside their firms because their firm- specific, contextual skills are hardly applicable to other firms. In the U.S., on the other hand, the active stock market has a close bearing on the liquid labor markets based on the job system, where the wide range of workers are employed on the basis of their a priori acquired specialized, yet standard skills. The wages are mostly determined in the competitive labor market and secured for the period reflected in the employment contract. Usually, the employment contract does not last for such a long time as a lifetime as in the Japanese firms and, in the event of its expiration, the workers are prompted to renew their contracts or await the next employment chance. The concept of institutional complementarity and how such complementarity is distinctively formed between the two economies suggest two possibilities for policy outcome. One is that the policy intending for an organizational change in one market is limited in its effect because such an organization change is severely restrained by another market that functions in a complemental fashion with the market targeted by the policy makers. The difficulty involved in transplanting the market system of one economy to another can be well explained by this. The other one, diametrically opposite to that, is that such a policy triggers a chain effect, thereby giving rise to sequential organizational changes in other markets as well. No matter which one of these may be the case, the present author recommends and introduces the most conservative policy norm for the Japanese financial market reform. That is, the policy norm should be set to maintain or inherit the mode of institutional complementarity established between the Japanese financial and labor markets as observed above. The main reason for this is the prohibitively expensive cost that is expected to arise from the conflicts between the financial and labor markets if the reform sets a breach against their complementarity. More specifically, the financial reform, if introduced in such a way as to give precedence to the securities market over the banking sector, will necessarily incur the expenses on the side of the labor market in such forms as the creation of the function-segmented labor market, as 253 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. in the U.S. economy, and large reshuffling of the Japanese internal labor markets. Second, and probably more importantly, with regard to the current political economic climate both domestic and overseas, there seems no impending reason to alter the establishment of the complementarity between the two markets. There occasionally were those pundits and economists who, especially during the post-bubble economy period, claimed that the Japanese financial system is delayed in adapting to the global trend, probably keeping in mind the Anglo-American securities system, but their arguments often lack solid theoretical foundations. From the viewpoint of the institutional framework set out by Aoki, as explicated previously, such a global trend itself should be understood in the framework of strategic complementarity. What they referred to as a “trend,” in this regard, is hardly anywhere, as there is globally no convergence of any equilibrium regarding the modes of effective and efficient monitoring and production as of today. To see this, suffice it to point out two examples of the U.S. automobile and investment trust industries, which had both intensively adopted the Japanese modes of automobile production and long-term financial monitoring in the early 1980s and the early 1990s, respectively (Aoki 1994c: pp. 138-9; Aoki 1995a: pp. 6-7, 167; Aoki and Okuno 1996; Asanuma 1989; Asanuma 1992). The policy norm thus laid out, the last point to check is if the above-confirmed, complemental market relations hold true for small and medium-sized firms. This needs to be checked because Aoki’s model/concept, whether it is the “contingent governance of team” or “institutional compelmentarity,” seems to be built to a large extent on his observation of the main bank system that was consolidated during the high growth period of the 1950s-1970s, and because his model does not explicitly consider the inter-corporate, asset-specific transactions. Fortunately, the above-confirmed market characteristics are generally common across small and medium-sized firms as well: first of all, the financing of small and medium-sized firms is mostly through banks and/or their parent firms, with the latter making investment in widely differing degrees of 0-100%, and not through the open securities market. Thus, the governance over small and medium-sized firms is, in most cases, carried out by banks and/or their parent firms utilizing these financial 254 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. nexuses, as depicted in the model of the contingent governance of team. Also, the workers employed in the small and medium-sized firms are generally skill-specific to the products they produce and thus homogeneous inside the same firms (Koike 1997: pp. 28-9, 32-4, 122, 13 8-41).3 0 The homogeneous nature of labor employed in small and medium sized firms is particularly pronounced when the firms are created as a result of spinning-off from their parent firms (Aoki 1987), Shimotani (1996). As a matter of course, the inter-corporate ties between small and medium sized subsidiaries and their parents are generally tight, as the latter continuously subscribes to the asset-specifically related goods from the former. In addition to these general market characteristics, Asanuma (1989, 1992), who conducted the fieldwork on the electronic and automobile industries, showed how small and medium-sized firms in these industries were governed by their parent firms in the similar fashion to the contingent governance of team:3 1 typically, in these industries, there are constantly a selected few tier 1 subsidiaries maintaining relatively long-term transaction relations with their large parent firm. While assigned the routine work of producing the identical product at a fixed proportion among them, they are often placed into harsh competition by their parent over cost reduction, quality improvement, and customized-order production (“two-vendor policy”). The subsidiary that can beat others in this competition is often rewarded by the parent in the form of contract renewal, often at better terms, and/or the status promotion that enables the subsidiary to partake in the new product development process. Those which could not, on the other hand, are doomed to lose such a chance, and, even worse, may be forced out from their tier 1 subsidiary status.3 2 With the policy norm introduced, and the institutional complementarity between the financial and labor markets confirmed to be properly inherited in the inter-corporate relation vertically arranged, it is almost obvious at this stage that the policy proposals should be made in such a manner that both large firms and banks can nourish the vertical transactions by further extending their financial capabilities. 255 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 4.4 A Case Study: The Big Bang Financial Reform and Its Problems But before proceeding to the specific policy recommendations that are in line with the policy norm set up above, this section examines the so-called “Big Bang financial reform” (the term is abbreviated hereafter as the “Big Bang.”) of Japan, a spate of the extensive financial market reform measures that were successively introduced between 1997-2001. Sharing the interest of financial market reform and being the most extensive kind in postwar Japan, skipping over the analysis of the Big Bang does not seem sensible. This section overviews how it came to the fore, and then examines it from the perspectives of institutional complementarity and TCE in that order. The Big Bang financial reform started with then Prime Minister Hashimoto’s announcement in November 1996 of the primary objectives: (1) better use of 1,200 trillion yen of personal assets and (2) preventing hollowing out of the Tokyo financial market through the removal of its heavy regulations. Simultaneously put forward by Hashimoto were three principles along which the new financial system is expected to function: (1) “free”: the new financial system is based on market principles, (2) “fair”: the new financial system is transparent and rule-based, and (3) “global”: the new financial system operates under legal, accounting, and supervisory regimes respectively consistent with international norms. In all, the Big Bang aimed to re-establish the Tokyo financial market to be comparable to those of New York and London in both scale and sophistication of financial services (Dekle 1998: p. 237; Horiuchi 2000: pp. 233-4; Hoshi and Kashyap 2001: p. 289; Mukai 1997: pp. 13-5; Nihon Keizai Shimbunsha 1997: pp. 4-7; Royama 2000: pp. 257-8). Albeit titled after the original Big Bang of the U.K., the purview of the Japanese version was much wider, covering areas concerning the Foreign Exchange Law, the Banking Law, 33 the Securities Exchange Law, and the Insurance Business Law. With the foreign exchange reformation merely another step in a series of liberalizations,3 4 and with the insurance industry reformation practically absorbed in the issue of how to cope with the newly promulgated insurance 256 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. law of 1995, consequent to the Japan-U.S. negotiations of liberalization of the Japanese insurance industry (Royama 2000: pp. 259-60), reformation of the securities market was the largest part of the Big Bang. In preparing Big Bang proposals starting shortly after Hashimoto’s announcement, the Securities and Exchange Council (SEC) clearly envisioned the exceeding role of the securities or capital market over banks in the forthcoming regime, which would attach more importance to the effective allocation of financial resources (Royama 2000: pp. 258-9). In the interim report of November 1996, the SCE listed item “[to] make the securities market the lead actor in the fund intermediation” as one of the five intermediate goals in accordance with the Big-Bang directive 35 (Royama 2000: p. 260). In June 1997, the SEC’s effort in mapping out the details was materialized in the final report, which presented the comprehensive reform plan intending to break away from the traditional style of the securities exchanges administration by the MOF’s Securities Exchange Bureau. More concretely, to the end of achieving the efficient allocation of funds in the securities market, the report advised improving market competition, which was then expected to be achieved via inviting new entrants and promoting the innovation of financial commodities and services, including those traditionally banned by the Securities Exchange and other related laws. To realize (1) the removal of fixed commissions on stock transactions, (2) the shift from the licensing to registration system in the securities market, (3) the off-exchange transactions and (4) the revision of the related laws such as the civil, criminal, and commercial laws, were respectively proposed.3 6 Besides, corporate disclosure was strongly put forward, along with auditing by certified public accountants, in order to enhance the transparency of transactions. After putting forward all those market-enhancing proposals, the SEC revealed an unsavory opinion on cross-shareholdings: [CJross holdings of shares should not inhibit the market mechanism, as that will be the detriment of optimal allocation of resources and efficient use of capital. From this standpoint, even for cross-held shares, shareholders should constantly reevaluate the return on investment and exercise checks on the management of firms. (Royama 2000: p. 265) 257 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Those SEC proposals, along with those others submitted by the other three Councils, were mostly incorporated in “the Laws for Financial System Reformation” enacted in June 1998 (Horiuchi 2000: p. 237; Hoshi and Kashyap 2001: p. 290; Royama 2000: p. 269), whose key elements are summarized in Table 66. Promoting the securities market, relative to the banking sector and thereby increasing the importance of its disciplinary role, the Big Bang can reasonably be evaluated as encouraging the Japanese system to move toward the Anglo-American system, as Horiuchi (2000: p. 250) so commented. Table 66. Key Elements of the Big Bang 1. Expansion in the choice o f instruments fo r investors i. Enhancements to investment trusts (Dec. 1998) ii. Full liberalization o f securities derivatives (Dec. 1998) iii. Enhanced attractiveness o f stocks (Jun. 1997) iv. Smaller minimum investment lots for stocks (Jul. 1997) V. Streamlining o f foreign equity listing by using depository receipts (Dec. 1998) vi. Improved access to trading and quotation information (Dec. 1998) 2. Expansion o f the options in corporate fund-raising i. Introduction o f new corporate bond products (Dec. 1998) ii. Promotion o f asset-backed securities (Sep. 1998) iii. Promotion o f medium-term notes (May 1997) iv. Facilitation o f listing and initial public offering (Sep. 1998) v. Revision o f listing standards (Dec. 1998) vi. Enhancement o f over-the-counter market (Dec. 1998) (also included as 4-iii) vii. Deregulation o f unlisted and unregulated equities market (Sep. 1997) (also included as 4-iv) 3. Allowance offinancial institutions to provide a wider variety o f services i. Elimination o f business restriction on securities companies (Dec. 1998) ii. Liberalization o f brokerage commissions (Oct. 1999) iii. Reform o f Non-life insurance Rating Organization regime (Jul. 1998) Table continues next page 258 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 66 (continued). Key Elements of the Big Bang 3 (continued). Allowance offinancial institutions to provide a wider variety o f services iv. Promotion o f asset investment business (Dec. 1998) V. Allowing banks to issue straight bonds (Oct. 1999) vi. Allowing other financial companies to issue bonds (May 1999) vii. General shift in the rules from prior licensing and application for approval to after the fact notification (Dec. 1998) viii. Promotion o f competition across business boundaries (Mar. 2001) ix. Allowing insurance companies to enter the banking business (Oct. 1999) X. Allowing banks to sell insurance products (Oct. 2000) xi. Allowing holding companies (Mar. 1997) 4. Creation o f an efficient market i. Review o f security exchanges operation rules (Dec. 1998) ii. Abolishing requirement o f consolidation o f order-flow for listed securities (Dec. 1998) iii. Enhancement o f over-the-counter market (Dec. 1998) (also included as 2-vi) iv. Deregulation o f unlisted and unregistered equities market (Sep. 1997) (also included as 2-vii) V. Creation o f share-lending system (Dec. 1998) vi. Improvement in clearing and settlement system (Dec. 1998) vii. Reduction o f settlement risk (Dec. 1998) 5. Creation o f a fa ir market i. Formulation of fair trading rules (Dec. 1998) ii. Strengthening the penalty for violation o f rules (Dec. 1998) iii. Regulations to prevent conflicts o f interest (Dec. 1998) iv. Establishment o f the system o f dispute settlement (Dec. 1998) V. Enhanced disclosure system (Mar. 2001) vi. Review o f securities taxation (Apr. 1999) 6. Improving the stability o f the financial system i. Capital adequacy requirements for security firms (Dec. 1998) ii. Enhanced disclosure for financial institutions (Dec. 1998) iii. Rules for bank subsidiaries and insurance subsidiaries (Dec. 1998) iv. Framework for the protection o f customers in the event o f failures (Dec. 1998) Source: Hoshi and Kashyap (2001: pp. 292-3). Note. 1-3 are concerned with the options available to savers, fund-raisers, and financial institutions. 4-5 relate to the efficiency and fairness o f the market. 6 is concerned with stability. Data shown in parentheses indicate that the majority o f the reforms in the area were implemented. 259 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. How can one evaluate the Big Bang from the standpoints of institutional complementarity, the TCE, and the policy norm introduced in the previous section? First, as already discussed to a considerable extent, the concept of institutional complementarity suggests that the Big Bang will most likely cause a functional conflict between the securities and labor markets and thus is considered as against the policy norm introduced in the previous section. Again, this is mainly because the liquid, competitive securities market, which will be brought about by the Big Bang, is not compatible with how the Japanese labor market operates. With the labor market chiefly characterized by its internal nature or, equivalently, the rank system, in the advent of the competitive securities market, shareholders will be totally dismayed to find out that they have no choice but employ those managers internally promoted for the board of directors. Even when they could coincidentally make the efficient employment contracts with those internally promoted, the problem will recur at the next lower level of the ranking hierarchy, where the directors thus employed need to make efficient employment contracts with the managers to be assigned at the position. This issue recurs all the way down to the shop-floor level, where most employees at the shop-floor are actually employed in the lump-sum fashion and are expected to move seasonally from one shop floor to another for the acquisition of the broad range of skills required in the whole production. Focusing the aspect of the inter-firm transactions, on the other hand, the TCE evaluates the Big Bang from the viewpoint of how it will affect such transactions. Particular attention is paid to the horizontal and vertical keiretsu. With the horizontal keiretsu, the Big Bang’s impact is considered generally unsavory. First of all, the liquid, competitive securities market and the cross shareholdings do not seem to co-exist harmoniously. Motivated more by trying to achieve as high a profit as possible, both corporate and individual investors will now actively partake in their corporate governance through the general stockholders’ meetings and wield their market disciplining influence over the firms’ decision-making. Also, they will actively partake in buying and selling in the securities market, thereby signaling their intent to the corporate directors in the form of stock prices. The profit maximization schemes designed by the directors put under those 260 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. pressures, how ever, do not necessarily lead to the facilitation o f the intra-group transactions o f the horizontal keiretsu: As Roe and Gilson aptly explained (1992: pp. 37-8), the investment pattern long held by the horizontal keiretsu is to make an additional investment as long as it can make an additional return even if investment levels of the member firms may not correspond to those of their profit maximization. This conversely implies that the profit maximizing investment levels of the member firms do not necessarily correspond to the optimal levels of the intra-group transaction. Second, the market-valuation accounting (jika kaikei), which was introduced as one of the Big Bang’s enactments in September 2001 (see item 5.v. of Table 66), is most likely to force some horizontal keiretsu members to reconsider the cross-shareholding practice, and eventually the trading relations with their partners, because the new accounting requires the re-evaluation of the cross-shared stocks by their market values at the end of every fiscal year, and does not allow the latent profit, which was long admitted under the acquisition-valuation accounting (shutoku genka 37 kaikei) during the postwar period. Third, and related to both the first and second points, investors are more likely to have the incentive to discharge the stocks of the horizontal keiretsu firms because their equity capital (jiko shiohn) is overestimated due to the cross-shareholdings (Tanaka 2003: pp. 66-7). If those stocks were sold off harshly by the investors, the firms involved would suffer much more seriously than the non-keiretsu firms, as their stock prices would likely fall much further than others.3 8 In the case of the vertical keiretsu, too, the evaluation of the Big Bang will not become savory. The primary reason is that the Big Bang grossly overlooks the importance of bank loans to the small and medium-sized firms. As “non-banks” or “financial subsidiary companies” are limited to a small number of large, blue-chip companies, bank loans are still the financial source more crucial than direct financing from the securities market for most of the small and medium-sized firms in Japan.3 9 This situation is clearly shown in Figures 36 and 37 (based on data from Sogo 261 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Seisaku Kenkyujo n.d.) where the borrowing ratio and the equity capital ratio have been constantly higher and lower respectively in the small and medium-sized firms since FY 1989. Borrowing Ratio 60 50 40 .2 30 20 10 0 —♦— L arge firms n S m all a n d m edium -sized firms Y e a r Figure 36. Borrowing ratio (defined as borrowing/annual sales). Unit: %. Equity Capital Ratio 25 Large firms — 0 — S m all a n d m ed iu m -size d firm s Y e a r Figure 37. Equity capital ratio (defined as equity capital/[equity capital + borrowing]). Unit: %. 262 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Granted, the Big Bang eased the conditions for listing stocks at the over-the-counter market and introduced the “market maker system” (December 1998), thereby having contributed to small and medium-sized firms financing in the stock market4 0 Despite such a reform, however, small and medium-sized firms have overall kept their footing deep in bank loans, and are not so enthusiastic about the securities market reform: according to the survey research conducted by National Life Finance Corporation (Kokumin Kin’yu Koko) targeting 2,475 small and medium-sized firms in 2001, 48.8% of them responded to have tapped into direct financing in the past three years, but most of the funds (44.0%) they raised were actually from their families (Kokumin Kin’yu Koko Sogo Kenkyujo 1999: p. 31; Yabushiita and Bushimata 2002: pp. 117-8)4 1 In another survey question asking if they were planning to raise funds through direct financing in the future, the majority of the respondent firms (87.5%) replied negatively, only 0.7% positively (Kokumin Kin’yu Koko 1999: p. 33; Yabushita and Bushimata 2002: p. 119). The primary reason (49.2%) for their not planning to utilize direct financing in the future was that “the borrowing from financial institutions is good enough” (Kokumin Kin’yu Koko 1999: pp. 33-4; Yabushita and Bushimata 2002: p. 120).4 2 Essential in interpreting those survey results is that those firms’ decision to shy away from the securities market in favor of bank loans is firmly based on economic rationale: those small and medium-sized firms have correctly understood that by choosing to borrow from their main banks, instead of numerous anonymous investors in the securities market, they could significantly reduce the costs involved in resolving the issue of asymmetric information lying between borrowers and lenders. Their holdings of their checking accounts at the lending banks, as well as (1) the loan on bills (tegata kariire), (2) overdraft (toza karikoshi), (3) the bill discount (tegata waribiki), and (4) the loan on deeds (shosho kariire) for further borrowing are, in fact, the least costly means to deliver their financial states to their lending banks 4 3 In contrast to bank loans, the costs involved in screening and disclosure requirements imposed by the securities market are, in general, prohibitively expensive for the small and medium-sized firms and cannot easily be the alternative to 263 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. bank loans (Yabushita and Bushimata 2002: p. 121).4 4 For instance, as pointed out by Kato and Matsuno (1999: p. 197), for most of the small-scale, start-up venture firms, almost all the merits of raising funds through the securities market vanish the moment they employ accountants for meeting the disclosure requirement. Besides, the issuance of bonds by small and medium-sized firms is often limited to private placement of less than 500 hundred million yen for less than 50 purchases, because, in that case, the firms are exempt from the strict disclosure requirement imposed by the Securities and Exchange Law (Yabushita and Bushimata 2002: p. 115)). All these problems show that, even if the Big Bang encourages small and medium-sized firms to partake in the securities market through its reform, it cannot ultimately overcome the issue of higher thresholds set against those firms, which are inherent in direct financing. In all, by any criterion, the Big Bang financial reform does not seem to benefit the Japanese economy. Instead, all the analyses above suggest that the banking reform, rather than that of the securities market, is much more conducive to the Japanese economy. In the economy where the innumerable firms with all sizes are engaged in mutual transactions, while interlocking their equities, the reform, if need be, would often bring about much more productive outcomes when administered in the banking sector. 4.5 The Policy Proposals Based on the policy norm set up and the recognition that the inter-corporate transactions generally shifted their weights from those among large firms centering around heavy and chemical industries to those between large firms and their subsidiaries, this section proposes economic policies deemed the most essential for implementation from institutional and legal perspectives. More specifically, these policies are (a) the introduction of the pure holding company system, (b) the introduction of the share buyback system, and (c) the cross-entry between banks and non- financial firms, all of which discussed below in order. Because these policies are closely related to 264 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. resolve the issue, it is essential for the government to adopt and administer these policies all at once. As Amyx (2004: pp. 235-7) emphasizes, this would require the policy coordination among ministries and other related government agencies beyond their jurisdictions. In order to make this feasible, the cabinet’s strong leadership would be very desirable. 4.5.1 The Pure Holding Company System In view of both the TCE perspective and the policy norm, the most fundamental policy ingredient for the low growth economy regime is to introduce the pure holding company (junsui mochikabu gaisha) system, which itself will also become the baseline for implementing other policies. The introduction of the system is actually a part of the Big Bang financial reform (see 3.x. of Table 66). It was already enacted in June 1997, through the revision of Section 9 of the Antimonopoly Law, and enforced in December of that year.4 5 Defined as the corporate entity exclusively focusing on controlling its first-order subsidiaries through the stockholdings, the pure holding company is the ultimate form of corporate governance that aims to govern both large firms and their subsidiaries. From the TCE viewpoint, first of all, the pure holding company, like the holding-operating company (jigyo mochikabu gaisha), is considered as a preferable corporate system in that it encourages the transactions in the vertical keiretsu. By holding more than 50% of the equities issued by their first-order subsidiaries for a long time, the pure holding company serves to maintain the long-term transaction relations, directly with those first-order subsidiaries, and indirectly with the second- or even lower-order subsidiaries, thereby contributing to sustaining the whole group of vertical keiretsu transactions. Once an anomalous event is detected somewhere in the ladder of the vertical keiretsu, the (operating) parent firm in charge is expected to swiftly respond to it as the largest shareholder. When the issue turns out to be unresolved even after that, the pure holding company will eventually take on the issue, exercising ultimate authority to replace the problematic 265 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. director(s) or by selling off the equities of the problem firm. What requires further notice of the pure holding company system is that, unlike the holding-operating company, which traditionally served as the holding-operating companies and thereby fulfilled the governance role over their subsidiaries as explained above, the pure holding company can extend its governance even to the large firms. Thus, the pure holding company is expected to fill in the governance lacuna brought about by the debilitating status of banks over the course of industrial transformation which, during the bubble economy period, led them to commit various speculations and scandals.4 6 In light of institutional complementarity, too, the holding company system is judged worthy of introduction. It stores a substantial part of the subsidiaries’ equities whose value represents the assets both physical and human specific to the firm and thereby contributes to preserving the Japanese-style of internal labor market of the subsidiaries. Even in the middle of the unprecedented depression in the postwar period, which rendered the Japanese-style employment system cast under deep skepticism, the system was firmly maintained and supported by a majority of Japanese firms. This was symbolically revealed by the facts that a large part of Japanese firms were unwilling to adopt the U.S. style “committee system” for the corporate governance tool, when it was made available after an overhaul of the Commercial Code in 2002, and that, even among those firms making the decisions to adopt the system, there was none, other than Hitachi Co., which could meet the quintessence of the system—the differentiation of two functions of the entire corporate decision makings and the day-to-day management (Gakushu Kenkyusha 2003: pp. 11-2; Milhaupt 2003: pp. 18-9; Okumura 2003: p. 192).4 7 Topping large firms, the pure holding companies in Japan are rather expected to strengthen directly and indirectly the internal labor market of their subsidiaries and to prepare the promotion ladder for their employees that extends from the lower, to the managerial, and eventually to the directorial positions in the holding companies themselves. 266 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Such an effective m eans notwithstanding, the current state o f the pure holding system in Japan still harbors one serious issue to be resolved—a subtle question of who is going to govern the pure holding companies themselves. This issue deserves serious consideration because the state of governance over holding companies will necessarily influence the two notable features of the Japanese labor market. If through their listings holding companies are directly exposed to the active securities market, these features will eventually come into dysfunction and eventually offset the advantages of the introduction of the pure holding company system. This scenario is not a mere product of pessimism: Table 67 shows a list of the firms which had either filed for setting up their pure holding companies with the Fair Trade Commission or officially announced to set them up by September 2000, along with the information of the type of the holding company they opened and their possibilities of listing them in the securities market. As shown, 19 out of 21 firms which set up 48 holding companies introduced their equities in the securities markets. In the past when those firms were engaged in cross-shareholdings or when the securities market was generally inactive, their introduction or listing of the equities did not seriously engender the problem stated above. With cross-shareholdings gradually unfolding, and the securities market, in parallel, transforming into the more active corporate market since the introduction of the Big Bang, however, the danger seems developing from mere pessimism to reality. When those firms realize the issue as quite impending and start to fear the possibilities of hostile takeovers, they may resume intensively engaging in cross-shareholdings, which, in turn, will cause another problem of inappropriate corporate 49 governance. 267 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 67. The Japanese Firms Planning to Set Up Their Pure Holding Companies as of September 2000 Firm ------- Type o f holding company Introduction of equities in the securities market Date o f introduction (in the case o f a pure holding company) Acmos P (stock swap) Y 1/2000 ADVANEX 0 Y N/A C&S P (corporate breakup) Y 3/2002 Chubu Electricity 0 Y N/A Daiei P (casting-off) Y 12/97 Daiohs P (corporate breakup) Y 1/2000 DaimlerChiysler Japan Holdings P (casting-off) Y 1/1999 Ebara 0 Y N/A Eiden P (share transfer) N 3/2002 Fujitec 0 Y N/A Hitachi Co. O Y N/A Hokuyaku 0 Y N/A HOYA 0 Y N/A INAX Holdings P (corporate breakup + stock swap) Y 10/2001 Ion O Y N/A Itochu Co. 0 Y N/A JFE Holdings P (share transfer) Y 3/2001 Kirin Beer 0 Y N/A Kobe Steel 0 Y N/A Matsushita Electric o Y N/A Mitsubishi Chemical 0 Y N/A Nichimen-Nissho-Iwai Holdings P (share transfer) Y 2/2004 Nikon 0 Y N/A Nippon Telecom Holdings P (corporate breakup) Y 8/2002 Nippon Valqua Industries O Y N/A Nisshin Flour P (corporate breakup) Y 7/2001 NTT P (casting-off) Y 7/1999 Table continues next page R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 67 (continued). The Japanese Firms Planning to Set Up Their Pure Holding Companies as o f September 2000 Firm Type o f holding company Introduction of equities in the securities market Date o f introduction (in the case o f a pure holding company) Okabe P (merger + share transfer) N 7/2001 Sanyo Electric 0 Y N/A Seiko P (casting-off) Y 7/2001 Shin Nippon Steel O Y N/A Shin-Nikko Holdings P (share transfer) Y 4/2002 Shosen Mitsui O Y N/A Shidax P (share transfer) Y 3/2002 Soft Bank P (casting-off) Y 10/1999 Sumida Corporation P (corporate breakup) Y 6/2002 Toa Valve Group P (share transfer) Y 3/2000 Toshiba O Y N/A Tostem Group P (stock swap + merger) Y 10/2000 Toyota Motors O Y N/A Yasukawa Electric O Y N/A Sources: Kosei Torihiki Iinkai Jimukyoku (2001b), Nihon Keizai Shimbunsha (2004). Note. “O” and “P” in the “Type o f holding company” respectively represent operating holding company and pure holding company. “Y” and “N ” in the “introduction o f equities in the securities market” respectively represent yes and no. Probably the best way to cope with this problem is to relax the restrictions on setting up “unlisted” pure holding companies (hijojo mochikabu gaisha), which will be completely blocked off from the havoc wrecked by the securities market. Under the regulation widely held by the securities exchanges in Japan, when parent firms are unlisted, their subsidiaries are also required to lose the listing status for the purpose of the investors protection (Kato and Matsuno 1991: pp. 133-4; Hakoda, Hocchi, and Otani 2000: pp. 145-8).5 0 By relaxing the requirement, the Japanese holding companies will be able to circumvent the excessive pressures from the securities market and thereby effectively preserve the transaction relations with their subsidiaries and their internal labor markets. Simultaneously, they will be able to keep their vantage point of exposing their subsidiaries to the 269 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. securities market, thereby allowing them to tap into direct financing and to nourish their competitive edge. This policy recommendation, of course, entails some cost that pure holding companies must incur. That is, in exchange for the relaxed conditions for gaining the unlisted status, pure holding companies will need to succumb to another requirement of protecting the investors to their subsidiaries. The best way to accomplish this is, on the one hand, to further encourage the institutional investors and outside directors to assume the monitoring task for the pure holding companies and, on the other hand, to place more stringent conditions on their disclosure requirements and fiduciary duty. 4.5.2 Share Buyback The above recommendation introducing the holding company system in Japan is fundamentally based on the recognition that since the oil crises the boundaries of the keiretsu “meta firms” drastically changed across a wide range of industries in the way that the inter-firm trade relations shifted from the horizontal ones among large-firms to the vertical ones between those large firms and their small or medium-sized subsidiaries. Thus, such a policy recommendation should be synchronized by another one facilitating the reduction of cross-shareholdings among the large-firms, especially those engaging in the heavy and chemical industries, to the appropriate levels reflecting their current transactions. The reality, however, went contrary to the wisdom of the TCE: As seen in Section 4.2.1, in the case of the six horizontal keiretsu, cross-shareholdings among the member firms has been a constantly increasing trend since the mid-1980s, and such a trend was reinforced by their unscrupulous equity financing in the late 1980s. This trend continued until various disastrous impacts came to the surface: immediately after the collapse of the bubble economy, the non- financial firms suffered enormous debts accruing from their equity financing, some of which, due to the decline of the stock prices, became ineligible for converting into equities. The banks, on their part, were caught by enormous bad loans, especially in the three non-production sectors. Other non- 270 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. keiretsu firms, too, went through some kind of anomaly, which often came to light in the development of corporate scandals. All these unprecedented accidents are largely due to the lack of proper corporate governance which, in turn, was brought about by the cross-shareholdings contributing more to insulating the corporate managers than strengthening the mutual monitoring (Nissei Kiso Kenkyujo 1997). In order to help reduce the cross-shareholdings among large firms, the “share buyback” (jishakabu shokyaku) system is the best means for allowing them to repurchase the equities they issued in the past. To fully understand this, it is essential to understand that the system shares with the cross-shareholding practice the effect of offsetting the capital raised between the cross-shared firms: the idea being the issuing and purchasing of the same values of equities among affiliate firms, the cross-shareholding practice is, in essence, tantamount to the intra-group equities exchange.5 1 Equivalently, the aggregate, real value of capital inside a whole keiretsu “meta” firm remains constant at the original level, no matter how many times the affiliate firms exchange their equities. The only difference between the cross-shareholding practice and the share buyback system is that, while the former works to incrementally increase the nominal value of the mutually held equities, the latter serves to decrease them. During the period when setting up a pure holding company was not allowed and large firms could not transfer their monitoring functions from their associate large firms to their own pure holding companies, the managers of those large firms might well have strengthened their cross-shareholdings in trying to protect their management autonomy, even though they no longer engaged in so many transactions as in the high growth era. But, with the pure holding company system now allowed, the share buyback system is expected to play the complementary role for reversing the past trend of cross-shareholdings. Even for the Mitsubishi Group, which suffered the least from the aftermath of the bubble economy by maintaining its close ties among members positioned in the heavy and chemical industries, trimming cross-shareholdings is considered beneficial, as it will prepare the way for the group firms to fine tune the levels of cross 271 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. shareholdings and to reassign the chief monitoring roles from their affiliates to their own pure holding companies.5 2 Preposterous as it sesems that the share buyback system had been consistently banned, while the cross-shareholding practice has gone virtually unchecked all the time, the system was finally introduced in October 1994 by amending Section 210 of the Commercial Code.5 3 It has since gone through various reformations for better use by firms (see Table 68). Furthermore, in November 2001, the treasury stock (kinko kabu) system, long banned on the same ground as the share buyback system, was also implemented so that firms can repurchase their equities and hold them with no quantitative limit.5 4 In consideration of the above analogy between the cross shareholding practice and the share buyback system, however, the system still missies an important point in resolving the issue inherent in cross-shareholdings.5 5 That is, the share buyback system recently developed in Japan only allows for the repurchase of stocks through the public stock markets (Kanda, Hideki, “Kabushiki shokyaku to onaji yoken de kinkokabu wo mitomeyo," Kin ’ yu Bijinesu, April 2001, pp. 28-9; Sheard 2001: p. 75), and as such there will be an imminent danger that the general market participants may purchase the cross-held stocks and sell them off after realizing that the values of the stocks are overestimated in the nominal term. Or knowing this possibility, the firms may become less willing to discharge their stocks to the markets, and cross shareholdings may consequently not be dissolved so much as desired at the outset. In fact, according to Inoue (2000: pp. 11-2), of the 430 firms announcing share buyback during 1996-99, there was no significant correlation observed between the rates of their cross-shareholdings and those of their share buyback. Besides, Tanaka (2003: pp. 66-73) pointed out that most of the stocks sold off by large firms were actually put in the “cross-trade,” only moving from one firm to another within the same enterprise (keiretsu) group: Never failing to overlook that the stock market has never crashed when the dissolution of cross-shareholdings was expected to climax, and the foreign investors are, in parallel, increasingly gaining the prominence as stockholders, Tanaka (2003) reckoned that those 272 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. “foreigners” are actually “brown-eyed”—Japanese banks and firms in disguise of foreigners—which first sold off their cross-shared stocks, but immediately repurchased them through the financial institutions of foreign soil.5 6 Table 68. The Reformations of the Share Buy-Back System Date o f implementation Concerned law Contents November, 1995 Special Measures Act concerning Temporary Taxation Temporary termination o f the deemed taxation on share buy-back until the end of 1998 June, 1997 Exceptional Provision o f the Share Buy-Back (1) Delegation o f the decision authority of share buy-back from the general stockholders’ meetings to the board o f directors through the modification o f corporate charters (2) Allowance o f share buy-back for the purpose o f stock option March, 1998 Exceptional Provision o f the Share Buy-Back Temporary allowance o f using legal reserves (shihon junbikin) for share buy-back until the end o f March 2000 (This deadline was later postponed to the end o f March 2002.) October, 1999 Law Concerning Revaluation o f Land Allowance o f using the latent profit o f land for share buy-back Source: Kuroki (2003: p. 7). In order to resolve these issues, therefore, it is imperative that the share buyback system be extended in such a fashion that large, cross-shared firms can directly re-exchange their cross-held stocks with no requirement to release them in the public securities market. If the extended system— which thereafter is dubbed “direct, mutual share buyback (DMSB) system”— is admitted, it should be carried out by the acquisition-values (boka). This is because the stock prices entered in the books are never influenced by any vicissitude in the industrial organization and thus serve as usefiil indices for the firms to refer to the prices they exchanged their stocks for in the past. The market value (jika), on the other hand, is problematic in the DMSB scheme because the changes in the industrial structure have most likely complicated the market values of the stocks across industries, and 273 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. therefore there is no guarantee that cross-shareholdings, especially those made prior to the oil crises, will be decreased to the appropriate levels when carried out by the current market values. Furthermore, cross-held stocks should be clearly given a definition in the DMSB scheme. This is because when firms are allowed to freely buy and sell their partners’ stocks any moment in the public market, they may not be able to reach the agreement regarding which portion should be 57 cross-held (Yoshida 2001: p. 104). For this DMSB scheme and the share buyback system to work out properly, it is indispensable that Japan’s Financial Service Agency (FSA) conduct the overarching task of policing the firms so that they will not commit any misconduct such as insider dealings and other forms of stock price manipulation. More specifically, the FSA’s policing task should include obliging participating firms to file both the application elaborating on various dealing conditions, such as date, price, and quantities, and the official summary reporting how their DMSB 58 plans were actually carried out. The FSA should also disseminate the contents of these documents to other participants in the securities market, so that they will be constantly updated with the information of the DMSB’s. By so doing, the FSA can restrain to a minimum of probability the mishaps that might occur by the sudden discrepancies between demands and supplies in the securities market. Getting right to the quintessence of cross-shareholdings and permitting the cross-shared firms to leap over the whirlpool of the public stock markets toward the direct, mutual deals, DMSB should be employed as the primary tool for facilitating the dissolution of cross-shareholdings throughout the nation’s economy. This does not mean, however, to exclude a possibility of the short-term policy. If the short-term policy is to be adopted to attain the policy outcome more efficiently, it will need to be devised in such a powerful manner that it can handle the issues that even DMSB may encounter. The ultimate issue that DMSB harbors in its achieving the policy goal lies primarily in its market-oriented nature. Simply put, DMSB still leaves the ultimate decision making up to the cross-shared firms, even though it has solved the issue of how those firms can skip 274 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. over the public securities market in carrying out their mutual share buybacks. Some of those firms, if left alone in the marketplace, might not be able to come to terms with their DMSB deals. For instance, there might occur the case where one of the mutually held firms refuses a DMSB deal proposed by the other, in fear of an increasing possibility of hostile takeover in the event of their dissipating cross-shareholdings. Encountering such a difficulty, the declined firm might finally acquiesce to the refusing partner. Consequently, a market failure obtains where the incumbents are entrenched under their cross-shareholdings, even though their transactions may be negligible. Alternatively, the declined firm might decide to dump the stock of the refusing partner in the public securities market, which itself runs counter to the spirit of DMSB. Against all those undesirable outcomes possible, if a government institution is devised in such a manner to be equipped with the function of purchasing stocks from the declined firm— let it tentatively be called “Cross shareholdings Purchase Corporation (CSPC)— that would serve as an effective means to break up those standstills.5 9 First, by serving as a temporary buffer for the stocks sold from the declined firms, the CSPC can circumvent the immediate fall of those stock prices. Second, and more importantly, through its purchases of the stocks from the declined firms, the CSPC can obtain a fraction of the voting rights against the declining firms, whereby to press them into the DMSB deal with their partner. In exercising voting rights, the CSPC should be equipped with its own criterion with which to judge, first, whether it should force the declining firm into the proposed DMSB deal and, if it should, how many stocks are to be brought from the declining firms. Out of many possible criteria, a prospective one would be a set of values comprising the cross-shareholdings rate and the sales made between the firms concerned prior to the emergence of the bubble economy, say, 1985. Under such a criterion, the CSPC’s purchasing scheme will be made of the following three cases: first, when both sides of the cross-shared firms are judged to own their partner’s stocks more than proportionally to their sales and increases between the present and base years, the CSPC should force a DMSB deal to both firms to eliminate their unproportionately held portions of stocks. If 275 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. there occurs the difference between the values of the stocks submitted to CSPS from both sides of the firms, the DMSB deal should be adjusted to the smaller one (the short-side principle of DMSB), so that the residual portion will be later bought up by the issuing firm by share buyback. In the case where only one of the cross-shared firms owns its partner’s stocks more than proportionately to their sales increase, on the other hand, the CSPC should accept the stock only from the excessively holding side and advise its partner to purchase that portion through share buyback. Finally, when none of the above two cases applies (i.e., none of the cross-shared firms owns its partner’s stock more than proportionately to its sales increases), the CSPC should leave the final decision to those firms.6 0 Having long suffered from its raison d’etre somewhat obscured through political turns and twists, “the Banks’ Shareholdings Purchase Corporation” (ginkonado hoyukabushiki shutoku kiko, BSPC) can be improved along the line of the CSPC proposal above. First and foremost, it is important to recall that, when the original plan was devised in early 2001, leading to the establishment of BSPS, it was actually intended for the dissolution of cross-shareholdings across all industries (Yoshida 2001: p. 115, 140, 143). Greeted by generally indifferent reactions from the non-financial industries and subsequently subjugated as a part of “the Law concerning Restriction, etc. of Banks’ Shareholdings, etc.” (ginko kabushiki hoyu seigen-ho) (Yoshida 2001),6 1 however, the institution, when set up in January 2002 under the name of the “BSPC,” was assigned the limited function of purchasing and reselling the stocks held by banks only.6 2 Apparently, the flaw involved in this approach is that, as long as the stocks of banks are still held by the business firms, the stocks of business firms, which are first sold off to the BSPC by banks, and then to the market by the BSPC, will likely wreck havoc on the securities market. Such a wrong start nonetheless, the BSPC rearranged its purchasing condition within a short period and drew near to the ideal form: In December 2002, the BSPC purchased the stocks of banks from business firms if their stocks were already sold off to the BSPC by the issuing banks, to the value of 'A of the banks’ sales. 276 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Furthermore, in August 2003, the value was readjusted upward to the equivalent of the banks’ sales. Although the ceiling of the business firms’ sales was still unilaterally set by banks in this framework, these extensions no doubt greatly increased the possibility of DMSB between banks and business firms. Also, the abolition of the 8% toll charge set against banks in August 2003 lowered the threshold for banks to partake in the BSPC. While those developments successfully brought about the remarkable increase of the stock purchase by the BSPC,6 3 the CSPC proposal suggests that there is still room for further improvement. First, it is important that the BSPC resolves cross shareholdings, not on the open securities market, but through DMSB by clearly sorting out pairs of firms that are deemed as cross-shared with each other. In this regard, unfortunately, the BSPC seems to still stick to the idea of selling off the stocks to the open securities market, as it announced, in August 2003, to extend its planned closing period from 2012 to 2019, in view of trickling down the stocks in this extended time (Kuroki, October 2003, p. 6). Second, the BSPC should extend its purchasing scope to cope with the cross-shareholdings between business firms as well. Extending the BSPC’s purchasing scope this way, on the other hand, necessarily requires the modification of its bank-led nature, which is typically represented by its full staffing of bank-related officers and its more than 10 billion yen of founding contributions exclusively from the banking circle. Probably, the most reasonable solution for this is to refund the contributions to the banks from 200 billion yen of the government guarantee appropriated for the special account of the BSPC. There are three good reasons to support this solution. First of all, for a practical matter, it would be a daunting task for the BSPC to define, after the extension of its purchasing scope, which firms to charge the contributions. Second, if BSPS transforms itself into the DMSB promoter, it will no longer need so much guarantee as 200 billion yen for the preparation of the second losses that may occur by 2019, and it would, thus, be easier for the government to spend 10 million yen or so as the founding contribution for the BSPC. This is possible primarily because the government can save up the guarantee to a large extent by bypassing the market deals while promoting DMSB, and thus does not have to be concerned so much with the possibilities of the second losses. Third, the government can take 277 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. advantage of placing a firm grip on the BSPC to pursue its policy objective. Because the primary purpose of the BSPC lies in the adjustment of stockholding relations to the contemporary industrial organization, the government will need to staff the BSPC with the task force primarily chosen from the experts of both the FSA and the METI. In order for them to be coordinated with one another to achieve the BSPC’s end, the Council for Economic and Fiscal Policy (CEPP), which, in January 2001, was established for coordinating government agencies, will be expected to play the central role. It is essential to understand that the bottom line of the above BSPC reformation substantially lies in its allowing the government to effectively dissolve cross-shareholdings across industries through the DMSB deals and to save up the public fund largely by doing so. 4.5.3 Cross-Entry between Banks and Non-Financial Firms After clearing cross-shareholdings with large firms, Japanese banks need to redirect their effort to consolidate relationships with the small and medium-sized firms through financing and investment to a level comparable to that of non-financial business firms. As analyzed in the previous section, from the TCE perspective, the banks’ speculation in the three sectors of construction, real estate, and commerce was chiefly caused by their inability to properly finance and, most of all, invest in the small and-medium-sized firms, under the postwar legal framework that has never been modified to accommodate the significant industrial alternation after the oil crises. More specifically, the Antimonopoly Law (Section 11), the Banking Law (Section 16), and the Securities and Exchange Law (Section 65) contributed in their own ways to preventing banks from developing financial ties with small and medium-sized firms through financing and investment: the Antimonopoly Law prohibits banks from holding more than 5% of the stocks issued by non- financial firms. With this restriction, banks have not been able to secure substantial parts of the residual right of small and medium-sized firms to effectively govern them. The Banking Law, on the other hand, specified the areas of business where banks are permitted to enter and precludes banks 278 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. from running non-financial, business firms. Thus, even without the restriction of holding up to 5% of the business firms’ stock set by the Antimonopoly Law, the Banking Law staunchly blocks banks from running small and medium-sized firms. Thirdly, the Securities and Exchange Law prohibited banks from running the securities-related business. The implication of this is that banks have been consistently discouraged from developing their monitoring and screening capabilities of the business firms because, due to the law, they have had no chance to obtain the business firms’ stocks through their underwriting, but by indirectly purchasing them through the stock market. These restrictions being currently in effect, the relaxation of the first two laws (the Antimonopoly Law and the Banking Law), while keeping their applications to large firms, would provide banks with strong incentives to redirect their funds for small and medium-sized firms in the form of investment and to hone their competitive edge against other non-financial, business firms which have been constantly freed from legal restrictions regarding investment for their subsidiaries.6 4 Furthermore, if Section 65 of the Securities and Exchange Law is abolished, that will provide the solid condition for banks to improve their screening and monitoring capabilities of the small and-medium-sized firms via equipping them with the privilege of underwriting stocks issued by those firms, thereby serving as a good incentive for them to free themselves from the traditional collateral-based financing.65,6 6 Two important ramifications will follow from the banking reformation through the modification and abolition of the three laws. First, allowing banks to engage in the underwriting business will contribute to realizing Aoki’s “contingent governance” scheme of main banks in the domain of small and medium-sized firms, because that will enable the banks to check the cash flows of their underwritten small and medium-sized firms and thereby to take timely and appropriate actions for the underwritten firms. The implication of this is that, as surveyed in Chapter 1, the traditional Japanese employment pattern, as mainly characterized by the incomplete labor contract and the internal promotion, will be preserved substantially in the area of small and medium-sized firms.6 7 , 6 8 Second, banks’ underwriting-cum- stockholdings of small and medium-sized firms will 279 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. generally improve their capital-asset ratios. Currently, the FSA’s calculation formula, which reflects that established by the BIS (Bank for International Settlements), requires banks to enter the balance of financing in the denominator with 100% weight, which contributes to significantly lowering the capital-asset ratios of the Japanese banks. Besides, during the post-bubble period of the 1990s when they unanimously suffered from the need to accumulate the substantial reserves in disposing of their bad-loans, they were required to subtract those reserves out of their tier 1 capital entered in the numerator in the formula. By replacing investment, at least partly, for the traditional financing, their capital-asset ratios will be improved through the reduction in the denominator and the increase, by the same amount, in the numerator in the formula. Their reserve will also be reduced as their financing is replaced by investment. In the present transition period in which Japanese banks are recovering from bad-loans, the debt-for-equity swap will serve as the best technique for the Japanese banks to convert some of their loans into investment.6 9 For those very small-scaled firms which are not able to issue their stock due to various cost factors, as pointed out in Section 4.2.2., it is essential to treat their debts as de-facto capital, thereby allowing their main banks to enter their financing into the investment term in the calculation formula.7 0 To be fair with the banking industry and to accommodate the trend of the development of the vertical keiretsu across industries after the oil crises, it is also essential to process the banking reformation concurrently and symmetrically in the domain of the non-financial business industry. That is, it is equally important to allow the large, non-financial firms to set up their pure holding companies and to let these pure holding companies assume the corporate governance over their 71 subsidiaries, regardless of their sizes, by means of financing and investment. Along with the traditional way of large firms nurturing transaction ties with their small and medium-sized subsidiaries through investment and financing, this will further strengthen the inter-corporate ties of the vertical keiretsu. This policy, however, is still open to two crucial issues if not advanced with another policy: the first issue is that the financial availability of a pure holding company is 280 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. constrained by the revenue from the dividends accruing from the stockholdings of its subsidiaries. When this constraint is very severe, a pure holding company likely fails to properly govern its subsidiaries. The second is that a pure holding company is not capable of underwriting the stocks issued by its subsidiaries. All it can do is hold equities once issued by the subsidiaries. To overcome these issues, it will be necessary to allow a pure holding company to set up its banking and 7 2 securities subsidiaries. In this regard, the abolition of Section 65 of the Securities and Exchange Law is indispensable also in the area of financial reformation of the non-financial, business industry. Once the banking and securities businesses are equipped, a pure holding company will be less constrained by its original financial availability: its bank subsidiary will start to gamer deposits from the public and exercise ampler financing for the group firms to smooth out their transactions. Also, the securities subsidiary will utilize the superb capability of evaluating technologies long stored in the affiliated large firm and thereby make the underwriting job easier and more appropriate, especially for the small and medium-sized firms belonging to the same group.73,7 4 This is so true considering the fact that both traditional banks and business firms have been long prohibited from the investment trust and consulting businesses. At the same time as the above proposal is made, however, it will be required to introduce the regulation on the upper limits of financing and underwriting that the bank and securities business thus established can offer to the large firm from which their pure holding company originated. Such a regulation will be needed primarily because of the antimonopoly concern: with no such a regulation, there will be a potential that the gigantic financial establishment may emerge between the bank and securities business and the large firm all governed under the identical pure holding company. It is, thus, important to limit the exclusive financing and underwriting in the area of small and medium-sized firms. This is where the Japanese Fair Trade Commission (FTC) should be assigned a new oversight role after the various legal changes.7 5 281 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. In closing this section, let us consider more deeply the implication of the above governance mechanisms of banks and business firms by means of the two financial instruments—financing and investment—by placing equal emphasis on both, as the previous discussion was rather inclined to the latter at the expense of the former in the context of the argument of establishing a pure holding company. As Williamson (1988) pointed out, this consideration will lead to a better understanding of the distinctive roles played by financing (debt) and investment (equities or capital), especially in regard to the degree of asset-specificity of goods, which has, so far, not been discussed very explicitly. Also, as will be turned out shortly, such a consideration further will contribute to a better understanding of the current economic reform in Japan and to drawing more attention on what is still missing out of the reformation argument. As Williamson argued across his works, the focal point of TCE lies in the study of the governance mechanism to be designed between the transaction agents whereby the ex-post transaction maladaptation between them, arising from their bounded rationality of foreseeing the future economic events and their opportunism subsequent to such maladaptation, will be corrected at minimal cost. Williamson (1988) then contended that, in the case where governance needs to function in the case of vertical integration, how debt and equities should be used for the effective governance critically depends on the asset-specificity of goods under transaction.7 6 This is because, while maladaptation will be more (less) likely to happen as asset- specificity deepens (lessens), the governance cost of the two financial instruments increases (decreases), yet at a differing rate, with that of debt (equities) increasing much faster (slower) than equities. This, in turn, occurs because debt is more of a rule-based or market-like instrument, or, in Williamson’s term, “unforgiving” in that it will need to be repaid along with the interest within the predetermined interval, whereas equities is akin to managerial discretion or administration or again in his term, “more forging than debt.” Thus, concluded Williamson, the more (less) goods are asset- 77 specific, the more likely equities (debts) are preferred as the governance tool. In Japan, the modification/annihilation of the three laws (the Antimonopoly Law, the Banking Law, and the 282 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Securities and Exchange Law) in the way argued previously is considered to prepare the way for the vertical keiretsu to achieve the transaction-cost-saving governance envisaged by Williamson because such modification/annihilation of the legal framework will allow the bank-cum-securities function of the vertical keiretsu to adjust the levels of financing and investment freely according to the asset-specificity of the goods produced by their subsidiaries. As a result of the long trial-and-error efforts to resolve the bad-loan issue, the factors conducive to the establishment of such a new governance scheme are gradually burgeoning in the Japanese economy. From the late 1990s to quite recently, the Japanese government has introduced a series of innovative laws and policies, which then started to propel the private sector to reorganize the asset-specific, or inter-corporate relations, in the ways justifiable to the current industrial trend. What is so remarkable in this process, especially from the asset-specificity point of view, is that as the obsolete inter-corporate relations clear away and the bad-loans dissolve, simultaneous switchovers occur between debts and equities in the way Williamson argued. The first lending impetus to such a movement occurred late in 1998 when the government passed three laws (the Law Concerning Specified Cooperative Business on Real Estate, the Law Concerning Liquidation of the Special Assets by Special-Purpose Company, and the Securities Investment Trust Law) to expedite the liquidation of real estate amassed as investment, in the general contracting firms and as collateral 78 in banks, in the long haul of the post-bubble period. Among all, the second and third laws significantly impacted liquidating real estate for both sellers and buyers: On the sellers’ side, the Law Concerning Liquidation of the Special Assets by Special-Purpose Company allowed real estate owners, the majority of which are the general contracting firms, to set up their own special purpose company (SPC) to sell off their long-held real estate to it. The SPC, in its turn, takes on the task of developing the real estate so that it will eventually become asset attractive enough to serve for rental 79 and/or reselling purposes. This way, real estate started to obtain high liquidity and morphed into financial assets. On the buyers’ side, the Securities Investment Trust Law extended the definition of 283 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. the “securities” to incorporate the case of investment in real estate, thereby making possible “real estate investment trust” (REIT). REIT is, in fact, the most successful case of promoting the liquidation of real estate on a large scale by way of combining real estate with the financial 80 market. During the same period, another branch to ameliorate the asset-specific, inter-corporate relations also sprang up with the “Special Measures for Loan Management and Collection,” which was introduced as a part of the “Financial Reconstruction Law” (October 1998). The most notable feature of this law acknowledged licensed agents, often referred as “servicers,” allowed them to purchase bad-loans from banks without any requirement on the banks’ side to pay donation taxes in the selling process. With this advantage, the law contributed to banks selling off their huge-bad loans and thereby led to the creation of the spot market of bad-loans, where the future incomes of the debt-overridden firms are the most critical factor for the determination of the prices of the bad- loans. At the same time, the law paved the way for creating a liquid market for real estate, as disposing of bad-loans concurrently involves banks selling off their real estate collateral to third parties. To cap these legal innovations was the government’s introduction of the “Financial Reconstruction Program” in September 2002, which distinctively advocated an approach to take on the side-by-side reconstruction of both finance and industry. More specifically, the program featured (1) enforcing banks to strictly evaluate bad-loans, among all those classified under the “special attention,” by using the DCF (discounted cash flow) method, and to correspondingly accumulate the reserves, (2) permitting the treatment of deferred loans by banks to the small and medium-sized firms as equities, and (3) promoting “corporate turnaround funds,” whose main service lies in turning around financially ailing firms from the standpoint of large creditors or shareholders after purchasing their bad debts from banks. Under (1) and (2), the program could successfully drive banks to retract a substantial part of their financing from their large-firm customers, among all those positioned in the three non-production sectors (real estate, construction, and commerce) and to re- 81 extend it to small and medium-sized firms. On the other hand, promoting “corporate turn-around 284 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. funds” extended the scope o f the bad-loan-related business from the level o f m erely dealing w ith the bad-loan-cum-real estate issues through negotiations with banks to the involvement of recovering 82 ailing firms through a hands-on method to improve their production and sales management. Although further development of corporate turnaround funds is yet to be seen, its development together with that of the real state business so far shows the high correlation between the asset- specificity and the capital structure, which, in turn, is critical for understanding the role played by TCE in the economic reconstruction process. That is, while sharing the approach of reconstructing the problematic assets through improving their future cash flows, the corporate turnaround funds and the real estate development business deviate in how they deploy debt and equities in their improving problematic assets. In the case of the corporate turnaround funds, as predicted by Williamson (1988), equities play a more crucial role than debt in their corporate recovering process: Typically, after purchasing the bad-debts of the targeted, small and medium-sized firms from banks at the market prices derived through the DCF method, the funds first exercise the “debt-for-equity swap” to transform substantial parts of the purchased debts into equities in order to gain access to the boards of directors of the ailing firms. The funds can do so because they are immune from the 5% stockholding regulation set by the Banking Law (Fujiwara 2000: p. 62). After accessing the governance of problem firms, the funds typically proceed according to the following three steps: (1) analyzing various issues involved in the problem firms, such as (l.a) market conditions, or the demands and supplies of the goods they produce, (l.b) checking if they share the right position in the current industrial structure, and (l.c) their capital structures. (2) Searching for the M&A possibility. In other words, the funds serach for other firms which can be a sponsor and/or the parent or associate firm for the problem firms once their turnaround process completed. And (3) finally carrying out the legal process such as the Company Rehabilitation Law and the Civil Rehabilitation Law (Tasaku 2002: pp.14-42; Wada 2003: pp. 127-8). Among all the steps above, the most essential one, from the standpoint of the corporate recovering process, as Tomoo Tasaku stressed, is the M&A process, that is, (2) above (Tasaku 2002: pp. 157-8), which eventually entails transferring 285 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. substantial portions of the previously ailing firms’ equities to the sponsoring and/or associate firm. Dealing with real estate becoming more and more liquid once released from the collateral position held by banks, on the other hand, the real estate business, whether it is of RETI or the SPC, is characterized more by borrowing: when purchasing real estate, REIT and the SPC are both leveraged quite highly through bank loans, typically up to 50-70% of their investment in real estate, so that they can attain a much bulkier purchase than they could only with their own paid-in capital (The Nikkei Financial Daily Wednesday edition, June 23, 2004; Matsumura 2002). Furthermore, their loans are of the “non-recourse” type which, unlike the traditional corporate financing, delimits its recourse to those assets to which the loans are actually made (i.e., the real estate, in this case). Necessarily, banks need to consider the resalability of the real estate and its future earnings when they strike loan terms. The more reasonably the real estate is resalable and the higher its future earnings are, the more it will be able to attract financing from banks by their easing the lending 83, 84 terms. Once provided with both investment and banking functions as previously proposed, the Japanese holding companies or large firms will be able to attain the capacity to select the most proper governance forms to their subsidiaries through the best combined equities and debts in light of liquidity of the products they produce. When the need arises, they can exercise an “equities-for- debt swap,” vis-a-vis “debt-for-equities swap,” via share-buy back, to the degree they wish to convert their equities into debt. In particular, because of the consistent growth since the oil crises of the tertiary industry, in which businesses and products are best characterized by their mundaneness, 85 debt financing will most likely persist or even expand in the future Japanese financial sector. This conclusion vastly differs from that upheld by the majority of economists. For instance, in exptrapoloating from the contemporary plight of the Japanese banking sector which has long suffered from the bad loans and the continuously declining demand for their loans, Hoshi and Kashyap (2001: p. 306, pp. 313-27) predicted that the bank-dominated Japanese financial sector will 286 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. soon be transformed into a U.S style, capital market-leading one as the banking sector is gradually replaced by a liquid capital market. Their prediction obtains mainly because they fail to take into consideration the effect of the non-financial firms’ entry into the banking sector, and, in so doing, identify the bank loans as those supplied by traditional banks only. As more and more non-financial firms enter the banking sector, whether their purpose is for smoothing out their intra-group transactions or leveraging their REIT subsidiaries, it becomes less and less justifiable from the industrial organizational viewpoint that the traditional banks keep delimiting their businesses within those defined by the Banking and Securities and Exchange Laws.8 6 With banks now on the learning curve of how to turn around debt-overridden firms effectively making use of both debt and equities under the guidance of corporate turnaround funds, the time will eventually come when the revisions of the laws will be supported by a majority of legislators, and when banks again reveal their competencies (screening, monitoring, financing, and investment) endurable enough to compete against other vertical keiretsu groups through their full-fledged investment business for small and 87 medium-sized firms. 4.6 Concluding Remarks The advent of the low-growth economy after the two oil crises significantly morphed the transaction pattern of the horizontal keiretsu established during the high growth era: the member firms drastically reduced the intra-group transactions and instead shifted their transactions to those of their own subsidiaries. Proportionately, their reliance on the main bank borrowing declined. Furthermore, the capital liberalization of the 1980s allowed them to issue their bonds overseas at more favorable terms than their main banks offered, thereby accelerating this tendency. Getting around the traditional main bank relations, these firms could thus fulfill their borrowing needs while consolidating the transaction relations with their subsidiaries through both financing and investment. Left behind this huge economic vicissitude were the main banks. The Banking Law and Securities 287 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. and Exchange Law remained virtually unchanged since the time of heavy and chemical industrialization. Also, their collateral-dependent lending pattern was mostly preserved. Desperate from seeing their lending constantly shrinking over time to the member firms, the main banks re targeted their lending to a new customer base comprised mostly of small and medium-sized firms, with the excessive bias toward the three non-production sectors of real estate, construction, and commerce. In the vacuum of proper monitoring from their main banks, on the other hand, non- financial firms likewise committed excessive equities financing and other speculative actions. Interpreting from the TCE viewpoint how the bubble economy was created as replicated above, this chapter proposed three policies for correcting the fundamental problems in the Japanese financial system and industrial organization, with the consideration of the TCE and when necessary Aoki’s “institutional complementarity.” They are (a) the deregulation to permit (pure) holding companies, (b) the introduction of share buy-back, and (c) the cross-entry between banking and non- financial businesses: Deregulation was proposed not only because it prepares a way for the banks to set up their securities subsidiaries, but also because it allows large firms to set up their monitoring organ overhead, thereby contributing to strengthening the corporate governance over these large firms themselves which, in the lack of proper monitoring from their main banks, were induced to commit reckless, speculative actions during the bubble economy period. Share buyback was proposed because it gives a chance for large firms to disentagle or reduce their cross-held equities accumulated in the framework of the horizontal keiretsu during the high growth period to the levels that appropriately reflect their current frequency of transactions. In transferring the governance core from the horizontal to vertical level, it is essential to carry out deregulation and share buyback simultaneously. Finally, the proposition of cross-entry between the financial and non-financial sectors is based also on the recognition that the inter-corporate transactions shifted mostly from the horizontal to vertical level. To smooth out transactions between a large firm and its subsidiaries within the same vertical keiretsu, it is expedient for the large firms to be equipped with the banking function and thereby be prepared with ample working capital all the time. Equally important in 288 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. proposing the cross-entry is the consideration that such a combination between large firm/pure holding company and a bank will contribute to preserving the Japanese employment pattern in its small and medium-sized firms. Furthermore, on the flip side of all this argument, it is essential that large banks be allowed to invest in small and medium-sized firms by removing the fetter of the 5% investment regulation stipulated in the Banking and Securities and Exchange Laws. The primary reason for proposing this is that large banks have now substantially lost their customer base from their respective horizontal keiretsu groups, and that was the very reason they were all propelled to commit reckless lending to small and medium-sized firms, especially in the three non-production sectors of real estate, construction, and commerce during the bubble economy period. To achieve the most out of these policies, it is highly desirable to introduce them simultaneously, and, if so introduced, it would be imperative for the government to coordinate the relevant ministries. With only the cross-entry remaining to be achieved among the three policies proposed here as of this writing, the “Financial Reconstruction Program” of October 2002 seems to have lent a powerful impetus to its near achievement. First, the program allowed the “J-loans,” whereby non- financial business firms could raise funds in the form of investment trust and invest them to their small and medium-sized subsidiairies. Second, the program obliged banks to adopt the DCF method for assessing their targeted assets and thus prepared a way for them to invest in, rather than finance to, small and medium-sized firms. In fact, the immediate outcome of this policy was the growing of “non-recourse loans” and “project financing” in recent years, both of which are made solely on future cash flows the assets concerned generate, not on the values of their collateral. Third, by setting up their own or employing from outside corporate turnaround funds as the program advocated, banks were given plentiful chances to leam how to restructure debt-overridden small and medium-sized firms. Quite often, the whole turnaround process involves debt-for-equity swapping where the DCF calculation plays an essential role, owning more than 5% of the equities of the problem firms for the governance purpose, and finding their niche markets, which frequently requires M&A’s with other firms. When judged to have fully acquired the management skills of 289 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. small and medium-sized firms after a certain period, both banks and their pure holding companies should be allowed investment in this area without the 5% ceiling. Equally important to note is that, having come to an end in March 2005, the Financial Reconstruction Program should by no means be regarded as only a first-aid treatment that conduces the completion of the Big Bang financial reform. First and foremost, the program significantly differed from the Big Bang in explicitly considering small and medium-sized firms, most of which at the time suffered from excessive debts. The program then treated the two issues of bad loans and corporate turnaround as stemming from the identical root and thus encouraged solutions from both banks and their customer firms to resolve these two issues simultaneously. In this process, the program supported institutional complementarity that existed between the financial market ([main] banks) and the labor market (the internal labor market of the Japanese firms), which is obviously a missing factor in the Big Bang.8 8 Even when the solutions between banks and their client firms fail, corporate turnaround funds often make a foray into such cases and at the very end of the turnaround process promote M&A’s with sponsoring firms they recruit. In this whole process, there is virtually no role for the securities market to play. Once the purchase is done, the governance of the former ailing firms is endogenized into those sponsoring firms, as in the case of the main bank’s monitoring their customer firms. Thus, even if some discharge of recourses both human and physical may entail in the M&A process, the Japanese employment pattern persists at the former ailing firm in the new environment. It goes without saying that the J-loans also contribute to preserving the Japanese employment pattern. Whether banks start to ran non-financial firms or non-financial firms start to open banks, the financing and investment pattern will greatly change from the one observed during the high growth period. In particular, the equities culture will be sure to prevail across the Japanese economy and even in the domain of financing, the non-recourse loan or project financing will become more prevalent. Such a great change expected to arise in the near fixture notwithstanding, it is also considered inheriting a large part of the tradition established in Japanese goods, finances and labor 290 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. markets during the postwar era in that (a) financing and investment will be made mostly for the purpose of easing asset-specific transactions, (b) they will be made mostly by parent firms (pure holding companies, large firms and banks), not by the liquid securities market, and (c) the Japanese employment pattern will be tightly preserved. 291 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Endnotes for Chapter 4 1. Roe (1992) argued that the keiretsu or main bank system o f Japan and the universal banking system o f Germany have the (partial) integration o f finance with industry in common, and that that makes a striking contrast with American corporate financing, which is best characterized by arm’s length dealings between them. 2. The following table shows the economic positions o f the six gigantic horizontal keiretsu in the Japanese economy when measured by the indices (a)-(g). As can be readily confirmed, their economic capabilities— items (b)-(g)— are impressively huge relative to their sheer number o f firms (a). This alone would be a good reason for devising an economic policy for the six gigantic keiretsu. Table 69. Economic Positions of the Six Gigantic Horizontal Keiretsu in the Japanese Economy when Measured by the Indices (a)-(g) FY 1992 FY 1996 Index Billion yen Ratio (%) Billion yen Ratio (%) (a) Number of firms 164 firms 0.0073 270 firms 0.011 (b) Capital 9,479.5 15.29 10,784.0 15.05 (c) Gross asset 155,719.3 12.52 156,864.7 11.99 (d) Net asset 37,344.7 15.61 43,613.3 16.75 (e) Sales 201,995.9 13.79 190,799.0 13.17 (f) Operating profit 2,769.8 10.63 4,054.2 14.59 (g) Profit in the period concerned 1,288.3 16.52 1,735.6 19.59 Source: Kosei Torihiki Iinkai Jimukyoku (1998a: p. 4). 3. After identifying the similarity between shacho-kai and non-shacho-kai firms, Gerach (1992: p. 182) ventured into an inquiry o f what distinguishes shacho-kai member firms from non-shacho-kai firms. Analyzing the samples made o f 200 listed firms, he found that (1) shacho-kai firms are disproportionately represented in the producer-oriented industries and the intermediate product markets, that (2) shacho-kai firms have nearly twice the debt-equity ratios o f non-shach5-kai firms, that (3) firms with close group affiliations perform no better than other firms in profit earnings, and that (4) group member firms are much more likely to be diffusely held (Gerlach 1992: p. 182, 186). Taking into consideration the sufficiently small group coefficients (1-2%) in his model, he concluded that “history” is the main determinant o f the group membership (Gerlach 1992: p. 193). 4. The following are other reasons and their percentages: 40.9% replied for the stabilization o f stock prices, 31.8% for the business affiliation, 29.5% for the anti-buying-up purpose, 4.5% for consolidating capital, 4.5% for meeting the request from the main bank, and 2.3% for other reasons. The multiple replies were allowed in the questionnaires. See Kosei Torihiki Iinkai Jimukyoku (2001a: pp. 6-7). 292 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 5. In the past 10 years, the Japanese government spent approximately 130 trillion yen for fiscal policy. Also, since February 1999, the Bank o f Japan continuously held down its official rate to zero percent, but none o f these seems to have worked out. 6. Closely related to this, Gerlach (1992: p. 109, 141) argued that the waning o f the horizontal keiretsu today was mostly ascribed to the transformation o f the Japanese industrial organization away from heavy and chemical industries to more consumer-oriented, knowledge-intensive industries. 7. The Law on Temporary Measures for the Stabilization o f Specified Structurally Depressed Industry (the Structurally Depressed Industries Law, 1978-83) and the Law on Temporary Measures for Structural Reform o f Specified Industries (the Structural Reform Law, 1983-1988) designated some heavy and chemical industries as “structurally depressed” and facilitated the disposition o f their overcapacities. In a parallel move, keiretsu groups, among all their main banks and GTCs, also played significant roles in scrapping the overcapacities and spreading out among other member firms unemployed labor and the losses generated in the process. For details, see Sheard (1991c) and Uekusa (1987: pp. 490-9). 8. Selling stock by the market price became possible in the late 1960s and encouraged thereafter by the MOF. See Okumura (1998: p. 61) on this matter. Also, note that financial liberalization o f the 1980s also contributed to motivating banks to hold more stock issued by non-financial firms in order to stay in touch with them. See Miyamoto (2000: pp. 199-200). 9. Main bank rent is defined as the premium higher than the normal market interest rate which banks can obtain through directing their customer firms to achieve higher earnings (Aoki 1995: p. 132). 10. The TCE would agree with the first one, if it emphasizes the aspect o f leading to somewhat rigid lending rates, which in turn contributed to depriving banks o f their capacity to monitor firms. But there is actually a suggestion by Enkyo (1993: p. 118) and Suzuki (1987: p. 147) that the lending rates, both short- and long-term, were generally quite flexible, as they were not stipulated by law. The TCE agrees with the second and third ones because they failed to provide banks with the incentives to evaluate and monitor the small and medium-sized firms. The TCE cannot say anything about the fourth one as such a factor is not considered in the model. 11. It goes without saying that such a tendency was accelerated by non-financial firms’ enriching their internal funds and by a series o f financial liberalizations allowing them to raise capital directly from the capital market at favorable conditions, such as the relaxation o f collateral requirement in the bond market. 12. The original purpose o f Section 11 was to allow financial institutions to hold up to 5% o f stock issued by non-financial firms under the status o f trustee business. 13. Shimotani (1996: p. 29) argued that the co-existence o f Sections 10 and 9, which prohibited pure holding company (junsui mochikabu gaisha), served as the real factor for the former zaibatsu firms to form their enterprise groups or, in the terminology used in the present thesis, the horizontal keiretsu. 293 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 14. This argument above also casts doubt against the common view that the Japanese (main) banks served as the missing-link for the M&A market in Japan: the argument instead suggests that the banks were not so competent as non-financial firms in conducting M&A’s, especially those involving small and medium-sized firms. In fact, Ubukata (1986: p. 194) showed that M&A’s of the late 1980s were mostly carried out by the non-bank financial institutions set up by non- financial business firms. 15. The terms “non-bank” and “financial subsidiary company” seem to have been used inter- exchangeably by authors. They are meant to be the same financial entities in that they do not take in deposits from the public, but they engage in corporate or consumer lending without any regulation imposed on by the Banking Law. According to Ubukata (1986: pp. 43-44), there are four types o f non-banks or financial subsidiary companies set up by Japanese firms: (a) the group financing and the foreign exchange management, which is prevalent among home electric companies such as Matsushita Electric, Hitachi Co., SONY, and Toshiba, (b) group financing and the sales promoting financing, which is popular among automobile companies such as Nissan and Honda, (c) merchant banking, which facilitates trade financing, and which therefore is very popular among GTCs such as Mitsui Co, Mitsubishi Co., Sumitomo Co., Itochu Co, and Marubeni Co., and (d) the pure money-making purpose (zaiteku), which was the main strategy adopted by Shin-Nittetsu and NTT. 16. Reportedly, when GTCs opened their financial subsidiary companies in Europe, their main banks expressed their concerns. See Ubukata (1986: p. 80, 88, 114). 17. As a former employee in charge o f the small business financing at the Long-Term Credit Bank of Japan around 1980, Yanai (2002: pp. 54-6) recollected that at the time banks were in general under serious lack o f the financial data o f small and medium-sized firms and relied mostly on the data sets subscribed to from outside think tanks. 18. Reportedly, banks often recommended that small business owners invest in real estate assets as a means to save their inheritance taxes (Kimura 2003: pp. 152-5; Okumura 1999: p. 135, 164). According to a table created by Okumura (1999: p. 122), the small and medium-sized firms increased their loans from 32.2 trillion yen (1983-86) to 95.5 trillion yen (1987-90), out o f which 95.9% was supplied from banks. Correspondingly, those firms’ investment in various financial means increased from 33.0 trillion yen (1983-86) to 65.0 trillion yen (1987-90), out o f which 34.3% was allotted for the real-estate investment (24.5% for the bank deposit, 9.5% for the stock investment, 2% for other securities investment, and 29.7% for the investment in other assets). 19. It was not until January 1989 when the short-term prime rate and the long-term lending rate, respectively, terminated to move together with the official discount rate and the long-term debenture rate. Thereafter, banks introduced the new short-term prime rate, which was the weighted average between the official discount rate and the market borrowing rate, and the new long-term lending rate, which was derived by the short-term prime rate and the long-term risk premium. See Enkyo (1993: p. 124) for details. 20. One frequently used index measuring the ease o f borrowing terms for small and medium-sized firms is the “DI index,” which is defined as the number o f firms that perceived borrowing became easier than the previous period minus that which perceived otherwise. It is in the first quarter of 1984 that the DI index turned into positive at the first time in the postwar period (Yabushita and Bushimata 2002: p. 74). 294 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 21. Kokumin Kin’yu Koko’s survey research (August 1998) showed that, among the targeted small and medium-sized firms, 39.6% and 34.8% responded that their major lending institutions were regional banks and credit unions, while only 27.6% were city banks. In another o f Koko’s surveys that inquired about the three stages o f the borrowing terms (“easier,” “not changed much,” and “more difficult”), 54.2% o f the firms replied that borrowing from the city banks became “more difficult,” the highest value among all the cases (in the cases o f regional banks, the second-tire regional banks, credit co-ops, and credit unions, the corresponding values were 39.7%, 48.1%, 43.0%, and 39.2%, respectively). See Yabushita and Bushimata (2002: pp. 85-8). 22. O f course, those privileges granted to the banking industry must be shared by other non-financial firms. See Section 4.5.3. 23. While the concept o f institutional complementarity was on its way to completion, many researchers inquired about which period o f the Japanese history such institutional complenetarity emerged between banks and the labor market. They, then, found that the prototype o f the complementarity could be located in the wartime economic regime o f WWII. For details, see Aoki (1995a) and Okazai and Okuno-Fujiwara (1994). While admitting this, Aoki (1995a: p.21) cautioned against the simplified view that equalized the wartime prototype with the present-day main bank system and emphasized the importance o f the democratic transformation o f the wartime institutions in the postwar economic reforms. **“Okazaki and Okuno-Fujiwara” was added to bibliography I sent to you the other day. “Okuno-Fujiwara” sounds a bit strange, but is a last name.** 24. Aoki (1994d: p. 45, 51) showed two good examples o f how shared learning is promoted in Japanese firms. One was about a steel producing plant where an integrated engineering control room exists side by side with the engineering office for each workshop and where job rotation is frequently done between them to facilitate knowledge sharing and discourage the development of shop-centered interests. The other was about an auto manufacturer where on-site problems are solved by delegating the shared knowledge among neighboring workers, team leaders, and foremen, but not outside specialists such as mechanics. Koike (1999: p. 148) also pointed out that there hardly occurs a territorial conflict between the operators o f the lines and the mechanics in Japanese firms. 25. The rank system took its clear shape in the 1970s when the streamlined business became prevalent across industries (Miyamoto 1999: p. 19). In the system, the wages were clearly linked with rankings, which were granted by the personnel department to the workers in light o f their past performances (Koike 1997: pp. 100-8; Miyamoto 1999: p. 75). The composition o f basic pay (kihon-kyii) in a major automobile company, as shown below from Koike (1997: p. 102), reveals that the so-called “seniority wage system” (nenko chingin seido) is only a mythology, and how highly the basic pay is linked with the rank pay (shokuno-kyu) : (a) regular pay (30%), (b) age pay (25%), (c) rank pay (slightly less than 40%), and (d) performance pay (about 7%). Koike observed that the age pay declines when the workers become 53 years old or so (Koike 1997: pp. 104-5). 26. Aoki (1989a: pp. 350-1, 359, 1994d: p. 51, 55, 61) dubbed the complementarity between the incomplete governance market and the Japanese internal labor market “the first duality principle” and the two differing levels o f the human resource coordination o f the Japanese firms “the second duality principle.” Note that from these two duality principles and the argument made so far, it is often easy to explain other stylized facts pertaining to the Japanese labor market, such as its illiquid nature and the company-specific unions (see, for example, Sakakibara 1995: pp. 131-2). 295 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 27. Note that shareholder functions in the governance market of the U.S. include three essential characteristics o f the agency model o f the firm. They are (a) hierarchical decomposition o f control originating at stockholders, (b) market-conditioned incentive contracting, and (c) control o f the management decision according to the value maximization criterion. See Aoki (1989a: p. 351; 1994d:p.43,48, 5 5 ,6 9 ). 28. Some notable consequences o f U.S.’s adoption o f the job system is that unions are organized industry- or job-wise (Sakakibara 1995: p. 132), and that workers are not generally cooperative with each other in their being conscious o f the specific jobs assigned (Miyamoto 1999: p. 103). Obviously, these characteristics are diametrically opposite to the Japanese rank system. 29. Strictly speaking, the establishment o f the U.S. job system was in the 1980s. Before the period, the U.S. labor market was characterized more by an internal labor market (Miyamoto 1999: pp. 140- 3). Interesting enough to suggest Japan’s case, the U.S. economy before the 1980s was mainly characterized by the “managerial capitalism” (Alfred Chandler) which was in turn ascribed to the Berle-Means problem o f separation between ownership and management (Miyamoto 2000: pp. 109-10, pp. 123-4). When corporate raiders started to embark on hostile takeovers in the early 1980s, however, the internal labor market and managerial capitalism receded in parallel from the U.S. economy (Milhaupt 2003: pp. 29-30; Miyamoto 2000: pp. 128-9). 30. Koike (1997: pp. 28-9, 32.-4, 122, 138-41) pointed out that the labor market o f small and medium sized firms is mostly characterized by its internal nature where OJT training across various shop- floors is frequently exercised. Besides, he pointed out, similar to the large firms is the wage-curve which slopes upward till the age o f around the mid-forties and downward thereafter. Koike’s in- depth study showed that such a similarity was chiefly brought about by blue-collar class’s adoption o f the employment mode o f the white-collar class after WWII. See Koike (1997: p. 76, 101, 144). 31. Note that Asanuma (1989, 1992) did not use the term “the contingent governance o f team” in his work. For the issue o f parent firms’ evaluation o f the performance o f their subsidiaries, see also Yaginuma (1993: p.. 32, 36, 37). One subtle difference in the roles played by the main bank depicted by Aoki and the parent firms reported by Asanuma is that while the former pays attention exclusively to the earnings gained by its customer firm, the latter pays attention not only to its earnings and costs that need to be paid to their subsidiaries but also to the quality o f the products delivered from their subsidiaries. 32. “Two-vendor policy” in the vertical keiretsu is best compared to the competitive process in the horizontal keiretsu that arises from the division o f labor among member firms. 33. The main pillars o f the U.K.’s Big Bang were fivefold: (a) liberalization o f brokerage fees, (b) abolishment o f the title-grant system that separated dealers and brokers, (c) abolition o f the restriction on investing the securities exchange members (dealers and brokers), (d) introduction of the “market makers” for the improvement o f competition in the securities exchange, and (e) transition from the floor-based to screen-based trading (Mukai 1997: p. 156). Also note in the U.K.’s case, there was no reformative measure o f lifting the separation between the banking and securities sectors as there was traditionally no separation between the two sectors (Mukai 1997: pp. 156-7). 34. The Diet-amended Foreign Exchange Law in May 1997 to take effect in April 1998. The amendment abolished the authorized exchange bank system (Dekle 1998: p. 240; Floshi and Kashyap 2001: p. 290; Mukai 1997: pp. 128-9; Royama 2000: p. 259). 296 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 35. The other four items the SEC listed in the interim report were (a) to improve the functioning o f the Tokyo market, (b) to create a framework suited for the 21st century, (c) to open the market to everyone, and (d) to achieve a fair, transparent, and trusted market (Horiuchi 2000: p. 237; Hoshi and Kashyap 2001: p. 290; Royama 2000: p. 269). 36. For instance, the production o f “asset-backed securities” was long prohibited due to the criminal and commercial laws (Royama 2000: p. 263). 37. The vertical keiretsu is less vulnerable to the market-valuation accounting because the accounting is not applied to those firms which are regarded as subsidiaries. 38. As previously fully discussed, the horizontal keiretsu has generally waned since the two oil crises. This by no means is meant, however, to leave the member firms of the horizontal keiretsu suffering from the erroneous diagnosis given by the Big Bang. There is the right way to cure their issue, which is prescribed in the next section. 39. Ubukata (1986: pp. 196-9) showed that as o f 1985, there were only 80 financial subsidiary companies set up by 49 parent firms. 40. In December 1998, the Japan Securities Dealing Association (nihon shokengyd kyokai) lowered the profit condition at the over-the-counter market from more than twenty million yen to zero yen per annum (Kato and Matsuno 1999: pp. 188-9). Around this period, over-the-counter markets opened one after another in various securities exchanges: in December 1998 and November 1999, Osaka and Tokyo Securities Exchanges opened their own over-the-counter markets called “Shin Shijo” and “Tosho Mothers”, respectively. Later on, Nagoya and Fukuoka Securities Exchanges opened “Centrix” and “Q-Board”, respectively (Yabushita and Bushimata 2002: pp. 113-5). For details of their listing conditions, see Yabushita and Bushimata (2002). 41. For the rest o f the firms having tapped into the direct financing in the past three years, 8.2% responded to have raised funds from their staff, 3.1% from their parent or affiliated firms, 1.0% from the investors, and 0.1% from the venture capital (Kokumin Kin’yu Koko 1999: p. 31; Yabushita and Bushimata 2002: p. 118). 42. The second (24.4%) and third highest (24.3%) responses were that “there is no necessity for raising funds” and that “the firm is not large enough to raise funds from outside.” In this survey result, the responses to the three queries are also shown by the firm size measured by the number o f workers, but they are not very much different from the aggregated results. For details, see Kokumin Kin’yu Koko (1999: p. 31-5), Yabushita and Bushimata (2002: pp. 117-21). 43. This also benefits the lending banks in accumulating the information for efficient monitoring o f the borrowing firms. Thus, there emerges the incentive for both the borrowers and lenders to engage in long-term financial relations. In Japan, such relations are often observed between the regional banks and the credit co-ops and their customer firms. See Yabushita and Bushimata (2002: p. 45). 44. In this regard, Yabushita and Bushimata (2002: p. 115) pointed out that the securities market generally places much stricter standards on the redemption capability o f firms than banks. 45. It should be noted that the plan for introducing the holding company system had a totally different origin from other items in the Big Bang. It was first introduced in the “Three-Year Deregulation Plan” o f the Murayama Cabinet in 1995 under the promotion of the MITI. See Sheard (2001: p. 109). 297 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 46. Understanding the pure holding company system as only a natural extension o f the operating holding company system, Shimotani (1996) did not seem to acknowledge its significance. This seems to be primarily due to his underestimation o f banks’ weakening governance over the large firms over time. Also note that the governance by the pure holding company will take a similar form to banks in that it accomplishes the governance task mostly by checking the cash-flow o f its first-order subsidiary, (i.e., the large firm, not by checking the goods delivered from the subsidiaries. 47. About ten months prior to the implementation o f the “committee option,” a survey conducted by Japan Corporate Auditors Association in mid-2002, indicated that only 13 (0.5%) o f the 2,513 firms would switch to the committee-based structure (Milhaupt 2003: p. 19). On January 28, 2003, SONY announced that it would shift in the committee system if approved by the general shareholders meeting in June, but its original scheme showed that 11 out o f 15 directors would co hold the managerial positions (Gakushu Kenkyusha 2003: pp. 11-2). This is completely different from the U.S. style “committee system.” 48. Table 67 also reports that, despite their prior announcements, 19 out o f 41 firms have not yet set up their pure holding companies. While their reasons for not having done so are not clear, they might have found it not so beneficial to do so because o f the issue stated above. 49. For instance, Sheard (2001: p. 90) suggested the possibility that cross-shareholdings may remain between financial pure holding companies and non-financial pure holding companies. 50. One exception to this regulation is the case in which a parent firm acquires its subsidiary through a take-over-bid (TOB) and refrains from holding their shares more than a certain rate. For instance, the Tokyo Stock Exchange (TSE) sets the rate at 75%. Furthermore, in the case o f TOB, the TSE imposes a screening procedure on the parent firm to make certain that the subsidiary is secured o f their independence from the parent (Kato and Matsuno 1991: pp. 133; Hakoda, Flocchi, and Otani 2000: pp. 148). 51. On essentially the same ground, Tanaka (2003: pp. 66-9) pointed out the similarity between the cross-shareholding practice and the share buyback system. 52. The Mitsubishi Group finally admitted one scandal in September 2002, when Mitsubishi Motors was reported to have concealed a recall filed against its defective automobiles. This scandal implicitly informs o f the lack o f proper monitoring in Mitsubishi Motors, which used to be primarily delegated to Mitsubishi Bank and Mitsubishi Heavy Industry. 53. The primary reason for the ban against the share buyback system was in its engendering hollowing-out capital inside firms. By October 1994, two firms were indicted for committing share buybacks: Mitsui Mining (November 1978) and Katakura Industries (September 1981). For the complications o f these two incidents, see Ohashi and Helm (1995: pp. 55-86). 54. The main difference between the share buyback and treasury stock systems is that the former expunges stocks once purchased from their stockholders and that the latter does not and instead allows firms to keep the stocks for future needs o f raising capital. This way, the treasury system has an advantage o f allowing firms to dispense with various costs involved in issuing new stocks. Originally, the stocks repurchased under the treasury stock system were required to be entered in the asset account o f a balance sheet, but the new rule established in February 2002 requires them to be entered in the capital account, as in the case o f the share buyback system. 298 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 55. Note that cross-shareholdings have been often carried out through the allocation to a third party in the same group (daisansha zoshi). 56. See Section 4.4 for the “cross-trade” technique adopted many times over by Japanese banks in their cashing in the latent profits from the stocks they owned. As to the issue o f the so-called “foreign investors,” Tanaka (2003: pp. 71-2) cautioned against the “foreigners” breakouts in the shareholders lists because usually their true names are never publicized. To date, checking each order form kept in the foreign financial institutions is the only way to see the true investors sorted out as “foreigners,” according to a specialist o f the Tokyo Stock Exchange (Tanaka 2003: p. 71). 57. The Japan Federation o f Economic Organization (JFEO) proposed to define cross-shareholdings as those stocks held mutually more than 5 years (Inoue 1998: p. 9), but 5 years may be too short to cover the cross-shareholdings made during the high growth period. 58. These terms also would be useful for National Tax Administration to derive the deemed dividend taxes (minachi haito-zei) generated in the DMSB processes. As to the issue o f circumventing stock price manipulation under share buybacks, the new rules were introduced by Cabinet Law in March 2003: the ceiling o f the purchasing order should be 100% o f the average daily units o f the past four weeks, and the purchasing order can be placed within anytime a stock exchange is open. See http://www.tse.or.jp/listing /guideline/guideline naikaku.html. This Cabinet Law is also applicable to the DMSB case. 59. One prerequisite for the establishment o f the CSPC is for the Diet to reach consensus to allow the CSPC to use a certain amount o f the public funds for stock purchases, although it will eventually be repaid after the successful DMSB’s. CSPS in the present discussion is considered similar, yet opposite in its function, to Japan Cooperative Securities (Nihon Kyodo Shoken), as it was founded at the time o f the securities market crash o f 1964 for the purpose o f absorbing the stocks and redistributing them among banks and non-financial firms in line with cross-shareholding. The best sponsors for the CSPC would be the Financial Supervisory Agency (FSA) and the Ministry of Economy, Trade and Industry (METI). 60. Ideal as it seems, there might be difficulty in introducing the criterion as discussed. Notably, the most difficult part is obtaining sales data made between cross-shared firms in the targeted year. In the past, Japanese financial firms and non-financial firms alike were not required to report such data in their annual securities reports (yuka shoken hokokusho). If such data are unavailable, it is recommended, as the second best plan, that the CSPC simply adopts the cross-shareholdings rates o f the targeting year o f its choice. For its part, the FSA is urged to place an additional requirement on the Japanese firms to disclose the data on both their cross-shareholdings and trade amounts with their cross-shared trading partners. 61. Due to excessive stockholdings and the frequent use o f the cross-trade, Japanese banks became quite vulnerable to price volatility in the securities market. The market-valuation accounting, which was planned to be implemented in FY 2002, was also considered to accelerate such a tendency. Under such an adverse circumstance, the Law Concerning the Restriction, etc. o f Banks’ Shareholdings, etc. was legislated in November 2001. The law aimed to reduce the banks’ stockholdings to their tier 1 capital level so that banks would become less vulnerable to the price volatility o f the securities market. The following Table 70 shows the values o f stockholdings o f 9 major Japanese banks and trusts in 2000 and those of the stocks they planned to sell off in 2001. 299 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Table 70. Values o f Stockholdings o f 9 M ajor Japanese Banks and Trusts in 2000 and Those o f the Stocks They Planned to Sell O ff in 2001 Major Bank Planned selling amounts in 2001 (previous year) Stockholdings (ratio with the equity capital) Mitsui-Sumitomo 670 (400) 6,520.4(1.62) UFJ 670 (642.4) 6,112.2(1.54) Mizuho 570 (930) 7,917.7(1.27) Mitsubishi-Tokyo 555 (233.4) 6,062.5 (1.56) Asahi ......................... 500(350) 1,759.8(1.29) Chuo-Mitsui Trust 300(150) 1,864.2(2.39) Sumitomo Trust 200 (140) 1,251.0(1.64) Daiwa Trust 150(101.5) 1,265.9(1.56) Yasuda Trust 100(170) j 695.8(1.91) Sum 3,715(3,117.3) 33,449.5(1.51) Source: Norm Chukin Sogo Kenkytijo, August 2001, “Kin’yu kikan no kabushiki hoyu seigen to kabushiki kaitori kiko seturitsu,” p. 9. Note. Unit: billion yen. 62. There are two accounts in the BSPC: the general and special accounts. The former was for disposing of stocks purchased instantly through the creation o f EFT (Exchange Traded Fund), share buybacks, and other market transactions. The latter, on the other hand, was a safety net o f compensating the second losses that might occur with the stocks not categorized in the general account. The following argument only addresses the special account case, as the general account is equivalent to market deals. 63. From November 2003 to April 2004, BPSC’s purchases jumped to 666.8 billion yen, from 152.5 billion yen which was made during April to October 2003. See http://ww w.bspc.jp 64. When Section 11 o f the Antimonopoly Law is to be removed for the small and medium-sized firms financing, it would also be important to set the ceiling against the total investment allowed for banks. The Law Concerning Restricting Banks’ Stockholdings introduced in August 2003 is considered a step in the right direction, as it limits the banks’ stockholdings within their tier 1 capital. The definition o f small and medium-sized firms is summarized in the following table. Note that for individually-run firms, only the workforce criterion applies, whereas for corporations either the capital or workforce criterion applies, whichever is relevant. Table 71. Definition o f Sm all and M edium -Sized Firms Sector Capital (Yen) Workforce Production and others 30 million Less than 301 Wholesale 10 million Less than 101 Retail 5 million Less than 51 Service 5 million Less than 101 300 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 65. Actually, for these years, the Securities and Exchange Law has been intensively reexamined and modified. Part o f the modification was to allow the dealing and brokerage business to individuals and non-financial business firms in April 2004 and banks in December 2004. As o f this writing, the underwriting business is not yet allowed for non-financial firms and banks. Also, Yoshida (2001: pp. 119-120) pointed to a movement for relaxing the 5% restriction o f stockholding stipulated by the Antimonopoly Law. Recent success o f the corporate restructuring funds is also remarkable because it is freed from the 5% restriction when restructuring ailing firms. 66. Even if Section 65 o f the Securities and Exchange Law is abolished, it is still important to separate the financing and securities functions strictly by either placing a firewall between them or founding them as two separate corporate entities. A conflict o f interest always lies between the banking and securities businesses. In the latter case, there will be less concern for establishing a firewall than the former, but it is still important to establish the rule or legal framework regarding a personnel rotation between the banking and securities firms, as they still belong to the same holding company. 67. Note that the holding company, which itself is out o f the scope o f the Banking Law and the Securities and Exchange Law, plays a similar role to both investment and commercial banks, as it can supply both financing and investment to its subsidiaries (The holding company is out o f the scope o f the two laws because it does not offer any settlement service, one o f the essential functions o f banks). In fact, the prewar zaibatsu holding companies played such a role to its subsidiaries (Hocchi, Hakoda, and Otani 2000: p. 74; Muto 2003: pp. 123-4, 148-51). It is, thus, possible that the financial holding company, the parent firm o f a bank, can directly govern small and medium-sized firms through the stockholdings. Thus, the problem o f governance combination occurs among the holding company, bank, and subsidiaries (i.e., small and medium-sized firms), and it should be ultimately solved by the holding company in light o f the group strategy, including the importance o f the subsidiaries and the technical issues such as the financial availability at the holding company. 68. Although not from our interest o f corporate governance and industrial organization, Royama (2001) made an important point about the investment trust sold by banks, whereby individuals, not just banks, assume the risk involved in directly financing firms. He called such a style “the market-type indirect financing.” Apparently, the idea o f market-type indirect financing can constitute the banking reform policy proposed in the present section. Recently, under the “Financial Reconstruction Program” (September 2002) o f the Japanese government, the investment trust made o f the stocks issued by small and medium-sized firms was finally allowed. It is generically called the “J-Loan ” For the J-Loan, see Kimura (2003: pp. 159-160) and http://www.meti.go.jp/kohosys/ committee/summary/0001357/ (retrived on July 23, 2004). 69. This does not mean, however, that loans should be freely converted into capital in banks’ favor o f increasing their capital-asset ratios. There is actually an asset-specificity reasoning behind the conversion between loans (debts) and capital (investment), which is discussed shortly. 70. As Kimura (2003: p. 161) explained, the subordinate debt (retsugosai) and the ordinary stock share a similarity in that the interest and dividend payments are both pro-cyclical to the performance o f the firm. In February 2004, under the “Financial Reconstruction Program”, the Japanese government permitted banks to regard their subordinate loans to their small and medium-sized firms as capital in order to improve the banks’ capital-asset ratios. 71. This way, the TCE approach integrates the policy proposals o f the “universal banking” and the establishment o f holding company separably treated by Aoki (Aoki 1994c; 1995a: pp. 191-221), with more emphasis on the ties between banks and small and medium-sized firms. 72. Again, it is important to properly separate the banking and securities businesses under the same pure holding companies. See 66 above. 301 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 73. As part o f the Big Bang financial reform, the non-financial business firms entered the banking industry. In April 2001, IY Bank o f Ito Yokado and Sony Bank o f Sony entered the banking business. As for the revision o f the Securities and Exchange Law in April 2004, they also entered the securities dealings and brokerage businesses. As Section 65 still exists today, these banks are not currently permitted to engage in underwriting business. 74. As pointed out in 67 above, by introducing the banking and securities businesses in the non-financial industry, there will occur a problem o f the governance combination among bank-cum-securities firms, pure holding companies, and business firms (both large and small and medium-sized). Again, the issue should be solved by each group, in light o f the group strategies, the importance o f subsidiaries, and the financial availability at holding company. Also, for the same reason that is discussed in the case o f the banking industry reform, it is essential to set up a legal framework that allows the subordinated loans by banks to be treated as capital in the case o f a bank o f the business firm’s origin. 75. A similar argument holds when the pure holding company established by a bank starts to govern a large firm, or when its subsidiary (subsidiaries) becomes larger than a certain level (in terms o f asset, capital, and workforce). Note that while the new regulation prohibits financing and underwriting between banks and securities businesses and the large firm(s) within the identical vertical keiretsu, it allows such financial transactions as long as banks and securities business and the large firm belong to different keiretsu or enterprise groups. Thus, the limit o f financing and underwriting should be determined in consideration o f promoting the sound competition across various vertical keiretsu or enterprise groups. 76. This is one o f the points where TCE departs from the orthodox economics theories such as reflected in the Modigliani-Miller theorem, which argues that the average cost o f capital to any firm is completely independent o f its capital structure and is equal to the capitalization rate o f a pure equity stream o f its class (Modigliani, Franco, and Merton H. Miller 1958). 77. More specifically, Williamson (1988: pp. 580-1) expressed the governance cost o f debt and equities as D(k) and E(k), where k is an index o f asset-specificity, and postulated that D'(k)>E'(k) for any level o f k. Taking into consideration that D(0)<E(0) because debt is a simpler governance structure and, thus, requires a lower set up cost, he pointed out that a switchover occurs from D(k) to E(k) as asset-specificity increases (Williamson 1988: p. 580). Also, Williamson (1988: pp. 585-6) noted that many o f the target firms o f the leverage buyouts in the 1980s had a high proportion o f their total assets in tangible, mundane, less asset-specific properties, such as retailing, textiles, textiles, and soft drink bottles. 78. The three laws had to be legislated individually despite their sharing the purpose o f promoting the liquidation o f real estate. Such complication arose primarily due to their different origins: the Law Concerning Specified Cooperative Business on Real Estate (fudosan tokutei kyodo jigyo-ho) was issued from the Ministry o f Construction; the Law Concerning Liquidation o f Special Assets by Special-Purpose Company (SPC-ho), from the (present) Ministry o f Finance (zaimusho); and the Securities Investment Trust Law (shoken toshi shintaku-ho), from the (former) Ministry o f Finance (okurasho). All three laws gradually relaxed their original regulations between 1998-2000, typically by lowering the required capital, extending the definition o f asset, thereby improving the liquidity of real estate. In May 2000, the second law was absorbed into the third law, which itself was developed into the “Trust Law” in the same period, as a result o f broadening the definition o f securities. For the details o f the deregulation process o f these laws, see Oka (2000: p. 1). The first law has, so far, been infamous and unpopular. Inoue (1999: pp. 242-3) explained that that is primarily because (a) the certificate is not transferable among customers and (b) it requires relatively higher minimum investment o f 1 million yen. 302 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 79. The essence o f setting up a special purpose company (SPC) lies in its allowing the “bankruptcy remote” to build from the original real estate owner, thereby circumventing the forfeit o f real estate when the original owner bankrupts. Setting up the bankruptcy remote, however, often entails a complex process o f setting up another SPC overseas, typically in the Cayman islands, where the owner’s right is legally transferable from the SPC to a charitable trust. For other features and requirements o f the SPC, see Inoue (1999: pp. 248-55). 80. REIT is an investment fund acknowledged by the Securities Law, but not a stock company stipulated by the Commercial Code. Thus, in Japan the investment certificates issued by REITs are distinguished from other equities even when they are listed on stock exchanges. In September 2003, Nippon Building Fund (Mitsui-affiliated) and Japan Real Estate Fund (Mitsubishi-affiliated) were listed as Japan’s first REITs in the Tokyo Stock Exchange. Gaining popularity since then, as o f March 2005, the number o f REITs reached 50, and their net assets exceeded 700 billion yen, one- quarter o f that money management funds. See Nikkei Weekly April 12, 2005 and April 19, 2005. 81. In September 2004, the bad loans o f 11 mega banks shrunk to 4.7%, more than half the level at the time o f when the Financial Reconstruction Program started in October 2002 (Nikkei Shimbun, September 23). In parallel, the mega banks intensively expanded their financing to the small and medium-sized firms: in February 2005, the balance o f financing to small and medium-sized firms by 4 mega banks reached 224 billion yen, about as twice as the level o f the last year (Nikkei Shimbun, February 12, 2005). It should be noted that behind such a movement was the remarkable technological development o f the “credit risk database (CRD) system” which, through the evaluation o f financial conditions and other qualitative factors such as technological abilities and distribution channels, enabled banks to derive credit ratings and interest charges to each o f their customer small and medium-sized firms, with no recourse to real estate collateral. For more detail o f the CRD system, see Akiyama and Yabushita (2002: pp. 93-107) and Tanabe (2003). 82. Under the Financial Reconstruction Program, the government extensively promoted the corporate turnaround business through such channels as (a) setting up the Industrial Reconstruction Corporation o f Japan (IRCJ) in April 2003, (b) legislating the Small And Medium-Sized Firms Support Law in November 2002 to facilitate setting up the prefectural version o f the IRCJ across the nation, (c) adding the corporate turnaround service to the Resolution and Collection Corporation (RCC) in September 2002, and, most importantly, (d) promoting the private sector to set up the corporate turnaround funds. By February 2003, the number o f funds reached 73, out o f which 20 engaged in the cases involving the bad loan issues (Wada 2003: p.15). As M&A’s are often the most effective means to improve the corporate turnaround process, their number and value are exponentially increasing in recent Japan, as shown in the following table (Forbes Japan, October 2004 issue, p. 134): Table 72. N um ber o f M & As and Their Values (1998-2003) Year Number Aggregate Value (billion yen) 1998 9 0.051 1999 19 0.318 2000 1 1 0.404 2001 22 11.65 2002 39 10.72 2003 67 51.95 303 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 83. Due to its heavily leveraged nature, REIT is vulnerable to the interest rate hike, which is actually underway in recent Japan. See, for example, The Nihon Keizai Shimbun Tuesday edition, May 3, 2005. 84. As o f REIT, for instance, it is reported that Japan Tobacco (JT) set up its REIT subsidiary, and that JT, as its parent firm, financed the REIT subsidiary after the initial investment (Nikkei Weekly, August 27, 2005). 85. As shown in the following table listing, the industry-wide share of the GDP in fiscal 1950, 1970, and 1995, the growth o f the tertiary industry is remarkable from 1970 to 1995. Also, “modulization,” which in recent years has intensively prevailed, especially in the electric and automobile industries, is considered another example o f showing the importance o f debt financing today: “modules” are best characterized by a set o f semi-autonomous systems, which, once combined through their common rules, develop into a more complex module (such as a computer or an automobile). In this regard, each module thereof is considered less asset-specific and debt financing, rather than equities financing, is expected to be an effective means for the production process. For the various aspects o f modulization such as its impacts on the industrial organization, its relation with digitalization, and how it is being coped with by major Japanese industries, see Aoki and Ando (2002). Table 73. Industry-wide Share o f the GDP in Fiscal Years 1950,1970, and 1995 Year The primary industry The secondary industry The tertiary industry 1950 26.0 31.8 42.2 1970 6.1 44.5 49.4 1995 1.8 33.8 64.4 Source: Yoshikawa (1999: p. 115). 86. As o f this writing, there are still new entries occurring in the banking sector. Probably, the most notable case is Nomura Securities, the top Japanese securities house which, since early 2005, has sequentially expanded its business line to incorporate those that have not been treated as mainstays o f the securities business. They are (a) real estate-related business in collaboration with Nomura Investment Trust and bad loan-related businesses (February, June and August 2005), (b) M&A- related business under the affiliation with the Rothschild Group (February 2005), and, most interestingly, (c) banking business (August 2005). 87. It should be pointed out that the analysis o f this section also has a ramification for the accounting system. That is, the more (less) liquid a physical asset is, the more commensurate its securities become with the market-value accounting method (the acquisition-value accounting method). This is because when a physical asset is more liquid, its availability is more likely to be equated with its supply in the spot market and together with the spot-market demand, helps to establish the correct spot-market price. If a physical asset is more asset-specific and less liquid, on the other hand, the above condition will hardly be met. More straight to the point, there will be virtually no spot-market for the asset-specific goods. The values o f asset-specific goods, such as those o f subsidiaries, are rather determined by the cash-flows generated by the purchase from their parent firms. In that case, the adoption o f the acquisition-value accounting method contributes to stabilizing the transaction between the subsidiaries and the parent. In the case when the book value (the acquisition-value) of the subsidiary and its real value reflecting the cash-flow from its parent deviate significantly, the former should be adjusted to the latter after a certain interval. 304 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 88. In closing this chapter, a caveat is in order regarding two terms, “network” as used by Amyx (2004) and “institutional complementarity,” because they are occasionally used interexchangeably, and because some may cast doubt on the institutional complementarity discussed in this chapter due to the negative connotation loaded on the term “network.” At this point, it is crucial to note that the institutional complementarity discussed in this chapter is ascribed to pure economic factors, such as corporate governance, industrial organization, and transaction cost, and that they have no relevance to the venal politics, which occasionally arose within the “network” within the Japan’s financial circle, including the MOF, in the 1990s. In short, the constituents involved in the institutional complementarity discussed in this chapter is constantly exposed to harsh global competition. 305 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. B ib l io g r a p h y Abe, Tadahiko. (2000). Korekara no nihon no ridingu sangyo to sono ikusei [Leading industries in future Japan and their nourishment]. FRI Research Report No. 67, Fujitsu Research Institute. Aizawa, Koetsu. (1997). Yunibasaru banku to kin ’ yu mochikabu kaisha [The universal bank and the holding company], Tokyo: Nihon Hyoronsha. Aizawa, Koetsu. (1999). Nihongata kin’ yii shisutemu wo motomete [In pursuit of the most suitable financial system in Japan]. Tokyo: Toyo Keizai Shinposha. Albert, Michel. (1993). Capitalism vs. capitalism: How America’ s obsession with individual achievement and short-term profit has led it to the brink o f collapse. New York: Four Walls Eight Windows. Amyx, Jennifer A. (2004). Japan’ s financial crisis. Princeton and Oxford: Princeton University Press. Aoki, Masahiko. (1984a). The co-operative game theory o f the firm. Oxford: Oxford University Press. Aoki, Masahiko. (1984b). Shareholders’ non-unanimity on investment financing: Banks vs. individual investors. In Masahiko Aoki, The co-operative game theory o f the firm (pp. 193- 224). Oxford: Oxford University Press. Aoki, Masahiko. (Ed.). (1984c). The economic analysis o f the Japanese firm. Amsterdam, New York, Oxford: North-Holland. Aoki, Masahiko. (1987). The Japanese firm in transition. In Kozo Yamamura & Yasukichi Yasuba (Eds.) The political economy o f Japan: The domestic transformation (Vol. 1, pp. 263-88). Stanford: Stanford University Press. Aoki, Masahiko. (1988). Information, incentives, and bargaining in the Japanese economy. Cambridge: Cambridge University Press. Aoki, Masahiko. (1989a). The nature of the Japanese firm as a nexus of employment and financial contracts: An overview. Journal o f the Japanese and International Economies, 3, 345-66. Aoki, Masahiko. (1989b). Nihon kigyo no soshiki to joho [The organization and information of the Japanese firm], Tokyo: Toyo Keizai Shinposha. 306 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Aoki, Masahiko. (1990, March). Toward an economic model of the Japanese firm. Journal o f Economic Literature, 28, 1-27. Aoki, Masahiko. (1994a, August). The contingent governance of teams: An analysis of institutional complementarity. International Economic Review, 35, 657-96. Aoki, Masahiko. (1994b). The Japanese firms as a system of attributes. In M. Aoki & R. P. Dore (Eds.), The Japanese firm: Sources o f competitive strength (pp. 11-40). Oxford: Oxford University Press. Aoki, Masahiko. (1994c). Monitoring characteristics of the main bank system: An analytical and developmental view. In Masahiko Aoki and Hugh Patrick (Eds.), The Japanese main bank system: Its relevance fo r developing and transforming economies (pp. 109-141). New York: Oxford University Press. Aoki, Masahiko. (1994d). Toward an economic model of the Japanese firm. In Kenichi Imai and Ryutaro Komiya (Eds.), translation edited and introduced by Ronald Dore and Hugh Whittaker. Business enterprise in Japan: Views o f leading Japanese economists (pp. 39- 71). Cambridge, MA: The MIT Press. Aoki, Masahiko. (1995a). Keizai shisutemu no shinka to tagensei [Evolution and diversity of the economic system]. Tokyo: Toyo Keizai Shinposha. Aoki, Masahiko. (1995b). Unintended fit: Organizational evolution and government design o f institutions in Japan. Stanford University. CEPR Working Paper No. 434. Aoki, Masahiko, and Haruhiko Ando (Eds.). (2002). Mojuruka: Atarasii sangyo akitekucha no honshitsu [Modulization: The essence of the new industrial architecture], Tokyo: Toyo Keizai Shinposha. Aoki, Masahiko, and Ronald P. Dore (Eds.). (1994). The Japanese firm: Sources o f competitive strength. New York: Oxford University Press. Aoki, Masahiko, and Hyung-Ki Kim. (1995). Corporate governance in transitional economies. Washington, DC: The World Bank. Aoki, Masahiko, Kazuo Koike, and Iwao Nakatani. (1986). Nihon kigyd no keizaigaku [The economics of Japanese firms]. Tokyo: PHP Kenkyujo. Aoki, Masahiko, and Masahiro Okuno (Eds.). (1996). Keizai shisutemu no hikaku seido bunseki [The comparative institutional analysis of the economic system], Tokyo: Tokyo Daigaku Shuppankai. Aoki, Masahiko, and Hugh Patrick (Eds.). (1994). The Japanese main bank system: Its relevance for developing and transforming economies. New York: Oxford University Press. Aoki, Masahiko, Hugh Patrick, and Paul Sheard. (1994). The Japanese main bank system: An introductory overview. In Masahiko Aoki and Hugh Patrick (Eds.), The Japanese main bank system: Its relevance for developing and transforming economies (pp. 3-50). New York: Oxford University Press. 307 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Asahi Shimbun Keizaibu. (1999). Kin ’ yu doran [Financial turmoil]. Tokyo: Asahi Shimbunsha. Asanuma, Banri. (1989). Manufacture-supplier relationships in Japan and the concept of relation- specific skill. Journal o f the Japanese and International Economies, 3, 1-30. Asanuma, Banri. (1992). Japanese manufacture-supplier relationships in international perspective: The automobile case. In P. Sheard (Ed.), International adjustment and the Japanese firm (pp. 99-124). St. Leoneards, NSW, Australia: Allen & Urwin in association with the Australian Research Centre, the Australian National University. Auerbach, Alan J. (Ed.). (1987). Corporate takeovers: Causes and consequences. Chicago: The University of Chicago Press. Beller, Alan L., Tsunemasa Teirai, and Richard M. Levine. (1992). Looks can be deceiving—A comparison of initial public offering procedures under Japanese and U.S. securities laws. Law and Contemporary Problems, 55, 77-118. Beston, George, J. (1996). The origins of and justification for the Glass-Steagall Act. In Anthony Saunders and Ingo Walter (Eds.), Univeral banking: Financial system design reconsidered (pp. 31-69). Chicago: Irvin. Campbell, John. Y., and Yasuhiko Hamao. (1994). Changing patterns of corporate financing and the main bank system in Japan. In Masahiko Aoki and Hugh Patrick (Eds.), The Japanese main bank system: Its relevance fo r developing and transforming economies (pp. 325-49). New York: Oxford University Press. Caves, Richard E., and Mass Uekusa. (1976). Industrial organization in Japan. Washington, DC: The Brookings Institution. Coase, Ronald H. (1937). The nature of the firm. Economica, IV (13-16), 386-405. Coffee, John C. (1991). Liquidity versus control: The institutional investor as corporate monitor. Columbia Law Review, 97(6), 1277-1368. Corbett, Jenny. (1998). Changing corporate governance in Japan. In Morten Balling, Elizabeth Hennessy, and Richard O'Brien (Eds.), Corporate governance, financial markets and global convergence (pp. 113-136). Boston: Kluwer Academic. Chusho Kigyocho. (Various issues). Kigyd Jittai Kihon Chosa [Small and Medium Enterprise Agency]. Tokyo: Chusho Kigyocho. Dai-Ichi Kangyo Ginko. (1992). Dai-Ichi kangyo ginko nijunenshi [The twenty-year history of the Dai-ichi Kangyo Bank], Tokyo: Dai-Ichi Kangyo Ginko. Dekle, Robert. (1998). The Japanese “big bang” financial reforms and market implications. Journal o f Asian Economics, 2 ,237-249. Diwa Shoken. (1963). Daiwa shaken 60 nenshi [The 60 year-history of Daiwa Security]. Tokyo: Daiwa Shoken. 308 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Dore, Ronald P. (1986). Flexible rigidities: Industrial policy and structural adjustment in the Japanese economy 1970-1980. Stanford: Stanford University Press. Dore, Ronald P. (2000). Stock market capitalism: Welfare capitalism: Japan and Germany versus the Anglo-Saxons. New York: Oxford University Press. Economic Planning Agency (Japan). (1984a, March). Annual report on national accounts. Tokyo: Government Planning Office. Economic Planning Agency (Japan). (1984b, various issues, monthly). Japanese economic indicators. Tokyo: Government Planning Office Enkyo, Soichi. (1993). K in’ yu jiyuka nyumon [An introduction to financial liberalization], Tokyo: Nihon Keizai Shimbunsha. Fried, Joel, and Peter Howitt. (1980). Credit rationing and implicit contract theory. Journal o f Money, Credit and Banking, 12(3), 471-487. Fuji Ginko. (1982). Fuji Ginko hyakuhenshi [The hundred-year history of Fuji Bank]. Tokyo: Fuji Ginko. Fujiwara, Soichiro. (2000). Kigyo saisei to M&A [Reconstruction of firms and M&A], Tokyo: Toyo Keizai Shinposha. Fukao, Mitsuhiro. (1998). Japanese financial instability and weakness in the corporate governance structure. Seoul Journal o f Economics, 11(4), 381-422. Fukao, Mitsuhiro. (1999). Koporeto gabanansu nyumon [Introduction to corporate governance], Tokyo: Chikuma Shobo. Fukao, Mitsuharu, and Yasuhiko Morita (Eds.). (1997). Kigyo gabanansu no kokusai hikaku [An international comparison of corporate governance], Tokyo: Nihon Keizai Shimbunsha. Fukuda, Atsuo, and Shin’ichi Flirota. (1996). Main bank relationship and capital structure in Japan. Journal o f the Japanese and International Economies, 10, 250-261. Gakushu Kenkyusha. (2003). Korenara wakaru koporeto gabanansu [Quick overview of corporate governance]. Tokyo: Gakushu Kenkyusha. Gerlach, Michael L. (1992). Alliance capitalism: The social organization o f Japanese business. Berkley, Los Angeles, Oxford: University of California Press. Gibson, Michael S. (1998, September). ‘ Big bang’ deregulation and Japanese corporate governance: A survey o f the issues. International Finance Discussion Papers, No. 624, Board of Governors of the Federal Reserve System. Gilson, Ronald J. (1991, October). The political ecology o f takeovers: Thoughts on harmonizing the European corporate governance environment. Stanford Law School Working Paper No. 84. 309 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Gilson, Ronald J., and Mark J. Roe. (1992). Understanding the Japanese keiretsu: Overlaps between corporate governance and industrial organization. John M. Olin Program in Law and Economics, Stanford Law School, Working Paper No. 97. Goto, Akira, Kazuo Serizawa, and Juro Teranishi. (1975, November). Shikin shijo to kigyo gurupu [The financial market and the enterprise groups], Keizai Hyoron, 32-43. Tokyo: Nihon Hyoronsha. Hakoda, Jun’ya, Toshio Hocchi, and Hayao Otani. (2000). Mochikabu-gaisha no jitsumu [The practice of running a holding company]. Tokyo: Toyo Keizai Shinposha. Hashimoto, Juro. (1995). Sengo no nihonkeizai [The postwar Japanese economy], Tokyo: Iwanami Shoten. Hashimoto, Juro, and Haruhito Taketada (Eds.). (1992). Nihonkeizai no hatten to kigyo shudan [Development of the Japanese economy and the enterprise groups]. Tokyo: Tokyo Daigaku Shuppankai. Higashitani, Satoshi. (1998). Gurdbaru sutandado no wana [The trap of the global standard]. Tokyo: Nikkan Kogyo Shimbunsha. Higashitani, Satoshi. (2000). Kin 'yucho ga chushokigyd wo tsubusu: Naze chushokigyo ni shikin ga mawaranaku nattanoka [The Financial Services Agency forces the small and medium sized enterprises to bankruptcy: Why has the fund become short for the small and medium sized enterprises?]. Tokyo: Soshisha. Hirota, Shin’ichi. (1990). Nihon ni okersu mein banku no hoken teikyo kino ni tsuite: Jissho bunseki [On the insurance provision mechanism of the Japanese main banks: Empirical research], Doshisha Daigaku Keizai Ronso, 41, 329-352. Hiwatari, Nobuhiro. (1991). Sengo nihon no shijo to seiji [The politics and market in postwar Japan]. Tokyo: Tokyo Daigaku Shuppankai. Holmstrom, Bengt. (1987). Comment. In Alan J. Auerbach (Ed.), Corporate takeovers: Causes and consequences (pp. 56-61). Chicago: The University of Chicago Press. Horiuchi, Akiyoshi. (1998). Kin’ yu shisutemu no mirai [The future of the Japanese financial system], Tokyo: Iwanami Shoten. Horiuchi, Akiyoshi, Frank Packer, and Shin’ichi Fukuda. (1988). What role has the “main bank” played in Japan? Journal o f the Japanese and International Economies, 2, 159-180. Horiuchi, Akiyoshi, and Quing-Yuan Sui. (1992). Mein bank kankei no keizai bunseki: Tenbo [A survey of the economic analysis of the main bank system], Tokyo Daibaku Keizaigakubu discussion paper, 92-J-l. 310 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Horiuchi, Toshihiro, and Eiji Murakami. (1991). Wagakuni ni okeru mein banku torihiki no jittai: Anketo kekka kara mita kigyo kin’yu no sugata [Questionnaire responses of main bank transactions and corporate financing]. In Toshihiro Horiuchi (Ed.), Jiyuka kokusaika jidai no kigyo kin’ yu no henbo: Mein banku no kino no hensei [Transformation of corporate financing in the era of liberalization and internationalization: The function and transformation of the Japanese main bank]. Tokyo: Kenky hokoku 75. Nihon Keizai Kenkyu Senta (Japan Center for Economic Research). Hoshi, Takeo. (1994a). Clearing up the balance sheets: Japanese experience in the post-war reconstruction period. Research Report 94-09, Graduate School of International Relations and Pacific Studies: University of San Diego. Hoshi, Takeo. (1994b). The economic role of corporate grouping and the main bank. In Masahiko Aoki and Ronald P. Dore (Eds.), The Japanese firm: Sources o f competitive strength (pp. 285-309). New York: Oxford University Press. Hoshi, Takeo. (1995a). Evolution of the main bank system in Japan. In Mitsuaki Okabe (Ed.), The structure o f the Japanese economy (pp. 287-322). New York: St. Martin’s Press. Hoshi, Takeo. (1995b). Financial deregulation and corporate financing in Japan. Working Paper, University of California, San Diego. Hoshi, Takeo, and Anil K. Kashyap. (2001). Corporate financing and governance in Japan: The Road to the Future. Cambridge, MA: The MIT Press. Hoshi, Takeo, Anil K. Kashyap, and Gary Loveman. (1993). Lessons from the Japanese main bank system fo r financial system reform in Poland. Research Report 93-01, Graduate School of International Relations and Pacific Studies, University of San Diego. Hoshi, Takeo, Anil K. Kashyap, and David Scharfstein. (1990). The role of banks in reducing the costs of financial distress in Japan. Journal o f the Japanese and International Economies, 2, 159-180. Hoshi, Takeo, Anil K. Kashyap, and David Sharfstein. (1991). Corporate structure, liquidity, and investment: Evidence from Japanese industrial groups. Quarterly Journal o f Economics, 106, 33-60. Hoshi, Takeo, Anil K. Kashyap, and David Scharfstein. (1993). The choice between public and private debt: An analysis o f post-deregulation corporate financing in Japan. Working Paper, University of California, San Diego. Hoshi, Takeo, and Hugh Patrick (Eds.). (2000). Crisis and change in the Japanese financial system. Boston: Kluwer Academic Publisher. Ide, Yasuo. (1999). Fudosan pureya no tameno yoi shokenka, warui shokenka: kakushite nihon-ryu fudosan shokenka ha hajimaru [Good and bad securitization for real estate players]. Tokyo: Forest. 311 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Ikeo, Kazuto. (2001, January). Sengo nihon no kin’yu shisutemu no keisei to tenkai, soshite rekka [Formation, expansion, and deterioration of the post-war Japanese financial system]. Finansharu Rebyu [Financial Review], 54, 155-177. Imai, Ken-ichi. (1976). Gendai sangyo soshiki [The contemporary Japanese industrial organization], Tokyo: Iwanami Shoten. Imai, Ken-ichi. (1982). Japan’s changing industrial structure and United States-Japan industrial relations. In Kozo Yamamura (Ed.), Policy and trade issues o f the Japanese economy: American and Japanese perspectives (pp. 47-75). Seattle: University of Washington Press. Imai, Ken-ichi. (1989). Kigyo gurupu [The Japanese enterprise groups]. In Ken-ichi Imai and Ryutaro Komiya (Eds.), Nihon no kigyo [The Japanese firms] (pp. 131-161). Tokyo: Tokyo Daigaku Shuppankai. Imai, Ken-ichi. (1992). Japan’s corporate networks. In Shumpei Kumon and Henry Rosovski (Eds.), The political economy o f Japan: Cultural and social dynamics (Vol. 1, pp. 198-230). Stanford, CA: Stanford University Press. Inoue, Hideaki. (1998, October). Kabushiki mochiai no saiki no ugoki ni tsuite [On the recent move of cross-shareholdings]. Retrieved on May 23, 2003 from Nissei Kiso Kenkyujo at http://www.nli-research.co.jp Inoue, Hideaki. (1999, October). Kaisho ga kasokusuru kabushiki mochiai [Cross-shareholdings on further dissolution]. Retrieved on May 23, 2003 from Nissei Kiso Kenkyujo at http://www.nli-research.co.jp Inoue, Hideaki. (2000, October). Kaisho tsuzuku ‘ kabushiki mochiai’ : Kabushiki mochiai jokyo chosa 99-nendo ban [Cross-shareholdings on the continuous dissolution: Research on the state of cross-shareholdings for fiscal 1999], Retrieved on May 28, 2003 from Nissei Kiso Kenkyujo at http://www.nli-research.co.jp Ito, Takatoshi. (1993). The Japanese economy. Cambridge, The MIT Press. Itochu Shoji Chosabu (Ed.). (1992). Zeminaru: nihon no sogo shosha [A seminar on the Japanese general trading company], Tokyo: Toyo Keizai Shinposha. Iwai, Katsuhito. (1994). ‘Hojin’ to nihon shihon shugi (“Corporations” and Japanese capitalism). In Katsuhito Iwai (Ed.), Shinon shugi wo kataru (Talking about capitalism), Chapter 3. Tokyo: Kodansha. Iwai, Katsuhito. (2003). Kaisha wa korekara donarunoka [What will become of companies in the future?]. Tokyo: Heibonsha. Iwata, Kikuo. (2000). Kin ’ yu hotei [The financial court], Tokyo: Nihon Keizai Shimbunsha. Japan, Office of the Prime Minister, Bureau of Statistics. (1983). Report on the Survey o f Research and Development, 1982. Tokyo: Nihon Tokei Kyokai. 312 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Jensen, Michael C. (1986). Agency costs of free cash flow, corporate finance, and takeovers. The American Economic Review, 76(2), 323-329. Jensen, Michael C., and William Meckling. (1976). Theory of the firm: Managerial behavior, agency costs, and capital structure. Journal o f Financial Economics, 3, 305-360. Johnson, Chalmers. (1982). M /77 and the Japanese miracle: The Growth o f Industrial Policy, 1925- 1975. Stanford, CA: Stanford University Press. Kaizuka, Keimei, Yoshikazu Slhimura, and Shoichi Royama (Eds.). (1981). K in’ yu shoken koza [Lectures on Finance and Securities] (Vol. 1). Tokyo: Toyo Keizai Shinposha. Kaneko, Masaru. (1999). Sehutmetto no seijikeizaigaku [Political economy of the safety net]. Tokyo: Chikuma Shobci. Kano, Tadashi. (1999). K in’ yu saihen [Reorganization of the Japanese financial circle]. Tokyo: Bungei Shunjusha. Kaplan, Steven N., and Bernadette Alcamo Minton. (1994). Appointments of outsiders to Japanese boards. Journal o f Financial Economics, 34, 225-58. Kato, Masaharu, and Yuichiro Matsuno. (1999). Kabusihi kokai no chishiki: Shinban [The public offering: A new edition]. Tokyo: Nihon Keizai Shimbunsha. Kato, Toshihiko. (1975). Dai 3 sho: Shihon shijo to kin’yu shijo: kasituke shijo wo chushin ni [Chapter 3: The capital market and the financial market with special attention on the lending market]. In Tokyo Daigaku Shakai Kagaku Kenkyu-jo, (Ed.), Sengo Kaikaku 8: Kaikakugo no nihon keizai [The postwar reform 8: The Japanese economy after the reform] (pp. 109-138). Tokyo: Tokyo Daigaku Shuppankai. Kawakita, Hideo. (2001, May 23). Kabushiki mochiai jokyo chosa [Research on the state of cross- shareholdings], Retrived on May 28, 2003 from http://www.fsa.go.jp Kawamura, Mikio, and Masayoshi Hayashikawa. (1999). Sogdshdsha biggu ban [The big bang of the general trading company]. Tokyo: Toyo Keizai Shinposha. Keizai Chosa Kyokai. (Various issues). Keiretsu no kenkyii. Tokyo: Keizai Chosa Kyokai. Kester, W. Carl. (1991a). Japanese takeovers: The global contest fo r corporate control. Boston: Harvard Business School Press. Kester, W. Carl. (1991b, July). Governance, contracting, and investment time horizons. Harvard Business School Working Paper. Kester, W. Carl. (1992). Industrial group as systems of contractual governance. Oxford Review o f Economic Policy, 5(3), 2,4-44. Keynes, John Maynard. (1936). The general theory o f money, interest and employment. London: McMillan. 313 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Kikkawa, Takeo. (1992). Sengogata kigyoshudan no keisei [The formation of the postwar enterprise groups]. In Juro Hashimoto and Haruhiko Takeda (Eds.), Nihonkeizai no hatten to kigyo shiidan (pp. 255-304). Tokyo: Tokyo Daigaku Shuppankai. Kikkawa, Takeo. (1996). Nihon no kigyo shiidan: Zaibatsu tono renzoku to danzetsu [The Japanese corporate groups: The continuity with, and discontinuity from, zaibatsu]. Tokyo: Yuhikaku. Kimura, Takeshi. (2003). Takenaka puran no subete [All about the Takenaka plan]. Tokyo: Aski Communications. Kishi, Nobuhito, and Gendai Kin’yu Mondai Ken’kyu Gurupu. (1995). Kyodai ginko no tanjo: Mitsubishi-Tokyo ginko gappei no shinso to nihon no ginko no ashita [The birth of a mega-bank: The scenes behind the merger between Mitsubishi and Tokyo Banks and the future of the Japanese banks]. Tokyo: President. Klein, Benjamin. (1988). Vertical integration as organizational ownership: The Fisher Body- General Motors relationships revisited. Journal o f Law, Economics, and Organization, 4{ 1), 198-212. Kobayashi, Yoshihiro. (1980). Kigyo shiidan no bunseki [Analyses of Japanese enterprise groups], Sapporo: Hokkaido Daigaku Shuppankai. Koike, Kazuo. (1992). Nihon kigyo no jinzai ikusei: Fukakujitsusei ni taisho suru tame no nouhau [Human resource development in Japanese firms: The know-how to cope with future uncertainty]. Tokyo: Chuo Koronsha. Kokumin Kin’yu Koko Sogo Kenkyujo. (1989). Posuto biggu ban no chushokigyo kin’ yu [The post-big bang small and medium-sized firms financing], Tokyo: Chushokigyo Risachi Senta. Komoto, Keisho. (2005, April). Chuken, chusho kigyo nimade shinto suru M&A no riyd: Nissei kenkyujo anketo 2005nen lgatsu chosa kekka [NLI January 2005 report: M&A getting prevalent among small-and-medium sized firms]. Retrieved June 21, 2005, from http://www.nli-research.co.jp Kosei Torihiki Iinkai Jimukyoku (Ed.). (1989). Kigyo shiidan no jittai ni tsuite [On the enterprise groups in Japan], Tokyo: Kosei Torihiki Iinkai. Kosei Torihiki Iinkai Jimukyoku (Ed.). (1992). Keizai kozo no henka to sangyo soshiki [The changes in the economic structure and the industrial organization]. Tokyo: Okurasho Insatsu-kyoku. Kosei Torihiki Iinkai Jimukyoku (Ed.). (1998a). Kigyo shiidan no jittai ni tsuite: Dai rokuji chosa hokokusho [The latest report on the big six enterprise groups in Japan: The sixth investigation by the Fair Trade Commission], Retrieved July 19, 2003, from http://www. Jftc.admix.go.jp Kosei Torihiki Iinkai Jimukyoku (Ed.). (1998b). Kigyo shiidan no jittai ni tsuite: Dai rokuji chosa hokokusho [The latest report on the big six enterprise groups in Japan: The sixth investigation by the Fair Trade Commission], Tokyo: Kosei Torihiki Iinkai Jimukyoku. 314 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Kosei Torihiki Iinkai Jimukyoku (Ed.). (2001a). Daikibo jigyo gaisha to gurupu keiei ni kansuru jittai chosa hokokusho [A report on the current state of the large firms’ running their groups]. Tokyo: Kosei Torihiki Iinkai Jimukyoku. Kosei Torihiki Iinkai Jimukyoku (Ed.). (2001b). Kigyo shiidan no jittai ni tuite: Dai nanaji chosa hokokusho [The latest report on the big six enterprise groups in Japan: The seventh investigation by the Fair Trade Commission]. Retrieved July 19, 2003, from http://www. Jftc.admix.go.jp Kreps, David. M. (1990). A course in microeconomic theory. New Jersey: Princeton University Press. Kumon, Shumpei, and Henry Rosovsky (Eds.). (1992). The political economy o f Japan: Cultural and social dynamics (Vol. 3). Stanford: Stanford University Press. Kuroki, Fumiaki. (2001). Kabushiki mochiai kaisho no jokyo no jittai: Kabushiki mochiai jokyo chosa 2000-nendo ban [The state of the dissolving cross-shareholdings: Research on the state of cross-shareholdings for fiscal 2000]. Retrieved May 28, 2003, from Nissei Kiso Kenkyujo at http://www.nli-research.co.jp Kuroki, Fumiaki. (2002, October). Keizoku suru mochiai kaisho: 2001-nendo kabushiki mochiai jokyo chosa [The dissolving cross-shareholding on continuation: Research on the state of cross-shareholdings for fiscal 2001]. Retrieved on May 28, 2003, from Nissei Kiso Kenkyujo at http://www.nli-research.co.jp Kuroki, Fumiaki. (2003, October). Mochiai kaisho ni miru kigyo to ginko no kankei: 2002-nendo kabushiki mochiai jokyo chosa [The relations between banks and firms under the dissolving cross-shareholdings: Research on the state of cross-shareholdings for fiscal 2002], Retrieved June 1, 2003, from Nissei Kiso Kenkyujo at http://www.nli-research.co.jp Kurosawa, Yoshitaka. (1999). “ Kakuzuke” no keizaigaku [Economics of rating], Tokyo: PHP Kenkyujo. Kyuno, Masao. (2000). Umare kawaru ginko, yaburesaru ginko [Winning banks, losing banks]. Tokyo: Nihon Jitsugyo Shuppansha. Lawrence, Henry. (1996). Regulatory competition and the politics of financial market reform in Britain and Japan. Governance: An International Journal o f Policy and Administration, 9(3), 311-341. Leathers, Charles, and J. Patrick Raines. (1994). Future of Japan’s main bank system in deregulated financial markets. Rivista Internazionale di Scienze Economiche e Commerciali, 47(9), 741-752. Levich, Richard M. (1985). A view from the international capital markets. In Ingo Walter (Ed.), Deregulating Wall Street: Commercial bank penetration o f the corporate securities market (pp. 255-292). New York: John Wiley & Sons. Mainichi Shimbunsha. (1970-2005, various issues). Shukan ekonomisuto [Weekly economist]. Tokyo: Mainichi Shimbunsha. 315 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Matsumura, Tom. (2000, August). Fudosan shijo no honkakuteki seibi ni mukete: Motomerareru fudosan shijo no joho infura [Toward building up the infrastructure for the real estate market in Japan]. Retrieved June 21, 2005, from http://www.nli-research.co.jp Matsumura, Tom. (2001, May). Fudosan toshi ga kasoku saseru toshiteki torihiki: Fudosan toshi shijo keisei heno kitai [The real estate securitization and the formation of real estate investment market in Japan]. Retrieved June 21, 2005, from http://www.nli-research.co.jp Matsumura, Tom. (2002, December). Kakudai suru shibo-gata fudosan fando shijo: Sono bijinesu modern to shijo tenbo [The expansion of private-fund real estate investment: Its business model and prospect]. Retrieved June 21, 2005, from Nissei Kiso Kenkyujo at http://www.nli-research.co.jp Matsumura, Tom. (2004a, February). Ririkushita J-REIT shijo, sono seicho wo tenbo [J-REIT taking off: A forecast on its future growth]. Retrieved June 21, 2005, from Nissei Kiso Kenkyujo at http://www.nli-research.co.jp Matsumura, Tom. (2004b, November). Fudoosan toshi bumu to toshi kyoiku [The boom on the real estate investment and the education on investment]. Retrieved June 21, 2005, from Nissei Kiso Kenkyujo at http://www.nli-research.co.jp McDonald, Jack. (1989, April). The Mochiai effect: Japanese corporate cross-holdings. Research Paper No. 1045. Stanford University Graduate School of Business, Stanford. Megacompany Research. (2003). Saishin nihon no gydkai chizu [The up-to-date industrial contour in Japan], Tokyo: Chukei Shuppan. Milgrom, Paul, and John Roberts. (1992). Economics, organization and management. New Jersey: Prince-Hall. Milgrom, Paul, and John Roberts. (1994). Complementarities and systems: Understanding Japanese economic organization. Estudios econaomicos, 1(3), 3-42. Milhaupt, Curtis J. (1999). Japan’ s experience with deposit insurancedfailing banks: Implications fo r financial regulatory design? The Columbia Law School Working Paper No. 148. Milhaupt, Curtis J. (2001a). Creative norm destruction: The evolution o f nonlegal rules in Japanese corporate governance. The Columbia Law School Working Paper No. 190. Milhaupt, Curtis J. (2001b). On the (fleeting) existence o f the main bank system and other Japanese economic institutions. The Columbia Law School Working Paper No. 194. Milhaupt, Curtis J. (2003). A lost decade fo r Japanese corporate governance reform?: What’ s changed, what hasn’ t, and why. The Columbia Law School Working Paper No. 234. Milhaupt, Curtis J., and Geoffrey Miller. (1997). Cooperation, conflict and convergence in Japanese finance: Evidence from the “ Jusen” problem. Working Paper No. 97-3-1, Washington University School of Law. 316 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Mihaupt, Curtis J., and Mark D. West. (2001). Institutional change and M & A in Japan: Diversity through deals. The Columbia Law School Working Paper No. 193. Minsky, Hyman P. (1975). John Maynard Keynes. New York: Columbia University Press. Mitsubishi Sogo Kenkyujo (Ed.). (2000). 21 seiki nihon no dezain [The grand design of 21st century Japan], Tokyo: Mitsubishi Sogo Kenkyujo. Miwa, Yoshiro (1990). Nihon no kigyo to sangyososhiki [Firms and industrial organization in Japan]. Tokyo: Tokyo Daigaku Shuppankai. Miwa, Yoshiro. (1996). Firms and industrial organization in Japan. New York: New York University Press. Miyajima, Hideaki. (1992). Zaibatsu kaitai [The dissolution of the Zaibatsu]. In Juro Hashimoto and Haruhito Taketada (Eds.), Nihon keizai no hatten to kigyo shudan [Development of the Japanese economy and the enterprise groups] (pp. 203-254). Tokyo: Hosei Daigaiku Sangyo Joho Senta. Miyajima, Hideaki. (1994). The transformation of Zaibatsu to postwar corporate groups—From hierarchically integrated groups to horizontally integrated groups. Journal o f the Japanese and International Economies, 8, 293-328. Miyamoto, Mitsuharu. (1997). Nihongata shisutemu no shinso: meisosuru kaikakuron [The depth of the Japanese system: Reformation astray]. Tokyo: Toyo Keizai Shinposha. Miyamoto, Mitsuharu. (1999). Nihon no koyd wo dou mamoruka: nihongata shokuno shisutemu no yukue [Securing employment in Japan: The future of the Japanese job accomplishing system], Tokyo: PHP Kenkyujo. Miyamoto, Mitsuharu. (2000). Henbo suru nihon shihonshugi: Shijo genri wo koete [Japanese capitalism under transformation: Beyond capitalism]. Tokyo: Chikuma Shobo. Miyashita, Kenichi, and David W. Russel. (1994). Keiretsu: Inside the hidden Japanese conglomerates. New York: McGraw-Hill. Miyazaki, Yoshikazu. (1980). Excessive competition and the formation of keiretsu. In Kazuo Sato (Ed.), Industry and business in Japan (pp. 53-73). New York: M. E. Sharpe, Inc. Miyazaki, Yoshikazu. (1992). Fukugo fukyo: posuto baburu no shohosen wo motomete [The complex depression: in pursuit for the prescription of the post-bubble Japanese economy]. Tokyo: Chuo Koronsha. Miyazaki, Yoshikazu. (1966). Sengo nihon no keizai kiko [The economic institutions of postwar Japan], Tokyo: Shin Hyoronsha. Mizoguchi, Yukinobu. (1998). Zenekon daishitsugyo jidai ga yattekita [The time of mass unemployment ushering in the general contracting business in Japan]. Tokyo: Apple. 317 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Modigliani, Franco, and Merton H. Miller. (1958). The cost of capital, corporation finance, and the theory of investment. American Economic Review, 48, 261-98. Muto, Yasuaki. (2003). Mochikabu-gaisha keiei no jissai [The practice of running a holding company]. Tokyo: Nihon Keizai Shimbunsha. Nakahigashi, Masahumi. (2000). Kigyo saihen hosei no hensen to kongo no kadai [The recent changes in the corporate reorganization laws and the remaining issues], Chukyo Daigaku: Chukyd Hogaku, 1&2 issues, 25-64. Nakajima, Shuzo. (1991). Kabushiki no mochiai to kigyo-hd [Mutual shareholding and the corporate law], Tokyo: Shadan H<5jin Shoji Homu Kenkyujo. Nakamura, Ikki. (1997). Kakuzuke kikan ga yosoku suru kieru ginko, shoken, hoken [Disappering banks, securities and insurance companies: The forcasts by rating comanies]. Tokyo: Appuru Shuppansha. Nakamura, Naoto. (2001). Kaisha bunkatsu no susumekata [The procedures of corporate breakup]. Tokyo: Nihon Keizai Shimbunsha. Nakamura, Shuichiro. (1992). 21 seiki gata chusho kigyo [The small and medium-sized firm of the 21st century]. Tokyo: Iwanami Shoten. Nakaso, Hiroshi. (1999, July 1). Recent banking sector reforms in Japan. Federal Reserve Bank o f New York, Economic Policy Review. Nakatani, Iwao. (1984). The economic role of financial corporate grouping. In Masahiko Aoki (Ed.), The economic analysis o f the Japanese firm (pp. 227-58). Amsterdam: North-Holland Elsevier. Nihon Kakuzuke Toshi Joho Senta. (1998). Kakuzuke no chishiki. Tokyo: Nihon Keizai Shimbunsha. Nihon Keizai Shimbunsha. (1995). Giko tota: Mitsubishi- Togin gappei no shdgeki [The selection of banks: The impact of the merger between Mitsubishi and Tokyo Banks]. Tokyo: Nihon Keizai Shimbunsha. Nihon Keizai Shimbunsha. (1997). Donaru kin’ yu biggu ban [What will happen after the big bang?]. Tokyo: Nihon Keizai Shimbunsha. Nihon Keizai Shimbunsha. (2000). K in’ yu meiso no 10 nen [The 10 years of the Japanese financial circus going astray], Tokyo: Nihon Keizai Shimbunsha. Nihon Keizai Shimbunsha. (2001). Kensho baburu: han’ i naki ayamachi [Looking back at the bubble economy: A fiasco with no intent]. Tokyo: Nihon Keizai Shimbunsha. Nihon Keizai Shimbunsha. (2004, spring). Nikkei kaisha joho: 2004 II haru go [Nikkei corporate information: 2004II spring issue]. Tokyo: Nihon Keizai Shimbunsha. 318 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Nihon Koshasai Kenkyujo. (1995). Wagakuni saiken kakuzuke no ayumi [The history of credit rating in Japan], Tokyo: Nihon Koshasai Kenkyujo. Nihon Shoken Shimbunsha. (1997). Shoken biggu ban [The Big Bang on the securities industry]. Tokyo: Nihon Shoken Shimbunsha Niida, Hiroshi, Tsuruhiko Nanbu, and Akira Goto. (1987). Nihon keizaino kozo henka to sangyo soshiki. (Structural change of the Japanese economy and the industrial organization.) Tokyo: Toyo Keizai Shinposha. Nissei Kiso Kenkyujo. (1997). Fudosan shoken-ka no kadai: Ippan toshika nimo miryoku no aru honkakuteki shohin jitsugen ni mukete [Programs of real estate securitization: Toward making attractive commodities for general investors]. Retrieved June 21, 2005, from http://www.nli-research.co.jp Nissei Kiso Kenkyujo. (1999). Wagakuni ni okeru fudosan shoken-ka shinten no kanosei: Kadai mo ooiga furontia wo mae ni chucho sezu susumubeki [Securitization of real estate in Japan: Many obstacles to overcome on its way]. Retrieved June 21, 2005, from http://www.nil- research.co.jp/syoho/econo/eco9808a.html Nissei Kiso Kenkyujo. (2003, September 18). Kabushiki mochiai jokyo chosa 2002-nendo ban [Research on the state of cross-shareholdings for fiscal year 2002], Retrieved October 24, 2003, from http://www.nli-research.co.jp Nogi, Shoichi. (2001). Sumitomo Ginko ni yori 110 nen no “noren” ha kakushite hikisakareta: Itoman massatsu no kiseki wo hotei shogen wo majiete kensho [Sumitomo Bank thus failed 110 years of Itoman’s history: A personal retrospect and the court testimonies]. Retrieved March 12, 2005, from http://www. rondan.co.jp/html/dokusho/itoman Noguchi, Yukio. (1992). Baburu no Keizaibaku: nihon keizai ni nani ga okottanoka [An economic analysis of the bubble economy: What happened to the Japanese economy?]. Tokyo: Nihon Keizai Shimbunsha. Noguchi, Yukio. (1995). 1940-nen taisei [The 1940 regime]. Tokyo: Toyo Keizai Shinposha. Noguchi, Yukio. (1999). Nihonkeizai saisei no senryaku: 21 seiki eno kaizu [Strategies for reviving the Japanese economy: The chart for the 21st century], Tokyo: Chuo Koronsha. Noguchi, Yukio. (2002). Nihon keizai, kigyo karano kakumei: Daisoshiki kara shososhiki e [The revolutionary shift in corporate organizations: From large to small organizations]. Tokyo: Nihon Keizai Shimbunsha. Ohara, Katsuma. (1997). Purojekuto fainansu [Project finance], Tokyo: Kin’yu Zaisei Jijo Kenkyukai. Ohara, Yukiko. (2000). Ginko kakumei, kachinokoruo wa dareka: mega banku v ..v . igyoshu ginko [The bank revolution: Mega banks vs. their competitors], Tokyo: Kodansha. Ohashi, Keizo, and H. Christopher Helm. (1995). Kabushiki daihyd sosho [The shareholders’ derivative law suit], Tokyo: Chuo Koronsha. 319 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Oka, Masanori. (2000, April). Fudosan toshi shuho no kakudai to shintenkai: Wagakuni demo kitai sareru kaisha-gata fudosan toshi [Expansion and turnaround of the real estate market in Japan: The company-style real estate investment]. Retrieved June 21, 2005, from http://www.nli-reserch.co.jp Oka, Masanori. (2001, September). Fudosan un’ yo bijinesu no makuake: Fudoosan toshin un’ yo- gaisha no shido to kadai [The dawn of the real estate business in Japan: The beginning and problems of the Japanese real estate funds]. Retrieved June 21, 2005, from http://www.nli- research.co.jp Okazaki, Tetsuji. (1992). Shihonjiuka igo no kigyo shudan [The enterprise groups after the capital liberalization]. In Juro Hashimoto and Haruhiko Taketada (Eds.), Nihonkeizai no hatten to kigyo shudan (pp. 305-335). Tokyo: Tokyo Daigaku Shuppankai. Okazaki, Tetsuji. (1993a). The Japanese firm under the wartime planned economy. Journal o f the Japanese and International Economies, 7, 175-203. Okazaki, Tetsuji. (1993b). Kigyo shisutemu [Enterprise system]. In Tetsuji Ozaki and Masahiro Okuno (Eds.), Gendai nihon keizai system no genryu [The origin of the contemporary Japanese economic system] (pp. 219-44). Tokyo: Nihon Keizai Shimbunsha. Okazaki, Tetsuji. (1996). Sengo shijokeizai ikoki no seifu kigyo-kan kankei—sangyo gorika seisaku to kigyo [The relationship between the government and the firms toward market economy in the postwar transition period—the rationalization policy and the firms]. In Hideshi Ito (Ed.), Nihon no kigyo shisutemu [The Japanese corporate system] (pp. 317-48). Tokyo: Tokyo Daigaku Shuppankai. Okazaki, Tetsuji. (1999). Mochikabu gaisha no rekishi [History of the Japanese holding companies], Tokyo: Chikuma Shobo. Okazaki, Tetsuji, and Okuno-Fujiwara Masahiro (Eds.). (1994). The Japanese economic system and its historical orgins. Oxford, New York: Oxford University Press. Okumura, Hiroshi. (1976). Nihon no rokudai kigyo shudan [6 gigantic business groups in Japan]. Tokyo: Daiyamondosha. Okumura, Hiroshi. (1992). Kaisha hon'ishugi ha kuzureruka [Will Kaisha-ism collapse?]. Tokyo: Iwanami Shoten. Okumura, Hiroshi. (1993). Japan’s corporate capitalism in peril. Japanese Economic Studies, 21(4), 3-23. Okumura, Hiroshi. (1996). The fate of corporate capitalism: History and overview. Japanese Economic Studies, 24(5), 71-94. Okumura, Hiroshi. (1997). Sokaiya sukyandaru: Nomura shoken jiken no kozu [Scandals of corporate racketeers: How the Nomura Securities case had occurred]. Tokyo: Iwanami Shoten. Okumura, Hiroshi. (1998a). Kabunushi sokai [Stockholder’s meetings]. Tokyo: Iwanami Shoten. 320 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Okumura, Hiroshi. (1998b). Musekinin shihonshugi [The irresponsible capitalism]. Tokyo: Toyo Keizai Shinposha. Okumura, Hiroshi. (1998c, July). Mitsubishi ha shayo no michi wo tadoru [The Mitsubishi Group is on a decline]. Zaikai Tenbo [The Business World], pp. 34-5. Tokyo: Zaikai Tenbo Shinsha. Okumura, Hiroshi. (2003). Kaisha wo dou kaeruka [How can we change corporations]. Tokyo: Chikuma Shobo. Okumura, Yoshihiro. (1999). Gendai nihon keizai-ron: “ Baburu keizai" no hassei to hokai [The modem economy: The emergence and collapse of the bubble economy]. Tokyo: Toyo Keizai Shinposha. Olson, Mancur. (1965). The logic o f collective action: Public goods and the theory o f collective actions. Cambridge: Harvard University Press. Omura, Keiichi, and Makoto Masuko. (2001). Wagakuni kigyo no keiei pafomansu to koporetogabanansu [The performance of Japanese firms and their corporate governance], Finansharu Rebyu [Financial Review], 54, 178-209. Osano, Hiroshi, and Yoshiro Tsutsui. (1985, June). Implicit contract in the Japanese bank loan market. Journal o f Financial and Quantitative Analysis, 20(2), 211-229. Oshio, Takashi. (1999). Shijo no koe [The voice of the market]. Tokyo: Chuo Koronsha. Ostrom, Douglas. (2000, April 7). The keiretsu system: Cracking or crumbling? JEI Report, 14. Ozawa, Zen’ya. (2002). Hitome de wakaru jika, genson kaikei [A quick guide to market-price and impairment accountings]. Tokyo: Toyo Keizai Shinposha. Packer, Frank. (1994). The role of long-term credit banks within the main bank system. In Masahiko Aoki and Hugh Patrick (Eds.), The Japanese main bank system: Its relevance for developing and transforming economies (pp. 142-87). New York: Oxford University Press. Pauly, Louis W. (1990). Opening financial markets: Banking politics on the Pacific Rim. Ithaca, NY: Cornell University Press. Polanyi, Karl. (2001). The great transformation: The political and economic origins o f our time. Boston: Beacon Press. Posen, Adam S. (1998). Restoring Japan’ s economic growth. Washington, DC: Institute for International Economics. Ramseyer, J. Mark. (1994). Explicit reasons for implicit contracts: The legal logic to the Japanese main bank system. In Masahiko Aoki and Hugh Patrick (Eds.), The Japanese main bank system: Its relevance fo r developing and transforming economies (pp. 231-57). New York: Oxford University Press. 321 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Ramsayer, Mark J. (2000). Rethinking administrative guidance. In Masahiko Aoki and Gary R. Saxonhouse (Eds.), Finance, governance, and competitiveness in Japan (pp. 199-211). New York: Oxford University Press. Roe, Mark J. (1992). Some differences in corporate governance in Germany, Japan, and America. The Center for Law and Economic Studies, Columbia University School of Law, Working Paper No. 86. Rosenbluth, Frances M. (1989). Financial politics in contemporary Japan. Ithaca: Cornell University Press. Royama, Shoichi. (2000). The big bang in the Japanese securities market. In Takeo Hoshi and Hugh Patrick (Eds.), Crisis and change in the Japanese financial system (pp. 253-72). Boston: Kluwer Academic Publisher. Royama, Shoichi. (2001, March). ‘Shijo-gata kansetsu kin’yu’ joron [A preface to market type indirect financing]. Finansharu Rebyu [Financial Review], 56, 1-10. Rubin, Edward. L. (1997). Communing with disaster: What we can leam from the Jusen and the savings and loan crises. Law and Policy in International Business, 79-92. Rybczinski, Tad M. (1984). Industrial finance systems in Europe, U.S. and Japan. Journal o f Economic Behavior and Organization, 5(3-4), 275-86. Sakakibara, Eisuke (Ed.). (1995). Nichi-bei-ou no keizai shakai shisutemu [The socio-economic systems of Japan, America, and Europe]. Tokyo: Toyo Keizai Shinposha. Sakamoto, Kazuichi, and Masahiro Shimotani (Eds). (1987). Gendai nihon no kigyo gurupu (The enterprise groups in contemporary Japan). Tokyo: Toyo Keizai Shinposha. Sanwa Ginko. (1974). Sanwa ginko no rekishi [The history of Sanwa Bank]. Osaka: Sanwa Ginko. Sataka, Makoto. (1994). Sengo kigyo jiken-shi [The history of postwar corporate scandals]. Tokyo: Kodansha. Sato, Kazuo. (1987). Saving and investment. In Kozo Yamamura and Yasukichi Yasuba (Eds.), The political economy o f Japan: The domestic transformation (Vol. 1) (pp. 137-85). Stanford: Stanford University Press. Saunders, Anthony. (1985b). Conflicts of interest: An economic view. In Ingo Walter (Ed.) Deregulating Wall Street: Commercial bank penetration o f the corporate securities market (pp. 207-230). New York: John Wiley & Sons. Saunders, Anthony, and Ingo Walter (Eds.). (1996). Universal banking: Financial system design reconsidered. Chicago: Irwin. Schumpeter, Joseph A. (1996). The theory o f economic development: An inquiry into profits, capital, credit, interest, and the business cycle. New Brunswick, NY: Transaction Publishers. 322 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Seki, Mitsuhiro. (1993). Furusetto gata sangyo kozo wo koete: Higashi ajia shin-jidai no naka no nihon sangyo [Beyond the full-set industrial structure: the Japanese industry in the new age of the East Asia]. Tokyo: Chuo Koronsha. Sheard, Paul. (1985, December). Main banks and structural adjustment in Japan. Research Paper No. 129, Australia-Japan Research Centre. Sheard, Paul. (1989a). Communication: Japanese general trading company as an aspect of interfirm risk-sharing. Journal o f the Japanese and International Economies, 3, 308-322. Sheard, Paul. (1989b). The main bank system and corporate monitoring and control in Japan. Journal o f Economic Behavior and Organization, 11, 399-422. Sheard, Paul. (1991a). The economics of interlocking shareholding in Japan. Ricerche Economiche, 45 ,421-48 Sheard, Paul. (1991b). Reciprocal Delegated Monitoring in the Japanese Main Bank System. Journal o f the Japanese and International Economies, 8, 1-21. Sheard, Paul. (1991c). The role of firm organization in the adjustment of a declining industry in Japan: The case of aluminium. Journal o f the Japanese and International Economies, 5(1), 14-40. Sheard, Paul. (1992a). Japanese corporate finance and behavior: Recent developments and the impact of deregulation. In Colin Mckenzie and Michael Stutchbury (Eds.), Japanese financial markets and the role o f the yen. Sydney: Allen & Unwin. Sheard, Paul. (1992b). The role o f the Japanese main bank when borrowing firms are in financial distress. Stanford University. CEPR Working Paper No. 330. Sheard, Paul. (1994a, March). Delegated monitoring among delegated monitors: Principal-Agent aspects of the Japanese main bank system. Journal o f the Japanese and International Economies, 18, 1-2. Sheard, Paul. (1994b). Interlocking shareholdings and corporate governance in Japan. In Masahiko Aoki and Ronald P. Dore (Eds.) The Japanese firm: Sources o f competitive strength (pp. 310-49). Oxford: Oxford University Press. Sheard, Paul. (1994c). Main banks and the governance of financial distress. In Masahiko Aoki and Hugh Patrick (Eds.), The Japanese main bank system (pp. 188-230). New York: Oxford University Press. Sheard, Paul. (1994d). Reciprocal delegated monitoring in the Japanese main bank system. Journal o f the Japanese and International Economies, 8, 1-21. Sheard, Paul. (1996). Keiretsu and market access: An economics of organization approach. In Paul Sheard (Ed.), Japanese firms, finance and markets (pp. 21-53). Melbourne: Addison- Wesley. 323 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Sheard, Paul. (2001). Kigyo mega saihen: Shin-nihongata shihonshugi no makuake [The mega- reorganiztion of firms: The dawn of the new Japanese capitalism]. Tokyo: Toyo Keizai Shinposha. Shibagaki, Kasuo. (1975). Dai 2 sho: Sangyo kozo no henkaku [Chapter 2: The change of the industrial organization]. In Tokyo Daigaku Shakai Kagaku Kenkyu-jo (Ed.), Sengo kaikaku 8: Kaikakugo no nihon keizai [The postwar reform 8: The Japanese economy after the reform] (pp. 63-107). Tokyo: Tokyo Daigaku Shuppankai. Shibagaki, Kazuo. (1981). Nihon keizai no kozo tenkan [Structural change of the Japanese economy]. In Keimei Kaizuka, Yoshikazu Shimura, and Shoichi Royama (Eds.), K in’ yu shoken koza [Lectures on finance and securities] (Vol. 1) (pp. 39-81). Tokyo: Toyo Keizai Shinposha. Shimokawa, Koichi. (1990). Nihon no kigyo hattenshi: Sengo fukko kara no 50 nen [The developmental history of Japanese enterprises: 50 years from the postwar reconstruction], Tokyo: Kodansha. Shimokawa, Koichi. (1992, spring). Keiretsu transactions: Japan and the United States compared. Japanese Economic Studies (3), 49-66. Shimotani, Masahiro. (1993). Nihon no keiretsu to kigyo gurupu [The keiretsu and enterprise groups in Japan]. Tokyo: Yuhikaku. Shimotani, Masahiro. (1996). Mochikabukaisha kaikin [Removing the ban on the holding company in Japan], Tokyo: Chuo Koronsha. Shimotani, Masahiro. (1997). The history and structure of business groups in Japan. In Takao Shiba and Masahiro Shimotani (Eds.), Beyond the firm: Business groups in international and historical perspective (pp. 5-30). New York: Oxford University Press. Shiroyama, Saburo. (1975). Nezumi: Suzuki Shoten yakiuchi jiken [The assault on Suzuki Corp.], Tokyo: Bungei Shunju. Shiroyama, Saburo. (1980). Kanryd-tachi no natsu [The summer of the bureaucrats]. Tokyo: Shinchosha. Shiroyama, Saburo. (1981). Yasei no hitobito: Okura Kihachiro kara Matsunaga Yasuzaemon made [The wild entrepreneurs: From Kihachiro Okura to Yasuzaemon Matsunaga]. Tokyo: Bungei Shunju. Shleifer, Andrei, and Lawrence Summers. (1988). Breach of trust in hostile takeovers. In Alan J. Auerbach (Ed.), Corporate takeovers: Causes and consequences (pp. 33-56). Chicago: University of Chicago Press,. Showa Dojinkai. (1960). Kigyokan kyoso to gijutsu [Interfirm competitions and technologies], Tokyo: Toyo Keizai Shinposha. Shukan Ekonomisuto [Weekly economist]. (1990s, various issues). Tokyo: Mainichi Shimbunsha. 324 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. SomuchS Tokeikyoku (Statistics Bureau of the Ministry of Public Management, Home Affairs, and Posts and Telecommunication). (2002). Nihon hyojun sangyo bunrui (Japan standard industrial classification). Tokyo: Somucho Tokeikyoku. Retrived on June 5, 2003, from http://www.stat.go.jp/index/seido/sangyo/pdf/san3f.pdf Sumitomo Shoji Kabushiki Gaisha. (1972). Sumitomo shoji kabushiki gaisha shashi [The history of Sumitomo Corp.]. Osaka: Sumitomo Shoji Kabusihiki Gaisha. Suzuki, Ken. (1993). Nihon no kigyo-shiidan: Sengo nihon no kigyo to ginko [The Japanese enterprise groups: The postwar Japanese enterprises and banks]. Tokyo: Otsuki Shoten. Suzuki, Ken. (1998). Mein banku to kigyo shudan: Sengo nihon no kigyokan sisutem [The main banks and the enterprise groups: the post-war inter-firm system]. Tokyo: Minelva Shobo. Suzuki, Sadahiko, and Richard W. Wright. (1985). Financial structure and bankruptcy risk in Japanese companies. Journal o f International Business Studies, 97-110. Suzuki, Yoshio. (1987). The Japanese financial system. New York: Oxford University Press. Suzuki, Yoshio. (1993). Nihon no kin’ yu seisaku [The monetary policy in Japan]. Tokyo: Iwanami Shoten. Suzuta, Atsuyuki. (1976). Shin-kigyd shudan kenkyu: Dai-ichi kangyo gurupu no subete [All about the DKB group]. Tokyo: Nihon Jitsugyo Shuppansha. Takarajimasha. (2003). Gyokai saihen to shin-keiretsu chizu [The reorganizations and new configuration of the Japanese industries], Tokyo: Takarajimasha. Takasugi, Ryo. (1990-1991). Shosetsu: Nihon Kogyo Ginko [The Industrial Bank of Japan] (5 vols.). Tokyo: Kodansha. Takeda, Yasushi. (1995). Nomura shoho monogatari: Osaka shonin-do to daiyamondo keiei [The history of Nomura Securities: Its Osakaite way of running business and philosophy of the diamond management], Tokyo: Chuo Koronsha. Takeuchi, Kazuo. (2005, April). J-REIT ni okeru fuddsan toshi rimawari no doko [The recent yield trend of the J-REIT]. Retrieved June 21, 2005, from http://www.nli-research.co.jp Tamura, Tatsuya. (2002). Koporeto gabanansu: Nihon kigyo saisei heno michi [Corporate governance: The road to recovery of the Japanese firms]. Tokyo: Chuo Koronsha. Tanabe, Toshinori. (2003). Daigyakuten nihon kin ’ yu [The Japanese finance striking back], Tokyo: Chuo Koronsha. Tanaka, Hiroshi. (2003). Jika kaikei fukyo [The depression due to the current-value accounting]. Tokyo: Shinchosha. Taniuchi, Mitsuru. (1997). Recent developments in Japan’s financial sector: Bad loan and financial deregulation. Journal o f Asian Economics, 8(2), 225-44. 325 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Tasaku, Tomoo. (2002). Jigyo saisei [The reconstruction of business]. Tokyo: Kadokawa. Teramoto, Yoshiya (Ed.). (1997). Nih.onk.igyo no koporeto gabanansu. Tokyo: Seisansei Shuppan. Teramoto, Yoshiya, Naoto Iwasaki, and Toru Takai. (1999). Role of inter-organizational Networks: The case of Japanese corporate groups. In Daniel Dirks, Jean-Frangois Huchet, and Thierry Ribault (Eds.), Japanese management in the low growth era: Between external shocks and internal evolution (pp. 37-58). Berlin, Heidelberg, New York: Springer-Verlag. Teranishi, Juro. (1982). Nihon no keizai hatten to kin ’ yu (The Japanese economic development and financial system). Tokyo, Iwanami Shoten. Teranishi, Juro. (1990, September). Financial systems and the industrialization of Japan: 1900-1970. Banco Nacional del Lavoro Quarterly Review, 174. Teranishi, Juro. (1993). Mein banku shisutemu [The main bank system]. In Tetsuji Okazaki and Masahiro Okuno (Eds.), Gendai nihon keizai shisutemu no genryu [The origin of the contemporary Japanese economic system] (pp. 61-95). Tokyo: Nihon Keizai Shimbunsha. Teranishi, Juro. (1994). Loan syndication in war-time Japan and the origins of the main bank system. In Masahiko Aoki and Hugh Patrick (Eds.) The Japanese main bank system: Its relevance fo r developing and transforming economies (pp. 51-88). New York: Oxford University Press. Tokyo Daigaku Shakaikagaku Kenkyujo (Ed.). (1999). Rokudai kigyo shudan yushi keiretsu no kabusihi mochiai: 1974, 1984, 1994nen no kigyobetsu deta [The cross-shareholdings of 6 giant enterprise groups in Japan: The corporate data of 1974, 1984, and 1994]. Tokyo: Tokyo Daigaku Shakaikagaku Kenkyujo Tokyo Daigaku Shakaikagaku Kenkyujo. (Ed.). (2000). Keiretsu no keiretsu yushi: 1974, 1984, 1994nen no kigyo betsu deta [The keiretsu financing of 6 giant enterprise groups in Japan: The corporate data of 1974, 1984, and 1994]. Tokyo: Tokyo Daigaku Shakaikagaku Kenkyujo. Toyo Keizai Shinposha. (1994a). Nihon no rokudai kigyoshudan no jittai [The states of the six giant enterprise groups in Japan], Tokyo: Toyo Keizai Shinposha. Toyo Keizai Shinposha (Ed.). (1994b). Saishin nihon no rokudai kigyo shudan no jittai: Kosei torihiki iinkai dai goji chosa hokokusho "kigyo shudan no jittai ni tsuite" [The latest report on the big six enterprise groups in Japan: The fifth investigation by the Fair Trade Commission], Tokyo: Toyo Keizai Shinposha. Toyo Keizai Shinposha. (1994c). Zenzukai: Nihon no jinmyaku to kigyo keiretsu [The personal connections and Keiretsu in Japan: A comprehensive illustration]. Tokyo: Toyo Keizai Shinposha. Toyo Keizai Shinposha. (Various issues). Kigyo keiretsu soran [A survey of Japanese keiretsu], Tokyo: Toyo Keizai Shinposha. 326 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Toyo Keizai Shinposha. (Various issues). K in’ yu bijinesu [The financial business], Tokyo: Toyo Keizai Shinposha. Toyo Keizai Shinposha. (Various issues). Nihon no kigyo gurupu [The enterprise groups in Japan]. Tokyo: Toyo Keizai Shinposha. Toyo Keizai Shinposha. (Various issues). Shukan toyo keizai [Weekly toyo keizai], Tokyo: Toyo Keizai Shinposha. Tsuda, Kazuo. (1993). Kyodai ginko no kozo [The structure of gigantic banks]. Tokyo: Kodansha. Tsuru, Kotaro. (1994). Nihonteki shijdshisutemu: Tsuyomi to yowami no kensho [The Japanese market economy system: Strengths and weaknesses]. Tokyo: Kodansha. Tsuru, Kotaro. (2000). Japanese corporate governance in transition. Seoul Journal o f Economics, 13(3), 253-278. Ubukata, Yukio. (1986). K in’ yu kogaisha: Kokusai zaimu-senryaku no chukaku [Financial subsidiary companies: The nucleus of the international finance stratagem]. Tokyo: Nihon Keizai Shimbunsha. Uchida, Katsutoshi. (1976). Dai 7 sho: Sogoshosha no saihensei to atarashii yakuwari [Chapter 7: The reorganization of general trading compakanies and their new roles]. In Mataji Miyamoto, Yoshio Togai, and Yasuo Mishima (Eds.), Sogoshosha no keieishi [History of the formation of general trading companies]. Tokyo: Toyo Keizai Shinposha. Ueda, Kazuo. (1994). Institutional and regulatory frameworks for the main bank system. In Masahiko Aoki and Hugh Patrick (Eds.), The Japanese main bank system: Its relevance for developing and transforming economies (pp. 89-108). New York: Oxford University Press. Uekusa, Masu. (1987). Industrial organization: The 1970s to the present. In Kozo Yamamura and Ysukichi Yasuba (Eds.), The political economy o f Japan: The domestic transformation (Vol. 1). Stanford: Stanford University Press. Vogel, Steven K. (1996). Freer markets, more rules: Regulatory reform in advanced industrial countries. Ithaca and London: Cornell University Press. Vogel, Steven K. (1997). Kisei taikoku nihon no jirenma [Originally titled, The transformation of the Japanese economy—The political battle over deregulation], translated by Okabe Yoko. Tokyo: Toyo Keizai Shinposha. Wada, Ben. (2003). Kigyo saisei fando: Furyo saiken bijinesu no kyo to jitu [The corporate reconstruction fund: The bad loan-related business unmasked]. Tokyo: Kobunsha. Wada, Ben. (2004). Jigyo saisei fando: Kigyo, jigyo wo baishu shi, saisei suru hitobito [The business reconstruction fund: The people who buy out and reconstruct firms and businesses]. Tokyo: Diamond. Walter, Ingo. (1985). Deregulating Wall Street: Commercial bank penetration o f the corporate securities market. New York: John Wiley & Sons. 327 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Watanabe, Susumu. (1999). Changing image of Japanese small entrepreneurs. In Daniel Dirks, Jcan-Frangois Huchet, and Thierry Ribault (Eds.), Japanese management in the low growth era: Between external shocks and internal evolution (pp. 207-27). Berlin, Heidelberg, New York: Springer-Verlag. Watanabe, Takashi. (2001). Furyo saiken wa naze kienai [Why do the bad loans never disappear?]. Tokyo: Nikkei BPsha. Williamson, Oliver E. (1985). The economic institutions o f capitalism: Firms, market, relational contracting. New York: The Free Press. Williamson, Oliver E. (1987a). Antitrust economics: Mergers, contracting, and strategic behavior. Oxford and New York: Basil Blackwell Ltd. Williamson, Oliver E. (1987b). Comment. In Alan J.Auerbach (Ed.), Corporate takeovers: Causes and consequences (pp. 61-67). Chicago: The University of Chicago Press. Williamson, Oliver E. (1988). Corporate finance and corporate governance. The Journal o f Finance, XLIII (3), 567-91. Yabushita, Shiro, and Bushimata Tomoo (Eds.). (2002). Chushdkigyo kin’ yu nyumon [Introduction to financing of small and medium-sized firms]. Tokyo: Toyo Keizai Shinposha. Yaginuma, Hisashi. (1993, spring). The keiretsu issue: A theoretical approach. Japanese Economic Studies (3), 2-48. Yamaguchi, Yoshiyuki. (1997). K in’ yu biggu ban no genso to genjitsu [The ideals and realities of the financial big bang]. Tokyo: Jiji Tsushinsha. Yamaguchi, Yoshiyuki. (2002). Dareno tameno kin’ yu saisei ka: Furyo saiken shori no hijdshiki [For whom is the financial reconstruction being advanced? The irrationality of the disposition of bad loans]. Tokyo: Chikuma Shobo. Yamaguchi, Yoshiyuki. (2004). Keizai saisei ha “ genba” kara hajimaru: shimin, kigyo, gydsei no atarasii kankei [The economic reconstruction from the workplaces: The new relationship among citizens, firms, and administration]. Tokyo: Chuo Koronsha. Yamaichi Shoken. (1958). Yamaichi Shoken-shi [The history of Yamaichi Security]. Tokyo: Yamaichi Shoken. Yamanaka, Toyokuni. (1989). Sogo shosha: Sono hatten to riron [General trading companies: Their development and theory], Tokyo: Bunshindo. Yanai, Noboru. (2002). Mega banku no gosan: Ginko fukkatsu wa kano ka [The mistakes by the mega-bank: Can they revive?]. Tokyo: Chuo Koronsha . Yoshida, Haruki. 2001. Ginko to kabu: Daihakyoku heno shinario [Banks and stocks: The scenario toward a fiasco], Tokyo: Toyo Keizai Shinposha. 328 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Yoshida, Kei’ichi, and Toshikazu Nagayama. (1999). Sangyd kozo tenkan to chusho kigyo: Kudoka jidai eno taio [Transformation of industrial structure and the small and medium-sized enterprises: How to cope with the industrial hollowing-out]. Kyoto: Minerva Shobo. Yoshikawa, Hiroshi. (1999). Tenkanki no nihon keizai [The Japanese economy in the period of transition]. Tokyo: Iwanami Shoten. Yoshimura, Norihisa, Yasuhiro Ueno, and Tadao Kagano. (1999). Extemalization of organization and the dual governance structure. In Daniel Dirks, Jean-Frangois Huchet, and Thierry Ribault (Eds.), Japanese management in the low growth era: Between external shocks and internal evolution (pp. 21-35). Berlin, Heidelberg, New York: Springer-Verlag. Yoshino, Michael Y., and Thomas B. Lifson. (1986). The invisible link: Japan’ s sogo shosha and the organization o f trade. Cambridge: The MIT Press. Yoshitomi, Masaru. (1991). Japanese direct investment in Europe: Motives, impact and policy implications. Vol. 2: Joint project on Japanese investment in Europe. Aldershot, U.K., Brookfield, VT: Avebury. Zaikai Tenbo [The Business World]. (1997-2004). Various issues. Tokyo: Zaikai Tenbo Shinsha. Zaimu Sogo Seisaku Kenkyujo (Ministry of Finance, Policy Research Institute) (n.d.), Hojin Kigyo Nenpd. Retrieved on May 28, 2004, from http://www.mof.go.jp/lc002.htm 329 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.

Asset Metadata

Creator Suzuki, Shinichi (author)

Core Title The Japanese main bank system: A transaction cost approach

Contributor Digitized by ProQuest (provenance)

Degree Doctor of Philosophy

Degree Program Political Economy and Public Policy

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag business administration, banking,Economics, Finance,OAI-PMH Harvest,political science, general

Language English

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c16-433506

Unique identifier UC11336387

Identifier 3233846.pdf (filename),usctheses-c16-433506 (legacy record id)

Legacy Identifier 3233846.pdf

Dmrecord 433506

Document Type Dissertation

Rights Suzuki, Shinichi

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA