Manager or Politician? Effects of CEO Pay on the Performance of State-Controlled Chinese Listed Firms

Abstract

We examine the effects of chief executive officer (CEO) pay on two performance outcomes for central government-controlled Chinese listed firms: financial performance (return on assets, which is associated with the interests of all shareholders) and political performance (job creation, which reflects a primary objective of the controlling owner, the state). We posit that China’s mixed economy places state-owned enterprise (SOE) executives on dual career tracks—business and politics—and their career interests are differentially aligned with the divergent interests of minority shareholders and the state, respectively, depending on the firm’s position in the corresponding enterprise group. Based on a sample of 2,145 firm-year observations from 230 state-controlled Chinese listed firms between 2004 and 2014, we find that higher CEO pay leads to a higher return on assets. This relationship is strengthened when there are more hierarchical layers above the focal firm in the business group, when there is more potential political competition from other group affiliates, and when there is a nonstate second-largest shareholder. Higher CEO pay also leads to greater job creation, but this relationship is weakened by political competition and nonstate second-largest shareholder.

Keywords

CEO compensation SOE performance agency theory principal-principal conflicts business group China

Research investigating Chinese state-owned enterprises (SOEs) typically focuses on the principal-principal conflicts (PPCs) in these firms arising from the divergent interests of majority and minority shareholders (Bruton, Peng, Ahlstrom, Stan, & Xu, 2015; Dharwadkar, George, & Brandes, 2000; Li & Qian, 2013; Young, Peng, Ahlstrom, Bruton, & Jiang, 2008). The issue of principal-agent conflicts (PACs), which is traditionally central to agency theory (Fama & Jensen, 1983; Jensen & Meckling, 1976), does not receive much attention in this stream of research because it is assumed that the agent in SOEs has limited autonomy under central planning and bureaucratic control (Peng & Heath, 1996). In particular, top-level executives carry out the orders and serve the political objectives of the controlling owner—namely, the state—often at the cost of the minority shareholders (Bruton et al., 2015).

However, this assumption is not entirely consistent with the reality of China’s new economic order. A notable feature of the Chinese economy is that it is a mixed “socialist market economy” (Marquis & Qiao, 2020). SOEs are partly in a hierarchy, forming a nationwide, pyramidal group structure (Naughton, 2017), and partly in the market, coexisting with private firms, mixed-ownership firms, and foreign firms (Xu, Lu, & Gu, 2014). To compete with other types of firms, SOE managers have been given substantial incentives and autonomy so that they can respond to market imperatives (Tan & Tan, 2005; Xu & Shirley, 2001). Not surprisingly, the incidence of self-serving behavior has risen among SOE managers (Hu & Sun, 2019; Jia, Huang, & Zhang, 2018). Meanwhile, as SOEs are listed in the stock market to raise capital, their managers face the demands and requirements of the minority shareholders in their firms (Feinerman, 2007); their career interest at least partly depends on their ability to improve firms’ financial performance (Firth, Fung, & Rui, 2006a; Mengistae & Xu, 2004). Thus, neither the traditional focus on PACs nor the recent attention paid to PPCs alone can sufficiently address this reality.

In this study, we examine the effect of CEO pay on firm performance by considering both the PACs and PPCs in state-controlled Chinese listed firms. Executive pay is central to the principal-agent relationship, with an emphasis on the firm’s financial performance (Fama, 1980; Jensen & Murphy, 1990). In Chinese SOEs, executive pay also concerns the differing performance goals of the principals—while the minority shareholders mainly pursue financial interests, the controlling shareholder—namely, the state—may pursue a more diverse set of goals and place a higher priority on its political objectives (Bruton et al., 2015; Wang & Luo, 2019). Our purpose is to reconcile two different angles within agency theory. One posits that executives with higher pay are incentivized to deliver better financial performance to protect their own position and reputation (Fama, 1980), and the other—namely, the PPC perspective—suggests that different classes of owners value different performance outcomes including political performance (Jia et al., 2018), forcing managers to make trade-offs. The relevant question to ask, therefore, concerns when the monitoring mechanism—in our case, CEO pay—will produce performance outcomes that better serve the interests of one class of principals versus the other.

In answering this question, our central thesis is that China’s mixed economy places SOE executives on dual career paths—that is, a business career path and a political career path (Li & Andrew, 2001; Walder, Li, & Treiman, 2000)—and their career interests are differentially aligned with the interests of the minority shareholders and the state, depending on a set of structural factors. These factors—specifically, the number of hierarchical layers above the focal firm in the enterprise group, future competition on the executive’s political ladder, and the existence of a nonstate second-largest shareholder—serve as moderating influences on the relationships between CEO pay and an SOE’s financial and political performances.

Our study makes two intended contributions to theory and research. First, we provide an explicit integration of the PPC perspective and the broader agency theory by drawing attention to the PAC, which has often been downplayed in the PPC literature. Attempts at such integrations are not common, as the standard agency theory and the PPC perspective have different foci. To achieve this purpose, we focus on the conditions under which the agents’ interests are better aligned with those of one class of owners versus the other. Thus, the agent-principal relationship is extended to cover both agent-minority shareholder and agent-controlling owner relationships. Second, we contribute to a deeper understanding of managerial behavior in Chinese SOEs by examining how the dual career track system motivates or demotivates managers in a mixed economy. We relate SOE executives’ career interests to the vertical and horizontal structural positions of their firms in the SOE groups and highlight the role of nonstate second-largest owners of listed SOEs, which tends to be overlooked in the usual controlling owner-minority shareholder dichotomy.

Theoretical Foundation and Institutional Background

The central theme of agency theory is that there is a divergence of interests between the principal and the agent because of the separation of ownership and control, such that managers exhibit self-serving behavior at the cost of shareholders (Eisenhardt, 1989; Fama & Jensen, 1983; Jensen & Meckling, 1976). The classic agency framework involves a contractual relationship between the agent and principals, with no assumption of differing interests or preferences among the principals (Jensen & Meckling, 1976). Shareholder interests are typically interpreted in terms of financial interests or firm value, which concerns all shareholders, and therefore, the severity of the PAC, due to the agent’s self-serving behavior, is assumed to affect all shareholders, although only large shareholders are willing to pay the cost of monitoring (Shleifer & Vishny, 1986).

The growing PPC literature shifts the focus to the divergence of interests between two classes of principals—that is, the controlling owners and the minority shareholders (Chang, 2003; Dharwadkar et al., 2000; Young et al., 2008). A typical example is state-controlled listed firms in China, in which the interests of the controlling owner (i.e., the state) differ or conflict with those of the minority shareholders (Dharwadkar et al., 2000; Li & Qian, 2013; Su, Xu, & Phan, 2008). While minority shareholders’ primary goals are financial, the state has its own political objectives, such as supporting and sustaining economic growth, creating jobs, maintaining social stability and welfare, subsidizing underdeveloped regions, and meeting national strategic objectives (Bruton et al., 2015; Du, Tang, & Young, 2012; Holmes, Hoskisson, Kim, Wan, & Holcomb, 2018; Wang & Luo, 2019). A significant volume of research has documented that the Chinese government places high priority on its political objectives, especially on the objective of maintaining employment and hence social stability, thus compromising minority shareholders’ financial interests (Chang & Wong, 2009; Delios, Zhou, & Xu, 2008; Du et al., 2012; Li, Li, & Wang, 2019; Tihanyi et al., 2019). In essence, the government is using minority shareholders’ money to achieve its political goals.

While the PPC problem is viewed as a type of agency problem and the PPC perspective is considered a branch of agency theory (Dharwadkar et al., 2000), the agent does not play a large role from this perspective, because it is assumed that managers are related to or closely affiliated with the controlling owner and do not have independent interests (Young et al., 2008). However, we argue that SOEs possess abundant resources and substantial autonomy as a result of the development of a mixed economy (Tan & Peng, 2003; Tan & Tan, 2005; Zhou, Gao, & Zhao, 2017). Moreover, when SOEs become publicly listed firms, they are subject to the same corporate governance rules as those faced by nonstate-controlled listed firms (Feinerman, 2007), at least by the letter of the law. Therefore, SOE managers have a fiduciary duty to all shareholders and not just to the state. Extant research suggests that executives of Chinese listed firms, many of which are state-controlled, are financially rewarded for creating value for shareholders (Buck, Liu, & Skovoroda, 2008; Firth et al., 2006a; Mengistae & Xu, 2004), and their job security and career interests are also tied to firm performance, as in other parts of the world (Firth, Fung, & Rui, 2006b). Such new autonomy and responsibilities, coupled with managerial self-interest and the existence of the two classes of owners, are likely to give rise to both agent–minority shareholder and agent–controlling owner conflicts.

Career interest is the most pivotal aspect of managers’ self-interest in agency theory, reflected in the notion that managers are risk averse, as most of their human capital is invested in the firm (Beatty & Zajac, 1994; Eisenhardt, 1989). A well-known example is that managers will pursue value-destroying diversification to reduce firm risk—that is, to protect their own jobs—unless there is adequate monitoring by large shareholders (Amihud & Lev, 1999). A more optimistic view from one of the classic works on agency theory suggests that managers promote their career interests through good performance because they want to convey the value of their human capital to the market so that they can maximize their pay during the next contract negotiation (Fama, 1980).

The career interests of SOE managers are complicated because they face two career tracks: a business track in the corporate system and a political track within the Party and government. A prominent feature of publicly listed SOEs is that they are controlled by large, fully state-owned enterprise groups (Fan, Wong, & Zhang, 2007; Keister, 2009). Typically, the top managers themselves are also Communist Party members reporting to the Party Secretary, who is often the chairman of the board in the firm and who, in turn, reports to the higher levels of the Party organization in the enterprise group (Holz, 2007; Lin, 2017b). Top SOE managers are included in a Soviet-style nomenklatura, which is a system of important government and corporate posts filled by Party appointees (Brødsgaard, 2012; Chan, 2004). When they are at lower levels of the political hierarchy, managers typically only climb the business ladder within the enterprise group and are not transferred to political posts (Lin, 2017b; Walder, 1995a; Walder et al., 2000); at higher levels, political transfers become frequent, and they are promoted on the grounds of both political loyalty and their capability to manage economic activities (Walder, 1995b; Zuo, 2015).

At the top of the pyramid of corporate elites are the executives of approximately 100 central enterprise groups directly owned and controlled by the State Assets Supervision and Administration Commission (SASAC), representing the central government. These large enterprise groups are the “national champions” that receive full attention from the government (Brødsgaard, 2012; Lin & Milhaupt, 2013). The top executive positions in these groups correspond to the ministerial and provincial levels in the nomenklatura system. It is quite common that the CEO of a central enterprise group is transferred to the position of a central government minister or provincial governor (Brødsgaard, 2012; Lin, 2017b). For example, Qishan Wang, who is currently the Vice President of China, was the CEO and Party Secretary of China Construction Bank Corporation, a central SOE group, before becoming a Deputy Governor of Guangdong Province in 1997.

The existence of the two career paths poses unique agency problems in state-controlled listed firms. Because minority shareholders have primarily financial objectives, an alignment of interests between managers and minority shareholders can be achieved only if managers are properly rewarded for their financial performance in their business career. Although it is well understood that Chinese SOEs have political mandates, less informed small investors prefer to rely on direct, objective, and quantifiable performance measures for evaluating managers (Jia et al., 2018). However, to the Party-state, the primary objectives are political, especially for higher ranking posts (Cao, 2001). Correspondingly, although “both red and expert” are criteria for cadre promotion (Guo, 2019: 97), redness, namely, political loyalty to the Party and its ideology, has often taken precedence over expertise. In particular, maintaining social stability and, ultimately, the rule of the Communist Party is considered a performance criterion with “veto power” over other criteria (Wang & Luo, 2019: 12).

Thus, when SOE executives are unable to satisfy both the political and financial demands of the corresponding Party committees, either because these demands conflict with each other or because the managers have limited resources, capability, and attention (Ocasio, 1997, 2011), they are faced with a trade-off between the two career paths: They can either meet the political demands of the state, possibly at the cost of other shareholders’ financial interests, or make a true effort commensurate with their pay to maximize shareholder value, hence compromising the state’s political goals and their political careers. In the former case, they serve as agents of the state only and choose to climb the political ladder, since political performance becomes more important at higher hierarchical levels (Guo, 2019). In the latter case, they establish their credibility and reputation as dutiful professional managers and, even if their political career is terminated as a consequence, they can expect to advance their business career in the private sector (Allen, Qian, & Qian, 2005; Groves, Hong, McMillan, & Naughton, 1995). It is quite common to see media reports about SOE managers or government officials joining private firms (see, for example, Sina.com, 2014).

Hypotheses

CEO Pay, Financial Performance, and Political Performance

Executive pay is an important aspect of the principal-agent relationship, as it holds the key to the interest alignment between the two parties; an appropriate pay level and pay structure are expected to provide sufficient motivation for managers and lead to higher firm performance (Eisenhardt, 1989). The vast majority of studies on the relationship between CEO pay and performance examine the performance → pay link instead of the pay → performance link (e.g., Barkema & Gomez-Mejia, 1998; Chizema, Liu, Lu, & Gao, 2015; Wasserman, 2006). We focus on the latter, as we aim to understand how the manager’s differential interest alignment with the controlling versus minority shareholders influences the managerial behavior and firm performance, rather than how principals reward the agent.

Among the relatively few studies on the CEO pay → firm performance relationship, most have focused on the effect of the pay structure on performance (Brickley, Bhagat, & Lease, 1985; Carpenter & Sanders, 2002). Surprisingly, academic research about the effect of the executive pay level on performance is rather limited, with researchers paying particular attention to the weak association between the two, under the popular belief that CEOs are significantly overpaid (Jensen & Murphy, 1990; Tosi, Werner, Katz, & Gomez-Mejia, 2000). The level of executive pay in Chinese SOEs is generally lower than in non-SOEs (Lin, 2017a). It is certainly far more constrained than in firms in the United States and some other Western countries because of an ideological orientation that values egalitarianism more than economic efficiency (Aguilera, Duran, Heugens, Sauerwald, Turturea, & VanEssen, 2020). However, precisely because of such constraints, the extent of the marketization of managerial pay has always been viewed as an indicator of the success or failure of SOE reform since the 1980s (Groves et al., 1995; Mengistae & Xu, 2004). We examine the pay level → performance link as our baseline relationship because, unlike in U.S. firms, executive pay in China comprises almost exclusively salary and annual bonuses but not stock options (Buck et al., 2008), which is even more true in the case of state-controlled firms (Conyon & He, 2011). Thus, whether pay level can effectively motivate SOE executives to act in the interest of their shareholders becomes an important but underexplored research question.

Agency theory holds that agency problems can be mitigated through the external managerial labor market and internal monitoring among managers (Fama, 1980). A manager’s fundamental interest lies in the security and advancement of his or her career. As Fama (1980: 291-292) notes, “the managers of a firm rent a substantial lump of their wealth—their human capital—to the firm, and the rental rates for their human capital signaled by the managerial labor market are likely to depend on the success or failure of the firm.” The value of competent corporate executives relative to their pay, as reflected by the superior performance of the firm, will likely bring adequate future compensation for them. Although a managerial labor market has yet to fledge in China’s state sector, such a reputation effect does exist in the private sector (Allen et al., 2005; Groves et al., 1995) and is likely to also influence firms that are at the intersection of the state and market sectors, including the state-controlled listed firms. Thus, to prove that they are worth the pay that they received and secure higher pay in the next contract with the current or another firm, SOE executives will likely refrain from shirking and exert job effort that is commensurate with the level of their pay to achieve the objectives of all shareholders (Fong, Misangyi, & Tosi, 2010).

The objectives of a state-controlled firm, however, are more than financial. While both the state and minority shareholders are interested in seeing more profits, the state has political objectives that are not shared by minority owners. Among these political objectives, a persistent theme is to maintain social stability, which in turn translates into the need to create more jobs, and this is a political responsibility borne particularly by the remaining large, more resourceful SOEs in the aftermath of the large-scale privatization of and massive lay-offs by small- and medium-sized SOEs (Bai, Lu, & Tao, 2006; Bruton et al., 2015; Wang & Luo, 2019). As agents of the state, SOE executives will likely make an effort that is commensurate with their pay to achieve the primary political objective of the government, in addition to the financial objectives of all shareholders. Thus, we argue that to prove their worth in the eyes of both the Party-state and minority shareholders, SOE managers must fulfil both financial and political objectives, and the higher the salary that they are paid is, the more this is true, other things being equal.

Hypothesis 1a (H1a): There is a positive relationship between CEO pay and financial performance in state-controlled listed firms.

Hypothesis 1b (H1b): There is a positive relationship between CEO pay and political performance in state-controlled listed firms.

The Enterprise Group and the CEO Pay: Firm Performance Relationship

While SOE managers may endeavor to achieve both the financial and political objectives, the structural positions of their firms in the SOE groups have differing impacts on their business and political career interests, such that the managers are differentially aligned with the interests of the minority and controlling shareholders. Consequently, the strengths of the relationships between CEO pay and the financial versus political performance of the firm will diverge based on the structural factors, which we discuss below.

Hierarchical layers

The hierarchical position of a firm in the SOE group has important bearings on the relative weights of the two careers in the manager’s value calculation. The central SOEs have all formed large enterprise groups comprising many listed or unlisted firms, with a complex ownership holding structure. Through its corresponding group, each listed SOE is ultimately owned and controlled by the SASAC, but a varying number of layers exist between the listed firm and the top of the group pyramid (Fan, Wong, & Zhang, 2012; Keister, 2009; Xu, 2011). Figure 1 illustrates three different group structures. Figure 1a shows that the focal listed firm China Coal Energy is only one level below the top of the group China National Coal Group Corporation, which, in turn, is directly under the SASAC; Figure 1b shows that the focal listed firm Jiangling Motors is four levels below the top of the group China South Industries Group Corporation; Figure 1c shows that the focal listed firm CTS International Logistics is three levels below the top of the group China National Travel Service Group Corporation.

Figure 1

Ownership Charts of Representative State-Controlled Chinese Listed Firms: (a) China Coal Energy (2009), (b) Jiangling Motors (2005), and (c) CTS International Logistics (2012)

If an SOE executive is a typical business person, the hierarchical level of the firm would not affect his or her value calculation. Ostensibly, all listed firms are legally equal, independent market players (Feinerman, 2007; Firth et al., 2006a). A CEO position with a listed firm is the apex of a manager’s career and end of the business ladder. Regardless of the hierarchical level of their firms, CEOs cannot be “promoted” any further in terms of corporate positions. They can only maximize their business-career interests by making greater efforts on the job, because their firms—the listed firms—are the most market-oriented parts of the whole state sector. Any “promotion” to a higher level in the SOE group would be political, which means that the executive is leaving a CEO position with market-based pay and joining a wholly state-owned firm with a higher political status but lower pay, which is unattractive to the manager if he or she has no political ambition.

However, most SOE executives may be potentially interested in both careers because both are open to them. These executives may become interested in seeking promotion on the political ladder if they believe there are enough benefits to do so. We suggest that the hierarchical position of their firms affects the relative value of their political career versus business career. The political ladder in the enterprise group is hierarchical: A firm closer to the top of the group (e.g., China Coal Energy) is at a higher level, politically, than a firm more distant from the top (e.g., Jiangling Motors), according to the nomenklatura system. Given the tournament structure of political competition, advancement at a higher hierarchical level carries a greater political reward in terms of greater interrank spread compared to advancement at a lower level (Connelly, Tihanyi, Crook, & Gangloff, 2014; Conyon, Peck, & Sadler, 2001). Once SOE executives move along the political ladder to the top of the group hierarchy, the next step is for them to be transferred to Party or government positions, possibly at the deputy ministerial or deputy provincial-governor level, to fully capture the power and privileges associated with high-ranking political positions (Shirley & Xu, 1998).

Thus, the choice between the two careers varies with the hierarchical level. At a lower hierarchical level—namely, when there are more hierarchical layers above the focal managers—the political career carries fewer benefits for them. Their career interests are mostly tied to their position as the CEO of the company, and they can only maximize their own interests by promoting the financial interests of all shareholders. By doing so, they may offend the controlling owner and lose their current jobs if superior Party officials believe that they have disregarded their political responsibilities, but they will likely be compensated in the private sector since they have established their reputation as competent and trustworthy corporate executives (Allen et al., 2005; Groves et al., 1995). At a higher hierarchical level, managers’ career interests in business remain unchanged, but the relative importance of their business career interests becomes lower because the potential benefits from a political career are greatly amplified. To secure a political promotion, executives need to comply with the will of the Party-state and direct their own attention to a broader set of demands, particularly political objectives (Shirley & Xu, 1998; Xu, 2011).

Hypothesis 2a (H2a): The positive relationship between CEO pay and financial performance is strengthened when there are more hierarchical layers between a state-controlled listed firm and the top level of the enterprise group.

Hypothesis 2b (H2b): The positive relationship between CEO pay and political performance is weakened when there are more hierarchical layers between a state-controlled listed firm and the top level of the enterprise group.

Political competition

In addition to the relative value of a political career to SOE executives, the relative difficulty of advancement in a political career also affects their decisions. Just as in a business career, executives interested in political advancement also encounter competition along their political career path. Executives from other firms in the same enterprise group may constitute competition on a focal executive’s political ladder. Naughton (2017) describes the Chinese political system as a vast pyramid that includes the Party, government, and SOEs. Everyone wants to move up in the system, but promotion is difficult due to the limited number of slots. Officials in ministerial offices and regional governments thus become competitors in their respective tournaments, and so do executives in SOEs (Xu, 2011). Research suggests that with more contestants in a tournament, the odds of winning the competition are smaller, and the contestants are likely to decrease their effort (and presumably divert their effort to other opportunities) unless the reward in the current competition is increased (Connelly et al., 2014). Correspondingly, research in Chinese politics also suggests that too many officials accountable to the same principal leads to shirking (Lü & Landry, 2014).

While managers from all other firms in the enterprise group may become political contestants, not all of them are on a focal executive’s radar screen, depending on the latter’s career horizon. Some executives have a shorter career horizon and target only the next few steps on the political ladder, whereas others may have much higher, long-term ambitions (Wang & Luo, 2019). Consequently, different people target different reference groups as their respective competitors in their political careers. Regardless of a CEO’s political career horizon, the key executives within the enterprise group at the hierarchical level immediately above his or her level are likely viewed as the most relevant competitors for future advancement, and therefore, the CEO likely pays the most attention to those executives (Chen, 1996; Ocasio, 2011) and uses the number of group-affiliated firms at that level as the most relevant estimator of political competition in the foreseeable future.

The implication is that when managers perceive more competition at the level immediately above their own, they are less confident in their own chances of advancing to higher levels in the hierarchy and likely attach less value to their political career to limit potential loss (Chen, Ham, & Lim, 2011). Therefore, they place a higher emphasis on the alternative career path and become more diligent in their business duties. When there is less competition in managers’ political career, they perceive a higher probability of moving up the political ladder and become more devoted politically. Although one can argue that politically ambitious CEOs also need to deliver good financial results to boost their credentials, their effort is unlikely to exceed the efforts of executives with no such political motivation, who pay full attention to their business career.

Hypothesis 3a (H3a): The positive relationship between CEO pay and financial performance is strengthened when there are more affiliated firms in the enterprise group at the level immediately above the focal state-controlled listed firm.

Hypothesis 3b (H3b): The positive relationship between CEO pay and political performance is weakened when there are more affiliated firms in the enterprise group at the level immediately above the focal state-controlled listed firm.

Nonstate large shareholders

Ownership structure is an important determinant of both principal-agent and principal-principal problems. Large shareholders overcome the free-rider problem and serve as an effective monitoring mechanism for agent behavior, as they have both financial motivation and cost efficiency for monitoring (Shleifer & Vishny, 1986). Prior research finds that ownership concentration can limit the extent of self-serving by managers and, as a result, leads to lower levels of corporate diversification (Amihud & Lev, 1999). Meanwhile, the PPC literature also reveals that the ownership structure is a key mechanism in the expropriation of minority shareholder interests by the controlling owners (Chang, 2003; Singla, Veliyath, & George, 2014; Su et al., 2008). In this literature, whereas controlling owners typically face small minority shareholders (Dharwadkar et al., 2000), the existence of other large shareholders whose interests differ from those of the controlling owners could make a difference.

Among the large shareholders, the second-largest shareholder can most effectively counterbalance the power of the controlling owner (Dhillon & Rossetto, 2015; Laeven & Levine, 2008) and therefore is in a unique position to influence the CEO’s career choices. When the controlling owner is the state, the presence of a nonstate—namely, private or foreign, second-largest owner—is critical. Extant research shows that the performance of state-controlled firms is lower than that of private/foreign firms due to low efficiency and political influences (Chang & Xu, 2008; Li et al., 2019; Tihanyi et al., 2019). When mixed ownership is involved, the amount of state shares depresses firm value (Wei, Xie, & Zhang, 2005), and the performance rises if control is fully or partially transferred from the state to private investors (Gupta, 2005). Thus, one can expect that the existence of nonstate large shareholders will offset the political orientation of the controlling owner—namely, the state (Firth et al., 2006a). Particularly, a nonstate second-largest shareholder—such as a private entrepreneur, family-controlled firm, or other private entity—in a state-controlled firm may better align the managers’ interests with the firm’s financial objectives, as opposed to political goals, and encourage these managers to pursue a business career. The existence of another powerful principal on the board with a vested interest and diverse networks in the private sector serves as protection for the executives who are content to be professional managers rather than politicians. In short, although nonstate second-largest shareholders have less power compared to the controlling owners, they represent the private sector and an entirely different career path that is largely independent of the state.

Hypothesis 4a (H4a): The positive relationship between CEO pay and financial performance is strengthened when the second-largest shareholder is a nonstate owner.

Hypothesis 4b (H4b): The positive relationship between CEO pay and political performance is weakened when the second-largest shareholder is a nonstate owner.

Methods

Sample and Data

Our sample firms consist of the entire population of state-controlled Chinese public companies listed in either the Shanghai or Shenzhen Stock Exchange and affiliated with a central SOE group. To determine our sample, we first obtained the full list of central SOE groups during 2004 to 2014 from the SASAC website. We chose 2004 as the starting year because it is the first year when full disclosure of ownership charts in annual reports became available. The number of SOE groups ranged from 178 in 2004 to 112 in 2014. We identified all the listed firms of these groups during the same period, which varied from 149 in 2004 to 257 in 2014. We collected CEO- and firm-level data, including the firms’ ownership charts, from the China Stock Market and Accounting Research (CSMAR) database (Sun, Hu, & Hillman, 2016) and regional-level data from the National Bureau of Statistics (Li & Qian, 2013). Among the sample firms, 27 had missing data on job creation (8 firms), CEO education (12 firms), CEO pay (5 firms), or CEO duality (2 firms). After removing these 27 firms, we obtained a final sample of 230 firms. We performed mean comparison t tests and found no significant differences between the included and excluded observations, suggesting that the exclusion of these firms did not introduce sampling bias.

Measures

Dependent variables

There are two performance constructs in this study—financial performance and political performance—which are gauged by the return on assets (ROA) and job creation, respectively, and serve as the two dependent variables. ROA is the most commonly used performance variable in prior studies that examine the financial performance of firms in general and of Chinese SOEs in particular (Conyon & He, 2011; Mengistae & Xu, 2004; Tam & Hu, 2006). The ROA is measured as the percentage of the net income over the total assets of the firm, as adjusted by the relevant industry mean (i.e., the focal firm’s ROA minus the average ROA for all firms in the industry and year).

Whereas financial performance measures such as the ROA are well established in the literature, empirical studies directly addressing corporate political performance are relatively few. In the context of SOEs, a general consensus among researchers is that employment growth, or job creation, is an important political task for SOE managers worldwide (Bruton et al., 2015; Cuervo-Cazurra, Inkpen, Musacchio, & Ramaswamy, 2014; Dewenter & Malatesta, 2001; Tihanyi et al., 2019) and particularly in China (Delios et al., 2008; Du et al., 2012; Wang & Luo, 2019). This criterion directly affects promotions within the Party hierarchy. As Zheng and Deng (2018: 72) point out, “(o)f all the promotion criteria, one factor dominates—local employment rate. [. . .] Hence, to increase local employment will please one’s party boss one level higher to increase one’s chance to get promotion.” The pressure on high-level government officials for job creation is channeled through the state corporate system to their agents in the listed firms. Therefore, we use job creation as a proxy for political performance. The variable is measured as the difference between the numbers of total employees in the current year and the previous year, as adjusted by the industry mean (i.e., the focal firm’s job creation minus the average job creation for all firms in the industry and year). Since both profitability and labor intensity vary significantly across industries, adjusting the two dependent variables by the industry means increases the accuracy of the performance evaluation.

Independent variable

The independent variable of this study is CEO pay. Because compensation for Chinese SOE executives consists almost exclusively of cash components (Buck et al., 2008), we followed prior studies by measuring CEO pay as the natural log transformation of total annual salary and bonus (Boivie, Lange, McDonald, & Westphal, 2011).

Moderating variables

The first moderator is the number of hierarchical layers in the group. Each focal firm annually reports its ownership chart by presenting itself as the bottom level of the chart and moving upwards along the chain of ownership within the enterprise group. We counted the number of hierarchical layers between the focal listed firm and the top of the enterprise group. This measurement captures the “distance” between a focal CEO’s current political status and the top level of the group (see Figure 1). The second moderator is the near-future political competition of the focal CEO, which is measured as the number of affiliated firms in the enterprise group at the level immediately above the focal firm. For example, as shown in Figure 1a, the CEO of China Coal Energy has no competition at a higher level, and the variable is coded 0. As shown in Figure 1b, at the level above Jiangling Motors, there is only one firm within the group, namely, Jiangxi Jiangling Holdings, and the variable is coded 1. As shown in Figure 1c, there are three firms above CTS International Logistics, and the variable is coded 3. The third moderator is the existence of a nonstate second-largest owner of the focal listed firm through ultimate shareholding. These owners must be outside the SOE group and have at least 5% voting rights in the focal firm (Boivie et al., 2011). We assessed ownership on the basis of ultimate, instead of direct, shareholding, because we are addressing the potential expropriation of minority shareholder interests by the controlling owner through a pyramidic group structure (Faccio, Lang, & Young, 2001; La Porta, Lopez-de-Silanes, & Shleifer, 1999; Shleifer & Vishny, 1997). We used a binary variable coded 1 if the focal firm’s second-largest owner is a nonstate (private/foreign) owner and coded 0 otherwise. To illustrate, the firms shown in Figure 1a and c were assigned a value of 0 because they did not have any other large ultimate shareholders, but in Figure 1b, Jiangling Motors was given the value of 1 because its second-largest shareholder (outside the enterprise group) is a private/foreign firm (i.e., Ford).

Control variables

We included CEO-, firm-, and regional-level control variables that influence firm performance and/or the relationship between CEO pay and firm performance (Becker, Atinc, Breaugh, Carlson, Edwards, & Spector, 2015; Spector & Brannick, 2011). At the CEO level, we controlled for personal characteristics that have been found to have such influences (Daily & Johnson, 1997; Firth et al., 2006b; Haynes & Hillman, 2010; Hill & Phan, 1991; Zhong, 2015). These are CEO gender, which is a binary variable coded as 1 if the CEO is male and 0 otherwise; CEO age, which is based on the reported age of the CEO; CEO education, which is a continuous variable that varies from 1 to 5 to indicate the CEO’s education level from secondary school or below to a doctoral qualification; CEO tenure, which is the number of years that the CEO has held the position; and CEO duality, which is coded as 1 if the CEO is the same person as the chair of the board and 0 otherwise. By the same token, we controlled for a series of firm-level variables, including firm size, which is measured by the natural log of total assets; firm age, which is measured as the number of years since the firm’s establishment; debt, which is measured by the firm’s total debt; central state ownership, which is measured by the percentage of equity owned by the largest ultimate owner—the central government—based on the pyramidal ownership structure of the enterprise group (La Porta et al., 1999); and board independence, which is measured by the percentage of independent directors on the board (Hu & Tam, 2012). Further, our two dependent variables, ROA and job creation, also serve as firm-level controls in each other’s regression models. Lastly, at the regional level, we included the local GDP, which is measured by the natural log of the provincial GDP, and local unemployment rate, which is measured by the unemployment rate in each of the 31 provinces (Fan et al., 2007; Sun et al., 2016).

Statistical Analysis

A common issue in studies of the relationship between pay and performance is the need to address endogeneity, including reverse causality (e.g., Buck et al., 2008). Among various statistical models (e.g., maximum likelihood estimation and general cross-lagged panel models), we employed the Arellano and Bond system-generalized method of moments (system GMM) technique. The model enables the inclusion of at least one not strictly exogenous explanatory variable, unobserved fixed effects, heteroscedasticity, and serial correlation within firms; thus, it is particularly efficient in addressing various potential endogeneity problems (Arellano, 2003). We used the lagged values of the explanatory and control variables as instruments. Because CEO pay may also be affected by prior firm performance, we included a lagged dependent variable (ROA or job creation in year t−1) as an additional instrument to address the endogeneity issue (Arellano & Bond, 1991). All models were diagnosed for heteroskedasticity, nonlinearity, and outliers, and no problems were found. We mean-centered the independent and moderating variables to reduce the impact of multicollinearity caused by the interaction terms and to facilitate interpretation of these terms (Aiken & West, 1991). Time-invariant control variables such as industry and province dummies are not required in system GMM models (Arellano, 2003).

Results

Table 1 reports the means, standard deviations, and correlations of the variables. Table 2 presents the results for firms’ financial performance in a series of dynamic panel models. Upon running the Arellano–Bond system GMM models, serial correlation tests indicated that the residuals in the first differences (AR(1)) were significantly correlated; however, there was an absence of second-order serial correlation (AR(2)), thereby suggesting that the model specifications were valid. Further, the Hansen tests of overidentification were not significant in any model, and the Difference-in-Hansen tests of exogeneity showed that the subset of instruments used in the levels equations was exogenous for all specifications with insignificant p values, thus suggesting that our instruments were valid and our models properly defined (Hansen, 1982).

Table 1

Descriptive Statistics and Correlation Matrix

		M	SD	1	2	3	4	5	6	7	8	9	10	11	12
1	ROA^a	3.87	16.29	1.00
2	Job creation^a	854.02	5,669.61	0.06	1.00
3	CEO pay	12.95	0.83	0.24	0.09	1.00
4	CEO Pay × Hierarchical Layers	0.09	3.43	0.23	0.08	0.83	1.00
5	CEO Pay × Political Competition	0.02	0.43	0.21	0.10	0.73	0.50	1.00
6	CEO Pay × Nonstate Second-Largest Owner	0.50	2.55	0.12	−0.01	0.35	0.17	0.04	1.00
7	Hierarchical layers	3.10	1.07	0.05	−0.04	0.10	0.19	0.02	0.16	1.00
8	Political competition	1.18	0.58	0.06	0.01	0.10	0.14	0.05	−0.04	0.28	1.00
9	Nonstate second-largest owner	0.04	0.18	0.05	−0.01	0.11	0.15	0.03	0.20	0.22	−0.01	1.00
10	CEO age	49.69	4.99	0.01	0.03	0.11	0.12	0.06	0.07	−0.01	−0.01	0.01	1.00
11	CEO gender	0.95	0.16	−0.01	−0.01	−0.06	−0.06	−0.03	−0.11	0.00	−0.05	−0.03	0.02	1.00
12	CEO education	3.75	0.37	0.00	0.06	0.06	0.05	0.04	−0.01	−0.04	−0.02	−0.03	−0.16	−0.02	1.00
13	CEO tenure	3.03	1.20	0.01	0.03	0.00	−0.01	0.01	−0.01	−0.01	−0.01	0.01	0.05	−0.08	0.02
14	CEO duality	0.07	0.26	0.06	0.02	0.07	0.04	0.07	0.03	0.01	−0.06	0.05	0.04	0.05	−0.05
15	Firm size	22.53	1.74	−0.01	0.19	0.50	0.45	0.47	0.05	−0.14	−0.03	−0.03	0.14	0.00	0.12
16	Firm age	12.36	5.01	−0.04	−0.05	0.19	0.22	0.04	0.09	0.23	0.06	0.09	0.08	0.03	−0.02
17	Debt	21.78	2.02	−0.04	0.02	0.25	0.41	0.43	0.04	−0.12	−0.05	−0.05	0.13	−0.01	0.10
18	Central state ownership	40.97	16.31	0.06	0.07	0.07	0.02	0.15	−0.08	−0.37	0.07	−0.25	0.05	0.01	0.06
19	Board independence	36.41	6.11	−0.01	0.08	0.13	0.10	0.14	−0.02	−0.13	0.02	−0.07	0.02	0.00	0.02
20	Local GDP	9.59	0.77	0.01	0.03	0.44	0.42	0.33	0.18	0.22	0.02	0.11	0.04	−0.01	0.07
21	Local unemployment rate	3.11	1.07	−0.05	−0.04	−0.38	−0.29	−0.30	−0.14	0.04	−0.03	0.01	0.06	0.06	−0.08

		13	14	15	16	17	18	19	20
14	CEO duality	0.00	1.00
15	Firm size	0.03	0.10	1.00
16	Firm age	−0.02	0.04	0.10	1.00
17	Debt	0.03	0.08	0.47	0.14	1.00
18	Central state ownership	0.01	0.00	0.27	−0.36	0.23	1.00
19	Board independence	0.00	−0.09	0.29	−0.12	0.28	0.12	1.00
20	Local GDP	0.04	0.02	0.21	0.33	0.20	−0.08	0.06	1.00
21	Local unemployment rate	−0.05	0.00	−0.20	0.05	−0.16	−0.14	−0.10	−0.25

Note. N = 230 firms. Correlations ≥|0.05| are significant at p ≤ 0.05.

For ease of interpretation, we provide raw values of the firm ROA and job creation here, though we use the industry-adjusted values in our analysis.

Table 2

Arellano–Bond GMM Models for Firm Financial Performance (ROA)

	Model 1		Model 2		Model 3		Model 4		Model 5		Model 6
CEO pay			3.64 (0.29)	[0.000]	1.98 (0.57)	[0.000]	1.66 (0.58)	[0.004]	3.21 (0.31)	[0.000]	5.32 (2.19)	[0.015]
CEO Pay × Hierarchical Layers					0.49 (0.17)	[0.004]					1.09 (0.34)	[0.002]
CEO Pay × Political Competition							3.46 (0.82)	[0.000]			5.72 (0.89)	[0.000]
CEO Pay × Nonstate
Second-largest owner									3.09 (0.67)	[0.000]	4.36 (1.70)	[0.010]
Hierarchical layers	1.26 (0.28)	[0.000]	1.42 (0.31)	[0.000]	1.32 (0.34)	[0.000]	1.14 (0.31)	[0.000]	1.35 (0.31)	[0.000]	0.91 (0.36)	[0.012]
Political competition	0.61 (0.35)	[0.083]	0.08 (0.42)	[0.841]	−0.10 (0.44)	[0.822]	0.16 (0.39)	[0.677]	−0.01 (0.43)	[0.984]	−0.02 (0.44)	[0.957]
Nonstate second-largest owner	1.39 (0.47)	[0.003]	0.72 (0.48)	[0.131]	0.54 (0.58)	[0.353]	1.41 (0.47)	[0.003]	−0.63 (0.99)	[0.523]	1.13 (1.19)	[0.345]
CEO age	0.02 (0.02)	[0.223]	0.02 (0.04)	[0.499]	0.04 (0.04)	[0.358]	0.01 (0.04)	[0.748]	0.08 (0.04)	[0.031]	0.21 (0.05)	[0.000]
CEO gender	−3.31 (0.94)	[0.000]	0.55 (1.50)	[0.715]	1.29 (1.47)	[0.381]	−1.26 (1.56)	[0.417]	1.78 (1.46)	[0.223]	0.20 (1.42)	[0.887]
CEO education	0.43 (0.23)	[0.063]	1.17 (0.30)	[0.000]	1.17 (0.31)	[0.000]	1.08 (0.28)	[0.000]	0.87 (0.29)	[0.003]	0.98 (0.31)	[0.002]
CEO tenure	−0.19 (0.08)	[0.023]	0.01 (0.08)	[0.896]	0.07 (0.09)	[0.473]	−0.05 (0.09)	[0.561]	−0.07 (0.10)	[0.484]	0.10 (0.09)	[0.303]
CEO duality	−0.92 (0.28)	[0.001]	−0.47 (0.36)	[0.197]	−0.46 (0.35)	[0.190]	−0.38 (0.35)	[0.278]	−0.56 (0.36)	[0.120]	0.10 (0.32)	[0.748]
Firm size	4.33 (0.42)	[0.000]	3.38 (0.53)	[0.000]	3.48 (0.53)	[0.000]	2.97 (0.51)	[0.000]	3.20 (0.53)	[0.000]	1.01 (0.59)	[0.090]
Firm age	−0.41 (0.04)	[0.000]	−0.55 (0.05)	[0.000]	−0.55 (0.05)	[0.000]	−0.51 (0.05)	[0.000]	−0.56 (0.05)	[0.000]	−0.55 (0.05)	[0.000]
Debt	−3.25 (0.38)	[0.000]	−2.96 (0.49)	[0.000]	−2.99 (0.50)	[0.000]	−2.67 (0.46)	[0.000]	−2.61 (0.51)	[0.000]	−1.38 (0.51)	[0.006]
Central state ownership	−0.03 (0.01)	[0.056]	−0.04 (0.02)	[0.003]	−0.04 (0.02)	[0.004]	−0.04 (0.02)	[0.018]	−0.04 (0.02)	[0.005]	−0.03 (0.02)	[0.106]
Board independence	−0.08 (0.02)	[0.000]	−0.05 (0.02)	[0.024]	−0.04 (0.02)	[0.063]	−0.05 (0.02)	[0.019]	−0.04 (0.02)	[0.066]	−0.06 (0.02)	[0.006]
Job creation	0.00 (0.00)	[0.000]	0.00 (0.00)	[0.343]	0.00 (0.00)	[0.791]	0.00 (0.00)	[0.056]	0.00 (0.00)	[0.277]	0.00 (0.00)	[0.000]
Local GDP	−0.54 (0.27)	[0.044]	−2.27 (0.36)	[0.000]	−2.43 (0.38)	[0.000]	−1.83 (0.35)	[0.000]	−2.30 (0.37)	[0.000]	−2.54 (0.35)	[0.000]
Local unemployment rate	−0.32 (0.17)	[0.060]	0.02 (0.21)	[0.906]	0.07 (0.22)	[0.740]	0.10 (0.21)	[0.630]	0.04 (0.22)	[0.840]	0.26 (0.22)	[0.231]
Constant	−13.79 (3.13)	[0.000]	−36.93 (4.75)	[0.000]	−16.94 (9.24)	[0.067]	−10.11 (7.56)	[0.181]	−37.06 (4.63)	[0.000]	56.38 (16.85)	[0.001]
Number of instruments	172		161		161		161		161		161
Wald χ²	863.56	[0.000]	676.42	[0.000]	583.23	[0.000]	736.59	[0.000]	691.92	[0.000]	788.17	[0.000]
AR(1) test (p value)	(0.052)		(0.011)		(0.009)		(0.018)		(0.012)		(0.008)
AR(2) test (p value)	(0.484)		(0.333)		(0.316)		(0.345)		(0.307)		(0.228)
Hansen-Overid (p value)	(0.218)		(0.211)		(0.174)		(0.307)		(0.274)		(0.568)
Diff-in-Hansen Exogeneity (p value)	(0.414)		(0.510)		(0.347)		(0.672)		(0.572)		(0.608)

Note. 1,565 observations on 230 firms. Standard errors are in parentheses. p values are in square brackets (two-tailed tests).

In Table 2, Model 1 shows the control variables only; Model 2 enters CEO pay to test H1a; Models 3 to 5 enter the three moderating variables to test H2a, H3a, and H4a, respectively; and Model 6 presents the full model results. H1a predicts a positive relationship between CEO pay and financial performance. Model 2 shows that CEO pay is positively correlated with the ROA (b = 3.64, SE = 0.29, p = 0.000, CI_95% = [3.08, 4.21]), which supports H1a. H2a posits that hierarchical layers strengthen the positive relationship between CEO pay and ROA. Model 3 shows that CEO pay and hierarchical layers have a positive interaction effect (b = 0.49, SE = 0.17, p = 0.004, CI_95% = [0.16, 0.83]), supporting H2a. H3a posits that political competition strengthens the positive relationship between pay and performance. Model 4 shows that CEO pay and the number of affiliates have a positive interaction effect (b = 3.46, SE = 0.82, p = 0.000, CI_95% = [1.86, 5.06]), supporting H3a. H4a posits that having a nonstate second-largest shareholder strengthens the positive relationship between pay and performance. Model 5 shows that CEO pay and nonstate second-largest owner have a positive interaction effect (b = 3.09, SE = 0.67, p = 0.000, CI_95% = [1.78, 4.40]), thus supporting H4a. These results remain consistent in the full model (Model 6).

Table 3 presents the GMM results for political performance in a similar pattern to that shown in Table 2. Model 2 shows that CEO pay has a positive and significant effect on job creation (b = 1,159.79, SE = 205.88, p = 0.000, CI_95% = [756.27, 1,563.31]), which supports H1b. We predicted that this relationship would be weakened, rather than strengthened, by the three moderators. In Model 3, the interaction between CEO pay and hierarchical layers is not significant (p = 0.355). Thus, H2b is not supported. It seems that the CEO pay–political performance relationship remains unchanged across hierarchical levels of the firms, although the CEO pay–financial performance relationship is strengthened with more hierarchical layers. In Model 4, CEO pay and political competition have a negative interaction effect (b = −1,274.88, SE = 403.18, p = 0.002, CI_95% = [−2,065.10, −484.65]), supporting H3b. Similarly, in Model 5, the interaction between CEO pay and nonstate second-largest owner is negative and significant (b = −313.59, SE = 51.18, p = 0.000, CI_95% = [−413.89, −213.29]), supporting H4b. The results in the full model (Model 6) also provide support for H1b, H3b, and H4b, but not for H2b.

Table 3

Arellano–Bond GMM Models for Firm Political Performance (Job Creation)

	Model 1		Model 2		Model 3		Model 4		Model 5		Model 6
CEO pay			1,159.79 (205.88)	[0.000]	985.29 (321.74)	[0.002]	1,440.35 (296.68)	[0.000]	1,006.89 (197.94)	[0.000]	1,632.75 (453.32)	[0.000]
CEO Pay × Hierarchical Layers					56.23 (60.85)	[0.355]					−62.28 (80.17)	[0.437]
CEO Pay × Political Competition							−1,274.88 (403.18)	[0.002]			−1,852.66 (438.87)	[0.000]
CEO Pay × Nonstate
Second-Largest Owner									−313.59 (51.18)	[0.000]	−272.21 (70.05)	[0.000]
Hierarchical layers	303.99 (89.30)	[0.001]	−260.66 (111.34)	[0.019]	−284.58 (120.24)	[0.018]	−88.73 (119.32)	[0.457]	−416.22 (123.45)	[0.001]	−195.96 (138.50)	[0.157]
Political competition	−606.31 (180.97)	[0.001]	−440.98 (196.44)	[0.025]	−501.08 (194.34)	[0.010]	−480.32 (236.80)	[0.043]	−361.23 (223.29)	[0.106]	−349.24 (251.26)	[0.165]
Nonstate second-largest owner	85.29 (174.70)	[0.625]	74.18 (270.05)	[0.784]	20.23 (276.61)	[0.942]	−174.90 (297.03)	[0.556]	2,154.52 (576.77)	[0.000]	1,399.38 (693.97)	[0.044]
CEO age	−61.92 (4.46)	[0.000]	−45.39 (10.33)	[0.000]	−47.82 (10.69)	[0.000]	−52.25 (12.57)	[0.000]	−52.15 (10.95)	[0.000]	−56.54 (12.93)	[0.000]
CEO gender	588.62 (130.84)	[0.000]	312.04 (216.37)	[0.149]	284.46 (221.88)	[0.200]	523.01 (272.69)	[0.055]	299.48 (222.87)	[0.179]	539.99 (324.28)	[0.096]
CEO education	323.21 (93.59)	[0.001]	−64.52 (128.57)	[0.616]	−120.69 (139.65)	[0.387]	−151.41 (138.26)	[0.273]	94.88 (123.77)	[0.443]	−28.78 (154.22)	[0.852]
CEO tenure	64.08 (31.00)	[0.039]	406.93 (66.84)	[0.000]	424.83 (67.50)	[0.000]	342.28 (76.25)	[0.000]	414.62 (68.92)	[0.000]	352.23 (81.28)	[0.000]
CEO duality	2,971.34 (344.94)	[0.000]	2,585.17 (390.28)	[0.000]	2,552.28 (410.04)	[0.000]	2,358.65 (403.95)	[0.000]	2,546.01 (384.82)	[0.000]	2,489.13 (418.19)	[0.000]
Firm size	449.40 (186.69)	[0.016]	−746.40 (271.98)	[0.006]	−785.04 (277.94)	[0.005]	−582.41 (330.23)	[0.078]	−323.06 (291.01)	[0.267]	−115.06 (358.89)	[0.749]
Firm age	−115.78 (22.98)	[0.000]	−50.10 (26.60)	[0.060]	−54.21 (27.62)	[0.050]	−74.16 (31.92)	[0.020]	−47.00 (28.96)	[0.105]	−104.66 (34.19)	[0.002]
Debt	202.77 (159.71)	[0.204]	862.16 (237.42)	[0.000]	907.01 (240.67)	[0.000]	813.41 (295.23)	[0.006]	648.83 (244.72)	[0.008]	577.23 (308.22)	[0.061]
Central state ownership	−44.23 (4.02)	[0.000]	−40.19 (6.87)	[0.000]	−41.06 (6.99)	[0.000]	−47.56 (7.84)	[0.000]	−56.62 (7.18)	[0.000]	−64.59 (8.82)	[0.000]
Board independence	92.36 (11.85)	[0.000]	67.53 (13.32)	[0.000]	69.54 (13.53)	[0.000]	67.70 (14.06)	[0.000]	56.86 (13.17)	[0.000]	55.37 (15.28)	[0.000]
ROA	1,798.86 (988.34)	[0.069]	2,004.64 (1,438.92)	[0.164]	2,203.45 (1,486.80)	[0.138]	2,944.42 (1,593.55)	[0.065]	2,124.39 (1,418.80)	[0.134]	3,909.74 (1,669.80)	[0.019]
Local GDP	−916.28 (144.85)	[0.000]	−1,380.66 (210.13)	[0.000]	−1,380.45 (221.85)	[0.000]	−1,187.98 (243.05)	[0.000]	−1,382.43 (223.76)	[0.000]	−959.50 (310.92)	[0.002]
Local unemployment rate	−954.71 (87.46)	[0.000]	−756.91 (147.43)	[0.000]	−738.85 (150.35)	[0.000]	−853.07 (145.31)	[0.000]	−709.15 (142.04)	[0.000]	−802.54 (155.60)	[0.000]
Constant	−7,833.58 (1,623.62)	[0.000]	−4,555.57 (1,923.97)	[0.018]	−1,957.17 (3,624.37)	[0.589]	−11,167.74 (3,809.55)	[0.003]	−6,146.21 (2,086.14)	[0.003]	−20,077.30 (5,899.23)	[0.001]
Number of instruments	177		166		166		166		166		166
Wald χ²	698.88	[0.000]	284.95	[0.000]	290.88	[0.000]	265.13	[0.000]	386.85	[0.000]	314.28	[0.000]
AR(1) test (p value)	(0.003)		(0.003)		(0.003)		(0.003)		(0.003)		(0.003)
AR(2) test (p value)	(0.458)		(0.682)		(0.686)		(0.671)		(0.687)		(0.677)
Hansen-Overid (p value)	(0.530)		(0.989)		(0.987)		(0.998)		(0.913)		(0.976)
Diff-in-Hansen Exogeneity (p value)	(0.238)		(0.936)		(0.933)		(0.993)		(0.614)		(0.896)

Note. 1,565 observations on 230 firms. Standard errors are in parentheses. p values are in square brackets (two-tailed tests).

Robustness Tests and Additional Analyses

Using alternative variables and models

Table 4 reports a number of robustness tests using alternative variables and methods. First, following prior studies, we used alternative dependent variables to reestimate all the models. We used the return on equity (ROE) measured as the net income over total equity adjusted by the industry mean as the alternative dependent variable for the firm financial performance (Peng, Sun, & Markóczy, 2015). Model 1 of Table 4 shows the full model results using the Arellano–Bond GMM model. CEO pay and its three interaction terms are all significant (p = 0.000, p = 0.004, p = 0.000, and p = 0.004, respectively) and properly signed, supporting H1a, H2a, H3a, and H4a.

Table 4

Robustness Tests for Firm Financial and Political Performances

	Model 1		Model 2		Model 3		Model 4
	ROE		CSR		ROA		Job Creation
CEO pay	17.20 (4.86)	[0.000]	3.20 (1.37)	[0.019]	2.01 (2.30)	[0.383]	7,377.99 (990.31)	[0.000]
CEO Pay × Hierarchical Layers	1.07 (0.37)	[0.004]	−0.48 (0.21)	[0.022]	1.23 (0.43)	[0.004]	−441.52 (101.56)	[0.000]
CEO Pay × Political Competition	31.59 (5.13)	[0.000]	−5.62 (1.85)	[0.002]
CEO Pay × Political Competition in Industry					11.24 (2.98)	[0.000]	−13,548.01 (1,714.69)	[0.000]
CEO Pay × Nonstate Second-Largest Owner	13.98 (4.86)	[0.004]	0.43 (0.33)	[0.187]	1.67 (1.37)	[0.220]	−244.31 (83.56)	[0.003]
Hierarchical layers	2.99 (0.88)	[0.001]	−0.02 (0.15)	[0.901]	2.27 (0.31)	[0.000]	−300.20 (157.93)	[0.057]
Political competition	4.14 (1.42)	[0.004]	0.22 (0.25)	[0.385]
Political competition in industry					−0.03 (0.01)	[0.000]	3.37 (2.84)	[0.237]
Nonstate second-largest owner	−12.78 (4.09)	[0.002]	−5.87 (4.41)	[0.183]	1.14 (1.80)	[0.526]	1,560.04 (680.31)	[0.022]
CEO age	0.68 (0.13)	[0.000]	0.01 (0.02)	[0.620]	0.10 (0.05)	[0.037]	−74.72 (13.81)	[0.000]
CEO gender	−11.15 (4.92)	[0.024]	1.31 (0.67)	[0.050]	−0.44 (1.44)	[0.758]	−276.69 (321.78)	[0.390]
CEO education	0.73 (0.83)	[0.381]	−0.44 (0.28)	[0.113]	1.14 (0.31)	[0.000]	−98.37 (158.68)	[0.535]
CEO tenure	−0.60 (0.28)	[0.034]	−0.18 (0.12)	[0.140]	−0.02 (0.10)	[0.880]	423.59 (85.48)	[0.000]
CEO duality	−0.20 (0.98)	[0.841]	0.56 (0.42)	[0.186]	0.26 (0.32)	[0.418]	2,708.25 (391.61)	[0.000]
Firm size	1.51 (1.40)	[0.282]	0.24 (0.38)	[0.521]	1.66 (0.60)	[0.006]	−872.63 (392.95)	[0.026]
Firm age	−2.07 (0.15)	[0.000]	−0.04 (0.03)	[0.218]	−0.49 (0.06)	[0.000]	−1.58 (34.14)	[0.963]
Debt	−0.62 (1.32)	[0.639]	−0.19 (0.31)	[0.554]	−1.89 (0.54)	[0.001]	1,289.18 (321.70)	[0.000]
Central state ownership	−0.19 (0.04)	[0.000]	−0.00 (0.01)	[0.729]	−0.04 (0.02)	[0.028]	−31.94 (10.15)	[0.002]
Board independence	−0.65 (0.09)	[0.000]	0.01 (0.02)	[0.648]	−0.05 (0.02)	[0.019]	94.30 (17.29)	[0.000]
Job creation	0.00 (0.00)	[0.002]			0.00 (0.00)	[0.001]
ROA			−3.64 (2.04)	[0.074]			6,110.79 (1,642.49)	[0.000]
Local GDP	−5.19 (1.09)	[0.000]	−1.10 (0.45)	[0.014]	−2.59 (0.33)	[0.000]	−979.43 (258.68)	[0.000]
Local unemployment rate	1.15 (0.43)	[0.007]	0.50 (0.42)	[0.242]	0.25 (0.20)	[0.224]	−705.63 (156.23)	[0.000]
Constant	114.35 (40.71)	[0.005]	−31.09 (18.70)	[0.096]	93.25 (33.57)	[0.005]	−96,092.32 (12,453.23)	[0.000]
Number of instruments	162				162		166
Wald χ²	4,364.43	[0.000]	102.70	[0.000]	1,296.04	[0.000]	288.62	[0.000]
AR(1) test (p value)	(0.009)				(0.013)		(0.003)
AR(2) test (p value)	(0.179)				(0.284)		(0.678)
Hansen-Overid (p value)	(0.168)				(0.138)		(0.981)
Diff-in-Hansen Exogeneity (p value)	(0.115)				(0.129)		(0.855)
N	1,494		1,713		1,499		1,498

Note. N = 230 firms. Standard errors are in parentheses. p values are in square brackets (two-tailed tests). ROE = return on equity; CSR = corporate social responsibility; ROA = return on assets.

Next, we used a firm’s CSR engagement as an alternative variable for political performance. The Chinese government seeks to promote and improve CSR because they view CSR as an important aspect of building a “harmonious society” and sustaining social and economic development (Gao, 2011; See, 2009: 1). Thus, SOEs have an incentive to engage in CSR as part of a political agenda (Raynard, Lounsbury, & Greenwood, 2013). As CSR is encouraged but not mandatory, we used a binary variable coded 1 if the focal firm engaged in any CSR activities in a given year and coded 0 otherwise. The data were obtained from CSR reporting in the CSMAR-CSR database (Wang, Cao, & Ye, 2018). Upon running the panel logit model, the results in the full model (Model 2) show that CEO pay has a positive effect, and its interactions with hierarchical layers and political competition have negative effects, on CSR engagement (p = 0.019, p = 0.022, and p = 0.002, respectively), thus supporting H1b, H2b, and H3b. However, the interaction between CEO pay and nonstate second-largest owner in Model 2 is not statistically significant (p = 0.187), lending no support for H4b.

Third, we used an alternative variable to measure political competition. As the current measurement focuses on political competition within the enterprise group in the near future, we developed a new variable to measure present political competition in industry. That is, we expanded the focal executive’s reference group of political competitors to include executives in other SOE groups in the same industry by counting the number of all central government-controlled listed firms at the same hierarchical level as the focal firm. This alternative variable enables us to capture the industry-wide political competition at the same level as the CEO of the focal firm. The interaction of CEO pay and political competition in industry is positive in Model 3 (p = 0.000) and negative in Model 4 (p = 0.000), suggesting the alternative political competition measure has the same moderating effects as the original measure.

Additional analyses of the impact of other large shareholders

In the main analyses for H4a and H4b, we focused on the existence of a nonstate second-largest owner with at least 5% ownership. In our additional analyses, we developed several other measures to capture the impact of other large shareholders more fully. First, we used a binary variable, other large owner, which is coded 1 if the focal firm has at least one large shareholder with a 5% stake or more, other than the central government, and coded 0 otherwise. These large shareholders include nonstate owners as well as local governments and local SOEs. The results for this variable are reported in Models 1 and 2 of Table 5. The significant interactions found in both models (p = 0.037 and p = 0.029, respectively) suggest that when there is a large shareholder other than the central government itself, CEO pay becomes a more effective mechanism for improving the financial performance but reducing the political performance.

Table 5

Additional Analyses of the Impact of Other Large Shareholders

	Model 1		Model 2		Model 3		Model 4		Model 5		Model 6
	ROA		Job Creation		ROA		Job Creation		ROA		Job Creation
CEO pay	7.69 (2.15)	[0.000]	8,079.67 (980.66)	[0.000]	−1.19 (1.36)	[0.382]	5,717.70 (880.76)	[0.000]	7.00 (1.28)	[0.000]	7,856.35 (997.96)	[0.000]
CEO Pay × Hierarchical Layers	1.02 (0.33)	[0.002]	−431.77 (100.10)	[0.000]	0.82 (0.33)	[0.012]	−234.69 (89.86)	[0.009]	0.91 (0.21)	[0.000]	−454.13 (112.38)	[0.000]
CEO Pay × Political Competition	5.65 (0.86)	[0.000]	−14,822.80 (1,620.00)	[0.000]	4.28 (0.62)	[0.000]	−11,636.66 (1,536.11)	[0.000]	6.71 (0.56)	[0.000]	−14,439.95 (1,629.53)	[0.000]
CEO Pay × Other Large Owner	3.82 (1.83)	[0.037]	−108.18 (49.60)	[0.029]
CEO Pay × Nonstate Second-Largest Equity					0.22 (0.10)	[0.024]	−10.56 (15.82)	[0.504]
CEO Pay × Number of Other Large Owners									0.91 (0.30)	[0.002]	279.79 (221.82)	[0.207]
Hierarchical layers	0.85 (0.36)	[0.018]	−80.24 (138.10)	[0.561]	0.98 (0.36)	[0.007]	−312.78 (138.87)	[0.024]	1.11 (0.20)	[0.000]	−221.39 (149.89)	[0.140]
Political competition	0.18 (0.44)	[0.684]	8.32 (3.00)	[0.006]	0.02 (0.42)	[0.961]	5.35 (2.56)	[0.037]	0.01 (0.30)	[0.972]	6.30 (2.93)	[0.031]
Other large owner	−54.31 (25.03)	[0.030]	28.38 (312.02)	[0.928]
Nonstate second-largest equity					−3.16 (1.33)	[0.017]	87.18 (207.61)	[0.675]
Number of other larger owners									−9.48 (3.95)	[0.016]	−4,254.02 (2,948.02)	[0.149]
CEO age	0.20 (0.05)	[0.000]	−50.77 (13.34)	[0.000]	0.14 (0.04)	[0.001]	−64.24 (13.06)	[0.000]	0.20 (0.03)	[0.000]	−51.02 (13.28)	[0.000]
CEO gender	−0.53 (1.48)	[0.719]	−185.50 (335.34)	[0.580]	−0.52 (1.46)	[0.720]	25.37 (256.63)	[0.921]	3.17 (1.03)	[0.002]	−324.08 (335.33)	[0.334]
CEO education	1.22 (0.31)	[0.000]	−65.83 (151.95)	[0.665]	0.95 (0.27)	[0.000]	−238.94 (147.87)	[0.106]	0.79 (0.19)	[0.000]	−103.85 (153.85)	[0.500]
CEO tenure	0.14 (0.09)	[0.115]	321.05 (79.17)	[0.000]	0.15 (0.10)	[0.123]	370.59 (72.70)	[0.000]	0.21 (0.07)	[0.003]	375.99 (75.86)	[0.000]
CEO duality	0.30 (0.33)	[0.361]	2,878.36 (380.56)	[0.000]	0.45 (0.33)	[0.172]	2,852.21 (369.55)	[0.000]	−0.27 (0.25)	[0.274]	2,572.41 (377.59)	[0.000]
Firm size	1.03 (0.57)	[0.070]	−1,316.48 (413.84)	[0.001]	1.67 (0.53)	[0.002]	−1,001.84 (368.12)	[0.006]	0.10 (0.39)	[0.793]	−1,210.15 (416.73)	[0.004]
Firm age	−0.57 (0.05)	[0.000]	−5.05 (33.02)	[0.878]	−0.55 (0.05)	[0.000]	−39.04 (33.86)	[0.249]	−0.52 (0.04)	[0.000]	3.60 (32.96)	[0.913]
Debt	−1.57 (0.47)	[0.001]	1,600.24 (339.23)	[0.000]	−1.91 (0.47)	[0.000]	1,409.13 (304.26)	[0.000]	−0.60 (0.35)	[0.086]	1,548.46 (350.66)	[0.000]
Central state ownership	−0.03 (0.02)	[0.031]	−20.44 (9.66)	[0.034]	−0.04 (0.01)	[0.003]	−26.16 (8.38)	[0.002]	0.01 (0.01)	[0.317]	−20.02 (9.77)	[0.040]
Board independence	−0.08 (0.02)	[0.001]	98.43 (15.91)	[0.000]	−0.07 (0.02)	[0.001]	97.20 (14.87)	[0.000]	−0.06 (0.02)	[0.001]	93.78 (16.58)	[0.000]
Job creation	0.00 (0.00)	[0.000]			0.00 (0.00)	[0.002]			0.00 (0.00)	[0.000]
ROA			4,494.23 (1,613.18)	[0.005]			5,743.56 (1,521.76)	[0.000]			5,141.76 (1,581.36)	[0.001]
Local GDP	−2.49 (0.34)	[0.000]	−1,392.08 (271.42)	[0.000]	−2.48 (0.35)	[0.000]	−794.33 (266.68)	[0.003]	−1.96 (0.26)	[0.000]	−1,256.36 (276.79)	[0.000]
Local unemployment rate	0.30 (0.22)	[0.178]	−964.18 (159.65)	[0.000]	0.20 (0.21)	[0.357]	−888.57 (141.81)	[0.000]	0.23 (0.17)	[0.158]	−839.28 (156.56)	[0.000]
Constant	51.95 (16.78)	[0.002]	−10,0196.63 (12,105.77)	[0.000]	38.52 (16.63)	[0.021]	−76,038.45 (11,675.47)	[0.000]	52.79 (9.43)	[0.000]	−99,000.31 (12,388.92)	[0.000]
Number of instruments	162		166		162		166		162		166
Wald χ²	904.44	[0.000]	299.22	[0.000]	874.76	[0.000]	333.89	[0.000]	1,065.89	[0.000]	285.80	[0.000]
AR(1) test (p value)	(0.007)		(0.003)		(0.011)		(0.003)		(0.012)		(0.002)
AR(2) test (p value)	(0.256)		(0.670)		(0.262)		(0.682)		(0.234)		(0.673)
Hansen-Overid (p value)	(0.502)		(0.987)		(0.425)		(0.961)		(0.392)		(0.982)
Diff-in-Hansen Exogeneity (p value)	(0.529)		(0.904)		(0.367)		(0.827)		(0.739)		(0.900)
N	1499		1498		1499		1498		1499		1498

Note. N = 230 firms. Standard errors are in parentheses. p values are in square brackets (two-tailed tests).

Second, we used nonstate second-largest equity, which is the shareholding percentage of the nonstate second-largest owner, as an alternative variable. We removed the 5% ownership threshold to include all nonstate second-largest owners in each firm, which increases the total number of firm-year entries with a nonstate second-largest shareholder from 86 to 94. The results of this analysis are reported in Models 3 and 4 of Table 5. The significant interaction in Model 3 (p = 0.024) indicates that the shareholding of a nonstate second-largest owner positively moderates the relationship between CEO pay and financial performance.

Third, we calculated the number of other large owners, which is the total number of all other large shareholders (outside the SOE group but within the ownership chart) of each firm. For example, Figure 1b shows that Jiangling Motors has two other large owners (i.e., Ford and Nanchang Government). The logic here is that a greater number of large shareholders (who have a significant stake of at least 5%) other than the central government may provide an incentive for executives to focus more on a generally accepted performance criterion—namely, firm profit. The average number of other large owners in our sample is 0.15, since most of the firms do not have any other large shareholders. The results for this variable are reported in Models 5 and 6 of Table 5. The significant interaction in Model 5 (p = 0.002) indicates that having a greater number of other large shareholders enhances the positive effect of CEO pay on financial performance.

Discussion and Conclusions

In this study, we examine the effects of CEO pay on the financial and political performances of Chinese state-controlled listed firms and how these effects are influenced by the dual career considerations of top executives. Based on both agency theory and the PPC perspective, we find that, on average, a higher level of executive compensation leads to both higher ROA and more job creation. Furthermore, the CEO pay–financial performance relationship is strengthened when the firm is at a lower hierarchical level in the enterprise group, when there is more political competition in the group, or when the second-largest shareholder is a nonstate owner. However, the CEO pay–political performance relationship is weakened when the firm has more political competition in the enterprise group or when a nonstate owner is the second-largest shareholder of the firm.

Contributions

Our study makes two major contributions to theory and research. First, we provide a closer integration of agency theory and the PPC perspective by highlighting the role of the agent in a PPC context. Extant PPC literature has mainly covered two scenarios of PPC: (1) PPC between minority shareholders and a dominant family who controls and manages the firm (Chang, 2003; Faccio et al., 2001), and (2) PPC between minority shareholders and the state as the controlling owner (Li & Qian, 2013; Su et al., 2008). In the former case, managerial self-interest is assumed away, as the firm is managed by the family itself. In the latter case, managerial discretion is assumed away because managers are considered political appointees and have limited autonomy. In contrast, the agents and their career interests take center stage in our study. We establish that SOE executives have both self-interest in terms of their dual career paths and managerial autonomy as a result of SOE reform to make decisions that differentially impact the two classes of firm owners. In our framework, the agents may conform to the state’s demands at the cost of minority shareholders’ interest, or they may choose to side with the minority shareholders and disregard the state’s political interest. The conflict between controlling owners and minority shareholders occurs through the decisions of self-interested managers. Thus, we have integrated both PACs and PPCs in a unifying framework to advance our understanding of more complex agency problems in a transition economy.

A relatively small but emerging stream of research addresses multiple agency relationships in firms (Arthurs, Hoskisson, Busenitz, & Johnson, 2008). So far, research on this topic has a limited scope and mainly involves the dual agent-owner roles of venture capitalists and investment bankers, typically in the context of dispersed ownership and mild conflicts between institutional investors in developed markets (Arthurs et al., 2008; Connelly, Tihanyi, Certo, & Hitt, 2010; Hoskisson, Hitt, Johnson, & Grossman, 2002). In contrast, our scenario is one with majority shareholders controlling firm outcomes and expropriating minority shareholder interests in an environment of weak legal protection for minority owners (Dharwadkar et al., 2000; Young et al., 2008). The agent in our study is faced with a trade-off between two career interests that are associated with the two classes of owners, respectively. In a way, our study also constitutes an enrichment and extension of multiple agency theory.

Second, we contribute to research on Chinese SOEs by examining how managers’ dual career paths impact their behavior and performance. Extant research on Chinese firms often views SOEs as backward, inefficient, poorly governed, and lacking autonomy (Allen et al., 2005; Morck, Yeung, & Zhao, 2008; Park, Li, & Tse, 2005; Zhou et al., 2017). As such, researchers are unable to explain why, despite the decreasing number of small- and medium-sized SOEs, the larger SOEs are becoming even larger and more powerful (Meyer, 2011; Xu et al., 2014), other than attributing such phenomena to various kinds of government support, such as better access to information, financing, networks, and legal protection (Morck et al., 2008; Peng & Luo, 2000; Xu, Zhou, & Phan, 2010). However, government support for SOEs alone would not have sustained China’s strong, continuous economic growth. Our study sheds light on the SOE myth and ties back to agency theory and the PPC perspective in several ways.

First, we provide contextualized theorizing about SOE managers’ dual career interests by relating these career interests to the structural positions of their firms in the enterprise groups in terms of both vertical (number of layers) and horizontal (number of affiliates at a certain level) features. The double identities of SOE managers are often overlooked in prior research, as SOE managers are either regarded as Party cadres who are exclusively pursuing a political career or are treated as average managers engaging in a business career (Holz, 2007). Researchers who emphasize the transitional or hybrid nature of the Chinese economy are attracted by other mixed-ownership firms, such as Township-Village Enterprises (Che & Qian, 1998) and Collectively Owned Enterprises (Xu et al., 2014), but rarely pay attention to how SOEs and their top managers cope with a mixed economy. Our emphasis on the managers’ dual career paths offers a new angle for examining their motivation and behavior.

Second, our focus on other large shareholders reveals a more complex ownership structure than what is typically shown in prior SOE research. Agency theory has traditionally addressed the monitoring function of large shareholders in the context of relatively dispersed ownership in the West (Shleifer & Vishny, 1986). The PPC literature, in contrast, highlights the scenario of a dominant, controlling state owner versus small minority shareholders (Dharwadkar et al., 2000; Laeven & Levine, 2008). By highlighting the role of other large shareholders, we reveal the complexity of PPCs. On the one hand, a divergence of interests may occur among these shareholders. On the other hand, some of them may share common interests among themselves against the controlling owner. The existence of a nonstate second-largest owner, in particular, provides an incentive for SOE executives to maximize profit and downplay the government’s political goals.

Third, the results of our baseline hypotheses establish a positive effect of executive compensation on the performance of Chinese SOEs. The pay → performance link is underresearched in the agency theory literature compared to the vast number of studies on the performance → pay link. Although a previous study has detected a positive executive pay → financial performance relationship for Chinese listed firms in general (Buck et al., 2008), little research can be found on such a relationship for SOEs. We have provided corroborating evidence for the effectiveness of the market in disciplining SOE managers, conjectured by other researchers (Allen et al., 2005; Groves et al., 1995) and have detected a positive relationship between executive pay and firms’ political performance, which is rarely covered in previous research.

Limitations and Future Research

A major limitation of our study is that, to some extent, our theorizing relies on the unique context of China’s mixed economy and Chinese SOEs. Today, China is the only major economy with a ruling Communist Party, and this limits our generalizability in a way. However, we argue that the sheer size of the Chinese economy and the dominant role that SOEs play in that economy are grounds for studying the topic. If we do not build on their uniqueness, we could perhaps never fully understand Chinese SOEs. Moreover, we can easily foresee the relevance of our theoretical framework in other contexts. One potential application of our theory is in research on business groups across the world. This research suggests that controlling families have created complex group structures to ensure the private benefits of control and conceal money tunneling (Chang, 2003; Johnson, La Porta, Lopez-de-Silanes, & Shleifer, 2000). Because the size of a core family is limited, however, executive positions in the firms at the bottom of the group pyramid are likely to be filled by noncore family members who have separate interests, and they may likely diverge from the interests of the controlling family in an attempt to establish their own career reputation as qualified managers. Another possible application of our dual career path analysis may be in research on family businesses with second-generation family owners, who face the choice of taking over their parents’ role in the firm or developing their own career interests (Le Breton-Miller, Miller, & Lester, 2011; Stavrou & Swiercz, 1998). Future research can therefore test our theory in these different contexts to better assess the generalizability of our framework.

A second limitation of our study is its reliance on job creation (and CSR engagement in the robustness tests) as the proxy measure of political performance. Whereas the ROA is a generally accepted, standard measure of financial performance, the decision to increase hiring within a firm may be influenced by both business and political considerations. We believe that the set of control variables that we included—firm profitability (ROA), size, leverage, local unemployment rate, and others—have captured the major business factors; therefore, CEO pay and its interactions with the moderating variables provide additional explanations for job creation, as we outline in the Hypotheses section. Future research can further pursue this theoretical angle by employing other measures of political performance.

A third limitation of this study is its lack of more detailed analyses of the roles of specific types of other large shareholders such as local governments, banks, and other institutional investors. Different types of large shareholders may behave differently in ways for which our various measures in the main and additional analyses cannot account. In particular, local interests and competition among regional governments may incentivize economic growth (Jin, Qian, & Weingast, 2005 Walder, 1995b; Xu, 2011). Thus, regional governments have a stronger interest in seeing high profits achieved by the SOEs in which they have shares compared to the central government. On the other hand, local governments, just like the central government, are concerned with local employment. Our study, including the additional analyses, has not isolated the impacts of local government ownership. Future research with a greater focus on ownership categories could conduct more thorough investigations into the impacts of local governments and institutions.

Our results differ from those of a previous study on Chinese state-controlled listed firms. Leung and Cheng (2013) find that top-executive remuneration does not affect the firm value in central state-controlled listed firms. However, their overall message is consistent with ours. They find that executive pay does increase the financial performance in firms controlled by local governments or nonstate owners as well as in a whole sample combining all central, local, and nonstate firms. Through additional analyses, we also find that local governments and nonstate owners can enhance the positive relationship between CEO pay and firm financial performance. The different main effects of CEO pay on the performance of the central government-controlled firms between their study and our study are probably due to the differences in the sample, timeframe, and measurement, including direct versus ultimate shareholding when determining ownership. Further research is needed to improve our understanding of this issue.

Conclusions

CEO pay has a positive impact on both the financial and political performances of state-controlled SOEs in China, suggesting that a higher compensation level motivates Chinese executives to fulfill shareholder objectives. Nevertheless, due to the existence of two career tracks—business and political—for SOE managers, the interests of these managers are differentially aligned with those of the Party-state and minority shareholders. Structural factors—namely, the hierarchical position of the firm within the enterprise group, the extent of competition on the CEO’s political ladder, and the existence of a nonstate second-largest shareholder—are in general effective moderators that change the strengths of the CEO pay → financial and political performance relationships. Thus, our study integrates theories of both PACs and PPCs in a unifying framework to account for the new phenomena of a mixed economy.

Footnotes

Acknowledgements

Both authors contributed equally to this article. We thank Associate Editor William Wan and three anonymous reviewers, including a JOM methods reviewer, for their helpful comments and thoughtful suggestions. We also thank Gary Biddle, Don Lange, and Laszlo Tihanyi for their valuable comments on earlier versions.

ORCID iDs

Helen Wei Hu

Dean Xu

References

Aguilera

Duran

Heugens

Sauerwald

Turturea

VanEssen

2020. State ownership, political ideology, and firm performance around the world. Journal of World Business. Advance online publication. doi:10.1016/j.jwb.2020.101113

Aiken

L. S.

West

S. G.

1991. Multiple regression: Testing and interpreting interactions. Thousand Oaks, CA: SAGE.

Allen

Qian

2005. Law, finance, and economic growth in China. Journal of Financial Economics, 77: 57-116.

Amihud

Lev

1999. Does corporate ownership structure affect its strategy towards diversification? Strategic Management Journal, 20: 1063-1069.

Arellano

2003. Panel data econometrics. Oxford, UK: Oxford University Press.

Arellano

Bond

1991. Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations. Review of Economic Studies, 58: 277-297.

Arthurs

J. D.

Hoskisson

R. E.

Busenitz

L. W.

Johnson

R. A.

2008. Managerial agents watching other agents: Multiple agency conflicts regarding underpricing in IPO firms. Academy of Management Journal, 51: 277-294.

Bai

C.- E.

Tao

2006. The multitask theory of state enterprise reform: Empirical evidence from China. American Economic Review, 96: 353-357.

Barkema

H. G.

Gomez-Mejia

L. R.

1998. Managerial compensation and firm performance: A general research framework. Academy of Management Journal, 41: 135-145.

10.

Beatty

R. P.

Zajac

E. J.

1994. Managerial incentives, monitoring, and risk bearing: A study of executive compensation, ownership, and board structure in initial public offerings. Administrative Science Quarterly, 39: 313-335.

11.

Becker

T. E.

Atinc

Breaugh

J. A.

Carlson

K. D.

Edwards

J. R.

Spector

P. E.

2015. Statistical control in correlational studies: 10 essential recommendations for organizational researchers. Journal of Organizational Behavior, 37: 157-167.

12.

Boivie

Lange

McDonald

M. L.

Westphal

J. D.

2011. Me or we: The effects of CEO organizational identification on agency costs. Academy of Management Journal, 54: 551-576.

13.

Brickley

J. A.

Bhagat

Lease

R. C.

1985. The impact of long-range managerial compensation plans on shareholder wealth. Journal of Accounting and Economics, 7: 115-129.

14.

Brødsgaard

K. E.

2012. Politics and business group formation in China: The party in control? China Quarterly, 211: 624-648.

15.

Bruton

G. D.

Peng

M. W.

Ahlstrom

Stan

2015. State-owned enterprises around the world as hybrid organizations. Academy of Management Perspectives, 29: 92-114.

16.

Buck

Liu

Skovoroda

2008. Top executive pay and firm performance in China. Journal of International Business Studies, 39: 833-850.

17.

Cao

2001. Careers inside organizations: A comparative study of promotion determination in reforming China. Social Forces, 80: 683-711.

18.

Carpenter

M. A.

Sanders

W. M. G.

2002. Top management team compensation: The missing link between CEO pay and firm performance? Strategic Management Journal, 23: 367-375.

19.

Chan

H. S.

2004. Cadre personnel management in China: The nomenklatura system, 1990-1998. China Quarterly, 179: 703-734.

20.

Chang

E. C.

Wong

S. M. L.

2009. Governance with multiple objectives: Evidence from top executive turnover in China. Journal of Corporate Finance, 15: 230-244.

21.

Chang

S. J.

2003. Ownership structure, expropriation, and performance of group-affiliated companies in Korea. Academy of Management Journal, 46: 238-253.

22.

Chang

S. J.

2008. Spillovers and competition among foreign and local firms in China. Strategic Management Journal, 29: 495-518.

23.

Che

Qian

1998. Institutional environment, community government, and corporate governance: Understanding China’s township-village enterprises. Journal of Law, Economics, & Organization, 14: 1-23.

24.

Chen

Ham

S. H.

Lim

2011. Designing multiperson tournaments with asymmetric contestants: An experimental study. Management Science, 57: 864-883.

25.

Chen

M. J.

1996. Competitor analysis and interfirm rivalry: Toward a theoretical integration. Academy of Management Review, 21: 100-134.

26.

Chizema

Liu

Gao

2015. Politically connected boards and top executive pay in Chinese listed firms. Strategic Management Journal, 36: 890-906.

27.

Connelly

B. L.

Tihanyi

Certo

S. T.

Hitt

M. A.

2010. Marching to the beat of different drummers: The influence of institutional owners on competitive actions. Academy of Management Journal, 53: 723-742.

28.

Connelly

B. L.

Tihanyi

Crook

T. R.

Gangloff

K. A.

2014. Tournament theory: Thirty years of contests and competitions. Journal of Management, 40: 16-47.

29.

Conyon

M. J.

2011. Executive compensation and corporate governance in China. Journal of Corporate Finance, 17: 1158-1175.

30.

Conyon

M. J.

Peck

S. I.

Sadler

G. V.

2001. Corporate tournaments and executive compensation: Evidence from the U.K. Strategic Management Journal, 22: 805-815.

31.

Cuervo-Cazurra

Inkpen

Musacchio

Ramaswamy

2014. Governments as owners: State-owned multinational companies: Editorial. Journal of International Business Studies, 45: 919-942.

32.

Daily

C. M.

Johnson

J. L.

1997. Sources of CEO power and firm financial performance: A longitudinal assessment. Journal of Management, 23: 117-297.

33.

Delios

Zhou

W. W.

2008. Ownership structure and the diversification and performance of publicly-listed companies in China. Business Horizons, 51: 473-483.

34.

Dewenter

K. L.

Malatesta

P. H.

2001. State-owned and privately owned firms: An empirical analysis of profitability, leverage, and labor intensity. The American Economic Review, 91: 320-334.

35.

Dharwadkar

George

Brandes

2000. Privatization in emerging economies: An agency theory perspective. Academy of Management Review, 25: 650-669.

36.

Dhillon

Rossetto

2015. Ownership structure, voting, and risk. Review of Financial Studies, 28: 521-560.

37.

Tang

Young

S. M.

2012. Influence activities and favoritism in subjective performance evaluation: Evidence from Chinese state-owned enterprises. Accounting Review, 87: 1555-1588.

38.

Eisenhardt

K. M.

1989. Agency theory: An assessment and review. Academy of Management Review, 14: 57-74.

39.

Faccio

Lang

L. H. P.

Young

2001. Dividends and expropriation. American Economic Review, 91: 54-78.

40.

Fama

E. F.

1980. Agency problems and the theory of the firm. Journal of Political Economy, 88: 288-307.

41.

Fama

E. F.

Jensen

M. C.

1983. Separation of ownership and control. Journal of Law and Economics, 26: 301-325.

42.

Fan

Wong

Zhang

2007. Politically connected CEOs, corporate governance, and post-IPO performance of China’s newly partially privatized firms. Journal of Financial Economics, 84: 330-357.

43.

Fan

J. P. H.

Wong

T. J.

Zhang

2012. Institutions and organizational structure: The case of state-owned corporate pyramids. Journal of Law, Economics, and Organization, 29: 1217-1252.

44.

Feinerman

J. V.

2007. New hope for corporate governance in China? China Quarterly, 191: 590-612.

45.

Firth

Fung

P. M. Y.

Rui

O. M.

2006a. Corporate performance and CEO compensation in China. Journal of Corporate Finance, 12: 693-714.

46.

Firth

Fung

P. M. Y.

Rui

O. M.

2006b. Firm performance, governance structure, and top management turnover in a transitional economy. Journal of Management Studies, 43: 1289-1330.

47.

Fong

E. A.

Misangyi

V. F.

Tosi

H. L.

2010. The effect of CEO pay deviations on CEO withdrawal, firm size, and firm profits. Strategic Management Journal, 31: 629-651.

48.

Gao

2011. Philanthropic disaster relief giving as a response to institutional pressure: Evidence from China. Journal of Business Research, 64: 1377-1382.

49.

Groves

Hong

McMillan

Naughton

1995. China’s evolving managerial labor market. Journal of Political Economy, 103: 873-892.

50.

Guo

2019. The politics of the core leader in China: Culture, institution, legitimacy, and power. Cambridge, UK: Cambridge University Press.

51.

Gupta

2005. Partial privatization and firm performance. Journal of Finance, 60: 987-1015.

52.

Hansen

L. P.

1982. Large sample properties of generalized method of movements estimators. Econometrica, 50: 1029-1054.

53.

Haynes

K. T.

Hillman

2010. The effect of board capital and CEO power on strategic change. Strategic Management Journal, 31: 1145-1163.

54.

Hill

C. W. L.

Phan

1991. CEO tenure as a determinant of CEO pay. Academy of Management Journal, 34: 707-717.

55.

Holmes

R. M.

Hoskisson

R. E.

Kim

Wan

W. P.

Holcomb

T. R.

2018. International strategy and business groups: A review and future research agenda. Journal of World Business, 53: 134-150.

56.

Holz

C. A.

2007. Have China scholars all been bought? Far Eastern Economic Review, 170: 36-40.

57.

Hoskisson

R. E.

Hitt

M. A.

Johnson

R. A.

Grossman

2002. Conflicting voices: The effects of institutional ownership heterogeneity and internal governance on corporate innovation strategies. Academy of Management Journal, 45: 697-716.

58.

H. W.

Sun

2019. What determines the severity of tunneling in China? Asia Pacific Journal of Management, 36: 161-184.

59.

H. W.

Tam

O. K.

2012. Independent directors in China and India: A comparative assessment. Company and Securities Law Journal, 30: 453-470.

60.

Jensen

M. C.

Meckling

W. H.

1976. Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics, 3: 305-360.

61.

Jensen

M. C.

Murphy

K. J.

1990. CEO incentives: It’s not how much you pay, but how. Harvard Business Review, 68: 138-153.

62.

Jia

Huang

K. G.- L.

Zhang

C. M.

2018. Public governance, corporate governance, and firm innovation: An examination of state-owned enterprises. Academy of Management Journal, 62: 220-247.

63.

Jin

Qian

Weingast

B. R.

2005. Regional decentralization and fiscal incentives: Federalism, Chinese style. Journal of Public Economics, 89: 1719–1742.

64.

Johnson

La Porta

Lopez-de-Silanes

Shleifer

2000. Tunneling. American Economic Review, 90: 22-27.

65.

Keister

L. A.

2009. Interfirm relations in China: Group structure and firm performance in business groups. American Behavioral Scientist, 52: 1709-1730.

66.

Laeven

Levine

2008. Complex ownership structures and corporate valuations. Review of Financial Studies, 21: 579-604.

67.

La Porta

Lopez-de-Silanes

Shleifer

. 1999. Corporate ownership around the world. Journal of Finance, 54: 471-517.

68.

Le Breton-Miller

Miller

Lester

R. H

. 2011. Stewardship or agency? A social embeddedness reconciliation of conduct and performance in public family businesses. Organization Science, 22: 704-721.

69.

Leung

N. W.

Cheng

M. A.

2013. Corporate governance and firm value: Evidence from Chinese state-controlled listed firms. China Journal of Accounting Research, 6: 89-112.

70.

Andrew

G. W.

2001. Career advancement as party patronage: Sponsored mobility into the Chinese administrative elite, 1949-1996. American Journal of Sociology, 106: 1371-1408.

71.

Wang

2019. The liability of opaqueness: State ownership and the likelihood of deal completion in international acquisitions by Chinese firms. Strategic Management Journal, 40: 303-327.

72.

Qian

2013. Principal-principal conflicts under weak institutions: A study of corporate takeovers in China. Strategic Management Journal, 34: 498-508.

73.

Lin

2017a. Behind the numbers: State capitalism and executive compensation in China. University of Pennsylvania Asian Law Review, 12: 140-185.

74.

Lin

2017b. Reforming China’s state-owned enterprises: From structure to people. China Quarterly, 229: 107-129.

75.

Lin

Milhaupt

C. J.

2013. We are the (national) champions: Understanding the mechanisms of state capitalism in China. Stanford Law Review, 65: 697-759.

76.

Lü

Landry

P. F.

2014. Show me the money: Interjurisdiction political competition and fiscal extraction in China. American Political Science Review, 108: 706-722.

77.

Marquis

Qiao

2020. Waking from Mao’s dream: Communist ideological imprinting and the internationalization of entrepreneurial ventures in China. Administrative Science Quarterly. Advance online publication. doi:10.1177/0001839218792837

78.

Mengistae

L. C.

2004. Agency theory and executive compensation: The case of Chinese state-owned enterprises. Journal of Labor Economics, 22: 615-637.

79.

Meyer

M. W.

2011. Is it capitalism? Management and Organization Review, 7: 5-18.

80.

Morck

Yeung

Zhao

2008. Perspectives on China’s outward foreign direct investment. Journal of International Business Studies, 39: 337-350.

81.

Naughton

2017. Is China socialist? Journal of Economic Perspectives, 31: 3-24.

82.

Ocasio

1997. Towards an attention-based view of the firm. Organization Science, 18: 187-206.

83.

Ocasio

2011. Attention to attention. Organization Science, 22: 1286-1296.

84.

Park

S. H.

Tse

D. K.

2005. Market liberalization and firm performance during China’s economic transition. Journal of International Business Studies, 37: 127-147.

85.

Peng

M. W.

Heath

P. S.

1996. The growth of the firm in planned economies in transition: Institutions, organizations, and strategic choice. Academy of Management Review, 21: 492-528.

86.

Peng

M. W.

Luo

2000. Managerial ties and firm performance in a transition economy: The nature of a micro-macro link. Academy of Management Journal, 43: 486-501.

87.

Peng

M. W.

Sun

S. L.

Markóczy

2015. Human capital and CEO compensation during institutional transitions. Journal of Management Studies, 52: 117-147.

88.

Raynard

Lounsbury

Greenwood

2013. Legacies of logics: Sources of community variation in CSR implementation in China. In Lounsbury

Boxenbaum

(Eds.), Research in the sociology of organizations, 39, Part A: 243-276. Bingley, UK: Emerald Insight.

89.

See

G. K. H.

2009. Harmonious society and Chinese CSR: Is there really a link? Journal of Business Ethics, 89: 1-22.

90.

Shirley

M. M.

L. C.

1998. Information, incentives, and commitment: An empirical analysis of contracts between government and state enterprises. Journal of Law, Economics, and Organization, 14: 358-378.

91.

Shleifer

Vishny

R. W.

1986. Large shareholders and corporate control. Journal of Political Economy, 94: 461-488.

92.

Shleifer

Vishny

R. W.

1997. A survey of corporate governance. Journal of Finance, 52: 737-783.

93.

Sina.com. 2014. Retrieved from http://finance.sina.com.cn/china/20140127/162218099164.shtml. Accessed November 3, 2020.

94.

Singla

Veliyath

George

2014. Family firms and internationalization-governance relationships: Evidence of secondary agency issues. Strategic Management Journal, 35: 606-616.

95.

Spector

P. E.

Brannick

M. T.

2011. Methodological urban legends: The misuse of statistical control variables. Organizational Research Methods, 14: 287-305.

96.

Stavrou

E. T.

Swiercz

P. M.

1998. Securing the future of the family enterprise: A model of offspring intentions to join the business. Entrepreneurship Theory and Practice, 23: 19-40.

97.

Phan

P. H.

2008. Principal—Principal conflict in the governance of the Chinese public corporation. Management and Organization Review, 4: 17-38.

98.

Sun

H. W.

Hillman

A. J.

2016. The dark side of board political capital: Enabling blockholder rent appropriation. Academy of Management Journal, 59: 1801-1822.

99.

Tam

O. K.

H. W.

2006. Supervisory boards in Chinese corporate governance. In Ho

L. S.

Ash

(Eds.), China, Hong Kong and the world economy: Study on globalization: 327-347. New York: Palgrave-Macmillan.

100.

Tan

Peng

M. W.

2003. Organizational slack and firm performance during economic transitions: Two studies from an emerging economy. Strategic Management Journal, 24: 1249-1263.

101.

Tan

2005. Environment-strategy co-evolution and co-alignment: A staged model of Chinese SOEs under transition. Strategic Management Journal, 26: 141-157.

102.

Tihanyi

Aguilera

R. V.

Heugens

Van Essen

Sauerwald

Duran

Turturea

2019. State ownership and political connections. Journal of Management, 45: 2293-2321.

103.

Tosi

H. L.

Werner

Katz

J. P.

Gomez-Mejia

L. R.

2000. How much does performance matter? A meta-analysis of CEO pay studies. Journal of Management, 26: 301-339.

104.

Walder

A. G.

1995a. Career mobility and the communist political order. American Sociological Review, 60: 309-328.

105.

Walder

A. G.

1995b. Local governments as industrial firms: An organizational analysis of China’s transitional economy. American Journal of Sociology, 101: 263-301.

106.

Walder

A. G.

Treiman

D. J.

2000. Politics and life chances in a state socialist regime: Dual career paths into the urban Chinese elite, 1949 to 1996. American Sociological Review, 65: 191-209.

107.

Wang

Luo

X. R.

2019. Retire in peace: Officials’ political incentives and corporate diversification in China. Administrative Science Quarterly. Advance online publication. doi:10.1177/0001839218786263

108.

Wang

Cao

2018. Mandatory corporate social responsibility (CSR) reporting and financial reporting quality: Evidence from a quasi-natural experiment. Journal of Business Ethics, 152: 253-274.

109.

Wasserman

2006. Stewards, agents, and the founder discount: Executive compensation in new ventures. Academy of Management Journal, 49: 960-976.

110.

Wei

Xie

Zhang

2005. Ownership structure and firm value in China’s privatized firms: 1991-2001. Journal of Financial and Quantitative Analysis, 40: 87-108.

111.

2011. The fundamental institutions of China’s reforms and development. Journal of Economic Literature, 49: 1076-1151.

112.

J. W.

2014. Organizational forms and multi-population dynamics: Economic transition in China. Administrative Science Quarterly, 59: 517-547.

113.

Zhou

Phan

P. H.

2010. A real options perspective on sequential acquisitions in China. Journal of International Business Studies, 41: 166-174.

114.

L. C.

Shirley

M. M.

2001. Empirical effects of performance contracts: Evidence from China. Journal of Law, Economics, and Organization, 17: 168-200.

115.

Young

M. N.

Peng

M. W.

Ahlstrom

Bruton

G. D.

Jiang

2008. Corporate governance in emerging economies: A review of the principal-principal perspective. Journal of Management Studies, 45: 196-220.

116.

Zheng

Deng

2018. State failure and distorted urbanisation in post-Mao’s China, 1993-2012. London: Palgrave Macmillan. https://link-springer-com-443.web.bisu.edu.cn/book/10.1007%2F978-3-319-92168-6

117.

Zhong

2015. Corporate governance of Chinese privatized firms: Evidence from a survey of non-listed enterprises. Journal of Comparative Economics, 43: 1101-1121.

118.

Zhou

K. Z.

Gao

G. Y.

Zhao

2017. State ownership and firm innovation in China: An integrated view of institutional and efficiency logics. Administrative Science Quarterly, 62: 375-404.

119.

Zuo

2015. Promoting city leaders: The structure of political incentives in China. China Quarterly, 224: 955-984.