State Ownership and Corporate Governance in China: An Executive Career Approach STATE OWNERSHIP AND CORPORATE GOVERNANCE IN CHINA: AN EXECUTIVE CAREER APPROACH Li-Wen Lin* China's state-owned enterprises (SOEs) now comprise over sixty percent of the largest 500 companies in China and more than ten percent of Fortune Global 500 companies in the world. Despite their importance to China's domestic economy and foreign investment strategy, many governance characteristics of the SOEs remain a black box, including the SOEs' executive composition and recruitment development. This Article shifts away from the typical focus on how corporations function (i.e., corporate structure) to an examination of the people in charge. Such an approach is important for understanding corporate governance and economic development in countries with weak legal institutions. Using this approach, this Article investigates the legal guidelines for SOE executive recruitment and the evolution of the educational, political, and career attributes of the CEOs of China's large SOEs over the past decade. Furthermore, this Article utilizes legal, historical, sociological, and comparative methods to explain the change and stability of executive composition in China's large SOEs. In sum, SOE executive recruitment shows an orientation toward politically- bounded and firm-specific professionalism, as well as some faint potential for bottom-up and competition-driven marketization. SOE recruitment guidelines and the empirical findings in this Article raise * Assistant Professor, University of British Columbia Faculty of Law. Helpful comments on earlier drafts were received from Curtis Milhaupt, Bruce Carruthers, Martin H6pner, James Mahoney, Saskia Freye, David Stark, Josh Whitford, and Gil Eyal, as well as participants at the Max Planck Summer Conference and workshops at the University of British Columbia Faculty of Law and the Wharton School of the University of Pennsylvania. questions about the adequacy and capacity of existing international laws and enforcement mechanisms to cope with the rise of Chinese SOEs, the challenges to improving Chinese corporate governance, and the underlying forces that form apparent similarities in elite composition across countries. I. Introduction ....................... ........ 744 II. The Organization and Governance of China's SOEs ..750 III. Institutional Reforms and Executive Recruitment.....753 IV. A Typology of Executive Career Pathways ................. 758 A. Pathways to CEO ................. ..... 759 B. Post-CEO Career Development.... ............ 765 V. Rules and Realities: Empirical Analysis ..... ..... 768 A. Descriptive CEO Attributes ............... 769 B. Executive Attributes in Comparative Perspective. ........................... 777 C. Post-CEO Status and Governance ..... ..... 785 D. Implications, Challenges, and Future Research. 788 1. Implications for International Legal Systems. ........................... 789 2. Challenges facing Chinese SOE Governance Reform ........... ........ 792 3. Directions for Future Research ............. 795 Appendix 1: Model Explanation for Table 3 ....................... 798 Appendix 2: Model Explanation for Table 4 ....................... 800 I. INTRODUCTION China's once-dilapidated state-owned enterprises (SOEs) have grown into powerful giants. After three decades of reform, China's SOEs now comprise over sixty percent of the largest 500 companies in China' and more than ten percent of Fortune Global 500 companies in the world.2 Pervasive 1 Wang Zhiyong, China Unveils New Top 500 Companies List, CHINA.ORG.CN (Sept. 3, 2011), http://www.china.org.cn/business/2011- 09/03/content_23344983.htm (last visited Dec. 2, 2013). 2 See Fortune Global 500 2013, CNNMONEY.CoM, http://money.cnn. com/magazines/fortune/global500/2013/full_1ist/ (last visited Dec. 2, 2013). 744 COL UMBIA B USINESS LA W RE VIE W [Vol. 2013 No. 3:743] STATE OKWERSHIP AND CORPORATE GOVERNANCE IN CHINA 745 state ownership has continued with no sign of vanishing as a salient feature of Chinese corporate governance. When analyzing China's SOEs, scholars have typically measured their governance attributes against international standards of corporate governance and have generally concluded that Chinese governance institutions are lacking or dysfunctional.3 This typical approach tends to focus on the function of things (i.e., rules and structures) and often overlooks the character of humans. This approach seeks corporate governance by the rule of law in lieu of the rule of man and promises functional corporate governance that can minimize arbitrariness exercised by human agents. The converse of this underlying philosophy, however, suggests that the personal attributes of corporate leaders significantly affect the quality of corporate governance-especially when legal institutions are weak, as in China. As a result, simply focusing on rules or structures without probing into leadership insufficiently grasps the full picture of the governance of China's SOEs. The political institutions in China further complicate the importance of leadership attributes in SOE governance. A Chinese state-owner is not an ordinary controlling shareholder. The Chinese Communist Party ("CCP" or the "Party") is the real hand in the glove of state ownership in China. As the single ruling party, it controls all the important institutions in politics, business, media, academia, and effectively every sphere of public life in China. The Party's chief control mechanism is its sophisticated but 3 See generally Sonja Opper & Sylvia Schwaag-Serger, Institutional Analysis of Legal Change: The Case of Corporate Governance in China, 26 WASH. U. J.L. & POL'Y 245, 245 (2008) ("[C]hanges in the law have not sufficiently been matched by changes in institutional conditions, explaining why, in spite of far-reaching judicial reform, Chinese corporate governance still displays significant weaknesses in practice."); Zhong Zhang, The Shareholder Derivative Action and Good Corporate Governance in China: Why the Excitement Is Actually for Nothing, 28 UCLA PAc. BASIN L.J. 174, 174 (2011) ("The judiciary is weak, unsophisticated, and riddled with corruption . . . [meaning] it is unrealistic to expect that . . . private enforcement of law in general, can play a significant role in corporate governance."). opaque personnel management over key positions in important institutions, including SOEs.4 As one commentator notes, "[t]he Party's control over personnel was at the heart of its ability to overhaul state companies, without losing leverage over them at the same time."8 By directly managing executives' careers, the Party shapes managerial incentives and in turn influences the corporate behavior of China's SOEs. Recent studies have insightfully suggested that the Party's executive career management fundamentally explains why many Chinese SOEs' practices diverge from traditional principles of corporate law or securities regulation. For example, scholars have found that, in addition to monetary compensation, political promotion acts as another important incentive mechanism.' Moreover, empirical evidence shows that stock options granted to SOE executives are forged to fool foreign investors because such a compensation scheme is incompatible with China's indigenous executive career management system.' The institutionalized personnel rotations between China's SOEs and other government units restrict the exercise of stock options, which can drastically enlarge the pay gap between SOEs and the civil servant system. Relatedly, Professor Katharina Pistor has argued that in China's financial industry, the Party's tight control over the financial cadre's 4 See, e.g., John P. Burns, Research Note, Strengthening Central CCP Control of Leadership Selection: The 1990 Nomenklatura, 138 CHINA Q. 458 (1994) (discussing the party personnel system); Hon S. Chan, Cadre Personnel Management in China: The Nomenklatura System, 1990-1998, 179 CHINA Q. 703 (2004). 5 RICHARD MCGREGOR, THE PARTY: THE SECRET WORLD OF CHINA'S COMMUNIST RULERS 69 (2010). 6 See Jerry Cao et al., Political Promotion, CEO Incentives, and the Relationship between Pay and Performance (Wharton Fin. Inst. Ctr., Working Paper No. 11-53, 2011), available at http://papers.ssrn.com/sol3/papers.cfm?abstractid=1914033. 7 See Zhihong Chen et al., Are Stock Option Grants to Directors of State-Controlled Chinese Firms Listed in Hong Kong Genuine Compensation?, 88 ACCT. REV. 1547. 746 COL UMBIA B USINESS LA W RE VIE W [Vol. 2013 No. 3:743] STATE OWNERSHIP AND CORPORATE GOVERNANCE IN CHINA 747 careers appears to be the dominant governance mechanism over ownership ties and legal rules.' Although scholarship to date has recognized that the Party's control over executive careers has significantly shaped the governance of China's SOEs, its control over personnel management remains obscure to outsiders.' Several fundamental questions regarding the Party's executive management still remain: Who are the top managers? What career pathways have guided the top managers of China's SOEs? What kinds of attributes are advantageous and desired in the executive labor market of China's SOEs? How cohesive is the elite at the highest echelon of China's largest companies? How has executive recruitment evolved over time? Have corporate governance reforms, such as the introduction of boards of directors, changed executive composition? From a perspective of comparative corporate governance, how do Chinese executives differ from their international counterparts? Is the composition of the Chinese business elite moving toward that of the shareholder-oriented model or the stakeholder- oriented model? Answering these questions requires a method for tracing the backgrounds and career paths of executives, termed an "executive-career approach" in this Article. Using this approach, this Article systematically investigates the biographies of the CEOs of China's large non-financial SOEs between 2002 and 2010.10 The emphasis on CEOs assumes that the managerial culture of Chinese companies is highly 8 See Katharina Pistor, The Governance of China's Finance, in CAPITALIZING CHINA 35 (Joseph P.H. Fan & Randall Morck eds., 2013). 9 A number of prominent sociologists have noted that the composition of the business elite remains an important yet absent area of scholarship on Chinese corporate governance and the national economy. See Andrew G. Walder, From Control to Ownership: China's Managerial Revolution, 7 MGMT. & ORG. REV. 19 (2011); Neil Fligstein & Jianjun Zhang, A New Agenda for Research on the Trajectory of Chinese Capitalism, 7 MGMT. & ORG. REV. 39 (2011). 10 Financial and non-financial SOEs are governed by different regulatory regimes in China. This Article focuses on non-financial SOEs. COLUMBIA BUSINESS LAW REVIEW hierarchical and paternalist, with decision-making power concentrated in the highest echelon of the corporate hierarchy.n Moreover, many large Chinese SOEs have not yet established a board of directors and the management power remains concentrated in the top leader ("yibashou") of the corporate entity. The period of investigation (2002-2010) in this Article was chosen to evaluate the evolving dynamics in executive composition during a period of recent institutional reform. Over the past decade, the Chinese government has introduced a variety of rules and guidelines intended to professionalize and marketize executive recruitment. These rules and guidelines help shed light on Chinese corporate governance, but they remain unexplored in the existing literature. The executive recruitment guidelines provide a roadmap for identifying potential changes in executives' attributes, including educational credentials, political qualities, and career experience. In addition to using these regulatory schemes as a basic analytical framework, this Article draws upon three sources of knowledge to analyze the empirical findings on executives' educational, political, and career attributes. The first source comes from China's political and business organization history, which aims to contextualize these findings. Because climbing to the top echelon of the corporate hierarchy usually takes decades, the composition of the contemporary elite is largely a consequence of institutional changes accumulated in the past. The second source derives from sociological theories that have been frequently used to explain career patterns 11 See GORDON REDDING & MICHAEL A. WITT, THE FUTURE OF CHINESE CAPITALISM: CHOICES AND CHANCES 7 (2007); S.H. Kong, An Empirical Investigation of Mainland Chinese Organizational Ideology, 5 ASIAN Bus. & MGMT. 357, 370 (2006); C.J. Zhu et al., Development of HR Practices in Transitional Economies: Evidence from China, 19 INT'L J. HUMAN RES. MGMT. 840 (2008); Michael A. Witt, China: What Variety of Capitalism? 6 (INSEAD, Working Paper No. 2010/88/EPS, 2010), available at http://papers.ssrn.com/sol3/papers.cfm?abstract-id=1695940 ("The available evidence on China suggests ... decision-making is in essence the exclusive prerogative of the very top of the pyramid."). 748 [Vol. 2013 No. 3:743] STATE OWNERSHIP AND CORPORATE GOVERNANCE IN CHINA 749 and achievements. In particular, this Article applies sociologist Ronald Burt's idea of "brokerage and closure" in network theory to explain the comparative advantages of certain executive career pathways and their implications for Chinese corporate governance and the national economy. 12 The third source is executive career studies in the comparative corporate governance field. This literature illustrates how Chinese executives differ from those in other corporate governance regimes, and helps solve the puzzle of how China fits within the taxonomy of comparative corporate capitalism.13 Using this analytical framework, this Article shows that China's executive composition over the past decade reveals indications of both stability and change under institutional reforms. This combination of effects indicates mixed signals for Chinese corporate governance development. Generally, executive recruitment is oriented toward politically-bounded and firm-specific-knowledge professionalism, and indicates a potential trend of bottom-up and competition-pressure- driven marketization. It is a system that strongly favors insiders over outsiders and generates a high degree of closeness and cohesion. While cohesion among the elite may facilitate national policy implementation, it poses challenges to corporate governance improvement due to an increased tendency for groupthink and the perpetuation of old practices, which tend to undermine the implementation of governance reforms. Moreover, through the comparative analysis of the business elite, this Article finds that China looks similar to countries in the stakeholder-oriented model, and obviously different from those in the shareholder- oriented model. However, the apparent similarities in the 12 See generally RONALD S. BURT, BROKERAGE AND CLOSURE: AN INTRODUCTION TO SOCIAL CAPITAL (2005) (explaining that brokerage, which concerns the activity of people exposed to broad social networks, intersects with closure, which describes coordinated interactions within a closed network). 13 See, e.g., REDDING & WITT, supra note 11 (analyzing how China's norms fit into the standard comparative capitalism paradigms); Lin & Milhaupt, infra note 14. elite composition between China and countries in the stakeholder-oriented model are most likely formed by different underlying causes that vary by country. This Article proceeds as follows. Part II briefly describes the ownership and governance structure of China's large non-financial SOEs. It provides an organizational blueprint to contextualize executive recruitment practices and to construct the executive career pathways discussed in the following Parts. Part III discusses the executive recruitment reform rules over the past decade and the implied changes in executive attributes. Part IV explores how executives may have come to power by constructing six types of career pathways based on organizational structure and relational distance. This Article hypothesizes the potential development patterns of each career pathway under institutional reforms and discusses the implications of each career pathway for corporate governance and individual career attainment. Part V empirically examines the evolution of executive attributes under formal institutional reforms. It draws upon the specific institutional setting, social network theory, and comparative corporate governance literature to analyze these empirical findings. Part VI concludes with the legal implications for international regulators, as well as the challenges for and the future research of executive recruitment and corporate governance in China. II.THE ORGANIZATION AND GOVERNANCE OF CHINA'S SOES China's large state-owned non-financial enterprises are typically organized as vertically-integrated corporate groups. Each corporate group has a holding company (known as a ''core company") perched at the top of the ownership hierarchy. The Chinese non-financial SOEs on the Fortune Global 500 list, such as China National Petroleum Corporation, China Mobile Communication Corporation, and China Datang Corporation, are all core companies of a corporate group. Below the core company are a large number of subsidiaries, including listed companies, finance 750 COL UMIA B USINESS LA W RE VIE W [Vol. 2013 No. 3:743] STATE OWNERSHIP AND CORPORATE GOVERNANCE IN CHINA 751 companies, research institutes, and many other related firms along the production chain.14 The ultimate controlling shareholder of the core company in the corporate group is an ownership agency of the central or local government, known as the State-Owned Assets Supervision and Administration Commission ("SASAC"). In addition, there are SASACs at the provincial and lower-government levels. 15 The core company acts as an intermediary between SASAC and group firms that engage in production. This Article focuses on executive recruitment related to the core company in the corporate group. Established in 2003, SASAC is legally tasked with consolidating the shareholder control rights formerly dispersed among various government agencies.16 In practice, SASAC's controlling shareholder status is often overshadowed by the persistent old power structure. SASAC's appointment right is eclipsed by a deep-rooted institutional practice in China; the Organization Department of the Chinese Communist Party controls the human resources management of all important organizations, including SOEs." As a result, SASAC and the Party's Organization Department often jointly release executive management regulations and personnel announcements. In cooperation with the Party's Organization Department, SASAC has introduced several regulatory rules intended to improve the quality of SOE executive teams. In 2008, the Chinese government passed landmark legislation on SOE 14 See Li-Wen Lin & Curtis J. Milhaupt, We Are the (National) Champions: Understanding the Mechanisms of State Capitalism in China, 65 STAN. L. REV. 697 (2013) (analyzing corporate group structure in China). 15 See China State-owned Assets Management System Reform Entering New Stage, SASAC.GOV.CN, (May 22, 2003), http://www.sasac.gov.cn/n2963340/n2964727/n2974401/2976097.html (last visited Dec. 2, 2013). Unless specifically referring to "the central SASAC" or "the local SASACs", this Article simply uses "SASAC" to refer generally to this type of ownership agency in the Chinese government system. 16 For a brief discussion on the role of SASAC, see Lin & Milhaupt, supra note 14, at 734-45. 17 See, Burns, supra note 4, at 462-63. governance, the Law on State-Owned Assets of Enterprises ("SOE Assets Law").' 8 The SOE Assets Law formally defines how SASAC, as the agency authorized to exercise controlling shareholder rights, should manage SOE matters such as executive personnel management.19 According to the SOE Assets Law, SASAC can appoint and evaluate the SOEs' directors and top managers, including CEOs, vice CEOs, CFOs, and other executives.20 This appointment authority blatantly strips away the board of directors' most important power: selecting and evaluating top managers. Thus, while SASAC officially advocates corporate governance reform in the form of establishing SOE boards of directors, the SOE Assets Law (coupled with other related regulations) shows virtually no sign that SASAC (and, ultimately, the Party) will relinquish its authority to appoint the top positions in the most important business enterprises. 21 Without the ability to use appointment authority as a monitoring weapon, it is doubtful that an SOE board of directors can meaningfully improve the quality of 18 Qiye Guoyou Zichan Fa [Law of the People's Republic of China on the State-Owned Assets of Enterprises] (promulgated by the Standing Comm. Nat'l People's Cong., Oct. 28, 2008, effective May 1, 2009) 2008 STANDING COMM. NAT'L PEOPLE'S CONG. GAZ. 19 PAUL HASTINGS, ENACTMENT OF THE ENTERPRISE STATE-OWNED ASSETS LAW (May 2009), available at http://www.paulhastings.com /Resources/Upload/Publications/1309.pdf. 20 Qiye Guoyou Zichan Fa [Law of the People's Republic of China on the State-Owned Assets of Enterprises], art. 22. The core companies of the large state-owned business groups in China are typically 100% owned by the national or local government, with SASAC exercising the shareholder control rights. 21 See Lin & Milhaupt, supra note 14, at 739 n.122 ("In 2008, [the central] SASAC and the Organization Department of the Communist Party promulgated Guidance Opinions on Top Manager Appointments by the Board of Directors of Central Enterprises. These Opinions for the first time gave some appointment power to boards. However, the CEOs of the top fifty-three central enterprises are not covered by the Opinions. Even with respect to other enterprises, the nomination committee of the board is required to 'fully consult' with the Party Committee and SASAC before nominating a CEO. The preliminary appointment must be filed with SASAC before the appointment becomes final."). [Vol. 2013752 COL UMBIA B USINESS LA W RE VIE W No. 3:743] STATE OWNERSHIP AND CORPORATE GOVERNANCE IN CHINA 753 executive management teams. Even with the appearance of boards of directors, the Party truly orchestrates SOE governance in China.22 III.INSTITUTIONAL REFORMS AND EXECUTIVE RECRUITMENT Since 2003, SASAC and the Party's Organization Department have introduced rules and guidelines intended to improve the quality of SOE executive teams. These reform policies aim to professionalize executive teams and broaden the executive labor market.23 In turn, the evolution of executive recruitment reforms may change the composition of the business elite. This Part discusses how professionalization and marketization reforms may cause changes in three dimensions: political attributes, educational backgrounds, and career pathways. Due to the particular complexity of career pathways, Part IV discusses this topic in depth. Professionalization. State ownership often raises concerns of politics overriding professionalism. Seniority, personal connections ("guanxi"), and political loyalty are important factors in employment and promotion 22 See Nicholas Howson, China's Restructured Commercial Banks: Nomenklatura Accountability Serving Corporate Governance Reform?, in CHINA's EMERGING FINANCIAL MARKETS: CHALLENGES AND GLOBAL IMPACT 123, 140 (Zhu Min, Cai Jinqing & Martha Avery eds., 2009) (arguing that despite the use of boards of directors, the Party Committee-which is subject to control of the Party system-is the "real power" in China's largest state-owned commercial banks); Lin & Milhaupt, supra note 14, at 737 (noting that SASAC exercises its power to select and compensate top managers "in the shadow of party control"). 23 Rather than reviewing in detail each regulatory scheme on executive recruitment, this Article provides a summary of key points derived from the relevant rules. See, e.g., Provisional Rules on Corporate Leaders of Central SOEs (promulgated by the St. Council, effective 2009); Provisional Measures on Comprehensive Evaluation of Corporate Leadership Teams and Leaders of Central SOEs (promulgated by the St. Council, effective 2009); Guidance on Public Recruitments for Senior Managers of Central SOEs (promulgated by the St. Council, effective 2004). Local SASACs also have similar rules. opportunities in China, especially for state-affiliated organizations like SOEs.24 This Part addresses how professionalization reforms could occur and what factors would subsequently change. SASAC executive reform rules suggest that professionalization will only occur in a politically-bounded fashion because political allegiance to the Chinese Communist Party remains paramount when selecting and evaluating top managers of SOEs.25 The corporate governance system fortifies political quality control. Chinese SOEs have two parallel systems in personnel management: the regular corporate management system and the party system.2 6 Corporate management system positions, similar to those commonly found in firms elsewhere, include a CEO, vice CEO, chief accountant, and-if the company has a board of directors-board chairman and independent board members. Party system leadership teams include the secretary (and several deputy secretaries) of the party committee, a secretary of the discipline inspection commission (e.g., an anticorruption office), and other members of the party committee. Formal policy governing both corporate and party leadership teams dictates that a corporate manager of a given rank typically holds a position of equivalent rank in the party system.27 The link between 24 See Andrew G. Walder, Career Mobility and the Communist Political Order, 60 AM. Soc. REv. 309, 309 (1995) (discussing the communist party's role in executive promotion within China's divided career path system, and noting that the path that considers political credentials leads to positions with greater authority and more material privileges than the path that only considers educational credentials). 25 See Provisional Rules on Corporate Leaders of Central SOEs, art. 4(1); Provisional Measures on Comprehensive Evaluation of Corporate Leadership Teams and Leaders of Central SOEs. 26 See Provisional Rules on Corporate Leaders of Central SOEs, art. 9. 27 See Lin & Milhaupt, supra note 14, at 737 n.118 ("In 2004, the Organization Department of the Chinese Communist Party and the Party Committee of SASAC released Guanyu Jiaqiang he Gaijin Zhongyang Qiye Dangjian Gongzuo de Yijian [Opinions Concerning Strengthening and Improving the Party Construction Work in the Central Enterprises] (promulgated by the Org. Dep't Communist Party of China & Party COL UMBIA BUSINESS LA W RE VIE W [Vol. 2013754 No. 3:743] STATE OWNERSHIP AND CORPORATE GOVERNANCE IN CHINA 755 the two leadership teams aims to ensure consistency between corporate decision-making and party policy. With political loyalty to the Chinese Community Party entrenched as a primary requirement, this suggests that the Party has no intention of relinquishing control over the largest economic organizations in China. Professionalism is unlikely to reach areas requiring or triggering sensitive political reforms. Rather, permissible reforms are expected to occur in less politically sensitive areas, such as age, education, specialization, work experience, and moral integrity, as envisioned in SASAC's executive reform guidance. 28 Part of the professionalization scheme is to bring young managers into-and retire old managers out of-leadership teams.29 Young managers are desirable because they may be Comm. SASAC, Oct. 31, 2004). A key principle of the Opinions is the policy of 'bilateral entries and cross appointments.' The term bilateral entries means that members of the Party Committee can serve on the board of directors, the supervisory board, and the top management team, while board members and top managers who are party members can join the Party Committee. The term cross appointments means that, if the company has a board of directors, the secretary of the Party Committee and the board chair can be assumed to be the same person. If the company does not have a board of directors, then the secretary of the Party Committee can be assumed to be the CEO, and the Vice-CEO can be assumed to be the deputy secretary of the Party Committee.") (emphasis added) (citation omitted). 28 See infra notes 29-34 and accompanying text. 29 In December 2004, the central SASAC issued a regulatory order to the fifty-three most important central SOEs regarding the mandatory retirement age of the top managers. According to the order, ministerial- rank managers were strictly required to retire upon reaching the age of sixty-five, and the vice-ministerial-rank managers upon reaching the age of sixty. Local SASACs declared even more aggressive requirements. For example, according to the Provisional Rules on Corporate Leaders of SOEs under Heifei City SASAC, the chief-position executives, such as the board chairman and CEO, should not exceed the age of fifty, and the vice- position executives, such as vice-CEO, should not exceed forty-five years old. Furthermore, the reappointed chief-position executives should not exceed the age of fifty-five and the reappointed vice-position executives should not be older than fifty-two. See He Dan, Most Opposed to Increasing Retirement Age, CHINA DAILY (Sept. 1, 2013, 11:23 PM), http://usa.chinadaily.com.cn/china/2013-09/01/content_16935872.htm COLUMBIA BUSINESS LAW REVIEW more active and innovative, while less influenced by old traditions. The professionalization reform also sets educational requirements. Since academic credentials are a proxy for intellectual ability, the executive reform rules require executives to have a minimum of a bachelor's degree. 0 Moreover, while Chinese elites have traditionally been trained in engineering, 31 non-engineering-trained executives are expected to be on the rise as the division of labor grows increasingly complex in large business organizations. In terms of career experience, organizations should prefer business-related work experience to political career experience. Traditionally, China's SOEs were managed similarly to other government units, and incompatibly with modern business management. To address this concern, SASAC's reform rules require executives to have at least ten years of work experience in business organizations.32 Besides the general requirements of age, education, and work experience, moral integrity is a specific quality that needs significant improvement. Because China's SOEs have been plagued by corruption,33 executive reform policies repeatedly emphasize that executives should have rectitude, a law-abiding nature, and a good public image.34 (nearly 95% of the 25,300 polled voiced opposition to increasing China's mandatory retirement age, which currently stands at sixty for male employees, fifty-five for female officials, and fifty for female workers). 30 See Provisional Rules on Corporate Leaders of Central SOEs, art. 5. 31 See generally CHENG LI, CHINA'S LEADERS: THE NEW GENERATION 37, 40 (2001) (noting that, of the college-educated Chinese elites who received promotions in the 1980s, most were trained in engineering and natural sciences, including 73.8% of ministers and vice ministers in 1988 and 74.6% of mayors in 1986). 32 See Provisional Rules on Corporate Leaders of Central SOEs, art. 5. 33 See Michele Lee, Note and Comment, Franchising in China: Legal Challenges When First Entering the Chinese Market, 19 AM. U. INT'L L. REV. 949, 954 (2003). 34 See Provisional Rules on Corporate Leaders of Central SOEs, art. 1; Provisional Measures on Comprehensive Evaluation of Corporate Leadership Teams and Leaders of Central SOEs (promulgated by the St. Council, effective 2009); Regulations on Honest and Ethical Business 756 [Vol. 2013 No. 3:743] STATE OWNERSHIP AND CORPORATE GOVERNANCE IN CHINA 757 Marketization. Previously, the SOE executive headhunting process used to be entirely shrouded in mystery. Employment information and opportunities were available only to people inside the system. As a result, applicants were all system-insiders who perpetuated old habits because they lacked access to new management skills. To address this problem, the central and local SASACs- with the Party's endorsement-have openly solicited job applications for hundreds of executive positions since 2003.35 These positions include CEOs, vice CEOs, and chief accountants (equivalent to CFOs) of the core companies under their supervision. Those interested in the executive openings may submit their applications and undergo paper- based qualification reviews, standardized written examinations, and face-to-face interviews.36 This executive headhunting targets both China's domestic labor market as well as overseas talent. The political, educational, and work experience requirements discussed above are relaxed for executives recruited from overseas through this process.37 In 2010, thirty-one senior managers of central SOEs were successfully recruited through this process, plucked from 1,410 applicants worldwide. 38 The Chinese government has praised this new recruitment strategy as a bastion of "openness, fairness, competitiveness, and meritocracy." 39 In theory, marketization of the recruitment process suggests that political affiliation with the Chinese Practice Behavior of SOE Corporate Leaders (promulgated by the St. Council, effective 2009). 35 See Press Release, SASAC, SASAC's Press Release on Central State- Owned Enterprises' Public Recruitment for Top Managers (June 14, 2006) (on file with author). 36 See Guidance on Public Recruitments for Senior Managers of Central SOEs (promulgated by the St. Council, effective 2004). 37 See Provisional Rules on Corporate Leaders of Central SOEs, art. 6. 38 See Press Release, The First Bureau of Corporate Leader Management of SASAC of the State Council (April 1, 2011) (on file with author). 39 See Press Release, Recruitment Decisions by SASAC of the State Council (May 10, 2005, Aug 28, 2006, and May 22, 2007) (on file with author). Communist Party may become a less important factor for executives recruited through this process, since the value of professional experience accumulated outside the state system may compensate for a lack of political affiliation. In other words, it opens up opportunities for outsiders- professionals who build up their careers in private or foreign companies-to parachute directly into the top of the SOE system. This public recruitment process also suggests a potential increase in the number of executives who have been educated abroad. IV.A TYPOLOGY OF EXECUTIVE CAREER PATHWAYS Tracing executive career pathways is a useful method for investigating what aspects of career experience have changed under the professionalization and marketization reforms. This Article constructs six types of career pathways based on the prevailing organizational structures of China's SOEs and the "organizational distance" between the SOE at issue and the organizations that its CEO worked for previously. I measure this distance by whether the prior organization is related to a given SOE in terms of supervisory authority in the government system and whether the prior organization is an institution outside the state system (i.e., a private or foreign company). Organizations that fall outside the supervisory or state system are considered more distant from the relevant SOE. This typology allows for evaluation of the degree of personnel integration between SOEs and other government units, as well as the degrees of professionalization and openness of the executive labor market. Based on this methodology, this Article assesses the features of each career pathway in attaining the CEO position and the corresponding implications for corporate governance. Moreover, rather than simply focusing on how CEOs came to power, this Article considers what they do after leaving their executive positions. Tracking post-CEO status generates a more complete picture of the integration between SOE personnel and other government units. 758 COL UMBIA B USINESS LA W RE VIE W [Vol. 2013 No. 3:743] STATE OWNERSHIP AND CORPORATE GOVERNANCE IN CHINA 759 Furthermore, analyzing post-CEO status helps to better understand managerial incentives and to evaluate the legal system's role in punishing top managers' illegal behavior. A. Pathways to CEO The Single-Group Track. China's large non-financial SOEs are typically organized as corporate groups registered with the state.40 A corporate group is legally required to be comprised of a parent company and at least five controlled subsidiaries. When an executive has spent his or her whole career within one corporate group before being appointed CEO of that group, the executive's career follows a single- group track. The theory underlying the single-group track is centered on climbing the corporate ladder from the bottom to the top. If this track represents the mainstream route, it suggests that a group-based seniority system operates to select the top managers of China's SOEs. According to social network theory, executives who rise via the single-group pathway possess a high volume of firm-specific knowledge and insider network resources. 4 1 In circumstances where firm-specific knowledge and insider social connections are valued, managers following the single-group track wield great advantages in attempting to reach the CEO position. From a corporate governance perspective, the single-group track assures some measure of professionalism, given the CEO's rich firm-specific knowledge. It also presents, however, an increased risk of excessive power concentration, especially 40 See Provisional Rules on Business Group Registration (promulgated by St. Council, effective 1998) (discussing how, in order to form a group, the parent company must have registered capital of at least 50 million RMB (about $7.5 million) and at least five subsidiaries. The total registered capital of the parent and its subsidiaries must be at least 100 million RMB ($15 million)). 41 For the early leading literature exploring the importance of social networks to job opportunities, see generally MARK GRANOVETTER, GETTING A JOB: A STUDY OF CONTACTS AND CAREERS (2d ed., 1995) (arguing that the best jobs are found through personal contacts and developing a causal model to explain how job information is acquired through such contacts). because the board of directors is usually dysfunctional (or absent) as an internal monitoring mechanism and China's external governance institutions remain weak. The Multi-Group Track. This career pathway is characterized by routes where the executive has worked in other ' state-owned business groups prior to CEO appointment. At first blush, SOE executives following the multi-group track appear similar to those changing jobs across employers in other executive labor markets. But unlike other labor markets normally governed by the invisible hand, China's SOE executive market is centrally managed by the party-state. Thus, these job movements are not as free as in other labor markets. Setting aside personal reasons for job change across organizations, an important institutional reason for forming the multi-group track is the party-state's personnel rotation management system. Personnel rotation is a legally institutionalized system for managing civil servants in Chinese government units; because the boundary between government units and SOEs is porous, the system in practice spills over into SOEs.4 2 The party-state frequently rotates top managers across business groups of the same industry. For example, in April 2011, the state rotated the CEOs of the three central petroleum SOEs in China, each of which is a Fortune Global 500 company. 43 From the party-state's perspective, these executive rotations serve two institutional functions. First, since institutionalized corporate oversight organs, such as boards of directors, have yet to fully develop, executive rotations can help reduce excessive concentration of managerial power in a business group. Second, personnel rotations facilitate 42 According to the Provisional Measures on Rotations of Civil Servants, enacted in 1996 and still effective today, a civil servant who has served in a leadership position for five or more years should rotate. According to the regulation, personnel rotation facilitates operation efficiency and prevents corruption. See Zou KEYUAN, CHINA'S LEGAL REFORM: TOWARDS THE RULE OF LAW 137 (2006). 43 China Moves Top Bosses in Oil Sector, WALL ST. J. (Apr. 9, 2011, 12:01 AM), http://online.wsj.com/news/articles/SB1000142405274870 4503104576250243224429846. 760 COL UMIA B USINESS LA W RE VIE W [Vol. 2013 No. 3:743] STATE OWNERSHIP AND CORPORATE GOVERNANCE IN CHINA 761 management skills sharing among SOEs.44 Personnel rotations allow executives who gain useful skills in one business group to share their experience with another group. For instance, one purported reason for rotating CEO Chengyu Fu from China National Offshore Oil Corporation ("CNOOC") to Sinopec Group-two of China's three petroleum giants-was that Fu's rich overseas business experience with CNOOC could alleviate Sinopec's frustrations with overseas expansion. s The Supervisory-Bureau Track. An SOE executive comes to power through the supervisory-bureau pathway by primarily developing his or her career in the SOE's supervisory bureau(s). China's large SOEs were carved out of industry-based government ministries or bureaus in the corporatization process, which transformed government organs into joint-stock companies. For industrial matters, these corporatized SOEs remain supervised by government ministries (bureaus). Because of the historical integration in organizational structures and the continuing supervisory relationships in business, personnel exchanges between the SOEs and their supervisory bureaus are quite frequent.4 6 Government officials can gain firm-specific knowledge and build social connections with employees by supervising SOEs. In addition, supervisory government officials tend to have industry-wide information and social connections. This social and informational capital can increase the usefulness of the supervisory-bureau track in securing executive positions. Anecdotal evidence indicates that many government officials take top-tier positions in SOEs in preparation for retirement. Such transitions are attractive because large SOEs usually offer better monetary compensation than other 44 See Pistor, supra note 8, at 42-45 (similarly advocating the benefit of managerial rotations among the state-owned financial institutions in China). 45 Yiming Zhang, Sinopec and CNOOC Welcome New Heads, Top Leader Rotations Facilitate Cooperation, CHINA EcoN. TIMES (April 11, 2011). 46 See Lin & Milhaupt, supra note 14, at 701-02. COLUMBIA BUSINESS LA WREVIEW government units do. In sum, it is practicable for senior officials in the supervisory bureaus to adopt this path because they have accumulated relevant social and knowledge capital. Due to the late-stage nature of retirement, executives coming through the supervisory- bureau pathway are expected to be senior in age. This retirement transition pattern, in fact, is not unique to China. In Japan, there is an institutionalized practice known as amakudari, where senior bureaucrats who reach mandatory retirement age (usually between fifty and sixty years old) retire to join private companies or SOEs linked with, or under, the jurisdiction of their ministries or agencies. The practice of amakudari has often been criticized as corrupt and obstructive to regulatory reforms.4 8 For China's SOEs, the Japanese experience suggests that the supervisory- bureau pathway should be restricted in order to further executive professionalization and corporate governance reform. The Unrelated-Government- Units TRack. An executive follows the unrelated-government-units track when, prior to CEO appointment, his or her career primarily develops in government-affiliated organs other than for-profit SOEs and supervisory bureaus. Such government units are relatively irrelevant to the focused SOE in terms of the type of organizational identity or the nature of business matters. For example, Biting Chen first worked his way up in the Party system to become the Party Secretary of the Youth 47 See Ulrike Schaede, The "Old Boy" Network and Government- Business Relationships in Japan, 21 J. JAPANESE STUD. 293, 293 (1995). 48 See, e.g., Lonney E. Carlile & Mark C. Tilton, Is Japan Really Changing?, in Is JAPAN REALLY CHANGING ITS WAYS? REGULATORY REFORM AND THE JAPANESE ECONOMY, 197, 210-16 (Lonney E. Carlile & Mark C. Tilton eds., 1998); see also Elizabeth Norville, The "Illiberal" Roots of Japanese Financial Regulatory Reform, in Is JAPAN REALLY CHANGING ITS WAYS? REGULATORY REFORM AND THE JAPANESE ECONOMY, 111, 111-14 (Lonney E. Carlile & Mark C. Tilton eds., 1998) (contending that financial regulatory reform in Japan needs more than just "fair and transparent rules"). 762 [Vol. 2013 No. 3:743] STATE OWNERSHIP AND CORPORATE GOVERNANCE IN CHINA 763 League Committee of Anhui Province.49 He later became the mayor of a city in Jiangsu Province, chief secretary of Jiangsu Province, and eventually the vice-governor of the province. Finally, he was appointed as the CEO of Shenhua Group, one of the largest energy SOEs in China and also a Fortune Global 500 company.50 Executives with this career pathway tend to possess more skills in political operation rather than firm-specific or industry-specific knowledge. If the unrelated-government-units track is the prevailing career pathway, it suggests that the management logic of China's SOEs is oriented more toward political than business purposes. Under professionalization reform, the unrelated- government-units track will likely decline in prevalence. The Multi-Sphere Track. An SOE executive comes to power through the multi-sphere track when the executive's career spans multiple types of organizations, such as unaffiliated SOEs, supervisory bureaus, and other unrelated government units. For example, Shulin Su first worked his way up from an entry-level technician of a subsidiary up to the vice CEO position in the core company of petroleum giant and SOE Sinopec.51 He was then transferred to the Party Standing Committee of Liaoning Province and was later appointed as the CEO of China National Petroleum Corporation, one of the largest oil companies in China. 52 This type of career pathway is essentially a combination of the previous four types. Compared to single-group executives, multi-sphere executives are more likely to have system-wide knowledge and diverse personal connections, which can facilitate career advancement. The prevalence of the multi-sphere track would suggest high personnel 4 Press Release, China Ming Yang Wind Power Grp. Ltd., MY Announces Board Change, Further Enhances Board Diversity (June 16, 2011) (on file with author). 50 Id. 51 Erica Downs & Michal Meidan, Business and Politics in China: The Oil Executive Reshuffle of 2011, CHINA SECuRITY, no. 19, 2011, at 8-9. 52 Ellennor Grace M. Francisco, Petroleum Politics: China and Its National Oil Companies, at 23 (June 26, 2013) (unpublished masters thesis, European Institute of High International Studies). COLUMBIA BUSINESS LAW REVIEW integration between SOEs and other government units. This phenomenon may be detrimental to corporate governance because SOEs would be at a higher risk of being managed in ways similar to other government units. Nevertheless, the multi-sphere track may generate greater cohesion through shared career experiences among the political and business elite, which can facilitate economic coordination and policy implementation at the national level. The System-Outsider Track. All of the aforementioned career pathways fall strictly within the boundaries of the state system. Executives traveling on these pathways are savvy system-insiders. Conversely, system-outsiders accumulate their career experiences in organizations unaffiliated with the state, such as private or foreign companies. Unlike system-insiders embedded in a closed network, system-outsiders possess "brokerage ties" of connectivity outside the state system that tap into new ideas and resources.53 Executives coming from the system-outsider path are institutional brokers importing new management knowledge into the SOEs, which otherwise tend to be trapped in an outdated mentality. The participation of system-outsiders in the SOE system represents improvement in corporate governance. Although system-outsiders can use th*eir brokerage advantages to break into the SOE system and win top management positions, it is uncertain whether these advantages can overcome the "liability of foreignness"-the competitive disadvantages that foreigners face when entering into a local system. 54 China's SOEs rank employees 53 For the leading analysis on the concept of forming brokerage ties outside a closed system, see BURT, supra note 12, at 12-13. In a simple sense, "brokerage" is defined as a connection between two or more unrelated agents or groups. 54 The "liability of foreignness" term is most commonly used in business literature, and refers to foreign firms that are less competitive compared to local firms when entering a local market. See, e.g., Srilata Zaheer, Overcoming the Liability of Foreignness, 38 ACAD. MGMT. J. 341, 341 (1995). 764 [Vol. 2013 No. 3:743] STATE OWNERSHIP AND CORPORATE GOVERNANCE IN CHINA 765 according to a strict administrative hierarchy, which parallels the government's civil service administration. Despite several attempts to abolish this system, it continues to persist today. A system-outsider who parachutes into an executive post, therefore, would disrupt internal promotion expectations based on administrative ranks. Diverging from institutionalized experiences would also cause a legitimacy problem. Moreover, system-outsiders often lack local personal connections, such as guanxi, which restricts their access to job information and opportunities within the state- affiliated organizations. As a result, system-outsiders may need to build outstanding external reputations for professionalism in order to overcome their liability of foreignness. SASAC's marketization reform of the SOE executive labor market is expected to lead to more executives from the system-outsider path, particularly in industries where competition depends on innovation and efficiency. Competition pressures may increase the likelihood that new strategies will be adopted that diverge from old recruitment practices." But due to suspicion of outsiders, the system- outsider track may be less likely to take hold in industries of critical national security. Rather, the Chinese SOEs that are likely to embrace system-outsiders tend to be in industries relatively open to the private sector, such as steel, automobile manufacturing, and light industries. B. Post-CEO Career Development The government's executive recruitment guidelines set forth not only criteria for selecting SOE executive candidates, but also standards for removing executives. Grounds for dismissal include, for example, failing to meet performance targets in the absence of objective causes, reaching mandatory retirement age, having health problems, committing serious violations of law such as corruption, and 55 Christine Oliver, The Antecedents of Deinstitutionalization, 13 ORG. STUD. 563, 574 (1992). undertaking other job responsibilities." These dismissal standards raise questions about post-CEO career development, which is an important dimension of the executive personnel system. For instance, what types of CEOs are more likely to step down due to legal liabilities? Is there any path-dependent effect-i.e., staying in the same pathway after a CEO post-in the executive training system of China's SOEs? The existence of path-dependent effects signals some predictability in how the state oversees SOEs' top managers, while the absence increases the difficulty of understanding the executive training system. As explained previously, the single-group track CEO has firm-specific knowledge and insider social connections. This intellectual and social capital can help one attain both the CEO position as well as other leadership positions in the same group, such as the chairmanship of the board. Thus, the single-group CEO may be likely to stay in the same group after leaving the CEO position. If so, there is a path- dependent effect in the single-group track that impacts subsequent career moves. Similarly, the multi-group track and the unrelated-government-units track may also have path-dependent effects, as CEOs of these types possess intellectual and social capital across multiple groups or spheres, which may give them advantages in moving across spheres in the government system. The path-dependent effect may also exist in the supervisory-bureau track. However, since CEOs of the supervisory-bureau track tend to be senior in age, retirement would be another common career outcome. The post-CEO status of those from the system-outsider track is uncertain. It may depend on how the system- outsider fares in the state system. Anecdotal evidence suggests that system-outsider managers often leave the SOE very soon after arrival due to poor acclimation to the state 56 See Provisional Rules on Corporate Leaders of Central SOEs (promulgated by the St. Council, effective 2009) ch. 8. [Vol. 2013766 COL UMBIA B USINESS LA W RE VIE W No. 3:743] STATE OWNERSHIP AND CORPORATE GOVERNANCE IN CHINA 767 system.57 Because system-outsiders often retreat so quickly, it greatly complicates efforts to improve SOE governance through reliance on external human resources. Finally, in addition to staying on the same path or switching to a different trajectory, post-CEO career development can be disjointed and miserable. As discussed in Part III, supra, corruption control is a major theme in executive reform schemes. The Chinese party-state has the power not only to promote CEOs to other, higher positions in the government system, but also to punish them severely through its judiciary machinery. While the probability of ending up in jail seems low, the possibility remains quite real-as evidenced by a number of recent cases where high- profile executives of China's leading SOEs were ousted and presented with serious criminal liabilities related to their executive duties.58 Because career outcomes are often the result of an accumulation of previous experiences, CEOs who come to power through different career pathways may face different risks of criminal behavior such as corruption. For example, an intuitive concern is that the unrelated-government-units track may be more likely to produce corrupt CEOs, due to their politically-tainted backgrounds and lack of business management experience. In this regard, the single-group track-with its rich, firm-specific knowledge, but lack of 57 Zai Guoqi Shuitu Bufu Fansi Haigui Shifuo Yuewei [Rethinking the Use of Overseas Returnees Given the Problem of Acclimatization in State- Owned Enterprises], Directors and Boards (Feb. 4, 2006), http://news.xinhuanet.com/overseas/2005-02/05/content2551055.htm. 58 Prominent cases include (1) Tonghai Chen, the former CEO of Sinopec Group, who was sentenced to the death penalty for corruption in July 2009 (Beijing Court Convicts Ex-Sinopec Chief of Bribery, N.Y. TIMES DEALBOOK (July 15, 2009, 5:17 AM), http://dealbook.nytimes.com/2009/ 07/15/beijing-court-convicts-ex-sinopec-chief-of-bribery/); and (2) Rixin Kong, the former CEO of China National Nuclear Corporation, who was sentenced to life in prison for corruption by the Beijing's First Intermediate People's Court in November 2011 (Top 10 Toppled Executives, CHINA DAILY (Dec. 24, 2009, 9:11 AM), http://www.chinadaily.com.cnIbusiness/2009-12/24/content_9225428. htm#name6. external ties-may seem less problematic. However, the single-group track may still be susceptible to corruption because it may invite the excessive concentration of authority and abuse of power, especially when there is no effective internal monitoring mechanism, such as a board of directors. V.RULES AND REALITIES: EMPIRICAL ANALYSIS Parts III and IV propose potential changes to current executive recruitment practices. As always in China, however, formal institutional reforms do not guarantee actual changes. To assess the real development of China's SOE executive recruitment under these institutional reforms, this Part examines the biographical backgrounds of CEOs of the SOEs among the largest 500 companies (by revenue) in China, according to annual rankings by the China Enterprise Confederation and China Enterprise Directors Association.59 More than sixty percent of the largest 500 companies in China are SOEs.60 Using these rankings is advantageous because it accounts for the fact that large enterprises in China are organized as business groups, in which core companies control a large number of subsidiaries (including listed firms). This Part focuses on the CEOs of the core companies in the large state-owned business groups. In order to track the effects of the institutional reforms over the past decade, this Part examines data from three distinct years (2002, 2005, and 2010) spanning a period before and after major reform measures. The graph in Appendix 1 illustrates the sequence of the data years and the important institutional reform events discussed supra in Parts II and III. The CEOs' biographical information derives from multiple sources, including corporate prospectuses, annual reports, corporate websites, government documents and websites, industrial 59 Top 500 Enterprises of China, http://www.cec-ceda.org.cn/china- 500/english/ (last visited Dec. 2, 2013). 60 See Zhiyong, supra note 1. COL UMBIA B USINESS LA W RE VIE W [Vol. 2013768 No. 3:743] STATE OWNERSHIP AND CORPORATE GOVERNANCE IN CHINA 769 association websites, and news reports. After excluding missing data, the sample size of CEO biographical profiles is 272 in 2002, 274 in 2005, and 273 in 2010.61 The total sample includes 612 distinct CEO profiles. A. Descriptive CEO Attributes Table 1 summarizes the descriptive attributes and patterns of change in education, political affiliation, and career pathways from 2002 to 2010. 61 If no data were missing, there would be 352 observations in 2002, 333 in 2005, and 301 in 2010. 770 COLUMBIA BUSINESS LAW REVIEW [Vol. 2013 TABLE 1: CEO ATTRIBUTES OF LARGE STATE-OWNED NON- FINANCIAL ENTERPRISES IN CHINA, 2002-20 10 2002 2005 2010 General Attributes Male 96.7% 96.4% 97.4% Main Birth Place (i.e. Shandong Province) 13% 14.1% 11.5% (N=185) (N=206) (N=182) Educational Attributes Graduate Degree 37.5% 49.3% 57.1% Foreign Degree 2.9% 3.3% 4.4% Tsinghua University 2.9% 2.9% 5.5% 09 League 15.8% 12.8% 11.4% Engineering 61.8% 62.8% 66.3% Political Attributes Party Membership 1. Communist 83.5% 85.4% 80.6% 2. Other .4% .4% .7% 3. Unknown 16.1% 14.2% 18.7% Average Age When Joining the Communist 26.4 25.7 25.3 Party (CPC) (N=51) (N=66) (N=77) Member of National Congress, National 8.1% 11.7% 19.4% Consultative Assembly, or CPC National Congressa Career Attributes Average Age When Becoming CEO 46.4 46.3 46.8 Main Starting Position Title ("Technician") 36.1% 33.5% 27.8% (N=158) (N=170) (N=169) Career Pathways 1. Single-Group Track 52.9% 56.2% 52% 2. Multi-Group Track 23.5% 22.6% 24.2% 3. Supervisory-Bureau Track 15.8% 9.1% 12.1% 4. Unrelated-Government Unit Track 3.7% 2.9% 2.2% 5. Multi-Sphere Track 4% 9.1% 9.2% 6. System-Outsider Track 0% 0% .4% Tenure 8.8 8.1 NA Status After CEO Positionb NAe 1. Chairman/Director/Executive of the 27.8% 23.4% Group 2. Senior Government Official (e.g., 9.2% 4.7% Minister, Governor, Committee Member) 3. Director/CEO/Senior Manager of 5.9% 6.6% other Groups 4. Convicted Prisoner 4% 2.6% 5. Still CEO of the Group 15.8% 39.1% 6. Inactive Retiree/Decedent 13.6% 6.6% 7. Unknown 23.9% 17.2% N (Default, if not specified otherwise) 272 274 273 a The membership is counted as whether the CEO was a member of the political bodies of that year. b The post-CEO statuses were tracked until October 2011. 0 Because many of the CEOs just took office and 88.6% of the 2010 CEOs are still in power, this measure is not properly applicable. The data show that the professionalism reforms have noticeably affected the educational dimension. The No. 3:743] STATE OWNERSHIP AND CORPORATE GOVERNANCE IN CHINA 771 percentage of the CEOs with a graduate degree significantly increased from 37.5% in 2002 to 57.1% in 2010. The percentage of the CEOs with a foreign degree also rose, albeit only slightly, to 4.4% in 2010. Those with foreign study experience usually received their degrees from academic institutions in the United States. Regarding academic discipline, more than sixty percent of the CEOs were still trained in engineering, representing a slow diversification in specialization areas. The dominance of engineering-trained CEOs is a result of China's industrial structure and political history. The Chinese government has aggressively sought personnel with technological expertise, but this knowledge cannot be achieved on the job, unlike financial or marketing experience. Moreover, when China's education system was battered by the storms of Maoism, engineering was a politically safer field of study than most. These factors increased the tendency of students to study engineering. As for elite school education, while the number of CEO graduates from the C9 League (the so-called "Ivy League of China")6 2 declined to 11.4% in 2010 from 15.8% in 2002, the Tsinghua alumni network expanded to 5.5% in 2010. Thus, while the elite school network has shrunk overall, it has simultaneously become more concentrated in Tsinghua, the leading engineering school in China. Regarding political membership, the data show that in 2010, more than eighty percent of CEOs were members of the CCP. This number is significantly higher than the overall employee party membership rate (around thirty 62 C9 League, formed in 1998, is an alliance of nine prestigious universities handpicked by the Chinese government, including Fudan University, Harbin Institute of Technology, Nanjing University, Peking University, Shanghai Jiao Tong University, Tsinghua University, University of Science and Technology of China, Xi'an Jiao Tong University, and Zhejiang University. Michael Sainsbury, China Establishes Group of Ivy League Universities, THE AUSTRALIAN (Nov. 4, 2009, 12:00 AM), http://www.theaustralian.com.aulnews/world/china- establishes-group-of-ivy-league-universities/story-e6frg6so- 1225794050411. COLUMBIA BUSINESS LAW REVIEW percent) throughout the SOE corporate hierarchy.'3 While it may seem that CCP membership has declined, this is an unsupported interpretation, given the increase in unknown/missing data.64 The CCP-affiliated executives joined the Party at a relatively early age (around twenty-five or twenty-six years old). This suggests that an early pledge of political loyalty boosts career development in the state system. Early party membership signals a higher degree of political commitment and allows for a prolonged process of scrutiny, cultivation, and training, which together increase the likelihood of subsequent promotion into elite positions.6 5 While most CEOs are CCP members, a minimal number of the remaining CEOs are affiliated with the China National Democratic Construction Association, which is an ally of the CCP.66 The marginal representation of non-CCP party members in the SOE elite adds merely symbolic, rather than political, diversity to the SOE system. In addition to the predominance of CCP affiliation, the integration of executive personnel into China's political system seems to be on the rise. The Chinese government has an institutionalized practice of selecting top managers for its representative national political bodies, including the National People's Congress (the government's symbolic legislative body), the National People's Political Consultative Conference (an advisory body composed of representatives of different social and political groups), and the National 63 SASAC, CHINA'S STATE-OWNED ASSETS SUPERVISION AND ADMINISTRATION YEARBOOK (2010) (discussing how approximately one- third of the employees in the national SOEs are members of the Party; as of the end of 2009, 3.03 million of the 9.36 million employees of the central SOEs were party members). 64 When a CEO's party affiliation cannot be clearly identified in the multiple information sources, the CEO's political affiliation is coded as unknown. 65 Bobai Li & Andrew G. Walder, Career Advancement as Party Patronage: Sponsored Mobility into the Chinese Administrative Elite, 1949-1996, 106 AM. J. Soc. 1371, 1395 (2001). 66 USC US-CHINA INSTITUTE, WHITE PAPER ON CHINA'S POLITICAL PARTY SYSTEM, 2007 (2007), available at http://www.china.usc.edu/ShowArticle.aspx?articleID=855. 772 [Vol. 2013 No.3:743] STATE OWNERSHIP AND CORPORATE GOVERNANCE IN CHINA 773 Congress of the Chinese Communist Party (the Party's general assembly). While members usually lack substantive power, membership in such political bodies represents a social and political status or a mark of legitimacy formally recognized by the party-state. The data show that the number of CEOs who were members of these national political bodies increased from 8.1% in 2002 to 19.4% in 2010. The increased representation suggests the growing importance of SOEs in China's national political system, but it also deepens the concern about the SOE management's autonomy. Overall, the political attributes here confirm that political loyalty to the CCP remains a paramount requirement because SOE executive personnel have become more integrated into the national political bodies. The rising political integration diverges from the principle of separating SOEs from the government that many Chinese corporate governance reform laws have purportedly declared." Turning to career attributes, the CEOs on average came to power when they were around forty-six years old- relatively young compared to executives in other corporate governance regimes.68 This junior CEO phenomenon confirms the party-state's executive professionalization rules as discussed in Part III, as young executives are typically more enthusiastic and less constrained by old traditions. "Technician" has been the most common initial job title in the CEOs' careers, although it has declined from 35% in 2002 to 27.2% in 2010 due to added diversity in job titles. This observation is closely related to their educational training, mainly in the engineering discipline. With respect to career pathways, the single-group track has consistently been the mainstream path. More than half of the CEOs came to power by climbing the corporate ladder. The stable majority of single-group track CEOs suggests that firm-specific knowledge is valued in China's SOE 67 See Qiye Guoyou Zichan Fa [Law of the People's Republic of China on the State-Owned Assets of Enterprises] (promulgated by the Standing Comm. Nat'l People's Cong., Oct. 28, 2008, effective May 1, 2009) 2008 STANDING CoMM. NAT'L PEOPLE'S CONG. GAz. art. 6. 68 See infra Part V.B. management and insider connections are helpful to secure CEO positions. The second-most common career pathway is the multi-group track, accounting for more than twenty percent. In sum, most of the CEOs developed their careers completely within the SOE system rather than moving around in different government spheres. It suggests that the party-state maintains a certain degree of separation in human resources training between the political and business spheres within the state system. The supervisory-bureau track is the next most popular pathway, followed by the multi-sphere track and the unrelated-government-units track. There exists only a marginal-and declining-percentage of CEOs coming to power through the unrelated-government-units track. This suggests that the party-state does not favor executive candidates who are unfamiliar with SOEs' business matters, a notion that aligns consistently with its professionalization reform rules. None of the CEOs in 2002 or 2005 followed the system- outsider track. A sign of change emerged in 2010, however, when one CEO came to power through the system-outsider track: Dazong Wang, the CEO of Beijing Automotive Industry Corporation, an SOE under Beijing SASAC's supervision. After completing his Ph.D. at Cornell University, Wang joined General Motors ("GM") and swiftly worked his way up to a top-level position of engineering design at GM. After working at GM for twenty-one years, Wang became vice president of SAIC Motor Corp., a publicly traded subsidiary of Shanghai Automotive Industry Group, which is an SOE owned by the Shanghai SASAC. In 2008, after less than two years in office at SAIC Motor Corp., Wang parachuted into the CEO position at Beijing Automotive Industry Corporation. Wang's narrative has several noteworthy implications. First, this case occurred in the automobile industry, which is not monopolized by the state and remains relatively open to the private sector and foreign companies. Consistent with the assumption in Part IV, competitive pressure in a non-monopolized industry can drive SOEs to recruit system-outsiders to improve 774 COL UMBIA B USINESS LA W RE VIE W [Vol. 2013 No. 3:743] STATE OWNERSHIP AND CORPORATE GOVERNANCE IN CHINA 775 innovation capacity. Second, it is a local, rather than central, SOE that took the initiative to embrace a system- outsider. A partial explanation is that central SOEs usually bear national security importance and are thus less open to system-outsiders. It also reflects a typical institutional reform pattern in China: experiments starting from the local level.69 Third, the details of this case provide insight into what incentivizes a system-outsider to join SOEs and the paths they can travel to arrive there. As Wang explained, personal sentiment, family education, and patriotism played a significant role in his decision to join the SOE system. 70 His aspirations were realized with the help of external opportunities. As GM sought rapid expansion in China, Wang seized the opportunity to transfer to a GM joint venture in China and worked there for 3 years until 1997. During his work in China, he became acquainted with Maoyuan Hu, the then-vice president of Shanghai Automotive Industry Group, and the two thereafter maintained a relationship. This social connection paved Wang's way into the Chinese SOE system. His experience indicates that system-insiders' endorsements are very helpful to system-outsiders' entry into the state system, which supports the common finding that personal 69 See ANN M. FLORINI ET AL., CHINA EXPERIMENTS: FROM LOCAL INNOVATION To NATIONAL REFORM 45-46 (2012). 70 Interview by Qicheren with Dazong Wang, CEO, Beijing Auto. Indus. Corp. (Nov. 2008) ("A man should not forget his own roots. A man who does not know his own roots is a pathetic one. . . . I am a traditional Chinese. . . . From a personal perspective, I am really grateful to my father. My life philosophies are all inherited from my father. I seldom see any person who is as patriotic as my father. I always remember my father said repetitively at his death bed that we should learn in the United State and apply in China. . . . My father studied in Japan. Upon returning to China, he established the first ceramics high school and the first ceramics factory for home appliances in China, becoming the founder of China's contemporary ceramics industry. The old intellectual's scientific patriotism is deeply passed down to his next generation. What leads me is such philosophy. Like my father said, I can contribute by applying in China and simultaneously broaden my career horizon."). 71 Id. COLUMBIA BUSINESS LAW REVIEW connections (guanxi) play an important role in the Chinese government system.72 The average CEO tenure is about eight years. Since the CEOs on average came to power at around forty-six years old, by the end of their tenure they are on average only in their mid-fifties-an age still capable of active work. This raises questions about where they go after their time as CEO. Table 1 shows that about a quarter of CEOs in 2002 and 2005 followed the so-called "apprentice model," in which the CEOs were promoted to chairman, director, or other executive positions of the same business group. The apprentice model allows the CEOs to continue contributing their firm-specific knowledge and help guide their successors before transitioning to retirement. After serving as CEO, it is also common to transfer to other government units, usually as a minister, governor, or government committee member. This post-CEO transition, however, seems to be on the decline, down from 9.2% in 2002 to 4.7% in 2005. Close to 6% of the CEOs took up senior manager positions in other business groups, which suggests the influence of the government's personnel rotation practice in post-CEO careers. Although SASAC's executive recruitment reform rules repeatedly emphasize executives' moral integrity, 73 post- CEO status shows that some CEOs ended up in jail on charges related to their executive responsibilities. The data show that 4% of the CEOs in 2002 and 2.6% in 2005 faced criminal liability connected to their executive duties. The typical criminal charges were corruption and embezzlement. The decline in criminal convictions seems to suggest some 72 See Walder, supra note 24, at 309; Yanjie Bian, Getting a Job Through a Web of Guanxi, in NETWORKS IN THE GLOBAL VILLAGE: LIFE IN CONTEMPORARY COMMUNITIES 255 (1999). 73 Provisional Rules on Corporate Leaders of Central SOEs (promulgated by St. Council, effective 2009) art. 1; Provisional Measures on Comprehensive Evaluation of Corporate Leadership Teams and Leaders of Central SOEs (promulgated by the St. Council, effective 2009); Regulations on Honest and Ethical Business Practice Behavior of SOE Corporate Leaders (2009). 776 [Vol. 2013 No. 3:743] STATE OWNERSHIP AND CORPORATE GOVERNANCE IN CHINA 777 improvement in moral integrity, but this is inconclusive since many factors-most notably politics-could affect the probability of prosecution in China. While a number of CEOs were found guilty of corruption or embezzlement under Chinese criminal law, none of the CEOs in the dataset ever became defendants in litigation involving the breach of fiduciary duties under Chinese corporate law. The Chinese government, as controlling shareholder, seems to govern the top managers in a way similar to government bureaucrats, who are subject to liabilities under public law such as criminal law. As of the end of data tracking, 15.8% of the CEOs from 2002 and 39.1% from 2005 were still in office. At least 13.6% in 2002 and 6.2% in 2005 were inactive retirees or deceased. In addition, 23.9% of CEOs in 2002 and 17.2% in 2005 had an unknown post-CEO status. This is due to the fact that biographical information tends to become obscure when CEOs leave office, especially for those not moving to prominent positions. In summary, executive educational quality shows some improvement, but the political and career attributes generally have remained stagnant under the professionalism and marketization reforms. While the marketization reform in particular relaxes political affiliation and welcomes career diversity, the empirical evidence suggests that openness in the executive labor market is confined to the laws on the books rather than the laws in action. The executive labor market remains a game for system-insiders who have the right political and social connections. B. Executive Attributes in Comparative Perspective Scholars of comparative corporate governance recognize that corporate governance regimes may differ in their legal arrangements as well as the composition of the local elite.7 4 74 See generally MAIRI MACLEAN, CHARLES HARVEY & JON PRESS, BuSINESs ELITES AND CORPORATE GOVERNANCE IN FRANCE AND THE UK (2006) (examining how social and cultural norms in different countries explain how individuals ascend to the business elite). COLUMBIA BUSINESS LAW REVIEW Different legal rules can produce different corporate elites, and different corporate elites can generate different rules. Thus, this raises questions about how the CEOs of China's large SOEs differ from corporate executives in other countries. Based on the descriptive findings in Table 1, this Article compares Chinese SOE CEOs with executives of the archetype countries in comparative capitalism and comparative corporate governance literatures, including France (which promotes state activism); Japan and Germany (which practice the coordinated market and stake-holder oriented models, respectively); and the United Kingdom and United States (which utilize the liberal market and shareholder-oriented models, respectively). A general summary of the executive attributives in comparative perspective is reported in Table 2. 75 See VARIETIES OF CAPITALISM: THE INSTITUTIONAL FOUNDATIONS OF COMPARATIVE ADVANTAGE (Peter A. Hall & David W. Soskice eds., 2001); Vivien A. Schmidt, French Capitalism Transformed, Yet Still a Third Variety of Capitalism, 32 EcON. & Soc'Y 526, 527 (2003). 778 [Vol. 2013 No. 3:743] STATE OWNERSHIP AND CORPORATE GOVERNANCE IN CHINA 779 TABLE 2: EXECUTIVE ATTRIBUTES IN COMPARATIVE PERSPECTIVE State Act1vismn Coordinated Market / Liberal Market / Stakeholder Oriented Shareholder-Oriented China (SOEs France Japan Germany UK USA only) Educational Attributes Dominant Engineering or Engineering or Engineeri Engineering Business. Business- Educational natural science natural science ng or or natural Related Related Discipline (more than 60% (more than 30% natural science with an as of 2005), science (more than increasing trend) followed by (more than 50% as of business, 30% as of 2005) economics or 2011); low administration economics Elite School C9 League, Grandes Scole, Tokyo, No specific Cambridge Ivy Concentration particularly particularly Kyoto, concentration and Oxford League Tainghua Polytechnique Waseda, and a University and Aoole and Keio number of Nationale elite d Administration schools (about 13 in total), with Harvard University as the lead Degree of Elite Low High High Low Moderate. Moderate School (about 11% as of (about 75% as of (about High (more Concentration 2010) 2005) 60% as of (about 40% as than 20% 2009) of 2007) as of 2011) Career Attributes Prevailing Single-group Lengthy careers Single. Single-group Multi- Multi. Career pathway within in state service group pathway employer employer Pathway the state system before transfer pathway (more than pathway pathway, (more than 50%) to business 55% as of less than 2005) one-third lifetime employees Work When broadly Moderate-High Low- Low Low Low Experience in defined (SOEs as (more than 50% Moderate Government part of in 1998 down to government): 38% in 2005) Very High (more than 99%) When narrowly defined (only government units other than SOEs): Moderate (about 20%) Approximate Mid-Late 40s Early 50s Mid-Late Early 50s Early 50s Early 50s Age of 50s Appointment Average About 8 years, About 8 years About 6 About 8 years, About 6-7 About 8 Tenure with a declining (s of 2011) years in with a years (as of years, trend 2007, declining 2011) down down from down from trend down from 9.6 in 10 years 8 years in from 12 years 1995 in 2010 1995 in 1980 Apprenticeship Low-Moderate Lowt High Lowt Lowt Moderate- Model in (About 25%) High (43% Training CEOs as of 2011) For data on France, see MAIRI MACLEAN, CHARLES HARVEY & JON PRESS, BUSINESS ELITE AND CORPORATE GOVERNANCE IN FRANCE AND THE UK (2006); Taeyoung Yoo and Soo Hee Lee, In Search of Social Capital in State Activist Capitalism: Elite Netwvorks in France and Korea, 30 ORG. STUD. 529 (2009); Vivien A. Schmidt, A Profile of the French CEO, 35 INT'L EXEC. 413 (1993); Booz & Company, CEO SUCCESSION REPORT: 12TH ANNUAL GLOBAL CEO SUCCESSION STUDY (2012), available at http://www.booz.commedialfile/BoozCo.CEO- Succession-Study-201 1_Extended-Study-Report.pdf. For data on Japan, see Booz & Company, CEO SUCCESSION REPORT: 12TH ANNUAL GLOBAL CEO SUCCESSION STUDY (2012); Ken Favaro, Per-Ola Karlsson & Gary L. Neilson, CEO SUCCESSION 2000-2009: A DECADE OF CONVERGENCE AND COMPRESSION (2010), available at http://www.booz.com/global/home/what-we- think/reports-white-papers/article-display/succession-2000-2009-decade-convergence; Booz & Company, CEO SUCCESSION 2008: STABILITY IN THE STORM (2009), available at http://www.strategy- business.com/article/09206?pg-all; Booz & Company, CEO SUCCESSION 2003: THE PERILS OF "GOOD" GOVERNANCE (2004), availableat www.boozallen.com/medialfile/139656.pdf. For data on Germany, see PAUL WINDOLF, CORPORATE NETWORKS IN EUROPE AND THE UNITED STATES (2002); Booz & Company, CEO SUCCESSION REPORT: 12TH ANNUAL GLOBAL CEO SUCCESSION STUDY (2012); Saskia Freye, Germany's New Top Managers? The Corporate Elite in Flux, 1960-2005 (Max Planck Institute for the Study of Societies, Discussion Paper 10/10, 2010), available at www.mpifg.de/pulmpifg-dp/dp10-10.pdf. For data on United States, see Peter Cappelli & Monika Hamori, The New Road to the Top, 83 HARV. BUS. REV. 25 (2005); Booz & Company, CEO SUCCESSION REPORT: 12TH ANNUAL GLOBAL CEO SUCCESSION STUDY (2012); Jason D. Schloetzer & Matteo Tonello, The 2011 CEO Succession Report (The Conference Bd. Research Report No. R-1478-11-RR, 2011), available at http://papers.ssrn.com/sol3/papers.cfm?abstract id= 2032227; Jeffrey S. Sanders, The Path To Becoming a Fortune 500 CEO, FORBES (Dec. 5, 2011, 12:26 AM), http://www.forbes.com/sites/ciocentral/2011/12/05/the-path-to-becoming-a-fortune-500-ceo/; Richard S. Tedlow, Courtney Purrington & Kim Eric Bettcher, The American CEO in the Twentieth Century: Demography and Career Path (Harvard Bus. Sch., Working Paper No. 03-097, 2003), available at http://papers.ssrn.com/sol3/ papers.cfm?abstractid=383280. t Booz & Company (2012) reports at the overall European level, rather than on specific countries. Table 2 shows that CEOs of the state activism and the stakeholder-oriented models are commonly trained in engineering, while those of the shareholder-oriented model tend to be trained in business-related disciplines. The dominance of engineering-trained CEOs is most salient in China and Germany. With respect to the alumni network, China and Germany are both on the low end of elite school concentration, while France and Japan are on the high end, with the United States and the United Kingdom in between. As to career pathways, the prevailing career route in China, Japan, and Germany is climbing the corporate ladder to the top within a single-business group, while the popular career track in the United Kingdom and the United States is to follow an external labor market strategy. In France, a significant number of CEOs started their careers in the public sector before transferring to the corporate sector, which means that government work experience is quite common in France. In Japan, while many executives follow the so-called amakudari career pathway, most of them do not assume CEO positions. Chinese SOE CEOs seem closest to French CEOs in terms of their career connections with the government. As to the average age of appointment, Chinese SOE CEOs come to power at a younger age (around forty-six years old) than CEOs in the other countries (usually in their fifties). Nevertheless, there is no sharp difference in terms of tenure, which is currently in the range of six to eight years with a trend of shortening in length. Finally, unlike Japan and the 780 COL UMBIA B USINESS LA W RE VIE W [Vol. 2013 No.3:743] STATE OWNERSHIP AND CORPORATE GOVERNANCE IN CHINA 781 United States, where outgoing CEOs are commonly appointed as the chairman of the same company, China uses this apprentice model less frequently when training and guiding CEOs, much as its European counterparts. The moderate use of the apprentice model in China may be partially due to the fact that many Chinese SOEs have not yet established boards of directors. Overall, Table 2 shows that Chinese SOE CEOs bear a resemblance to executives in countries of the stakeholder- oriented and the state activism models, although different historical or political forces probably underlie this resemblance. The comparison highlights the similarities shared by Chinese and German CEOs, while noting the many distinctions between Chinese CEOs and their British and American counterparts. This comparison clearly places China far outside the category of the liberal market economy or the shareholder-oriented model. While the Chinese SOE executives present many characteristics dissimilar to Western executives, more refined data analysis suggests some convergence in terms of the nature of career pathways. Scholars find that, in Europe and the United States, CEOs who have spent the bulk of their careers climbing the ranks within a single company spent significantly less time climbing to the top compared to those who moved across multiple employers.76 With ordinary least squared regression, Table 3 finds that Chinese executive career pathways show signs of converging toward these Western findings." 76 Monika Hamori and Maria Kakarika, External Labor Market Strategy and Career Success: CEO Careers in Europe and the United States, 48 HUMAN RES. MGMT. 355 (2009). 77 See infra Appendix 1. COLUMBIA BUSINESS LAW REVIEW TABLE 3: OLS REGRESSION ANALYSIS OF EXECUTIVE CAREER PATHWAYS AND AGE OF ATTAINMENT, 2002-2010 Dependent Variable: Age When Becoming the CEO 2002 2005 2010 (1) (2) (3) (4) (5) (6) Career Pathways 1. Single-Group Track 2. Multi-Group Track 3. Supervisory-Bureau Track 4. Unrelated-Government- Units Track 5. Multi-Sphere Track 6. System-Outsider Track- Educational Backgrounds C9 Lengue Engineering Graduate Degree Foreign Study Organizational Controls Board of Directors (Yes-1) Ownership (Central Gov.=1) Log Revenues Log Employees Return on Assets Constant Observations R._nrd ----------------------- Omitted as Reference Category. - --.. --- 1.316 .950 1.406 1.425 4.143*** (.999) (.944) (.932) (.925) (.776) 3.903** 3.549** 3.766' 3.029* 4 .4 80 *** (1.123) (1.046) (1.443) (1.209) (1.127) 2.487 1 .630 1.999 1.854 5.191** (1.071) (1.465) (2.149) (1.571) (1.804) 2.742 1.462 2.264* 2.109t 3.313* (1.508) (1.759) (1.104) (1.103) (1.406) 11.105*** (1.314) -.567 (1.020) 3.087*** (.823) -2 .83 6 * (.788) 4.881t (2.605) -1.109 (1.030) 1 .249 (.832) -3.300*** (.727) 3.322 (2.393) .439 (1.017) 1.434 (.829) -.711 (.749) .913 (1.711) -.185 (1.127) .248 (.826) .634 (.730) -. 140 (1.178) 1.270 -1.264 (1.017) (.949) 2.973* 1.495 (1.222) (1.097) .499 .292 (.443) (.522) 1.201*** .892* (.328) (.399) -.115 -.143 (.117) (.103) 44.367*** 2 6 . 2 1 1 *** 44.693*** 33.311*** (.910) (4.729) (.939) (6.368) 272 263 274 272 .122 .284 .049 .157 -.532 (.846) ,564 (.790) -1.024 (.708) -1.839 (1.350) 45.193*** (.908) 273 .145 The table presents unstandardized coefficients with robust standard eerors in parentheses. Significance Level:t p<.l * p <.05; **p <.01; *** p <.001. The lower the significance level, the stronger the evidence. * The system-outsider category is omitted for the years of 2002 and 2005 because there is no observation falling into this category. Table 3 shows that the nature of career pathways has changed in terms of age when reaching the CEO position. In 2002, the CEOs coming to power via the supervisory-bureau pathway were significantly older at the time of appointment than the CEOs through the single-group pathway." The CEOs via the multi-group, unrelated-government-units, and multi-sphere pathways were older than their single-group colleagues, as well (although this difference is not 782 [Vol. 2013 3.679*** (.774) 4.443*** (1.070) 4.620** (1.635) 2.691* (1.177) 11.465* (1.374) .1.363 (.872) -.167 (.741) -.418 (.684 -2.364f (1.370) -1.616 (.375) 1.049 (.964) .502 (.505) .607 (.375) -.147 (.124) 33.139* (6.046) 270 .250 78 See Table 3 (finding that the supervisory-bureau track coefficient is 3.903 (p< .01) in Model 1, and 3.549 (p<.01) in Model 2). No. 3:743] STATE OWNERSHIP AND CORPORATE GOVERNANCE IN CHINA 783 statistically significant)." Similarly, in 2005, the single- group career CEOs reached the executive position at a younger age than the CEOs of all other career pathways, but the results for the multi-group and unrelated-government- units pathways are not statistically significant.so These trends became clearer in 2010. The CEOs coming to power through the single-group pathway were significantly younger than the CEOs of all other types of career pathways."' Among all of the types, the system-outsiders were the oldest when ascending to power.82 These results support the hypothesis that system-outsiders need longer to establish the requisite external reputation and high professional status, and therefore must usually be older to overcome the liability of foreignness when entering the state system. Finally, the CEOs of the supervisory-bureau track were consistently older when they landed in the executive position, which can be explained by the retirement-transition strategy as discussed in Part IV.8 In short, the career pathway results suggest that firm-specific knowledge and firm-specific social networks are helpful assets when securing executive positions, similar to the findings about Western executives. Table 3 shows that elite school attendance was not a significant factor in ascending to the CEO position. 84 79 See Table 3 (finding that in Model 2, the multi-group track coefficient is .950, the unrelated-government unit track coefficient is 1.630, and the multi-sphere track coefficient is 1.462). 80 See id. (finding, that in Model 3, the supervisory-bureau track coefficient is 3.766 (p<.05) and the multi-sphere track coefficient is 2.264 (p<.05). In Model 4, the supervisory-bureau track coefficient is 3.029 (p<.05) and the multi-sphere track coefficient is 2.109 (p<.10, marginally significant)). 81 See id. (finding that all of the career pathway coefficients in Models 5 and 6 are positive and statistically significant). 82 See id. (using a system-outsider track coefficient of 11.105 (p<.001) in Model 5 and a system-outsider track coefficient of 11.465 (p<001) in Model 6). 83 1See id. (using both positive and statistically significant supervisory- bureau track coefficients in Models 1-6). 84 See id. (using all statistically insignificant C9 League coefficients in Models 1-6). COLUMBIA BUSINESS LAW REVIEW Neither was an engineering degree a significant factor when controlling for organizational factors.85 The 2002 data suggest that CEOs with graduate degrees reached their positions at a younger age, but this advantage was weaker in 2005 and 2010.86 In part, the value of graduate degrees has diminished as more and more executive candidates have obtained such degrees. This could also be because pursuing a graduate degree postponed the entry into the workforce or hampered their work performance and thus delayed promotions. For organizational variables, while recruiting younger executives and establishing boards of directors are both important SOE reform measures, Table 3 shows no significant association between the appointment of younger CEOs and the existence of boards of directors.8 8 Moreover, Table 3 indicates that executives of SOEs owned by the central government tend to reach the CEO position at an older age compared to those of SOEs owned by the local governments, though the result is significant only for 2002. An institutional explanation for this finding is that it requires more time to ascend to executive positions at central SOEs, which are high-status positions. Table 3 also shows that CEOs of the SOEs with larger numbers of employees tend to be older at the time of appointment, which may be related to higher competition in the internal labor market.90 85 See Table 3 (using insignificant engineering coefficients in Models 2, 4, and 6). 86 See id. (using negative and statistically significant graduate coefficients in Models 1-2 and insignificant graduate coefficients in Models 3-6). 87 Rather than completing all education credentials before entering the workforce, many of the CEOs pursued their graduate degrees while they were at work. 88 See Table 3 (using insignificant board of directors coefficients in Models 2, 4, and 6). 89 See id. (using an ownership coefficient of 2.973 (p<.05) in Model 2, 1.495 in Model 4, and 1.049 in Model 6). 90 See id. (using a (log) number of employees coefficient of 1.201 (p<.001) in Model 2, .892 (p<.05) in Model 4 and .607 in Model 6). 784 [Vol. 2013 No. 3:743] STATE OWNERSHIP AND CORPORATE GOVERNANCE IN CHINA 785 C. Post-CEO Status and Governance Using logistic regression models, Table 4 shows results regarding post-CEO development (discussed in Part IV.B).91 TABLE 4: LOGISTIC REGRESSION ANALYSIS OF POST-CEO STATUS, 2002-2010 Post-CEO Career Status Chairman/DirctorEx Director/Executive of Senior Government Convicted Prison ecutive of the Same Other Groups Official Group (1) (2) (3) (4) (5) (6) (7) Carer Pathways 1. Single-Group -.689 -.379 -.575 -.644 er 8) Track REF (.620) (.663) (.603) (.660) REF-- --- 2. Multi-Group 439 -.350 -.500 -.365 .566 1.302 Track (.340) (.364) REF (.645) (.690) (.678) (.778) 3. Supervisory. -.808 -.628 .762 .869 .104 .824 Bureau Track (.483) (.510) (.739) (.768) REF (.907) (.982) 4. Unrelated- -1.364 -1.455 1,919 2.623" .424 .610 -14.721 .17.021 Government- (.754) (.799) (.805) (.917) (.870) (1.025) (.638) (4909.2) Units Track 5. Multi-Sphere -.398 -.535 1.250 1.326 -.599 .872 .104 2.166 Track (.574) (.621) (.784) (.848) (.936) (1.047) (.920) (1.293) 6. System. Outsider Track- Educational Backgrounds C9 League 1.095* 1.210* .173 .190 .1.311 -1.018 -.827 .16.387 (.436) (.480) (.633) (.654) (.769) (.808) (1.260) (2647.8) Engineering -.016 -. 167 110 .233 .665 .421 -1.548 -2.170" (.314) (.343) (517) (.564) (.433) (.480) (.574) (.746) Graduate Degree .699* .618 -.180 -. 125 .071 .042 -.995 -1.419 (.302) (.313) (.520) (.537) (.384) (.408) (.677) (.803) Foreign Study -1.1291 -.899 3.088** 2.659" -.944 -.445 -13.847** -14.717 (.674) (.752) (.766) (.826) (1.094) (1.140) (.729) (5246.9) Organizational Variables Board ofDirectors 1 .2 3 4 " -. 796 -1.032* .739 (Yes=l) (.398) (.593) (.510) (.880) Ownership (Central .580 .155 -1.398* .807 Go-1) (.447) (.688) (.652) (1.113) Log Revenues -.011 .150 .185 -.430 (.155) (.242) (.214) (.361) Log Employees .166 5.183 .447* .578 (.154) (.200) (.217) (.335) Return on Assets .060 -.075 -.100 .059 (.042) (.078) (.074) (.069) Life course control Age at Appointment -.055* -.071* -.070 -.087* -.074* -.097* .046 .070 (.025) (.028) (.040) (.044) (.032) (.038) (.047) (.060) Constant 2.395 .468 ..028 1.196 1.304 -3.472" -3.912 .5.541 (1.230) (1.995) (1.558) (3.142) (1.467) (2.683) (2.182) (4.526) Observations 232 229 232 229 232 229 232 229 x2 (df) 30.22 43.32 31.37 33.56 12.80 27.05 15.12 26.58 (9) (14) (9) (14) (9) (14) (9) (14) p-value .000 .000 .000 .002 .186 .019 .088 .022 The table presents unstandardized coefficients with standard errors in parentheses. Significance Level: t p<.1; * p <.05; " p <.01; ** p <.001, The lower the significance level, the stronger the evidence. a The system-outsider category is omitted because there is no observation falling into this category. 91 See infra Appendix 2 (explaining the models in more detail). COLUMBIA BUSINESS LAW REVIEW The career pathway variable shows that there is no clear path-dependent effect on post-CEO career development. The CEOs of the single-group track are not significantly more likely to stay in the same group. 92 Meanwhile, the data show that the CEOs of the multi-group track are not more likely to transfer to other groups. 93 Similarly, the CEOs who had work experience in other government units are not significantly more likely to return to government bureaus. 9 4 The lack of clear path-dependent effects underscores the complexity and unpredictability of the SOE elite training system. Interestingly, the data show that CEOs of the unrelated- government-units track are least likely to be subject to criminal charges at the end of their tenure (though the results are not statistically significant for other career paths)." This finding can be interpreted in two ways. On one hand, it might relieve a common concern that CEOs of the unrelated-government-units track may be more likely to mismanage corporate governance and incur criminal liability for corruption due to their political backgrounds and lack of business experience. On the other hand, this finding could suggest that these types of CEOs may simply be more politically savvy in managing their careers in the state system and may have more political resources to escape legal liabilities, which is not a good sign for corporate governance. Table 4 also shows that the CEOs who graduated from elite schools are more likely to stay in the same group96 and less likely to end up in prison." Engineering-trained CEOs 92 See Table 4 (using statistically insignificant career pathway coefficients in Models 1-2). 93 See id. (using all of the career pathway coefficients in Models 3-4). 94 See id. (showing that all of the career pathway coefficients are statistically insignificant in Models 5 and 6). 95 See id. (finding that the unrelated-government units track coefficient is -14.721 in Model 7 and -17.021 in Model 8). 96 See id. (finding that the C9 League coefficient is 1.095 (p<.0 5) in Model 1 and 1.201 (p<.05) in Model 2). 97 See id. (finding that the C9 League coefficient is -16.387 in Model 8). 786 [Vol. 2013 No. 3:743] STATE OWNERSHIP AND CORPORATE GOVERNANCE IN CHINA 787 are also less likely to be criminally convicted after serving in the CEO position." CEOs with a graduate degree are also less likely to be imprisoned (though the result is not statistically significant)." Foreign-educated CEOs are more likely to transfer to positions in other business groups 00 and less likely to be held for criminal liabilities.10 1 In general, Table 4 shows that better education is negatively associated with criminal liability at the end of the CEO tenure. Thus, executives' improving educational credentials may be a good sign for the quality of corporate leadership. An important question in corporate governance is whether the board of directors plays a role in executive career development and behavior. Table 4 shows that CEOs of SOEs with a board of directors are more likely to assume the chairman, director, or other executive positions in the same business group once their tenure ends. 102 Moreover, the presence of a board of directors reduces the likelihood of a CEO becoming a senior official in government bureaus, which suggests the board might play a role in cutting political connections. 103 However, there is no evidence that a board of directors can effectively reduce the likelihood of executive criminal behavior.10 This calls into question whether boards of directors, even when extant, monitor executives proficiently. Finally, Table 4 shows that CEOs of central SOEs are less likely to transfer to government 98 See Table 4 (finding that the engineering coefficient is -2.170 (p<.01) in Model 8). 99 See id. (finding that the graduate degree coefficient is -.995 in Model 7 and -1.419 in Model 8). 100 See id. (finding that the foreign study coefficient is 3.088 (p<.001) in Model 3 and 2.659 (p<.01) in Model 4). 101 See id. (finding that the foreign study coefficient is -13.847 (p<.001) in Model 7 and -14.717 in Model 8). 102 See id. (finding that in Model 2, the board of directors coefficient is 1.234 (p<.01)). 103 See id. (finding that in Model 6, the board of directors coefficient is -1.032 (p<.05)). 104 See id. (finding that in Model 8, the board of directors coefficient is .739). COLUMBIA BUSINESS LAW REVIEW bureaus than are CEOs of local SOEs.as As such, the post- CEO personnel integration between SOEs and other government units seems higher at the local than at the central level. D. Implications, Challenges, and Future Research This Article analyzes the executive composition and recruitment evolution of China's large non-financial SOEs. It shifts away from the typical focus on how things operate to which people operate those things, adopting a novel approach to understanding corporate governance and economic development in countries with weak legal institutions. This Article shows some improvement in educational credentials and general stability in political affiliation and career patterns. The executive recruitment reform of China's large SOEs is oriented toward politically-bounded and firm- specific knowledge professionalism, as well as potential bottom-up and competition-driven marketization. It also shows similarities and dissimilarities between Chinese SOE executives and their foreign counterparts. Moreover, this Article shows the complexity of the Chinese elite training system, where there is no clear path-dependent effect on post-CEO status and there is no conclusively negative association between political career experience and executive criminal liabilities. Finally, this Article finds little evidence that boards of directors of China's SOEs play any significant role in affecting executive career development or avoiding serious executive criminal behavior. Based on these findings, this Article discusses the adequacy of existing international legal rules and enforcement concerning SOE governance, the challenges ahead for SOE governance, and some directions for future research. 105 See Table 4 (finding that in Model 6, the central ownership coefficient is -1.398 (p<.05)). 788 [Vol. 2013 No.3:743] STATE OWNERSHIP AND CORPORATE GOVERNANCE IN CHINA 789 1. Implications for International Legal Systems Chinese SOEs are not only important actors in China's domestic economy, but also active players in the global investment market. This global expansion raises immediate questions for international regulators about the adequacy and capacity of existing legal rules and enforcement to cope with such unconventional corporate entities. This section demonstrates why an investigation of Chinese SOE executive backgrounds is helpful in reexamining existing legal regimes governing the SOEs' international activity. One global expansion strategy of Chinese SOEs is to acquire control of foreign companies. Such acquisitions often raise national security concerns and trigger regulatory reviews in the host country. In the United States, for example, parties to a transaction that could result in a foreign person controlling a U.S. business may file a notice with the Committee on Foreign Investment in the United States ("CFIUS") to determine whether this transaction presents any national security risks.106 This determination is based on several factors: the extent to which the foreign acquirer's investment decisions are based solely on commercial grounds; whether corporate governance structures are in place to ensure management's independence from the controlling government; and the degree of transparency in the foreign acquirer's investment purposes, institutional arrangements, and financial information.107 Understanding executives' political attributes and career patterns is helpful in evaluating the degree of management independence and government involvement in corporate investment decisions. As this Article shows, although having a board of directors makes the SOE governance structure resemble international standards, there is little evidence that those SOEs exhibit 106 Guidance Concerning the National Security Review Conducted by the Committee on Foreign Investment in the United States, 73 Fed. Reg. 74,567, 74,572 (Dec. 8, 2008). CFIUS does not, however, issue advisory opinions. See id. 107 Id. at 74,571. higher-quality corporate leadership. This Article also shows that there is great complexity in the executive personnel system-having a CEO without prior career experience in government bureaus (i.e., a single-group or a multi-group track CEO) does not indicate more managerial independence from the government, because a political career connection can emerge during the post-CEO career stage. Moreover, although the Chinese government has reduced its ownership stake in many SOEs over the past decade, executive personnel integration with important government organs has increased at the same time. The size of the ownership stake alone insufficiently captures the potential degree of state involvement in SOE management; other factors such as executive personnel management should be considered as well. The characteristics of SOE executives also have legal implications for securities regulations in international capital markets, since SOEs have actively traded shares on overseas markets. Do Chinese SOEs' corporate disclosures to international investors provide an adequate and accurate depiction of their actual governance practices? Chinese SOEs may significantly downplay their connections to the Chinese government and its ruling political party in their corporate disclosures. In fact, the government's role expressed in the disclosure is often so simple as to be virtually meaningless. 108 the true role of the Chinese Communist Party is virtually never explained. Their prospectuses and annual reports often omit the top managers' affiliation with national representative political bodies, and rarely explain their responsibilities as members of the party committee. 109 The executive biographies in such 108 Xu Xuelei & Xu Xin, Information Disclosure of State-Owned Enterprises in China, 4 TSINGHUA CHINA L. REV. 1, 22 (2011) ("There is no regulation on the types of information that are required to be disclosed, and no regulation regarding which SOEs are obliged to disclose."). 109 For example, Sinopec, PetroChina, China Mobile, China Telecom, ChinaUnicom, and Huaneng Power are leading Chinese SOEs with shares listed on the New York, London, and Hong Kong Stock Exchanges. The chairman of each of these companies is a member of the Party's National 790 COL UMBIA B USINESS LA W RE VIE W [Vol. 2013 No. 3:743] STATE OWNERSHIP AND CORPORATE GOVERNANCE IN CHINA 791 disclosures are focused on business experiences, but are devoid of political backgrounds, intentionally making the top executives look similar to their Western counterparts. These companies also never disclose the fact that the Party holds appointment power, not the board of directors. Furthermore, there is no disclosure that the top managers are prohibited from freely exercising stock options because of personnel integration between SOEs and other government units.110 In short, a significant amount of material information is omitted or misrepresented. While the lack of adequate disclosure is mainly a result of the secretive culture of the Chinese government, it is also partly due to calculated advice by Western investment bankers, law firms, and accounting firms to make Chinese SOEs attractive to international investors;111 international securities regulators generally loosen enforcement against foreign companies, whether state-owned or not.112 Since political interference remains a major concern in SOE governance, existing disclosure practices regarding Chinese SOEs' political connections should be reassessed to ensure that international investors are adequately informed of all material information. Congress, China's Parliament, and/or China's National Political Consultative Conference. However, such political affiliation information is not disclosed in the companies' annual reports. See SINOPEC 2012 ANNUAL REPORT AND ACCOUNTS 57 (2012); PETROCHINA, 2012 ANNUAL REPORT 77 (2012); CHINA TELECOM ANNUAL REPORT 2012 14 (2012); CHINA UNICOM ANNUAL REPORT 2012 38 (2012). 110 See Chen et al., supra note 7, at 14. inl See RICHARD MCGREGOR, THE PARTY: THE SECRET WORLD OF CHINA'S COMMUNIST RULERS 48 (2010). See also CARL E. WALTER & FRASER J.T. HOWIE, RED CAPITALISM: THE FRAGILE FINANCIAL FOUNDATION OF CHINA'S EXTRAORDINARY RISE 163-84 (2011); FRASER J.T. HOWIE, PRIVATIZING CHINA: INSIDE CHINA'S STOCK MARKETS (Carl E. Walter ed., 2nd ed. 2006). 112 See Jordan Siegel, Can Foreign Firms Bond Themselves Effectively By Renting U.S. Securities Laws?, 75 J. FIN. ECoN. 319, 321 (2004) (finding that SEC enforcement actions against Mexican firms are practically nonexistent and suggesting that this trend extends to all foreign firms). 2. Challenges facing Chinese SOE Governance Reform China's Company Law is the central object of study in the literature of Chinese corporate governance. Although the Company Law provides the fundamental legal structure of corporate entities, including state-owned companies,113 the SOE governance regime, to a great extent, is built on a set of special laws, such as the Law on State-Owned Assets of Enterprises (2008)114 and a large body of administrative rules and guidelines. At first glance, the Company Law appears objective, betraying no trace of political proclivity toward the Chinese Communist Party. However, the large body of administrative rules and guidelines governing SOE management unabashedly reveal strong political preferences serving the Party's interests. The executive recruitment guidelines examined in this Article represent an illustrative example. The state-owner's dual role as a shareholder and regulator can easily construct a legal system in favor of its interests.1 This is particularly true for the Chinese state- owner as the single-ruling party. Compared to high-profile legislation such as the Company Law (where there is generally more public oversight), the administrative rules and guidelines implementing the state-owner's rights and responsibilities have very low transparency and face paltry scrutiny. The low-profile administrative rules can be handy 113 Gongsifa [Company Law] (promulgated by the Standing Comm. Nat'l People's Cong., Dec. 29, 1993, effective Dec. 29, 1993) STANDING COMM. NAT'L PEOPLE'S CONG. GAz. (amended Oct. 27, 2005, effective Jan. 1, 2006). 114 Qiye Guoyou Zichan Fa [Law of the People's Republic of China on the State-Owned Assets of Enterprises] (promulgated by the Standing Comm. Nat'l People's Cong., Oct. 28, 2008, effective May 1, 2009) 2008 STANDING CoMM. NAT'L PEOPLE'S CONG. GAZ. 115 See Mariana Pargendler, State Ownership and Corporate Governance, 80 FORDHAM L. REV. 2917, 2944-45 (2012) (finding in practice that Chinese corporate law serves to protect the government's interest as a controlling shareholder while denying significant legal rights to minority shareholders). 792 COL UAMIA B USINESS LA W RE VIE W [Vol. 2013 No. 3:743] STATE OWNERSHIP AND CORPORATE GOVERNANCE IN CHINA 793 instruments for the state-owner to undercut the corporate governance reform goals stated in the high-profile laws. Besides the problem of shadowy rules, the substantive management of Chinese SOEs may suffer from being excessively closed. The empirical evidence in this Article shows that the elite of large Chinese SOEs is a very cohesive group in terms of educational training, political affiliations, and career experiences. The current executive labor market remains a system-insider's game. The elite's cohesion presents advantages as well as challenges to China's corporate governance and national economy. According to sociologist Ronald Burt's analysis of "brokerage and closure" in social structure, high cohesion can help consensus formation and policy implementation, but runs the risks of closure, groupthink, and a lack of creativity.116 Building brokerage ties that reach outside the closed system can import new ideas and stimulate innovation; as competition in the global economy has become more dependent on innovation, Chinese SOEs need inputs of new talent and management skills in order to gain a competitive edge. The Chinese government seems aware of the potential close- mindedness problem and has adopted some administrative measures to open the SOE executive labor market. In the Party's eyes, neither complete closure nor complete openness is desirable, because complete closure harms competitiveness and complete openness invites loss of control. It will take time for the Chinese government to learn how to strike a balance between the two approaches. Any openness in the executive labor market will most likely start with lower-level managers and move to high-level executives; from business groups owned by the local governments to those controlled by the central government; and from companies faced with fierce competitive pressure to companies in monopolized sectors. The evidence in this Article, however, suggests that the whole process will develop slowly, since personnel management remains highly sensitive and the Chinese government still views it as the 116 See BURT, supra note 12. most important way to control large enterprises as the government reduces its ownership stake. Moreover, while China's SOEs might demand professionals trained outside the state system, the executive reform process can be tempered by the supply side of talent-whether system-outsiders are interested in joining the SOEs. For example, in several instances, SASAC officials have extended offers to system-outsiders, only to have the candidates decline the offers because the pay was below prevailing market standards."' To resolve this problem, the Chinese government is experimenting with a dual system of executive compensation. Under the dual system, the compensation of system-insiders, as is tradition, is unilaterally set by SASAC, while system-outsiders are paid based on market prices through contract negotiation. The latter compensation is usually much higher than the former. It is unclear whether such duality will work well, as it stands to reason that such pay discrepancy can brew resentment among system-insiders.11 In addition, an SOE management culture tainted with political complexities may make system-outsiders shy away from joining the SOE system. As a result, the executive labor market probably will continue to be dominated by system-insiders for a long time, which increases the risks of perpetuating old practices that deviate from international standards. The lack of substantive openness in the composition of the SOE corporate leadership can be a fundamental cause for the common dysfunction of many corporate governance institutions, such as the board of directors and independent directors. 117 See, e.g., Richard McGregor, Soaring Profits Maintain Chinese State's Grip, FIN. TIMES (Dec. 4, 2006, 11:56 PM), http://www.ft.com/intl/cms/s/0/22c8O926-83ee-1ldb-9e95-0000779e2340 .html#axzz2mEiKcPn8 (noting that SASAC "cannot offer competitive salaries"). 118 See id. 794 COL UMBIA B USINESS LA W RE VIE W [Vol. 2013 No. 3:743] STATE OWNERSHIP AND CORPORATE GOVERNANCE IN CHINA 795 3. Directions for Future Research To date, scholars of comparative corporate law have paid scant attention to comparative state-owned enterprises. 19 This research gap should be filled, given that SOEs continue to play a significant role in many international economies. The Organization for Economic Co-operation and Development ("OECD") has recognized the importance of this topic and has initiated surveys on the legal frameworks and governance mechanisms of SOEs in OECD member countries.120 It also has adopted the first international standards for SOE governance.121 The OECD surveys show various cross-country variations and similarities in the SOE governance rules, including the executive appointment process.'22 The findings in the OECD reports and in this Article both urge further investigation of the underlying causes of SOE governance similarities and differences. For instance, this Article shows that Chinese SOE executives bear many similarities with executives in countries favoring 119 Very recently, a small number of corporate governance scholars have begun to approach this topic. See Aldo Musacchio & S~rgio G. Lazzarini, Leviathan in Business: Varieties of State Capitalism and Their Implications for Economic Performance (Working Paper No. 12-108, June 4, 2012) available . at http://papers.ssrn.com/sol3/papers.cfm? abstractid=2070942. 120 As examples of these surveys, see ORG. FOR ECON. CO-OPERATION AND DEV., CORPORATE GOVERNANCE OF STATE-OWNED ENTERPRISES: A SURVEY OF OECD COUNTRIES (2005), available at http://www.oecd.org/corporate/ca/corporategovernanceofstate- ownedenterprises/corporategovernanceofstate-ownedenterprisesasurvey ofoecdcountries.htm; ORG. FOR ECON. CO-OPERATION AND DEV., STATE- OWNED ENTERPRISE GOVERNANCE REFORM: AN INVENTORY OF RECENT CHANGE (2011), available at http://www.oecd.org/corporate/ca/corporate governanceofstate-ownedenterprises/state- ownedenterprisegovernancereformaninventoryofrecentchange.htm. 121 See ORG. FOR ECON. CO-OPERATION AND DEV., OECD GUIDELINES ON CORPORATE GOVERNANCE OF STATE-OWNED ENTERPRISES (2005), available at http://www.oecd.org/corporate/calcorporategovernanceofstate-owned enterprises/oecdguidelinesoncorporategovernanceofstate- ownedenterprises.htm. 122 See supra Table 2 and accompanying text. the state activism and stakeholder-oriented models, and especially with German executives. How do we explain the apparent similarities? Are the underlying causes for the prevalence of engineering-trained CEOs in China similar to those in Germany? Are the underlying causes that have made the single-group track the prevailing career pathway in China the same as those in Germany and Japan? To answer these questions, a context-rich comparison of the legal system, political institution, industrial structure, occupation training system, social organization, and cultural values is required. An investigation into the underlying causes would also help to evaluate the usefulness and appropriateness of the international SOE governance standards proposed by the OECD, since the legal transplant literature has warned against adopting internationally standardized corporate governance rules without a context- rich analysis. 12 3 In addition to comparative study across countries, future research on China might consider comparing corporate leaders in the state-owned sector with those in the country's private sector. Are they a cohesive group-and if not, what specific attributes distinguish them? Moreover, this Article has shown that SASAC and the Party's Organization Department have formulated standards on executive qualifications: what sources do they consult to derive such standards? Do they consult the experiences of China's private sector or do they look to emulate another country's model? It can be useful to explain the apparent similarities in executive attributes with other countries and find out if there is diffusion of a specific executive model. Qualitative research methods, such as interviews with Chinese government officials and top managers, might be helpful avenues of research for this inquiry. Future research might also seek to track data further back into the past, as well as extended into the future, in 123 See Katharina Pistor, The Standardization of Law and Its Effect on Developing Economies, 50 AM. J. COMP. L. 97, 98 (2002) (finding that, to be effective, the law must be understood and embraced by law enforcers and legal customers). 796 COL UMBIA B USINESS LA W RE VIE W [Vol. 2013 No. 3:743] STATE OWNERSHIPAND CORPORATE GOVERNANCE IN CHINA 797 order to reveal a bigger picture of how elite composition has evolved and how it has impacted micro-corporate behavior as well as macro-institutional changes. For example, a preliminary test (unreported in this Article) exploring the relationship between types of career pathways and profitability suggests that SOEs with CEOs coming to power through the unrelated-government-units track seem to offer the worst return on assets.124 Similarly, if in the future more system-outsiders join the SOE system, there will be a sufficiently large pool of observations (currently only one exists in the sample) to statistically examine what kinds of SOEs are likely to hire system-outsiders and whether such system-outsiders contribute to corporate behavior such as profitability or organizational culture change. 124 Future research is needed to scrutinize the relationship between executives' career attributes and different dimensions of corporate performance. APPENDIX 1: MODEL EXPLANATION FOR TABLE 3 The dependent variable in Table 3 is age when appointed as the CEO. The personal attributes include career pathways and educational backgrounds. Types of career pathways as one of the independent variables are measured as dummy variables, with the single-group track as the reference category. The statistical interpretation for using the single-group track as the reference category is that compared to the CEOs of the single-group track, the CEOs of another track (e.g., the multi-group track) tend to reach the executive position at an older/younger age. Educational credentials are measured as dummy variables of whether the executive is a graduate of C9 League (yes=1), whether the executive has an engineering degree (yes=1), a graduate degree (yes=1), and a foreign degree (yes=1), respectively. In addition to personal factors, different organizational environments may affect career outcomes. Some of China's large SOEs have recently experimented with launching a board of directors as an institution to monitor executives. The board of directors is included as a dummy variable (yes=1) to test whether SOEs that established a board of directors may be more active reformers, and thus more likely to have younger CEOs, who tend to be less constrained by old traditions. Moreover, given that SOEs owned by the central government tend to have greater national importance in the economy and thus have stricter standards for executive quality than SOEs owned by local governments, central ownership is included as a dummy variable (yes=1) in the models. The firm revenues variable is included to control for firm size, the number of employees stands for internal labor market size, and return on assets stands for profitability. The variables including revenues and number of employees are log transformed. Data concerning executives' personal attributes (including career and education) derives from multiple sources, including corporate prospectuses, annual reports, corporate websites, government documents and websites, 798 COL UMIA B USINESS LA W RE VIE W [Vol. 2013 No. 3:743] STATE OWNERSHIP AND CORPORATE GOVERNANCE IN CHINA 799 industrial association websites, and news reports. The data concerning organizational attributes are from the China Enterprise Confederation and China Enterprise Directors Association. Models 1, 3, and 5 in Table 3 include personal attributes only (i.e., career pathway and educational backgrounds), while Models 2, 4, and 6 are full models that include all independent and organizational attributes. COLUMBIA BUSINESS LAW REVIEW APPENDIX 2: MODEL EXPLANATION FOR TABLE 4 Table 4 uses binary logistic regression models to examine how personal and organizational attributes affect post-CEO status. Binary logistic regression is a type of regression analysis where the dependent variable is a dummy variable, coded 1 (yes=1) or 0 (no=O). The sample (with missing data excluded) counts 232 CEOs who left office during the period of examination. The dependent variables are four types of immediate post-CEO statuses and have been constructed as dummy variables (yes=1): (1) the chairman/director/other executive of the same business group; (2) a director/executive of other business groups; (3) a senior official in other government units; or (4) a convicted prisoner. Types of career pathways as one of the independent variables are measured as dummy variables with the single- group track as the reference category. Other independent and control variables are similar to those used in Table 3. Models 1, 3, 5, and 7 include personal variables only (i.e., career pathway and educational variables), while Models 2, 4, 6, and 8 are full models that include all independent and controlling variables. 800 [Vol. 2013