In recent years, China has initiated a new marketization campaign with the introduction of strategic private capital and innovative corporate governance in the Chinese state sector.Footnote 1 The “mixed ownership” and classification reform has placed professional management practices such as open recruitment and fixed-term labour contracts for SOE managers on the SOE reform agenda. This round of SOE reforms aims at changing state–Party–business relations within SOEs, with the State-owned Assets Supervision and Administration Commission of the State Council (Guowuyuan guoyou zichan jiandu guanli weiyuanhui 国务院国有资产监督管理委员会, SASAC hereafter) attempting to give equal weight to bureaucracy and the market. It has, however, proved difficult for SOEs to strike a balance between market-oriented incentives and their government-controlled status. Currently, most of the managerial elites in central and local SOEs are still controlled through administrative ranks.Footnote 2 They are ambiguous semi-officials, rotating between SOEs and government departments. Zhihong Chen, Yuyuan Guan and Bin Ke refer to this phenomenon as China's closed pyramidal managerial labour market.Footnote 3 In another interesting case, Chen Donghua and colleagues find that local institutional forces can render market-style incentive mechanisms dysfunctional within China's SOEs.Footnote 4 Comparatively, local SOEs are subject to more government intervention as local government officials pursue personal political interests.Footnote 5 Hence, the appointment, performance evaluation and leadership tenure of managerial elites can differ across central and local SOEs.
In this paper, we attempt to uncover the practices of personnel management in the publicly listed subsidiaries of China's central and local SOEs. We focus on the determinants of SOE managers’ career mobility and performance evaluation. Specifically, we investigate the following research questions. First, how do the managerial elites of SOEs move within China's state system?Footnote 6 How frequently do they rotate jobs? Second, is firm economic performance a more important factor in the evaluation of SOE leaders than political connection? Third, is personnel management structured differently across SOEs at the central, provincial, municipal and county levels? Fourth, how do bureaucracy and market jointly influence the incentive practices of SOEs over the long term?
To answer these research questions, we first gathered data on the executive turnover in China's listed SOEs, categorizing the departure destinations of SOE leaders. We then compared the career mobility and performance evaluations of central SOE leaders with those of local SOE leaders. Third, considering that SOE leaders are subject to different appointment authorities, we considered how performance evaluation varies between the leaders of SOE publicly listed subsidiaries and those of SOE group companies. Overall, we found that the leaders of central and local SOEs have similar career mobility, but their performance evaluation criteria are different. Vertical intragroup mobility is common, and this lock-in effect is more pronounced for central SOEs. On average, more than two-thirds of SOE leaders leave office within three years. Roughly, one-third of departing SOE managerial elites are demoted, one-third transfer to lateral positions and the remaining leaders are promoted. Furthermore, we found that economic performance and political connections play different roles in the evaluations of SOE leaders. Local SOE leaders with close political connections to local governments benefit from more advantages as they climb the career ladder. In contrast, only central SOE leaders with a record of good economic performance are offered more opportunities. Through additional analyses, we further found that this holds for the leaders of SOE group companies at both central and local levels.
Our paper is closely related to the literature on SOE leader political mobility and its determinants.Footnote 7 For example, Kjeld Erik Brødsgaard and colleagues compare the factors that influence the career progressions of executives who work in SOEs where the state holds direct ownership with those that influence the careers of executives in SOEs with indirect ownership. Wendy Leutert and Samantha Vortherms investigate if central SOE leaders retain their original positions. The current paper, however, differs from previous studies in that it covers a broader range of SOEs and extends beyond SOE group-level leaders of central SOEs and the classification of SOE leaders’ career moves into promotion, lateral transfer and demotion. To our knowledge, our research is the first to conduct a comprehensive analysis of the personnel management of China's SOEs at different administrative levels. We expect to reveal how the dual identity of SOE leaders develops with bureaucracy and the market.
Our paper mainly makes three contributions. First, it provides a complete picture of SOE leaders’ career mobility at different levels within the hierarchy, encompassing all the possible destinations open to SOE leaders: SOEs, government departments, private firms, NGOs, retirement, or departure for health reasons, criminal activity or scandal. We offer comprehensive analyses on how SOE leaders change positions within or across SOE groups. Second, our findings expand the knowledge of the determinants influencing career mobility and the performance evaluations of central and local SOE leaders. Importantly, we find that economic performance and political connections have different roles in the evaluations of central and local SOE leaders. Third, our investigation into SOE group companies and SOE publicly listed subsidiaries enhances understanding of how SOE leaders move within China's closed pyramidal managerial labour market. Overall, we believe that our findings have important implications for the future reform of China's SOE personnel system.
The remainder of the paper is organized as follows. The following section reviews the literature related to our study. The paper then introduces the institutional background to China's SOE personnel system before going on to describe the data construction and methodologies. The sections following that report our analyses of the career mobility and performance evaluations of central and local SOE leaders. Further discussion on SOE group companies and SOE publicly listed firms is then presented before the paper concludes and considers future research directions.
Literature Review
In line with the extant research on the political mobility of government officials, more inquiry should be made into SOE managerial elites.Footnote 8 Prior studies have examined the political mobility of Chinese government officials and the underlying determinants. For example, Jean Oi and Maria Edin both provide in-depth analyses of rural politics in China,Footnote 9 while other scholars pay more attention to the selection and appointment of political elites at the top-to-middle echelons within the Chinese bureaucracy.Footnote 10 There has also been extensive research into whether economic performance or political connectedness (also referred to as “clientelism” or “patronage”) drives the political mobility of government leaders at provincial, municipal and even county levels.Footnote 11
Despite the fruitful results of research on the career mobility of government officials, more knowledge is needed about the paths taken by managerial elites in central and local SOEs. Brødsgaard studies top leaders’ rotations in China's SOE business groups, showing that the nomenklatura system facilitates the Party's control of powerful state-owned business groups.Footnote 12 Leutert also states that the career mobility of SOE leaders, especially in local SOEs, is still an important research question in extant literature.Footnote 13 Some studies focus on the career pathways of the managerial elites in central SOEs or SOEs in specific industries. In addition, some papers emphasize individual career dynamics rather than the SOE personnel system. Leutert is the first to systematically investigate the biographical information of the central SOEs’ managerial elites under the Hu Jintao 胡锦涛 administration, concluding that lateral transfers frequently occurred along their pathways. Li-Wen Lin looks into the different career patterns of managerial elites in non-financial SOEs and state-controlled banks, investigating a variety of human resource practices used by these SOEs.Footnote 14
We also need to know more details about the turnover of SOE leaders, such as the determinants of promotion. Evaluation-related information on SOE leaders, including leadership tenure, economic performance and political connectedness, can provide a better understanding of SOEs’ personnel management.Footnote 15 Although many empirical papers have studied turnover performance sensitivity, an investigation into this question is still valuable.Footnote 16 In the extant literature, the relationship between economic performance, political connections and leadership turnover in SOEs can be categorized as: (1) no relation;Footnote 17 (2) economic performance is more important to SOE leaders than political connectedness;Footnote 18 or (3) the multiplicity of government objectives renders the economic performance of SOEs subordinate to political pressures and social connections in personnel decisions.Footnote 19 A study by Fei Du, Guliang Tang and Mark Young, in particular, identifies political connections, social responsibilities and the political rank of SOEs as the primary determinants of SASAC evaluation scores.Footnote 20
Personnel Management in China's SOE
Central SOEs and local SOEs
Both central SOEs and local SOEs have multiple objectives that encompass political, social and economic tasks.Footnote 21 Policy burdens, such as extra strategic investment or employment stability, can lead to operational losses and financial distress.Footnote 22 Since central SOEs and local SOEs are controlled by different levels of administration with divergent political goals and interventions,Footnote 23 the performance evaluations for different SOE managers vary. Owing to fiscal decentralization and tax reform, local governments have a vested interest in leveraging SOEs to achieve social and political goals and are more inclined to intervene in local SOEs. In addition, local SOE leaders deal with more policy burdens because of their closer ties to local governments. In contrast, central SOEs are usually business giants and in the public eye; central SOE managers need to demonstrate that their SOEs are in robust economic health in order to satisfy the public's high expectations. Economic performance is thus an important factor when evaluating central SOE managers. It is therefore necessary to empirically clarify how political and economic factors feature in personnel control practices such as career mobility and performance assessment across central and local SOEs.
Central and local SOEs are both subject to the pyramid structure of China's state-owned business groups.Footnote 24 Figure 1 provides two examples of this structure that relate to central and local SOEs. Under this system, the government and CCP bodies retain control over SOEs through ownership ties and the nomenklatura system. The government and CCP manage the appointment of SOE leaders through a cadre management system that transfers/rotates SOE leaders between government and Party agencies according to their political rank. Owing to the unique business culture and work habits inside SOEs, most SOE managers spend their entire career within the state system. Donghua Chen and colleagues argue that SOE leaders work within a closed managerial labour market that is characterized by its own internal hierarchy.Footnote 25 SOE leaders working within this closed labour market can find positions in group companies, parent companies or subsidiary companies. In addition, they can be laterally transferred to other state-owned business groups or the government. Demotion is also a possibility.
Two Examples of the Pyramid Structure of Central and Local SOEs
Source: Data collected by the authors.
Appointment
According to the Law of the People's Republic of China on State-owned Assets of Enterprises (2008) (Zhonghua renmin gongheguo qiye guoyou zichan fa 中华人民共和国企业国有资产法), the central and local SASACs can appoint, assess or recommend leaders for SOEs that are fully or partially controlled by the state. The authority to appoint central SOE executives is divided into two parts. Specifically, the CCP Central Organization Department (Zhongguo gongchandang zhongyang weiyuanhui zuzhibu 中国共产党中央委员会组织部), in consultation with the Enterprise Leaders No. 1 Management Bureau of the SASAC (qiye lingdao renyuan guanli yiju 企业领导人员管理一局), appoints and evaluates the Party secretaries, chairs and CEOs for 51 central SOEs with vice-ministerial administration ranks. The central SASAC manages the other executives in these large SOEs. Additionally, the Enterprise Leaders No. 2 Management Bureau of SASAC (qiye lingdao renyuan guanli erju 企业领导人员管理二局) appoints and appraises the senior executives in another 47 central SOEs. In sum, the central SASAC participates in the appointment and appraisal of all the leading executives in the central SOEs. Similarly, the senior executives of local SOEs are jointly managed by the local SASAC and the local Party committee organization department.
In general, the party-state's control over SOE leaders results in bureaucratic appointments being widely applied. SOE managerial elites are directly appointed by the government and have administrative ranks on a par with China's political hierarchy. To encourage more managerial talent to join SOEs, recruitment has been conducted through a more market-oriented process since 2004, and in 2014, following the mixed-ownership and classification reform, SOEs hired more professional managers. Nevertheless, the government, with ultimate control rights, still exerts a significant and direct influence over the management of SOEs – particularly on the appointment and dismissal of SOE leaders.Footnote 26
Here, we offer the Shenzhen SASAC as an illustration of SOE manager appointment. The policy document, “Interim provisions on the selection and appointment of leaders of municipal state-owned enterprises” (shishu guoyou qiye lingdao renyuan xuanba renyong zanxing guiding 市属国有企业领导人员选拔任用暂行规定) (2015), stipulates that the Shenzhen SASAC has the authority to appoint and appraise the chair, CEO and Party secretary of the SOEs and their affiliated listed firms under its jurisdiction. For example, He Jianfeng 何建锋, the former chairman of the Shenzhen Agricultural Products Group Co., Ltd. (Shenzhen shi nongchanpin jituan gufen youxian gongsi 深圳市农产品集团股份有限公司), was recently appointed chairman of the Shenzhen Food Group Co. Ltd (Shenzhen shi shipin wuzi jituan youxian gongsi 深圳市食品物资集团有限公司) by the Shenzhen SASAC (Shenzhen SASAC [2020] No. 9). Figure 2 shows that before joining the listed SOE, He was a government official in the Shenzhen SASAC for a long time (① and ③ in Figure 2). Through a series of “revolving door” transfers, He moved between the government and Shenzhen's SOEs (②, ④ and⑤ in Figure 2). The sequential numbers ①, ②, ③, ④ and ⑤ represent He's career sequence.
Career Mobility of a Local SOE Leader: He Jianfeng
Source: Data collected by the authors.
Notes: ① Government official, leader of Division one in the Shenzhen SASAC; ② Deputy leader of Shenzhen Sez Construction and Development Group Co. Ltd; ③ Government official, chief economist of the Shenzhen SASAC; ④ Chairman of Shenzhen Food Group Co. Ltd, chairman and Party secretary of its listed subsidiary (000061); ⑤ Chairman and Party secretary of Shenzhen Investment Holdings Co. Ltd.
Performance evaluation and termination
The current SOE managerial promotion guidelines explicitly include firm performance in the CEO evaluation matrix. Since 2003, the SASAC has promulgated a series of “Interim provisions on business performance evaluations for persons-in-charge at central enterprises” (Zhongyang qiye fuzeren jingying yeji kaohe banfa 中央企业负责人经营业绩考核办法) (2003, 2006, 2009, 2012, 2016, 2019) that set new evaluation methods for central SOE leaders.Footnote 27 Articles 2, 53 and 54 of the 2019 edition indicate that the provisions apply to the leaders of the group company and publicly listed subsidiaries. Moreover, these articles indicate that local SOE leaders are also essentially subject to the provisions. The 2019 document stipulates that SOE leaders of different levels must sign two types of liability letter regarding operational performance. These letters specify the key indices to be used for evaluations, such as annual indicators and service term indicators. For instance, indices can include total annual profits, returns on net assets, the ratio of maintenance and appreciation of state-owned assets, and revenue growth rate. Based on the results of their appraisals, SOE managerial elites are divided into five grades: A, B, C, D and E.
As political businesspeople, SOE leaders are faced with multiple tasks that include meeting economic, political and social targets such as maintaining and increasing the value of state-owned assets or implementing redundancies. Thus, the performance evaluation indicators for SOE leaders reflect principals’ priorities.Footnote 28 Driven by strong political incentives, local government officials are more likely to assign political tasks to local SOE leaders and intervene in the appointment of SOE managerial elites. Therefore, personnel management practices can differ in central SOEs and local SOEs in response to bureaucratic and market demands.Footnote 29
Data and Methodologies
Career information of SOE leaders
We gathered information on 1,250 SOE leaders in 585 Chinese A-share listed companies from 2003 to 2012. Leaders’ positions included chair, CEO and Party secretary.Footnote 30 We hand collected the data from the SASAC websites, corporate websites, biographies, annual reports, departure announcements and the China Stock Market and Accounting Research (CSMAR) databases, as well as financial websites such as Sina.com.
First, we classified executives’ origins and destinations into four categories: “public sector,” “non-state sector,” “other” and “unknown.” The “public sector” category refers to job rotations within the state system including SOEs and government departments. The “non-state sector” category includes job moves from the state system to private firms or non-profitable organizations (NGOs). The “other” category refers to executive turnover through retirement, health problems, criminal acts or scandal. The “unknown” category refers to turnover without any extra information. The classification scheme is shown in Figure 3.
Origins and Destinations of Career Mobility of China's SOE Leaders
Source: Drawn by the authors.
Second, in light of previous literature and the change in administrative levels,Footnote 31 we further categorize the career moves of SOE leaders into different groups: political promotion, non-political promotion, lateral transfer and demotion. Figure 4 shows that political promotion refers to the situation where SOE leaders are promoted to government departments. Non-political promotion refers to appointment to a higher position in the state sector, including (a) a new position in the parent SOE or SOEs of a higher administrative rank, or (b) a higher position in the same SOE or SOEs of similar administrative rank. Lateral transfer refers to when leaders move to a position in another SOE without changing their administrative rank. SOE leaders are demoted if they (a) transfer to a subsidiary SOE or SOEs of a lower administrative rank, or (b) are dismissed for their involvement in criminal activity or scandals.
Promotion, Lateral Transfer and Demotion of China's SOE Leaders
Source: Drawn by the authors.
Publicly listed SOEs
We chose publicly listed SOEs for our research sample for two reasons. First, the choice is consistent with the previous literature, such as work undertaken by Donghua Chen and colleagues and Xiping Cao and colleagues, which uses data on the promotion of executives in listed SOEs.Footnote 32 Second, ownership, performance and executive biographical data are more accessible for publicly listed firms. Using the CSMAR database and public information, we divided all the public firms on the Chinese Stock Market into different ownership categories: (1) private firms controlled by non-government units, (2) local SOEs controlled by local governments (for example, local SASACs), and (3) central SOEs controlled by the central government (for example, the Ministry of Finance and the central SASAC). To ensure that sample firms were comparable over the sample period, we retained only the listed firms that were consistently state-owned from 2003 to 2012.
Career Mobility of Central and Local SOE Leaders
Sample description
The sample of SOE managerial elites includes 561 CEOs, 630 board chairs and 59 Party committee secretaries. None of the 1,250 SOE leaders individually hold all three positions simultaneously; however, 47 leaders hold the positions of both CEO and Party secretary, 37 leaders are both CEO and chair, and 12 leaders hold the positions of chair and Party secretary at the same time.Footnote 33 Figure 5 shows that about 52.61 per cent of SOE leaders in our sample work in both parent SOEs and listed SOEs. Among the 580 SOEs in our sample, 40.86 per cent are provincial SOEs, 33.28 per cent are central SOEs and 25.86 per cent are prefectural SOEs. In addition, 59.5 per cent of the sample SOEs are located in the east of China, 21.05 per cent in the west and 19.25 per cent in central China. While 64.96 per cent of the sample SOEs are from competitive industries, the others are from monopolistic industries.Footnote 34 A large number of the SOEs are in the petroleum and petrochemical industry (12.40 per cent), the machinery industry (16.16 per cent) and the metallic and electricity industry (9.76 per cent).
Concurrent Positions of China's SOE Leaders
Source: Drawn by the authors.
Age and educational background, illustrated in Table 1, are also important factors in the career mobility of SOE leaders. The mandatory retirement age is 60 for officials, and this rule also applies to SOE leaders. On average, central and provincial SOE leaders are appointed at the age of 47, while city-level or county-level SOE leaders assume their positions at the age of 44 or 42. Central and provincial SOE leaders generally leave office at the age of 50, whereas city-level or county-level SOE leaders vacate their positions at the age of 47 or 44. The age difference between assuming and leaving office is reasonable because political promotions usually take three to five years. The age distribution is consistent with the political tournament requirements in China, which suggest that leaders’ political ranks are positively correlated with their age. If SOE leaders have not been promoted by a certain age, their future prospects for promotion are diminished. It is noteworthy that later retirement exists within our sample: 8.96 per cent of central SOE leaders, 7 per cent of provincial SOE leaders and 4 per cent of city-level SOE leaders retire after the age of 60.Footnote 35 We do not find this to be the case for county-level SOE leaders.
Age and Educational Background of Central and Local SOE Leaders
Source: Data collected by the authors.
Different levels of education degrees result in different career opportunities.Footnote 36 We find that the majority of SOE leaders have bachelor's degrees. Additionally, in central and provincial SOEs, many leaders have master's or doctoral degrees (41.6 per cent and 40.24 per cent respectively). Of the 1,250 SOE leaders, 96 have a doctoral degree.Footnote 37 Although an MBA degree seems to be critical for business, our findings indicate that only 68 leaders hold an MBA degree (5.4 per cent). The evidence suggests that Party school training experience or a postgraduate degree is more important for SOE leaders.Footnote 38
Leadership tenure
We now turn to turn to how frequently SOE leaders rotate their positions. Fortune's CEO data show that the 500 largest companies in the US have a median CEO tenure of 4.9 years. Different from the contractual employment of the market economy, the majority of SOE leaders are appointed by SASACs, and each tenure term is stated in a three-year letter of liabilities. Further, employment tenure is relatively uncertain because bureaucratic appointments are bound by fewer contractual constraints.
Since each SOE leadership term generally cannot exceed three years, we set three years as the minimum term, with intervals of three years, and compared the leadership tenures of managerial elites in both central and local SOEs. Figure 6 shows that more than 60 per cent of SOE leaders serve fewer than three years. In other words, most of the SOE managerial elites will be “shifted” within three years, indicating that their career pathway is characterized by frequent job rotations. In addition, fewer than 30 per cent of SOE leaders serve three to six years, and fewer than 10 per cent of leaders serve six to nine years. None of the county-level SOE leaders serve more than nine years. The leadership tenures of different level SOEs follow a similar pattern: the average tenure is 3.12 years and the standard deviation is about two years. Therefore, turnover is more frequent for SOE leaders than for Fortune CEOs. The results suggest that SOE personnel management is distinct from the contractual employment of the external labour market.
Tenure Distribution of Central and Local SOE Leaders
Source: Drawn by the authors.
Departure destinations
In China, SOEs and government agencies constitute the state sector and the career pathways of SOE leaders are shaped by a closed, hierarchical managerial labour market.Footnote 39 SOE managers can move jobs across different SOEs and government agencies because their political rank facilitates their career mobility. Leaving the state sector incurs high switching costs associated with the loss of benefits and high social status provided by SOEs and government agencies. Consequently, the majority of SOE leaders choose to remain working within the state sector.
Table 2 summarizes the departure destinations of central and local SOE leaders. We find that the public sector is the most popular destination for SOE leaders. In total, 59.04 per cent of SOE leaders stay within the same SOE group. In addition, 6.08 per cent of SOE leaders move outside their previous SOE group, and 6.56 per cent of leaders move to posts in government departments. The results suggest that job rotation across different groups is not frequent, as SOE leaders typically prefer working in the same SOE group. Only 4.4 per cent of leaders move to the non-state sector, suggesting that a strong “lock-in” effect exists in the state system. Some SOE leaders have to leave their posts for retirement, health reasons or because of criminal activity.
Departure Destinations of Central and Local SOE Leaders
Source: Data collected by the authors.
We also investigate the differences between central, provincial and prefectural SOEs in Table 2. Although the majority of SOE leaders are inclined to move to jobs within the same SOE group, prefectural SOEs offer a lower likelihood of intra-group mobility than do central and provincial SOEs. The “lock-in” effect also becomes stronger as SOEs have higher political ranks.
Promotion, lateral transfer and demotion
To summarize the factors behind career mobility, we divide the pathways into three categories – promotion, lateral transfer and demotion – by comparing the changes of position of SOE leaders. The definitions of the three categories are provided in the section, “Data and methodologies,” above. In Table 3, we focus on 959 SOE leaders who rotate jobs within the public sector. The table shows that 41.4 per cent of the SOE leaders experience promotion; 29.82 per cent of the leaders are laterally transferred; and 28.78 per cent of the leaders face demotion. The share of central SOE leaders who are demoted (22.64%) is smaller than that for local SOE leaders (28.98% or 39.51%). The evidence indicates that employment rigidity increases with political rank. In addition, more than 85 per cent of SOE demotions are owing to an involvement in criminal activity, punishment imposed by regulatory agencies and/or exposure to scandal.
Promotion, Lateral Transfer and Demotion of Central and Local SOE Leaders
Source: Data collected by the authors.
We also find that non-political promotion is more common than political promotion. Importantly, the likelihood of political promotion is associated with the political rank of the SOE. Given the pyramidal political structure of Chinese politics, SOE leaders can move between posts within the public sector based on their political rank. Compared to central and provincial SOEs, prefectural SOE leaders are more likely to be politically promoted. As the political rank of the SOE rises, the proportion of SOE leaders receiving political promotion declines, especially for central SOE leaders. The comparison suggests that higher political ranks result in fewer opportunities for political promotion, because high-level government positions are rare.Footnote 40
Performance Evaluation of Central and Local SOE Leaders
Model specification
The career mobility of SOE leaders is dependent on the outcome of their performance evaluation by central and local SASACs. In accordance with the SASAC policy documents and prior work, we use the ordered logit model to examine the sensitivity of promotion to economic performance and political connection for central and local SOE leaders.Footnote 41 The model is specified below:
$$Promotion = \beta _0 + \beta _1Performance + \beta _2Politician + \beta _3Edu + \beta _4Age + \beta _5Tenure + \beta _6Size + \beta _7Chain + Market + Monopoly + Year + \varepsilon \;$$where Promotion is an ordinal variable that equals 1 if an SOE leader achieves promotion, 0 for lateral transfer and -1 for demotion. Performance measures the accounting performance of SOEs, which equals the return on assets (ROA). Politician is an indicator variable measuring the political connections of SOE leaders, equal to 1 if an SOE leader is a deputy of the National People's Congress, a member of the People's Political Consultative Conference, or a government official.
The other variables of Model (1) control for biographical, firm and industry characteristics. Representing the educational background of SOE leaders, Edu is an indicator variable which equals 1 if a leader has a master's or doctoral degree. Age measures the age of the executives and Tenure represents their service time. Size is calculated as the logarithm of total assets. Chain measures the number of control chains between the actual controller and the listed SOEs.Footnote 42 Market is an indicator variable equal to 1 if an SOE is in China's eastern regions. Monopoly is an indicator variable that equals 1 if an SOE is from a regulated industry. Year controls for the year fixed effects. Before regression analyses, we exclude financial industry SOEs and SOE leaders who retire or resign for unknown reasons. We also winsorize the continuous variables at the 1 per cent level to mitigate the impact of extreme value.
We predict that an SOE leader's promotion outcome is positively associated with economic performance and political connections. Both bureaucracy and the market can have an impact on an SOE leader's career mobility. Nonetheless, the relative roles of economic performance and political connections can change for SOEs at different administrative levels.
Empirical results
We first report a series of univariate analyses in Table 4. The mean difference test enables us to calculate the statistical differences of ROA (Performance) and political connections (Politician) between a variety of groups. For central SOE leaders, we find that both ROA and political connections are important to their evaluation process. Panel D shows that ROA is higher for the promotion group and the difference is statistically significant. In addition, Panels B and E indicate that political connections are mainly effective for the lateral transfer and political promotion groups. In contrast, for provincial and prefectural SOE leaders, their performance evaluation depends on their political connections. Panels D and E show that political connectedness is higher for the promotion or political promotion groups, and the difference is statistically significant. The effect of political connections is particularly strong for prefectural SOE leaders.
Univariate Analyses of Performance Evaluations
Source: Calculated by the authors.
Notes: *, ** and *** indicate statistical significance at the 10% level, 5% level, and 1% level, respectively.
Table 5 presents the regression results of model (1) for different types of SOEs. Column (1) shows that Performance has a positive and statistically significant coefficient. The results suggest that economic performance is more important for central SOE leaders’ performance evaluation. Additionally, we find that the coefficient of Performance in columns (2)–(4) is insignificant, while the coefficient of Politician is positive and has statistical significance. Importantly, the impact of political connections is stronger for municipal and county-level SOEs. These findings suggest that the career mobility of local SOE leaders largely depends on their political connections with local governments. The difference between central and local SOEs may derive from the fact that local governments need to maintain a tight connection with SOEs to achieve a variety of political and social goals.Footnote 43
Regression Analyses of Performance Evaluations
Source: Calculated by the authors.
Notes: To address the concerns about the potential similarities and differences between the Hu and Xi administrations, we hand collected the 2013–2017 career mobility data of SOE managerial elites under the Xi administration. Applying the same analysis of performance evaluation, we found that the test results under the Xi administration, which are not tabulated in the paper, are largely consistent with Table 5. t-values are reported in parentheses; *, ** and *** indicate statistical significance at the 10% level, 5% level, and 1% level, respectively.
Concerning other driving factors, Table 5 shows that in column (1) the coefficient of Edu is positive and statistically significant. The evidence indicates that central SOE leaders with higher academic degrees are more likely to achieve promotion. Columns (1), (2) and (3) report that Age has a negative and statistically significant coefficient, whereas the coefficient of Size is positive and statistically significant. The findings suggest that younger SOE leaders and the leaders of larger SOEs have more promotion opportunities. Overall, we find that the SASAC mainly takes economic performance, political connections and biographical characteristics into account when deciding on the promotional prospects of SOE leaders.
Discussion: Leaders of SOE Groups and Listed SOE Subsidiaries
Li-Wen Lin divides the career patterns of central and listed SOE leaders into single-group track, multi-group track and government unit track.Footnote 44 He finds that the single-group track, namely intra-group mobility, is prevalent in Chinese SOEs. SOE executives move up, down or laterally within the same SOE group, as the hierarchical ownership structure of the SOE group provides numerous career opportunities. To realize post-firm advancement, SOE managerial elites can also rotate between Party, business and government positions through the Party's personnel management system. We have shown how intra-group mobility and other types of personnel rotations take place and how economic and political factors influence the rotation.
Since SOE leaders are subject to different appointment authorities, we further distinguish the SOE leadership of group companies from that of publicly listed subsidiaries. To investigate whether the leaders of SOE groups and listed SOEs are comparable in pooled analyses, we separate the performance evaluation results of the SOE leaders with concurrent leadership posts and those of leaders with only firm-level leadership posts. Table 6 shows that in column (1), the coefficient of Performance is positive and statistically significant, whereas in columns (3) and (4), Politician has a positive coefficient of statistical significance. Therefore, for both central and local SOEs, the results of holding concurrent leadership posts are consistent with the findings of Table 5. The consistency can alleviate the concern that the evaluation criteria for different leadership posts are not comparable.
Comparison of Performance Evaluation of Different Leaderships
Source: Calculated by the authors.
Notes: t-values are reported in parentheses; *, **, and *** indicate statistical significance at the 10% level, 5% level, and 1% level, respectively.
This makes sense, as SOE groups and listed SOE subsidiaries have fundamental similarities. First, SOEs usually have multiple objectives that encompass economic, political and social goals. The targets set by the SASAC are assigned to different SOE groups and distributed to their subsidiaries. Although the leaders of parent SOEs and subsidiary SOEs are appointed by different authorities, the goals assigned to them by the SASAC are quite similar. Second, the leaders of SOE groups and listed SOEs face similar career mobility opportunities. They both move within a closed labour market that comprises SOEs and government agencies. Third, the performance evaluations of SOE groups and listed SOEs are similar. In addition to the group company's overall performance, the performance of listed subsidiaries is an important factor determining the appraisal of SOE group leaders, especially when the leaders assume the concurrent positions of both group companies and listed subsidiaries. Anecdotal evidence also indicates that SOE groups and listed subsidiaries tend to have similar evaluation criteria.Footnote 45
Taken together, we provide novel details on the intra-group mobility of SOE executives based on the vertical interlock of SOEs, shedding light on broader career mobilities related to the horizontal interlock between Party organizations and other corporate organs. Furthermore, our research highlights how performance evaluation varies between central and local SOE leaders. Our findings also suggest that the evaluations of different leadership posts are largely comparable. The similarities and differences regarding the career mobility and performance evaluation of SOE top executives enhances our understanding of the Party's control of personnel in China's SOEs.
Conclusion
It has always been challenging for the Chinese government to initiate SOE reforms through the Party's control over SOE personnel. Each round of SOE reform attempts to address this sticking point. In this paper, we provide a complete picture of personnel control in China's SOEs by showing how managerial elites rotate and how they are evaluated across central SOEs and local SOEs. Our findings suggest that the majority of central and local SOE leaders move within the state system. More than two-thirds of SOE leaders rotate within three years. Meanwhile, about one-third of departing SOE managerial elites are demoted, one-third experience lateral transfers and the rest are promoted. Fewer than 9 per cent of SOE leaders can achieve political promotion. Additionally, economic performance and political connections play different roles in central and local SOEs. Moreover, the results remain consistent for the leaders of SOE group companies and those of SOE listed firms.
Investigating both central and local SOEs broadens our understanding of the career mobility of SOE leaders within China's cadre management system. The findings on SOE leader evaluation indicate that different levels of governments have distinct preferences when governing the state sector. The leaders of SOE group companies and SOE listed subsidiaries are not exempt from the bureaucratic personnel system. These implications, taken together, suggest that any future reform of SOEs needs to consider the trade-offs between bureaucracy and the market, especially when government officials have strong political incentives.
Future research could further investigate how performance mechanisms influence the behaviour of SOE managers – for example, their management styles, risk preferences and decision-making patterns.Footnote 46 It is also important to draft a framework for analysing the differentiated behaviour patterns of central and local SOE leaders, or the behaviour of different types of SOEs, such as monopolistic industry SOEs, competitive industry SOEs and SOEs with a high degree of government ownership. Given the series of recent SOE reforms, future studies could explore how the system of personnel management for SOE leaders evolves under the new circumstances.
Acknowledgements
We are grateful to the editor and the anonymous reviewers for their insightful and detailed comments. All errors are our own. We acknowledge the funding from the National Natural Science Foundation of China (No. 72002011, 72232007).
Competing interests
None
Linlin ZHANG is an associate professor in the School of Economics and Management at Beijing Jiaotong University in Beijing, China. She received her PhD in 2017 from Sun Yat-sen University.
Feng LIU is a professor and the chief director of the Center for Accounting Studies at Xiamen University in Xiamen, China. He is also a member of the Advisory Committee of International Financial Reporting Standards (IFRS).
Xin YUAN is an assistant professor in the School of Accounting at Dongbei University of Finance and Economics in Dalian, China. He received his PhD in 2021 from Baruch College, City University of New York.
