--1」／「－/-/- --.－」11...--1【－--.-1－【1-1-4-－【1·．111.1릍F,,/1----111·1-1--1…－「．1-4【1/－【，--1--－떼－!--!---［「」」【－【1,-／「」［【1  Corporate Governance and Enterprise Reform in China  Q劇 黑：浮·呈 豐們畸■緘d杳 〕簾朧鬍！黑 華丰―鑒〔．萬〕藝 中叫尸•■馴■斤j久、凶 O凶螂馴 O的縴 Principal authors: Stoyan Tenev, Chunlin Zhang, and Loup Brefort. Copyright (D 2002 The World Bank and the International Finance Corporation 2121 Pennsylvania Avenue, N.W. Washington, D.C. 20433 USA www.ifc.org All rights reserved Manufactured in the United States of America First printing, March 2002 ISBN 0-8213-5136-2 The findings, interpretations, and conclusions expressed in this study are entirely those of the authors and should not be attributed in any manner to the World Bank, to its affiliated organizations, or to members of its Board of Executive Directors or the countries they represent. IFC and the World Bank do not guaran- tee the accuracy of the data included in this publication and accept no responsibil- ity whatsoever for any consequence of their use. To order additional copies by mail, write to the World Bank, RO. Box 960, Herndon, VA 20172-0960, USA; by phone, 1-800-645-7247 or 703-661-1580; by fax, 703-661-1501; by e-mail, www.books@woridbank.org. The material in this publication is copyrighted. Requests for permission to repro- duce portions of it should be sent to the Copyright Clearance Center, Inc., Suite 910, Rosewood Drive, Danvers, Massachusetts 09123, USA. Telephone: 978-750- 8400; fax 978-750-0569; Web address, www.publisbers@copyrigbt.com. Library of Congress Cataloging-in-Publication data have been applied for. Contents FOREWORD, Javed Hamid and Homi Kharas vii ACKNOWLEDGMENTS iX ABBREVIATIONS AND ACRONYMS X EXECUTIVE SUMMARY xi 1 INTRODUCTION I 2 THE EVOLUTION OF GOVERNANCE MECHANISMS IN CHINA'S STATE SECTOR 5 Corporate Governance in the Context of China's Overall Approach to Reform 5 From Danwei to the Modern Enterprise System 10 Agency Costs of Local Government Ownership and Enterprise Autonomy 20 Current Approach 24 Conclusion 28 3 THE CORPORATE GOVERNANCE OF TRANSFORMED SMALL AND MEDIUM ENTERPRISES 29 Ownership Transformation and Emerging Governance Issues 29 Role of Employees 40 Role of Banks 55 Role of Private Equity Investors 70 Conclusion 74 v CONTENTS 4 OWNERSHIP AND CONTROL OF LISTED COMPANIES 75 Ownership Concentration and Types of Investors 76 Ownership and Corporate Governance Issues 80 Board of Directors 83 Board of Supervisors 99 Agency Problem of the Controlling Shareholder 101 Conclusion 103 5 ROLE OF STOCK MARKETS AND INFORMATION DISLOSURE IN THE CORPORATE GOVERNANCE OF LISTED COMPANIES 105 Corporate Governance and Performance 105 Stock Market Role in Promoting Good Governance 110 Information Disclosure as a Tool of Corporate Governance 117 Conclusion 126 6 BUILDING A MODERN GOVERNANCE SYSTEM 127 Establishing Credible Penalties for Failure 127 Addressing the Agency Costs of Government Ownership 128 Strengthening Boards of Directors 135 Empowering Shareholders 142 Developing Capital Markets That Reward Good Corporate Governance 154 Improving Disclosure 156 Activating the Use of Various Corporate Governance Mechanisms 159 Conclusion 160 REFERENCES 163 vi Foreword Corporate governance has been identified by the Chinese government as the core element of the "modern enterprise system." The policy focus on corporate governance reflects the significant progress that China has made in building market institutions and the importance it attaches to changing corporate behavior. More than two decades of market-oriented reforms in China have created economic entities with a relatively high degree of autonomy. To date, however, ownership diversification and corporatization have had only a limited impact on corporate behavior. China's commit- ment to improving corporate governance practices reflects the authori- ties' growing concerns about the potential consequences of a high-level of nonperforming loans in the banking system, overcapacity in most of the industrial sector, and a highly volatile and speculative stock market. Externally, commitments under the World Trade Organiza- tion will expose Chinese companies to the opportunities and chal- lenges of globalization and add to the urgent need to tackle corporate governance issues in a comprehensive and systematic manner. In this context, Corporate Governance and Enterprise Reform in China explores the main corporate governance issues that China is encountering during the course of corporatization and ownership trans- formation of its enterprise sector. It makes a large number of recom- mendations concerning the policy and legal frameworks, procedures, and institutional capacity for improving corporate governance prac- tices in China. The study reflects the increasing emphasis that IFC and the World Bank place on improving corporate governance practices as part of the general effort to support the development of the market institu- tions needed for sustained growth and poverty reduction. In China, the World Bank's work over the years in support of government re- forms in the financial sector, corporate restructuring, accounting, and legal and judicial practices has contributed directly to the development vii CORPORATE GOVERNANCE IN CHINA of the institutions of corporate governance. At the company level, IFC is playing an important role in bringing Chinese companies closer to international standards in corporate governance through technical assistance, institution building in the area of financial markets, and incentives embedded in financial instruments. Current World Bank Group work in corporate governance emphasizes governance of fi- nancial institutions, capacity building through training for regulators, company directors, business owners, and investors; and dissemina- tion of best practices through the Global Corporate Governance Fo- rum, studies, and workshops. We hope that this study will provide all those with an interest in the corporate governance practices of Chinese companies with new insights into their status and new ideas for ways to support and par- ticipate in their future improvement. Javed Hamid Homi Kharas Director Chief Economist East Asia and Pacific Department East Asia and Pacific International Finance Corporation World Bank Vil1 Acknowledgments The study is a joint product of the IFC East Asia and Pacific Depart- ment and the Private Sector Development Unit, East Asia and Pacific Region of the World Bank. Background notes were prepared by Jean- Francois Arvis, Cally Jordan, Zhong jiyin (Integrity Consulting, Beijing), Klaus Lorch, Daochi Tong, Feng Tongqing (China Labor College), Jing Yiqing, (Deloitte & Touche, Beijing), Xu Xiaosong (China University of Politics and Law), and Junkuo Zhang (DRC). Arvind Gupta and Chau-Ching Shen made special contributions at various stages of the work. Harry Broadman, Cheryl Gray, and Rich- ard Newfarmer were peer reviewers. Extensive guidance and support were provided by Loup Brefort and Karin Finkelston. The Development Research Center (DRC) at the State Council supported and facilitated the study for its successful implementation. Chen Xiaohong, director general, and other DRC staff provided valu- able guidance and support throughout. A draft of the study was pre- sented and discussed at a workshop organized by the World Bank Group and DRC in May 2001 in Beijing. Additional presentations and com- ments were made by Wu Jinglian (DRC), Zhang Zuoyuan (CASS), Fang Liufang (China University of Politics and Law), Chen Xiaohong (DRC), Ding Ningning (DRC), Hu Ruyin (Shanghai Stock Exchange), Zhang Wenkui (DRC), Wang Dongjiang (SCORES), Chen Yanhai (SETC), Zhang Weiguo (CSRC), Li Xiaoxue (CSRC), Zhou Fangsheng (SETC), Chen Su (CASS), Liu Shijin (DRC), and Li Zhaoxi (DRC), who to- gether with the discussions at the workshop enriched the final study. The study also benefited from comments and insights from other World Bank and IFC staff, including Deepak Bhattasali, Olivier Fremond, Carmen Genovese, Sudarshan Gooptu, Mike Lubrano, Behdad Nowroozi, Djordjija Petkoski, Guy Pfeffermann, Peter Tay- lor, Jun Zhang, and Junkuo Zhang. Udayan Wagle and Mariko Higashi supported the field work through funding from IFC Trust Funds. Alice Faintich was the principal editor. Dana Lane organized the produc- tion of the book. Assistance was provided by Katie Shaw, Doris Chung, and Guo Zhenxuan. ix Abbreviations and Acronyms AMC Asset management company ASBE Accounting Standards for Business Enterprises CaIPERS California Public Employees' Retirement System CEO Chief executive officer CPA Certified public accountant CSRC China Securities Regulatory Commission D&O Directors and officers ESOP Employee stock ownership plan GDP Gross domestic product IAS International Accounting Standards IFC International Finance Corporation IPO Initial public offering M&A Merger and acquisition PBOC People's Bank of China SBIC Small business investment company SEC Securities and Exchange Commission (United States) SETC State Economic and Trade Commission SOCB State-owned commercial bank SOE State-owned enterprise ST Special treatment TVE Township and village enterprise WTO World Trade Organization x Executive Summary Over the past decade or so, China has made significant progress in developing the institutional foundations of a modern corporate gov- ernance system. More than 80 percent of all small and medium enter- prises have been transformed, with a significant portion sold to employees and outside investors. About 1,200 large companies have diversified their ownership through public listing. A basic legal frame- work underpinning the corporate form and including company law, contract law, accounting, and securities laws has been established. The financial system has become more diversified and independent of political influence. The regulators' capacity to enforce the new rules and prevent wrongdoings has been strengthened. In the past several years, the efforts of the authorities to improve corporate governance practices have intensified as exemplified by initiatives such as the sys- tem of independent directors for listed companies and the code of corporate governance for listed and nonlisted companies introduced by the China Securities Regulatory Commission and the State Eco- nomic and Trade Commission. Notwithstanding these impressive achievements, there is vast scope for further institution building to improve the corporate governance practices of Chinese companies. The following sections focus on remaining weaknesses, outstanding issues, and recommended priorities for policy actions. Summary Assessment The present structure of state ownership and control of enterprises accounts for some of their poor performance. This results from weak incentives for managers to maximize value for all investors and credi- tors and from protectionist practices of government agencies that shield firms from market discipline. The process of ownership diversifica- tion is itself often conducted in ways that inhibit the evolution of healthy corporate governance practices. In the case of listed companies, the xi CORPORATE GOVERNANCE IN CHINA initial public offering process has tended to select companies that have strong links with local governments and fuzzy boundaries with their parent groups. This has created strong incentives for the controlling shareholders to exploit companies' interdependence through related- party transactions. Implicit support by the government and parent companies, the franchise value of listing, and weak creditors' rights generate expectations among investors that they are engaging in low- risk investments. As a result, investors have few incentives to assess companies' fundamentals carefully or to demand good corporate gov- ernance. In the case of transformed small and medium enterprises, unrealistic valuation of assets, and the exclusion of land-use rights from the asset pool to circumvent the insiders' wealth constraint to taking a majority position are likely to make future access to capital markets more difficult, thereby preventing banks and outside investors from playing an important role in the governance of these enterprises. Banks and outside investors lack the capacity, the regulatory sup- port, and the incentives to actively monitor and influence companies' behavior. Bankruptcy of state-owned enterprises is largely an admin- istrative process, and the effective rights of creditor banks in cases of debtor default are weak. State-owned commercial banks generally suffer from similar corporate governance weaknesses as nonfinancial state-owned enterprises: their profit incentives are weak at best. The separation of commercial and investment banking means that banks cannot use ownership to supplement their creditor rights and exert more influence on firms. Local governments' practice of supporting their enterprises in difficult times makes credit decisions a function of an enterprise's implicit or explicit government support rather than of its merits, thereby reducing banks' incentives to evaluate and monitor companies' behavior. Private equity markets, especially venture capi- tal, are in an embryonic stage of development, and the state still plays a ubiquitous role as sponsor, investor, and fund manager. National regulations on venture capital and investment funds are still missing, although work on important legislation is in progress. In addition, China does not have an adequate legal framework for structuring con- tractual arrangements of particular importance to private equity in- vestors, such as convertible loans and options. Corporatization and ownership diversification have introduced new institutional forms for exercising corporate control without the dismantling of old representative bodies. The division of labor be- tween old and new governance structures is unclear and is further xi EXECUTIVE SUMMARY complicated by many companies' practice of combining such posi- tions as chair of the board of directors with secretary of the Party committee. As a result, key decision-making powers tend to be vested in informal mechanisms, and some institutions such as boards of su- pervisors have assumed largely decorative functions. In the case of listed companies, large shareholders often overstep the bounds of share- holders' meetings and boards of directors and exercise direct effective control. Relative to practices in other countries, boards are less inde- pendent, and some of their powers are, in effect, exercised by control- ling shareholders and government agencies. Chinese capital markets lack mature users of financial informa- tion, such as institutional investors and analysts. Financial reporting and disclosure are primarily oriented to satisfy the information needs of the taxation authorities. The interdependence between listed and parent companies creates strong incentives to distort information, particularly concerning related-party transactions. The quality of au- dits suffers from the narrow minimum requirements regarding cov- erage of the audit, the unclear liability of auditors, the challenges to the independence of many auditors from the state as the owner of audited enterprises, and a general shortage of well-skilled auditors at the local level. Recommendations Recommended priorities for action are based on the following guid- ing principles: * Corporate governance scandals in emerging and developed mar- kets indicate that there is no perfect corporate governance model. An effective corporate governance system should above all be capable of identifying weaknesses before they develop into systemic problems, of learning from failures, and of taking prompt corrective actions. Criti- cal ingredients of such a system are a credible threat of market failure and an effective regulation that builds on the incentives of market players in order to develop an effective system of checks and balances. * The institutional mechanisms of corporate governance com- prise a system that can employ alternative yet complementary instru- ments of control to effectuate changes in companies' behavior. An effective corporate governance system contains a multiplicity and cer- tain redundancy of control mechanisms. This principle implies that xiii CORPORATE GOVERNANCE IN CHINA priority should be given to mechanisms that (a) are relatively under- developed or altogether missing from a country's institutional arsenal of corporate governance mechanisms, and (b) exhibit strong syner- gism with other existing mechanisms. Based on these principles and on our assessment, the following areas emerge as recommended priorities for policy action: (a) alleviat- ing the negative impact of dominant state ownership on market disci- pline and on the regulatory capacity of the state; (b) building an institutional investor base; and (c) strengthening the role of banks in corporate governance. Many of the specific recommendations are con- sistent with and reinforce recommendations made in previous World Bank studies, particularly the 1997 report China's Management of Enterprise Assets and the 2001 report Bankruptcy of SOEs. Strengthening Market Forces and Regulatory Capacity. Dominant state ownership tends to erode the credibility of the threat of market failure and the regulatory capacity of the state. Given that the effectiveness of each and every corporate governance mechanism ultimately rests on a credible threat of market failure and a strong regulatory capacity, this underscores the point that sustainable improvements in corporate governance are unlikely without fundamental changes in ownership patterns. China could move more aggressively in experimenting with mecha- nisms for separating state control from state cash flow rights as a way to reduce political control over companies. Experiments with the man- agement of listed state shares by private institutional investors could promote a more market-based and value-maximizing approach. Modi- fying the nature of government equity claims by, for example, trans- forming them into preferred nonvoting shares is another approach. This would make the government's cash flow rights more like certain tax liabilities, thereby promoting greater consistency between the dif- ferent roles the government is playing with respect to government- owned firms. Such measures can be useful transitional mechanisms, as they could send a powerful signal that the government is commit- ted not to interfere with market forces. Various ways can be used to reduce the number of state-owned shares: state share placement, share repurchase, negotiated transfer, auctioning, and debt-equity transfers. An appealing way to reduce state shares is through institutional investors, because this has obvi- xiv EXECUTIVE SUMMARY ous synergies with capital market development and social welfare re- form. The Hong Kong experience with the Tracker index fund sug- gests a potentially useful method of divesting state shares with minimum disruption of market stability. State and legal person shares should gradually be allowed to become tradable so that market forces can begin to shape the ownership structure of listed companies. Given the magnitude of the regulatory challenge and the limita- tions imposed by dominant state ownership on the effectiveness of direct forms of regulatory interventions, the government will have to rely more on indirect methods of regulation including delegated moni- toring, self-regulation of professional organizations, and mobilizing civil society in the enforcement process. Indirect control over compa- nies' behavior through regulations of institutional investors and through accounting and legal firms that are independent of govern- ment and are not "too big to fail" will enhance regulatory efficiency. Empowering the "right" party, with an interest in certain regulations being enforced, implies enhancing the independence of associations, the media, self-regulatory bodies, and other members of civil society. The recent report on widespread market manipulation by China's 10 fund management companies in Caijing Monthly illustrates the enor- mous social benefits of independent civil discovery. Such practices should be encouraged. Developing an Institutional Investor Base. Institutional investors can play a catalytic role in activating the use and enhancing the effective- ness of many of the instruments of corporate governance. To facilitate shareholder activism by institutional investors, a priority should be to strengthen shareholders' rights, through, for example, a cumulative voting system or automatic rights for investors above a certain thresh- old shareholding to appoint a director; quorum requirements for share- holders' meetings based on outstanding shares; a proxy system, which through proxy contests can act as a partial substitute for the takeover process; and class action procedures. Although important in them- selves, such rights may not be sufficient to create active institutional investors. Based on international experience, three critical factors for active involvement of institutional investors in corporate governance are: (a) mitigation of conflicts of interests by restricting activities that may create excessive interdependence between companies and institu- tional investors; (b) making voting an integral part of institutional investors' fiduciary duties; and (c) allowing institutional investors to xv CORPORATE GOVERNANCE IN CHINA be named controlling parties in shareholder lawsuits against company management. Also of importance is the regulators' ability to supervise institutional investors and the corporate governance of domestic in- stitutional investors. In this context, privatization of existing institu- tional investors and, perhaps more important, accelerated new entry by domestic and international private institutional investors, should be considered. China has the option of importing regulatory and cor- porate governance capacities in this area by opening its capital mar- kets to foreign institutional investors and by promoting cooperation between foreign and domestic institutional investors in the form of joint ventures and technical assistance arrangements. Strengthening Banks' Role in Corporate Governance. Creditors are among the least effective instruments of corporate control in China, and strengthening their role in corporate governance should be a pri- ority. This is particularly important in the case of small and medium enterprises whose closely held nature precludes reliance on public monitoring. Legislation currently under preparation should take the opportunity to transform bankruptcy from a purely administrative process to a more market-driven one. This should involve consider- able strengthening of creditors' rights in the case of default and en- hanced options for banks to engage in reorganizations and restructurings of client companies. Allowing greater room for com- mercial bank involvement in investment banking activities, such as providing securities advice and custodial services that can lead to proxy voting by banks, will enhance banks' role in corporate governance. There is a strong economic rationale for allowing banks to hold quasi- equity and equity instruments, at least for a predefined maximum pe- riod, to facilitate restructuring. xvi 1 Introduction Corporate governance has moved to the center stage of enterprise re- form in China. The Fourth Plenum of the Chinese Communist Party's 15th Central Committee held in September 1999 adopted a "deci- sion" that calls for "strategic adjustment" of the state sector by "with- drawing what should be withdrawn." The decision identifies corporate governance as "the core" of the "modern enterprise system," the new system expected to prevail in the reformed enterprise sector. The cur- rent emphasis on corporate governance reflects the significant progress that China has made in building market institutions, but also the lim- ited success of past reform efforts in changing corporate behavior. Market-oriented reforms, including corporatization and owner- ship diversification, have brought corporate governance issues to the forefront. More than two decades of reforms have created economic entities with a relatively high degree of autonomy that are subject to significant market pressure and whose capacity to decide and struc- ture the parameters of their mutual interactions are growing. Most large and medium state-owned enterprises (SOEs) have corporatized themselves, although the process has not been completed. Ownership diversification has taken two main forms: listing on domestic and in- ternational stock exchanges in the case of larger SOEs, and sales to insiders, namely, management and employees, in the case of small and medium SOEs. In the process, new institutions for the exercise of corporate con- trol have emerged, such as boards of directors and supervisors. As a result, issues such as how to make these institutions more effective; what their composition and modus operandi should be; and what the appropriate division of labor should be between them and traditional representative bodies, such as trade unions, employee conferences, and party committees, have become important. Corporatization and own- ership diversification have also led to the emergence of new owners 1 CORPORATE GOVERNANCE IN CHINA and stakeholders, such as individual minority shareholders (about 60 million at present), institutional investors, and employee sharehold- ers. Their emergence has created the need to specify the rights of such stakeholders, clarify their role in corporate governance, and establish mechanisms to protect their interests. However, to date ownership diversification and corporatization have had only a limited impact on corporate behavior. The current policy focus on corporate governance reflects growing concern about the negative consequences of poor corporate governance practices. According to a recent People's Bank of China (PBOC) report, of the 62,656 enterprises that had completed transfers of ownership by the end of 2000, 51.2 percent had failed to repay their bank debts. The poor financial performance of a large number of SOEs, including state- controlled listed companies, continues to impose a severe burden on the banking system, Treasury, and stock market and is a potential threat to social stability. The nonperforming loan ratios in the finan- cial system are estimated at between 25 and 40 percent. Large excess capacity exists in manufacturing, but because of the structure of the labor market, many firms still carry excess labor on their books (Bhattasali and Kawai 2001). Unemployment concerns are slowing the pace of restructuring of loss-making SOEs. Commitments under the World Trade Organization add to the urgent need to tackle corporate governance issues in a comprehensive and systematic manner. As part of its accession negotiations, China has committed to a broad range of market access measures. Some will revolutionize the organization of business activity, thereby creating pressures for moving toward a rules-based, as opposed to relationship- based, investment environment and greater transparency in business and government activities consistent with international investment- related rules. Further trade liberalization in the context of WTO entry is expected to create significant pressure to reallocate productive re- sources in accordance with China's comparative advantages. These changes would be in addition to the resource reallocation trends that are already taking place as part of the transition from a planned economy to a market economy and from an agricultural to a manu- facturing and service-oriented economy. Corporate governance ar- rangements will determine to a large extent the way firms and other economic agents respond to these internal and external pressures. The current policy focus on corporate governance thus plays an important role in the internal dynamics of market-oriented reforms in INTRODUCTION China. While establishing and strengthening the new mechanisms of corporate governance are necessary to realize and solidify the benefits of past reform efforts, they are also likely to prepare the ground for further progress in institutional transformation. A good corporate gov- ernance framework is likely to facilitate the state's withdrawal from direct ownership. The topic of this study is the short- to medium-term corporate governance issues that are arising during the course of transformation of ownership in the Chinese state enterprise sector. The study looks at companies participating in the two main forms of ownership diversi- fication: listed companies and small and medium enterprises whose ownership structure is dominated by insiders. The focus is on the new mechanisms and stakeholders emerging during the process of owner- ship diversification and their role in corporate governance: boards of directors and supervisors, minority shareholders, shareholding em- ployees, creditors, information disclosure, and the capital market. While these issues are important for corporate governance in general, their relative importance differs in listed and nonlisted companies. Thus the study discusses the respective roles of boards of directors, minority shareholders, information disclosure, and capital markets in the context of listed companies. It discusses the role of employees, creditors, and outside private equity investors in the context of small and medium enterprises with insider-dominated ownership structures. However, many of the issues, observations, and recommendations extend to both types of companies. In the case of listed companies, the analysis, particularly of board structure and practices, is based on a survey of corporate governance practices among companies listed on the Shanghai Stock Exchange conducted in early 2000 by Integrity Management Consulting and the Research Center of the Shanghai Stock Exchange. A total of 10,560 questionnaires were sent to the directors, supervisors, and senior man- agers of all companies listed on the Shanghai Stock Exchange at that time, of which 9,600 were individual questionnaires, 480 were enter- prise questionnaires, and 480 were financial data questionnaires. The response rate was 41 percent for the individual questionnaires, 54 percent for the enterprise questionnaires, and 50 percent for the fi- nancial data questionnaires. Extensive information about corporate governance practices was thus obtained for 257 listed companies. Regarding transformed small and medium enterprises, the infor- mation came from three sources in the following order of importance: 3 CORPORATE GOVERNANCE IN CHINA in-depth interviews with government officials, workers, and manag- ers and detailed case studies of 14 enterprises in the towns of Jinhua in Zhejiang province and Zhucheng in Shandong province; interviews with enterprise and government officials in Beijing, Chongqing, Chengdu in Sichuan province, Shunde in Guangdong province, and other localities; and findings from surveys and research conducted by Chinese academic and government institutions. Localities were cho- sen based on considerations about political and economic importance, coverage of both interior provinces and coastal areas, and leadership in economic reforms. In particular, the cities of jinhua and Zhucheng were selected because they were among the first in China to launch comprehensive reform of their state enterprise systems and other prov- inces have emulated their approach. These cities' relatively long expe- rience with enterprise reform presents a valuable opportunity to observe the dynamics of ownership diversification at the local level in China and to draw conclusions that may be applicable to other localities that are at less advanced stages of reform. The study is set out as follows. Chapter 2 traces the main histori- cal developments in China's state enterprise reform from a governance perspective. It examines the evolution of the main governance prob- lems, from controlling the agency costs of increased enterprise au- tonomy to the emergence of a modern corporate governance framework. Chapter 3 discusses emerging ownership patterns in trans- formed small and medium SOEs and the roles of creditors, employees, and outside investors. This chapter also recommends how to strengthen the role of employees, creditors, and private equity Investors in corpo- rate governance. Chapter 4 looks at the ownership and control struc- tures of listed companies, focusing on boards of directors. The role of capital markets and information disclosure is examined in chapter 5. Finally, chapter 6 provides recommendations on corporate governance issues pertinent to listed companies. 4 2 The Evolution of Governance Mechanisms in China's State Sector For the purpose of this study, we define corporate governance as the set of instruments and mechanisms (contractual, legal, and market) available to shareholders for influencing managers to maximize share- holder value and to fixed claimants, such as banks and employees, for controlling the agency costs of equity (see box 2.1). This chapter places the current focus on corporate governance in the context of China's overall approach to market reforms, and traces the evolution of SOE reforms that led to the emergence of corporate governance as the core issue of the modern enterprise system. Corporate Governance in the Context of China's Overall Approach to Reform Although China adopted new policy direction without political liber- alization, the beginning of market-oriented reforms was accompanied by an important shift in ideology. A pragmatic approach focusing on development supplanted the fixation on how the revolution could be prevented from degenerating. The new growth imperative was ex- pressed most forcefully by Deng's (1994) proclamation that "devel- opment is the hard truth." China lacked a well-defined strategy or a clear blueprint of how exactly to promote development, but deliber- ate efforts were made early in the reform process to align government incentives at all levels with the new political focus on growth. The bureaucratic system was substantially transformed by intro- ducing a mandatory retirement program for the veterans of the revo- lution, promoting a drive for administrative and fiscal decentralization, and allowing bureaucrats to quit the bureaucracy and join businesses (Li 1998). Powerful incentives were added to promote local economic 5 CORPORATE GOVERNANCE IN CHINA BOX 2.1 CORPORATE FORM AND CORPORATE GOVERNANCE Modern, large-scale production involves inputs by multiple agents with divergent interests. Specialization typically extends to management and control, with the result that investors commit their resources to the con- trol of specialized agents. In doing so, investors compare the expected benefits of specialization with the agency costs associated with the di- vergence of interests and the risk that the resources they contribute may be squandered. Investors thus need some assurance that their interests will be protected, and such assurance usually takes the form of laws, contracts, discretionary authority, and informal arrangements. This set of institutional mechanisms governing the exercise of control over re- sources is the essence of governance of the production process. Under this general definition, governance issues arise in any economy where the division of labor extends to management and control. How- ever, the institutional mechanisms of governance can differ widely across economic systems. For example, a command economy relies exclusively on administrative mechanisms of control. Resources are combined by fiat and contracts and laws play an insignificant role, the autonomy of various parties is limited, the state sanctions all significant interactions, and risks and rewards are largely socialized. As a result, pecuniary in- centives are not emphasized. By contrast, a market economy allocates control over resources primarily through formal and informal volun- tary contracts between autonomous agents. The state provides a legal and regulatory framework for private arrangements and an enforce- ment mechanism for such agreements. The system is flexible and dy- namic, whereby different solutions emerge within a common framework as participants combine the basic components of a governance struc- ture to fit their own particular circumstances. The corporate form has evolved to solve the problems of incen- tives, monitoring, and information, or in other words, the problem of governance, that accompany the process of exchange for the purpose of joint production. The corporation is a set of contracts that allocate claims on income and control rights. It issues stock in exchange for an invest- ment. Shareholders bear the risk of failure and receive the marginal rewards of success. Equity investors are paid last, after debt investors, employees, and other investors with "fixed' claims. They have a re- sidual claim in the sense that they get only what is left over. Under normal circumstances, shareholders' risks are limited to the amount they have invested in the corporation. As residual claimants, equity in- 6 EVOLUTION OF MECHANISMS vestors bear the marginal consequences of their own decisions and have incentives to monitor the inputs of other participants and to make effi- cient economic decisions. Therefore allocating control rights to share- holders is efficient as long as the corporation is in a position to keep its promises in the form of fixed claims. However, when losses erode a corporation's equity, limited liability creates perverse incentives for eq- uity holders that can threaten the interests of fixed claimants. Thus fixed claimants have incentives to monitor these agency costs of equity for actions that may expose the corporation to significant risks. The corporate form thus embodies the basic structure of corporate governance, which largely concerns the mutual monitoring of share- holders and fixed claimants.' Corporate governance can therefore be defined as the set of instruments and mechanisms (contractual, legal, and market) available to shareholders for influencing managers to maxi- mize the value of shareholders' stock and to fixed claimants such as banks and employees for controlling the agency costs of equity. Share- holders' main mechanisms are the board of directors, direct shareholder activism, and the market for corporate control. Fixed claimants such as banks and employees rely mainly on elaborate contracts and a bank- ruptcy regime. All investors rely on information to protect their inter- ests to a varying degree. Thus the structure of information disclosure is a critical component of the institutional arsenal of corporate governance. While each of these mechanisms taken in isolation is an imperfect instrument for ensuring the efficient management of resources, in com- bination they can constitute an effective architecture. If the board of directors fails to take corrective action, shareholder activism can exer- cise pressure on the board. If the board of directors and shareholders are powerless to implement changes, and as a result the company con- tinues to underperform, it could become a potential takeover target. Finally, if none of these mechanisms can effectuate changes, the bank- ruptcy mechanism is supposed to facilitate changes in ownership, in the board of directors, and in the redesign of contractual arrangements. 1. Within this basic structure, agreements can be wonderfully diverse, matching the diversity of economic activity carried on within corporations. Shareholding structures may be extremely diffused or highly concentrated, managers some- times hold a great deal of a firm's stock, employees and banks may hold stock in addition to fixed claims, and so on. 7 CORPORATE GOVERNANCE IN CHINA development. These were in the form of a fiscal contracting system known by the nickname "eating from separate kitchens," which re- placed the previous system of "unified revenue collection and unified spending." The new system encouraged and rewarded local govern- ments for promoting development of their local economies. The growth and development of local economies became the main criteria for pro- moting local cadres. As a result, the bureaucracy functions as a "help- ing hand" for economic development, is directly involved in economic activity, pursues industrial policy, and often has close economic and family ties to entrepreneurs (Frye and Shleifer 1997; Walder 1995). Because of decentralization, the powerful incentives to promote development were supported by a significant capacity to design and implement policy initiatives at the local level. Since 1958 the Chinese economy has been organized around a geographical principle known as regional organization.' A regional system has the important advan- tage of flexibility: it can experiment with reforms locally because re- gional entities are self-contained and different ingredients of reforms can be tested without disrupting the organization as a whole. Thus in the absence of a clear blueprint for reforms at the na- tional level, and given the strong incentives to promote local develop- ment in the context of significant decentralization, China has developed an approach to market-oriented reforms that emphasizes gradual ex- perimentation at the local and sectoral levels (Gelb, Jefferson, and Singh 1993; Harold 1992). In line with this gradual approach, several years may elapse from the time a reform experiment starts in one of the provinces until the central government endorses it or other prov- inces imitate it. Another characteristic of reform has been the use of partial reforms within sectors, known as the dual-track approach. The first time this tactic was used was with two-tier pricing, which was introduced in rural areas in 1979 along with the household responsi- bility system. Later it was applied to other sectors: industry (through the contract management responsibility system), the national budget (through the fiscal contract responsibility system), external trade and payments (through the sharing of foreign exchange between central and local governments, trade contracting, and foreign exchange trad- ing centers), and labor markets (through the contract system for new hires in the state sector). 1. By contrast, organization in the former Soviet Union was much more central- ized, and was along sectoral lines (Qian 1999). 8 EVOLUTION OF MECHANISMS This dual-track approach is perhaps the most important aspect of Chinese reforms, because at the time it was an innovative solution to the political constraints on the direction and speed of reform. The adoption of a new policy direction without political liberalization and under the same political structure ruled out experiments that would have created losers on a large scale within the bureaucracy. Conse- quently, the experiments had to be of the dual-track type, so as to preserve the vested interests of the bureaucracy and a level of political stability. Although the reforms were controversial, the experimental dual-track method of introducing them enabled reformers to bypass the formal ideological debate usually required for public legislative sanction of reforms. In pursuing market-oriented reforms, China had to be creative in borrowing and applying market concepts that were relatively free from ideology and could easily be adapted to fit existing ideological con- straints. For instance, the reformers introduced market institutions such as stock markets and special economic zones by emphasizing their universal, technical, and pragmatic aspects (Deng 1994). New market institutions have been introduced without first dismantling old practices, and have often emerged from existing institutional ar- rangements. The coexistence of new and old institutions has been a distinctive feature of China's process of institutional transformation. Introduced partially and gradually and surrounded by institu- tional relics, the new institutions were imperfect, but were generally a sensible response to existing problems. Because China adopted new institutions in response to actual, narrowly defined problems, these institutions were being put to use as soon as they emerged. As they were being used, these institutional arrangements were evolving, gain- ing strength, and assuming new functions. In the process, new con- straints to the existing institutional structure were emerging. Over time, the ideological landscape has become more hospitable to new, mar- ket-oriented concepts and initiatives. In a similar vein, the current emphasis on corporate governance is another instance of creative borrowing of market concepts to de- sign policy responses to issues that emerged following partial reforms in the state sector. Corporatization and ownership diversification have not resulted in a fundamental change in ownership patterns, but have created a web of new agency problems and the rudiments of a corpo- rate governance structure. New institutions have emerged alongside old structures and are groping their way to becoming functional. In 9 CORPORATE GOVERNANCE IN CHINA the process they are creating a demand for new laws, regulations, and other institutional arrangements. These institutions are still weak and will remain imperfect in the presence of dominant state ownership and party control over managers; however, if they succeed they will prepare China for fundamental changes in these two areas. The significance of the current emphasis on corporate governance thus extends far beyond state enterprise reform. China's transition to a market economy is incomplete and will remain so if preserving a dominant share of state ownership remains an objective. As long as the constraints of dominant state ownership are imposed on the economy, China will not be able to have fully functioning factor mar- kets. As a result, the institutional arsenal of corporate governance will remain limited without efficient capital and labor markets. In this con- text, China's current corporate governance approach to enterprise reform can be viewed as the prelude to the final stage in its transition to a market economy. Given China's gradual approach to reforms, a historic perspective on the process of SOE reform that led to the cur- rent emphasis on corporate governance is useful. From Danwei to the Modern Enterprise System Based on distinctive levels of enterprise autonomy, we can distinguish three phases in the recent evolution of governance mechanisms: the collectivist prereform period, the second period from 1978 to 1992, and the third period from 1993 to the present. Vanished Economics: The Danwei. The main problem of the prereform system was its overwhelming reliance on administrative control and central planning. Given the insurmountable information problems associated with this system, it was inherently incapable of allocating resources rationally. Economic and efficiency considerations were sub- ordinated to political and social exigencies. The complete socializa- tion of risks and benefits implied a lack of incentives to discover and pursue business opportunities. The basic cell of economic life during the prereform period was the working unit or the danwei. The danwei system had multiple func- tions. It was foremost a political institution, one that extended the party's and state's presence to the grassroots level. It was also an ad- ministrative body that exercised control on behalf of the party and state. The danwei was an economic producer that provided social 10 EVOLUTION OF MECHANISMS welfare to workers and was attached to a central or local planning agency. The danwei had little economic independence and no clearly defined legal boundaries. They received production quotas, guaranteed outlets for products, and were given the necessary resources from the budget to reach production targets. Whatever "profits" the danwei made had to be remitted to the supervising state authority. The danwei system was designed not only to produce but also to deliver goods to members of its community. Having SOEs function as social welfare providers both attenuated the gravity of the short supply of goods and underlined the workers' need to rely on the danwei system. Permanent employment and membership in the social community were implied. 1978-92: Reintroducing Incentives. The focus during this first period of reforms starting in 1978 was on reintroducing markets and incen- tives within the domain of direct state ownership and control. Market forces started to operate alongside plans and administrative orders through a dual pricing system. In addition, the government allowed new structures such as collectives and private enterprises to develop and compete with state enterprises. Various types of performance con- tracts created a link between market success and compensation. Some rudimentary forms of penalties associated with failing the market test began to emerge: bank loans started to replace budget grants, a bank- ruptcy system was established for SOEs, and for the first time enter- prises could fire workers. Even though China did not introduce comprehensive price liber- alization like some East European countries did, it created a two-tier system under which companies could sell output produced in excess of the plan, initially at prices up to 20 percent above planned prices. Since 1985 companies have been able to sell their excess output at prices determined by markets. Markets for industrial products ex- panded in the early 1980s. By the late 1980s the share of output di- rectly marketed by firms had become quite substantial for most industrial producer goods, and even higher for consumer durables. Accompanying the rise of the share of the market was a decline in the proportion of goods subject to allocation under the state plan. By the second half of the 1980s market prices for industrial producer goods were common in numerous cities, and the evidence indicates that these prices were not subject to systematic controls (Byrd 1992, p. 7). Overall, even though the adjustment of government-controlled prices was usually limited and was hindered by the opposition of vested 11 CORPORATE GOVERNANCE IN CHINA interests, China made considerable progress in releasing prices from central planning and control. The government further liberalized the prices of producer and consumer goods in the early .1990s. As table 2.1 shows, 80 percent of producer goods were sold at market prices in 1994, compared with only 36 percent in 1990. The share of consumer goods sold at market prices also increased dramatically, from only 3 percent in 1978 to 53 percent in 1990 and 94 percent in 1993. The introduction of market forces had to be accompanied by appropriate reforms in the financial incentives of SOEs. Early reforms focused on restoring enterprises' ability to retain profits. On the basis of earlier local experiments, notably in Sichuan province, the govern- ment introduced a profit retention scheme in mid-1979 whereby it allowed a few enterprises to retain a share of the profits made on sales of items that were part of their government-mandated quota, but in other cases they could only keep profits on those sales they made after they had fulfilled their quota. Although profit retention was only in- tended to be a limited experiment, it spread rapidly, and by the end of 1980, 6,600 industrial SOEs, accounting for 60 percent of total in- dustrial output and 70 percent of total profits, had instituted some form of profit retention. However, incentives were still weak because of low and unstable retention rates (Byrd 1992, pp. 3-4). In 1981-82 the government instituted various forms of the "eco- nomic responsibility system," under which enterprises contracted to TABLE 2.I PRODUCER PRODUCT MARKET, MARKET VERSus REGULATED PRICES, 1990-94 (percent) Year Fixed prices Guided prices Market prices 1990 44.6 19.0 36.4 1991 36.0 18.3 45.7 1992 20.0 - 80.0 1993 15.0 5.0 80.0 1994 14.7 5.3 80.0 - Not available. NOTE: When market participants are free to determine prices within administratively set parameters, these are referred to as guided prices. a. Includes guided prices. SOURCE: IN/I (1996). 12 EVOLUTION OF MECHANISMS hand over only a certain percentage or fixed amount of their incre- mental profits. This system specified targets for profits that enterprises turned over to the government, with high retention rates for above- quota profits that often amounted to 60 to 80 percent, and sometimes even 100 percent. In 1983 the government introduced a scheme that substituted tax payments for profit remittances, and SOEs paid a profit tax at a uni- form rate of 55 percent. However, because of price distortions and other "objective factors" that caused profits to vary across firms and industries, the government also imposed an enterprise-specific adjust- ment tax on most large and medium enterprises. The early profit retention schemes were "soft" and negotiable, and weakened rather than strengthened financial discipline. Different rates of tax or profit retention led to considerable bargaining between enterprises and the government and weakened the incentives gener- ated by profit retention. China was the first of the transition economies to introduce per- formance contracts.2 Beginning in 1987, the government introduced a variety of contracts under the "contract responsibility system," which included the leasing of smaller firms, the contract management re- sponsibility system, the enterprise management responsibility system, and the asset management responsibility system. While the details of these programs varied, they shared some common elements. First, all of them involved a contract-based relationship between the enterprise, usually represented by its director, and its supervisory agency. Second, the directors faced substantial risks and rewards as a result of partici- pating in these programs, because their performance was linked to their enterprises' performance. Third, these schemes involved open selection (as opposed to direct administrative appointment) of enter- prise directors. Finally, most of these systems had multiyear targets and incentives in order to weaken ratchet effects. 2. Shirley and Xu (2001) analyzed China's experience with performance contracts in roughly 500 SOEs. They found that, on average, performance contracts did not improve performance and may have made it worse. However, they noted that China's performance contracts were not uniformly bad, and actually improved productivity in slightly more than half of the cases. The negative effects of perfor- mance contracts were caused by the large losses associated with poor design. SUc- cessful performance contracts featured sensible targets, stronger incentives, longer terms, and managerial bonds and were in more competitive industries. 13 CORPORATE GOVERNANCE IN CHINA Directors of enterprises that had entered into these contract re- sponsibility systems were given greater control over their enterprises' operations in return for meeting profit remittance targets. Many con- tracts also gave enterprises greater autonomy over sales and permitted managers to give bonuses to their employees and to hire temporary labor. Beginning in 1986, most newly hired workers in SOEs were given fixed-term, usually three-year, contracts. This measure was intended to put an end to the "iron rice bowl" system, under which workers were effectively guaranteed the right to keep their jobs for their entire careers, regardless of their performance. Under this new system workers whose performance was unsatisfactory could, in principle, be terminated when their contracts expired. The new system was also expected to increase labor discipline and strengthen performance-based incentives for work- ers. In practice, however, workers were rarely terminated and fixed- term contract renewal became largely automatic (Byrd 1992, p. 8). A more important development occurred in the mid-I 980s, when the government gave managers the authority to rationalize their work force by allocating surplus labor from production to other tasks or to training. In 1992 a government directive stipulated that contracts under the responsibility system could give managers additional autonomy, including the rights to make production decisions, negotiate prices for outputs and inputs, purchase goods and materials, make investment decisions, hire workers, and determine wages and bonuses. The contracting responsibility system achieved some success in that it increased the autonomy of SOE directors and improved their ability to manage effectively. It might have provided incentives for good performance, but it failed to penalize bad performance (Pannier 1996, p. 15), although some elements of hard budget constraints be- gan to emerge during this period.3 In relation to funds for investment in SOEs, budgetary financing and subsidies began to give way to bank financing and to financing from retained profits. Since 1979 bank loans have increasingly replaced budgetary grants, and interest charges have gradually increased. In 1983 the government formally established the PBOC as the country's central bank by removing its commercial banking activities. Four specialized state-owned banks were created to take over the functions involved in financing enterprises. 3. For more details about performance contracts in general see World Bank (1995), and for more details about performance contracts in China specifically, see Shirley (2000). 14 EVOLUTION OF MECHANISMS China has made progress in curtailing subsidies to SOEs: budget- ary subsidies fell from 7.5 percent of gross domestic product (GDP) in 1992 to 2.3 percent in 1994. However, other forms of subsidies con- tinued, including soft loans from state banks and subsidies from local governments, which control the majority of SOEs (Broadman 2001 b; World Bank 1996, p. 15). The Bankruptcy Law for SOEs was enacted in 1986 and became effective in 1988. In 1991 the Civil Procedures Law introduced rudimentary provisions for the bankruptcy of legal persons in general. Nevertheless, the period 1988-93 averaged only 277 bankruptcy cases per year. The government introduced incentives without making fundamental changes in the ownership of SOEs. However, it allowed new forms of ownership to develop. Township and village enterprises (TVEs) domi- nated the growth of the nonstate sector throughout the 1980s. "Ap- pearing from nowhere," as Deng Xiaoping was reported to have said in 1987, TVEs became important players in the economy. China's market reforms during this period also resulted in the emergence of a signifi- cant private sector. Private business was revived after the Cultural Revo- lution as a quick way to respond to the mounting pressures of unemployment and economic stagnation. It was first allowed on the fringes of the economy and was initially regarded as a supplement to the state and collective sectors (Gregory, Tenev, and Wagle 2000). For- eign investment, especially foreign direct investment, began to flow into China in the early 1980s, when it started to open up its economy to foreign investors. Investment from Chinese expatriates, mostly from Hong Kong, dominated the initial flow of foreign investment. The competition from these new ownership forms has become an important component of market control over SOEs. More important, the coexistence of various ownership forms and growing state enter- prise autonomy have created the conditions for a hybridization of state and nonstate enterprises. This hybridization has become a dis- tinct feature of China's market-oriented reforms. The process has taken the form of breaking up existing enterprises to form "secondary legal entities" (or subsidiaries), often disguised as collectives; joint ventures with foreign and/or domestic partners; limited liability companies; and joint stock companies.' This has been one way for SOE managers to gain further autonomy from supervising government agencies. 4. Over time, hybridization has resulted in complex organizational structures (see Broadman 2001b). 15 CORPORATE GOVERNANCE IN CHINA Initial experiments using stock to raise funds also led to some ownership diversification. In 1984, 11 SOEs became shareholding enterprises though a process called gufenhua or shareholding trans- formation. By the late 1980s another handful of SOEs had undergone gufenhua. Stock exchanges emerged in a number of cities, and in 1990 and 1991 the first official stock exchanges were established in Shang- hai and Shenzhen as an experiment. 1993-Present: Reemergence of the Corporate Form. The period since 1993 has been marked by important changes in China's overall ap- proach to reforms. While experimentation continued, a coherent strat- egy of transition to a market system began to emerge. An important development was the reemergence of the corporate form, a relatively ideology-free concept that the government found useful for redefining the broad relationship between economic actors, including between autonomous economic entities and the state. The introduction of the corporate form was associated with further ownership diversification, rapid capital market development, and the beginnings of unified treat- ment of state and nonstate economic entities. This, in turn, created a demand for a broad legal and regulatory framework consistent with a rules-based environment. The new wave of economic reform started with Deng Xiaoping's visit to south China in 1992, when he called for a continuation of the reform effort. In November 1993 the Third Plenary Session of the 14th Party Congress issued the Decision on Issues Concerning the Establishment of a Socialist Market Economic Structure. This deci- sion outlined a 50-point agenda for economic reform to be under- taken through 1999, including some important policies regarding SOE reform (Broadman 1995, p. 27). Two points were particularly impor- tant. The first was the creation of a "modern enterprise system," with its corporate structure, governance, and management based on the principle of corporatization, and with provisions for full separation of the state's exercise of ownership rights from the enterprise's exercise of legal person property rights. The second encouraged the develop- ment of diversified forms of enterprise ownership, including "privately owned, individually owned, and foreign-invested" enterprises. The Company Law, promulgated in November 1993, provided the legal underpinnings for the concept of a modern enterprise sys- tem. The new legislation provided, for the first time, a firm legal foun- dation for the establishment and operation of companies. It provided 16 EVOLUTION OF MECHANISMS rules for the incorporation of all enterprises of different ownership types into limited liability and limited liability shareholding compa- nies and specified governance structures, rules regarding the transfer and sales of shares, and procedures for mergers and bankruptcy. The introduction of the corporate form built on the shareholding experiments of the 1980s and the existence of the two stock exchanges. Local governments operating under increasingly hard budget con- straints appreciated the feature of limited liability as an opportunity to distance themselves from continuously underwriting SOEs' liabili- ties. The government assuaged ideological concerns through such means as controlling state ownership stakes, not allowing state shares to be traded, and molding old institutions into the corporate form. It this way it could present the diversification of ownership through corporatization as a mechanism for the state sector to play a leading role in relation to other sectors at the enterprise level in a mixed economy. With the fast growth of listed companies in the 1990s, the corporate form became widely accepted, culminating in 1997 with the broadening of the official definition of public ownership to in- clude publicly held private companies. The government subsequently introduced a series of comple- mentary reforms to build the institutional mechanisms for control consistent with the corporate form. While increasing the autonomy of SOE management, the government was also seeking to strengthen the supervision of state property, but in ways consistent with the new form of enterprise autonomy. In 1994 the government issued supervision regulations that provided the legal basis for the emerg- ing network of state-owned bodies designed to supervise SOE prop- erty (Broadman 2001a; World Bank 1997, p. 23). The tendency, although not yet fully realized, was to move toward an indirect, del- egated form of control in line with the tenet of separation between ownership and management. The government introduced accounting reforms to ensure that owners, boards of directors, and managers were provided with reli- able information for monitoring company performance. In July 1993 the Ministry of Finance issued the Accounting Standards for Business Enterprises (ASBE). These standards embody general principles mod- eled on internationally accepted practices. They also distinguish be- tween taxable income and profits, and are thus designed to measure corporate efficiency and performance rather than revenue generation (World Bank 1997, p. 65). 17 CORPORATE GOVERNANCE IN CHINA Increased enterprise autonomy required corresponding changes in the banking sector, which was still an extension of the administra- tive apparatus. Furthermore, local governments had strong incentives to use the national banking system as a vehicle for localizing the ben- efits and socializing the risks of local investment projects. The bank- ing reform program consisted of four major components: (a) separating policy lending from commercial lending by setting up policy banks, (b) deregulating the banking sector and establishing new banks, (c) improving the legal framework of the financial system, and (d) devel- oping financial markets. The government established three policy banks in 1994, desig- nated as the main vehicles for policy-based lending in the future. In addition, it started to deregulate the banking sector and lower barri- ers to entry. This resulted in the establishment of new nonstate com- mercial banks. In 1996 the All-China Federation of Industry and Commerce, an association of private enterprises, created the Minsheng Bank, China's only national, private commercial bank. More foreign banks and financial institutions entered China's market, and some of them were permitted to conduct domestic currency transactions. In 1998 the PBOC underwent significant restructuring, aimed at reducing provincial and local government intervention in credit alloca- tion and monetary policy and improving the soundness of the financial system by strengthening financial supervision. The most important change was the replacement of the 31 provincial branches with 9 re- gional branches. The old provincial branch network had been based on administrative jurisdiction. Under that system, provincial governments had a strong influence over the decisions made by the provincial branches under their jurisdiction. The move from a provincially based branch system to a regionally based branch system was expected to minimize such influence and improve the central bank's independence. In 1998 the government also took a major step in the reform of credit allocation by phasing out the credit quota system that was ap- plied to the four state-owned commercial banks and replacing it with asset liability management. The new system applies to both fixed as- sets and working capital loans (World Bank 1999, p. 34). The credit plan has been a powerful tool for controlling the money supply and allocating credit to SOEs and high priority sectors. As the economy becomes increasingly market oriented and the nonstate banks make up a bigger share of the banking system, controlling the total credit of state-owned commercial banks has become less effective. 18 EVOLUTION OF MECHANISMS The 1997 Asian financial crisis raised concerns among policy- makers about the possibility of a banking crisis in China because of the large volume of nonperforming loans and the low level of capital. To deal with the problems of a large number of nonperforming loans in the state-owned commercial banks and the high leverage of SOEs, the government established asset management companies (AMCs) for the four state-owned commercial banks. Despite the large number of loss-making SOEs, labor concerns continued to prevent many bankruptcies. In 1996 the conflict between social security and creditors' rights became apparent in decree num- ber 492, which gave labor and pension expenses priority claims on land use rights even when the SOE had mortgaged these rights, not- withstanding the basic principles of the Bankruptcy Law. Policies de- signed to deal with the social dimensions of bankruptcy have the potential of exacerbating financial instability and underscore the ur- gent need to establish a modern welfare system (World Bank 2001a). Both the central and municipal levels have made progress in de- linking social safety net functions from commercial operations by pooling pension, unemployment, and health obligations and transfer- ring them to government agencies. At the central government level, the process of unbundling took off in earnest in March 1998 with the creation of the Ministry of Labor and Social Security. The ministry inherited responsibility for social insurance for urban workers from the Ministry of Labor and authority over social insurance for civil affairs agencies, rural insurance, social security insurance, and medi- cal insurance from other ministries. The Ministry of Labor and Social Security was established with a view to introducing a comprehensive new social security system and facilitating welfare transfers through the redistribution of resources. Following the 1997 State Council Decision on the Establishment of Unified Pension Insurance for Enterprise Employees, the government has adopted a three-pronged approach to pension reform, namely: (a) a mandatory pooled fund to include all SOE employers administered by cities or provinces, (b) mandatory individual accounts managed by cities or provinces and funded by employer and employee contribu- tions (transferable and fully vested in 15 years and fully funded after 40 years), and (c) voluntary supplemental accounts set up by enter- prises. In February 2001 the government established the National Social Security Fund and the National Social Security Council to create, in effect, a supplementary pension system. The National Social Security 19 CORPORATE GOVERNANCE IN CHINA Fund will be professionally managed. There are plans to sell state- owned shares of listed companies to raise funds, in which under an initial public offering (IPO) the government would sell its shares (up to 10 percent of an IPO), and these proceeds would be used to fund the National Social Security Fund. This initiative would establish an important potential link between social security reforms and capital market development. The implementation of these plans has encountered a number of obstacles and issues. To begin with, the implied pension deficit has been rising. The existing system has not generated enough reserves nationwide for the transition to a fully funded system, and because of mounting arrears many SOEs have withdrawn from current pension systems. In addition, a comprehensive national system is still not in place because the central government offered cities a choice of differ- ent options, and because economic growth rates and labor market conditions vary across regions. The provinces and cities remain in charge of all new programs. Administration is another issue. The National Social Security Council will administer some national pro- grams, local social security offices will administer others based on a combination of national and local regulations, and enterprises will continue to have a role in pension administration. Finally, because of pressure from financial institutions with stock market exposure, the government had to suspend sales of state shares in IPOs to fund the National Social Security Fund. Agency Costs of Local Government Ownership and Enterprise Autonomy The agency costs associated with local government ownership and enterprise autonomy have been an important factor in determining the speed and direction of the market reforms that led to the current emphasis on corporate governance. Because of China's distinctive approach to reforms, local governments emerged as dominant own- ers and powerful regulators of companies tinder their jurisdiction. While bureaucratic entrepreneurialism at the local level generated much of the growth dynamism in the early years of reforms, ten- sions between powerful local incentives and national interests have been growing. One area where such conflicts of interest have become apparent is the national banking system. Local governments have used their power to influence credit decisions in order to localize the benefits 20 EVOLUTION OF MECHANISMS and socialize the risks of investment projects. To some extent the high level of nonperforming loans in the banking system reflects this dis- crepancy in incentives. Local protectionism has also become rampant. Local governments have been enacting internal trade barriers through their tax and price policies. This has resulted in higher prices, less efficient investment, and excess capacity in many sectors. Interprovin- cial trade has fallen from 37 percent of national retail trade in 1985 to about 25 percent today. Despite a huge expansion of the national trans- port infrastructure, the average distance traveled by a freight ship- ment fell from 395 kilometers in 1978 to 310 kilometers in 2000. Some economists have argued that double-digit export growth partly reflects local companies' inability to sell domestically, just as China's huge inflows of foreign investment partly reflect the diversion of much of the country's savings into inefficient state enterprises (World Bank 2001b). Local protectionism is also evident in discriminative employ- ment policies in almost every large Chinese city, and results in ineffi- cient allocation of labor resources. Interprovincial investment has been deterred by biases in the ju- dicial system. Prosecutors and judges overwhelmingly favor compa- nies in their districts. According to an analysis by Pei Minxin, a scholar at the Carnegie Endowment for International Peace, a local litigant enjoys a two-to-one advantage in Chinese courts over a nonlocal party (May 1999). Even when a local company is found guilty, it often can escape the consequences, because local authorities have appointive and financial power over judicial and law enforcement departments, and may obstruct the enforcement of court judgments (Yang 2000). Many of the reform initiatives at the national level are intended to reduce the negative impact of such agency costs related to local gov- ernment ownership and development incentives. Another problem of China's market-oriented reform was how to increase management's autonomy while making managers account- able to the state as the owner of the assets, or in other words, how to control the agency costs of enterprise autonomy. Over time, SOE re- forms have resulted in a significant degree of insider control as SOE managers have gradually acquired considerable discretion over the use of state assets. The agency costs of this increased autonomy have manifested themselves in various incentives for managers to maintain or acquire private benefits of control through on-the-job consump- tion and other rents related to investment and expansion. Among these incentives the tendency for overinvestment is perhaps the most impor- tant from an economic standpoint. 21 CORPORATE GOVERNANCE IN CHINA In the Chinese context, the incentive to invest has been enhanced by the motivation of SOE managers to gain further autonomy from their supervising agencies, typically through a series of organizational transformations, for example, by breaking up existing enterprises to form subsidiaries, joint ventures with foreign and/or domestic part- ners, limited liability companies, or joint stock companies. Additional motivations could include the possibility of undertaking new business opportunities without losing existing connections to and benefits from the state, shifting bad debts and surplus labor burdens onto parent companies, and so on. The coexistence of different ownership forms has created addi- tional incentives and opportunities for managers to realize private benefits from their control over state assets. With the rapid develop- ment of the nonstate sector, managers or their relatives and friends often have their own businesses, which provides opportunities for di- verting state assets to private benefits. A large body of anecdotal evi- dence indicates that asset stripping, or siphoning resources into structures where the controller has both majority control and income rights, is widespread. Furthermore, the "grafting" of nonstate prop- erty onto the state sector also offers opportunities for asset stripping, for instance, by using the appraisal and valuation process to form joint ventures or using bankruptcy to liquidate state assets at low prices and divert them for private use, which explains why some enterprises are enthusiastic about bankruptcy. Many key aspects of China's approach to market and SOE re- forms have been partly motivated by the need to control such agency costs. For example, initially and for some time thereafter enterprise managers were given the right to use state assets and to enjoy some of the income generated, but no formal rights to dispose of state assets. Also, the sequencing of market development, with product markets being allowed first and factor markets later, played an important role in controlling agency costs (Naughton 1995). This discouraged redis- tribution, because the only way to transform factors of production into income was through new production and not through specula- tive activities. In addition, siphoning resources into privately controlled activities was discouraged by discriminating against private sector activities. Also, activities that are particularly prone to and effective in redistribution and rent-seeking, such as distribution and trade, are still out of the reach of most private entrepreneurs. Furthermore, the Communist Party's control over the rights to appoint and dismiss top SOE managers may have served as an important counterbalance to 22 EVOLUTION OF MECHANISMS managerial discretion (Qian 2001). While this type of control has other distorting impacts on managerial incentives, it may have restricted the opportunities for egregious asset stripping, as insider control in China has never reached the degree it attained in Russia and other transition economies (box 2.2). BOX 2.2 EVOLVING ROLE OF THE COMMUNIST PARTY IN SOE MANAGEMENT The Communist Party's role in SOE management has changed over time. In the prcreform period the party secretary of the factory was the ulti- mate decisionmaker and managed the factory's day-to-day operations. Managers have been given greater autonomy in managing factory op- erations since 1984. Even though early reforms gave firms greater lati- tude in making decisions, the question of who would exercise this power remained unanswered. The general trend of reform since the mid-i 980s has been to have the factory director become the ultimate clecisionmaker and represent the enterprise when dealing with outside agencies. In 1986 the government issued the Regulations on the Work of Factory Directors in State-Owned Industrial Enterprises, which gave factory directors final operational authority. Most important, directors were given decisionmaking power over personnel within their enter- prises. However, SOEs' party committees still played important roles, especially as concerned personnel issues. In theory, managers in the corporatized SOEs are either elected or appointed by the board of directors, but in practice, they are more often appointed by the Communist Party's Organizational Department. For large SOEs or enterprise groups, the general manager is often also the secretary of the party committee. The party's standing committee usu- ally consists of the general manager and senior managers, and is the ultimate decisionmaking body for important issues. The Communist Party Central Committee or provincial committees still control and de- termine the appointment, promotion, or dismissal of senior managers of large SOEs or enterprise groups. The central government's and party's willingness to permit decen- tralization and delegate control power to enterprises is probably related to the fact that the party's control over appointments and dismissal is being preserved. However, direct control and ownership by the party have been largely eliminated. In late 1998 the government ordered all party and government administrative organs to sever their links with the enterprises they control (World Bank 1999, p. 32). 23 CORPORATE GOVERNANCE IN CHINA China has found it difficult to control the agency costs associated with the incentives to overinvest. Initially, inefficiencies were limited by the low level of development and the niches created by the com- mand economy: identifying positive net present value projects was relatively easy. Competitive product markets and competition from the nonstate sector also played a positive role. However, the dynamics of the process were unsustainable. Over time inefficiencies tended to accumulate and began to manifest themselves in overcapacity and a high level of nonperforming loans in the banking system. The need to restructure in response to these challenges has prompted additional reform measures: the corporate form offered the promise of limited liability to the government; the local govern- ments found that privatization was the only way to staunch the bleed- ing of small and medium state-owned and collective enterprises; the listing of enterprises on the stock exchanges seemed like a conve- nient way to raise funds for restructuring; and the financial system is also being reformed to introduce market principles in this area. These are some of the initiatives that led to a diversified ownership struc- ture and have, in the process, transformed the initial problem of how to make managers accountable to the state as an owner into one of corporate governance. Current Approach The system for managing state assets has been evolving in the direc- tion of separating ownership and regulation, as well as streamlining the administration of state ownership rights. The current oversight system is not completely unified and is still undergoing experimenta- tion. Reforms aim at consolidating fragmented oversight and the coil- flicting goals of stakeholders with the intention of providing a single reporting line to SOE managers by integrating party and government functions into one agency. Many experiments have taken place, and decisions have reflected compromises made between different centers of power within the government (box 2.3). The system for managing state assets at the central government and local government levels is evolving differently, with several large city governments more advanced than the central government in In- troducing a more streamlined and unified system of SOE oversight. The central government has been encouraging local governments to experiment with different approaches to managing state assets and to 24 EVOLUTION OF MECHANISMS BOX 2.3 EVOLUTION OF THE STATE ASSET MANAGEMENT SYSTEM Between 1987 and 1998 the government started the process of separating the management of ownership from the management of government. In 1987 Shenzhen, one of the earliest special economic zone cities, set tIp the nation's first specialized state assets management institution, the Shenzhen Municipal Investment Administration Company, to manage all the city's assets. In 1988 the central government established a separate state asset management agency, the National State Asset Administration Bureau, which reported to the State Council and was under the guidance of the Ministry of Finance. Both the bureau and the ministry were named as the general representatives of the owner of state assets. In the early 1 990s, almost at the same time, some big cities, includ- ing Beijing, Shenzhen, and Shanghai, began to deepen their state asset management system by separating regulatory matters and ownership oversight from governance and management. For example, in 1992 Shenzhen set up the State Asset Management Commission, which spe- cializes in the supervision and governance of the ownership of state assets. The former Municipal Investment Administration Company has been transformed into the State Asset Management Company to man- age state assets. At the central level, the efforts to gradually eliminate the ownership functions of line ministries and to transfer their powers to the National State Asset Administration Bureau and the Ministry of Finance to estab- lish a unified and separate agency responsible for state assets met with strong resistance from line ministries. Several comprehensive government ministries, such as the State Economic Restructuring Commission and the State Economic and Trade Commission (SETC), also tussled with each other and the bureau to become the unified agency for asset man- agement. In addition, debate on what models of state asset management system China should follow has been ongoing. A compromise was reached in 1998 during the government reorganization, whereby, on the one hand, most line ministries were merged into the SETC and their ownership func- tions were removed, and on the other hand, the bureau was dissolved and merged into the Ministry of Finance. Ownership functions were then sepa- rated and transferred to different comprehensive government ministries. From the point of view of a recombination of regulatory and ownership functions, this was a retrograde step. SOURCE: Zang (2001). 25 CORPORATE GOVERNANCE IN CHINA continue to follow the path of separating ownership from the government's regulatory function. Several relatively developed cities, including Shanghai, Shenzhen, Qingdao, Wuhan, and Xiamen, pur- sue a system that can be referred to as the Shenzhen-Shanghai model. In the Shenzhen-Shanghai model the municipal government is autho- rized by the next highest level of government to manage the state as- sets assigned to the city, and a three-tier management structure is set up to manage the state assets. The first tier is a state management com- mission, which consists of major city-level leaders and heads of relevant municipal government departments, and functions as the general board of directors for the city's state assets. The second tier comprises holding companies and enterprise groups authorized by the government to man- age state assets. These are special corporations that enjoy ownership rights to the state assets under their jurisdiction. The third tier consists of enterprises wholly or partly owned by the state. At the central level the state management system is still fragmented, with many government ministries involved in managing SOEs. The main features of the current system for oversight of SOEs at the central level are described in box 2.4. From the standpoint of facilitating effective SOE oversight, the system has the following major defects. First, own- ership, and as a result responsibility, have been fragmented. Under the current system ministries can issue orders to enterprises according to their role, but none is fully responsible for the results. Second, the com- prehensive ministries, such as the Ministry of Finance, the State Eco- nomic and Trade Commission (SETC), the State Development and Planning Commission, and the PBOC, are also the most important regu- lators of the whole economy. Thus the potential for conflicts of interest is considerable. These government agencies could apply policies and resources in a way that was favorable to SOEs and deter the creation of a level playing field for nonstate enterprises. Third, from the viewpoint of enterprises, they now face multiple principals in the line ministries. Transaction costs increase because they need to negotiate different man- agement matters with different ministries. However, enterprises can also avail themselves of loopholes and escape from owners' supervision and control more easily. Finally, too many SOEs exist for direct control to be effective. Currently 4,000 enterprises are directly under the control of the central government, including 600 key enterprises and groups, making effective control a difficult task. At the Chinese Communist Party's 15th Congress held in Sep- tember 1997, the Central Committee endorsed both the corporatization 26 EVOLUTION OF MECHANISMS BOX 2.4 SOE MANAGEMENT AT THE CENTRAL GOVERNMENT LEVEL The SOE oversight function at the central level is fragmented, as dem- onstrated by the following: * The Financial Working Commission, the Economic Working Commission, and the Ministry of Personnel manage the appointments and dismissals of senior officers. Both the commissions are newly estab- lished subordinates under the Chinese Communist Party Central Com- mittee, while the Ministry of Personnel is a government ministry under the State Council. These three organizations are responsible for the per- sonnel issues of different groups of SOEs. The Financial Working Com- mission is responsible for state financial institutions, including state banks, insurance companies, and security companies; the Economic Working Commission is responsible for 39 of the most important and largest enterprise groups; and the Ministry of Personnel is responsible for other important and large enterprises or groups. * The Ministry of Finance is responsible for the administration of capital, for example, the definition and registration of property rights of state assets and the supervision of state asset evaluations and transac- tions, along with the supervision of revenue collection. * The newly established Special Inspector's Commission is respon- sible for financial auditing. Even though the commission is headquar- tered in the Ministry of Personnel, it is an independent agency that reports directly to the State Council. Usually one special inspector works with five or six assistants and is in charge of five or six enterprises. Special inspectors are supposed to audit enterprises at least twice a year and are not allowed to interfere in enterprise management. After auditing, spe- cial inspectors provide an assessment of the management's performance and make suggestions about the retention, promotion, or dismissal of incumbent senior officers. Special inspectors' reports are sent to the SETC for review before being passed on to the State Council. * The SETC and the State Development and Planning Commis- sion are responsible for the supervision and control of enterprises' busi- ness activities, including financing, investment, sale of assets, and internal restructuring. SOURCE: Zang (2001). 27 CORPORATE GOVERNANCE IN CHINA of large SOEs and the restructuring of small SOEs, and decided to accelerate the speed of reforms. The central component of the reform program was zhuada, fangxiao, or "grasping the big, and enlivening the small" (see Broadman 2001 a for more details). The notion of grasp- ing the big involves two related sets of reforms. First, the government is creating a number of large enterprise groups with extensive cross- ownership by encouraging mergers in core industries. Second, the government is encouraging those enterprises performing better to be listed on the stock exchange to promote ownership diversification and raise funding for restructuring. By the end of 2001 about 1,200 com- panies were listed on the Shanghai and Shenzhen stock exchanges. The concept of enlivening the small provides for further experimenta- tion with ownership reform in small and medium SOEs. The govern- ment has used various mechanisms to let go of small enterprises, including restructuring, entering into alliances, encouraging mergers and acquisitions (M&As), and forming shareholding companies (World Bank 1999, p. 3 1). In August 2001 the SETC announced that by the end of 2000 more than 81 percent of the 63,490 small SOEs that had existed at the end of 1996 had been reformed, mainly through sales. Conclusion China's current emphasis on corporate governance is largely a policy response to the issues that emerged following partial reforms in the state enterprise sector. Market reforms have transformed the problem of state control over the agency costs of enterprise autonomy into one of corporate governance. The current approach takes the process of emancipation of economic agents from the state and politics to a higher level by focusing on the institutional framework within which partici- pants can shape the governance structure of production through mar- ket transactions. While the potential for institution building is vast, the effectiveness of the new institutions of corporate governance is likely to be limited in the context of dominant state ownership and of party control over managers. However, based on the dynamics of China's reform experience, the new institutions will likely facilitate changes in the areas of state ownership and political control over eco- nomic activities. Such changes are necessary for China to develop fully functioning factor markets, without which the effectiveness of corpo- rate governance institutions will remain limited. In this context, the current emphasis on corporate governance can be seen as a prelude to China's final stage in its transition to a market economy. 28 3 The Corporate Governance of Transformed Small and Medium Enterprises This chapter examines the main corporate governance issues arising in the process of insider-centered ownership diversification of small and medium SOEs. The closely held nature of these corporations is associated with the absence of an active market in shares, which pre- cludes reliance on public monitoring. This chapter therefore focuses on the role of employees, creditors, and outside equity investors in such corporations. Ownership Transformation and Emerging Governance Issues The ownership diversification of small and medium SOEs has taken a variety of forms (see table 3.1).' In some cases SOEs have been sold directly to individuals or private firms. While such direct sales face resistance from insiders, the process is becoming politically more ac- ceptable, especially since the 15th National Congress. Progress in so- cial security system reform in recent years has also made it less difficult for local governments to sell their enterprises to private owners. Be- cause of their heavy reliance on bank financing since the mid-1980s, 1. In Sichuan, for example, the provincial government embarked on a restructur- ing program in 1994. By the end of 1998, the process had been completed for 69 percent of the 42,681 firms in the program. Among those transformed, 45.1 per- cent became employee-owned companies, 13.1 percent became employee-owned cooperatives, 14.3 percent were sold, 7 percent were contracted out to individu- als, 8.5 percent were leased out, 7 percent filed for bankruptcy, and 5 percent were absorbed by other firms. 29 CORPORATE GOVERNANCE IN CHINA TABLE 3-1 TRANSFORMATION OF SMES BY 2OOO No. of Trans- Method (%) enter- formed Region prises (%) R M L C ]SC Sale B Other Coast" 17,629 83 17 13 11 9 22 8 8 12 Centralb 20,713 83 14 11 14 9 22 9 11 10 West' 21,068 80 20 12 9 8 19 9 11 12 NoTEs: R-restrUcturing, M-merger, L-leasing, C-contracting, JSC-joint-stock company, B-bankruptcy. a. Liaoning, Hebei, Beijing, Tianjin, Shandong, Jiangsu, Shanghai, Zheijang, Fujian, Guangdong. b. Heilongjiang, Jilin, Shanxi, Henan, Hubei, Anhui, liangxi, Hunan, Hainan. c. Inner Mongolia, Shaanxi, Ningxia, Gansu, Qinghai, Xinjiang, Tiber, Sichuan, Guizhou, Yunnan, Guangxi. SOURCE: SETC. most SOEs are overindebted and many are insolvent (see table 3.2). An assets evaluation organized by the government in the mid-1990s found that nearly 40 percent of the 302,000 nonfinancial SOEs were empty shells, in the sense that their debt obligations exceeded their assets (Wu 1998, p. 26). The percentage would certainly be much higher if assets were recorded at their market value and off-balance- sheet liabilities were taken into account. Although rare, liquidation procedures have been used to transfer ownership rights over physical assets such as land, buildings, and equipment to nonstate owners. Debt for equity conversions are beginning to introduce new owners to small and medium SOEs. In the vast majority of cases, the corporatization of small and medium SOEs has been accompanied by the allocation of ownership rights to insiders such as managers and employees. Emerging Ownership Patterns. The diversification of the ownership of small and medium SOEs has been driven by local governments, largely in response to the poor financial performance of firms under their control (see box 3.1). In 1995, for example, 72 percent of the firms owned by local governments were in the red. In Zhucheng, a comprehensive audit of state assets in April 1992 revealed that of the 150 enterprises belonging to the municipality and responsible for their own economic profits and losses, 103 were sustaining losses. The preference given to employee ownership reflects a number of factors: financial problems, which make direct sales difficult; de facto 30 TRANSFORMED ENTERPRISES TABLE 3.2 SOME INDICATORS OF FINANCIAL PERFORMANCE OF SMEs Loss- Net loss as making share of gov. Debt No. of enterprises revenues Ratio Region companies (%) (%) (%) Beijing 2,046 27 4 62 Tianjin 951 55 10 77 Hebei 2,125 29 3 76 Shanxi 1,480 26 7 75 Inner Mongolia 662 23 n.a. 70 Liaoning 1,946 34 6 69 Jilin 1,398 33 8 78 Heilongjiang 1,476 37 4 83 Shanghai 1,439 31 2 61 Jiangsu 1,833 37 3 62 Zhejing 1,067 39 2 55 Anhui 892 50 4 71 Fujian 1,115 40 3 63 Jiangxi 2,006 45 8 79 Shandong 1,437 28 1 72 Henan 2,043 33 5 77 Hubei 2,642 35 5 77 Hunan 1,901 45 7 75 Guangdong 2,480 30 3 70 Guangxi 1,560 51 5 66 Hainan 182 51 5 87 Chongqing 588 61 8 74 Sichuan 1,424 41 5 72 Guizhou 951 41 5 75 Yunnan 1,024 54 5 65 Tibet 199 33 10 36 Shaanxi 1,232 50 9 78 Gansu 984 29 7 60 Qinghai 350 42 26 86 Ningxia 138 45 11 48 Xinjiang 1,047 49 11 74 SOURCE: SETC, China Statistical Yearbook 2001, and authors' calculations. control by insiders, acquired during the process of enterprise reform; systemic financial problems, which call for systemic solutions locally as opposed to an enterprise-by-enterprise approach as in the case of direct sales and bankruptcies; and political feasibility, particularly with respect to procedures, that is, less strict evaluation and pricing of assets 31 CORPORATE GOVERNANCE IN CHINA BOX 3.1 PRIVATIZATION OF SMALL SOEs: SOME CASES Pricing Formula of an Industrial City The city uses the formula M = A - BI - B2 - B3 + Cl + C2 + C3 where M is the government's reservation price; A is the net assets con- firmed by evaluation; Bi is the number of redundant workers multiplied by Y 8,000 per person; B2 is the number of senior retirees multiplied by Y 6,000 per person; B3 is the number of other retirees multiplied by Y 12,000 per person; C1 is the value of preferential tax policies enjoyed by the enterprise; C2 is the value of land use rights; and C3 is the value of intangible assets, such as brand names and goodwill. If M is less than zero, the implication is that the firm should be liquidated rather than sold. Getting Workers' Agreement before Selling to a Manager The Agricultural Machinery Manufacturing and Repair Factory started making losses when its fiscal subsidy was removed and stopped production in 1990. A new manager, Mr. Wen, was appointed to turn it around in 1991, and did so by introducing the TVE management system. However, in 1993 the factory was in trouble again when RMB 6 million in receiv- ables became uncollecrable because of macroeconomic conditions. In 1994 the factory stopped production again and each worker received a monthly income of only Y 60. After demonstrations by workers, in 1995 the munici- pal government decided to sell the factory to the manager; however, it took no less than 20 meetings with workers before the government succeeded in getting their agreement to the sale. An asset evaluation concluded that the factory had a total of RMB 9.14 million in assets and RMB 7.27 million in debts. The manager agreed to buy the factory for RMB 1.94 million, which he was allowed to pay in installments over two years. Ensuring That Employees Can Afford the Price The Textile Machinery Factory, a medium-sized SOF, was founded in 1971 and ran into financial difficulties in 1994. At the time of can be politically acceptable and the social impact of privatization can be less severe. As aptly summarized by local government officials in jinhua, employee ownership satisfied three constraints: government officials' fear of making political mistakes, managers' fear of losing power, and workers' fear of losing jobs. 32 TRANSFORMED ENTERPRISES privatization in 1997 it had suffered losses for three years. An assets evaluation showed RMB 24.7 million in total assets and RMB 15.1 million in total debts. A number of deductions were made to enable employees to rent the land use rights for RMB 7.57 million, to contrib- ute RMB 0.42 million to the pension pool for retired workers, and to provide RMB 1.34 million as an employees insurance fund. Employees then paid RMB 0.197 million to buy the factory, and the new firm raised RMB 2 million in additional funds by issuing new shares to employees. Escaping from the Sinking Ship by Small Boat The Paper Factory was established in 1969 and has been in difficulties since then, partly because of a heavy social burden and large volume of social assets. In the mid-1990s the municipal government decided to carve out productive assets from the factory to set up a new entity that would be privatized. The new entity was evaluated and its net worth was set at RMB 1.34 million. All 350 employees were asked to sub- scribe to shares, and compulsory minimum amounts of contributions were set at Y 6,500 per person for top managers, Y 5,500 each for mid- level managers, and Y 4,500 each for other employees. The privatized new entity was incorporated as a limited liability company, while the old factory remained as an independent legal entity. The old factory rented its social assets to the new company to collect cash so that it could look after retirees and injured and sick employees. Privatizing through Leasing to Bypass Liquidity Constraints The net worth of the Pharmaceutical Factory was determined as RMB 9.37 million, and the employees collectively leased the factory from the Yuhuan county government for 10 years. The government required them to put down a 10 percent deposit and maintain all the jobs un- less some workers chose to leave. The employees were supposed to buy the factory within 10 years through installment payments made from their profits. Different localities have followed similar procedures. Typically, managers and employees first put forward plans for employee owner- ship for discussion with and approval by the municipal government. An outside accounting firm carries out the valuation, at best with only formal independence from the provincial government. In some 33 CORPORATE GOVERNANCE IN CHINA localities, such as Zhucheng and Jinhua, land and social assets were excluded from the valuation to bypass the insiders' wealth constraint to assuming a controlling position. In such cases, enterprises would typically lease the land from the government. In other localities, for instance, in Shunde, land was included in the valuation. The inclusion of land in the deal has a significant implication for a firm's ability to borrow in the future. In any event, land was the asset most at risk of being undervalued. However, as local governments typically had to take over the firm's net debt if its net assets were negative, this pro- vided some limit to the extent to which certain assets could be under- valued. The exact structure of ownership was agreed upon during a series of discussions at conferences with employee representatives. Decisions about the allocation of shares were often based on such factors such as number of years of employment and rank in the mana- gerial hierarchy. Additional share purchases were allowed. Typically there were no stipulations as to how big the difference in individual shareholdings, particularly between managers and ordinary workers, should be. The process of ownership diversification reflected a combination of bargaining, coercion, and persuasion. In some cases, local govern- ments had to give managers and employees a put option to make the risks more palatable. Proceeds from the purchase of shares have often been given back to enterprises in the form of government loans, which could also be viewed as a performance bond on the government's put option. Explicit protection for employees was often included in the agree- ments; for example, in Shunde no more than 5 percent of the work force could be fired in the three-year period following transformation. Under the new ownership structure employees emerged as the most important shareholders. Jinhua is a typical case. In the trans- formed enterprises in Jinhua natural persons held 76 percent of the shares. Second in importance were shares owned collectively by em- ployees. Most of the enterprises that changed their ownership emerged as 100 percent employee owned. The share of senior management was relatively high, and on average amounted to 20 to 30 percent of all employee shares. In principle, senior managers were allowed to hold significant blocks of shares, but they often opted not to do so because of concerns that other employees might disapprove. How- ever, in a number of enterprises top managers were able to amass significant blocks of shares, in some cases exceeding 50 percent of all shares issued. Overall, the initial ownership structure was character- 34 TRANSFORMED ENTERPRISES ized by a significant dispersion of ownership. The largest sharehold- ers held between 3 and 25 times as many shares as the smallest share- holders and 2 to 10 times as many shares as the average shareholder. State and legal person shares were in the minority. The size of net assets seems to explain the presence of state shares, which were found only in the largest enterprises in terms of net assets (and some of the most profitable). Corporate Governance Issues. New legal and organizational forms consistent with the corporate form were introduced without disman- tling the old representative bodies. The conference of shareholding employees, the board of directors, and the board of supervisors were established as new governing structures. An important issue was how to divide functions between these new institutions representing the shareholders and the traditional organizations of social control, such as the workers' congress, the party committee, and the trade unions. The standard approach has been not to disband the traditional instruments of social control, but instead to make them compatible with the new management structure by introducing new procedures. This was typically achieved by combining the leadership or functions of various institutions. The usual practice was to combine sharehold- ers' and workers' congress meetings and to have the same person serve as chair of the board of directors and secretary of the party commit- tee. Many enterprises held joint meetings of various representative bodies. The intent was to streamline the administrative structure, re- duce overstaffing, and avoid duplication. In reality, meetings prolifer- ated and considerable confusion arose about the division of functions and methods of decisionmaking, especially initially. Determining which issues were operational, to be decided at shareholders' meetings using the one share one vote method, and which issues concerned employee benefits and had to be decided at the workers' congress using the rule of one person one vote was often difficult. Two main approaches emerged to deal with the situation. One was the design of detailed internal procedures for new institutions. In most enterprises shareholders' meetings and conferences of employee representatives (workers' congress) had to adhere to detailed internal regulations concerning the number of attendees, the information to be disclosed, the decisions to be made, and the voting method to be used. The activities of the board of directors were relatively standard- ized, while the functioning of the board of supervisors was generally 35 CORPORATE GOVERNANCE IN CHINA perceived to be relatively poor. Supervisory boards tended to meet less often than the boards of directors, and their activities were less struc- tured. Most of those interviewed believed that the main reason for this was that the chair of the board of directors tended to be more senior and had been at the enterprise longer than the chair of the board of supervisors. However, the situation also depended on the personal relationship between the two chairs and the former tradi- tions in the enterprise. The other approach was to resort to informal mechanisms to economize on decisionmaking costs. Many enterprises used meetings of large shareholders to review important proposals and make decisions. Often key financial information had to provided to the large shareholders first, before being disclosed, with their ap- proval, to other shareholders and employees. A typical contractual structure regulating internal rights and ob- ligations included a responsibility contract signed by the sharehold- ers' organization and the board of directors, while a traditional collective contract would be signed by the management and the trade union. A responsibility contract would usually cover profits, increases in net assets, and tax targets linked to salaries. To support these con- tractual arrangements, the disclosure of information became more important than in the past. Some enterprises even introduced trans- parency policies. In some enterprises shareholding employees routinely reviewed business entertainment expenses and evaluated the manage- ment. In some cases, however, enterprises had to limit the disclosure of information to prevent the leakage of business and technical se- crets. Overall, the channels through which shareholders/employees could monitor senior management increased and were often institu- tionalized by means of specific procedures. Despite some positive changes in enterprise behavior following corporatization and ownership diversification, serious issues emerged relatively early in the process in relation to incentives and governance practices. Surveyed enterprises reported the prevalence of a short-term outlook, manifested by excessive dividend distribution accompanied by a lack of direct links between profitability and income growth. As a result, enterprises were unable to accumulate sufficient resources for long-term growth. During the initial period, shareholding employees were prima- rily interested in the distribution of dividends. In the enterprises sur- veyed in jinhua and Zhucheng employees recovered their investments in three or four years. A common phenomenon among transformed 36 TRANSFORMED ENTERPRISES enterprises was an excessive distribution of dividends at the initial stage of reform. Well-performing enterprises distributed all their prof- its in the form of dividends. In some localities municipal govern- ments had to limit dividend distribution in the context of widespread deterioration of enterprises' economic performance. Even though dividend distribution began to slow down after shareholders had recouped their initial investments because of government interven- tion and deteriorating performance, wages continued to increase steadily and remained significantly higher than before the enterprises had been transformed. In the transformed enterprises in Zhucheng, for example, dividends amounted to about 25 percent of employees' average annual salaries, and employees were receiving about 2.4 times as much in total compensation as they had before the change, with- out a corresponding increase in enterprise profitability. This situa- tion highlights the lack of monitoring by creditors, who under normal circumstances have the incentives and the tools to monitor and con- trol excessive dividend distribution. Respondents also reported that once shareholding employees had recouped their initial investments, their incentives to monitor com- pany performance were reduced. Given the diffused ownership struc- ture, employees were not motivated to spend the required time and effort to inform themselves about factors affecting enterprise perfor- mance. In any case, ordinary employees in the surveyed enterprises felt that their influence on the enterprise's decisionmaking was lim- ited, either as workers or as shareholders. Some managers and techni- cal personnel who were shareholders could affect the enterprise's decisionmaking, but the average shareholder's power and benefits were insufficient to justify taking greater business risks. Moreover, in the context of China's monetary tightening in 1996 and the Asian finan- cial crisis in 1997, workers felt that the enterprise's economic perfor- mance depended on many factors, most of which were beyond their control. This further weakened incentives to monitor performance and participate in decisionmaking. Surveyed enterprises reported numerous instances where the dif- fused ownership structure was inconsistent with the actual distribu- tion of power and control over key resources. For example, in some enterprises human capital in the form of knowledge and business con- nections was the critical resource, and was concentrated in a few key managers and technical personnel. Without control rights these em- ployees had the incentives to quit the enterprise, withdrawing these 37 CORPORATE GOVERNANCE IN CHINA key resources in the process. In one company, some technical staff left their jobs to become private entrepreneurs, eventually becoming mil- lionaires. In another company, the "talented took away the technol- ogy and the client connections, left the enterprise, and competed with the original enterprise." The original enterprise languished. Under such circumstances the new ownership structure could not protect the firm's integrity. Giving controlling ownership rights to people with the power to withhold key resources is one way of ensuring the continued exist- ence of such enterprises (Zingales 2000). Finally, the low ownership concentration affected the efficiency of the decisionmaking process and caused problems related to missed business opportunities and low management efficiency. The heteroge- neous nature of the work force in most enterprises, combined with an institutional framework for decisionmaking that combined various organizational forms with different objectives (organizational and social), led to higher decisionmaking costs. Employees tended to view their shareholder rights primarily as a tool for enhancing their job security. Despite some efforts to downsize and streamline operations, the transformed enterprises did not lay off staff or reduce overall employment. Most of the enterprises actually increased the size of their labor force. A review of employment num- bers shows an insensitivity to overall market conditions and individual enterprise performance. According to data provided by the Zhucheng System Reform Committee, in 1992, before the enterprise reform, municipality-owned enterprises employed 15,624 people and the TVEs employed 35,105. By the end of 1998, the formerly municipality-owned enterprises had 15,686 staff and the TVEs had 39,712. Some sur- veyed enterprises in jinhua did not engage in large-scale cutbacks de- spite poor economic performance, and most enterprises actually increased the number of employees. While some enterprises did re- duce their staff numbers, this occurred mainly as a result of natural attrition. In most localities, the process of ownership transformation failed to produce a radical change in the relationship between enterprises and the government, although some positive changes have reportedly taken place. Local governments retained some key powers that should have been transferred to the new owners, of which perhaps the most important was the right to appoint enterprises' top management. In most cases the municipal government had to approve all appointments of senior managers. Surveyed enterprises presented numerous examples 38 TRANSFORMED ENTERPRISES of government behavior inconsistent with the autonomy of a priva- tized enterprise. For example, policies on taxable salaries and divi- dends were under the direct control of the supervising municipal government departments, and local governments continued to act as arbitrators in the case of internal conflicts and disagreements. Continued government involvement can be related to continued government ownership in some enterprises and to the contingent li- abilities in the form of explicit or implicit guarantees that the govern- ment continued to hold with respect to transformed enterprises. It also reflects the absence of effective monitoring by market players such as banks, who in normal circumstances would have the incen- tives and the instruments to restrain excessive wage and dividend pay- ments. The transformed enterprises themselves sometimes actively sought government support. Some enterprises continued to develop their relationships with local governments in attempts to use their administrative power to promote their own narrow interests. Trends in Ownership Structure. As a result of the aforementioned gov- ernance problems, a perception that the existing employee ownership structure was not conducive to the long-term development prospects of transformed enterprises was widespread. Local governments and enterprise management saw the solution as lying in the concentration of ownership, and initiated further ownership changes to move in that direction. In contrast with earlier reforms, the driving force behind the second wave of ownership transformation was management. Typi- cally, boards of directors would put forward plans to nurture large shareholders, primarily managers and highly regarded employees, to strengthen the driving force behind the enterprise. Plans for sources of new equity included personal savings, bank loans, and investments by other companies. The sense of crisis created by enterprises' poor fi- nancial condition often facilitated the acceptance of such plans. While the second wave of ownership transformation did not achieve radical changes in ownership patterns, it did result in some ownership concentration, a higher percentage of ownership by man- agement, and in some cases the introduction of outside investors. For example, in Jinhua the overall capital increase in the 201 enterprises participating in the second wave of ownership transformation was RMB 600 million, of which RMB 150 million was from investment by employees, RMB 250 million was from bank credit used to pur- chase shares, RMB 160 million was from allocations of accumulated 39 CORPORATE GOVERNANCE IN CHINA retained earnings, and RMB 40 million was mobilized from other cor- porations. Under the new ownership structure in Jinhua, in 57 enter- prises the percentage of shares held by the chair of the board of directors exceeded 10 percent and the percentage held by managers and board members exceeded 25 percent. Based on observed trends, the owner- ship structure is likely to evolve in the direction of a higher concentra- tion of ownership by management and the introduction of outside investors. Under the new ownership structure the primary challenge is safe- guarding the interests of minority shareholders, especially workers, against possible expropriation by managers who are also controlling shareholders. Another important problem is managerial entrenchment, that is, the difficulty of replacing incompetent or poorly performing managers who are also significant shareholders. Employee ownership complicates the corporate governance characteristics of insider own- ership. Employee entrenchment can reduce the corporation's flexibil- ity to adjust to changes in the environment if employees use their shareholder rights to pursue their narrow interests as employees, even to the detriment of the corporation. Subsequent sections will discuss the role of employees, creditors (banks), and private equity investors in alleviating these agency problems. Role of Employees Employees usually invest in firm-specific human capital and can be viewed as residual claimants in situations of financial distress. As such, they have a collective interest in monitoring the agency costs of equity, particularly with respect to important decisions that could affect the enterprise's long-term prospects. Contractual and legal rights usually protect employees' fixed claims, including in the case of bankruptcy. However, their firm-specific investments are often poorly protected by formal contracts and regulations. As a result implicit contracts, or in the words of Chinese economist Wu Jinglian, "promises made in the past" to provide some form of insurance for human capital investments often complement formal arrangements.2 2. The noncontractible interests related to firm-specific human capital investments draw a parallel between employees and shareholders. Even though both employ- ees and shareholders can bc seen as having implicit contracts with the firm, for- mal control rights are typically given to shareholders. The usual explanation for 40 TRANSFORMED ENTERPRISES Various forms of workers' participation in residual control or re- sidual income rights have emerged to provide additional protection for employees' firm-specific investments, particularly in the context of economic transitions. Thus employees' contractual, legal, and own- ership rights should be examined in parallel to analyze workers' role in corporate governance. Worker Participation in Corporate Governance. Chinese workers have a number of legal rights to protect their interests. Collective contracting and negotiations typically govern narrow employee in- terests related to compensation, firing, social benefits, working con- ditions, and so on. In addition, workers' congresses and trade unions have extensive rights to consultation and information regarding pro- duction plans, use of public welfare funds, and other matters that could affect employees' interests. In some types of enterprises, namely, limited liability companies with the government as a controlling share- holder and joint stock companies, trade unions have the right to organize workers to oversee and assess the virtues, ability, diligence, and achievements of the chair of the board of directors, general managers, and high-level management personnel. According to SETC regulations, SOE managers are obligated to report to the employee conference on various business-related entertainment expenditures every six months. In addition to the collective rights exercised through workers' congresses and trade unions, employees can be represented on boards of directors and supervisors. Articles 45 and 68 of the Company Law stipulate that a proper proportion of workers' representatives should be elected as board members in limited liability companies established with investment from two SOEs or two state investment holding enti- ties, or in state-funded companies. According to articles 52 and 124 of the Company Law, the boards of supervisors in limited liability companies and joint stock companies should also contain a proper proportion of workers' representatives. Employees are represented to such a practice is that employees' implicit contracts are more likely to be self- enforcing because employees are making continuous firm-specific investments and the firm wants them to do so (Gordon 1999). In contrast, shareholders contribute capital only infrequently, and as a result their implicit contracts are not self- enforcing, thereby creating the need for special governance mechanisms (voting) to provide credible protection against expropriation. 41 CORPORATE GOVERNANCE IN CHINA a significant extent on the boards of directors and supervisors of com- panies that have corporatized and transformed their ownership. A 2000 survey of 1,229 enterprises in Henan province revealed that 48 percent of the enterprises had employee representatives, typically union leaders, on their boards of directors, and 69 percent had employee representatives on their boards of supervisors. In a similar survey con- ducted in 1,669 enterprises in Hebei province, the figures were 92 and 98 percent, respectively. Employees can also participate in corporate governance in their capacity as owners. As discussed earlier, employee ownership emerged as the dominant form of ownership transformation of small and me- dium SOEs under the control of local governments, though some limi- tations exist on employee ownership in public shareholding companies. A July 1993 regulation issued by the State Economic Mechanism Re- form Commission set a limit of 2.5 percent on employee-held stocks in the stock-issuing. However, local regulations soon superseded this limit. In 1994 Shanghai promulgated the Experimental Measures on Issuing Employee-Held Stocks, which allowed 10 to 30 percent of employee-held stocks. That same year the Shenzhen Special Economic Zone released the Regulation on the System of Employee-Held Stocks, which allowed up to 30 percent of employee-held stocks. This pro- portion later rose to 50 percent and then exceeded this figure. Later other provinces, municipal cities under central administration, and autonomous regions followed suit in passing similar rules. In addition to individual employee shareholding, collective ve- hicles have emerged to exercise employees' shareholder rights. Since 1994 some enterprises have adopted an internal employee stockhold- ers' system and set up employee stockholders' unions, similar to em- ployee stock ownership plans (ESOPs) in some market economies. Local governments and trade unions in Beijing, Guangxi, Jiangsu, Jilin, Shaanxi, Shanghai, and Tianjin jointly formulated preliminary regu- lations for internal employee stockholders' unions. The cities of Dalian, Shenzhen, and Shijiazhuang, which were authorized to experiment, also issued their own local regulations. In addition, enterprises in the building, pharmaceutical, textile, and metallurgical industries in Shang- hai experimented with various approaches. In October 1997 the Min- istry of Civil Affairs, the Ministry of Foreign Trade and Economic Cooperation, the State Commission for Restructuring the Economy, and the State Administration for Industry and Commerce jointly is- sued the Provisional Regulations Concerning the Registration and 42 TRANSFORMED ENTERPRISES Management of Unions of Employee Stockholders at Experimental Foreign Trade and Economic Enterprises. These regulations state that an employee stockholders' union is an organization that manages employees' stock capital, subscribes to stocks from the company, ex- ercises the power of the stockholders, performs the obligations of the stockholders, and safeguards the legitimate rights and interests of the employee subscribers. However, no corresponding national policy or law regulating such practices is in place. Stipulations regarding the nature of employee stockholders' unions are of two types. In Beijing and Tianjin, employee stockholders' unions are separate corporate legal entities. In Jiangsu, Shaanxi, and Shanghai employee stockholders' unions are not separate legal entities, but are under the auspices of the trade unions. Various regulations specify a number of sources of funds that shareholders' unions can use to acquire stocks, namely: cash, annual bonuses, awards to outstanding employees, part of the company's prof- its if all shareholders consent, and other legitimate sources agreed on at stockholders' meetings. The regulations in Jilin province stipulate that a company could purchase part of its state-owned stocks or legal persons' stocks to resell them to its employees. A listed company can also purchase a corresponding part of its own stocks on the secondary market to resell them to its employees. Employees who have the right to purchase stocks are typically full-time employees who have worked in a company, one of its subsidiaries, an associated company, or in the company's representative offices for at least a year; the company's directors, supervisors, and managers; and retirees who receive their pension from the company. Regulations typically prescribe a limit for shareholding by such unions to between 10 and 50 percent of the total capital, depending on the size of the company. Some regulations also limit the maximum amount of shares that senior management can purchase. Issues Concerning Workers' Role in Corporate Governance. The ex- tent of some important worker rights is a function of enterprises' own- ership structure. Workers' rights are typically most extensive in collective enterprises and enterprises with dominant state ownership. Workers' rights vary significantly in TVEs, foreign-invested enter- prises, and private companies. This correlation between workers' rights and the form of ownership can generate resistance to ownership changes. For example, the reduction in the number of enterprises that 43 CORPORATE GOVERNANCE IN CHINA are predominantly state-owned has generated demands to extend employee rights in SOEs to enterprises that reduce or eliminate state ownership. Use of the legal system to protect workers' rights is still limited. Regulations typically give employees the right to request intervention by the appropriate local government department if they consider that their rights have been violated. Such practices create conditions for continued involvement by government entities in enterprises' internal operations. The rights and obligations of institutions that can represent work- ers' interests overlap significantly, and many believe that this has a negative influence on enterprises' management and operations. An example is the controversy surrounding the relationship between the so-called "three new committees" and "three old committees." Dur- ing the reform process, many SOEs established a shareholder commit- tee, a board of directors, and a supervisory committee: the three new committees. These coexist with the long-standing party committee, workers' congress, and trade union: the three old committees. The proliferation of representative bodies has made the exercise of certain rights and functions merely a formality. In many instances, corporate governance is still carried out using the traditional methods and in- struments, such as party and administrative meetings. Boards of di- rectors still function in a perfunctory manner, despite the adoption of rules and procedures. Having nonfunctioning supervisory boards where employees are heavily represented is common. The existence of so many representative bodies with overlapping functions has complicated employees' exercise of their shareholder rights. It has created conditions whereby employees perceive their share- holder rights merely as an additional instrument for furthering their narrow interests as employees. Employees still think of themselves as permanent workers, implicitly protected against layoffs, and shareholding often reinforces this perception. Such a conflict of inter- est hinders employees' capacity to play a constructive role in corpo- rate governance. Many localities have issued regulations that, in effect, buy em- ployees permanent worker status (see box 3.2). According to these regulations, enterprises pay employees a certain allowance or com- pensate them if they terminate them or rescind the original labor con- tracts signed between the employees and the enterprise. The main objective is to break the permanent association between workers and 44 TRANSFORMED ENTERPRISES BOX 3.2 IDENTITY SWAP OF SOE EMPLOYEES IN CHANGSHA In late 1999 and early 2000, three medium to large SOEs in Changsha, the capital of Hunan province, conducted a reform experiment that in- volved an equity swap, which referred to the ownership reform of these enterprises, and an identity swap, which referred to the substitution of employees' permanent affiliation to a particular SOE by selection through the market. The identity swap is, in a sense, a farewell to the "iron rice bowl" system. It also extends to managers, who through the identity swap lose the status of government officials and the corresponding pro- motion opportunities, including the right to be assigned to another en- terprise in case of dismissal, liquidation, and so on. The municipal government of Changsha prepared and offered the following compensation package in exchange for "permanent employee status": * Former permanent employees recruited before 1984 should be compensated RMB 500 per year for a period of I to 10 work years and RMB 900 per year for more than 10 work years, but the total accumu- lated amount cannot exceed RMB 20,000. * Employees who are less than five years short of retirement age can take early retirement after approval from the Labor Department. * Salaries (basic living expenses), welfare payments, and social security expenses that the company owes to employees should be paid together with the compensation. * Compensation for staff currently on the payroll should be paid in the form of a preferential price for the purchase of shares. Unexpectedly, the identity swap faced many difficulties as workers and managers were reluctant to accept full exposure to market forces and the threat of unemployment. their enterprises and to substitute a relationship based on market se- lection. Some localities such as Jinhua have also used this approach to create the conditions necessary for concentrating ownership in the hands of managers and key technical staff. In many instances, employee ownership has been merely a fundraising exercise, and it has not been accompanied by genuine changes in corporate governance mechanisms. Smaller enterprises have 45 CORPORATE GOVERNANCE IN CHINA often been able to avail themselves of this captive source of funds through a combination of coercion and persuasion, including by local governments. Surveys have found that employees have little aware- ness of the risks associated with investing in stocks. One important implication is that enterprises were able to isolate themselves from the discipline of capital markets. Furthermore, the transformation of em- ployees into shareholders could further soften enterprises' budgetary constraints by reducing the discipline imposed by workers' fixed claims on wages and other forms of compensation. Future of Employee Participation in Corporate Governance. Contin- ued discretionary involvement by local governments in dividend and wage controls, managerial appointments, and arbitration of internal negotiations is incompatible with the new ownership status of trans- formed small and medium enterprises and is likely to discourage the development of market-based corporate governance practices. Such functions should be apportioned between outside creditors and inves- tors, shareholders, and internal representative bodies. Creditors, for instance, should increasingly be in a better position to monitor wage and dividend payments and to enforce discipline in this area through covenants, repayment clauses, refusal to renew working capital facili- ties, and other means. This type of externally imposed discipline is likely to make it easier for managers to resist unwarranted pressures from employees for wage increases and dividend payments and to create, in turn, incentives for employees to monitor managers' perfor- mance and compensation. However, government involvement is often actively sought by managers and employees who are reluctant to part with the paternal- istic protection of the state, including their association with the civil service and permanent employee status. In this context, complete de- linking of SOE managers from the civil service system will facilitate ownership transformation and will promote the development of a la- bor market for managers. The reluctance of managers of transformed enterprises to part with their status as government officials exempli- fies the limitations of the pilot experimental approach to divesting SOEs in the absence of civil service reform that redefines the scope and nature of civil service in line with the economy's changing owner- ship structure. In March 2001 the SETC, the Ministry of Personnel, and the Ministry of Labor and Social Security issued Proposals on Deepening the Reform of the Internal Systems of Personnel, Labor, 46 TRANSFORMED ENTERPRISES and Distribution of State-Owned Enterprises. The proposals recom- mend the following: (a) removing the administrative rank of enter- prises; (b) changing the status of SOE managers so they are no longer considered to be government officials; (c) eliminating the strict demar- cation between cadres and workers; and (d) transforming identity management into position management, meaning that managers' sal- ary and benefits would be a function of their current positions and not depend on their personal attributes. If fully implemented and extended to all types of SOEs, these measures are likely to have profound effects on corporate governance. They will represent an important step toward further depolitization of the business process, will create the conditions for the development of a managerial labor market, and will positively affect management- labor relationships by facilitating movement between these two cat- egories of employment. However, the current reluctance of SOE managers to part with their status as government officials reflects a number of structural rigidities in the system. For example, even with civil servants' indirect control over managers through boards of direc- tors, dominant state ownership perpetuates an implicit benchmarking of managerial compensation to civil service pay levels. As a result, there is no significant differentiation between managerial and civil service remuneration to compensate for higher job security, mobility within the civil service system, and other prerogatives associated with government employment. In addition, many of the factors that cur- rently limit labor mobility in China, such as the absence of pension portability or the easy transfer of social benefits, apply to civil ser- vants and discourage separation from the civil service system. Finally, the link between managers and the civil service will persist as long as the government and the party continue to be involved in managerial appointments. Separating legal labor rights from the form of ownership will en- hance workers' role in corporate governance. The current links be- tween legal labor rights and ownership form create a bias in favor of the status quo and discourage workers from looking at contractual mechanisms to structure and protect their rights. In general, the links make ownership transformation more costly, as workers have to be compensated in some way for changes in ownership that they per- ceive as reducing their bundle of rights. Thus what is needed is a uni- form approach to labor rights that treats all employees in a similar fashion irrespective of ownership form. Given the variety of ownership 47 CORPORATE GOVERNANCE IN CHINA forms and the associated bundles of legal rights, as well as the wide spectrum of forms that employee participation in corporate gover- nance can take in general, the question arises as to what the com- mon denominator behind such uniform treatment of legal labor rights should be. The wide diversity of ownership forms and of mechanisms for worker participation in corporate governance (see box 3.3) suggests BOX 3-3 THREE STYLIZED REGIMES FOR WORKER PARTICIPATION IN STRATEGIC MANAGEMENT Charny (1999) distinguishes between three stylized regimes that govern workers' exercise of managerial power: "hard," "soft," and "no par- ticipation" regimes. Hard regimes are based on legally mandated and regulated mecha- nisms for worker participation in corporate governance. Germany is an example. The German system has successfully divided the responsibility for different aspects of the workplace regime: board representation pro- vides workers with a means of gathering information and communicat- ing their views to the board, sector and public (legislative) bargaining resolves basic compensation issues, and councils can deal with plant- level issues of enforcement and work structure. Soft regimes work through mechanisms that are not explicitly man- dated by the legal order or by legally enforceable contracts. The typical case is Japan. The Japanese industrial relations system enables workers to participate in strategic decisionmaking both on the shop floor and at the top layer of corporate management, but it does so largely through informal, nonlegal mechanisms. Japan's enterprise unions have been effective in moderating wage demands in exchange for an implicit com- mitment by the firm to share the benefits of long-term growth by means of lifetime employment and gradual wage increases linked to seniority. Several factors account for the high degree of cooperation between labor and management in this system: (a) a board that is relatively free from direct pressure from shareholders seeking to maximize their profits and is instead loyal to keiretsu members, corporate affiliates, and creditors; (b) the domination of boards by insiders, including those who have risen through the corporate ranks, thereby ensuring that at least some board members have personal connections and familiarity with labor interests, (c) the presence of labor unions at the enterprise level that are capable of resolving problems of collective action among subgroups. 48 TRANSFORMED ENTERPRISES that legally embedded labor rights should be focused on the core tra- ditional labor issues of compensation, working conditions, social ben- efits, collective action, and so on, leaving the rest to voluntary, legally enforceable arrangements. The various forms of worker participation in ownership and control should not be viewed as substitutes for strengthening the traditional contractual and legal mechanisms for protecting workers' core interests. The no participation regime has no standard mechanisms for worker participation in strategic management. The United States is perhaps the most representative of this regime. Some of the reasons why systematic forms of worker participation in strategic decisionmaking have not evolved in U.S. enterprises may be (a) fragmented unions organized along sectors and sharply defined job slots; (b) an ideology of managerial autonomy; and (c) shareholders' reliance on the stock market for monitoring mana- gerial performance, which discourages the development of long-term, cooperative, implicit contracts between workers and managers. A comprehensive and definitive evaluation of the relative merits of these systems may well be impossible. Each has demonstrated that it is highly effective for certain aspects of worker participation and under particular circumstances. Furthermore, the effectiveness of these sys- tems cannot be examined in isolation from factors such as the level of public versus private provision of social services; the type of financial system, that is, market based versus bank and relationship based; and the development of other supportive institutions, such as unions and public institutions for social welfare entitlements. However, in relation to adaptation to large-scale industrial change, such as divesting from sectors with overcapacity and developing new growth areas, the Ameri- can system has shown a remarkable responsiveness and flexibility rela- tive to the other two systems. An important lesson from the United States is that worker cooperation in the sense of acquiring new skills or making new firm-specific investments may be induced without the pro- tective participatory devices of either the hard or soft regime. In reality, under conditions of rapid change in production, the hard and soft re- gimes may actually become a barrier to worker cooperation. SOURCE: Charny (1999). 49 CORPORATE GOVERNANCE IN CHINA Through individual and collective bargaining and contracting, firms and workers can design their rights and responsibilities regard- ing duties, compensation, access to information, representation on boards, and so on.' In the process of monitoring the execution of these contractual arrangements, workers can play a meaningful and important role in corporate governance. Collective action is a neces- sary counterpart to collective bargaining and contracting, and could be enhanced by strengthening the independence of trade unions. Thus in normal times, the standard contractual and legal tools for protect- ing employees' core interests are also potentially the most powerful instruments for worker participation in corporate governance. Shareholding and board representation often add little to the extent to which workers' rights are protected, and in reality may reduce the effectiveness and applicability of traditional instruments such as exit, collective action, and contractual arrangements related to compensa- tion. Shareholding, for example, softens employees' fixed claims and thus relaxes their disciplining effect on the company's management, while board representation often amounts to little more than work- ers' accepting responsibility for decisions that are not under their con- trol. In addition, continuously bringing administrative bodies into labor negotiations and disputes undermines labor rights and company inde- pendence in the long run. One particular form of worker involvement in corporate gover- nance is co-determination, which typically provides for workers' par- ticipation in control but not in the distribution of residual earnings. Some elements of co-determination are currently present in China, al- though their effectiveness in providing workers with the incentives and instruments to monitor management and the agency costs of equity is questionable. International experience does not provide any conclusive evidence on the effectiveness of co-determination in enhancing employ- ees' role in corporate governance, although it could potentially have a serious impact on corporate behavior (see box 3.4). Given this uncer- tainty, a relatively safe course is to make co-determination in its various 3. Contractual arrangements can also protect employees' firm-specific human capi- tal, albeit imperfectly. Firm specificity of skills makes workers more difficult to replace and should therefore increase their bargaining power. Severance payments and other forms of compensation linked to tenure and seniority can approximate protection of firm-specific human capital investments. 50 TRANSFORMED ENTERPRISES BOx 3.4 IMPACT OF CO-DETERMINATION ON FIRM BEHAVIOR The experience of European countries such as Germany, the Nether- lands, and the Scandinavian countries that have adopted some form of co-determination does not provide any conclusive lessons. Some ob- servers have argued that while such systems do not give workers enough power to fundamentally change firms' behavior, they do play an infor- mational role, providing a credible source of information from the firm to the workers (and vice versa) in support of collective bargaining by the unions and decisionmaking in the works councils, where workers exert their influence in practice. Thus co-determination can be viewed as a useful supplement to traditional contractual arrangements. However, recent studies of German firms using co-determination (see, for example, Gorton and Schmid 2000) found that it does affect firms' behavior. They found that co-determination empowers employ- ees, and that they use this power in ways that contradict the desires of shareholders, that is, they change the firm's objective function. Co- determination gives employees bargaining power by effectively transfer- ring some of the control rights to them in the form of seats on the supervisory board. With employees on the supervisory board, firm re- sources are directed differently, decreasing the returns on assets and the market to book ratio. Gorton and Schmid found that co-determination reduces market to book value by 27 percent, return on assets by 5 per- cent, and return on equity by 2 percent. Other studies have found that co-determined firms are more likely to resist restructuring and that share- holders use the capital structure to mitigate the impact of co-determina- tion though higher leverage. None of this answers the question of whether co-determination is socially optimal or not. forms optional, as in France, and to leave it to companies to negotiate the extent of worker representation, if any, on boards. Employee ownership can take a variety of organizational forms, and the particular mechanisms for worker participation in corporate governance can be critically important to the firm's economic success. While direct employee ownership is relatively rare across countries and sectors (see box 3.5), it may have special value during a time of eco- nomic transition associated with restructuring. An ownership struc- ture dominated by management and employees can give rise to various 51 CORPORATE GOVERNANCE IN CHINA BOX 3.5 INTERNATIONAL EVIDENCE ON DIRECT EMPLOYEE OWNERSHIP A significant amount of evidence is available on the effectiveness of employees as shareholders. Around the world, employee ownership is the exception rather than the norm. The types of industries in which employee-owned firms are found and the structures those firms assume are remarkably similar. Employee-owned firms are rare in the industrial sectors but are quite common in the service sector, especially in the ser- vice professions, such as legal, accounting, investment banking, man- agement consulting, advertising, architectural, engineering, and medical firms or practices. Successful employee-owned firms frequently convert to investor ownership. Employee ownership has often emerged following the restructuring of investor-owned firms that have developed severe financial difficul- ties. Financial distress has been a significant factor in the emergence of insider-dominated ownership in China. Selling distressed firms as a whole or in part to their employees has a variety of potential benefits. First, it offers a way for the employees, and especially their union, to accept the substantial concessions necessary for the firm to continue to operate, such as layoffs, severely reduced wages, and changed work rules, be- cause it gives them a benefit (the stock in the reorganized firms), albeit of uncertain value, to set off against their reduced wages and benefits. This lowers the net magnitude of employees' losses and hence makes them easier to accept psychologically (Hansmann 1996 ). Second, it is a credible way for the firm's investor-owners and its managers to signal to the workers how serious the firm's financial difficulties are and the con- sequent necessity for employee concessions, thereby averting costly bar- gaining. Finally, it is a credible way to assure the employees that, if the firms survives and prospers, the fruits of their concessions will not go disproportionately to the firm's current investor-owners. conflicts of interests. For example, managers can use the hire and fire process to solidify their control and expropriate minority sharehold- ers who are also employees, while employees can use their position as shareholders to pursue their narrow interests and resist restructuring that could be beneficial for the company. In market economies, various institutional and contractual solu- tions to these problems have emerged in the context of closely held 52 TRANSFORMED ENTERPRISES corporations. For example, high voting and quorum requirements, as well as employment and compensation agreements that make it hard for managers to act without the consent of minority shareholders, can be used to protect shareholding employees. However, as minority share- holders become more powerful, company agreements can make greater use of arbitration, voting trusts, or third parties who have the right to vote only to break deadlocks. Some of these mechanisms, such as vot- ing trusts, can also alleviate the problems of collective action by a large number of heterogeneous workers. China has recently adopted a new trust law that will facilitate the use of such mechanisms. While direct employee ownership is rare, partial (that is, mainly allowing for participation in residual earnings, but not in control) and indirect (that is, through collective investment vehicles) forms of worker participation in corporate governance are widespread and increasing rapidly. Some of these forms could be useful in the Chinese context, both in alleviating some of the incentive issues associated with insider- dominated ownership structures and in facilitating transitions away from direct employee ownership. ESOPs, in particular, could play a useful role in the evolution of corporate governance practices in trans- formed small and medium enterprises, provided that certain condi- tions are satisfied. Although the international evidence on the extent to which ESOPs enhance workers' role in corporate governance is inconclusive (see box 3.6), they can facilitate collective action by shareholding employees in China provided they are independent of company management and capable of acting as the representatives of independent shareholders by separating shareholders' interests from narrow workers' interests. Both conditions can be approached through regulations and various forms of delegation, including through trust arrangements. One promising way for Chinese labor to play a role in corporate governance is through institutional investors such as union pension funds and labor-oriented investment funds. For China, establishing an institutional investor base is an important priority in capital mar- ket and corporate governance reforms. Evidence from the United States indicates that in the 1990s unions became the most aggressive of all institutional shareholders. Unions, union pension funds, individual union members, and labor-oriented investment funds are using the corporate voting process to push for changes in corporate governance. Observers have interpreted this as a new alignment of interests between shareholders and workers, as op- 53 CORPORATE GOVERNANCE IN CHINA BOX 3.6 ESOPs Do NOT TYPICALLY ENHANCE WORKERS' ROLE IN CORPORATE GOVERNANCE Employees' capacity to exercise their rights as individual shareholders are intrinsically limited by their heterogeneity and the collective action problem. Various collective vehicles have emerged for the exercise of employees' shareholder rights. For example, since the 1970s many American firms have adopted ESOPs under which most or all of the firm's employees receive a portion of their compensation in the form of stock in the firm. Roughly 90 percent of all ESOPs are in privately held firms. The rapid proliferation of ESOPs in some developed market econo- mies is not, however, an unbiased indicator of their efficiency, because they became popular when they were granted substantial federal tax subsidies. China is actively debating whether a regulatory framework including tax incentives should be established to promote ESOPs. The numerous studies of ESOPs to date, while not conclusive, have failed to present clear evidence of improvements in either employee pro- ductivity or firm profitability once they have allowed for tax subsidies. Note also that ESOPs generally provide for participation only in earn- ings, but not in control. Only rarely have they been structured to give employees a significant voice in the firm's governance. The fact that employees typically do not participate in the governance of these firms suggests that those responsible for structuring them believe that any reduction in agency costs that might result from making management directly accountable to the firm's employees, even though the employ- ees are already the firm's beneficial owners, would be outweighed by the costs-whether in the form of inefficient decision or high process costs-that would be engendered by the political process required for such accountability. However, this could also be interpreted as corpo- rate managers trying to preserve or increase their own autonomy, that is, protecting themselves both from hostile takeovers and from direct accountability to the firm's employees. posed to the alignment between workers and managers in the 1980s during the wave of takeovers in the United States. Typical proposals advanced by U.S. unions in the mid-1990s included abolishing anti- takeover devices established by management, prohibiting conflicts of interest by directors, separating the positions of chief executive officer (CEO) and board chair, and linking directors' pay to the company's 54 TRANSFORMED ENTERPRISES performance. Labor representatives have used some innovative ap- proaches to participate in corporate governance, such as making pro- posals from the floor at shareholders' meetings and amending the corporation's by-laws to restrict certain discretionary powers of the board of directors. On occasion, these institutional investors have mobilized individual shareholders who are union members to actively use their shareholder rights. Thus unions and union pension funds are showing a capacity to act as typical institutional investors motivated by increased shareholder value. Labor unions often face a potential conflict of interest when they act as shareholders. Forces that could constraint workers' opportun- ism include fiduciary obligations of pension fund trustees; workers' needs to persuade other shareholders to vote for their shareholder initiatives; product and factor markets competition; capital structure, especially bank borrowing; and laws and regulations that allow only corporate governance proposals for which all shareholders as a group share the same interests. Role of Banks One would expect creditors, banks in particular, to play an important role in the corporate governance of transformed Chinese small and medium enterprises, which are typically highly leveraged, have no immediate access to public equity markets, and often experience finan- cial difficulties. Yet banks still play an extremely limited role in the governance of these enterprises. They have not generally been involved in the restructuring of ownership patterns and they have not been able to control the agency costs of equity, as demonstrated by exces- sive dividend distribution and wage growth. Furthermore, firms' ac- cess to the captive pool of employees' savings has somewhat reduced their dependence on debt, thereby limiting banks' leverage, at least, for the moment. Certain features of the transformation process, such as the exclusion of land use rights, are likely to affect the role banks can play in the governance of these enterprises in the future. The fol- lowing sections discuss the main mechanisms for the exercise of con- trol by banks, the factors accounting for banks' limited role in corporate governance, and some approaches toward strengthening that role. Main Mechanisms for Exercise of Control by Banks. Creditor banks can exercise control over corporate governance in debtor enterprises 55 CORPORATE GOVERNANCE IN CHINA in several ways (Gray 1997). Foremost, banks can exert influence by means of credit. By giving or denying credit or by making credit terms more or less favorable, the bank allows its debtor to strengthen or expand or obliges it to resort to a less attractive, alternative source of financing. The extent of this influence depends largely on the avail- ability of such alternative sources, while the quality of the influence depends to a large extent on the bank's credit decision process. Creditors exercise influence over borrowers based on laws and contracts. Legislation or contract covenants often give creditor banks the right to receive information, impose audits, require prepayment, veto certain strategic decisions, and so on. Some of these rights apply in the normal course of business, and others refer to the use of the bank credit or to developments that increase financial risk. In addi- tion, creditors may exert influence in informal ways, such as being consulted about major decisions, invited to shareholder meetings, appointed to boards, or asked to second bank staff to work at the firm. Debtors may accept such practices if they have few alternative sources of financing because of weak competition among banks and little chance of accessing nonbank financing. One of the most powerful tools available to banks to exercise influence over corporate governance is their special rights in the case of default. Once a debtor defaults on repayment or other credit condi- tions, the bank can trigger court actions, such as foreclosure on col- lateral, liquidation, or reorganization of the firm. The mere threat of such action may allow the bank, either individually or jointly with other creditors, to force a range of actions on the defaulting firm's managers and owners. The creditors may even formally become the new owners through a debt/equity swap or by accepting assets in pay- ment. The effectiveness of these methods of influence depends largely on legislation and the court system. A creditor bank may also exercise control if it engages directly, or through affiliates, in the investment or securities business. For instance, an affiliate may hold equity in the debtor. Through its actions, the bank provides signals to third parties, in- creasing its direct influence. just the leakage of the bank's assessments of the enterprise could influence other stakeholders, given their aware- ness of the bank's ability to obtain and analyze information, and be- cause of the potential consequences of the bank's actions as a creditor. Countries vary significantly in the extent to which banks use different channels of influence over corporate governance (see box 3.7). 56 TRANSFORMED ENTERPRISES BOx 3.7 SOME CROSS-COUNTRY PATERNS IN BANKS' ROLE IN CORPORATE GOVERNANCE Industrial economies commonly have some means whereby creditors can influence debtors, such as the actual credit award, basic covenants in credit contracts, and special clout in the case of insolvency. However, the extent of such influence differs widely. Germany and Japan have long relied heavily on bank credit relative to securities finance, and main bank systems have evolved in these countries.' In Germany this is un- derpinned by equity holdings and proxy voting by the universal banks. Limited competition and consensus among the banks resulted in con- siderable bank influence even over many firms that are not joint stock corporations. In Japan main banks are often members of business groups with multiple cross-holdings, and the size of these groups and their de- pendent satellite suppliers have meant that these banks have a wide influence. German and Japanese main banks enjoy access to extensive information, board membership, and frequent consultation by their debtors, and in the German case a creditor-friendly insolvency system. The United States tends to be at the opposite end of the spectrum, with less extensive bank finance, a separation of commercial and investment banking, and arm's-length bank-client relationships. Some trend toward convergence is apparent, however. German banks are spinning off or selling off their industrial holdings, and German enterprises are access- ing financial markets. In the United States the strict separation of coin- mercial from investment banking is being relaxed, thereby permitting more comprehensive bank-client relationships. Developing countries have often followed the traditions of their former colonial powers. For instance, Hong Kong (China), Malaysia, and Singapore have pursued a broadly U.K. and U.S. approach, albeit in Malaysia with some government use of banks to guide the real sector. In the Republic of Korea, in the past the government had mandated that each chaebol indicate a main bank for at least the core firms of the group, and recently assigned one commercial bank to each distressed 1. There was an initiative to impose a main bank system in China. In 1996 the PBOC issued the Provisional Measures for the Administration of a Main Bank System for medium and large SOEs, primarily the better performing ones, in seven municipalities. The initiative has not been expanded beyond this pilot. (continues) 57 CORPORATE GOVERNANCE IN CHINA BOx 3.7 (CoNTNUIED) chaebol to lead its workout. Overall, however, Korean banks have fol- lowed government guidance in lending and have exerted little direct governance influence. The role of creditors in governance also varies in the transition economics. Russia is heavily reliant on bank financing, Russian banks' own stakes in industry through shares-for-loans swaps and other ma- neuvers, and influential banks are present at the center of several large business groups. However, Russian banks are rarely original sources of governance, but rather a convenient conduit for business tycoons ("oli- garchs") to exert their control. In several other countries of Eastern Europe and Central Asia banks also became vehicles for making early speculative gains, attracting business talent, and building up political clout. From this basis they played a major role in acquiring control in privatized enterprises by buying shares outright, swapping debt into equity, creating voucher investment funds, or lending to other buyers of enterprises. This governance role of banks in Eastern Europe and Cen- tral Asia was commonly through ownership in the enterprises, not as creditors. This mirrored the single-minded focus of policymakers and their advisors on improving governance through ownership changes alone without regard for a supplementary role by creditors and other stakeholders. Credit approval or denial. The approval or denial of credit by a bank can have a significant impact on a firm's strategic development. This is particularly true in China, where firms' access to securities markets is de facto tightly rationed and various other forms of non- bank financing have yet to develop. Thus a pertinent question in relation to corporate governance is whether the banks' credit decisions are sound. Until the mid-1 990s, the lending decisions of state-owned commercial banks (SOCBs) were often not driven by borrowers' creditworthiness and the financial merits of their projects, but rather by persuasion by local or national authorities and by personal rent-seeking on the part of bank person- nel. Efforts to rectify this situation have intensified in the last five years by such means as curtailing instructed lending, measuring bank performance more on the basis of profits than of lending targets, over- 58 TRANSFORMED ENTERPRISES hauling banks' credit approval procedures, and expanding training in credit analysis. Nevertheless, the scope for improvement is considerable. While branch managers may be evaluated in part based on the performance of credits that they approved, the underlying portfolio classification may be manipulated by those same managers because internal control practices and regulatory supervision are still weak. Banks still consider project analysis a secondary matter, given their exclusive reliance on collateral and third-party guarantees. In addition, the financial performance of state-controlled enterprises still depends largely on government decisions about industry ratio- nalization, technological upgrading, and other forms of support. This limits the usefulness of assessing borrowers and their projects on their own merits. Enterprise accounting and auditing still suffer from serious short- comings. Banks, especially SOCBs, rarely insist that their clients im- prove their accounting systems, use specific auditors trusted by the bank, or disclose related party transactions. Even bank managers who do recognize the need for changing their stance in this regard find implementing such changes difficult because of the current surge of competition among banks. With the support of the PBOC, the newly created Association of Banks could play a useful role in helping banks demand more reliable financial information. Covenants and consultation. Banks have few rights to influence their clients' strategic decisions. The standard credit contracts include few covenants that permit real involvement. On such matters as ma- jor financial or asset restructuring, ownership changes, or changes in business lines, at best they usually require "information" rather than consultation, let alone approval. Neither do they require strict main- tenance of key financial ratios. The only strict covenants tend to con- cern repayment terms, credit security, use of the credit funds, and perhaps adherence to government programs related to the credit. To date banks have had little direct influence on their borrowers' strategic decisions through means other than legal and contractual requirements. Client enterprises do not generally consult the principal bank before making major decisions. More often than not, they even make investment decisions before approaching the bank for credit. Publicly listed companies do not tend to invite key bank person- nel to their annual shareholders' meetings, and bank managers are 59 CORPORATE GOVERNANCE IN CHINA rarely appointed as external members of their client companies' boards of directors. While an increasing number of external members do sit on company boards, they are usually representatives of major share- holders, including local governments. In some cases bank managers avoid seeking board memberships, because they believe it violates the prohibition of second jobs by bank managers under the Commercial Banking Law. Many interviewees emphasized the lack of trust between banks and enterprises, based apparently on the banks' ongoing transition from soft budget agents to profit-oriented competitive players and the changes in perceptions, relationships, and capabilities that this entails. The state- owned banks, awash with deposits that bear an implicit guarantee but increasingly less inclined to fund high-risk clients, have difficulties find- ing good borrowers; yet they are unfamiliar with how to deepen their relationships with good clients. As to enterprises, those that are not creditworthy and face this new rigor on the part of their banks feel let down, and in many cases become antagonistic and try to mobilize the support of government officials against the reluctant bank. Among the good borrowers, some try to conform to the new circumstances and see no need for anything other than an arm's-length relationship with their banks. In between these two extremes, those enterprises with less than sterling creditworthiness should eventually appreciate the advantages of dealing more openly with banks. Special rights in cases of debtor default. If their borrowers de- fault, creditor banks' rights are weak (World Bank 2001 a). Banks have less influence on defaulting debtor enterprises than in industrial mar- ket economies. The lack of a credible threat of bank influence in the case of defaults has also weakened the banks' overall clout under nor- mal circumstances. China has a detailed Law on Credit Security, and the Commer- cial Bank Law requires banks to secure credit for all but their best borrowers. However, foreclosures do not give the banks much lever- age for several reasons: for example, self-help is limited; judgments by courts with limited independence and sometimes weak skills are unre- liable; enforcing court judgments can be difficult; irregularities in the valuation and registration of securities can occur; and in many SOE bankruptcies, the government appropriates mortgaged land use rights to give first priority to settling workers' entitlements. 60 TRANSFORMED ENTERPRISES Banks have limited influence on their debtors when the debtors declare bankruptcy. Under the Trial Bankruptcy Law for State-Owned Enterprises, filing for bankruptcy requires approval of the debtor's line bureau. The Liquidation Comniission represents the debtor's owners, regulators, and the local community, and creditors have little involvement. The old management usually remains in place until liq- uidation has been completed. Municipal bodies tend to control asset valuation and disposal, and under the Capital Structure Optimization Program the often most valuable asset, land use rights, is used to settle workers' and pensioners' entitlements as a first priority. The bank- ruptcy of enterprises other than SOEs is governed by the Company Law and Civil Procedures Law. Such bankruptcies tend to suffer from a number of weaknesses, including a lack of specificity in these laws, inadequate information provided to creditors, limited independence of the courts, and little recourse against court decisions. The incomplete and vague legal framework for bankruptcy re- mains a severe constraint to the influence that creditor banks might otherwise have over insolvent debtors. As one banker put it, the only hope for substantial recovery of unsecured credit is to "react immedi- ately to early distress signals and obtain immediate payment or addi- tional collateral by using all kinds of threats against the firm and its managers and owners as long as they are legal." The drafting of a new bankruptcy law is in its final stages. Its adoption will be an important step toward rectifying the situation. Banks have also played a limited role in reorganizations and workouts. Court-supervised reorganization is rare for SOEs, because the authorities prefer liquidation with another enterprise taking on the entire asset bundle, cleansed of debt. Mergers of distressed com- panies are often carried out administratively, with little creditor in- volvement, under a debt restructuring formula predetermined by the Capital Structure Optimization Program. To date creditor banks have managed to force workouts on only a few of their many nonperform- ing debtors. The reasons for this include the weakness of the bank- ruptcy threat, the government's concerns about the socio-political implications of layoffs, the banks' lack of skills in and experience with workouts, the lack of incentive for bank managers to disclose loan quality problems, the shallowness of markets for disposing of certain assets, and the tight tax limits to loan loss provisioning and write-offs. 61 CORPORATE GOVERNANCE IN CHINA Ownership stakes. The Commercial Bank Law prohibits coin- mercial banks from ownership in nonfinancial institutions. Thus they cannot directly hold shares in client enterprises, thereby supplement- ing their creditor rights with ownership rights and having a greater influence on the firm. In addition, the commercial banks are not di- rectly engaged in providing advice on securities management, which could lead to significant proxy voting by these banks. However, through their ownership of securities and investment houses, the larger commercial banks have indirectly had the potential to exert some ownership influence on debtor enterprises. In the after- math of the Asian financial crisis, a PBOC regulation discouraged commercial bank investments in nonbank financial institutions, and any such ties are now being severed. The trust companies of the SOCBs became the legal predecessors of the asset management companies, now owned by the Treasury, and their securities operations have been spun off and merged into new entities independent of the SOCBs. The China Construction Bank still owns a major stake in an investment banking joint venture with Morgan Stanley, but this firm engages in venture capital financing and investment advice rather than investing in clients of the China Construction Bank. Other SOCBs have stakes in investment and securities houses in Hong Kong (China) or over- seas. The extent to which these banks own mainland companies is unclear. Finally, some of the smaller commercial banks are members of diverse business groups and lend to affiliated firms within the group, but rather than exerting influence on these firms, the reverse is the case. The Everbright Group included both commercial and invest- ment banking operations but recently sold the latter to the State De- velopment Bank. Spearheaded by the China International Trust and Investment Corporation, some groups are now considering clarifying their structures and establishing financial holding companies that would formally own both commercial banks and investment and se- curities houses. China seems to be relaxing the strict separation between com- mercial and investment banking. The PBOC recently issued the Provi- sional Regulations on Intermediary Businesses of Commercial Banks, which state that following PBOC approval, commercial banks can engage in financial derivatives, securities, investment fund trusteeships, and information and financial consulting. This opens up new mecha- nisms for bank involvement in corporate governance. 62 TRANSFORMED ENTERPRISES A special case of creditors with equity stakes are the four large AMCs. As nonbank financial institutions, they can own equity in in- dustrial firms but also hold the old credit claims transferred from the SOCBs. While majority stakes by a single AMC are an exception, AMCs frequently hold more than 25 percent of the equity, which to- gether with the stakes of other AMCs in the same firm brings their total close to a majority. In many of these firms, the AMCs also retain some credit claims. If given the opportunity, how actively the AMCs would engage in the governance of these equity holdings is unclear. Government offi- cials have proclaimed that the AMCs would be free to exercise such governance as they deemed necessary to maximize assets, but also reminded the AMCs that the objective of the swaps was to realize immediate improvements in the firms' financial picture rather than giving ownership control to the banks in return for the banks' accept- ing a lower-ranking claim. Given such ambiguous signals, AMC man- agers seem reluctant to pursue an active governance role, and many firms whose debt had been swapped sought to prevent AMCs from taking an active governance role. Signals to third parties. Creditors can influence the governance of enterprises through the signals they send to other stakeholders. If such stakeholders perceive the banks as having strong analytical skills and the right incentives for making credit decisions, then simply the award of credit can be a powerful signal. In the past, observers viewed loans by SOCBs as a signal of continued government support for an enter- prise. As more bank credit is awarded without government guidance, this interpretation is less common; however, confidence that bank lend- ing now reflects true firm creditworthiness is only gradually emerging. In the past, China's banks shared information about borrowers more readily than they do now, given the increasingly competitive environment. For instance, banks' internal credit ratings are not com- monly shared between banks and in principle are not available to third parties. Borrowers may be able to find out their own ratings, but if they pass the information on to third parties, the latter have difficulties getting banks to confirm it. Similarly, in principle banks do not share information about clients' repayment records or guar- antees obtained in return for credit, though in practice much infor- mation is obtained informally at the local level. However, the 63 CORPORATE GOVERNANCE IN CHINA increasing tendencies to rotate staff among different branches and give them incentives to compete might gradually reduce such infor- mal information sharing. To discourage information sharing, in 1999 the PBOC introduced a system of "borrower passports." In addition to the latest financial statement, the passport is required to show any bank loans, the collat- eral put up for these loans, and the status of debt service. It is even supposed to show trade paper discounted at banks. Bankers are find- ing this system extremely useful, because it reveals excessive bank borrowing, multiple use of the same assets as collateral, and repay- ment performance. The passports can also reveal contingent liabilities in the form of guarantees the borrower has given to others, although this information may not be complete. Gaps may be uncovered by cross-checking the passports of several entities. The success of this borrower passport initiative will depend on the PBOC's capacity to discourage free-riding behavior by participating lenders. In some locations PCOB offices have recently published lists of defaulters. Such lists can have a major effect when these debtors seek new bank or trade credit. The lists may also signal to creditors and the firms' other stakeholders that the PBOC would not oppose legal ac- tion taken against these defaulters. In some cases PBOC branches even prohibited new bank lending to listed defaulters. The transfer of loans to AMCs should theoretically also send pow- erful signals to other stakeholders of debtor enterprises. In practice, the signal has been blurred by the transfer of performing loans for debt to equity swaps. In interpreting loan transfers to AMCs, the ex- pectation is widespread that nonperforming loans would simply be warehoused at AMCs until the debtors were eventually bailed out rather than facing foreclosure, restructuring, or liquidation. Constraints to the Exercise of Governance by Creditor Banks. In the context of constraints to banks' exercise of governance, the question is whether their own stakeholders motivate creditor banks to pursue maximum, risk-adjusted, long-term profitability. In the case of the SOCBs, state representatives now frequently ask for a profit orienta- tion. In practice, however, the SOCBs' objective function is blurred. First, they are subject to continuous demands to support various state enterprise sector policies, both at the national and local levels. Sec- ond, because they are SOEs, the SOCBs do not issue dividends to the state; therefore the Treasury focuses on the banks' tax payments, thus 64 TRANSFORMED ENTERPRISES in a sense making this their bottom line. Third, shortcomings in ac- counting in general, and in loan loss provisioning in particular, mean that financial reports do not reflect genuine profitability.' Fourth, an understanding that the SOCBs and some other banks may be too large to fail results in moral hazard, which regulatory supervision cannot fully resolve. SOCBs also tend to face similar corporate governance issues as nonfinancial SOEs. Like some other large SOEs, the SOCBs have re- cently been assigned supervisory councils consisting of special repre- sentatives of the state as owner. The extent to which this will improve their governance remains to be seen. Calls for corporatizing the SOCBs, and perhaps even listing them publicly, are increasing. The latter, in particular, could have a positive effect on their governance if the own- ership distribution and governance arrangements gave nonstate share- holders the opportunity to restrict the state's nonfinancial interests in these banks. In the second-tier commercial banks the key shareholders tend to be local governments and state-owned or state-controlled enterprises. Only one commercial bank, Minsheng Bank, is referred to as a nonstate entity. The licensing of additional nonstate banks under genuine pri- vate control and the expansion of foreign-invested banks into local lending could result in more creditor banks that are themselves under effective corporate governance. Strengthening Creditors' Role in Corporate Governance. The existing ownership structure of banks and borrowers, which is still largely dominated by the state, imposes limitations on the extent to which the role of banks in corporate governance can be strengthened. Explicit and implicit government guarantees associated with state ownership make companies' viability relatively independent of their own efforts and of project quality. This in turn reduces banks' incentives to screen 4. In a significant move, in 2001 the Ministry of Finance reduced the tax burden on banks and is allowing greater flexibility in writing off bad debts. The business tax, applicable to gross revenues, was reduced from 8 to 7 percent and is to be reduced further to 5 percent over the next two years. The 1 percent limit on tax- deductible provisions was changed to allow banks the flexibility to provision against loan losses. The financial supervisors have the authority to ask for additional provisions, if needed. The new rules specified a five-year framework within which financial institutions are encouraged to absorb historical losses through progres- sive provisioning (World Bank 2001c). 65 CORPORATE GOVERNANCE IN CHINA projects carefully and to monitor firms' performance. State owner- ship weakens banks' rights in the case of debtor defaults, especially when the courts are financed by local governments that are also sig- nificant owners of the defaulting enterprises. Finally, options to strengthen banks' profit motives and introduce credible penalties for failures are limited when banks are still state owned and are perceived as too big to fail. In this context, an important step would be to strengthen banks' profit incentives through private ownership and competition. At present, the ownership structures of the real sectors do not align with those of the financial sectors. Private ownership in the financial sector is practically nonexistent. The government should allow the entry of new domestic, private financial institutions, especially in view of China's WTO membership, which will open up entry opportunities to foreign financial institutions. To alleviate regulatory concerns, par- ticularly in light of recent financial crises in Asia and worldwide, stricter entry and prudential requirements could initially be applied to new private financial institutions. Private institutions are likely to be more independent of political considerations and more profit oriented. They are not likely to compete directly with existing state-owned banks, although increased competition through new entry will have an in- vigorating effect on the state-owned financial sector. However, more competition may not necessarily result in banks' adopting tougher lending criteria. When competing for market share, individual banks may find it disadvantageous to introduce more ex- tensive credit covenants, impose more detailed audits, insist on disclo- sure of related party transactions, and so on. Foreign-invested and domestic private banks will probably become important creditors by the second half of the decade; however, in the absence of a level play- ing field and some form of collective action, Chinese banks may real- ize that their laxity with respect to lending conditions could actually give them a competitive advantage over the more stringent foreign bankers. With PBOC support, the newly created Association of Banks could play a useful role in disseminating information about standard international practices and discouraging free-riding in their applica- tion by Chinese banks. Regulators should also emphasize training and education to enhance the use of traditional tools such as covenants, independent audits, monitoring and supervision practices, and so on. The large state-owned banks are likely to dominate the domestic financial landscape for the foreseeable future. Strengthening the profit 66 TRANSFORMED ENTERPRISES incentives of these banks would therefore significantly strengthen banks' role in corporate governance. Corporatization, listing, and stra- tegic partnering with foreign financial institutions are some options for reaching this objective. Even without fundamentally changing their ownership struc- ture, commercial banks in China can do a great deal to improve their own corporate governance. They can become more transpar- ent by using International Accounting Standards (IAS) and repu- table external auditors. Banks can improve board practices by setting up various committees and appointing independent directors to chair sone of these committees. Better corporate governance will, in turn, help banks play a more important role in the corporate governance of their borrowers. As banks adopt modern corporate governance approaches, their credit decisions are likely to become more sound, and those with better corporate governance will find that attracting strategic partners is easier. A stronger capital base will allow such banks to take a longer-term approach to their strategic and lending decisions. Some Chinese commercial banks have made significant progress in improving their corporate governance and financial per- formance through entering into technical assistance arrangements with reputable financial institutions and attracting strategic inves- tors. As the experience of Bank of Shanghai (see box 3.8) demon- strates, this can lead to better corporate governance, more rigorous risk management practices, and greater profitability. Even in transformed small and medium enterprises, ownership transfer often occurs in ways that impede enterprises' access to bank loans and other forms of financing in the future, thereby limiting the role banks and other financiers can play in the corporate governance of these enterprises. For instance, the valuation process often excludes land use rights or severely undervalues them so that insiders can ac- quire controlling stakes more easily. This reduces the borrowing ca- pacity of transformed enterprises. Preferential treatment of insiders in the form of huge discounts on the purchase price of shares or deferred payments for shares might make attracting outside investors more dif- ficult in the future. This could affect enterprises' long-term prospects, and the state is likely to continue to perform some of the monitoring functions that banks and outside investors would normally do. In this context addressing deficiencies in the credit security and insolvency re- gime becomes especially important. Market participants hope that the new legislation will transform bankruptcy from a purely administrative 67 CORPORATE GOVERNANCE IN CHINA BOX 3.8 BANK OF SHANGHAI: A LEADER IN CORPORATE GOVERNANCE The Bank of Shanghai was established in 1995 through a merger of urban credit cooperatives as part of the reform and development of China's financial sector. The bank's shareholders include the Shanghai municipal government and 13 district governments, which have a 30 percent stake; 11 large SOEs in Shanghai, which hold 8 percent; more than 2,000 small and medium enterprises that hold 28 percent; and 38,000 individuals, including most of the bank's 4,500 employees, with 34 percent. The bank's lending is oriented primarily toward support for local small and medium enterprises. The Bank of Shanghai's strategy for transforming itself into a mod- ern banking institution managed according to international standards and banking best practices focuses on entering into technical assistance arrangements with reputable international banks and attracting strate- gic investors. The International Finance Corporation (IFC) has supported the bank's efforts since 1995 through technical assistance and direct investments. A total of US$1.3 million in wide-ranging technical assis- tance in the areas of corporate governance and risk management was provided through grants from the government of Japan, the European Union, and IFC and was executed by Allied Irish Bank, ABN/AMRO Bank, and IFC. In 1999, IFC made a US$22 million equity investment in the Bank of Shanghai, representing 5 percent of the bank's expanded share capital. Corporate governance practices in the bank have improved dra- matically following the technical assistance and IFC's equity invest- ment. Independent directors were appointed to the board, and the board has become more engaged in active discussions with the man- process subject to quotas and government approvals to a more mar- ket-driven one. Restoring the normal priority of secured creditor claims is critically important. Court-appointed liquidation commissions should not consist mainly of agencies representing enterprises' own- ers (typically the local government) and employees' interests. Options for out-of-court reorganization and secured creditors' self-help should be enhanced. However, banks may still be reluctant to use bankruptcy as a tool if weak capital bases and inadequate loss provisioning rules hinder loss recognition. 68 TRANSFORMED ENTERPRISES agement on the strategic development of the bank. The frequency of board meetings has increased from twice a year to at least four times a year. The board has set up three committees: an audit committee and a compensation committee (both chaired by independent directors), and a risk management committee. Board meetings now include dis- cussions on specific subjects relating to the bank's management and seminars to inform the directors about modern banking concepts and trends. The management team has introduced various improvements in all the operational areas, particularly in credit risk management and internal controls. In 2001 the bank generated profits of about US$120 million. In 2001 the Bank of Shanghai was able to attract two foreign stra- tegic investors: the Hong Kong and Shanghai Banking Corporation and the Shanghai Commercial Bank, the first time foreign commercial enti- ties have invested in a Chinese domestic bank. In connection with this capital increase, the Shanghai municipal government will transfer its entire shareholding to its wholly owned Shanghai State Asset Manage- ment Company, a move that is consistent with the central government's requirement for local governments to transfer their direct equity hold- ings in commercial entities to such state asset management companies. The expectation is that the involvement of the Hong Kong and Shang- hai Banking Corporation and the Shanghai Commercial Bank will ac- celerate the Bank of Shanghai's progress in corporate governance, management, and operations. The evidence indicates that other Chi- nese commercial banks are contemplating similar approaches to strengthen their corporate governance. Allowing creditor banks to hold additional equity stakes, directly or indirectly through affiliates, would deepen banks' governance role and would be in line with global trends toward more universal bank- ing groups. There is a strong economic rationale for allowing banks to convert debt into equity in the case of financial distress. In the presence of underdeveloped capital markets, wider use of subordi- nated debt and quasi-equity instruments such as convertible loans also makes sense to better align banks' incentives with those of sharehold- ers. Such instruments are also likely to promote a more active role by 69 CORPORATE GOVERNANCE IN CHINA banks in corporate governance. Given the information advantages of creditor banks, they could, in theory, develop governance services as a business line. External board membership could be one example. Banks could also offer to help small stockholders with monitoring, proxy voting, and external board membership. Role of Private Equity Investors Highly indebted and often under financial pressures, many transformed Chinese companies are looking for outside financing to realize growth opportunities. Outside equity investors, such as venture capitalists, can play an important role in mitigating agency problems in such closely held corporations. Such investors typically take an active role in struc- turing financial contracts, carrying out preinvestment screening, and providing postinvestment monitoring and advice. The equity alloca- tion, which is typically sizable, provides private equity investors with the incentives to engage in costly support activities that increase up- side values, rather than just minimizing potential losses. Private inves- tors are thus an important complement to the role of creditors in shaping the incentive structure of closely held corporations. Private equity investors, especially venture capitalists, typically use an elaborate set of contracts and instruments to allocate cash and control rights. Empirical analysis of the financial contracting such in- vestors use (Kaplan and Stromberg 2001) revealed that they rely heavily on convertible securities and combinations of multiple classes of com- mon stock and straight preferred stock. Cash flow rights, voting rights, control rights, and future financing are frequently contingent on mea- surable financial and nonfinancial performance. Voting and board rights are frequently structured in a way that gives private investors complete control in the case of poor performance. Contracts also pay a great deal of attention to methods of resolving conflicts of interest by management through vesting and noncompete clauses. Underdevelopment of Private Capital Markets. Private equity mar- kets, especially venture capital, are in an embryonic stage of develop- ment in China. Indeed, offshore venture capital appears to be a far more important source of capital for smaller companies than domes- tic venture capital. At present, there are no regulatory guidelines de- fining the legal and organizational structures for establishing private equity funds. As a result, would-be fund promoters, generally local 70 TRANSFORMED ENTERPRISES governments interested in developing their high-tech sector, often set up limited liability corporations as investment vehicles. These corpo- rations issue shares in exchange for investment, and funds are then pooled and managed by a fund manager. The corporation must abide by the Company Law, which does not permit more than 50 percent of capitalization to be invested in subsidiaries or other legal entities. While this rule was instituted to prevent the siphoning off of company as- sets, it prevents the corporations from investing more than half of their assets in anything other than cash-equivalent securities. Insurance companies and pension funds are not permitted to in- vest in nonlisted securities. Increasingly, large SOEs are among the most active domestic investors in smaller firms. Some of these, espe- cially listed companies, have stepped in to provide venture capital, primarily for high-tech growth companies. Securities firms, asset ex- changes, and trust and investment companies currently play a limited role in facilitating private equity financing. By 2000, China had ap- proximately 180 such venture capital firms with more than RMB 15 billion under management (VCChina, 2001). More than half of the Chinese venture capital firms are established by government entities and one third by state-owned industrial or financial companies. Al- most two-thirds of the Chinese venture capital firms are located in the three cities of Beijing, Shanghai, and Shenzhen. Most venture capital firms invest in companies at different stages of development, but only 6 percent of the Chinese and 11 percent of the foreign funds consider the provision of seed capital. Developing Private Equity Markets. Private equity funds should be developed within a comprehensive legal framework, and transitional arrangements can speed up the process. In November 1999 China issued regulations on Establishing a Venture Investment Mechanism, Several Opinions, which set forth a conceptual framework within which a venture capital industry could develop. The regulations recognize that the current legal and regulatory framework is unsuitable for pro- moting development in this area. Some local governments, led by Shenzhen, have promulgated regulations on venture capital invest- ment. However, no regulatory guidelines are available at the national level that define the legal and organizational structures that can be used to establish private equity funds, known in China as industrial investment funds. In addition, some of the existing laws do not pro- vide an enabling environment for the development of private equity 71 CORPORATE GOVERNANCE IN CHINA markets. In particular, limited partnerships, which have been a popu- lar organizational form of venture capital activities, are not sanc- tioned by the Partnership Law, and the current law restricts all partners to natural persons only. According to the Company law, capital injections have to be paid in full at the time of registration: the concept of callable capital is not explicitly recognized. This de- prives investors in private equity funds of much needed flexibility and a powerful tool to provide incentives for fund managers and control them. The newly drafted Investment Fund Law and the re- cently adopted Trust Law are likely to facilitate the development of private equity and venture capital regulations. The state still plays a ubiquitous role as fund sponsor, investor, and manager. As a result, the pressure to make profits on government- supported investment funds is minimal, which is not in accordance with the principles by which the venture capital industry is supposed to operate, that is, high risks, high returns. Lack of strong profit ori- entation on behalf of government-supported venture capital firms is discouraging private investments in the industry. Without the partici- pation of large amounts of private capital, it is difficult to expect a rapid growth in the industry. The investment industry would benefit considerably if the state acted less as a patron of the companies in which it invests and more as a protector of efficient competitive mar- kets. The Several Opinions regulations broadly define the state's role in promoting a venture capital environment. They clearly emphasize the concept of separating government from business and the government's intent to discourage SOEs from directly investing in high- risk areas. The regulations also refer to the development of high-tech and new technology zones, industrial parks, and other government initiatives. As a transitional step, the state could use indirect mecha- nisms to ensure that venture capital flows are strong, stable, and ac- cessible to a wide range of companies-particularly at the seed stage of development, where the lack of private equity capital is most ap- parent. The experience of the Small Business Investment Corporation program in the United States exemplifies one transitional mechanism for ensuring that small companies with attractive futures, but not the high returns private venture capitalists demand, also have access to prelisting equity capital. Private equity investors and investors in pre-IPO companies in particular, need latitude to structure transactions so that they are op- timal both to issuing companies and to themselves. Risk and return 72 TRANSFORMED ENTERPRISES preferences vary, as does the appropriateness of different securities. This requires the ability for companies to structure investments using a range of securities. A security essential to many private equity transactions but not permitted in China is preferred stock. The cumulative preferred share satisfies the balance of risk and return acceptable to some investors at a level between that of ordinary common shareholders and of bank lenders. The lack of provision for such different classes of shares seems to deny the needed flexibility to financial arrangements by enterprises looking to attract outside investors. In limiting ownership to a single class of shares, policymakers originally intended to create a simple and transparent shareholding environment and to prevent a control- ling class from abusing the rights of others. However, the failure to acknowledge different classes of shares and the different rights associ- ated with such classes will thwart efforts to provide sophisticated fi- nancing strategies for investors in Chinese issuers and hinder the establishment of a basis upon which minority shareholders or inves- tors giving different value can be recognized and protected. Issuers also need to be able to offer investors quasi-debt securities that provide current income as well as the potential for equity appre- ciation in the future. Securities of this kind include bonds that are convertible into shares, currently permitted only for listed companies; bonds that carry "warrants," meaning the right to purchase a fixed amount of shares at a predetermined price in the future; and stock options. The Company Law does not provide any basis for issuing share options and warrants. In particular, it lacks specific regulations regarding authorized but unissued shares or authorized capital in- creases. To the contrary, any single capital increase is subject to gov- ernment approval at the time it is effected. Consequently, a company cannot reserve unissued shares and grant vested rights to acquire such shares in the future. Similarly, there is no obvious way to provide debt obligations to lenders that can be converted by their terms into equity claims against a company. The state of private equity markets depends largely on the level of development of the public equity market, mainly through the provision of exit mechanisms for private equity investors. In this context, the es- tablishment of the Second Trading Board with somewhat relaxed list- ing requirements, simplification of share buybacks, and reduced restrictions on the sale of sponsors' shares is likely to have a profound effect on private equity markets (Gregory, Tenev, and Wagle 2000). 73 CORPORATE GOVERNANCE IN CHINA Closely held corporations with growth opportunities will need to improve their corporate governance practices to take advantage of such opportunities. Some of them may contemplate listing do- mestically and internationally, and should realize that adopting some of the corporate governance practices required for listed companies ahead of time can be tremendously helpful. Thus closely held corpo- rations should strive to improve transparency and maintain simple and transparent organizational structures, maintain a clear business focus, move their accounting practices closer to international stan- dards, use external auditors, and strengthen their boards' indepen- dence by establishing board committees and inviting outside and independent directors. Efforts along these lines will make attracting strategic investors (including foreign strategic partners) easier, and these will, in turn, facilitate further improvements in corporate gov- ernance practices. Not only will strategic partnerships position com- panies better for increased competition in the context of WTO membership, but they are also likely to have a profound impact on other important relationships. For example, they can help control opportunistic behavior on the part of managers and employees, and they make attracting debt financing easier. Conclusion The corporatization and ownership diversification of small and me- dium SOEs have resulted in an ownership structure dominated by managers and employees. The process has not yet transformed insid- ers into genuine owners fully exposed to market pressures. The state remains involved, employees and managers are reluctant to part with their status as state employees and the associated implicit job assur- ance, and the captive source of employee savings has shielded these enterprises from the discipline of capital markets. More important, certain features of the ownership transformation process are likely to make future access to capital markets more difficult and to prevent banks and outside investors from playing an important role in the governance of these enterprises. This underscores the importance of strengthening the core legal rights of employees, creditors, and out- side investors so that they can play a positive role in the corporate governance of small and medium enterprises. 74 4 Ownership and Control of Listed Companies Today's ownership and governance characteristics of listed compa- nies in China are largely shaped by the past incentives structure of the listing process. The government introduced stock markets partly as a means of reforming the state sector, and under the quota system, local governments were responsible for selecting which companies were to be listed. Local governments tended to give preference to companies that were under their control, urgently needed capital infusion, or were otherwise socially or economically important. Such criteria would not necessarily result in the selection of the most dynamic, successful, and high-growth companies. They also created a bias against private sector companies. While the Company Law stipulated various criteria for listing modeled after regulations in successful developed markets, the crite- ria were insufficient to play a screening role. For example, the law allowed issuers who were divested from SOEs or large and medium SOEs to use pro-forma profit records. This provided incentives to es- tablish SOEs for the specific purpose of listing, a trend that came to be known as "packaging for listing." The packaged shell companies of- ten did not have a meaningful track record, and their business models were at times ad hoc. Thus the companies that are listed on China's stock exchanges are mostly SOEs. They have strong links with the government, especially local governments, and their boundaries with their parent groups are relatively new and often artificial. This chapter examines the ownership and control structure of Chinese listed companies and the main corporate governance issues associated with it. It focuses on the board of directors as the main corporate governance mechanism. The discussion is based on a sur- vey of 257 companies listed on the Shanghai stock exchange (see Xu and Wang 1997 for a similar survey). 75 CORPORATE GOVERNANCE IN CHINA Ownership Concentration and Types of Investors Chinese company shares are classified as A shares, B shares, and H shares.' In our sample, 252 companies issued A shares, 21 companies issued A and B shares, 5 companies issued only B shares, and 4 com- panies issued A and H shares. Shares of listed companies are further classified into state shares, legal person shares, and tradable shares. Each type accounts for about one-third of all shares. Shares of the same kind carry the same rights. State shares are held by central and local governments, which are represented by local financial bureaus, state asset management com- panies, or investment companies. State shares can also be held by the parent of the listed company, typically an SOE. They are not tradable. By 1999, 42 percent of the largest shareholders in the sample held state shares, as did 5.1 percent of the second largest shareholders. The state therefore tends to be the controlling shareholder, and is rela- tively rarely the second or third largest shareholder. Domestic institutions such as industrial enterprises, securities com- panies, trust and investment companies, foundations and funds, banks, construction and real estate development companies, transportation and power companies, and technology and research institutes hold legal person shares. These institutions are further classified according to their ownership structure as SOEs, state-owned nonprofit organi- zations, collectively owned enterprises, private enterprises, joint stock companies, and foreign-funded companies. Legal person shares are not tradable. In 1999, in 57 percent of the companies in the sample, the largest shareholder was holding legal person shares. Almost all the largest legal person shareholders are industrial SOEs (table 4.1). Thus in more than 95 percent of the cases, the state is directly or indirectly (through industrial SOEs) in control of listed companies. State and legal person shares can be transferred to domestic insti- tutions upon approval of the China Securities Regulatory Commis- 1. A shares are issued by domestic companies and are held and traded in RMB by domestic investors only. B shares are stocks issued by domestic companies regis- tered on the mainland, but traded in hard currency by foreign investors, including overseas Chinese and individuals and institutions from foreign countries as well as from Hong Kong (China), Macao, and Taiwan (China). Individual domestic investors have been allowed to trade B shares since February 2001. H shares arc issued and listed by domestic companies in Hong Kong (China). 76 OWNERSHIP OF LISTED COMPANIES TABLE 4-1 NATURE OF THE THREE LARGEST SHAREHOLDERS, AT IPO AND 1999 (percentage of all firms) First largest Second largest Third largest Shareholder At IPO 1999 At IPO 1999 At IPO 1999 Industrial SOE 55 57 32 27 26 22 State asset management company 9 7 0 0 0 0 Natural person 0 0 12 8 17 15 Diversified agribusiness 7 8 5 6 5 6 Transportation and tele- communications co. 6 5 3 3 0 0 Commerce entity 5 6 10 8 10 9 Construction and real estate company 4 7 5 3 4 4 Trust and investment company 2 1 12 11 11 I Securities firm 0 0 4 16 5 11 Bank 0 0 0 0 4 3 Foundation or fund 0 0 8 11 8 9 Other 12 9 9 7 10 10 SOURCE: Survey. sion (CSRC). In about 47 percent of the sample companies, nontradable shares accounted for 70 to 90 percent of total shares, and in 41 per- cent of the sample, nontradable shares accounted for 50 to 69 percent of the total. In only 8 percent of sample firms did tradable shares represent more than 50 percent of all shares. Domestic individuals and institutions hold tradable A shares. About 30 percent of all shares are tradable.2 At the end of 1999, of the 30 percent of tradable shares, individuals held 25 percent and institutions held 5 percent.' In addition to state, legal person, and tradable shares, there are the so-called employee shares. The company sells employee shares to 2. According to the Company Law, the shares issued to the general public will amount to 25 percent or more of total shares issued. 3. Article 46 of the Provisional Regulations for Issuance and Trading of Securities specifies that an individual cannot hold more than 5 percent of the shares issued by a single listed company. However, some natural persons indirectly control listed companies through the legal persons shares of parent companies. 77 CORPORATE GOVERNANCE IN CHINA management and employees, typically at a significant discount, at the time of going public. These shares have to be held for 6 to 12 months after an IPO, and can then be sold on the stock exchanges following approval by the securities regulatory authorities. In 1998 the regula- tory authorities issued a circular in relation to discontinuing the issu- ance of employee shares. As a result, the number of employee shares is gradually falling. Some features of the ownership structure are correlated with en- terprise size, the nature of the largest shareholders, and the sectoral affiliation of the listed company. As table 4.2 shows, the percentage of state shares tends to increase with company size and that of individual shareholders tends to decrease. The proportion of tradable shares seems to be higher in companies in which the first largest shareholder holds state shares. For example, among the listed companies with more than 50 percent tradable shares, in only- 6 percent is a legal person (indus- trial SOE) the largest shareholder. In terms of sectors, the state has higher shareholdings in manufacturing and in energy and power. Le- gal persons have controlling positions across all sectors, but their con- trol is especially pronounced in chemicals and conglomerates. Individual investors are relatively better represented in retail and chemi- cals, while institutional investors are conspicuously avoiding retail and chemicals and pharmaceuticals. Ownership in China's listed companies is relatively highly con- centrated. Data from 1999 indicate that the three largest shareholders held, on average, about 58 percent of total shares, of which the aver- age shareholding of the largest shareholders is about 47 percent, of the second largest is 8 percent, and of the third largest is 3 percent. In TABLE 4-1 OWNERSHIP STRUCTURE OF COMPANIES AND SIZE, L999 Indi- Institu- Company size Legal Employ- vidual tional (RMB) State persons ees (tradable) (tradable) < 500 million 13 50 3 29 5 500 million - 1 billion 24 40 2 28 5 > 1 billion - 1.5 billion 26 44 3 20 6 > 1.5 billion 31 40 2 22 4 SOURCE: SurveV. 78 OWNERSHIP OF LISTED COMPANIES almost 49 percent of sample firms, the three largest shareholders ac- counted for 60 to 80 percent of all shares. This high concentration of ownership, combined with the relatively small portion of tradable shares, implies that few, if any, of China's listed companies have con- testable control. Overall, the ownership structure is relatively stable. The con- centration, although quite high, has shown a slight tendency to de- crease: the share of the three largest shareholders declined from 61 percent at IPO to 58 percent in 1999. However, the share of compa- nies with highly concentrated ownership has shown a tendency to increase in recent years, although companies with the state as con- trolling shareholder have tended to decrease over time. For example, in 47 percent of the listed companies the largest shareholder was holding state shares at IPO, and in 49 percent of the listed compa- nies the largest shareholder was holding legal person shares (table 4.3). By 1999 the percentage of companies with the state as largest shareholder had dropped to 42 percent and the percentage of com- panies with legal persons (industrial SOEs) as the largest shareholder had increased to 54 percent. The total shares of listed companies are about equally divided between state shares, legal person shares, and tradable shares. Be- tween the time of the IPO and 1999, the percentage of state shares has TABLE 4-3 DIsTRIBUTIoN OF SURVEYED COMPANIES By TYPES OF LARGEST SHAREHOLDERS, AT IPO AND 1999 (percent) First largest Second largest Third largest Share type At JPO 1999 At IPO 1999 At IPO 1999 State shares 47 42 6 5 1 4 Legal person shares 49 54 64 58 17 53 Employee shares 0 0 4 1 4 1 A shares 3 2 22 26 28 31 B shares 0 0 3 5 3 6 H shares 0 0 0 1 1 1 Foreign legal persons 0 1 0 0 1 0 Other 0 0 0 4 0 4 SOURCE: Survey. 79 CORPORATE GOVERNANCE IN CHINA FIGURE 4.1 TRENDS IN COMPANY OWNERSHIP STRUCTURE, AT 11O AND 1999 Percent 100 O State 80 - Legal persons i Eniployecs 60 InstitUions Individuals 40- 20 - 0 IP1O 1999 SOURCF.: Survey. fallen while the percentage of legal person shares and tradable shares has increased (figure 4.1). In 28 percent of the companies surveyed, the largest shareholder had changed since the IPO, with most of the changes taking place between 1998 and 1999. In almost 82 percent of these companies, the change was associated with the replacement of the chair of the board of directors, and in 44 percent with the replacement of the general manager. In addition to the transfer of control, new share issues are an- other channel through which the ownership structure evolves. For example, one possible reason for the decline in state shareholding is that most listed companies pay dividends not in cash but in rights issues. In general, legal persons are financially able to accept rights issues, while the government or its agencies prefer cash payouts. Ownership and Corporate Governance Issues The information in tables 4.4 and 4.5 allows us to compare the own- ership structure of Chinese listed companies with that in some West European and East Asian countries and the United States. In terms of 80 OWNERSHIP OF LISTED COMPANIES TABLE 4.4 CONCENTRATION OF COMPANY OWNERSHIP, SELECTED COUNTRIES, 1998 (percentage of shareholding) 2nd 3rd 4th-10th Country Largest largest largest largest Austria 82.2 9.5 1.9 6.5 China 47.0 8.0 3.0 - France 56.0 16.0 6.0 5.0 Italy 52.3 7.7 3.5 5.1 Netherlands 28.2 9.2 4.3 7.1 Spain 38.3 11.5 7.7 10.3 United Kingdom 14.0 8.3 6.1 9.2 2nd and 3rd 4th and 5th 6th-I 0th Country Largest largest largest largest Belgium 55.8 6.9 0.6 0.2 Germany 59.7 8.6 2.6 0.3 United States 22.8 9.5 7.5 3.8 Ist-5th Country Largest largest Indonesia 48.2 67.5 Korea, Rep. of 20.4 38.5 Malaysia 30.3 58.8 Philippines 33.5 60.2 Thailand 28.5 56.6 - Not available. SOURCEs: ADB (1999); Tricker (1999). concentration, China is somewhere in the middle. The ownership struc- ture is concentrated, but at levels similar to those in most West Euro- pean countries. What differentiates China's ownership structure is the identity of the controlling shareholders (table 4.5). A similarity with ownership structures in some West European and East Asian countries is the dominant position of other corporate entities as shareholders. In terms of types of largest shareholders, China is differentiated by the absence of significant ownership by individu- als and families, the negligible role of financial institutions and insti- tutional investors, and the large state role. These features have a direct bearing on the types of corporate governance issues that China faces. 81 CORPORATE GOVERNANCE IN CHINA TABLE 4.5 TYPES OF LARGEST SHAREHOLDERS, SELECTED COUNTRIES, 1997 (percentage of shareholders) Indi- Holding viduals Insur- Invest- and in- and ance ment dustrial Company Country families Banks companies funds companies State directors Austria 38.6 5.6 0 0 33.9 11.7 0 Belgium 15.6 0.4 1.0 3.8 37.5 0.3 0 China 0 0 0 1.0 57.0 42.0 0 France 15.5 16.0 3.5 0 34.5 1.0 0 Germany 7.4 1.2 0.2 0 21.0 0.7 0 Italy 68.6 7.2 0 0 24.2 0 0 Korea, Rep. of 60 8.5 2.0 6.1 - 1.5 - Malaysia 4.8 - - - 45.6 17.2 - Netherlands 10.8 7.2 2.4 16.1 10.9 1.3 0 Philippines 5.5 2.1 0.2 13.5 4.9 2.6 - Spain 21.8 6.6 8.8 0 32.6 0 0 Thailand 35.2 2.2 - - - 2.5 - United Kingdom 2.4 1.1 4.7 11.0 5.9 0 11.3 - Not available. SOURCE: ADB (1999); Tricker (1999); survey. Perhaps the most important implication of the dominant role of state ownership in China's listed companies is the control the govern- ment can exert over management appointments and incentives, and thereby over companies' behavior. Most corporate managers still aspire to a civil service rank and are concerned about how their superiors in the political and administrative hierarchy assess their performance. This assessment may be quite arbitrary or subjective and be based on such indicators as profits, political correctness, and the discharge of social obligations. Furthermore, local governments may have incen- tives that are not aligned with the plans of companies that operate on a national or international scale. As noted, the dominant position of corporate entities as control- ling shareholders is not unique to China. However, in China, the na- ture of the listing process in the past compounds the risks of conflicts between controlling and minority shareholders. Listing and parent companies are often in the same business sector and may compete with each other, have business transactions with each other, or share resources and functions. In some cases, the listed company may de- 82 OWNERSHIP OF LISTED COMPANIES pend on the rest of the group for distributing products or supplying raw materials. Senior managers often work for both the listed and unlisted parts of the group. This type of interdependence between listed companies and their parent firms creates fertile ground for agency problems. Corporate control mechanisms and shareholder activism can do little to alleviate such agency problems under the existing ownership structure. The high degree of ownership concentration and the nontradability of more than two-thirds of the shares imply a low contestability of control. In addition, tradable shares are held largely by individuals, who have few incentives and resources to perform monitoring functions. Board of Directors The board of directors is the critical link between ownership and cor- porate governance. This is the setting where governance takes place and where some of the solutions to corporate governance problems can be sought. Ownership and Control. According to the survey, shareholders ap- point 76 percent of the directors of listed companies (table 4.6). Hold- ers of state-owned legal person shares are the most influential, selecting 48 percent of all directors, followed by owners of state shares at 21 percent. Thus directly or indirectly, the state is in absolute control, selecting almost 70 percent of all directors. State enterprises appoint 45 percent of all the directors appointed by shareholders (table 4.7). In line with their insignificance as share- holders, financial institutions play a relatively minor role in the selec- tion of directors. Trust and investment companies, securities companies, and banks together account for the selection of 1.3 percent of execu- tive directors and 9.3 percent of nonexecutive directors. Private enter- prises appoint 1.5 percent of all directors appointed by shareholders. Those holding state shares and tradable shares tend to appoint a larger proportion of executive than nonexecutive directors.4 As table 4.7 4. Public servants from government departments are not permitted to become directors of listed companies and banks are not allowed to be shareholders. How- ever, directors from these institutions do exist. To some extent this reflects the lack of effectiveness of the regulations. It also reflects some banks' recent practice (in the past two years) of engaging in debt-equity swaps. 83 CORPORATE GOVERNANCE IN CHINA TABLE 4.6 COMPANY OWNERSHIP AND CONTROL (percent) Control Shareholder type Ownership (board seats) State 24 21 Legal persons 44 48 Employees 2 3 Tradable shares 30 4 Total 100 76 SOURCe: Survev. shows, the owners of state shares select a higher portion of executive directors, while the owners of state-owned legal persons shares, who typically represent the interests of the parent SOE, select the majority of nonexecutive directors. Like ownership, control is highly concentrated. The largest share- holder accounts for slightly less than 50 percent of all shares but con- trols more than 50 percent of board seats. The average share of the three largest shareholders is 59 percent, but they appoint 79 percent of the directors. Furthermore, considerable disparity is apparent TABLE 4-7 SELECTION o, DiRECTORS AND SUPERVISORS OF LISTED COMPANIES (percent) Directors Non- Super- Total Fxecutive executive visors Owners 76 - - 50 State shares 28 36 16 25 State-owned legal person shares 45 44 54 44 Public legal person shares 18 13 27 12 Internal employee shares 3 1 11 Publicly circulating shares 6 5 2 7 - Not available. SouRcE: Survey. 84 OWNERSHIP OF LISTED COMPANIES FIGURE 4.2 DISCREPANCY BETWEEN CASH fLow AND CONTROL RIGHTS IN LISTED COMPANIES Control rights (percent) 60 Largest shareholder 50 - 40 - 2nd 30 - largest 3rd share- largest holder 20 - share- holder 10 - 0 0 10 20 30 40 50 Cash flow rights (percent) SOURCE: Survey. between the shares of board seats and the ownership of different share- holders, that is, between control and cash flow rights. A comparison of the control and cash flow rights of the largest three shareholders shows that the discrepancy is higher for the second and third largest shareholders than for the largest shareholder, conditional on the latter's having obtained majority control (figure 4.2). The marginal value of control diminishes after majority control has been obtained. This can be seen in table 4.8, which compares the distribution of the single largest shareholder with board control. The share of board seats con- trolled by the largest shareholder is higher than that shareholder's ownership share, but this situation is reversed after the shareholder obtains a majority stake (more than 50 percent). Thus to a significant extent, parent companies control the boards of listed companies. Few board seats are available for nonshareholders, and the notion of independent directors is new to most listed compa- nies. The directors appointed by nonshareholders account for 24 per- cent of total directors. The executive directors appointed by nonshareholders are mainly recommended by company staff, but are 85 CORPORATE GOVERNANCE IN CHINA TABLE 4.8 DISCREPANCY BETWEEN SIZE OF THE LARGEST SHAREHOLDING AND CONTROL RIGHTS Percentage of Average directors Number of size of the appointed Largest share companies largest share by largest (percent) in sample (percent) shareholder > 80 5 87 66 50-80 77 63 62 20-49 78 34 46 < 20 11 16 31 Total 171 48 53 SouRcE: Survey. sometimes appointed by the government. The independent directors are mainly recruited by the companies. They tend to be former advi- sors to the companies and are usually elected because of their reputa- tions and professional expertise. Board Composition and Trends. According to the Company Law, a board of directors should consist of 5 to 19 directors. The average size of boards in 1999 was 9.9, compared with 10.1 in 1996 (table 4.9). Two-thirds of all directors are executive directors. Only 54 directors, or 3.1 percent of all directors, have some degree of independence (fig- ure 4.3).' About half of the executive directors take senior manage- 5. The U.S. National Association of Corporate Directors (1996, pp. 9-10) defines an independent director is one who (a) has not been employed by the company in an executive capacity within the last five years, (b) is not affiliated with a com- pany that is an adviser or consultant to the company, (c) is not affiliated with a significant customer of or supplier to the company, (d) has no personal services contracts with the company or with a member of the company's senior manage- ment, (e) is not affiliated with a not-for-profit entity that receives significant con- tributions from the company, (f) has not had any business relationships with the company other than service as a director within the last five years, (g) is not employed by a public company for which an executive officer of the company serves as a director, (h) has not had any of the relationships described above with any affiliate of the company, and (i) is not a member of the immediate family of any person described above. 86 OWNERSHIP OF LISTED COMPANIES FIGURE 4.3 DISTRIBUTION OF COMPANY DIRECTORS BY TYPE OF APPOINTMENT, 1999 Executive directors - Executive directors appointed by appointed by shareholders nonshareholders 43% 24% Nonexecutive directors appointed by shareholders Independent 30% directors 3% SOURCE: Survey. ment positions. Comparatively fewer directors hold professional po- sitions such as chief engineers, advisers, and economists. In about 22 percent of the companies surveyed, the board chair is also the general manager. In some 33 percent of the sample companies the chair is not a company employee. In the majority of listed compa- nies, about 45 percent, the board chair and general manager are two different people, but the chair does hold a position in the company. Table 4.9 shows trends in board size and composition. The aver- age share of executive directors decreased from 56 percent in 1996 to 49 percent in 1999, and the average share of directors appointed by shareholders increased from 65 percent in 1996 to 70 percent in 1999 (table 4.9). A significant number of directors come from connected companies, and their percentage representation appears to be increas- ing. This is also true for associated companies that either hold shares of the listed company or provide credit to or have business transac- tions with the listed companies. While the average number of executive directors on boards has been on the decline since 1996, as figure 4.4 shows, the concentration of control, as measured by the percentage of seats controlled by the three largest shareholders, has increased over time (see figure 4.5). The increase has been especially pronounced for small boards. 87 CORPORATE GOVERNANCE IN CHINA TABLE 4.9 TRENDS IN COMPANY BOARD SIZE AND COMPOSITION, 1996-99 Category 1996 1997 1998 1999 Board size (average number) 10.1 9.9 9.9 9.9 Director affiliation (average number) Holding management position 5.6 5.3 5.0 4.9 Appointed by shareholders 6.6 6.5 6.8 6.9 Representing nonbank financial institution (shareholder) 1.4 1.5 1.4 1.3 Connected companies (shareholder) 3.8 4.3 4.9 4.9 Connected companies via business transactions 2.5 2.9 3.5 3.2 SouRcil: Survey. Directors' Qualifications and Careers. Almost 60 percent of the direc- tors of listed companies have a graduate or higher degree (table 4.10). Nonexecutive and independent directors tend to be more highly edu- cated than executive directors appointed by shareholders. About half of all executive directors, two-thirds of nonexecutive directors, and FIGURE 4.4 CHANGES IN SIARE OF EXECUTIvE DIRECTORS AND BOARD SIZE, 1996-99 Percent geo iectos 60 50 30 10 .-10 1 199 -1 1996 ~1997 1:1 1998 1999 Board size Souic: Survev. 88 OWNERSHIP OF LISTED COMPANIES FIGURE 4.5 PERCENTAGE OF BOARD SEATS CONTROLLED BY THE TIREE LARGEST SHAREHOLDERS AND BOARD SIZE, 1996-99 11evce"t 80 70 If .50 40 30 20 10 5-9 0 10-14 1996 - 1999 Board size SOURCE: SUrvey. TABLE 4.10 EDUCATION OF COMPANY DIRECTORS (percentage of directors) Executive directors appointed Other Non- Inde- Highest level by share- executive executive pendent of education holders directors directors directors Total Postgraduate or higher 14 10 20 43 16 Graduate 43 37 47 33 42 College degree 35 42 28 22 34 High school graduate 7 10 6 2 7 Middle school and lower 0 1 0 0 1 SOURCE: Survey. 89 CORPORATE GOVERNANCE IN CHINA almost all independent directors hold degrees in management-related subjects or finance and accounting. An insignificant number of direc- tors have studied law. In terms of background, most directors have worked in the fields of engineering, marketing, and sales. Independent directors have more experience in technological research, education, and Chinese Coin- munist Party and Chinese Communist Youth organizational work. Most executive and nonexecutive directors are former SOE employ- ees, while independent directors are drawn mostly from government departments, research institutions, and universities (table 4.11). This accounts for the description of listed companies as "old wine in new bottles," that is, on the surface a new corporate governance frame- work appears to be in place, while in reality companies still operate more or less like the old SOEs. Most executive directors were with the listed companies at the time of the IPO, and most of them had previously worked in the firm's parent company (table 4.12). Regarding nonexecutive directors, 32 TABLE 4.11 COMPANY DIRECTORS' FORMER EMPLOYERS (percentage of directors) Executive directors appointed Other Non- Inde- by share- executive executive pendent Employer holders directors directors directors Macroeconomic government department 9.0 5 16 24 Government ministry 15.0 12 24 32 Research institute or university 15.0 9 15 33 Financial institution 3.0 1 13 24 SOE 76.0 73 73 43 Collectively owned enterprise 9.0 8 7 6 Private enterprise 6.0 5 8 4 Joint venture 11.0 6 10 4 Other listed company 2.0 1 6 2 Foreign company 0.4 1 4 2 No-E: Numbers exceed 100% because individuals can have different employers over the course of their career. SOURCE: Survey. 90 OWNERSHIP OF LISTED COMPANIES percent were with the company at the time of the IPO and 16 percent had worked in the parent company. More than one-third of all direc- tors also sit on other boards. This percentage is higher for nonexecutive and independent directors (56 and 43 percent, respectively), and lower for executive directors not appointed by shareholders (25 percent). Director Selection. Chinese corporations still lack nominating com- mittees for directors and corporate governance committees. Listed com- panies do not disclose their procedures for nominating directors or their corporate governance principles. Most directors believe that their companies have internal criteria for selecting directors. When surveyed they cited management experience followed by professional expertise and reputation as the main criteria. Other criteria cited include share ownership and personal connections. The Company Law stipulates that the shareholders' general meet- ing is responsible for selecting and removing directors, but it does not stipulate who is responsible for nominating directors. International practices typically include a nominating committee under the board- TABLE 4.12- CAREER ROUTES OF COMPANY DIRECTORS (percentage of directors) Executive directors appointed Other Non- by share- executive executive Category holders directors directors With the company at IPO 78 87 32 With the predecessor company 59 69 16 Used to be manager of the predecessor company 32 27 9 Used to be an official in the relevant government department 8 6 10 Was employed at the shareholding company 48 26 53 Used to be a manager in another company in the same industry 18 14 19 Used to be a manager in another industry 18 14 32 NOTE: Numbers exceed 100% because individuals can have different employers over the course of their career. SOURCE: Survey. 91 CORPORATE GOVERNANCE IN CHINA that is composed mainly of outside or independent directors who do not hold a position in the company other than as board directors- that formulates selection criteria and nominates new directors. In China, large shareholders nominate new directors in 57 percent of listed companies, the board of directors does so in 34 percent of com- panies, the chair of the board in 6 percent of companies, and existing directors in 3 percent of companies. The May 2001 CSRC rules on independent directors stipulate that the board of directors, the board of supervisors, or any share- holders who separately or jointly hold 5 percent of the shares in the listed company can nominate candidates for independent directorships. In all cases the nominees must consent to being nominated. The final decision is made by a vote at the shareholders' meeting. Compensation and Other Incentives. Most directors are paid less than RMB 50,000 per year (table 4.13). None of the independent directors receive more than RMB 50,000 per year. Listed companies provide most executive directors with such benefits as a company car and a house and independent directors with allowances. Few nonexecutive directors receive benefits from the company. Bonuses average 24 per- cent of directors' annual incomes, more for executive directors and less for independent and nonexecutive directors. Simple tests of the survey data failed to identify any statistically significant correlations between directors' compensation and various measures of company performance. TABLE 4.13 COMPANY DIRECTORS' ANNUAL INCOME, 1999 (percentage of directors) Executive directors appointed Other Non- Inde- by share- executive executive pendent Salary holders directors directors directors Total < RMB 25,000 40 47 70 89 46 RMB 25,000- RMB 50,000 36 38 22 11 35 > RMB 50,000- RMB 80,000 14 7 5 0 11 > RMB 80,000 9 8 3 0 8 SoURCE: Survey. 92 OWNERSHIP OF LISTED COMPANIES TABLE 4-14 COMPANY DIRECTORs' SHAREHOLDINGS, AT IPO AND 1999 Executive directors Not Appointed appointed Non- Inde- by share- by share- executive pendent Time Shareholding holders holders directors directors At IPO RMB value of shares 48,410 40,976 2,153,316 15,711 Percentage of shares 0.004 0.006 0.180 0.002 1999 RMB value of shares 116,569 180,638 2,401,174 33,371 Percentage of shares 0.006 0.008 0.140 0.001 Shares/annual income 2.9 4.9 77 1.3 SOURCE: Survey. Nonexecutive directors hold more shares than the other three types of directors, both at IPO and subsequently (table 4.14), while independent directors hold the smallest amount of shares. Directors' shareholdings are increasing rapidly, with the highest increase for ex- ecutive directors and a moderate increase for nonexecutive directors. Ilowever, this largely reflects the internal distribution of shares prior to an IPO, a practice that has recently been restricted. Functioning of Boards of Directors. Listed companies average 4.2 board meetings per year, significantly less than their counterparts in industrial countries.' In developed market economies, listed compa- nies generally establish special committees under the board of direc- tors to ensure that the board functions properly.7 The survey revealed 6. Based on a sample of 1,700 public U.S. companies, Bhagat, Carey, and Elson (1999) found that U.S. firms averaged 7.2 meetings per year, with smaller firms having fewer meetings than larger firms. 7. Typically these committees are (a) a nominating committee responsible for rec- ommending candidates for election as directors; (b) a compensation committee that reviews and approves compensation arrangements for management; (c) an audit committee responsible for reviewing the audited financial statements pro- duced by independent auditors and overseeing the company's financial reporting process; (d) a finance committee that reviews the company's investments, capital requirements, and resource allocations; and (e) an executive committee respon- sible for approving major decisions taken between board meetings. Many compa- nies also have public policy, public responsibility, and environmental protection committees. 93 CORPORATE GOVERNANCE IN CHINA that only 5.4 percent of the companies have established such commit- tees and only 14 percent plan to set up such committees. In those companies that do have committees, these are usually an investment or finance committee, an audit committee, a financial management committee, and/or a strategy committee. Among the committees established, investment committees and financial management committees are composed predominantly of executive directors, while strategy committees have a higher percent- age of nonexecutive and independent directors. The main functions of the committees focus on decisions concerning major investment projects. Their supervisory and auditing functions are at an early stage of development. Most listed companies do not have a system in place for estab- lishing board committees. The main reasons for this are (a) the boards of directors' relative lack of independence, (b) the prevalence of in- sider control, and (c) the lack of independent directors who are famil- iar with the legal aspects of business operations. Some special committees cannot function meaningfully in the absence of indepen- dent directors. Table 4.15 compares board structures and directors' characteristics' in selected economies, including China. China differ- entiates itself by the relatively low percentage of nonexecutive direc- tors and the extremely low percentage of companies with board committees. Recent measures to strengthen the independence of boards of di- rectors focus on increasing the number of external and independent directors, on upgrading directors' professional qualifications, and on standardizing the functioning of boards. For instance, as of May 30, 2001, CSRC rules on establishing an independent board of directors in listed companies require that at least one-third of the board consist of independent directors, including at least one accounting profes- sional. The CSRC also requires the independent directors to work for the listed company for no less than 15 hours a year. The board of directors, the board of supervisors, or any shareholders who sepa- rately or jointly hold 5 percent of the shares of the listed company can nominate independent director candidates. Furthermore, two or more independent directors can now convene extraordinary general meet- ings, and in the case of companies listed overseas, they can report directly to the shareholders' meetings, the CSRC, and other relevant government department regarding the affairs of the listed company. Independent directors are now explicitly instructed to safeguard the 94 TABLE 4.15 COMPANY BOARD STRUCTURE AND DIRECTOR CHARACTERISTICS, SELECTED ECONOMIES Percentage of com- Manage- Chairman/ Non- Average panies Nomi- Remuner- ment and CEO Average executive Worker age of with Audit nation ation executive separate board directors represen- directors board commit- commit- commit- commit- Economy people size (%) tation (years) committees tee tee tee tees Board of directors Australia Very high 8 75 No 55 85 Yes Yes Yes No Belgium High 15 78 - 56 65 Yes No Yes No Brazil Low 6 60 No 57 Low Yes No Yes Yes Canada 66% 13 80 No - 100 Yes Yes Yes Yes China 79% 10 50 Yes 47 5 Yes No No Yes France Nil 13 82 Yes 59 77 Yes Yes Yes No Hong Kong, China Very low 8 15 No - Low Yes No No No Italy 100% 11 73 No 57 40 No No No Yes Singapore Low - - No - Low Yes No No - South Africa 50% 13 60 - 53 100 Yes No Yes No Spain 37% 12 71 No 54 57 Yes No No Yes Sweden High 9 85 Yes 56 Low No Yes No No Switzerland 63% 5 89 No 60 67 No No No Yes Taiwan, China Low - - No - - - - - United Kingdom 90% 12 50 No 56 100 Yes Yes Yes No United States 15% 13 77 - 61 100 Yes Yes Yes No Supervisory board France 100% 12 92 Yes 59 Low Yes No No No Germany 100% 15 100 Yes 59 None No No No Yes Netherlands 100% 7 100 Yes 61 54 Yes No Yes Yes - Not available. SOURCE: Tricker (1999); Survey. CORPORATE GOVERNANCE IN CHINA company's overall interests, especially the rights and interests of small and medium shareholders. According to a recent report (Shanghai Securities News, July 24, 2001), 204 listed companies have already hired 314 independent directors. Compared with practices in other markets, Chinese boards have relatively little decisionmaking power within the existing legislative framework, while government ministries and commissions and secu- rities regulatory authorities enjoy substantial decisionmaking power. As table 4.16 shows, for about 20 percent of China's listed compa- nies, the board has full powers in matters regarding finance and in- vestment, while in about half the companies the board has decisive influence over these matters. Some boards, typically those dominated by insiders, are trying to increase their autonomy from shareholders in relation to financial matters (box 4.1) One of the major functions of a board of directors is to design and enforce management contracts. Table 4.17 shows average annual incomes and shareholdings of senior management in listed compa- nies. Salary is still the main form of management compensation, and most corporate managers still aspire to a civil service rank. Thus their remuneration is explicitly or implicitly benchmarked against civil ser- vice wages and they aspire to being promoted to higher civil service positions (although this attitude does appear to be changing, espe- TABLE 4.16 THE ROLE OF COMPANY BOARDS IN FINANCIAL AND INVESTMENT DECISIONS (percentage of listed companies) Sub- Decisiue stantial Sone Full influ- influ- influ- No Conpany action powers ence ence ence power Adjustment of equity structure 10.5 52.9 12.8 3.9 1.6 Dividend distribution 10.9 59.1 13.6 1.9 0.5 New capital investment 21.8 49.8 9.7 2.7 0.7 Sale of assets 20.6 48.6 10.9 1.6 0.8 Borrowing for fixed asset investments 25.7 38.1 14.8 3.9 0.8 Mergers and acquisitions 10.9 56.4 12.1 1.2 0.9 Company's strategic planning 35.4 44.4 6.6 1.2 0.4 SOURCE: Survey. 96 OWNERSHIP OF LISTED COMPANIES BOX 4.1 EXPANDED ROLE FOR BOARDS Recent reports from listed companies suggest that more boards are ask- ing shareholders for greater autonomy in finance and investment mat- ters. The rationale is typically found in the dynamic nature of markets, which require quick responses to business opportunities. For other com- panies that are reorganized from SOEs, the main reason is to simplify procedures. In turn, some boards have authorized general managers to handle asset disposal, investments, and guarantees if individual deals are below some prcspecified amount, typically RMB 10 million. As boards become more empowered, more comprehensive supervision of boards by the relevant regulators and the market community is called for to prevent board members from abusing their positions and harm- ing the interests of medium and small shareholders. SOURCE: Shanghai Securities News (January 14, 1999). cially in Guangdong). As a result, salaries of different categories of managers are generally low and undifferentiated between, as well as within, companies. Not surprisingly, tests on the survey data for pos- sible correspondence between management compensation and com- pany performance as measured by net assets per share, return on assets, and. Tobin's Q failed to identify any relationship.' Senior management hold mostly employee shares issued at the time of the IPO. With the authorities discouraging the issuance of employee shares, bonuses have become practically the only incentive component in the compensation contract. While some listed compa- nies and shareholders have conducted some experiments (see box 4.2), the lack of legislative support has meant that stock-based incentives, including stock options, are not being developed. Given the lack of strong incentives linked to share performance, corporate managers are particularly sensitive to their performance 8. Tobin's Q is defined as the ratio between a company's market value and the replacement value of its physical assets. 97 CORPORATE GOVERNANCE IN CHINA TABLE 4.17 MANAGERIAL COMPENSATION AND SHAREHOLDING AT SURVEYED COMPANIES Executive deputy Deputy Assistant General general general general Category manager manager manager manager Compensation (percentage of managers) < RMB 25,000 28 41 39 53 RMB 25,000 - < RMB 50,000 36 39 37 29 RMB 50,000 - RMB 80,000 16 8 15 9 > RMB 80,000 20 12 9 9 Shareholding RMB value of shares at IPO 40,859 53,474 52,583 26,127 RMB value of shares in 1999 136,860 112,155 87,471 21,686 SOURCE: Survey. assessments by their superiors in the political and administrative hier- archy. In many cases the government still evaluates companies based on their total profits and taxes paid. Thus managers' incentives are not linked to their companies' return on equity or earnings per share growth. As a result, Chinese companies want to issue new shares as often as possible to increase investment, and correspondingly, total profits and taxes. Listed companies also tend to overstate their earn- ings as they aim to fulfill political expectations, for example, the profit and tax targets set by city governments. This behavior is quite differ- ent from the behavior of private companies, which typically try to understate profits so they can evade tax payments. The board exerts full powers in the appointment of general man- agers in about 70 percent of the companies surveyed. Board secretar- ies' perceptions that the board of directors does not appoint all the general managers is probably because the secretaries consider that large shareholders or government agencies as large shareholders in- tervene. While the Company Law spells out the relationship between shareholders' meeting, boards of directors, and management relatively well, in reality, large shareholders often overstep their boundaries and exercise effective control. Board chairs usually have good government connections and exert genuine control over shareholders' meetings as 98 OWNERSHIP OF LISTED COMPANIES BOX 4.2 STOCK AWARD EXPERIMENT, WUHAN In July 2000, the Wuhan State Asset Management Company gave shares to the chairs of three listed companies as part of their annual compensa- tion. For example, the income of the chair of Wuhan Zhong Shang was RMB 167,000, with RMB 70,000 in 8,000 company shares. According to the internal regulations of Wuhan State Asset Man- agement Company, a company chair is compensated in the form of an annual salary that consists of the following four components: * Basic salary, which is based on the company's previous year's results and typically amounts to RMB 18,000 to RMB 42,000 for a company with a profitable track record. * Seniority income, which usually ranges from RMB 2,400 to RMB 19,200 and depends on the length of service. * Special annual bonus, which is a reward for consistently good company performance, that is, a 20 percent increase in net profits for three consecutive years. * Risk income, which is also a reward for the company's perfor- mance based on reaching predetermined targets. About 30 percent of the risk income is in cash, and the rest is retained for conversion to company shares. This 70 percent of the risk income will be converted to company shares based on the average share price the month after the release of the annual report. For an agreed period the shares are not tradable and the State Asset Management Company exercises the vot- ing rights. However, the recipient is entitled to dividends and rights issues, if any. representatives of the largest shareholder. The state's dual role as owner and regulator implies that political intervention is likely. Managers and boards of directors tend not to resist arbitrary government intru- sion into all areas of business operations. Board of Supervisors As conceived of in China, the board of supervisors is unique: a mix- ture of the German-style supervisory committee and China's tradi- tional concept of employees as masters of enterprises. The Company 99 CORPORATE GOVERNANCE IN CHINA Law does not specify the proportion of shareholders' representatives and employees' representatives on supervisory committees, but just requires that corporate charters properly stipulate the proportion. Theoretically, shareholders could specify a low proportion of employee representatives on the supervisory committee. However, in reality, because of the SOEs' traditional business practices, shareholders ap- point only about half of the supervisors, but especially in the case of holders of legal person shares (SOEs), appoint significantly more chairs and vice chairs. Leaders of party committees tend to assume the posi- tions of chair and vice chair in almost all listed companies. Unions are not represented on supervisory boards to any significant extent. The majority of supervisors hold political science degrees, with few holding degrees in economic management or engineering. Like directors, most supervisors took office at the time of the IPO, and many used to have positions in the parent companies. The proportion of supervisors coming from government authorities that supervised the parent companies is higher than in the case of directors. Overall, supervisors have less business and outside company experience than directors. Like directors, most supervisors come from SOEs. Supervisors' average salaries are slightly lower than those of di- rectors, and supervisors typically hold shares. The shareholdings of supervisory board chairs and vice chairs have tended to increase, while those of other supervisory board members decreased slightly in 1999 compared with their holdings at the time of the IP"O. Supervisors generally meet less often than boards of directors and their meetings are less well attended. Supervisory boards are more "decorative" than functional. Their published announcements indicate that they rarely contest decisions made by boards of direc- tors and company executives. The lower-quality and less professional experience of supervisors compared with directors and managers has led to supervisors' inability to actually supervise directors and man- agers. Supervisors' main sources of information are attendance at board meetings as nonvoting participants and the working reports of the chair of the board of directors and the general manager. In addition, boards of supervisors lack the finance and audit commit- tees that are usual in industrial countries. The Company Law does not stipulate that boards of directors and management have to re- port regularly to the supervisory board. In addition, supervisors are not involved in the selection of directors and managers and have no means of disciplining them. 100 OWNERSHIP OF LISTED COMPANIES Agency Problem of the Controlling Shareholder As noted earlier, two main corporate governance problems arise from the current ownership structure of listed companies in China. One, the agency costs of the controlling shareholder, is related to the con- centration of ownership. The other, the government's poor capacity to maximize shareholder value, is related to the identity of owners. Most recent initiatives have focused on attempts to control the first problem. Relatively few, if any, measures have directly addressed the second issue. However, as the two issues overlap because the govern- ment is directly or indirectly the controlling owner in most listed com- panies, measures aimed at the controlling shareholder problem have indirectly affected the ability of government agencies to exercise dis- cretionary powers in their capacity as owners. Documented abuses by controlling shareholders include soft loans from listed companies on a long-term basis; the use of listed companies as guarantors to borrow money from banks; and the sale of assets to listed companies at unfair prices, usually without an appraisal by an independent evaluator. Given the historical relationship between listed companies and unlisted parent companies, the latter implicitly assume that listed companies will and should help a parent company if the need arises. Similarly, if the listed company comes under pressure-for ex- ample, if it has to satisfy an earnings requirement for a share placement- it may call on the parent company for help, for instance, by buying assets from the listed company to create exceptional gains for the latter. Interference by the parent company may also include transferring and appointing listing company personnel at will. This type of interference has often resulted in major difficulties for listed companies. As some listed companies have tens of thousands of individual investors, abuses of this sort can have a major impact on social stabil- ity and on market confidence. The CSRC has been the most active in tackling the problem from the side of listed companies. Recent CSRC measures have focused on segregating the management of listed com- panies from the management of their controlling institutional share- holders, on prohibiting or restricting certain connected transactions, 9. Note, however, that government agencies have a wide arsenal of instruments to affect a company's behavior. 101 CORPORATE GOVERNANCE IN CHINA and on strengthening managements' independence from their control- ling shareholders and the boards of directors' independence from com- pany management and main shareholders. CSRC regulations now require that the management of listed com- panies (including boards of directors, senior management, and finan- cial and marketing departments) be segregated from the management of their institutional controlling shareholders. They include provisions stipulating that listed companies cannot have more than two senior officers from the management of the controlling shareholder acting as the chair, vice chair, or executive director of the listed company at any time. They also prohibit the appointment of any officer from the man- agement of the parent company to a senior management position in the listed company. Since March 1999, separation from the parent company and the absence of connected transactions are conditions for new rights issues by all listed companies. Several CSRC stipulations, including informing the CSRC and the public, aim at making the procedures for transferring and remov- ing listed company senior management more cumbersome. The stipu- lations explicitly state that for listed companies to perform well, the positions of directors and managers should be stable. Other measures are aimed at making transactions between listed companies and controlling shareholders more costly. For instance, a June 2000 CSRC notice restricts listed companies from using their assets to provide guarantees for their shareholders or for connected companies. Many of these requirements are even stricter than in Hong Kong (China) and other developed markets. They underscore both the severity of the problem and the authorities' determination to pro- tect minority shareholders. A March 1999 CSRC measure (Further Standardizing the Op- erations and Reform of Companies Listed Outside China Opinion) emphasized directors' use of professional consultants, advisors, and experts and the formation of specialist committees. Some CSRC opin- ions have also included provisions in relation to supervisory boards, for instance, recommending an increase in the number of independent supervisors and suggesting that supervisory boards should have the power to request information and to be kept fully informed about listed companies' operations and financial status. While these measures indicate significant progress in strengthen- ing boards of directors' independence, without other supporting mea- sures their effectiveness may be limited. Such supporting measures 102 OWNERSHIP OF LISTED COMPANIES could emphasize directors' awareness and education and clarify the exact rights and responsibilities of directors. Such clarity would need to be supplemented by a proper system of rewards and penalties, with protection under the business judgment rule being an integral part of the system of rights and responsibilities. Some of the current measures have the effect of creating different classes of directors with different rights and responsibilities. This con- travenes the principle that all directors should have the same types of fiduciary responsibility toward all shareholders. China's approach to strengthening the independence of boards of directors implicitly assumes a system of unitary boards; however, one could argue that the concept of independent directors is irrelevant in a dual board system such as China's. Under current circumstances, strengthening the supervisory board is an alternative to establishing a system of independent directors. However, applying a symmetric ap- proach to the two types of boards can produce too much of a good thing. Recent measures reinforce the relative strength of the board of directors versus the supervisory board. These measures rely exclusively on regulatory control for enforce- ment. They do not emphasize giving powers to minority shareholders directly and lowering the costs to minority shareholders of exercising their ownership rights. For example, the new independent director system gives every shareholder with more than a 5 percent stake the right to nominate independent directors. Under the current owner- ship structure, in practice this means that only the first and second largest shareholders have the right to nominate independent direc- tors, yet these shareholders already have greater control rights rela- tive to their cash flow rights. One could argue that nomination by these shareholders takes away independent directors' independence, and that they are therefore unlikely to look out for the interests of small and minority shareholders. Conclusion The ownership structure of listed companies in China is characterized by the absence of families and individuals as significant shareholders, the negligible role of financial institutions and institutional investors, and the large role of the state. In more than 95 percent of listed com- panies the state is directly or indirectly (through industrial SOEs) in control. Ownership is highly concentrated. In almost 50 percent of 103 CORPORATE GOVERNANCE IN CHINA sample firms the three largest shareholders account for 60 to 80 per- cent of all shares. The high concentration of ownership combined with the relatively small portion of tradable shares implies that control is contestable in few, if any, listed companies. These ownership features have a direct bearing on the type of cor- porate governance issues that China currently confronts. Perhaps the most important implication of the dominant role of state ownership in listed companies is the government's control over management appoint- ments and incentives, and thereby over companies' behavior. While the dominant position of corporate entities as controlling shareholders is not unique, in China large shareholders often overrule shareholders' meetings and boards of directors and exercise direct control. Recent measures to improve the functioning of boards of direc- tors have increased the costs of actions by controlling shareholders that may hurt minority investors. They also limit the capacity of gov- ernment agencies as controlling shareholders to exploit minority in- terests. However, such measures, are insufficient to control the government's distorting impact on managers' incentives and career concerns. 104 5 Role of Stock Markets and Information Dislosure in the Corporate Governance of Listed Companies An efficient stock market rewards better corporate governance with lower funding costs. The stock market processes information disclosed by listed companies, but it also creates incentives for acquiring infor- mation about firms and, by acting on this information, for dissemi- nating it to other market participants. Efficiently provided information allows company owners to link management incentives with stock prices, which helps align the interests of owners and managers (Levine 1997). Efficient stock markets also create incentives to identify poorly managed companies, facilitate the takeover of such companies, re- place their management, and make other changes in the companies' corporate governance. This chapter looks at some of the structural characteristics of Chinese stock markets that affect their capacity to promote good corporate governance, with a focus on information dis- closure. Corporate Governance and Performance Despite their spectacular development in little more than 10 years, Chinese stock markets are not yet capable of promoting good corpo- rate governance. While rigorous testing of whether market efficiency promotes good governance practices is inherently difficult, some ba- sic tests combined with a rule of thumb approach may provide an indication. For example, using Black's (1986) definition of an effi- clent market as one in which the stock price is within a factor of two 105 CORPORATE GOVERNANCE IN CHINA of the value, that is, the price is more than half but less than twice the value provides some evidence of inefficiency in Chinese stock markets (table 5.1).' judged by this measure, Chinese stock markets are clearly an outlier among emerging markets: the market to book value ratio is among the highest and does not show any clear trend of convergence toward more normal levels. Regarding corporate governance specifi- cally, the Credit Lyonnais Securities Asia corporate governance watch fails to establish a robust relationship between various indicators of corporate governance and the performance of Chinese companies on the stock market (CLSA 2001). Because ascertaining the performance of Chinese listed compa- nies with a reasonable degree of confidence is difficult given existing TABLE 5-1 MARKETro BOOK VALUE, SELECTED ECONOMIES, 1996-zooT Economy 1996 1997 1998 1999 2000 2001 Argentina 1.2 1.0 1.0 1.0 1.0 0.9 Brazil 1.6 1.4 1.3 1.4 1.4 1.4 Chile 1.6 1.5 1.5 1.5 1.6 1.5 China 3.7 3.5 3.6 3.7 3.9 3.3 Czech Republic 1.0 1.0 0.8 0.8 0.8 0.8 Egypt 1.8 1.6 1.4 1.5 1.5 1.4 Greece 3.9 3.8 3.0 2.8 2.6 2.3 Hungary 2.6 2.1 2.0 2.1 1.9 1.7 India 2.7 2.7 2.2 2.2 2.2 1.9 Indonesia 1.8 1.8 1.6 1.5 1.7 1.8 Korea 1.0 0.9 0.9 0.9 1.1 1.0 Malaysia 1.6 1.6 1.4 1.3 0.8 0.8 Mexico 1.9 1.8 1.7 1.8 2.1 2.1 Philippines 1.2 1.1 1.0 1.0 1.0 1.0 Poland 2.2 1.9 1.8 1.9 2.0 1.8 Russia 0.6 0.6 0.6 0.7 0.8 0.8 South Africa 2.2 2.2 1.9 2.1 2.3 2.2 Taiwan, China 2.2 2.0 2.1 1.9 1.8 1.7 Thailand 1.6 1.6 1.4 1.4 1.6 1.7 Turkey 3.6 3.0 2.6 4.0 3.9 3.9 SOURCE: Standard & Poor's emerging markers database. 1. Book value is perhaps the most unsophisticated hut not necessarily the most unreliable estinate of value. 106 ROLE OF STOCK MARKETS accounting and disclosure problems, our basic tests for correlation between governance and performance in our sample of listed compa- nies adopt the following approach. First, we compare the corporate governance patterns in special treatment (ST) and non-ST companies (table 5.2). By definition, ST companies have shown three consecutive years of losses, and their performance is thus unquestionably bad. We look for any corporate governance features common to ST compa- nies. Second, we look at whether the stock market takes these features into account. TABLE 5.2. OWNERSHIP STRUCTURE AND CORPORATE GOVERNANCE IN SPECIAL TREATMENT AND NON-SPECIAL TREATMENT LISTED COMPANIES, 1999 ST Non-ST Category companies companies Ownership structure (%) Nontradable shares/tradable shares 2.6 2.9 State shares/total shares 27.6 21.2 Legal person shares/total shares 36.7 38.0 Shares held by institutions (tradable)/total shares 0.9 2.4 First three largest shareholders/total shares 46.9 58.1 Shares held by largest shareholder/total shares 38.6 47.3 Performance RMB earnings per share -0.4 0.1 Net assets per share (RMB) 0.8 2.6 Return on net assets (%) -5.2 5.7 Market capitalization/total assets 0.6 1.5 Market capitalization/net assets 1.9 2.4 Board structure Executive directors/board size (%) 63.3 48.5 Directors appointed by largest shareholder/ board size (%) 25.5 51.4 Directors appointed by the three largest shareholders (%) 42.2 67.9 Size of board (number) 9.9 9.8 Chairman = general manager (%) 25.0 16.9 Meetings held per year (number) 2.6 2.5 NOTE: Market capitalization = share price x number of tradable shares + net assets per share x nontradable shares. SOURCE: Survey; authors' calculations. 107 CORPORATE GOVERNANCE IN CHINA As table 5.2 shows, the two categories of companies differ little in terms of ownership structure; however, ST companies have a higher percentage of state shares and a smaller share of tradable shares held by institutional investors than non-ST companies. Also the ownership of ST companies is less concentrated than that of non-ST companies. In terms of board structure, ST companies have more executive direc- tors than non-ST companies, which indicates higher insider control; however, the largest shareholder and the three largest shareholders have fewer seats on the board in ST companies than in non-ST coin- panies, while the chair of the board of directors is also the general manager in more ST companies than non-ST companies. One way to interpret these results is that control structures dlomi- nated by insiders and company management are associated with poorer economic performance, and that in markets with a developing legal framework and enforcement mechanisms, companies with a highly concentrated ownership structure may perform better than those with a less concentrated ownership structure, that is, companies controlled by management. A possible explanation for these relationships is the lack of legal and market mechanisms for exercising control over man- agers and insiders. A high percentage of executive directors in the context of dysfunctional supervisory boards could lead to too much insider control, and consequently to disappointing performance. A higher proportion of nonexecutive directors appointed by sharehold- ers could provide an effective restraint on insiders. An alternative explanation of these observations is that they are simply an accounting artifact brought about by the capacity of large shareholders and parent companies to manipulate the books of the companies they control or to enter into related party transactions to provide temporary assistance that disguises bad economic performance. Under this interpretation, ST companies simply lack a shareholder that is big enough or motivated enough to provide assistance in times of difficulty or have an ownership structure that makes it difficult for shareholders to agree to provide such support. While these explanations are not necessarily mutually exclusive, we can gain further insight by comparing the decisionmaking powers of the boards of directors in ST and non-ST companies. Table 5.3 shows only those areas where significant differences are apparent be- tween the two types of companies. Overall, the boards of directors of non-ST companies are more powerful with respect to key decisions, particularly in the areas of borrowing and forming strategic alliances. This gives some credence to the first hypothesis that management/ 108 ROLE OF STOCK MARKETS TABLE 5.3 DECISIONMAKING POWER OF BOARDS OF DIRECTORS IN SPECIAL TREATMENT AND NON-SPECIAL TREATMENT LISTED COMPANIES (percentage of companies) ST companies Non-ST companies Sub- Sub- stantial Some A little stantial Some A little Category power poiver power power power power Financial investments 62.5 0 0 72.2 8.5 2.8 Equity investments 62.5 0 0 72.6 10.5 2.0 Sale of assets 62.5 12.5 0 70.2 t0.5 2.4 Borrowing for fixed assets investments 37.5 12.5 0 65.3 14.5 4.8 Borrowing for working capital 25.0 12.5 12.5 52.5 16.9 12.1 Collateral borrowing 37.5 25.0 0 70.2 9.7 3.2 Guarantee for other companies' borrowing 25.0 25.0 0 69.7 10.1 3.6 Decision on amount of donation 37.5 12.5 12.5 65.7 11.3 5.2 Use of donated money 37.5 0 25.0 47.2 14.5 9.6 Tender for major projects 60.0 0 40.0 49.1 13.3 8.8 M&As of other companies 50.0 12.5 12.5 68.6 11.7 0.8 M&As by other companies 37.5 0 25.0 60.5 11.7 2.4 Planning of long-term goals of company 50.0 0 12.5 79.8 8.9 1.2 Strategic planning 50.0 0 12.5 81.1 6.9 1.2 Strategic alliance 25.0 0 25.0 58.5 12.5 8.0 Change of business direc- tion, entry into new industry or market 50.0 0 12.5 72.6 10.9 2.4 SOURCE: Survey; authors' calculations. insider control is the key factor. We found that higher state shares tend to be associated with more executive directors. To the extent this results in greater insider/management control, it provides a potential link between state ownership and poorer economic performance. Next we correlated the corporate governance features identified as potentially important with measures of economic and market per- formance over the entire sample (table 5.4). The results confirm our comparison of ST and non-ST companies: the higher the percentage of executive directors, the worse the company's performance, and the 109 CORPORATE GOVERNANCE IN CHINA TABLE 5.4 CORRELATIONS BETWEEN BOARD CHARACTERISTICS AND SELECTED MEASURES OF COMPANY PERFORMANCE Performance Net profit Net profits! Market value! Board characteristics per share net assets book value Number of board Not statistically Not statistically Not statistically members significant significant significant Executive directors/ Negative. sta- Not statistically Not statistically overall directors tistically sig- significant significant nificant Directors appointed Positive. sta- Positive, sta- Not statistically by the largest three tistically sig- tistically sig- significant shareholders/over- nificant nificant all directors Directors appointed Positive, sta- Positive, sta- Not statistically by the largest share- tistically sig- tistically sig- significant holder/overall dir- nificant nificant ectors SOURCE: Authors' calculations. higher the percentage of directors appointed by key shareholders, the better the company's performance. However, the lack of a statistically significant correlation between these corporate governance features and the market's evaluation of company performance as shown by the market to book value suggests that the market does not reflect these aspects of corporate governance. Stock Market Role in Promoting Good Governance Chinese companies possibly still view listing as a privilege and as a fund-raising mechanism for ailing SOEs. Market participants gener- ally believe that the quality of listed companies is poor2 but that inves- tors are protected, either because the government is likely to provide 2. For example, in a recent interview (Institutional Investor, May 1, 2001) an investment banker mentions that he finds that only 15 of the 114 B share comipa- nies that he had visited were of good quality. The manager of China Heartland Fund believes that of the more than 1,000 stocks available onl the A share marker, only about 10 percent are investable based on Western standards. 110 ROLE OF STOCK MARKETS direct support or the company will find a "white knight" interested in backdoor listing (the purchase of a significant portion of shares in a listed company by a nonlisted company so as to gain indirect access to the public equity market). Given the perception of poor firm quality and implicit investor insurance, market participants have little incen- tive to pay attention to fundamentals, including corporate governance. Supply Side. The government is introducing measures that are likely to change this situation. It has eliminated the quota system for IPOs, and share issuance and future decisions about which companies will access the market and when and where they will do so will be based on market principles. The CSRC's guidelines for assessing IPOs re- quire committee members to pay special attention to corporate gover- nance issues, such as whether the company's shareholders' meetings, board of directors, and board of supervisors have been discharging their duties and exercising their rights independently according to the law; whether the company's management structure is complete; whether the company is engaged in frequent related party transac- tions; whether its assets, personnel, governing organs, and financing are separated from its parent company; and whether a competitive relationship exists between the company and some of its sharehold- ers. Requirements related to corporate governance aspects also ex- plicitly enter the decisionmaking process in the case of secondary offerings. However, lifting the quota system may be insufficient to ensure a supply of high-quality companies. Attracting high-quality companies is likely to remain difficult given the immature nature of the stock markets. While companies may be able to enjoy better valuations in China than elsewhere, in the absence of a significant institutional in- vestor base, the amount of money that can be raised in the A and B share markets is limited.' In addition, the high volatility of the A mar- ket makes executing a substantial fund-raising exercise difficult. As a result, larger and better companies prefer to list overseas. Recent regulations that make it easier for private and foreign- invested enterprises to list domestically are likely to improve over time the overall quality of listed companies. Allowing companies to issue 3. According to a representative of China Eastern airlines, which issued 300 mil- lion shares in Shanghai in 2000, the money raised was not enough to buy a single aircraft. 111 CORPORATE GOVERNANCE IN CHINA Chinese depository receipts and permitting dual and multiple listings are alternative ways to introduce better quality companies to Chinese stock exchanges. The hope is that listing companies such as Unicom and Mobil would establish benchmarks for disclosure, performance, and corporate governance. However, until the idea of Chinese deposi- tory receipts becomes practical, some important issues related to the convertibility of the renminbi, the disclosure regime for issuers of Chinese depository receipts, the promulgation of special regulations, and so on need to be resolved. To ensure that listed companies are of good quality, market exit is as important as market entry. The new rules for delisting and the first delisting of PT Narcissus show that an exit mechanism has been set up. The new regulations give the power to delist to the CSRC as opposed to the stock exchanges as is the normal practice and stress profitability as the main reason for delisting. International exchanges have a broader list of factors that may include, for example, failure to observe good accounting practices and the creation and perpetuation of conflicts of interest. In an important move the new regulations mandate detailed disclosure of the company's financial status over the years and the reasons for the financial difficulties that led to the de- listing, including related party transactions. They also give sharehold- ers the right to request such disclosure without, however, specifying the procedures for exerting such rights. Despite the breakthrough, delisting is likely to remain a difficult issue, given that local governments are significant owners of listed companies and are likely to resist the delisting of companies under their jurisdiction. In addition to delisting, bankruptcy and privatization will be the next tests for whether listed companies have ceased to be "sacred cows." So far, even though thousands of companies have gone bankrupt, not a single listed company has gone bankrupt or been priva- tized. Local governments are prepared to grant tax rebates, offer fis- cal subsidies, and buy obsolete inventory from listed companies to prevent their bankruptcy. Delisting is a difficult but significant step, because ST and particular transfer methods rarely result in a success- ful restructuring. Buyers often buy stocks of failing companies expect- ing a government bailout, a safety net not found in most other markets. Consequently, ST and particular transfer stocks often outperform the market, which makes the use of stock market performance as part of managers' incentive mechanisms impractical. In addition to the quality of listed companies, the market liquid- ity of shares is important for market efficiency. In a larger and more 112 ROLE OF STOCK MARKETS FIGURE 5.1 TOTAL FREE FLoAT/GDP RATIOS, SELECTED EcONOMIES Percent 160 140 120 100 80 60 40II. United Hong United Canada Japan France Germany China China Kingdom Kong, States (1997) (2000) China SOURCE: CSRC; IFC emerging markets database. liquid market, market participants may have greater incentives to ac- quire information about firms, as it becomes easier for agents who have acquired such information to disguise their information and make money. Liquidity is a function of free float and market turnover. Chi- nese markets are characterized by a low free float and excessive turn- over (volatility). Despite the rapid growth of China's stock markets, the total free float market capitalization/GDP ratio still lags signifi- cantly behind that in some industrial countries (figure 5.1). A low free float tends to increase volatility and therefore to reduce the informa- tion content of stock prices. Healthy development of the market re- quires increasing the amount of tradable shares. Demand Side. Institutional investors have a small market presence and cannot as yet play a stabilizing role. By February 2001 China had 33 investment funds, which should increase the market's sophistica- tion. The government has allowed closed-end funds to enter the mar- ket. Regulations for open-end funds have been released, and the first open-end funds were allowed in September 2001. Insurance compa- nies still have a limited presence on the stock market-less than 9 percent of their investments are in stocks-but the growth potential is huge. They have been allowed and encouraged to invest in the stock 113 CORPORATE GOVERNANCE IN CHINA market.' Furthermore, the government has begun to explore the links between social security reform and the development of the stock market. The CSRC's April 2000 regulations on secondary offerings re- quire public offerings of new shares to reserve a certain percentage of shares for institutional investors. This stipulation seeks to encourage institutional investors to become long-term strategic investors. Fur- thermore, the government is considering plans to introduce a quali- fied foreign institutional investor scheme similar to that in Taiwan (China). Funds and other institutional investors are likely to be tightly regulated, which is appropriate given the immature nature of the mar- ket and the lessons learned from the experience of other transition economies (Johnson and Shleifer 1999). The lack of a strong institutional investor base is related to the lack of professional stock market analysts. The largest brokerage houses have hired many analysts, but their interactions with listed companies and market coverage are often superficial. Analysts and fund managers do not scrutinize companies sufficiently. This is hardly surprising, given that until recently the prices of new issues were arti- ficially fixed at low levels compared with the price-earnings ratios prevailing in the secondary market. These pricing policies made the underwriting of new shares a risk-free activity. Consequently, under- writers have had little need or incentive to analyze companies and the market. The CSRC's new regulations on share issuance are likely to change that. These regulations introduce market-oriented pricing and impose greater liabilities on listed companies and lead underwriters regarding disclosure and earnings forecasts. The six-month holding period for strategic investors will expose institutional investors to greater risks and would make thorough due diligence worth their while. For securities companies, consignment sales will also create incentives for more rigorous analysis of companies and the market. 4. As of February 2001, 6 insurance companies have been allowed to invest up to 15 percent of their assets in stocks, 2 companies have been allowed LP to 12 percent, 11 companiCs up to 10 percent, and 3 companies up to 5 percent (UBS Warburg 2001). 5. Circular on Further improving Methods of Issuing Shares (1999) and Public Share Offerings by Listed Companies Tentative Procedures (2000). 114 ROLE OF STOCK MARKETS Compers and Metrick (1998) found that institutional investors prefer stocks that are, on average, larger and more liquid, and possess value characteristics. Thus a shift from individual to institutional in- vestors will increase demand and, other things being equal, will in- crease prices for large, liquid stocks. This would tend to lower the cost of capital for such companies and encourage them to list or issue shares on the exchange. Preliminary data on the behavior of existing Chinese funds indicate similar behavior, which in the current market results in high concentration in funds' portfolios. According to the data (UBS Warburg 2001), Chinese fund investments favor the elec- tronics and telecommunications equipment sectors. On average, funds' 10 largest holdings account for 64 percent of their total portfolios. Market for Corporate Control. The existing ownership structure of Chinese listed companies makes successful hostile takeovers unlikely. However, hundreds of M&As take place each year that involve the transfer of nontradable A shares through private agreements. As noted earlier, the largest shareholder has, changed since the IPO in 28 per- cent of the surveyed companies. Table 5.5 shows the mode of change of the largest shareholder. In these private transactions, shares trade at a fraction of the price of liquid shares. Thus if a buyer were able to find a controlling share- holder willing to part with unlisted shares, the price of acquiring con- trol would likely be low compared with a market takeover. The different TABLE 5-5 MODE OF CHANGE OF LARGEST SHAREHOLDER Number of Percentage of Mode companies companies Transfer of state shares by agreement 24 46 Transfer of legal person shares by agreement 15 28 Transfer gratis 10 19 Change of administrative department of state shares 2 4 Transferred right issues 1 2 Total 52 100 SoURCE: Survey. 115 CORPORATE GOVERNANCE IN CHINA markets for different classes of shares and the resulting price differen- tial has led to the perverse outcome of nonlisted companies buying listed com,,t3nies, when valuations of A share companies would pro- vide a strong economic rationale for acquisitions to occur in the op- posite direction.' The reasons for M&As vary. In addition to the traditional moti- vation, Chinese companies may need to sell assets to meet profit tar- gets specified by the government or to generate exceptional gains to meet the 10 percent return on equity required to issue new shares. Increasingly, M&As are related to forced asset foreclosure when debt- ors become unable to honor their commitments. In many cases, large equity stakes, and even majority control, are auctioned off when stake- holders default on bank loans and other debts. While some of these changes are likely to be beneficial for the company in the long run, they do not necessarily align with managerial incentives. In our sample, in 86 percent of cases a change in the controlling shareholder resulted in replacement of the chair of the board of directors and 46 percent of the time in replacement of the general manager. Thus when the con- trolling shareholder is in financial difficulties, senior managers are likely to feel threatened, and therefore have incentives to take action to avoid a change of majority shareholder. Another question is whether existing corporate governance struc- tures encourage enough M&As, because the number that take place is insignificant compared with China's restructuring needs. For example, the level of industry concentration is low compared with other econo- mies of a similar size, such as the United States. According to official statistics, of 46 industries only 8 had a concentration level greater than 40 percent based on sales, while most had a concentration level of less than 20 percent (HSBC 2001). Small, uneconomical plants, most of which would be unable to survive without implicit or explicit government subsidies, dominate most industries, with capacity utili- zation running at about 60 percent in manufacturing. This low indus- 6. A number of A share companies have been acquiring nonlisted companies. Tsingrao Brewery acquired 2 factories in 1997, 5 in 1998, and 15 in 1999, all near bankruptcy. Shanghai Jinling, a company that produces and distributes elec- tronics and communications products, has adopted the growth strategy of acquir- ing troubled SOEs in the sector and revitalizing them with improved incentive schemes for management. Some listed mining companies, for example, Yanzhou Coal Mining, have also acquired or are planning to acquire existing mines. 116 ROLE OF STOCK MARKETS try concentration accounts for the presence of excessive competition: there is excess supply for more than 90 percent of production inputs and consumer products. This industrial structure means that China is not ready to open its markets to foreign competition. Consolidation on a national scale is needed, and M&As are a more effective mecha- nism to achieve this than, for instance, bankruptcies and foreclosures. Voluntary M&As could be easier to accomplish in sectors such as petrochemicals and aviation, where a common ultimate shareholder base exists at the national level. However, achieving significant M&A activities in sectors such as power, toll roads, and automobiles, where ownership by local governments is significant, would be difficult, and this is the case for most sectors represented on the stock exchanges. Local governments are likely to oppose M&As, especially if they would result in rationalization of capacity, layoffs in their localities, and tax revenue losses. Thus significant local government ownership is likely to be the major impediment toward establishing truly national mar- ket players in China. China cannot integrate itself into the world economy successfully without an integrated national market and na- tional economic players. Several other factors also limit M&As. Chinese M&As do not typically involve investment banks and are carried out using cash. This is partly because of the lengthy and complicated process of ob- taining approvals for issuing new financial instruments, and partly because of the immature nature of the market. Furthermore, foreign and private participation in M&As is still subject to hurdles, and the government's discretionary approach toward M&A applications is a problem. Often which agency must authorize an M&A transac- tion is unclear, and the responsibilities of different agencies often overlap and change. The strongest incentive for M&As seems to be to gain access to the stock market. As a result, they are not based on any economic rationale, do little to strengthen the fundamentals of acquired compa- nies, and tend to perpetuate incentives for abuse of majority shareholding positions. Information Disclosure as a Tool of Corporate Governance This section examines the demand for disclosure, the incentives for listed companies to supply reliable information, and the role of regu- lations and of intermediaries such as external auditors. 117 CORPORATE GOVERNANCE IN CHINA Demand for Information Disclosure. Given the stage of development of Chinese capital markets, mature users of financial information, such as institutional investors and analysts, are in short supply.! As a result, the market is not yet ready to exercise a supervisory function in rela- tion to auditing and accounting professionals and listed enterprises' disclosure practices. Financial reporting, accounting practices, and disclosure are currently oriented primarily toward satisfying the in- formation needs of taxation authorities and not of investors. Separate reporting for tax and accounting purposes does not exist. Tax laws and regulations directly affect accounting practices. For example, many enterprises do not accrue bad debt provision, as ob- taining approval from the tax authorities to do so is difficult. In the case of discrepancies, tax laws and regulations typically prevail over accounting standards, and in many instances this results in an over- statement of revenue. Such practices have important implications. Because tax regulations drive accounting methods and are designed to prevent taxpayers from minimizing their tax liabilities, some pru- dent accounting practices that defer income or increase allowable de- ductions are disallowed. Thus to some extent, the rigidity of accounting norms and regulations derives from the fact that the government, par- ticularly the tax authorities, are the main users of financial informa- tion. This explains the apparent paradox that despite a rigid accounting system, enterprises have strong incentives to use loopholes in the sys- tem to meet profit targets. Incentive to Supply Information. China's financial markets do not yet reward better and more transparent companies with lower-cost funds. Until recently, criteria for market access were relatively independent of the quality of companies, which were allowed to list without a meaningful track record. Numerous cases have revealed that the net assets of packaged companies were significantly overstated, which led to losses for minority shareholders. Subsequent to listing, the interdependence between listed and parent companies creates strong incentives to distort information, 7. The existence of sophisticated users of financial information is directly linked with the integration of Chinese capital markets with the global economy. Analysts do take note of Chinese companies listed on the B share market or on interna- tional exchanges. 118 ROLE OF STOCK MARKETS particularly information about related-party transactions, which are frequently used to adjust operating results and financial positions. The Shanghai Stock Exchange recently reviewed 2,000 annual reports by listed companies. The results revealed a number of disclosure prob- lems, including the following: * The disclosure of related party transactions did not fully ac- cord with relevant accounting standards. The pricing policy, the rea- sons why a related party was using company funds, and the agreements pertaining to assets provided to or occupied by the related party were not fully disclosed, and certain significant financing activities in con- nection with related parties were not disclosed on time. * Some listed companies did not fully disclose how they used funds raised from the public. They also did not provide enough expla- nation about any significant differences between anticipated profits and actual net income. * The quality of audits by certified public accountants (CPAs) differed, which affected both the reliability of audited accounts and the ease with which they could be compared. Xiao (1999) found that the overall level of compliance with re- porting requirements among listed companies was high, which is to be expected given the mandatory nature of these requirements. Listed companies often provide additional voluntary disclosure of such mat- ters as annual general meetings, performance of subsidiaries, contin- gent liabilities, and projected earnings. The likely reasons for such voluntary disclosures include (a) old habits from the days of a planned economy; (b) lack of awareness of potential liability, litigation, and confidentiality problems; (c) desire to impress or please investors and regulators; and (d) imitation of foreign practices. In addition to the shortcomings identified by the Shanghai Stock Exchange, Xiao found inadequate disclosure of line segment information, accounting poli- cies, impact of extraordinary items, and effects of changes in govern- ment policies. Other common problems relate to the reporting of important investments, capital commitments, consolidated statements, fair value, and contingencies. The confusion arising from the existence of multiple bases for preparing and auditing financial statements also affects the quality of information. Some companies follow IAS, other use U.S. generally accepted accounting principles, and still others follow domestic standards. Further confusion exists in companies following domestic 119 CORPORATE GOVERNANCE IN CHINA standards, because they may use China's ASBE or industry-specific rules. Often the holding company follows ASBE while subsidiaries use industry-specific accounting systems. Similar problems exist with respect to auditing standards. As a result, comparing information about different firms is difficult. A number of reasons account for the deficiencies in reporting, including internal auditors' lack of independence in performing their functions. The revised Accounting Law makes some progress toward increasing independence by clearly stating that company management is responsible for accounting information and should assure the truth and completeness of financial reporting. The law also requires that enterprises establish and strengthen internal accounting supervision. However, the procedural underpinning of these provisions is still lack- ing, and many listed companies have not set up an independent inter- nal audit function to monitor their operating activities because they are not legally required to do so. Thus if carried out at all, internal audits normally concentrate on compliance with laws and regulations and detection of any cheating, and not on a review of business performance. In addition to inadequate disclosure, selective disclosure is an important problem given the underdeveloped and speculative nature of Chinese capital markets. In the securities market, for example, se- lectively disclosed information has often resulted in significant changes in share prices or higher than usual trading volumes, to the disadvantage of small and medium shareholders. Companies listed on the A share market are not expected to distribute their financial re- ports to shareholders. Intermediaries. Improvements in the quality of accounting and audit- ing services have lagged behind the recent rapid growth of profes- sional accounting firms. Despite the dramatic development of accounting education, China still lacks a sufficient number of well- educated accounting professionals, and most internal auditors per- form their duties without sufficient training.' More experienced accountants usually train new staff using on-the-job training. Com- 8. Accounting as a major was reintroduced in universities in 1978. In 1996 more than 120,000 students studied accountancy. In addition, the International Inter- nal Audit Committee has developed a national examination program to improve the quality of internal auditors. 120 ROLE OF STOCK MARKETS pared with international accounting firms, the overall audit quality of local firms, especially smaller firms, is questionable. Some local me- dium or large enterprises can only produce an audit report after two or three days of fieldwork by two or three audit staff. Most local accounting firms have no proper procedures for performing risk as- sessments when accepting clients. They seldom perform internal con- trol assessments as part of the audit to understand the control environment and identify potential risks. Auditors do not apply a sys- tematic and standardized audit approach even within their firms. Quality control policies and procedures to ensure that all audit work fulfills the requirements of independent auditing standards are often lacking. While competition in the industry is intense-for example, in 1999 about 100 large accounting firms competed to provide services for some 1,000 listed companies-this does not seem to exert enough selection pressure to improve overall standards. To increase their mar- ket share, many firms compete on price, which leads to a vicious circle of poor audit quality. Traditionally, auditor-company relationships have been difficult in China. As is the case in many other countries, real and perceived independence is a fundamental problem that both internal and exter- nal auditors face, an issue addressed in relevant laws and regulations. Article 5 of the Chinese General Standard on Professional Ethics re- quires auditors to remain independent. Accountant firms and CPAs cannot undertake audits or other attestation functions if they have vested interests in the clients' performance and face a number of re- strictions during the course of their work. However, no limitations are placed on the provision of nonaudit services, such as bookkeeping, preparing financial statements, or consulting on taxation and man- agement issues, nor is any restriction placed on the percentage of a CPA's total income derived from one client. To enhance auditors' in- dependence, some countries prohibit auditors from providing certain services. For example, in Japan and the Netherlands an accounting firm cannot perform tax and consulting services for the same clients for which it also serves as an auditor; in Switzerland fees from a single client cannot exceed 10 percent of the firm's total income; and in Mexico auditors are economically associated with corporations if they are the source of more than 40 percent of the auditors' income in one year (Lin and Chan 2000). In China, in addition to independence in relation to clients, the issue of independence takes the dimension of independence from the 121 CORPORATE GOVERNANCE IN CHINA government. An important positive step in this regard was the 1998 delinking procedure, as a result of which external auditors have sev- ered their links with their sponsors or government authorities in the areas of personnel, finance, and business strategies. However, genuine independence will also require a change in ownership, and is not likely to be achieved without the development of a truly independent, self- regulating accounting body. Currently, the Chinese Institute of Certified Public Accountants is still in its infancy, and its basic role is likely to be monitoring and supervising rather than providing services to CPA firms. To enhance their legal binding power, all standards developed by the institute are promulgated by the Ministry of Finance. However, as Chinese inde- pendent auditing and professional standards are only considered ac- counting profession standards and the Chinese Institute of Certified Public Accountants is not a government entity, other government en- tities have not accepted these standards. In lawsuits involving account- ing firms and CPAs, Chinese independent auditing and professional standards are not regarded as legal evidence or commonly acceptable practices from the legal perspective and are therefore inadmissible, while accountants and auditors are generally not invited to partici- pate. Such practices further diminish the responsibilities, and conse- quently the independence, of accountants and auditors. Even though one of the large international accounting firms set up a representative office in China in 1981, the Chinese accounting industry has not yet fully opened up to the big international account- ing firms. The government has adopted an incremental approach to opening up the market for accounting services, encouraging coop- eration between international and domestic firms instead of compe- tition. While the "big five" are gradually becoming part of China's accounting community, their impact has been relatively insignificant. This has retarded the introduction of best international practices, but has allowed the evolutionary development of an indigenous ac- counting industry. As part of its WTO negotiations with the United States and the European Union, China made some important concessions in the ar- eas of accounting and management consulting. For example, foreign accounting firms will be permitted to affiliate with Chinese firms. These firms must be represented by CPAs licensed by Chinese authorities. CPA licenses will be issued on a national treatment basis. Accountants will be allowed to provide taxation and management consultancy ser- 122 ROLE OF STOCK MARKETS vices under the same conditions as accounting services, and will no longer be required to partner with domestic players. The Chinese de- cree that imposes burdensome requirements that might affect the con- fidentiality of market research reports will be substantially amended. Reports will no longer be examined by Chinese authorities before being given to the client, and firms will merely have to send copies of ques- tionnaires (not copies of the replies and results) to the authorities. Regulations and Standards. The ASBE consist of a general standard and a series of specific standards. Since 1993 the Ministry of Finance has promulgated 14 ASBE. Chinese standards are generally consistent with respective IAS but have adapted them to local conditions. The sequencing of their introduction and the differences with IAS also re- flect the regulator's priorities. Appropriately, the Chinese standard specifies the disclosure of more details in connection with related enter- prises, including their principal business and ownership proportion, while the IAS only require disclosure of the relationships. However, the IAS define related parties more broadly as based on "significant influence" from any principal individual investors, key management personnel, or the close family members of such individuals, whereas the Chinese standard focuses more on control. On a related issue, the ASBE require revenue to be measured at the price specified in the agreement between the transacting parties, whereas the IAS specify that revenue should be measured at the fair value of the consideration received or receivable. In this sense the IAS are more reasonable than the ASBE, because adjusting profits by in- tentionally choosing an improper transaction price can be effectively minimized under the IAS. However, the new comprehensive Account- ing Regulations for Business Enterprises, adopted by the Ministry of Finance in January 2001, require an explanation about the fairness of any related party transaction in which the transaction price is higher or lower than the normal transaction price. Chinese standards are generally more restrictive and leave less room for the exercise of professional judgment, partly because of the lack of qualified accounting professionals. This seems to be motivated by the regulator's desire to reduce flexibility during the current forma- tive period of the Chinese accounting profession. One example is the treatment of revaluation. Chinese standards adopt historical cost ac- counting as the main accounting convention. Revaluation is generally prohibited except in rare cases in which approval by the relevant 123 CORPORATE GOVERNANCE IN CHINA authorities is required. IAS, by contrast, permit revaluation, notably for fixed assets, and regularly evaluate fixed assets for any impair- ment. Revaluation would make financial information more reliable and relevant. Room for abuse could be limited by specifying the cir- cumstances under which revaluation would be appropriate. Thus while Chinese accounting standards and regulations are generally in line with IAS, an enterprise's financial statements could differ significantly from those it would produce under IAS (see box 5.1), and many individual IAS have no ASBE equivalents. However, the need for IAS is becoming more urgent with the rapid development of Chinese capital markets, the increasingly important role of M&As, and WTO membership. The disclosure required by the CSRC includes requirements for annual reports, interim reports, and quarterly reports, as well as in- formation about significant events, such as restructuring, acquisition, or sale of significant assets. Special disclosure guidance is provided to govern information disclosure for companies in specific industries, for instance, banking or real estate. BOX 5.1 CHINA MINSHENG BANK: AN EXAMPLE OF THE APPLICATION OF DIFFERENT STANDARDS The China Minsheng Bank issued public shares and listed on the Shang- hai Stock Exchange at the end of 2000. In early 2001 it disclosed its annual financial report prepared both under Chinese statutory require- ments and lAS. The following significant differences were apparent: RMB Chinese Category requirements IAS Bad debt provisions 470 1,400 Net profit 429 151 Retained earnings 365 -321 The bank explained these differences by noting that figures pro- vided under Chinese statutory requirements conformed to relevant rules and requirements, while under IAS professional judgment was used. 124 ROLE OF STOCK MARKETS Disclosure in annual reports is governed by detailed requirements that focus on short-term rather than long-term objectives and strate- gies, do not sufficiently emphasize business opportunities and risks, are lenient on segment reporting and corporate governance, and dis- courage the use of projections. The regulations do not require the distribution of annual reports to shareholders. Given the focus of disclosure requirements, the CSRC's regula- tions often introduce new concepts and accounting treatments with- out providing definitions and detailed explanations or guidelines. In issuing such disclosure requirements, the CSRC primarily takes into account the regulatory needs of the stock market and may not coor- dinate closely enough with accounting standards and regulation- setting bodies. The lack of such cooperation often results in inconsistent information disclosure practices, which reduces the comparability and the ease of interpretation of the financial statements of listed companies. Furthermore, significant accounting distinctions based on type of industry or form of business enterprise persist, for instance, SOEs or foreign-invested enterprises. Thus during the current transition period, for investors to compare performance across different indus- tries and types of enterprises is difficult. While the regulatory regime is still problematic, the government is continually working to improve the situation. In recent years, regu- lators' investigative powers and expertise have been strengthened, as has their authority to enforce penalties. Important examples are the requirement for supplying the CSRC with an internal control assess- ment report at the time of new share issues and the requirement for stricter and more detailed disclosures in relation to restructuring, re- lated party transactions, and M&As. To ensure the fairness and reli- ability of audited statements, the Ministry of Finance and the CSRC require all listed financial institutions appointing international account- ing firms to carry out audits under IAS. Moreover, the focus has shifted noticeably from uncovering wrongdoing to preventing it. While in most fraud cases the penalties are small, in some high-profile cases the regulators have imposed weighty penalties, including prison sentences for senior officials.' Perhaps the main benefit of the regulatory emphasis 9. As recently as October 2000, two minister-level officials were removed from their posts for their involvement in two separate IPOs where significant wrongdo- ing, including actions related to disclosure, were uncovered. 125 CORPORATE GOVERNANCE IN CHINA on mandated disclosure during this period of the development of a corporate governance framework in China derives from the principle that the very act of observing or recording an event changes the na- ture of the event being observed or recorded. This could in itself have a positive impact on corporate governance by raising awareness and increasing the costs of bad corporate governance. Conclusion This chapter has identified some of the structural characteristics of the Chinese stock market that prevent the market from playing its role in promoting good corporate governance practices. The market is still incapable of identifying and rewarding good corporate gover- nance practices, the supply of good quality companies is still limited, and the lack of tradability of most stocks has resulted in a volatile market populated largely by individual investors looking for short- term speculative gains. Chinese capital markets lack mature users of financial informa- tion, such as institutional investors and analysts. As a result, the market is not yet ready to exercise a supervisory function in relation to auditing and accounting professionals and listed enterprises' disclosure practices. Financial reporting and disclosure requirements and accounting prac- tices are primarily oriented toward satisfying the information needs of the taxation authorities rather than those of investors. The confusion arising from the existence of multiple bases for preparing and auditing financial statements also affects the quality of information. In addition to inadequate disclosure, selective disclosure is an important problem given the underdeveloped and speculative nature of Chinese capital markets. Audit quality suffers from such factors as narrow minimum requirements about the coverage of au- dits, unclear liability of auditors, challenges to the independence of many auditors from the state as owner of audited enterprises, and a general shortage of skilled auditors at the local level. While Chinese standards are generally consistent with respective IAS, a significant gap between IAS and ASBE is still apparent. 126 6 Building a Modern Governance System The institutional mechanisms of corporate governance discussed in the previous chapters comprise a system that can employ alternative instruments of control to effectuate changes in companies' behavior. The various instruments of corporate governance may be imperfect in isolation, but in combination they can constitute a powerful architec- ture. At the same time, fundamental weaknesses in individual instru- ments can undermine the effectiveness of the entire structure. For example, weak creditor rights will inevitably be reflected in the cost of equity capital and share prices, and will undermine the disciplining role of stock markets. The effectiveness of the various instruments of corporate gover- nance depends largely on the incentives of market players to use them. Effective regulation builds on their incentives, strengthening some and weakening others, thereby establishing an effective system of checks and balances. Without the support of markets and incentives, an over- emphasis on regulations and rules may be a triumph of form rather than of substance. Establishing Credible Penalties for Failure The effectiveness of the modern corporate governance system rests ultimately on rigorous market tests of success or failure. Without a credible threat of failure in the form of loss of market share, bank- ruptcy, delisting, or hostile takeover, most instruments of corporate governance will remain unused or their effectiveness will be limited. The strength of creditors' rights in bankruptcy, for example, un- derpins all other instruments that banks and other creditors have at their disposal to affect companies' behavior. Without a real threat of 127 CORPORATE GOVERNANCE IN CHINA bankruptcy, banks are likely to lend only to prime customers or to resort to short-term financing. As a result, their role in corporate gov- ernance will remain limited. Weak creditors' rights will in turn be re- flected in stock prices, resulting in excessive valuations. Other things being equal, this will lead to excessive demand for listing and share issuance and a distorted capital structure. Excessive valuations will create stakeholders that are likely to oppose strengthening creditors' rights and imposing discipline on the stock market. With weak credi- tors' rights and without a credible threat of delisting, stock market participants will not have the incentives to base their investment be- havior on information on companies' business fundamentals, includ- ing corporate governance practices. As a result, stock prices will not be able to serve as useful market signals in designing compensation contracts, assessing managerial performance, and guiding takeover activities. The threat of failure (and actual failure) has the unique capacity to focus the attention of investors, regulators, lawyers, and so on on taking corrective action, on discovering wrongdoing, and on ascer- taining what their respective rights are and how they can use them to protect themselves. It will therefore activate the use of the various mechanisms of corporate governance, thereby triggering an evolution- ary process of further improvement and fine-tuning. Addressing the Agency Costs of Government Ownership The threat of failure cannot become fully credible in China given the dominant role of state ownership. The government, and local govern- ments in particular, is likely to continue to rely on noneconomic con- siderations and use its powers to affect decisions on bankruptcies, delistings, and takeovers of firms under its control. As owners local governments are more likely to continue to extend support to failing firms in ways that run counter to the principles of market order, such as providing explicit or implicit subsidies, engaging in local protec- tionism, and lobbying. As owners they are also more likely to inter- fere with the application of the rule of law when they have to enforce laws, regulations, and court decisions against companies under their control. The tools available to address some of these issues can be divided into three broad groups: aligning incentives and building regu- latory capacity, separating government control rights from cash flow rights, and reducing state ownership. 128 BUILDING A MODERN SYSTEM Aligning Incentives and Building Regulatory Capacity. For manag- ers and company directors to pursue any legal means to advance the fortunes of their companies, including seeking the support of local governments, is not inconsistent with the principle of fiduciary du- ties. In the same vein, a responsible local government is expected to use its powers legitimately to advance the prospects of firms under its jurisdiction. Indeed, local governments' incentives to promote regional development, and in this context, the support they have extended to local firms, have been a key factor in China's economic dynamism, as discussed in chapter 2. Over time, the evolutionary dynamics of the development process inevitably tend to generate ten- sions and conflict between incentives at the company, regional, and national levels. China's regulatory challenge is how to align these incentives better while preserving the positive forces at the company and local government levels that are consistent with the operation of market forces. At the company level, national regulations related to the fidu- ciary duties of directors, managers, and controlling shareholders are intended to align the interests of directors and officers more closely with those of shareholders, and in the process to affect the capacity of government agencies as significant shareholders to use control for the production of political goods and to abuse the interests of indepen- dent (minority) shareholders. China has borrowed concepts from the Western corporate law tradition that had been developed to protect absentee, "inactive, and irresponsible" shareholders (Berle and Means 1999, p. 311) by, in effect, shifting the costs of monitoring from the principal (shareholders) to the agent (management and controlling party) and to the regulator and the courts. According to Berle and Means (1999, p. 242), the global trend is for corporate law "to be- come in substance a branch of the law of trusts." Having been de- signed with the interests of passive and absentee owners in mind, the principles of fiduciary duty, if successfully enforced, tend to make the identity of the owner and the ownership structure in general less im- portant in terms of their impact on economic efficiency. In addition, complete separation of listed companies' managers from the civil service system will reduce the scope of political and government control over managerial appointments and will promote the development of a managerial labor market. Better alignment of the incentives of managers and minority shareholders through stock options and compensation based on stock performance will make 129 CORPORATE GOVERNANCE IN CHINA managers more likely to resist requests from state shareholders to pro- mote noneconomic objectives. At the local level, many of the central government's regulatory initiatives and ideas have been directed at reducing local governments' incentives and capacity to engage in protectionist practices, and in the process usurping the controlling functions of markets. In April 2001 the State Council issued regulations that outlaw regional protection- ism and establish penalties for government officials engaging in pro- tectionist conduct. Price deregulation has not been extended to the subnational levels; however, steps are under way to strictly implement Article 18 of the Price Act to limit subnational governments' discre- tionary power to fix prices. Legal scholars have proposed the passage of an interstate commerce clause like that in the U.S. Constitution. These measures are expected to hasten the development of a unified national market and reduce the scope for local protectionism. Impor- tant initiatives are also under way in the area of taxation. The Tenth Five-Year Plan envisages major improvements to the tax system. The goal is a unified system and a reduction in the numerous taxes and surcharges levied by local governments. In October 1999 the Supreme People's Court introduced a five-year plan for comprehensive court reforms. The reform plan notes that China's legal institutions and mechanisms face a severe test because of, among other things, the rise in local protectionism. The five-year plan emphasizes that the lower courts belong not to the localities but to the state. The Supreme People's Court has also started to study reform of the funding system so that the lower courts can be funded through the central government. These regulations and initiatives are intended to build a unitary, fair, and orderly market system, and if successfully implemented will increase the costs for local governments in using ownership control for the production of political goods. They are also expected to strengthen the state's regulatory and implementation capacity. However, while well intentioned, some of the reform proposals may run against the natural incentives and interests of market players. For example, the State Plan- ning Commission has suggested revamping the fiscal system so that local officials do not collect tax payments directly from local branches of state enterprises as a way to curtail the incentives for local protec- tionism. This proposal, in addition to implicitly assuming that taxation rights create stronger incentives for local protectionism than owner- ship, raises the important issue of how such a revamping of the fiscal 130 BUILDING A MODERN SYSTEM system would affect local governments' incentives to use ownership to derive noneconomic benefits. Another idea has been to stop linking the promotion of local cadres to the economic growth of their localities (World Bank 2001b). Regulations often rely for enforcement on signals from agents that may be conflicted about the very issue they are asked to help regulate and enforce. Recent regulations on independent di- rectors, for example, in effect create two different classes of directors and the opportunity for antagonism in the board room. In many areas regulations amount to no more than prohibitions, and this is insuffi- cient to make them effective given the powerful forces working in favor of maintaining the status quo. The complexity of the issues and the magnitude of the task pose significant challenges to the government's regulatory capacity, espe- cially at the national level. The government will have difficulties meet- ing these challenges if it relies only on its own resources and on direct methods of regulation. Box 6.1 illustrates the successful expe- rience of the U.S. Securities and Exchange Commission (SEC), which is based on self-regulation and on extending support to the "right" party that has an interest in regulations being enforced. China has made some progress in self-regulation and in mobilizing civil society to participate in the enforcement process, but the unrealized poten- tial in this area is enormous. In the Chinese context, empowering the "right" party in relation to specific regulatory issues will often mean enhancing the independence of associations, the media, self-regulatory bodies, and other members of civil society. In the future, China's trading partners are likely to play an important role in the enforce- ment of some of these regulations by seeking to bring WTO disci- pline to bear on local internal barriers to trade and other forms of local protectionism. On a more fundamental level, building a strong regulatory capacity is likely to enhance the government's confidence that it can manage the economy effectively without the tools of di- rect ownership. Separating Government Control Rights from Government Cash Flow Rights. China could move more aggressively in experimenting with various mechanisms for separating control from cash flow rights as a way to reduce political control over companies' behavior. Regula- tions along the lines discussed earlier are likely to reduce the benefits of control associated with government ownership. As a result, the 131 CORPORATE GOVERNANCE IN CHINA BOX 6.1 THE EARLY, INDIRECT APPROACH OF THE U.S. SECURITIES AND EXCHANGE COMMISSION An important constraint that shaped the SEC's strategy in its initial years was the resource constraint: if the SEC had decided to do its job entirely on its own, its resources would have had to be enormous. Just the polic- ing of corporate financial reporting would require thousands of audi- tors and accountants, each under some form of supervision. Thus the strategy of direct control had obvious drawbacks. Instead, the SEC decided to work through existing private struc- tures and, where necessary, to create new ones. Its philosophy was to manipulate private incentives to serve public ends, while preserving its independence and not being manipulated to serve private interests. The SEC's strategy would be presented to accountants, bankers, and bro- kers as an attractive plan for "self-regulation." The heart of the regula- tory system would be a careful shaping and bending of the incentive structures so that each of the major players would voluntarily carry out SEC policies. This was based on an appreciation of the need to manipu- late incentives implicit in the industry, so as to give those involved a self- interest in obeying and strengthening the law. As a result of its efforts the SEC was able to obtain the cooperation not only of the exchanges but also of brokerage houses, investment bankers, and corporation ex- ecutives, who in turn recognized that their efforts to improve financial practices were now buttressed by the strong arm of the government. The SEC's first major success in co-opting interest groups was with accountants. Accountants had an obvious interest in cooperating, be- cause the SEC's regulations and laws were creating a huge demand for accounting services. Regulation built along these lines welded together existing self-regulation and direct control by the government. The phi- losophy of the SEC's interventions was to restore checks and balances by the government's adding its weight to the right party rather than substituting for private parties. SOURCE: McCraw (1984). government will increasingly begin to associate ownership with cash flows and monetary values. This will increase the incentives to trade shares and will make various mechanisms for separating control from cash flow rights more attractive. 132 BUILDING A MODERN SYSTEM For example, the government could conduct experiments to de- termine whether having private (including foreign) institutional inves- tors manage listed state shares would promote a more market-based and value-maximizing approach. One way to do this would be to trans- fer state shares to trusts administered by trust investment companies. In this way state shares would be allowed to circulate without putting them directly into private hands. This may help demonstrate an im- portant point: a portfolio approach may achieve value preservation and enhancement more successfully than holding controlling, but il- liquid, packets of shares in specific companies. The recently adopted Trust Law and Trust Investment Companies Measures creates a legal framework for this. Another method for separating control from cash flow rights could be to modify the nature of government equity claims by, for example, transforming government equity claims into preferred nonvoting shares. This would transform the nature of the government's cash flow rights so that they would become more like some forms of tax liabilities, thereby promoting greater consistency between the different roles the govern- ment is playing with respect to government-owned firms. A wide vari- ety of preferred nonvoting shares is available, some of which allow for conversion back into voting shares under certain circumstances. Both these measures-making the government a beneficiary owner and transforming the nature of the government's equity claim-could be useful transitional mechanisms, as they could send a powerful signal that the government is committed not to interfere with market forces. Reducing State Ownership in a Gradual Fashion. The government should not view the aforementioned measures and approaches as a substitute for further ownership reform. The government realizes that despite the significant progress it has made to date in develop- ing the various instruments of corporate governance, fundamental and sustainable improvement cannot be achieved without reducing state ownership. A number of different methods are available for gradually reduc- ing state ownership. China has been considering different mechanisms for liquidating state assets in the capital market, both to help support the social security system and to fulfill conditions for SOE reform. In September 2000 the government unveiled a plan to reduce the pro- portion of state-owned shares from 68 to 30 percent in two stages. The plan specified five ways to reduce the amount of state-owned 133 CORPORATE GOVERNANCE IN CHINA shares: state share placement, share repurchase, negotiated transfer, auctioning, and debt-equity transfers. Other proposals for reducing state shares include transferring state shares to employees and to in- stitutional investors. Some experiments have already taken place. All existing plans agree that the process of divesting state shares has to be gradual. The process is likely to be influenced by changes in ideologi- cal thinking, concerns about market stability, pressure to restructure, and social safety net needs. Perhaps the most appealing way of reducing state shares is through institutional investors. Institutional investors, including foreign- invested institutions or those under foreign management, can provide the liquidity and sophistication to absorb a large supply of govern- ment shares without disrupting market stability. Disposing of state- owned shares in the form of an index fund, for example, has some attractive features. An index-linked fund, similar to the Tracker Fund of Hong Kong (China), could be established to sell state-owned shares. Along with the development of a domestic pension fund, this would reinforce the development of the Shanghai main board market and would also facilitate the creation of other index-driven products. An attractive feature of an index fund is that it is likely to have a rela- tively small destabilizing effect on the market. Ideally, the fund would be expected to issue units to tap cash from retail and institutional investors, including insurance firms and pension funds. The ownership structure is likely to undergo fundamental changes as a result of new share issues and new listings, which are likely to be the dominant approach in the near to medium term. For example, a provision already exists for selling up to 10 percent of the state share in an IPO to replenish the social security funds. The July 1999 CRSC Circular on Further Improving Methods of Issuing Shares, which ap- plies to companies with registered capital of more than RMB 400 million (US$48.3 million), is likely to have a significant impact on the liquidity of legal person shares. Under the circular the prices of shares in private placements are equal to those in public offerings, and pri- vate placements are also subject to similar disclosure requirements as public placements. Legal person shares issued under the circular can be publicly traded subject only to a six-month or three-month holding restriction. As more state and legal person shares become tradable, market forces and corporate governance institutions will begin to shape the ownership structure of listed companies. 134 BUILDING A MODERN SYSTEM Strengthening Boards of Directors Chinese regulators have emphasized strengthening boards of direc- tors as part of their efforts to improve corporate governance practices in listed companies. However, many of the aspects of good board of director practices are part of the internal operations of a company and may fall outside the domain of laws and regulations. Therefore promising approaches might combine regulatory initiatives with peer pressure mechanisms to disseminate good practices. To this end China has adopted the Code of Good Corporate Governance Practice for both listed and nonlisted companies. Assessing Whether Independent Directors Add Value. International evidence suggests that corporate boards need directors who are not just independent of management, but who are accountable to share- holders. A rapidly growing literature focuses on the relationship be- tween board composition, governance, and performance; however, empirical evidence broad enough to cover all the major issues is avail- able only for the United States. Thus the literature reflects relation- ships in a highly developed and sophisticated market economy populated by large firms that already have a significant number of independent directors on their boards. While we believe that many of the results have some general validity, they do not emphasize absolute relationships, but rather whether existing average board structures in U.S. companies are efficient, and whether small changes from the sta- tus quo would help or hinder corporate governance and performance. A number of studies look at the relationship between board inde- pendence and observable board actions, such as firing a poorly per- forming CEO, setting the level of CEO compensation, and committing financial fraud. Weisbach (1988) reported that boards with more than 60 percent independent directors are more likely than boards with fewer independent directors to fire a poorly performing CEO. He also found that boards independent of a majority stockholder are faster in firing the CEO if observable performance measures such as stock price and earnings are poor. Other studies (Bhagat and Black 1999; Bhagat, Carey, and Elson 1999) have found that firing a CEO under such cir- cumstances increases the firm's value. Critics have questioned these find- ings on the grounds that they may be limited to a particular period in U.S. corporate history when takeover activity was exceptionally high. 135 CORPORATE GOVERNANCE IN CHINA If that were the case, the results suggest that boards independent of majority stockholders are faster in taking disciplinary action when the threat of a takeover is real. Taken together, the various findings seem to confirm that independent directors behave differently than inside directors in the decision of whether to replace a poorly per- forming CEO, but that the differences are marginal. Regarding CEO compensation, little evidence suggests that inde- pendent directors do a better job of setting CEO pay than inside direc- tors. Several studies (see, for example, Borokhovich, Parrino, and Trapani 1996) report that the higher the proportion of independent directors, the more the CEO is paid. Independent directors in U.S. companies are not doing a good job of developing incentive compen- sation to induce better performance. This could be because many in- dependent directors are current or former CEOs who are prone to compensate the CEO in the manner they would like to be compen- sated themselves. With respect to the likelihood of committing financial fraud, Dechow, Sloan, and Sweeney (1996) reported that firms with a major- ity of inside directors and without an audit committee are more likely to commit financial fraud than a control group matched by industry and size. Overall, findings seem to suggest that independent directors help to control financial fraud, but the reverse causality is also pos- sible, that is, that managers prone to financial fraud resist oversight by independent boards. Finally, a number of studies explore various aspects of the rela- tionship between board composition and performance. Rosenstein and Wyatt (1990) found that, on average, stock prices increase about 0.2 percent when companies appoint additional outside directors. Kline (1999, as quoted in Bhagat and Black 1999) examined whether the existence and staffing of board committees affected performance and found little evidence that monitoring committees dominated by inde- pendent directors affect performance. Companies with inside director representation on a board investment committee tend to exhibit supe- rior performance. This suggests that companies with a large majority of independent directors may perform worse because they have too few inside directors to perform this role. Looking at the direct link between performance and board com- position, a number of recent studies, the most influential of which is Bhagat and Black (1999), find a negative correlation between the pro- portion of independent directors and firm performance-the exact 136 BUILDING A MODERN SYSTEM opposite of conventional wisdom. This may indicate that U.S. boards have too many outside directors. One implication of these findings is that favoring independent directors regardless of their backgrounds may not be the best approach. Firms need boards capable of balancing independent directors' lack of in-depth knowledge about the business with the combination of knowledge and conflict of interest that is characteristic of insiders. Above all, corporate boards need directors who are not merely inde- pendent of management but are accountable to shareholders. Making Directors More Accountable to Shareholders. Directors' rights and responsibilities need to be clarified and their enforceability made credible. Existing laws and regulations do not set out directors' exact rights and responsibilities clearly enough. The Company Law does not stipulate any disclosure obligation on the part of directors or any specific liabilities assumed by directors who fail to perform their obli- gations. Regulations such as the new guidelines on independent direc- tors provide more details, but still fail to clearly describe directors' main duties and responsibilities. China could follow the examples of countries like Australia and New Zealand, which have translated du- ties related to company loyalty into statutes. Increased legal liability often follows on the heels of increased responsibility. Thus directors also need a reasonable degree of assur- ance that if they follow established standards of behavior, they will be relatively protected from litigation. Protection under the business judg- ment rule should be an integral part of the system of rights and re- sponsibilities, translated into statutes as some countries are currently doing. One change in Australia has been the introduction of a statu- tory business judgment rule, which provides a "safe harbor for honest business decisions that turn out badly providing they satisfy certain criteria" (Hockey 2001). The statutory formulation should provide a clear presumption in favor of directors' judgment. Protection under the business judgment rule should be linked to directors' qualifica- tions; functional responsibilities on the board; and the procedures that govern board functioning, such as the use of experts. With sufficient clarity concerning directors' rights and duties complemented by corresponding shareholders' rights, legal action could become a way to monitor and enforce regulations. Lack of compliance with regulations by boards of directors and directors' failure to live up to their obligations could result in legal action against a company and 137 CORPORATE GOVERNANCE IN CHINA its board. In China, enforcement emphasizes administrative and crimi- nal processes rather than derivative civil actions. According to Article 166 of the Criminal Law, any employees who take advantage of their positions, turn management of their units over to their relatives or friends, purchase commodities from units managed by their relatives or friends at prices higher than the market price, sell commodities to such units at prices lower than the market price, or purchase com- modities that are not up to standard from units managed by their relatives or friends face up to seven years in prison. However, these regulations do not apply to the directors and managers of non-SOEs, thereby creating an uneven playing field. A major reason for China's emphasis on criminal sanctions is because under current conditions, few company officers or directors could afford to pay a substantial civil judgment for breach of fiduciary duty. Thus under current conditions, directors' liabilities are not cred- ible, and legal action as a controlling mechanism will not emerge with- out some sort of liability insurance for directors and officers (D&O). The D&O insurance market has expanded and evolved rapidly in the last few years at the same time that directors' obligations have in- creased (see box 6.2). With globalization, the directors of some Chi- nese companies will be exposed to litigation under U.S. securities laws, as they increasingly seek to raise capital in the United States. One objection to D&O insurance is that it introduces moral haz- ard and does little to improve corporate governance. However, stan- dard techniques developed in the insurance industry to tackle the issue of moral hazard are applicable in the case of D&O insurance, for example, deductibles, coinsurance, and policy conditions to restrict the coverage of high-risk individuals. In effect, the incentive for tak- ing care is provided by the insurer rather than by the threat of having to pay compensation, a risk that is now largely transferred to the in- surer. Insurers, while diversified and with few incentives for close monitoring, can be powerful forces in introducing standards and rat- ings. The price of D&O insurance can be a barometer of corporate governance. In addition, the presence of D&O insurance may influ- ence the likelihood of legal action taken, for example, by sharehold- ers. The presence of D&O insurance will certainly make directors more likely targets of shareholder activism. Promoting Board Independence. Board independence could be pro- moted through a more flexible approach that focuses on procedures, 138 BUILDING A MODERN SYSTEM BOX 6.z DIRECTORS' AND OFFICERS' LIABILITY INSURANCE D&O liability insurance is a form of malpractice insurance for corpo- rate boards of directors. It provides coverage for acts, errors, and omis- sions by individual company D&O to the extent those acts are committed during the course of their employment. Such insurance does not pro- vide coverage for the company itself. D&O insurance is a key compo- nent of the protection that every director and officer should have. Some developed markets have statutory schemes permitting corporate indem- nification. The overriding goal of these statutes is to encourage capable individuals to serve as D&O secure in the knowledge that they would be insulated from personal liability if corporate actions, taken in good faith, were attacked by way of legal proceedings. For many years D&O policies have been structured to insure not only the directors but also the companies when the latter have indemnified directors. Such indem- nification may arise by virtue of a contractual right of the directors, by order of a court, or by means of a voluntary payment that the company has the power to make and the funds to support. North America remains the biggest market for D&O insurance, accounting for about 83 percent of premium income compared with around 13 percent for Europe. Research (Monteleone and Conca 1996) reveals that the tendency to buy D&O coverage increases with firm size, as do the size and frequency of D&O claims. In the United States around 90 percent of major industrial corporations carry D&O insurance, and coverage is almost universal (97 percent) among utilities, large banks, and insurance companies, which seem especially risk averse. In the United Kingdom about 55 percent of companies with annual turnover in ex- cess of 100 million buy D&O insurance, but only about 1 in 10 com- panies with an annual turnover of less than £5 million do so. In general, firms more likely to encounter financial distress are more likely to pur- chase D&O coverage. rights, and responsibilities. The CSRC's current approach is to pre- scribe a somewhat rigid board structure. Rigidity inevitably implies some degree of arbitrariness in the prescribed number of independent directors. Companies differ tremendously in terms of their size, their affiliations with other companies, the contestability of their market positions, and so on. These are all factors that affect the potential for 139 CORPORATE GOVERNANCE IN CHINA conflict of interest and the need for board independence. Thus an alternative approach would be to regulate board composition and in- dependence indirectly through regulations on, for instance, related party transactions and boards' and directors' duties and responsibili- ties. For example, mandating the approval of large related party trans- actions by noninterested directors (suitably defined) implicitly restricts board composition while leaving companies the flexibility to adjust their board structures to their unique circumstances. Similarly, impos- ing requirements on the functions and responsibilities of auditing, nomination, and compensation committees has direct implications on the composition of boards of directors. Additional measures, such as disclosure by directors and specific board procedures, can also improve a board's independence. For in- stance, some directors may be obliged to the company or to its current CEO in ways too subtle to be captured by customary definitions of independence. This possibility is consistent with the evidence that di- rectors who are appointed during the current CEO's tenure are more generous in determining the CEO's compensation. One way to im- prove transparency would be for the CSRC to require the disclosure of financial and personal ties between directors (or the organizations they work for) and the company and its CEO. Procedures may be more important than exactly specified board composition. Measures for improving board performance focus on procedures that facilitate monitoring, such as holding an annual meet- ing of outside directors without the presence of inside directors, peri- odically reviewing the performance of the CEO and the directors, or appointing a lead director. Examples of good board practices adopted in some of the economies in the region include reducing the number of principal directorships (up to six in Singapore); setting new perfor- mance benchmarks based on shareholder value added (economic value added), especially for government-sponsored companies; requiring that boards meet no less than once a quarter (Republic of Korea); evaluat- ing the performance of management; making board and outside di- rectors' reports public; and requiring directors to be accredited (Hong Kong Institute of Company Directors). Providing Independent Directors with Incentives. Independent direc- tors need the direct economic incentive of substantial stock owner- ship to actively monitor management. However powerful a disciplining device the threat of regulatory and legal action may be, it cannot sub- stitute for the power of positive incentives. Independent directors have 140 BUILDING A MODERN SYSTEM little incentive to actively monitor management without the direct eco- nornic incentive of substantial stock ownership, and should therefore be required to purchase significant amounts of stock before serving on the board. High fixed compensation that does not depend on stock performance can act as a disincentive for management monitoring. The stock ownership will serve as a performance bond. For example, Bhagat and Black (1999) found that the greater the amount of stock individual outside directors owned, the better the company performed and the higher the likelihood of management turnover based on disci- plinary grounds. Professionalizing Corporate Directors. Continued training and educa- tion constitute an essential component of any measures supporting im- provements in the functioning of boards of directors. Internationally, a growing trend toward professionalizing corporate directors has become apparent. The U.K.'s Institute of Company Directors, for example, is introducing a chartered director qualification that it claims will be the world's first professional standard for directors (see box 6.3). This may eventually bring directors in line with practitioners in other professions, such as the law and accounting. Thus in the future directors will in- creasingly be expected to have completed some form of training, passed certain examinations, and achieved a required level of competence. In this regard, China needs an institute of company directors in- dependent from the regulator. This institute would provide training for directors; maintain a database of individuals who are potential Box 6.3 CHARTERED DIRECTOR QUALIFICATION IN THE UNITED KINGDOM The candidates are required to be at least 28 and have at least three years of board experience. They are expected to take a three-hour ex- amination in addition to other tests. On successful completion of the assessments, candidates will receive chartered director status, which will give them the right to put the letters "C. Dir." after their names. Char- tered directors are expected to subscribe to a code of professional con- duct and agree to 30 hours of training a year. The introduction of the chartered director qualification is expected to put board membership on a level playing field and detach it somewhat from the use of criteria based on titles, connections, and so on. 141 CORPORATE GOVERNANCE IN CHINA candidates for directorships and help companies fill directorship po- sitions; accredit directors; and disseminate information to establish useful benchmarks for remuneration, functions of boards of direc- tors, and so on. Moving toward a Single-Tier Board Structure. The original intent of Chinese legislation was to limit the power of directors through the presence of supervisory boards: the main function of the supervisory board seemed to be to ensure directors' accountability. Shortly after the promulgation of the Company Law, some scholars expressed the view that supervisory boards are providing ex ante director account- ability more effectively than any equivalent regime under company law elsewhere. They claim that supervisory boards are even more im- portant in the Chinese context, because the enforcement of court judg- ments is comparatively weak in China (Ong and Baxter 1999). However, in recent years, countries have tended to move to- ward a legal regime that strongly favors a single-tier board that is relatively small, and that contains some insiders as well as a major- ity of outside directors. Mandatory two-tier board structures seem to be going out of favor with companies and regulators based on the argument that they promote weaker and less responsive boards. According to Cha (2001): It is sometimes argued that more authority should then be given to the supervisory board, which sits on top of the boards of our listed companies. However, experience has shown that this system of supervision is not effective as it is often un- clear whose interest is being represented by the supervisory board. In many cases the supervisory board duplicates the authority of the board itself but without corresponding re- sponsibilities. In fact the presence of a supervisory board may give the illusion of certain checks and balances in the listed company when none existed. China seems to be moving in this direction. Recent measures strengthen the power and independence of boards of directors, which has the effect of further emasculating supervisory boards. Empowering Shareholders While regulatory activism and more independent boards can play an important role in protecting minority investors, they cannot substi- 142 BUILDING A MODERN SYSTEM tute for directly empowering small and independent shareholders to protect and further their own interests. Stipulating and Enforcing the Fiduciary Duties of Controlling Share- holders. Recognizing that existing ownership structures and IPO prac- tices have made related party transactions a critical issue for corporate governance, the regulator has made significant progress in improv- ing disclosure; introducing various internal procedural mechanisms, such as disclosure of interests and disinterested voting; and strictly prohibiting certain activities, for example, guarantees for other com- panies. However, further progress is needed to align definitions of related party transactions with international practice. More impor- tant, as the fiduciary duties of controlling shareholders are not stipu- lated, their liabilities in relation to losses incurred by minority shareholders are not clear. Over the last 25 years corporate law has come to recognize that shareholders in a corporation owe fiduciary duties to each other akin to those partners in a partnership owe each other. One of the most important developments of the concept of fiduciary duties has been the recognition of the duty of fair dealing by majority shareholders in relation to minority shareholders. Recent regulations implicitly intro- duce this principle without, however, spelling out liabilities, penalties, and procedures. In a fiduciary duty context those in control carry the burden of proof in establishing that their actions were taken in good faith and were fair to the minority shareholders' interests. Given the lack of sophistication in fiduciary duties matters, the appropriate ap- proach for China seems to be the paternalistic one, where statutes should set out the full range of fiduciary obligations. The statutes could ensure flexibility by specifying the types of duties that shareholders could opt out of, or these could be specified in individual operating agreements. The doctrine of piercing the corporate veil is not well established in China, but if developed and applied could be effective in discourag- ing certain forms of abuse by controlling shareholders.' Although this 1. The Company Law is piercing the corporate veil with respect to the ownership rights of the state as shareholder when declaring in Article 4 that "the ownership of state-owned assets in a company shall reside with the state" (see Howson 1997). This provision creates legal uncertainty as to the real rights of a corporation and the validity of the corporate form in the presence of significant state ownership. 143 CORPORATE GOVERNANCE IN CHINA doctrine is primarily directed at providing a remedy for creditors, it has the effect of protecting minority shareholders because of its deter- rence effect on large shareholders. China has seen some cases where the courts have demonstrated that under certain circumstances, typi- cally related to inadequate capitalization, they are willing to pierce the corporate veil (Feng 2001). In developed marked economies the factors determining courts' decisions about whether to pierce the cor- porate veil are quite broad and may include (a) a failure to observe corporate formalities, (b) the siphoning off of corporate funds by the dominant shareholder, (c) the nonfunctioning of officers or directors, (d) the use of the corporation as a facade for the operations of the dominant shareholder, (e) the mixing of the parent company's assets with the assets of a newly established firm, and (f) the exercise of direct control by the parent company by ignoring the subsidiary's cor- porate structure. Such practices are quite common in China today and tend to hurt creditors and minority shareholders. A broader ap- plication of the doctrine of piercing the corporate veil may be an ef- fective approach to discouraging them. Conglomerates pose special issues with respect to related party transactions. The duties and restrictions placed on controlling share- holders in conglomerates should be clearly spelled out and the appro- priate checks and balances should be implemented. The disclosure of interested director contracts is not enough. German corporate law has developed mechanisms designed to protect the minority share- holders of subsidiaries in corporate groups or conglomerates. Strengthening the Rights of Minority Investors. Shareholders' meet- ings are a corporation's most powerful authority; however, Chinese laws and regulations do not specify shareholders' rights clearly. The power to elect and remove directors is one of the indirect methods whereby shareholders can influence a company's behavior, but Chi- nese minority shareholders have limited capacity to ensure that they are represented on the board. Reportedly, a small number of listed companies have tried the cumulative system-a complicated proce- dural mechanism that is designed to facilitate minority shareholder representation on boards, but one that is open to abuse if not prop- erly regulated-however, the lack of enabling regulations limits its applicability to listed companies. The new system of independent directors stipulates that they should pay special attention to the rights of medium and small share- 144 BUILDING A MODERN SYSTEM holders. However, the stipulation of a 10 percent minimum share- holding for nomination means that minority shareholders are unlikely to be in a position to nominate independent directors, much less ap- point them. Without this, expecting independent directors to have a special duty of loyalty to minority shareholders, as stipulated in the regulations, is unlikely. Under the current ownership structure, the right to nominate independent directors is worth little if not supple- mented by methods that enhance the chances of minority sharehold- ers to elect directors, for example, cumulative voting. However, a better approach might be that proposed in Brazil, where minority share- holders holding a certain percentage, for instance, 15 percent, would have the right to appoint a board member. Company law and related regulations make removing directors difficult, primarily to strengthen their independence in relation to con- trolling shareholders. The Company Law provides that before direc- tors' terms of appointment expire, shareholders' meetings shall not be permitted to remove directors from their post without due cause. The law does not specify what reasons could lead to their removal, al- though the CSRC's guidelines in the Articles of Association include the provision that directors who fail to attend two successive board meetings, whether in person or by delegating their duties to other di- rectors, are deemed incapable of performing their duties, and at the shareholders' meeting the board of directors can propose removing such directors. The law and regulations do not specify removal by court order. Removal of directors by court order for a specific reason can be an important device for protecting minority investors in cases where a director is also an important shareholder. The provision that shareholders' meetings cannot remove di- rectors without a cause before their term expires acts as an antitakeover device, in that it limits the capacity of a new control- ling owner immediately to effect changes in the board of directors. However, the provision may cause less resistance to changes in con- trol on the part of management. Modern statutes in other countries have dramatically expanded shareholders' powers to remove direc- tors without cause. Provisions governing the removal of directors should be drafted so as to prevent the majority from undermining the outcome of cumu- lative voting or other devices designed to protect minority sharehold- ers. For instance, if a corporation uses cumulative voting, then the statutes will normally provide that directors cannot be removed if the 145 CORPORATE GOVERNANCE IN CHINA number of votes cast against their removal would have been enough to elect them. A quorum requirement could be a powerful protection device for minority shareholders and should be strengthened in the case of share- holders' meetings. It could be a simple majority or some other mini- mum percentage of outstanding shares. Specific, usually stricter, quorum requirements for specific issues, such as amending the articles of association, should also be based on outstanding shares. The cur- rent system, which requires a simple or qualified majority of share- holders in attendance, significantly strengthens the power of controlling shareholders. The development of a proxy system, essential for the exercise of shareholders' rights, will have to accompany the adoption of a quo- rum requirement. A proxy system provides shareholders with the means of submitting proposals to fellow shareholders and regulates the na- ture and content of information flows between management and share- holders. For example, annual reports are not currently sent to shareholders, but will have to be if a quorum is required and all share- holders are given equal chances to exercise their voting rights. Thus the focus of a proxy system is to provide investors with adequate in- formation before they exercise their rights. Proxy contests are an im- portant complement, and in some sense a substitute, for the takeover process. They permit partial changes in corporate governance prac- tices without necessarily undergoing changes in control. They also allow a competitive process in deciding on specific proposals and is- sues. In theory, a proxy contest is cheaper than a hostile takeover bid and makes obtaining control without owning a majority of the shares possible by simply winning a majority of the board seats. A critical factor for the functioning of a proxy system is the cost of interactions of listed companies with their shareholders. Given China's predominantly retail and fluid investor base, this could present a serious obstacle to a functioning proxy system. At least initially, regulations could provide for extra time for vote solicitation in the case of large, dispersed firms and highly liquid stock. However, the advent of the Internet has provided more options for maintaining ac- curate shareholder lists and reducing the costs of interaction with share- holders. Institutional investors can greatly facilitate the implementation and enforcement of proxy regulations. The power to request a meeting is a key shareholder right. Cur- rently, only shareholders holding 10 percent or more stock have this 146 BUILDING A MODERN SYSTEM right. This could be expanded to shareholders that together represent 10 percent or more stock, thereby giving trustees, institutional inves- tors, and individual investors the opportunity to solicit representation through the proxy system. With respect to shareholders' rights to pro- pose agenda items for shareholders' meetings, the CSRC's guidelines in the Articles of Association specify that a shareholder or sharehold- ers holding an aggregate 5 percent or more of the voting rights of the company may propose agenda items. Many jurisdictions impose restrictions on shareholders' pre- empting or challenging management, for example, in putting forward shareholder proposals, calling meetings, requesting information, or taking certain kinds of legal action. These restrictions are usually in the form of a percentage shareholding threshold. Given the difficul- ties facing any shareholder initiatives in China and the impracticali- ties of litigation, such restrictions are probably unwarranted. In the interests of curtailing excessive management power, some observers have suggested that shareholder approval be required for a wider range of activity. In situations with a controlling shareholder and an align- ment of interests between this shareholder and management, requir- ing the approval of a majority of the minority shareholders may be advisable. This runs counter to the doctrine of majority rule in corpo- rate law matters, but has been invoked in some countries, and may be justified in China under certain circumstances. The right to inspect the company's books and records is a funda- mental shareholder right. For shareholders to be able to decide intel- ligently whether to continue to be shareholders of the company; to bring legal action against the company, a controlling shareholder, or a director; or to take other action in relation to their investment they need information. Article 110 of the Company Law gives sharehold- ers the right to consult the company's articles of association, minutes of shareholders' meetings, and financial and accounting reports, and also the right to make inquiries about the company's business opera- tions. However, these rights are not codified and are at best rudimen- tary. The types of records to which shareholders have an automatic right of access is not broad enough and does not include all written communications to every shareholder, the names and addresses of current officers and directors, and annual reports or financial state- ments. No provisions allow for qualified rights, based on a definition of a proper purpose, for shareholders to inspect more sensitive infor- mation, such as the list of shareholders. The law also does not specify 147 CORPORATE GOVERNANCE IN CHINA if the beneficiaries of shareholders can inspect company materials. Shareholders' rights in relation to information access are significantly less than the information listed companies are required to give the government. The information parity of minority shareholders with respect to the government, major shareholders, and analysts is impor- tant for the existence of a proper system of checks and balances. Existing laws and regulations do not specify penalties for corpo- rations and officers that obstruct shareholders' rights to access infor- mation. To encourage more stable shareholdings, the rights of inspection could be strengthened for shareholders who have held their shares for a certain time or for larger shareholders. One area of weakness in the present legal structure is the lack of civil remedies for investors (Neoh 2000). The only remedy the Com- pany Law provides to minority shareholders is that they may apply to the courts to prevent the continuation of unlawful conduct by direc- tors and majority shareholders. At the same time, the Securities Law is unclear as to whether investors can take civil action against direc- tors and investment professionals for false or negligent disclosures that resulted in losses. The Civil Procedure Law allows collective ac- tion, but so far none has been taken in the securities field because of the uncertainties about remedies. The authorities need to clearly es- tablish these rights and introduce the necessary procedural laws. Chinese shareholders have the right to sue for an injunction and damages if a board decision violates laws and administrative rules. In such cases the law does not stipulate a threshold in terms of minimum shareholding. How the court determines who the defendant is in cases of this sort-the shareholders' general meeting, the board of direc- tors, an individual, or the company-is not clear. Often individual investors do not wish to incur the expense of a complex lawsuit to recover small, private losses. The need for broad civil discovery and class action suits is especially critical in this re- spect. The judicial system must also permit proof of wrongdoing to be based on circumstantial evidence. Rules that shift the burden to insid- ers to prove or disprove fairness once suspicious circumstances have been established can be highly valuable. Class actions or other ways to combine many individual small claims are important. Contingent fee arrangements are a useful supplement to class action procedures, but are probably not essential. Institutional investors can play an ac- tive role in securities class action suits by representing and organizing individual shareholders. In the United States, for example, the pas- 148 BUILDING A MODERN SYSTEM sage of the Private Securities Litigation Reform Act in 1995 increased the frequency with which institutional investors serve as lead plain- tiffs in securities class action lawsuits. U.S. courts and legal scholars generally agree that as fiduciaries, investment managers have an affir- mative duty to determine whether pursuing litigation is in investors' interests (see Fried and others 1996). Shareholders' rights would be significantly enhanced if derivative suits were permitted, that is, if shareholders or directors were allowed to commence proceedings on behalf of the company when the com- pany is unwilling or unable to do so. Current law does not make the distinction between direct or derivative action by shareholders. A number of countries that did not previously permit suits initi- ated by shareholders against directors and managers are now doing so. Germany recently reduced the ownership threshold that qualifies shareholders to demand legal action (to be brought by the supervisory board or special company representative against managing directors) from a 10 percent equity stake to a 5 percent stake or DM 1 million stake, whichever is smaller, if dishonesty or illegality is suspected. Meanwhile Japan has altered its rules on posting bond to remove dis- incentives for litigation, while the United States is moving back the other way by beginning to rein in the strong incentives for potentially opportunistic litigation. Mobilizing Civil Society to Enhance Protection for Minority Inves- tors. The recent publication of a report on widespread market ma- nipulation and cheating by China's 10 fund management companies in the country's respected business magazine, Caijing Monthly, illus- trates the enormous social benefits of independent public monitoring of stock market practices by the media. Some interesting proposals to encourage such practices are circulating, some of them modeled after successful experience in Taiwan (China), including establishing asso- ciations of minority shareholders with the powers to monitor com- pany behavior and bring legal action in the case of malpractice. Assessing Whether Institutional Investors Can Be Active Minority Shareholders. The literature has described the increase in monitoring by traditionally passive minority investors, such as institutional inves- tors and individuals, as shareholder activism. Theoretical (Admati, Pfleiderer, and Zechner 1992) and empirical (Smith 1996) evidence supports the thesis that large and diversified shareholders may have 149 CORPORATE GOVERNANCE IN CHINA the incentives to expend resources on monitoring despite the presence of free-riding by other shareholders. Institutional investors may have incentives to engage in monitoring when the expected benefits exceed the costs (see box 6.4). Investigators have found that the level of shareholder activism by institutional investors depends on whether they behave like affiliated or unaffiliated investors. When institutional investors have other links with a company, for example, an insurance company that holds a significant portion of a corporation's stock and concurrently acts as its primary insurer, they are unlikely to be active shareholders (Pound 1988). Because of conflict of interest, institutional investors are likely to vote with management. Furthermore, empirical surveys (for example, Ayres and Cramton 1993), have found that institutional investors are more likely to target larger firms, firms that underperform the market, and firms with sig- nificant outside ownership. Institutional investors that commit to hold- ing a firm's equity have increased credibility and influence in monitoring that firm's management. Other factors that influence the level of ac- tivity and the rate of success are ownership levels and stock liquidity. The higher the ownership share, the more seriously management and other shareholders take proposals by an institutional investor. By con- trast, higher stock liquidity makes it more difficult for an institutional outside investor to identify shareholders and solicit their support in a proxy fight. Several factors seem to be important in determining the level of shareholder activism by institutional investors and their role in corpo- rate governance. Other things being equal, limiting opportunities for conflicts of interest by institutional investors, especially pension funds and insurance companies, is likely to result in a higher level of in- volvement in corporate governance. The corporate governance of the institutional investors themselves could affect their incentives to exer- cise shareholders' rights. For example, introducing higher disclosure requirements for institutional investors, such as a provision for fidu- ciaries to disclose how they vote, may increase the cost of collusion with managers.2 The efficiency of the proxy system could be an im- portant factor, because it allows reputable institutional shareholders 2. The lack of such a provision creates an imbalance between the information available to management, who typically know how major shareholders vote, and beneficial owners, that do not. 150 BUILDING A MODERN SYSTEM Box 6.4 CALIFORNIA PUBLIC EMPLOYEES' RETIREMENT SYSTEM AND SHAREHOLDER ACTIVISM The leader in shareholder activism in the U.S. equities market is the California Public Employees' Retirement System (CalPERS). CalPERS is one of the largest public pension funds in the United States and has had an organized shareholder activism campaign since 1986. To en- courage shareholder activism by interested institutional investors, in 1984 CalPERS helped found the Council of Institutional Investors and has been involved in public policy formation at the federal and state levels. Firms that CalPERS considers targets for activism are in its inter- nally managed portfolio. In 1986 CalPERS identified 47 firms held in its portfolio that had implemented poison pills without shareholder ap- proval. Of those 47, CalPERS identified a subset in which it was one of the largest shareholders and in which the level of institutional owner- ship was high, typically greater than 60 percent. Ten firms met these criteria and were selected to be targets of shareholder resolutions re- questing the rescission of the poison pills. During the 1987 and 1988 proxy seasons CaIPERS targeted firms based primarily on their corporate governance structures. In 1987 CalPERS expanded its governance structure criteria to include firms making greenmail payments and firms not using confidential shareholder voting systems. Based on these criteria CalPERS identified seven target firms for the 1988 proxy season. The target selection process changed in 1988 when the primary selection criterion shifted from governance structure to firm performance and the selection process became more sophisticated. The process now begins in June of each year when CalPERS ranks firms based on their last five years' stock returns. It then takes the bottom quartile of approximately 250 firms for further analysis. Firms in the "Bottom 250" are eliminated as potential targets if they have high levels of inside ownership, large ESOPs, or low levels of institu- tional ownership, or if CalPERS is not one of the largest shareholders. The result of this filtering is an annual list of approximately 50 firms, referred to internally as the "Failing 50." Firms in this group are then analyzed further and the Investment Committee identifies approximately 12 targets and 1 corporate governance structure issue for each target that it will pursue in the form of a shareholder resolution. Shareholder resolu- tions have included creating shareholder advisory committees, changing (Box continues) 151 CORPORATE GOVERNANCE IN CHINA Box 6.4 (CONTINUED) the composition of the board of directors and its committees, and restruc- turing executive compensation. The first step in notifying targets is to file shareholder resolutions with the target firms. In 1992 CalPERS tested what it considered to be a less adversarial approach, in which it did not file resolutions with target firms. CalPERS sends a letter to the chairman of the board, the CEO, or both requesting a meeting with CalPERS officials to discuss ways in which the company can meet CalPERS' governance structure goals without a shareholder resolution. If management adopts the proposal or reaches a suitable compromise, CalPERS withdraws the resolution, it does not appear in the proxy statement, and hence it is not voted on. CalPERS prefers settle- ments, because even if shareholder resolutions receive a majority of votes they are often nonbinding. During the first two years, when CalPERS was selecting targets based on governance structure, only 1 of the 15 targeted firms adopted the resolution or made changes sufficient to warrant a settlement. During this period, surveys have found that CalPERS activism had a negative impact on firm market value, possibly because CalPERS may have pro- duced information that had not been incorporated in the stock price prior to targeting. During 1989-93, 26 of the 36 targeted firms either adopted the resolution or settled the first year they were targeted. Dur- ing this period successful targeting was associated with a positive im- pact on the stock price and vice versa. This suggests that activism is beneficial for shareholders if the activist is able to change the organiza- tional control structure of targeted firms. During the period the total wealth increase for CalPERS was approximately $19 million while the estimated cost of activism was about $3.5 million. Thus activism ap- pears to have resulted in a net benefit. SOURCE: Smith (1996). to garner the support of other shareholders relatively easily. Allowing institutional shareholders to make pledges to purchase additional shares contingent on the passage of proposals may increase the credibility of their stated intentions and the chances of successful proxy contests. Factors such as making voting an integral part of institutional inves- tors' fiduciary duties and facilitating coalition building among institu- 152 BUILDING A MODERN SYSTEM BOX 6.5 THE FACTORS BEHIND THE RISE IN U.S. SHAREHOLDER ACTIVISM In the United States, several factors seem to have contributed to the more active role that institutional investors are currently playing in the corporate governance of investee firms. Perhaps the most important was the 1988 ruling by the U.S. Department of Labor (the so-called Avon letter) stating that decisions on voting by pension funds were fiduciary acts of plan asset management under the Employee Retirement Income Security Act, which must either be made directly by trustees or del- egated wholly to external managers. Yet despite the growing impor- tance of mutual funds, no such requirement is in effect for them. In the early 1990s, a ruling by the SEC liberalized coalition building among institutional investors. This has encouraged activism among U.S. share- holders, because institutional shareholders were able to gain influence by acting together without the need for each to amass significant shareholdings. Under the lead plaintiff provision of the U.S. Private Sc- curities Litigation Act of 1995, large shareholders can seek to be named controlling parties in class action shareholder lawsuits against company management. SOURCE: Davis and Steil (2001). tional investors can also play an important role, as evidenced by the U.S. experience (see box 6.5). Clearly the degree of shareholder activism by institutional inves- tors will depend on the size of their shareholdings. The larger the size the more difficult exit becomes and the stronger the incentives to as- sume an active role in corporate governance. The investment policies of institutional investors are among the most important factors deter- mining their approach to corporate governance. In most countries institutional investors have been adopting investment policies based on passive indexation to measure performance, achieve diversifica- tion, and lower transaction costs. However, passive indexation strate- gies have reduced institutional investors' incentives and capacity to affect the behavior of poorly performing companies.' Some have ar- 3. Monks (1997) has argued that indexation strategies have actually encouraged activism by, in effect, forcing institutional investors to hold shares in large compa- nies that form the index, thereby restricting the exit option. 153 CORPORATE GOVERNANCE IN CHINA gued that the use of cumulative voting for board elections is the most powerful tool for allowing institutional shareholders to elect directors that are truly independent and to play an active role in protecting the interests of minority investors (Vittas 1998). Institutional investors have been also active in using the media as a disciplining mechanism in corporate governance. Open public criti- cism of such corporate practices as empire building, excessive mana- gerial compensation, and adoption of antitakeover defenses by institutional investors has been instrumental in mobilizing action by other shareholders and regulators. Collective action by large institu- tional investors through informal groups or associations has proved to be an effective tool of corporate governance. Developing Capital Markets That Reward Good Corporate Governance Chinese capital markets should develop the capacity to create pres- sure and incentives for improvements in corporate governance prac- tices. To this end it is critical to establish a credible threat of market failure and to introduce market players that demand better gover- nance and can contest control. Giving Institutional Investors a Stronger Role in the Future. Institu- tional investors are key to developing a Chinese capital market that rewards companies with good corporate governance practices. Insti- tutional investors operating in a global environment should be able to develop the capacity to recognize, demand, and expect good corpo- rate governance in the form of reputable auditors, timely disclosure of material events, composition and quality of boards, and so on. The development of an institutional investor base will also encourage the development of the investment analyst profession. In addition, institutional investors are expected to play a direct role in improving corporate governance by making available share- holders' rights effective, using the proxy system, and assisting the take- over mechanism. Institutional investors are also expected to provide the extra liquidity needed for the gradual disposal of state-owned and legal person shares. Most important, their presence will allow the regu- lator to adopt and apply a multilayered system of regulations and to enforce regulations more efficiently and effectively through the insti- tutional investors rather than interacting directly with the listed com- 154 BUILDING A MODERN SYSTEM panies. In this way, the hope is that overspeculation and volatility will be reduced. Development of a strong institutional investor base has obvious synergies with other reform objectives, such as reforming SOEs, deep- ening and broadening capital markets, and establishing a social safety net. China could make relatively rapid progress in increasing the role of institutional investors in its capital market development simply by opening up its markets to foreign institutional investors. A plan to liberalize various aspects of the capital account, unify stock markets, and allow greater foreign presence in financial services is gradually emerging. Developing a Market for Corporate Control. Institutional investors can play another important role in corporate governance indirectly through their financing of M&As. M&A activity should help elimi- nate some of the more notable mispricing in the stock markets. For example, striking differences are still apparent between the valuations of H shares and their corresponding A shares and between B shares and their corresponding A shares. These unjustified differences might be substantially reduced, and even eliminated, through M&As. In this way, M&A activities could be an instrument for a gradual conver- gence and unification of China's segmented capital markets. China's legal regime has been evolving in the direction of reduc- ing the costs of acquisitions. For example, the section on acquisitions in the 1993 Stock Trading Provisional Regulations requires anyone who acquires control to offer to buy the shares of all shareholders at the average trading price of the stock in the last six months prior to the acquisition. Because the acquisition of control has invariably in- volved listed companies with state-owned majority shareholders, and because these listed companies invariably need injections of capital by a new majority shareholder, to date the CSRC has waived the need to make a general offer. The Securities Law, which came into effect on July 1, 1999, has cut the cost of M&As by allowing partial bids when more than 30 percent of the shareholdings of a listed company are acquired through trading in the stock markets. The regime will rely strongly on disclosures by the incoming majority shareholders. This focus on transparency will help the market take a more informed view of listed companies and their controlling shareholders. The CSRC is currently preparing a takeover code that is expected to be made pub- lic soon. 155 CORPORATE GOVERNANCE IN CHINA However, the M&A regime will always be imperfect and ineffec- tive if two-thirds of the shares are nontradable. Thus improving the tradability of shares is an essential precondition for developing an effective M&A regime. Furthermore, the existing ownership structure hinders the growth of M&A activities. As powerful owners of listed companies, local governments often oppose M&As that could be as- sociated with a loss of their tax base, layoffs, and closings. However, M&A activity should help break the old model of local government ownership. As a result, even without the rapid divestiture of state as- sets, the industry landscape should become more business oriented than government oriented. M&As will inevitably lead to higher par- ticipation in business by private and foreign investors, thereby intro- ducing management discipline and incentives. The Company Law has several legal provisions that make M&As costly. One is the requirement that all creditors have the right to re- quest to be paid out in the event of a merger or division. In most other markets creditors do not have the automatic right to call for repay- ment if the merged entity can clearly show that it can pay all debts when they come due. In the event that the merged entity cannot pay its debts in full, then most markets adopt an absolute majority rule subject to approval by a judicial or regulatory authority. A mature capital market requires company law that enables M&As to be under- taken efficiently and fairly. Another inhibiting factor is the require- ment in the Company Law that directors cannot be replaced without cause prior to the expiration of their terms. As mentioned, interna- tional practice has moved in the direction of allowing shareholders to replace directors without cause. Improving Disclosure A strong disclosure regime can have a powerful effect on companies' behavior. To improve disclosure by Chinese listed companies, devel- oping a more investor-oriented accounting and disclosure system and strengthening the accounting profession are important. Developing a More Investor-Oriented Accounting and Disclosure Sys- tem. Accounting rules and disclosure requirements should strike a sen- sible balance between users' needs for information, the cost of providing the information, and companies' concerns that giving detailed infor- mation to investors also means giving valuable information to corn- 156 BUILDING A MODERN SYSTEM petitors. Good accounting rules should be designed so as to provide information in a form that is helpful to such users as tax authorities, regulators, and different types of investors, whereas in China they are heavily influenced by the information needs of the tax and regulatory authorities, whose information needs take priority at the expense of investors. Examples relate to provisioning rules and the lack of flex- ibility for using professional judgment, which often results in over- stated profits, and consequently in excessive tax payments. As the Chinese accounting and auditing profession continues to develop its expertise and sophistication, greater flexibility and use of professional judgment within an international accounting frame- work will increasingly be needed. The need for flexibility is also re- lated to the costs of information disclosure and the different and evolving needs of various users. For example, abundant evidence suggests that individual investors, even in developed financial mar- kets, do not base their investment decisions on financial disclosures made in registration or other statements. Analysts and institutional investors are heavy users of financial and accounting information, but their requirements are often different from those mandated by securities laws and regulations.' With the rapid evolution of China's economic environment, im- portant new economic actors are likely to emerge with different infor- mation needs. For instance, the continued liberalization and sophistication of China's capital markets would, over time, increase the importance of risk factors related to movements in interest and exchange rates. Institutional and portfolio investors will become more important. M&As are likely to assume a more prominent role as re- structuring accelerates following entry into the WTO. Intangible as- sets will become more important determinants of overall firm value. These changes will create demands for different types of information. For example, institutional investors attribute significant weight to in- dicators of volatility. Disclosure about derivative and off-balance-sheet 4. Analysts develop their own ways of analyzing and interpreting financial state- ments. For example, according to an ING Barings analyst based in Hong Kong (China), the best valuation tool for companies in China is to focus on their debts, which means disregarding not only their share prices but also their income state- ments, and instead focusing on the ratio of current assets to current liabilities (The Economist, May 24, 2001). In Taiwan (China) the information on research and development expenditures is most valuable (The Economist, May 24, 2001). 157 CORPORATE GOVERNANCE IN CHINA positions will become important. Reporting on good will, brand value, and so on will provide valuable information. Reporting on business combinations is likely to assume increasing importance in an environ- ment where M&As are the main form of restructuring. Many of these standards are not well developed or are lacking in China. With the increased emphasis on flexibility and professional judg- ment, incentives for disclosure will become more important. Listed companies already provide a significant amount of disclosure on a voluntary basis. Understanding the motivation for such practices is critical in order to build on them. Strengthening the Accounting Profession. Perhaps the single most im- portant requirement of all professions is the ability to perform their jobs independently. Prior to 1998, government authorities owned or managed most local firms, and these firms are still struggling to be- come independent business entities and improve their quality. Account- ing firms should make an effort to improve their staff quality, increase risk control, and establish control procedures within the firm. Ac- counting firms should assume modern corporate or partnership forms of business organization. Private ownership is a necessary condition for real independence. The firm's legal form and ownership would have significant implications for its accountants' professional conduct. For instance, current practice is that the CPA who signs the report takes nearly the entire responsibility in the case of litigation. Such a practice reduces senior management's incentives to focus on improv- ing quality control. The Chinese Institute of Certified Public Accountants should gradually weaken its monitoring function while at the same time strengthening its self-regulation function. It should increase the ser- vices it provides to its members, such as training on audit ethics and continuing education on accounting updates and auditing techniques. It should lead the development of the auditing industry and promote the establishment and improvement of relevant laws and regulations to provide a healthier environment for the auditing industry. This pro- fessional, self-regulated body should set up its own code of ethics and conduct to regulate its members and enforce the code strictly. Good self-regulation will not only enhance the independence of its members but will also strengthen the public perception of professional indepen- dence. In China the Ministry of Finance tends to write accounting rules, which primarily reflect the need to provide the information re- quired to collect taxes. Writing good accounting rules requires good 158 BUILDING A MODERN SYSTEM knowledge of how companies operate and how they use loopholes in the rules to portray their performance as better than it really is. This offers some reasons to vest rule writing in a quasi-public organization run by accountants rather than in a government agency. Unlike in other countries, communication and cooperation be- tween external and internal auditors at the time of auditing are mini- mal in China. This is mainly the result of the perceived lack of independence of internal auditors. Various measures could be adopted to strengthen the independence of internal auditors. A minimalist ap- proach would include, for instance, audit committees composed of independent directors with good skills in finance and accounting. Expanding the scope of the internal audit function is also important to focus not only on protecting the interests of shareholders and en- suring compliance with legal requirements, but also on reviewing and improving business performance and internal control functions. The legal system should play an important role in promoting good accounting practices; however, current practices do not promote such a role. The Ministry of Finance approves all standards for the purpose of enhancing their legal binding power, but as Chinese Independent Accounting Standards are considered to apply only to the accounting profession and the Chinese Institute of Certified Public Accountants is not a government agency, other government entities do not comply with these standards. In lawsuits where accounting firms and CPAs are involved, Chinese Independent Accounting Standards are typically not regarded as legal evidence or common acceptable practice from the legal perspective, and therefore cannot be referred to. Accounting and auditing professionals are generally not invited to participate in legal proceedings. Developing a liability insurance system for account- ing and legal firms would greatly enhance the use of the legal system as a controlling mechanism. At present, the market for accounting services does not promote better accounting practices and accountants' independence in relation to their clients. A more rigorous licensing system might assure quality while at the same time creating franchise value and increasing the risks resulting from corrective actions. Activating the Use of Various Corporate Governance Mechanisms Boards of directors, shareholders' rights, well-functioning capital markets, and disclosure are the necessary ingredients of a modern corporate governance system. However, without the existence of 159 CORPORATE GOVERNANCE IN CHINA agents with the need and incentives to use them, these mechanisms are likely to remain mere formalities. Institutional investors can play a catalytic role in activating the use and further development of these instruments. Institutional investors can play an important role in promoting an efficient capital market by providing the depth, the liquidity, and the incentives to discover and exploit arbitrage opportunities. A strong institutional investor base is likely to make the job of the stock market regulator somewhat easier by enhancing its capacity to supervise mar- ket players and promote market stability. Institutional investors are in the best position to develop an investment culture that expects, de- mands, and rewards good corporate governance. The development of an institutional investor base is intrinsically linked to the develop- ment of the investment analyst profession and the associated demand for transparency and disclosure. Under a plausible set of conditions, some types of institutional investors are likely to be active sharehold- ers that make active use of their shareholder rights, and in the process provide some protection to other minority shareholders. China is making progress in developing an institutional investor base by drafting enabling legislation, establishing a pension system, licensing closed and open-end investment funds, reforming its invest- ment trust sector, promoting the insurance industry, and making com- mitments to open up its capital markets to foreign institutional investors. Key factors for building institutional investors' potential to play a strong role in corporate governance are the regulator's capacity to supervise institutional investors and the corporate governance of institutional investors themselves. The pace of building these capabili- ties will determine the speed at which institutional investors will be allowed to enter the market. China has the option of importing many of these ingredients by opening its capital markets to foreign institu- tional investors, and by promoting cooperation between foreign and domestic institutional investors in the form of joint ventures and tech- nical assistance arrangements. Conclusion The role of honest and capable courts and regulators is critical to the evolution of corporate governance in China. However, without the support of markets and incentives, the overemphasis on regulations and rules carries the risk of a triumph of form over substance. Laws 160 BUILDING A MODERN SYSTEM and regulations should focus on giving content to procedural rights and establishing credible liabilities and penalties and effective incen- tive structures so that an evolving system of corporate governance is established that is capable of generating selection pressure for improved corporate governance. This would involve parallel emphasis on in- centives, markets, and the legal system in addition to regulations. The implementation of regulations should itself rely on the proper align- ment of incentives and be based on existing or newly created agencies and associations of accountants, analysts, directors, and others. How- ever, fundamental and sustainable improvements in corporate gover- nance will not take place without fundamental changes in the ownership structure of listed companies. 161  References ADB (Asian Development Bank). 1999. Country Studies under RETA. Manila. Admati, Anat, Paul Pfleiderer, and Josef Zechner. 1992. "Large Share- holder Activism, Risk Sharing, and Financial Market Equilibrium." journal of Political Economy 102(6): 1097-1130. Ayres, Ian, and Peter Cramton. 1993. "An Agency Perspective on Relationship Investing." Unpublished manuscript. Stanford Univer- sity, Stanford Law School, Palo Alto, California. Berle, Adolf, and Gardiner C. Means. 1999. The Modern Corpora- tion and Private Property. New Brunswick, New Jersey: Transac- tion Publishers. Bhagat, Sanjay, and Bernard Black. 1999. "The Uncertain Relation- ship between Board Composition and Firm Performance." Busi- ness Lawyer 54(3). Bhagat, Sanjay, Dennis Carey, and Charles Elson. 1999. "Director Ownership, Corporate Performance, and Management Turnover." Business Lawyer 45(3): 885-916. Bhattasali, Deepak, and Masahiro Kawai. 2001. Implications of China's Accession to the World Trade Organization. Washington, D.C.: World Bank. Black, Fischer. 1986. "Noise." Journal of Finance 41(July): 529-43. Borokhovich, Kenneth, Robert Parrino, and Teresa Trapani. 1996. "Outside Directors and CEO Selection." Journal of Financial and Quantitative Analysis 31(3): 337-55. Broadman, Harry. 1995. Meeting the Challenge of Chinese Enterprise Reform. Discussion Paper no. 283. Washington, D.C.: World Bank. --. 2001a. "The Business(es) of the Chinese State." World Economy 24(7): 849-75. 163 CORPORATE GOVERNANCE IN CHINA 2001 b. "Lessons from Corporatization and Corporate Gov- ernance Reform in Russia and China." Paper presented at the inter- national conference on Corporate Governance Development in Vietnam, October 11-12, Hanoi. Byrd, William, ed. 1992. Chinese Industrial Firms under Reform. Oxford, U.K.: Oxford University Press. Cha, Laura. 2001. "The Future of China's Capital Markets and the Role of Corporate Governance." Speech delivered at the China Business Summit, April 18, Shanghai. Charny, David. 1999. "Workers and Corporate Governance: The Role of Political Culture." In Margaret Blair and Mark Roe, eds., Em- ployees and Corporate Governance. Washington, D.C.: Brookings Institution Press. CLSA (Credit Lyonnais Securities Asia). 2001. "CG Watch. Corpo- rate Governance in Emerging Markets." Compers Paul, and Andrew Metrick. 1998. "Institutional Investors and Equity Prices." Working Paper no. 6723. National Bureau of Economic Research, Cambridge, Massachusetts. Davis, Philip, and Benn Steil. 2001. Institutional Investors. Cambridge, Massachusetts: MIT Press. Dechow, Patricia, Richard Sloan, and Amy Sweeney. 1996. Causes and Consequences of Earnings Manipulation. Contemporary Ac- counting Research. Deng, Xiaoping. 1994. "Excerpts from Talks Given in Wuchang, Shenzhen, Zhuhai, and Shanghai." In Selected Works, vol. Ill. Beijing: Foreign Language Press. Feng, Jinwei. 2001. "Piercing the Corporate Veil." China Law and Practice 7(February): 64. Fried, Frank, Harris, Shriver, & Jakobson. 1996. A Plaintiff Wail: The New Role of Institutional Investors under the Securities Litiga- tion Reform Act. Available on: http://www.ffhsj.com/cmemos/ 0075295.htm#43. Frye, Timothy, and Andrei Shleifer. 1997. "The Invisible Hand and the Grabbing Hand." American Economic Review, AEA Papers and Proceedings 87(May): 354-58. Gelb, Alan, Gary Jefferson, and Inderjit Singh. 1993. "Can Commu- nist Economies Transform Incrementally? The Experience of China." In National Bureau of Economic Research Macroeconomics An- nual. Cambridge, Massachusetts: MIT Press. 164 REFERENCES Gordon, Jeffrey. 1999. "Employee Stock Ownership in Economic Tran- sitions: The Case of United and the Airline Industry." In Margaret Blair and Mark Roe, eds., Employees and Corporate Governance. Washington, D.C.: Brookings Institution Press. Gorton, Gary, and Frank Schmid. 2000. "Class Struggle Inside the Firm: A Study of German Co-Determination." Working paper no. 7945. National Bureau of Economic Research, Cambridge, Massachusetts. Gray, Cheryl. 1997. "Creditors' Crucial Role in Corporate Gover- nance." Finance and Development 34: 29-32. Gregory, Neil, Stoyan Tenev, and Dileep Wagle. 2000. China's Emerg- ing Private Enterprises: Prospects for the New Century. Washing- ton, D.C.: International Finance Corporation. Hanks, Sara. 1999. "Foreign Company Listings in the United States." International Journal of Accounting 34(1): 1-10. Hansmann, Henry. 1996. The Ownership of Enterprises. Cambridge, Massachusetts: Belknap. Harold, Peter. 1992. China's Reform Experience to Date. Discussion Paper no. 180. Washington, D.C.: World Bank. Hockey, Joe. 2001. "Corporate Governance and the Government: Why Sound Governance Means Sound Regulation." Address to the Aus- tralian Institute of Company Directors, Brisbane. Available on: http:// www.minfsr.treasury.gov.au/content/speeches/200'1/. Howson, Nicholas. 1997. "China's Company Law: One Step Forward, Two Steps Back? A Modest Complaint." Columbia Journal of Asian Law 11(1): 127-73. HSBC (Hong Kong and Shanghai Banking Corporation). 2001. "China Strategy Research, China Economic Forum." Beijing. IMF (International Monetary Fund). 1996. People's Republic of China-Recent Economic Developments. Staff Country Report no. 96/40. Washington, D.C. Johnson, Simon, and Andrei Shleifer. 1999. "Coase v. the Coasians." Working Paper no. 7447. National Bureau of Economic Research, Cambridge, Massachusetts. Kaplan, Steven, and Per Stromberg. 2001. "Venture Capitalists as Prin- cipals: Contracting, Screening, and Monitoring." Working Paper no. 8202. National Bureau of Economic Research, Cambridge, Mas- sachusetts. Lerner, Josh. 1996. "The Government as Venture Capitalist." Work- ing Paper no. 5753. National Bureau of Economic Research, Cam- bridge, Massachusetts. 165 CORPORATE GOVERNANCE IN CHINA Levine, Ross. 1997. "Financial Development and Economic Growth: Views and Agenda." Journal of Economic Literature 35(June): 688-726. Li, David. 1998. "Changing Incentives of the Chinese Bureaucracy." American Economic Review, AEA Papers and Proceedings 88(May): 393-97. Lin, Kenny, and K. Hung Chan. 2000. "Auditing Standards in China- A Comparative Analysis with Relevant International Standards and Guidelines." International Journal of Accounting 35(4): 559-77. May, Greg. 1999. Reality Check: Local Protectionism May Hamper China's WTO Compliance. Nixon Center. Available on: http://www. Nixoncenter.org/publications. McCraw, Thomas. 1984. Prophets of Regulations. Cambridge, Mas- sachusetts: Belknap Press. Monks, R. A. 1997. "Corporate Governance and Pension Plans." In M. E. Gordon, 0. S. Mitchell, and M. M. Twinney, eds., Position- ing Pensions for the 21st Century. Philadelphia: University of Penn- sylvania Press. Monteleone, Joseph, and Nicholas J. Conca. 1996. "Directors and Officers Indemnification and Liability Insurance: An Overview of Legal and Practical Issues." Business Lawyer 51(3): 573-634. National Association of Corporate Directors. 1996. NACD's Blue Ribbon Commission on Director Professionalism. Washington, D.C. Naughton, Barry. 1995. Growing Out of the Plan: Chinese Economic Reform 1978-1993. New York: Cambridge University Press. Neoh, Anthony. 2000. "China's Domestic Capital Markets in the New Millennium." China Online, Inc., Chicago. Available on: http:// www.Chinaonline.com. Ong, Kingsley, and Colin Baxter. 1999. "A Comparative Study of the Fundamental Elements of Chinese and English Company Law." In- ternational and Comparative Law Quarterly 48 (January): 88-126. Pannier, Dominique, ed. 1996. Corporate Governance of Public En- terprises in Transitional Economies. Technical Paper no. 323. Wash- ington, D.C.: World Bank. Pound, John. 1988. "Proxy Contests and the Efficiency of Shareholder Oversight." Journal of Financial Economics 29(2): 237-65. Qian, Yingyi. 1999. "The Institutional Foundations of China's Mar- ket Transition." Paper prepared for the World Bank's Annual Con- ference on Development Economics, April 28-30, Washington, D.C. 166 REFERENCES 2001. "Government Control in Corporate Governance as a Transitional Institution: Lessons from China." In Joseph Stiglitz and Shahid Yusuf, eds., Rethinking the East Asian Miracle. New York: Oxford University Press. Rosenstein, S., and J. G. Wyatt. 1990. "Outside Directors, Board In- dependence, and Stockholder Wealth." Journal of Financial Eco- nomics 26(August): 175-91. Shirley, Mary. 2000. Why Performance Contracts for State-Owned Enterprises Haven't Worked. Washington, D.C.: World Bank. Shirley, Mary, and Colin Xu. 2001. "Empirical Effects of Performance Contracts: Evidence from China." Journal of Law, Economics, and Organization 17(1): 168-200. Smith, Michael. 1996. "Shareholder Activism by Institutional Inves- tors: Evidence from CalPERS." Journal of Finance 51(1): 227-52. Tricker, Robert. 1999. "Corporate Governance and the Challenges Ahead." Corporate Governance International (Hong Kong, China) 2(2): 64-81. UBS Warburg. 2001. "China's Fund Managers and Investors." Glo- bal Equity Research. VCChina, The Private Equity Pulse, no. 1, July 2001. Vittas, Dimitri. 1998. "Institutional Investors and Securities Markets: Which Comes First?" Paper presented at the Annual Bank Confer- ence on Development Economics Latin American and Caribbean Conference, June 28-30, World Bank, Washington, D.C. Walder, Andrew. 1995. "China's Transitional Economy: Interpreting Its Significance." China Quarterly 144(December): 963-79. Weisbach, M. S. 1988. "Outside Directors and CEO Turnover." jour- nal of Financial Economics 20(January-March): 431-60. World Bank. 1995. "Bureaucrats in Business." Policy Research Re- port. New York: Oxford University Press. -- . 1996. The Chinese Economy: Fighting Inflation, Deepening Reform. Washington, D.C. -. 1997. China's Management of Enterprise Assets: The State as Shareholder. Washington, D.C. -- . 1999. China: Weathering the Storm and Learning the Les- sons. Country Study. Washington, D.C. -. 2001a. "Bankruptcy of State Enterprises in China." East Asia and Pacific Region, Private Sector Development Unit, Washington, D.C. 167 CORPORATE GOVERNANCE IN CHINA . 2001b. "China's Local Trade Barriers: A Hard Nut to Crack." Transition Newsletter. . 2001c. China Update. Beijing. Wu, Jinglian. 1998. Strategic Restructuring of the State Sector. Beijing: China Development Publishing House. Xiao, Zezhong. 1999. "Corporate Disclosures Made by Chinese Listed Companies." International Journal of Accounting 34(3): 349-73. Xu, Colin, and Yan Wang. 1997. "Ownership Structure, Corporate Governance, and Corporate Performance: The Case of Chinese Stock Companies." Policy Research Working Paper no. 1794. World Bank, Washington, D.C. Yang, Dali. 2000. Can China Overcome Legal Balkanization? Avail- able on: http://www.pnl.gov/china/balkaniz.htm. Zang, junkuo. 2001. "China's State Assets Management System: How the State Exercises Its Ownership." World Bank, Washington, D.C. Zingales, Luigi. 2000. "In Search of New Foundations." Working Paper no. 7706. National Bureau of Economic Research, Cambridge, Massachusetts. 168  Corporate Governance and Enterprise Reform in China Building the Institutions of Modern Markets Stoyan Tenev and Chunlin Zhang with Loup Brefort "The authors of this book, with unparalleled depth of knowledge on China's enduring experience in enterprise reform, provide an up-to-date analysis on tile issue that is central to its transition to market. They demonstrate how corporatiza- tion and ownership diversification, which introduced new institutional forms with- out the dismantling of old ones, have further complicated the universally complex problem of corporate governance. They make a number of recommendations for China's future reform that are economically sensible and politically feasible. I high- ly recommend this book to all who are interested in China's corporate governance reform." Yingyi Qian, Professor of Economics, UNIVERSITY OF CALIFORNIA, BERKELEY "Corporate Governance and Enterprise Reform in China is the most thorough and up- to-date analysis of the issues that China is grappling with as it enters tile World Trade Organization. It sets forth an ambitious agenda of reforms that are required to complete the transition to a modern market economy." Nicholas Lardy, Senior Fellow, BROOKINGS INSTITUTION "Corporate Governance and Enterprise Rcform in China is an extraordinarily rich study of an extraordinarily complex and dynamic topic. At one level, the study offers an essential empirical snapshot of the latest stage of Chinese reform, the effort to build the regulatory and governance mechanisms of a developed industrialized economy. At an even more profound level, the study-in its analysis of China-challenges us to reflect more broadly upon what constitutes the requisite institutional foundations of any modern market system. Given its wide-ranging data and thoroughgoing analysis, this study is 'must reading' for academics and practitioners alike." Edward S. Steinfeld, Associate Professor of Political Science, MASSACHUSETTS INSTITUTE OF TEcHNOLOGY 1 F C [ 15136 INTERNATIONAL lN T CORPORATION TEAwo o the W Id H.nk 2121 Pennsylvania Avenue, N.W. 9 780821 351369 Washington, D.C. 20433 www.ifc.org ISBN 0-8213-5136-2