RIETI Policy Symposium

Emerging Patterns of Corporate Governance among Japanese Firms - Converging to Any Specific Model? (Summary)

Information

Summary

On October 20, 2004, the Research Institute of Economy, Trade and Industry held an all-day policy symposium titled "Emerging Patterns of Corporate Governance Among Japanese Firms: Converging to Any Specific Model?" at Elizabeth Rose Hall at United Nations University in Tokyo. Participants presented the findings of their empirical studies on changes in Japanese corporate governance since the second half of the 1990s and the effect on corporate performance, and discussed a wide range of related issues.

Opening Remarks and Introduction

The symposium opened with remarks by RIETI President Masaru Yoshitomi . President Yoshitomi explained that the basic purpose of the symposium was to explore the relationship between the recent performance of Japanese business and the corporate governance reforms launched in the second half of the 1990s, and he outlined the following five basic questions to be addressed:

  1. What changes have occurred in the financing and cross-shareholding relationships between corporations and banks that have supported the "main bank system"? With the change in main-bank functions, can we discern a concomitant change in the keiretsu and other traditional relationships between business firms?
  2. With regard to bankruptcy and corporate rehabilitation why did the main banks' handling of corporate insolvency begin to break down in the 1990s? Is it possible to compare the effectiveness of the bailouts handled by main banks through the 1980s with resolution under the bankruptcy law that has been applied since the Civil Rehabilitation Law went into effect in 2000? How should we assess the Industrial Revitalization Corporation and the current Industrial Revitalization Law?
  3. Are domestic and foreign institutional investors replacing the main banks as the key instruments of external governance? What is the prevailing view of their exact role and their current exercise of influence via the mechanisms of exit, voice, etc? How is their role likely to evolve henceforth?
  4. Is the U.S. shift in weight from external governance to internal governance mechanisms, such as the insertion of outside directors-as demonstrated by CalPERS and other American institutional investors-likely to occur in Japan as well? What are the future prospects for the development of the "company with committees" system in Japan in view of the problems with U.S. corporate governance highlighted since the Enron scandal?
  5. How strong is the complementarity among the institutions and practices related to corporate governance? How should we assess the hybrid governance systems emerging now? (a) Do they represent a transitional phenomenon preceding the eventual convergence on a new model? (b) Do they represent a trend toward multiple equilibrium in corporate-governance models? (c) Or, are there in fact no "models" to begin with? What sort of theoretical framework should we use to make these judgments?

Next, Hideaki Miyajima (Faculty Fellow, RIETI, and Professor, School of Commerce, Waseda University) provided the following overview of the RIETI research project on which the papers being presented at the symposium were based, together with the topics of the scheduled symposium sessions.

  1. The performance of Japanese businesses has changed dramatically since the 1990s. Specifically, the return on assets (ROA) dropped precipitously between the early and late 1990s. At the same time, in the second half of the 1990s, we see a large increase in the standard deviation of corporations' consolidated ROA and the standard deviation of the difference between corporations' consolidated ROA and the industry average. That is to say, disparities among firms have become more pronounced, as have disparities among firms within the same industry.
  2. With this situation in mind, the research project made its first task an analysis of the way in which "Japanese style" governance, whose unique strengths were a focus of attention in Japan and abroad until the mid-1990s, worked to the detriment of Japanese business after that time, together with an assessment of the extent to which Japanese corporate governance has changed since then.
  3. Second, given that corporate governance of Japanese firms has diversified and grown more heterogeneous since the second half of the 1990s, how should this diversification be understood?
  4. Third, given the complementarity of the various elements of a corporate governance system-such as the structure of external governance, the structure of internal governance, and the relationship between employees and the board of directors-another major focus of concern is whether the diverse systems now emerging can coexist.
  5. From this basic perspective, empirical analyses were carried with respect to the four areas of (a) bank-firm relationships, (b) shareholding structure, (c) the board of directors, and (d) employment systems, in hopes of determining the following:
    1. What is happening now among Japanese companies?
    2. What do the changes signify?
    3. How do the changes interrelate, and what are the determinants that brought them about?
  6. The symposium was organized around four sessions, corresponding to the four abovementioned areas, for the presentation of research findings. The final session was to be devoted to summative discussion.

Session 1: Bank-Firm Relationships and Financial Distress

The session began with a presentation by Yasuhiro Arikawa (Faculty Fellow, RIETI, and Associate Professor of Finance, Waseda University) titled "The Changing Nature of Bank-Firm Relationships," which can be summarized as follows.

  1. An empirical analysis was conducted regarding recent changes in corporate governance by banks, which have long played a central role in Japanese corporate governance, and changes in the relationship between firms and banks.
  2. Among the sample of firms listed on the First Section of the Tokyo Stock Exchange that were used as the subject of this study, the level of bank borrowing as a percentage of total assets and total debt had actually risen, contrary to the popular perception that firms are weaning themselves from their dependence on banks.
  3. Generally speaking, dependence on bank financing was highest among firms with low growth expectations as measured by Tobin's Q ratio, an index that measures corporate performance in terms of growth. However, dependence on bank borrowing was also high among those businesses with very high growth expectations that were not yet recognized by the market.
  4. The ratio of main-bank borrowing to total assets rose from 4.48% in 1991 to 7.01% in 1999. In 1998 and 1999, in the wake of the financial crisis, a divide emerged between Japanese businesses increasing their dependence on main-bank financing and those reducing their dependence.
  5. Examining the relationship between the downgrading of Japanese bank ratings and corporations' excess profitability around the time of the banking crises, a close negative correlation was seen in firms in mature industries that had low bond ratings and a high rate of borrowing and main-bank dependence. During the banking crisis, the downgradings had the effect of encouraging the exit of firms with poor stockmarket performance.
  6. When the investment function for manufacturing industry, for which liquidity constraints and debt ratio are explanatory variables, was estimated on the basis of data for the decade of the 1990s, some suppression of investment was seen in firms with major growth opportunities but a high dependence on banks. Although a credit crunch existed for firms with an extremely heavy main-bank dependence, where our sample was concerned, it was not in evidence overall.
  7. In analyzing the correlation between bank-firm relationships and employment adjustment in companies requiring restructuring, it was found that firms with a heavy dependence on main-bank financing tended to lag behind in those restructuring efforts. This suggests that main banks have been following an "evergreen" lending policy, throwing good money after bad to prop up low-performing borrowers, thereby allowing them to delay restructuring measures.
  8. It is hoped that policies to restore the healthiness of Japan's banks will have the effect of putting an end to these evergreen policies.

Next, Xu Peng (Faculty Fellow, RIETI, and Professor, Faculty of Economics, Hosei University) gave a presentation titled "Corporate Governance in Financial Distress: The New Role of Bankruptcy," summarized below.

  1. There is a widespread perception that private workouts for financially distressed companies are more common in Japan than in the United States. In fact, in recent years the ratio of workouts has been lower in Japan: In the United States, by the end of the 1980s, there had been 51 cases of chapter 11 reorganization and 51 instances of workouts, while in Japan, between 1997 and 2003, there were 84 cases of legal bankruptcy and only 38 cases of private workouts. Underlying this trend is the fact that the pattern of corporate bailouts by main banks has been changing.
  2. Japanese banks are only committed to rescuing corporations when they have made them unsecured loans. In the case of the Sogo department store chain, the Shinsei Bank refused to go along with a private bailout plan that involved surrendering a huge amount of unsecured loans because the bank had a special agreement on cancellation rights with the government regarding bad loans. Thus, Sogo was forced into legal proceedings. Where restructuring of debt is concerned, banks will only forgive unsecured loans; they will not extend debt forgiveness for secured loans for fear of being sued by shareholders or being charged with breach of trust.
  3. However, notwithstanding the frequent characterization of the Japanese economic system as land capitalism, as of 1990 only 5% of Haseko Corporation's debt was secured. During the economic bubble of the 1980s, the banks extended loan after loan without adequate collateral, and the legacy of that practice is today's evergreen lending policy.
  4. In the United States, collateral takes many forms besides real estate, including inventory, receivables, and goodwill. Collateral is an important means of reducing banks' moral hazard.
  5. The number of legal bankruptcy cases in Japan has increased dramatically in recent years. From 1987 to 1996, there were only 10 cases; in the year 1997 alone, there were 6; and in 2002 there were 27.
  6. Under the old Corporate Reorganization Law, company executives strongly resisted bankruptcy, since custom demanded their resignation-unlike chapter 11 in the United States, which assumes DIP (debt in possession)-and this had the effect of delaying the initiation of bankruptcy proceedings. The Civil Rehabilitation Law that went into effect in 2000 made it possible for a company's executives to remain in their posts during bankruptcy and reorganization. In addition, the process has been simplified, and secured creditors need not participate in the rehabilitation plan.
  7. The duration of bankruptcy has also been drastically shortened since the enactment of the Civil Rehabilitation Law. When bankruptcy proceedings are initiated under the new Civil Rehabilitation Law, the average duration of bankruptcy is 0.6 years, much shorter than the 2.2 years under the old Corporate Reorganization Law. The average duration of bankruptcies under the Civil Rehabilitation Law and the revised Corporate Reorganization Law combined has been shortened to 0.7 years from 2.2 years. This is much shorter than bankruptcy duration in the United States, where the average has stood at 2.5 years since 1978, when the bankruptcy laws were overhauled and the present system instituted. In view of this, it is fair to say that Japan has extremely effective bankruptcy laws.
  8. One aspect of bankruptcy is removal of the insolvent corporation from the market; a system's effectiveness in this regard is extremely important. The resolution of insolvency can be regarded as the linchpin of corporate governance.
  9. The use of private workouts centered on main banks is analogous to over-the-counter cold remedies that only address the symptoms when they are still mild. Legal resolution is analogous to hospitalization.
  10. The current study compared the United States in the 1980s after the overhaul in U.S. bankruptcy law with Japan in the period since 1997 during which the number of legal bankruptcies has increased dramatically. To assess whether corporate governance in Japan is converging toward the U.S. model, it will be necessary to compare the two during the same time period.
  11. Although termed "the lost decade," the 10 years following the collapse of the bubble economy were extremely important in terms of hastening the exit of inefficient corporations. In the United States, financial innovations implemented during the 1980s enhanced the functioning of the country's financial institutions, an important factor behind Ripplewood Holdings' buyout of Japanese businesses. Financial innovation in Japan has just begun.

In response to these two papers Akiyoshi Horiuchi (Professor, Faculty of Policy Studies, Chuo University) offered the following comments, which were followed by discussion.

  1. Regarding the Arikawa-Miyajima paper:
    1. The diversity seen in Japan's corporate financing did not begin in the 1990s but has existed, albeit to a lesser degree, since the 1980s or earlier. The 1990s is only when this diversity became conspicuous.
    2. Corporate governance of the banks themselves is a major policy issue. This problem, which was inadequately addressed in the period of rapid economic growth and therefore emerged as a serious issue during the bubble economy, should be mentioned in the introduction.
    3. The diversity seen suggests that the relationship with the main bank is not inalterable but is a variable chosen by firms and banks. This point should be made explicit prior to analysis.
  2. Regarding the Xu paper:
    1. It is an exaggeration to call the resolution of insolvency the linchpin of corporate governance.
    2. It is necessary to examine whether analysis of management information regarding a corporation as a going concern (assessment of corporate value) has been handled properly during the public process of resolving insolvency cases since the adoption of the Civil Rehabilitation Law, and whether parties with information-analysis and communication capabilities on a par with those formerly possessed by the main banks have been able to participate effectively in the resolution process and perform those functions. These are questions that need to be answered in future studies.
    3. Since the 1990s, the functioning of Japanese banks has declined, leaving us without a smooth mechanism for private workouts of insolvency issues. This created an urgent need for a revision of bankruptcy law. An issue we need to look at henceforth is how to create a new mechanism for effective management information analysis in the resolution of insolvency cases.

Session 2: Changing Ownership Structure: Whose Corporation Is It?

The session began with a presentation by Hideaki Miyajima (Faculty Fellow, RIETI, and Professor, School of Commerce, Waseda University) titled "Dissolving Cross-shareholding: Why, How and Implications" which can be summarized as follows.

  1. A major characteristic of Japan's publicly traded companies is the predominance of "stable shareholding," achieved by means of extensive cross-shareholding between financial institutions and business corporations and among different business corporations. However, the ownership structure has changed dramatically since 1997, with a precipitous decline in cross-shareholding, particularly between businesses and banks, and an increase in the ratio of stock owned by non-Japanese investors. Furthermore, the increase in foreign investment and the disappearance of cross-shareholding have not advanced at a consistent pace throughout Japanese industry; instead, a divide has emerged between businesses that are changing rapidly and those that are not. What are the factors determining the dissolution of cross-shareholding?
  2. The price of bank shares was stable until 1995, and expected returns were consistent with the TOPIX index. However, after 1995, the fluctuation risk rose, and the rates of return fell far below the stockmarket average.
  3. In this environment, investors began unloading bank stock. A logit model regression analysis was performed on a sample of corporations listed on the First Section of the Tokyo Stock Exchange with the explained variable defined as whether or not a firm had sold off bank stock during the period from 1995 to 2002. It was found that corporations that had no need to worry about bond ratings or other market pressures, firms with low total market value that were at risk for a takeover, and companies with a high level of debt all tended to hold onto their bank stock.
  4. With respect to city banks' disposal of business firms' stock, before the banking crisis, the tendency was for banks to sell off high-risk corporate stock, whereas after the crisis (from 1999 on), the tendency was to sell off the easily disposable stock of high-performance business firms. The overall tendency was to hold onto the stock of firms with close connections to the bank.
  5. The effect of the shareholding structure on corporate performance (ROA, etc.) was analyzed on the basis of data from all publicly traded companies in Japan during the period since 1995. A very strong negative correlation was found between performance and the percentage of stock held by stable shareholders.
  6. With regard to shareholding structure, a divide is emerging between firms with a high ratio of stock held by foreign investors and a low ratio held by stable shareholders on the one hand, and firms with a low ratio of stock held by foreign investors and a high ratio in the hands of stable shareholders on the other hand. The former group has boosted performance through reform of corporate governance.

Next, Christina Ahmadjian (Professor, Graduate School of International Corporate Strategy, Hitotsubashi University) presented a paper titled "Foreign Investors and Corporate Governance," which covered the following points.

  1. The percentage of shares held by foreign investors has risen dramatically since the 1990s. However, during this time there has been an increase in investment by Western institutional investors-especially American and British pension funds-in overseas stocks worldwide, not just in Japan, for the purpose of international diversification. What sort of people are foreign investors? By what means do foreign investors exert influence?
  2. Some foreign investors are involved in direct investment, but the majority is institutional investors focused on portfolio investment.
  3. Foreign institutional investors tend to hold shares in well-known, high-performance large Japanese firms whose shares of exports in total earning are high.
  4. Most foreign investors are the custodians for various funds and bear the fiduciary duties. They are investing in Japan solely for the earnings. For this reason they are apt to insist on American-style corporate governance and push vigorously for corporate restructuring.
  5. The two ways for foreign investors to exert influence are exit and voice. Generally speaking, foreign investors engage in more active buying and selling of stock and thus exert more influence on prices than Japanese investors. This means they are able to use the threat of exit to influence Japanese firms. They exercise voice primarily on an informal level, often by conveying their views to the CEO and other executives in the meetings set for this purpose.
  6. An analysis was carried out using the JCG index, by which the corporate governance of a number of companies on the First Section of the Tokyo Stock Exchange has been rated on the basis of information submitted in a questionnaire survey. It was found that, as a group, those companies with a high JCG index had a higher proportion of stock owned by foreign shareholders than those with a lower JCG index. In addition, there was a strong correlation between foreign shareholding and a company's progress toward restructuring in the areas of employment and assets.

Hiroshi Osano (Professor, Institute of Economic Research, Kyoto University) responded with the following comments, which were followed by discussion.

  1. Regarding the Ahmadjian paper: Foreign investors use a wide variety of instruments, with hedge funds occupying an important position in their trading activities. There is also considerable diversity in their exercise of voice, including formal activism by large mutual funds and others. Funds that use indexing cannot exert influence by means of exit and so are inclined to rely on voice. It would be interesting to do a statistical study of this.
  2. Regarding the Miyajima-Kuroki paper: The risk of takeover has been higher in the last couple of years than during 1995-2002, the period chosen for analysis. There is a possibility that those Japanese companies that have restructured and put themselves on a firm financial footing will move actively to acquire other companies. In addition, foreign businesses are more interested in taking over Japanese companies now that the earlier financial concerns have receded. Analysis of cross-shareholding and takeovers will be of great significance from now on. Could hostile takeovers emerge as an important element of corporate governance in Japan? How is corporate governance likely to be affected by the threat of hostile takeovers and the use of poison pills in response? This area offers some interesting topics for future research.

Session 3: Corporate Board Reform

The session began with a presentation by Zenichi Shishido (Professor, Law School, Seikei University) titled "The Turnaround of 1997: Changes of Japanese Corporate Governance and Corporate Law," which can be summarized as follows.

  1. The traditional J (Japan) model is characterized by governance by the "company community," which contributes human capital. During the "lost decade," the traditional J model became dysfunctional.
  2. The year 1997 was a major turning point for the reform of various institutions relating to Japanese corporate law. The changes took two forms: demand-pull reforms initiated by the needs of the business sector, and policy-push reforms imposed by lawmakers.
  3. The demand-pull reforms were as follows: lifting of the ban on repurchasing of a company's stock by the company (1997), introduction of stock options (1997), simplification of merger procedure (beginning 1997), lifting of the ban on holding companies (1997), introduction of share-for-share exchanges (1999), establishment of a corporate divisions (2000), and adoption of limits on director liability in shareholder lawsuits (2001).
  4. The policy-push reforms were reform of the accounting system (beginning 1997), the mandatory inclusion of outside statutory auditors on boards of audit (2001), and the optional adoption of the "company with committees" system (2002).
  5. Corporate governance in Japan is converging with that of the United States on the formal level, i.e., in the sense that Japanese corporations can now opt for a system that is no different from the American system. However, because incentive patterns differ in Japan and the United States, structures of internal control are diverging on the functional level.
  6. There are three basic incentive patterns: "balancing," in which stakeholders exert influence through management; "monitoring," in which management acts as the agent of the shareholders; and "bargaining," in which the providers of monetary capital (shareholders, etc.) bargain with the providers of human capital (management and employees). The J model represents the third and last of these.
  7. The coexistence of a number of different internal governance structures may reflect the importance of relation specific investment for each industry.
  8. In the new J model, contingent governance based on the main-bank system has been abandoned, and the board has emerged as the system for bargaining between the inside directors, who represent the providers of human capital, and the outside directors, who represent the providers of monetary capital.

Next, Hideaki Miyajima (Faculty Fellow, RIETI, and Professor, School of Commerce, Waseda University) delivered a presentation entitled "The Changing Corporate Boardroom: Causes and Results," which can be summarized as follows.

  1. The purpose of the study was to analyze changes in the internal governance structure as represented by the board of directors and the impact of these changes on corporate performance.
  2. A Corporate Governance Score (CGS) gauging companies' commitment to corporate governance reform, an index for assessing policies and practices in 26 categories, was developed on the basis of information from a questionnaire survey of listed firms carried out by the Ministry of Finance in November 2002. In addition, two sub-indices were created: the CGSsh, which relates to the protection of shareholder rights; the CGSbr, relating to board reforms; and the CGSds, which rates disclosure of information.
  3. With regard to correlation between CGS and corporate performance, a significant relationship was found between such measures of performance as Tobin's Q ratio and ROA and both the CGS and CGSds. There are two possible explanations for the correlation between performance and disclosure of information: (1) disclosure lowers the cost to investors of collecting information, thereby lowers the cost of funds, and thus contributes directly to higher profits; (2) management's commitment to disclosure means that unfavorable information cannot be hidden; as a result, management works harder to prevent unfavorable events that would lead to the creation of unfavorable information that would have to be disclosed, and this leads to better performance.
  4. No significant relationship was found between the CGSbr index and corporate performance. This may be because the separation of management and implementation is on paper only, or it may be because corporate restructuring has not advanced to the point where the outside directors can perform their function.
  5. Firms with a high percentage of stock held by stable shareholders and a relatively heavy dependence on main-bank financing tended to have a low CGS. Firms with a high percentage of stock held by foreigners, a relatively heavy dependence on financing from corporate bonds, and bank relationships based on contracts tended to have a high CGS.
  6. Regarding the relationship between CGS and worker participation in strategic decision making, when companies under intense pressure from capital markets (companies with a bond rating of BBB or lower) and those not under such pressure were considered separately, a significant correlation was found in the former group.
  7. In recent years, it has been possible to classify Japan's large corporations into three different types on the basis of their employment and wage system. Type I consists of companies with a long-term employment system and seniority-based pay; Type II comprises companies that combine merit-based pay and a long-term employment system; and Type III companies are those that use merit-based pay and have no long-term employment system. An examination of the correlation between these types and CGS reveals a negative correlation for Type I and a significant positive correlation for Type II. A positive correlation can also be seen in Type III. Overall, companies that are shifting from seniority-based to merit-based pay, even while preserving long-term employment, are actively pursuing reform of corporate governance, including information disclosure.

Hiroyuki Yanai (Executive Director, Japan Association of Corporate Directors) responded to these presentations with the following comments, which were followed by discussion.

  1. The Japan Association of Corporate Directors will be three years old this December. One thing noticed in frequent discussions with members is the existence of fundamentally antithetical views of the market and the joint-stock corporation.
  2. With regard to market discipline, which markets should have priority-the capital markets or the consumer market for goods and services? For the capital markets, the ideal is a correspondence between executive gain and shareholder oversight, while for the consumer market, the ideal is a correspondence between executive gain and customer satisfaction as a result of market competition. These days, the impression is that leadership is shifting to the investor community, what with the growing presence of foreign investors, active pension funds, and shareholder ombudsmen.
  3. Each country has its own characteristic view of the company relative to the axes of Anglo-Saxon versus Latin, U.S. and U.K. versus continental Europe, and common law versus statute law. In Japan today, there are two basic views of the company: the "shareholder sovereignty" concept, an extension of primitive capitalism, and the "company as public institution" idea, an extension of the idea of employee sovereignty. The "public institution" theory holds that the company is a community built on a social contract, and that its executives are no more than stewards temporarily assigned to watch over things until the next caretakers relieve them.
  4. These antithetical concepts of market discipline and the company give rise to other opposing concepts. The shareholder is seen variously as the basis of governance or as just one important stakeholder. Executives are seen as either agents of the shareholders or holders of executive power. Some say executive power is legitimized by the executive's selection at a general meeting of shareholders, while others say it is legitimized by subsequent performance assessment through a fair and transparent system of governance.
  5. Regarding the role of the shareholders meeting, one issue is whether it should choose the directors or the executive. The prevailing notion is that in "companies with committees," it should choose the directors.
  6. Regarding the separation of management and director, there is the question of whether complete separation or partial separation is the goal. With complete separation, the image is that of a board of directors with not even one operating officer. In fact, one corporation tried this in the United States, but it was unable to function and the CEO was quickly reinstated on the board of directors.
  7. It seems that our image of a smart executive is one who-to use Professor Shishido's classification-achieves the concept of balancing in the guise of the monitoring board. In other words, the power of the executive in Japan is still very strong. When they are out of control, they cannot be stopped until the Industrial Revitalization Corporation takes over.
  8. To prevent executives from abusing their authority, it is not sufficient to implement such policies as making the majority of directors on the board independent directors, separating the posts of CEO and board chair, or disclosing individual executive salaries. A mechanism for assessing executives' business ethics is also needed.
  9. When determining the independence of an outside director, the focus tends to be on formal requirements, such as financial and familial relationships. But substantive factors need to be examined at the same time if not first. The most important is whether the person has experience as a CEO. If the current CEO cannot be appointed, then a former CEO is best.
  10. Rights, responsibilities, and compensation should coincide exactly in an executive. Half of the executives in Japan say that executives should not be highly paid. Only 10% say that they should be paid on a par with those of Fortune 500 companies. However, this kind of thinking may change naturally in Japan if the mobility of company executives continues to increase.

Session 4: Corporate Governance and Employment System

The session began with a presentation by Masahiro Abe (Faculty Fellow, RIETI, and Assistant Professor of Economics, Dokkyo University) titled "Corporate Finance and Human Resource Management," which covered the following points.

  1. Japan's employment system is undergoing major changes. Today, non-regular workers, and atypical workers account for almost 40% of the workforce, up 10 points from a decade ago. The pace of employment adjustment has also picked up since the beginning of the 1990s. Employee education and training is changing as well, with the emphasis shifting to optional self-development programs and training tailored to specific groups of employees.
  2. A theoretical proof was carried out on the basis of Tirole model, which focuses on the "agency problem" between the providers of monetary capital and managers. The four possible combinations of financial system and human-resource-management system were examined: (1) bank financing and in-house training, (2) stock-market financing and in-house training, (2) bank financing and self-training, (3) stock-market financing and self-training. With this model, it should be possible to demonstrate complementarity between bank financing and in-house training on the one hand, and between stock-market financing and self-training on the other.
  3. Using 1995 and 2001 results from the Human Resource Management Systems Survey, an annual questionnaire survey of large and medium-sized companies carried out by the Institute of Labor Administration, an analysis was carried out on 58 firms to determine the relationship between the adoption or non-adoption of 15 human-resource-management practices and patterns of financing as seen in such figures as the ratio of shares controlled by financial institutions, the ratio of shares held by foreign investors, and so forth.
    1. Companies that use a particular human-resource-management practice and those that do not were compared to determine whether there was a difference in their average ratio of bank ownership of stock and foreign ownership of stock. With regard to the practice of giving length-of-service awards, it was determined that corporations with a low foreign ownership were less likely to embrace the practice. With regard to annual salaries and domestic education programs, companies with high foreign ownership were more likely to have instituted these systems.
    2. A probit model regression analysis was carried out with financial structure (such as ratio of stock held by banks and ratio held by foreign investors) as the independent variables and the presence or absence of various human-resource-management practices as the explained variables. According to this analysis, firms with a high foreign ownership were less likely to adopt the practice of giving length-of-service awards, more likely to have domestic and overseas education programs, and more likely to have women in management positions.
  4. In may be concluded that in firms with a high foreign ownership, the human-resource-management system tends to depart from traditional Japanese practices.
  5. The following problems with this analysis should be noted with this analysis:
    1. Sample was small
    2. Analysis only dealt with the presence or absence of human-resource-management practices and did not examine their content. The interpretation of a practice may vary by company
    3. A time lag could exist between a change in financing and the impact on human-resource-management.

Next, Gregory Jackson (Visiting Fellow, RIETI, and Senior Lecturer, King's College London) gave a presentation on "Employee Participation, Adjustment and Distributional Conflict," summarized below.

  1. Japanese-style employment systems have been characterized in terms of long-term or lifetime employment, seniority-based pay schemes, and firm-specific training. Supporting factors include enterprise-based unions, labor laws with strong barriers to dismissal, and the absence of a strong welfare state like that of Germany. Complementarities have been posited between human-resource-management and corporate governance; for example, the main-bank system and stable shareholding protect investment in firm-specific skills.
  2. This is shifting toward a shareholder-value model that calls for stronger corporate competencies, equity-oriented performance targets, performance-oriented pay, disclosure, and market-oriented accounting.
  3. The current human-resource-management situation in Japan: While the lifetime employment system is basically stable, the system of seniority-based pay is undergoing modifications, and merit-based pay is being introduced.
  4. According to a Ministry of Finance survey, three types of employment system currently exist in Japan. Type 1 is the traditional model that combines lifetime employment with seniority-based pay (54%); type 2 combines lifetime employment with merit-based pay (29%); and type 3 has merit-based pay and no long-term employment (17%). Among firms with a high Corporate Governance Index, the percentage of type1 firms is dropping and the portion of type 2 firms rising commensurately. The percentage of type 3 firms is relatively constant.
  5. Regarding employment adjustment:
    1. 81% of companies expressed commitment to lifetime employment, but the "core employees" is shrinking.
    2. Between 2000 and 2003, 36% of companies carried out employment adjustment, cutting their workforce by 15% on average.
    3. The method of employment adjustment was overwhelmingly "benevolent": 54% of exits were by early retirement, 29% by hiring freeze, 5% by transfer, another 5% by spin-off, and only 4% through layoffs.
    Because employment adjustment was carried out throughout, analysis of data between 2000 and 2003 does not indicate a correlation between foreign ownership and probability of layoffs.

  6. In regard to pay systems, there is a negative correlation between the ratio of foreign ownership of stock and seniority-based pay, and a positive correlation between the percentage of executives promoted from the company ranks and seniority-based pay. However, unlike in Germany, no correlation can be found between the stock ownership structure or the board system and the adoption of performance-related pay (PRP).
  7. With regard to complementarity, no simple causal model can be found between corporate governance and employment system. It may be that Japanese-style employment is compatible with a fairly wide range of CG styles.

Mitsuharu Miyamoto (Professor, Graduate School of Economics, Senshu University) responded to the above presentations with the comments summarized below, which were followed by discussion.

  1. Abe-Hoshi paper:
    1. So far corporate governance has mainly been discussed from the standpoint of financing. The assumed causal relationship is "financing affects governance, which affects performance." However, while the impact of financing on governance has been determined, the impact of governance on performance is less clear.
    2. This is because performance depends in part on human-resource-management. Viewed in financial terms, the indicators of corporate performance are things like Tobin's Q and profit margin. Viewed in terms of human resources, the indicators are things like competitiveness and productivity. It is important to look at the causal relationship in terms of "financing affects governance, which affects human-resource-management, which affects performance."
  2. Jackson paper:
    1. In the past, it was assumed that there was complementarity between long-term employment and seniority-based pay, but it has been pointed out that nowadays companies that combine long-term employment and merit pay in some form are in the majority. There are indications that many of the firms that are attempting to maintain long-term employment are also committed to adopting performance-related pay. But can the two really coexist? There are indications that long-term employment can only be maintained in a system that combines core regular employees with long-term employment guarantees and peripheral part-time or temporary employees without such guarantees.
    2. There is a perception gap with regard to the likelihood that companies will be able to maintain long-term employment in the future: employees in their 20s and 30s regard it as unlikely, while executives regard it as likely.
    3. In reality, changes in compensation and employment are gradual and moderate. Concerning the introduction of PRP into the domestic labor market, it is questionable whether companies are implementing it in a form that differentiates enough to have an impact on the labor supply.
    4. The idea that product architecture and organizational architecture, which are based on corporate strategy, have an impact on corporate governance is important. Another possible path of causation is "corporate strategy affects the product/organizational architecture, which affects human-resource-management."
    5. From this, it may be possible to conclude that companies that boast an integrated type of organizational architecture based on corporate strategy and that depend on bonds market for financing adopt long-term employment plus PRP as their employment-compensation model. This represents a new J-type governance.

Session 5: Conclusions

Gregory Jackson (Visiting Fellow, RIETI, and Senior Lecturer, King's College London) gave the following presentation, titled "Corporate Governance in Japan: Institutional Change and Organizational Diversity," as a summation of the symposium.

  1. The Japanese-style firm, or J-firm, is characterized by the main bank relationship, stable shareholding, lifetime employment, and management by insiders who rose up through the ranks. It has been described as a community committed to long-term organization building.
  2. There are several theories regarding recent changes in the Japanese firm, including (1) convergence theories, which stress change toward a single "best" model, and (2) path-dependence theories, which argue that change is a bounded process.
  3. A number of forces led to change in the 1990s. The first was internationalization, which brought about major changes in finance, foreign direct investment, and accounting rules. However, the number of Japanese firms exposed to foreign investors, listing requirements of overseas stock exchanges, international bond ratings, and so forth, remains small. The second force was liberalization. In the 1980s, financial deregulation eroded the monitoring capacity of the main banks. The third force was a shift in organizational life cycles and architecture, relating to the promotion of new industries and the restructuring of older ones, as well as changes in knowledge and information and innovation systems.
  4. Major changes in corporate governance in Japan in recent years may be summed up as follows.
    1. With regard to ownership and financing, main-bank relationships have receded somewhat and stable shareholding has diminished, but neither has disappeared. The growing presence of foreign investors is strongly associated with recent changes in corporate governance, but their influence is limited at present.
    2. In response to financial distress, corporate restructuring is proceeding apace, particularly in respect to companies' business portfolios. With the overhaul of bankruptcy law, attention is focusing on the new role of legal resolution for insolvency cases.
    3. In the area of employment, lifetime employment has been modified but not abandoned, and more and more companies are adopting merit pay. It may be that complementarity between employment systems and corporate governance is weaker than would be expected on the basis of stylized theoretical models for Japanese companies.
    4. Corporate legislation has expanded options. Commitment to board reform varies greatly by firm.
  5. Among Japanese firms, corporate governance is diversifying, and hybridization is giving rise to a variety of blends.
  6. When a cluster analysis of Japanese businesses is carried out on the basis of a Ministry of Finance survey using 14 variables for measuring various aspects of corporate governance, the corporations fall into three broad groups:
    1. The J-firm type, characterized by keiretsu networks, strong bank dependence, and minimal commitment to CG reform (69% of firms)
    2. The hybrid type, characterized by institutional investors, higher dependence on bonds, and rapid CG reform (14% of firms)
    3. The independent firm, characterized by individual ownership, dependence on SME finance, and slow progress toward CG reform (17% of firms)
  7. However, apart from these, it is possible to discern nine important subtypes. Three of these are committed to corporate governance reform: the J-hybrid, A-hybrid, and progressive J-Type.
  8. Corporate performance depends not only on which CG system a firm adopts but also on how well the system it chooses fits the company's situation with respect to (1) industrial sector, (2) constraints originated from nation's unique characters, (3) international constraints, including pressure from foreign countries. The degree of complementarity between various elements of corporate governance also depends largely on the corporate and industrial context. Corporations' varying efforts to reconstitute corporate governance, reflecting these variables, have resulted in a multiplicity of hybrid models.
  9. Institutional change follows three basic patterns: One is "institutional exhaustion," typified by the main-bank system. The next is "institutional conversion," in which a system takes on new functions. This is exemplified by the modification of the lifetime employment system, which has been incorporated into a system of employment adjustment instead of a system for guaranteeing employment. The third pattern is "institutional layering," in which new options are added on to an existing system, as embodied in reform of the board system.
  10. As globalization progresses, global diversity of CG systems is likely to shrink but not disappear. Japan's hybrid types might adapt well. However, the Japanese firm must transform itself into an enterprise community embracing a wider range of stakeholders and improve transparency.
  11. Henceforth, the role of the state should be to pressure traditional J-firm type to reform, restore the banking system to health, nurture institutional investors, and respond to the challenge of M&A with an understanding of the merits and drawbacks.

Hideshi Itoh (Professor, Graduate School of Commerce and Management, Hitotsubashi University) offered the following comments in response to the Jackson-Miyajima paper.

  1. The central message of this presentation is that while CG is undergoing major changes in Japan, the change cannot be characterized as convergence to the Anglo-American model. There are considerable differences among firms, and heterogeneity is increasing, with the J-firm, J-type hybrid, A-type hybrid and others coexisting.
  2. The immediate questions that emerge from this are:
    1. Are these different types internally coherent?
    2. Is the reason corporate governance reform is not necessarily always reflected in performance that (i) optimization of CG is not global but only local? (ii) change is limited to isolated aspects of CG? (iii) many aspects of CG are changed at once, but the degree of change is small?
    3. Are the changes in the right direction from a normative standpoint?
    4. Is the lack of apparent institutional isomorphism a temporary or permanent phenomenon?
    "Theoretical analysis" is essential to assess these sorts of changes in corporate governance.

  3. First, to understand "complementarity" theoretically, we should go back to Milgrom and Roberts. According to Milgrom and Roberts, when complementarities exist, it makes change difficult because (1) a change in one dimension alone, however large, will not improve performance, and (2) a very small change, even if occurring in many dimensions simultaneously, will not be reflected in performance. However, even if performance is currently poor, there is a possibility that this is a temporary setback necessary for a major improvement in the future. This argues for reform through centrally coordinated innovation, not merely local experimentation.
  4. It is possible that bringing about significant change is even more difficult when the components of corporate governance are tightly coupled, as in Japan.
  5. What is corporate governance? Jackson and Miyajima view it broadly, as the institutional rules and shared beliefs that define the roles multiple stakeholders play in a company's decision-making process. In an ideal situation, the question would be how to raise efficiency as a whole while reflecting the welfare of multiple stakeholders, but in the real world of second best, the cost of achieving value for a wide variety stakeholders is a major issue.
  6. In other words, corporate governance deals with the problem of how to induce management-the governed-to pursue shareholder value, or else stakeholder value. In this context, the shareholder versus stakeholder issue is one that deserves more discussion. In the United States, the emphasis is on the board of directors and its function of appointing, firing, and monitoring the CEO. In this symposium, we have not considered management as the governed party. More discussion is needed on the issue of what sort of impact the structure of the board has on the profile and turnover of top management.
  7. There are new economic theories that have some bearing on these issues. One is the idea of the interaction between formal governance and informal and relational governance. Long-term employment, regarded as central system in Japanese corporate management, belongs to the second category. Systems like this are self-enforcing; as long as the long-term drawbacks of deviating from the system exceed the short-term merits, the organization will adhere to it of its own accord. The affect of formal governance on this arrangement-in other words, the relationship between formal and informal governance-is a fascinating issue.
  8. Another theoretical perspective is offered by the idea of self-disciplined governance. Japan's best companies are characterized by a corporate culture that places priority on the good of society and its members, and this has played a significant role.
  9. Returning to the issue of shareholder versus stakeholder value, focusing on shareholder value imposes a bias on decision making, whereas focusing on stakeholder value divides control and makes decision making difficult. A more realistic method is to adopt a shareholder-value approach but put in place a mechanism for protecting the stakeholders. Using detailed contracts, ensuring high mobility in the labor market to facilitate employee exits, and building systems immune to shareholder influence can have that effect.
  10. From this perspective, management's commitment to lifetime employment is one type of employee protection. A full-fledged shareholder-value approach paired with a commitment to lifetime employment could be an ideal combination. However, since lifetime employment is an informal, relational system, it needs to be given high credibility through self-disciplined governance committed to a corporate culture that places priority on the good of society and its members. Under these circumstances there is a good possibility that the combination would have internal consistency.

Next, Juro Teranishi (Professor, Institute of Economic Research, Hitotsubashi University) offered the following comments on the Jackson-Miyajima paper.

  1. The presentation considers the future of corporate governance in Japan and concludes that continued diversity and hybridization is a possibility. The basis of this conclusion is an analysis that focuses almost exclusively on efficiency and such political factors as regulation. However, there is a third important factor, namely institutional infrastructure.
  2. In East Asia many countries moved to deregulate finance as early as the 1980s, but even today the ratio of bank loans to total assets is high. In Japan, likewise, the level of bank deposits, bank loans, and cross-shareholding among nonfinancial corporations remains high. Government deregulation is not enough to ensure real reform.
  3. With regard to efficiency, looking first at allocational efficiency, banks contribute to efficient allocation at the company level through the corporate information they acquire in the course of their normal dealings, while the stock market contributes to allocational efficiency by incorporating the diverse views of stock market investors regarding new industries and technologies. Where organizational efficiency is concerned, banks contribute to lower agency costs through the main-bank system. The market contributes by being an organization that is able to provide efficiency by itself.
  4. One way of understanding institutional infrastructure is the "legal origins" approach of La Porta and others. La Porta notes that investor protection is strong in the common law countries, Britain and the United States, and weak in France, Germany, and other civil law countries. In East Asia, Malaysia, Singapore, and Thailand follow the common law model, while Japan follows the German model. However, when judged by a number of different indicators, Japan seems to fall in the middle of the East Asian continuum. La Porta's approach does not successfully explain the characteristics of Japan and other East Asian countries.
  5. Another approach to institutional infrastructure focuses on differences in institutional design principles. Japan and other East Asian countries emphasize organizational efficiency. Anglo-American firms tend to emphasize allocational efficiency . In other words, the difference between Japan and other East Asian countries on the one hand and the Anglo-American model on the other can be explained in terms of family ownership versus the public corporation, internal labor market versus external labor market, and bank dependence versus funding via capital markets.
  6. However, there is a trade-off between allocational efficiency and organizational efficiency. Dependence on the external labor market is unsuited to investment in company-specific skills but efficient in allocating a standard workforce. Dependence on capital markets entails added costs related to information processing but is efficient in selecting new industries and technologies.
  7. In view of this, there is a good possibility that a major difference exists between the Japanese / East Asian economic system and the A-model in terms of the principles of organization and system building.

In response to these comments, Hideaki Miyajima (Faculty Fellow, RIETI, and Professor, School of Commerce, Waseda University) gave the supplementary presentation outlined below, which was followed by discussion.

  1. Corporate governance of the traditional Japanese firm, formerly characterized in terms of long-term employment, seniority-based pay, the main-bank relationship, and cross share-holding, has split into three types under the impact of capital market pressures and the IT revolution. One of these, the "lock-in J-type," preserves the old system intact.
  2. The lock-in J-type has emergent necessity to be reformed, and it must be subjected to pressure. This could involve (1) rebuilding the main-bank system by making banks healthier, (2) encouraging institutional investors to exercise their voting rights, (3) the latent threat of merger and acquisition.
  3. As to whether those corporations oriented to the U.S. model are internally consistent, it is too soon to say. It is possible that these firms have only changed certain aspects of their corporate governance systems. Perhaps we should be speaking of different patterns of response to the external environment rather than types.
  4. In terms of providing a theoretical explanation for the meaning and potential of the progressive J-type category of firms, Professor Itoh's comments provide much food for thought.

(Summary compiled by Yuji Hosoya, Director of Research, RIETI)