Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Will political promotion expectation decrease the pay gap in state-owned enterprises in China?

Will political promotion expectation decrease the pay gap in state-owned enterprises in China? China Journal of aCC ounting StudieS , 2016 Vol . 4, no . 1, 53–78 http://dx.doi.org/10.1080/21697213.2016.1144969 Will political promotion expectation decrease the pay gap in state-owned enterprises in China?* a b c Danlu Bu , Chenyu Zhang and Teng Lin a b Southwestern university of f inance and economics, Chengdu, China; university of international Business and economics, Beijing, China; Shantou university, China KEYWORDS ABSTRACT a gency cost; pay gap; In this paper, we test the effects of executives’ promotion expectation political promotion; on compensation and the pay gap between executives and employees promotion expectation; in China. We measure the promotion expectation of current executives promotion rank by using their predecessors’ promotion ranks and direction. We find that there is a significantly negative relationship between promotion expectation and the pay gap, suggesting that senior executives with stronger expectation of political promotion are more willing to accept less compensation and therefore a smaller pay gap. Furthermore, we find that different types of promotion expectation have different effects on senior executives’ perceptions. Compared with intra- group promotion, political promotion expectation has greater effects on limiting executive compensation and narrowing the pay gap. However, we find that promotion expectation fails to mitigate agency conflicts between shareholders and managers and further motivates the executives to maximise firm value. 1. Introduction Company executives play an important role in wealth creation for shareholders. However, due to the agency problem, they do not necessarily align themselves fully with the share- holders. Thus, it is important to use various incentives to motivate managers in order to mitigate agency conflicts (Jensen & Meckling, 1976). There are several mechanisms that firms can apply to motivate executives. In addition to monetary compensation such as salaries, bonuses, stocks, options and perks, there is non-monetary compensation such as political promotion (Zhang, 1999). Within the special institutional background of China, the managers of state-owned enterprises (SOEs) in China all have political ranks. They have the same status as the government officials who have the corresponding political rank (Yang, Wang, & Nie,2013). The appointment and dismissal of SOE managers are the responsibility of the department of local government organisations. The nature of political promotion for the SOE manager gives us an opportunity to research the relationship between political pro- motion and compensation decision. CONTACT danlu Bu budanlu@swufe.edu.cn *Paper accept by heng Yue. © 2016 a ccounting Society of China 54 D. Bu ET a L. Since using monetary compensation may directly reduce the shareholders’ claim on the company’s wealth, agency problems still exist where managers cannot fully align their inter- ests with shareholders’ interests (Jensen & Meckling, 1976). In contrast to promotion accom- panied by monetary compensation, political promotion in China may not directly affect shareholders’ own wealth, but it could improve managers’ career development with the benefits of enjoying greater future economic and political power. Political promotion is often not explicitly based on official contracts. It not only links to performance (Cao, Michael, Pan, Qian, & Tian, 2011; Yang et al., 2013) but also has a relationship with some subjective and non-quantitative indicators, such as social welfare activities, employment opportunities, ecological construction, safe production and philanthropy (Campbell, 2008; Gibbs, Merchant, Van der Stede, & Vargus, 2004; Jia & Nie, 2015; Zheng, Li, Xu, Lin, & Zhao, 2012). Political promotion will not dilute shareholders’ wealth but it can bring benefits to managers. In China, political promotion does not lead directly to increased compensation. u sually, gov- ernors’ compensation is strictly regulated by government, therefore governors receive less remuneration than enterprises’ executives. We are not aware of any empirical evidence on the issue of whether the expectation of political promotion can influence executives’ demands for explicit compensation. Thus, this paper attempts to investigate the impact of political promotion expectation on executive compensation. a s disclosure of executives’ compensation has been mandated at an individual level since 2005 by the China Securities Regulatory Commission, the executives’ high compensation amount and the subsequent pay gap between them and the employees have attracted enormous public attention in China. In contrast to executives’ higher compensation, 448 chairmen in China received zero compensation in 2008, which accounts for approximately 37% of Chinese listed companies. From our analysis of the CSMaR database, during 2006– 2011, only about 21% of Chinese listed companies narrowed their pay gap in any year com- pared with the previous year. We also observed that in some firms the pay gap became smaller despite increasing firm performance. Theoretically, managers, including executives, have the power to influence compensation contracts (Bebchuk, Fried, & Walker, 2002; Quan, Wu, & Wen, 2011), which may increase the pay gap between executives and employees (Li & Hu, 2012). We investigate the motivation that drives managers to choose ‘Zero-payment’ and explain the phenomenon of the lower pay gap. We explain later in the paper that this is a specific feature of political promotion in China, whereby managers may choose to show their allegiance to political aims by accepting a lower level of remuneration in order to maintain a lower pay gap. We also ask whether there are other ways for an executive to pursue private benefits to complement insufficient compensation. Generally speaking, executives in China SOEs are appointed by the government and have certain official ranks, resembling those of government officials. Therefore, the SOE executives are entitled to similar political power, most of them expect to be government officials to gain governors’ power. That is because, according to the governors’ promotion route in China, government officials have greater scope for promotion and a longer political career (Yang et al., 2013). In 2006, the Central Committee of the Communist Party of China issued ‘the provisions on the official exchange between the enterprises and governments’, which clearly state the requirement of selecting outstanding SOE managers and promoting them into the government and the administrative system. Thus, the official exchange between SOEs and according to Sina f inance report, in 2008, 448 China listed company managers receive ‘zero compensation’, see: http://finance.sina.com.cn/roll/20090425/05032807627.shtml CHINa JOuRNaL OF aCCOuNTING STuDIES 55 government is widespread, and is mainly a unidirectional flow of executives in enterprises to be officials in the government (The Research Group of unirule Institute of Economics, 2011; Yang et al., 2013; Zheng et al., 2012). Therefore, the SOE executives have a strong incentive to advance their political careers (Yang et al., 2013; Zhang, 1999). Due to the political advantages, SOE executives pay more attention to political promotion. If executives’ political performance is outstanding, they can gain political promotion and hence enjoy more ben- efits (such as non-pecuniary compensation, Non-statutory income and political power) (Chen, Chen, Wan, & Liang, 2009). Meanwhile, the guiding principle of promoting equality and efficiency has shifted from efficiency to social equality since 2004. Thus, how to reverse the pay gap has become a primary concern of Chinese society. Therefore, the government initiated reforms in income allocation and issued regulations limiting pay to eliminate the pay gap. under the influence of ‘Compensation Cap’, senior executives accomplish their duty of closing the pay gap and send a positive signal to the public, which maintains a good image. Thus, this creates a research opportunity for investigating whether a promotion expectation would inspire executives, with the right to determine their own compensation, to voluntarily close the pay gap. Prior studies have focused on the incentive effect and economic consequences of the political promotion of government officials or SOE executives. For government officials, political promotion is more important than local fiscal revenue (Blanchard & Shleifer, 2001; Zhou, 2002). Moreover, this incentive brings growth of gross domestic product (GDP) to the province (Li & Zhou, 2005; Zhou, Li & Chen, 2005), and it will also lead to local protectionism and haphazard investment (Zhou, 2004). For SOE executives, there is a positive relationship between the incentive for political promotion and performance (Cao et al., 2011; Groves, Hong, John, & Barry, 1995; Jenter & Kanaan, 2015), but there are also aspects with a negative effect such as image project building and risk-taking behaviour (Zheng et al., 2012). However, the literature relating to executive political promotion generally tests the effect of political promotion on officials’ behaviour (Li & Zhou, 2005; Wang, Fu, Huang, & Wang, 2014; Yang et al., 2013). Such a method of testing may experience a problem of endogeneity. Commonly, if the former executive is being promoted to government departments, the current executive may also have the chance to be promoted to the government depart- ments. For example, Zongting Liu, the CEO of Fujian Mingdong Electric Power Limited Company (Stock: 000993), was promoted to the Deputy Secretary of Ningde Municipal Government in the year 2007. Bangheng He, the successor CEO after Liu, was promoted to the Forestry Bureau of Ningde in 2011. Lianxuan Zhang, the chairman of Tianjin Reality Development (Group) Co., Ltd (Stock: 600322), was promoted to the Deputy Director of Tianjin Municipal Construction and Management Commission in 2006. Yong Zhang, the t he unirule i nstitute of economics (unirule) is an independent, non-profit, non-governmental (ngo ) think tank, which was jointly initiated in July 1993 by five prominent economists, Professor Yushi Mao, Professor Shuguang Zhang, Professor hong Sheng, Professor g ang f an, and Professor Shouning t ang. unirule is dedicated to the open exchange of ideas in economics in general, with a particular focus on institutional economics, and maintains a highly prestigious status within academic circles. http://english.unirule.org.cn/ t he central government of China has issued some legislation on the compensation regulations, such as, ‘notice of the Ministry of f inance on the issues relevant to the management compensation in the State-o wned and State-Controlled financial enterprises’ (2009), ‘regulations on the executive compensation distribution in the State- o wned financial institutions’ (2009) ‘guiding opinions on further regulating executive compensation in the central-government controlled enterprises’(2009) and ‘notice of g eneral o ffice of the State Council on work division concerning further deepening reform of the income dis- tribution system’ (2013). t he results are given in t able 9. 56 D. Bu ET a L. successor chairman after Zhang, was promoted to the Deputy Secretary and mayor of Wuqing District, Tianjin in 2010. These examples are indications that both former and current executives who achieved promotion are present in our sample. Based on the above analysis, this paper measures the political promotion expectation of current executives by the political promotion status of former senior executives. If the former executives gain promotion to government or achieve a higher position in the group, we regard the current executives as having strong expectations of promotion. We use data from 2005 to 2012 in Chinese SOEs to test the effects of the executive’s promotion expectation on senior executives’ compensation as well as the pay gap between executives and employ- ees. We find that there is a significant negative relationship between promotion expectation and the pay gap. The senior executives who have stronger promotion expectations are more willing to accept less compensation and a lower pay gap. We further find that different types of promotion expectations have different effects on the behaviour of senior executives. Compared with intra-group promotion, political promotion expectation makes the pay gap smaller because the compensation of current executives is lower. The higher the official rank to which the former executives are promoted, the lower the compensation and the smaller the pay gap the current executives can accept. However, we find that the promotion expec - tation fails to mitigate the agency conflicts between the shareholders and managers and fails to motivate the executives to maximise firm value. Our study tries to provide an expla- nation for the ‘zero-payment’ phenomenon in China SOEs from the perspective of implicit incentives, and provides empirical evidence supporting incentive theory. Our results indicate that the executives tend to achieve promotion by using corporate resources for non-pro- ductive rent-seeking and they establish a personal social network to achieve their own personal interest, which intensifies the agency problem and weakens the profitability of the firm. Our paper contributes to prior studies in three aspects. First, we provide an explanation of the common ‘zero-payment’ phenomenon in China SOEs from the perspective of implicit incentives, and provide evidence for SOE regulators to set up management incentive con- tracts. a lthough there is prior literature testing compensation contracts based on optimal contract theory or management power theory (Bebchuk et al., 2002; Fang, 2009; Quan et al., 2011), few prior studies have investigated the ‘zero-payment’ phenomenon in China. Based on promotion expectation and implicit incentives, our paper provides new evidence and explanation for compensation contract theory. Second, compared with previous studies on the positive effect of political promotion (Cao et al., 2011; Wang et al., 2014; Yang et al., 2013), our paper provides evidence of negative effects of political promotion. This paper extends the framework of political promotion and provides evidence for regulators on how to design the executives’ incentive mechanism in SOEs. Third, our study systematically anal- yses types of executives’ promotion, distinguishing political promotion from intra-group promotion and classifying the promotion level of the executive. It helps us to understand the promotion route for China officials. Meanwhile, our setting avoids the endogeneity problem to some extent where promotion expectation is measured by the former execu- tives’ promotion. The rest of the paper is organised as follows. Section 2 discusses the insti- tutional environment that provides the setting for our analysis and the development of our hypotheses. Section 3 introduces our data sources and models. Section 4 presents our main empirical results and discussion. We run further analysis in Section 5, and conclude in Section 6. CHINa JOuRNaL OF aCCOuNTING STuDIES 57 2. Institutional background and hypothesis development 2.1. Institutional background The media coverage around executive compensation has attracted intensive public attention to the disclosure of the pay of individual executives in China since 2005. In addition, relatively transparent executive remuneration information reveals severe pay gap issues in a sensitive way. Examples like several millions executive compensation in financial institutions and monopoly industries have triggered heated public debates on executives’ pay. a ccording to the National Bureau of Statistics of PRC, China’s Gini coefficient reached 0.473 in 2013, which is larger than the internationally prescribed generally reasonable level and much larger than that of other countries at a similar level of development ( The World Bank, 2012). Thus, how to reverse the pay gap has become a primary concern of Chinese society. In 2004, the Fourth Plenary Session of the 16th Central Committee of the Communist Party of China first proposed the aim of building a ‘harmonious society’ and the determination to ‘take effective measures to solve the problem of an excessive pay gap’. In 2012, the Third Session of the 11th National People's Congress proposed ‘firmly reversing the widening trend of the income gap’ and the Fifth Session of the 11th National People’s Congress proposed ‘reversing the trend of the widening pay gap as soon as possible’. Meanwhile, in order to narrow the serious pay gap, the government began to promote reform targeting the rationalisation of income distribution with measures such as issuing a series of regulations restricting executive compensation and narrowing the pay gap. In 2009, the Ministry of Finance issued ‘Notice of the Ministry of Finance on the issues relevant to management compensation in the State-Owned and State-Controlled financial enterprises’, ‘Regulations on the executive compensation distribution in the State-Owned financial insti - tutions’, which clearly required that the executive compensation of the financial enterprises should be no more than CNY2.8 million, and that the executive compensation in 2008 should be no more than 90% of that of 2007. If the performance declines compared with 2007, the executive compensation should be further reduced by another 10% to 80% of that in 2007. In 2009, the Ministry of Human Resources and Social Security, the Ministry of Finance and another six ministries jointly issued ‘Guiding opinions on further regulating executive com- pensation in the central-government controlled enterprises’, which expresses a clear require- ment that SOE executives’ base salary should link to the employees’ average wages of the previous year. In 2013, ‘Notice of General Office of the State Council on work division con- cerning further deepening reform of the income distribution system’ proposed ‘to strengthen the management of executive compensation in SOEs. Moreover, executive compensation increases should be no higher than that of employee average increase’. Issued in a ugust 2014, under the leadership of Human Resources and Social Security of PRC and with partic- ipation by the Ministry of Finance as well as other ministries, another compensation regu- lation requests that compensation of executives in centrally controlled enterprises and Stated-owned finance firms should be decreased to about 30% of the present level and cannot exceed CNY 600,000. Moreover, the base salary of CEOs in central controlled http://news.xinhuanet.com/2014-01/20/c_119040570.htm t he gini coefficient was put forward by the i talian economist gini in 1922 and is an index that quantifies inequality among values of an income distribution. t he value of the gini coefficient is a number from 0 to 1. a value of zero means that all have the same income. a ccording to the international standard, if the coefficient is lower than 0.2, the disparity in income is low. if the coefficient is more than 0.5 then the disparities of income are relatively high. 58 D. Bu ET a L. enterprises should be twice the average annual pay of employees, the performance pay should be twice the base salary, and tenure incentives should be no more than 30% of the overall pay. a ll these restrictions signal that the government has been striving to narrow the increasingly widening pay gap. 2.2. Hypothesis development a compensation contract seeks to provide an optimal institutional arrangement to mitigate the conflicts between shareholders and agents arising from the separation of ownership and operation rights (Jensen & Meckling, 1976). With the gradual development and estab- lishment of China’s market economy system, aligned with further reform undertaken in SOEs, the government has issued a large number of laws and regulations on the enterprise com- pensation system, and a performance pay scheme also begins to take hold in China’s listed companies (Fang, 2009). However, SOEs still retain multiple non-profit goals in the process of market-oriented reformation. Specifically, the fiscal decentralisation reform left local gov - ernments with the right to gain more profit from enterprises located within the jurisdiction area, which would drive the local governments to retain influence over the enterprises within their jurisdiction area (Chen et al., 2009). a lthough a market economy mechanism and mod- ern corporate governance are gradually being established, the impact of administrative power on the resource redistribution still exists (Li, Qiu, & Gu, 2010). Due to the weak investor protection system and low efficiency of law enforcement, China is regarded as having an inefficient system of corporate governance (Firth, Fung, & Rui, 2006). The effect of dual governance environments (with both official governance and economic governance) (Li, Qiu, Niu, & Xu, 2010) on the incentive mechanism is seen in two ways. One is that although the official rank of executives is gradually being phased out, their appoint - ments, instead of being selected by the market, are still controlled by the government (Liu, 2001). The cross-border exchanges of SOE executives and government officials are wide - spread and many SOE executives are now, or have once been, government officials (Fan, Wong, & Zhang, 2007; Liang, Li, Chen, & Chen, 2015; Zhang, 1999). Hence, many of them would reasonably embrace expectations to be appointed as civil servants after accumulating experience in enterprises. Therefore, China’s SOE executives, who exhibit the characteristics of ‘half Officials half Entrepreneurs’, are more inclined to develop their political career rather than becoming professional managers (Xin & Tan, 2009). Meanwhile, the present overwhelm- ing political objectives of maintaining social fairness and stability, reducing local fiscal deficits but increasing fiscal surpluses, has led the compensation of SOE managers to be under strict government control (Chen, Chen, & Wan, 2005). a more diversified non-monetary compen- sation system, such as the economic incentive of perks and the administrative incentive of political promotion, may appear as a substitution for a direct monetary reward system when monetary pay-offs are restricted (Chen et al., 2009). Hence, managers’ motivation to pursue non-monetary interests such as political promotion would be strengthened when facing this unique institutional environment. Moreover, due to the impact of ‘official-oriented‘ ide - ology, political promotion is very attractive to executives, because it is accompanied by greater political power, more prestigious position, more perks and even rent-extraction (Chen et al., 2005), and governments (central and local) also tend to use political promotion as the primary way of motivating SOE executives (Zhang, 1999). Hence, political promotion CHINa JOuRNaL OF aCCOuNTING STuDIES 59 has become an important concern for SOE executives during the development of their professional careers (Yang et al., 2013). Official intervention and pressure from political goals have driven SOEs to target aims more than maximising profit (Lin, Liu, & Zhang, 2004). Consequently, there may be ambiguity in establishing a causal relationship between company performance and managers’ effort. The existence of multiple aims also weakens the effectiveness of a compensation contract based on performance. In this particular economic transition period, Chinese SOEs undertook multiple objectives, such as to obey the development strategies of the national economy, macroeconomic regulation, employment, and social stability as a result of the prevalent political intervention (Lin et al., 2004). under the institutional environment, where the pro - fessional destinations of executives in SOEs are mainly determined by the government, managers in SOEs would have to comply with the orders of government officials, who are influential and decisive in the selection, promotion and compensation setting of SOE man- agers, concerning the operation activities of the SOEs under their charge. Therefore, local government has the motivation and capacity to intervene in SOE production and business activities (Li & Zhou, 2005; Lin & Liu, 2000). The government intervention will often distort managers’ behaviour. What the managers care about is not company performance, but their own political careers. Theoretically, government officials’ assessment towards of SOE man- agers should focus on both economic and political performance (Zhang, 1999). However, sometimes, managers’ political performance will outweigh economic performance, that is, managers’ careers will continue to move upward rather than downward as long as they are politically doing well despite the general economic performance (Zhang, Liu, & Cai, 2015). This is because the multiple objectives of an SOE decrease the relevance of the manager’s effort and operating performance (Lin et al., 2004; Zhang, 1999). Thus, in order to ease the political burden of government at all levels, or to achieve the political promotion goal of government officials, or both, the government officials are more interested in the SOE man- ager’s political performance than economic performance (Zhang et al., 2015). In the case that government officials’ interests are more directly related to the political performance of SOE managers, the economic tournament incentive on executives will weaken sponta- neously (Zhang, 1999, 2014). under such an institutional background, government officials’ preference towards political gains will certainly lead enterprise managers to pay more atten- tion to political behaviour in that only SOE executives with a good political performance could be eligible for promotion in association with higher political rewards, more control power benets fi (such as in-oc ffi e consumption, hidden income), more political benefit (such as political promotion) and other forms of benefit (Chen et al., 2009). Since 2004, the guiding ideology around fairness and efficiency has changed from ‘giving priority to efficiency with consideration to fairness’ to ‘social fairness and efficiency are equally important’. With the purpose of narrowing the pay gap, the government began to reform income allocation by issuing a series of policies and regulations restricting excessive exec- utive compensation. under the background of a ‘compensation restriction order’, catering to the political goal of narrowing the pay gap can help corporations as well as executives to establish a good public image, to signal good news to outsiders vulnerable to information and to gain the public, media and government authority’s attention, which would in turn help executives’ personal interests, especially political promotions (Wang et al., 2014). Therefore, executives with promotion expectation are more willing to implement govern- ment policies and regulations. By doing so they can lay a solid foundation for future political 60 D. Bu ET a L. promotion and career development. Executives with an expectation of promotion are more inclined to narrow the pay gap in order to gain promotion advantages. Thus, based on the foregoing analysis, we propose our first hypothesis. H1: There is a negative correlation between promotion expectation and pay gap. a t present, the central government is vigorously promoting the reformation of separating enterprises from the government administration and conducting de-administration refor- mation in SOEs because executives in SOEs having government administrative ranks are still governed by the respective party committees. Specifically, most centrally controlled enter - prises are directly restructured from subordinate enterprises of the State Council’s professional ministries, and some of their top leaders are naturally shifted directly from the professional ministry leaders. a lthough central enterprise executives are not formal government officials/ civil servants they are managed and paid according to the corresponding governor scale ( The Research Group of unirule Institute of Economics, 2011; Yang et al., 2013). Obviously, all SOE executives are ‘quasi-officials’ being branded as administrative officials (Yang et al., 2013). Subsequently, these ‘quasi-officials’, similar to their government peers at the same level, receive similar political treatment and opportunities for political promotion (Yang et al., 2013). Most of the central enterprise executives personally expect to be promoted to the Party and to be government officials in power. Even if they achieve the status of ‘person in charge’ in the parent firm, their administrative level will be vice-ministerial level at most and there will be no further promotion thereafter (Yang et al., 2013; Zhang et al., 2015). Only by transferring from a ‘quasi-official’ to a ‘real official’ would it be possible to obtain a higher political promo - tion at administrative level by which to continue political life, such as being promoted to be the ‘person in charge’ in central ministries or local provincial government (ministerial level) after which an official would qualify to be a candidate for central leadership (Yang et al., 2013). In contrast to General Party and government cadres, the SOE executives equipped with a technical background and industrial experience are highly proficient in the market oper - ation and specialised in promoting the development of unique local industries. Therefore, the General Office of the Communist Party of China issued ‘Provisions on the exchange among the party and government cadres’ in 2006, clearly stating the requirement for select- ing outstanding SOE leaders to be appointed in the Party and government departments while recommending the Party and government cadres work in SOEs and similar institutions. This, however, provides ways for SOE executives to be promoted to the Party and government departments. For SOE executives there are two different types of promotion. One is to be promoted to an office in government. The other is to be promoted to a higher position in the firm. The differences are that being an official in government departments should be more beneficial than being an executive (Chen et al., 2009). On the one hand, it provides an opportunity for gaining higher political ranks as well as the continuation of political life. On the other hand, a government official will have easy access to scarce resources such as per - sonnel assignment rights and reallocation of other important materials (Yang et al., 2013). In contrast, intra-group promotion is a business position promotion. Even where an executive t he levels of civil servants regulated in China are nine in total, namely clerk, vice section, the section, vice department, the department, vice provincial department, the provincial department, vice-ministerial and ministerial in accordance with the civil servant rules. t he 53 top Central Soes in a list of central enterprise rankings issued by the SaSa C are ‘vice-ministerial central enterprises’. heads (chairman, general manager or secretary of the party committee) of these top 53 enterprises are appointed and managed jointly by the central organisation department and Bureau i Management managing corporation leaders subordinate to SaSa C in accordance with vice-ministerial official management codes. CHINa JOuRNaL OF aCCOuNTING STuDIES 61 is promoted to be the head of a central/parent enterprise, the corresponding administrative rank is vice-ministerial level. In many cases the executive is considered to have reached the ceiling of a business career. In that situation, their control of scarce resources is weaker rel- ative to that if politically promoted (Wang et al., 2014). The current executive appointment system in China SOEs implies a tournament among SOE executives who are motivated to strive for winning such a tournament (Li & Zhou, 2005; Zhou, 2004, 2007). The adoption of such a mode of appointment induces SOE executives to concentrate more on political promotion than economic ones, and in turn to obtain a higher likelihood of political promotion. Executives will often try their best to cater to political objectives, aiming at being more eligible for political promotion (Zhang et al., 2015). Such eo ff rts will, however, help to either relieve the political burden of local government, or make the political promotion goals of government officials come true, or both. This consideration, together with the background of a ‘compensation restriction order’, means it is reasonable for SOE executives to voluntarily reduce monetary compensation and eliminate pay differ - ences with employees. In doing so they may also build a favourable reputation that is helpful in their political promotion. Thus, compared with the expectation of intra-group promotion, the executive shows a stronger motivation to enter politics and is more likely to accept a lower monetary compensation and a smaller pay gap when facing the opportunity of polit- ical promotion. Thus, based on the analysis, we propose the second hypothesis. H2: Compared with internal/intra-group promotion expectation, the political promotion expec- tation has a larger negative effect on pay gap and makes the pay gap smaller. Gibbons and Murphy (1992) found that executives not only pay attention to the short-term benefits, but are also concerned about the performance that brings them competitive strength in the labour market. The career concerns can have important effects on incentives. The optimal compensation contracts neutralise career concern incentives by optimising the total incentives from the contract and from career concerns: explicit contractual incentives are high when implicit career concern incentives are low, and vice versa. The idea that optimal incentive contracts optimise total incentives can be applied to promotions (Baker, Jensen, & Murphy, 1998; Gibbons & Murphy, 1992). For Chinese SOEs, the higher the political rank of the company, the more positive the effect of career attraction for executives. So there is a substitute relationship for political ranks and compensation (Huang, Li, & Li, 2011). In China, there are two main factors determining the executive’s political promotion, one is the admin- istrative rank of the present ultimate controller of the SOE where the executive works, in which case the executive’s rank after promotion should be no higher than his peers whose ultimate controllers’ rank is in the same, and the other is the corresponding administrative rank of the executive’s specific position. The administrative level of the SOE’s ultimate controller is influ- ential over the executive’s political promotion space (Yang et al., 2013; Zhang, 1999). Since the administrative rank of the ultimate controller in the central enterprise is higher than that of the local enterprise, the executives in the central enterprise are endowed with easier access to government officials, therefore executives in the central SOEs focus more on political development and promotion (The Research Group of unirule Institute of Economics, 2011). a larger promotion space drives managers in centrally controlled enter - prises to obey the orders of the government at a higher level and restrain their own oppor- tunistic behaviour, compared with the managers in local government-controlled companies where executives are confined to a limited political career and a relatively smaller promotion 62 D. Bu ET a L. space (Yang et al., 2013). For a long time, state-controlled listed companies’ executives regarded the ‘Official Standard’ as the goal and the SOEs have implemented their compen- sation systems in such a way that the compensation has been significantly sensitive to the political rank, and rewarded executives differently according to their corresponding admin - istrative ranks (Zhang, 1999). In addition, executives at different administrative levels also receive different treatment in other aspects, such as in health care and welfare. It is obvious that a high administrative level itself and the opportunity of increasing administrative levels are incentives. Generally, the setting of executive compensation is based on the administra- tive level and average wages of workers in the same industry (Yang et al., 2013; Zhang, 2014; Zheng et al., 2012). It can be seen that there exists a different effect on the firm value of state-controlled companies, even of the same nature, as a result of their ultimate controllers’ executive ranks, which also exert different influences on executive compensation as well as agency cost. a ccording to the incentive theory (Maslow, 1954; Baker, Robert, & Kevin, 1994), the different political ranks will bring different benefits and generate different promotion motivations for an executive, and then the political ranks have different substitutive rela- tionships with compensation. However, the administrative ranks of executives in SOEs are not always the same. Take central SOEs for example. The top 53 central SOEs in a list of central enterprise rankings issued by the State-owned a ssets Supervision and a dministration Commission of the State Council (SaSa C) are ‘vice-ministerial central enterprises’, also known as the backbone of the central enterprises (Yang et al., 2013). Heads of these top 53 enterprises (chairman, general manager or secretary of the party committee) are appointed and managed jointly by the central organisation department and Bureau I Management of SaSa C in accordance with vice-ministerial official management codes. Separate from the top 53 firms, the other central enterprises in the list are the ‘provincial level central enterprises’, whose responsible persons are managed by the Bureau II Management of SaSa C according to the management of ‘provincial level’ government officials, which still needs formal approval from the central organisation department. SaSa C manages central enterprise leaders with reference to gov- ernment officials/civil servants who have the same political ranks (Yang et al., 2013). Since the establishment of SaSa C, the power of personnel appointment and dismissal of central enterprises has continued to shift from the central organisation department to the SaSa C, which forms a personnel management system with consolidated supervision on all leaders of central enterprises on the basis of SaSa C but supplemented by the central organisation department (Yang et al., 2013). In China, with the deepening of reformation of the govern- ment system, the central government has gradually delegated some power of administrative examination and approval to the local government, which enables the local government to exercise some powers (for example, approval authority and appointment power) (Zhang, 1999). Due to the difference in job position levels, government officials at different levels differ significantly in social connections, power and resources (Dai, Hong, & Pan, 2015). For example, compared with department-level officials, officials at the minister-level possess more relationship networks and power than the department-level officials. These differences, whether in power or resources, will cause different executive behaviour. The larger the polit - ical promotion space, the stronger the motivation to cater to the purpose of government t here are 112 central Soes who are supervised by SaSa C up to d ecember, 2015. in the list, the top 53 firms are ‘vice-minis- terial central enterprises’ according to management policy by SaSa C and the Communist Party Central Committee's organization d epartment. http://www.sasac.gov.cn/n1180/n1226/n2425/ CHINa JOuRNaL OF aCCOuNTING STuDIES 63 regulatory authorities and the incentives to implement corresponding policies ( Wang et al., 2014). Therefore, under the background of ‘compensation restriction requirement’, a broader political promotion space for executives serves as a stronger incentive for them to cater for political goals as well as accepting a smaller pay gap for the purpose of gaining political promotion capital. Thus, based on the foregoing analysis, our third hypothesis is stated as follows: H3: There is a negative relation between political promotion space and pay gap. 3. Research design 3.1. Variable definitions 3.1.1. Dependent variable: executive compensation (COMP), employee wage (WAGE) and pay gap (GAP) The executive compensation is mainly composed of monetary compensation and equity incentive. In the paper, we only investigate the monetary compensation of executives. Some executives receive a salary from the parent company and only receive an allowance (for expenses such as travel and conferences) from the listed firm. The independent directors only receive an allowance from the listed firm. We add up the executive compensation and exclude executives and independent directors who receive only an allowance, then we divide the result by the number of executives receiving compensation, and calculate the resulting average executive compensation. Following Li and Hu (2012) and Liu and Sun (2010) the employees’ average wage is equal to payments to employees, based on information in the Cash Flow Statement, and excluding the total compensation of directors, supervisors and executives, then scaled by the number of employees excluding the number of directors, supervisors and executives. The pay gap is the difference between the executives’ average compensation and employees’ average wage (Eriksson, 1999; Li & Hu, 2012; Siegel & Hambrick, 2005). We then take the natural logarithm of compensation to address the right skewness of the data. 3.1.2. Independent variable: promotion expectation The database, China Stock Market & a ccounting Research Database (CSMa R), does not track executive departure destinations after departure. We use the ‘Baidu’ search engine to find relevant information about the executives regarding their ultimate destination after depar- ture. We define the former executives as the chairman and general manager, and both of whom have great influence on strategic decisions and operating decisions, Moreover, their decision-making behaviour has a critical effect on performance (Li, 2000). Supervisors in Chinese companies are top managers, and they always receive higher compensation, therefore, we exclude them when we calculate employee payments. We follow prior Chinese literature about executive compensation to measure executive compensation as the average pay of directors, supervisors and executives. Such as the chairman Yanfeng Zhu of fa W Car (stock code: 000800), who worked as vice governor of Jilin Province after quitting the job in d ecember 2007. under the premise that we do not know anything about their destination, we enter ‘Zhu Yanfeng, fa W’ in the Baidu search engine, then the Baidu search engine will give us the relevant information about Zhu Yanfeng, and according to this, we can confirm that after Zhu Yanfeng left the job, he worked as vice governor of Jilin Province. 64 D. Bu ET a L. In the paper, we define promotion expectations as follows. (1) PROM . We consider departure from the company ( job-hopping) of a former exec- utive as a promotion if it happens to fall into at least one of the three following cases, (i) going forward to a government position; (ii) being the (vice) chairman, the general manager or the vice general manager (being only confined to the case where a general manager becomes the vice general manager of the parent firm while the transfer from chairman to general manager is treated as a non-promotion); (iii) becoming the chairman within the company (refers to change from general manager to chairman) or undertaking a similar position in companies at the same level. We use the promotion of the former executive as the proxy for the promotion expectation of the successor. When the executives have a promotion expectation, PROM is equal to one, otherwise it is zero. (2) PROM . We consider the destination of the former executive as the expectation of his successor. We give the value 2 to a promotion expectation, 1 to a lateral transfer, 0 to other and value –1 to a demotion. a ccordingly, PROM takes the value of 2, 1, 0, –1. (3) PROM . If the former executive acquires a government position after leaving the company, the PROM is equal to one, and zero otherwise. (4) PROM and PROM . PROM and PROM consider the levers of political promotion 4 5 4 5 expectation. PROM represents the executive levels of government department and PROM represents the executive levels of job ranks. Detailed explanations of the Promotion variables PROM and PROM are provided in Table 1. 4 5 3.1.3. Control variables Based on prior studies (Fang, 2009, 2011; Firth et al., 2006; Leone, Wu, & Zimmerman, 2006; Lu, Wang, & Zhang, 2012), we include the following control variables: (1) Ownership char- acteristics: Central SOE (C-SOE) equals one when a SOE’s ultimate controller is the SaS a C, the Treasury or other SOEs and zero otherwise; TOP is the stock proportion of the largest share- holder. (2) Board characteristic: DUAL equals one when the chairman also takes the position of the general manager, and zero otherwise; BOARD captures the size of the board; Independent Directors (INDP) is the number of independent directors on the board. (3) Financial characteristics: ROA is the ratio of total assets divided by net profit; SIZE is the natural logarithm of total assets; Leverage (LEV) is the ratio of total debt to total assets. (4) Industry (IND) and year dummy variables are included to control for the impact of industry and year. Detailed explanations of the variables in the paper are provided in Table 1. 3.2. Model We estimate the following equation to test the three hypotheses: GAP =  +  × PROM + Controls + (1) i,t i,t i,t i,t In model (1), the promotion variable is represented in turn by PROM , PROM , PROM , and 1 2 4 PROM . If the coefficient β of promotion is negative, it means that executives with a promo - tion expectation tend to narrow the pay gap, which is consistent with hypotheses 1 and 3. Whether the measurement of promotion expectation is reasonable or not should be examined. We test the rationality from two aspects which are given in 5.3. CHINa JOuRNaL OF aCCOuNTING STuDIES 65 Table 1. Variable definitions. Variable Definition COMP t he natural logarithm of executive average compensation WAGE employee average wage = t he natural logarithm of (cash paid to employees minus total compensation of directors, supervisors and executives)/(the number of employees minus the number of directors, supervisors and executives) GAP Pay gap = t he natural logarithm of (executives’ average compensation minus employees – average wage) PROM implies promotion type where former executives (either the chairman or the general manager) are promoted to (1) forward to a government position; (2) being the (vice) chairman, the general manager or the vice general manager (being only confined to the case where a general manager becomes the vice general manager of the parent firm while the transfer from chairman to general manager is treated as non-promotion); (3) becoming the chairman within the company (only refers to change from general manager to chairman) or undertaking a similar position in companies at the same level. We define the former executive’s promotion as a signal of promotion expectation for the successors, which is treated as a dummy variable (Promotion ) that is coded one if the executives have a promotion expectation, and zero otherwise. PROM We assign values 2, 1, 0, –1 to executive promotion expectation, lateral transfer, demotion and other respectively. PROM Coded one if the former executive (chairman or general manager) departed for a government department position, and zero otherwise. PROM l evels of government department where executive positions are are divided into central, province, city, county, town, within the group and other that are assigned the value 5, 4, 3, 2, 1, 0, –1 respectively, and the higher the value, the higher the level. PROM t he levels of civil servants regulated in China are nine in total, namely clerk, vice section, the section, vice department, the department, vice provincial department, the provincial department, vice-ministerial and ministerial in accordance to the civil servant rules, we assign the value 0 to promotions within the group and –1 to all other departures mentioned above. in light of such occasion, Promotion takes the value of –1, 0, 1, 2, 3, 4, 5, 6, 7, 8 correspondingly. expense ratio t he management expense scaled by sales asset utilisation ratio t he sales revenue scaled by total assets BOARD t he size of the board. INDP t he proportion of independent directors. DUAL When the chairman also takes a position of general manager, that is, a dual-position, the value is 1, otherwise 0. TOP t he largest shareholder’s share proportion. ROA t he total return on assets = net profit/ total assets LEV d ebt-to-asset ratio = total debt/total assets. SIZE t he natural logarithm of company total assets at the end of the year. C-SOE When the ultimate controllers are the SaSa C, the Ministry of f inance, other central agencies or the Soes directly report to central government, the value is 1, and otherwise 0. YEAR Year dummy variable. IND industry dummy variable. PC an indicator variable equal to one if the executive has past or concurrent work experience in the government or a political appointment, such as the People’s representative or Member of Chinese People’s Political Consultative Conference. EDUCATION an indicator variable equal to 5, 4, 3, 2, or 1 if the executive’ education is Phd , Master, Bachelor, College or senior middle school. AGE t he executive’s age in a given year. TENURE t he age of the executive being the chairman or general manager in a given year. To distinguish the effect of two types of promotion expectation, we include the interaction term of PROM1*PROM in model (1). The regression coefficient of political promotion expec - tation is (β +β ), and the intra-group promotion is β . a negative β means that the effect of 1 2 1 2 political promotion expectation on pay gap is more significant, which is consistent with Hypothesis 2. 66 D. Bu ET a L. 3.3. Data and sample Our sample consists of all the firms listed in the Shanghai and Shenzhen Stock Exchanges over the period of 2005–2012 where their ultimate controller is governmental or an SOE. a fter excluding financial firms and firms with missing financial information, our final sample consists of 6,475 firm-year observations. Information on promotion expectation is hand-collected. a ll other data are obtained from the China Stock Market and a ccounting Research (CSMaR) database. a ll continuous variables are winsorised at the 1% and 99% levels. Table 2 reports the sample distribution by year and industry. The annual trend in PROM and PROM suggests that the number of promotion expectations increased throughout our sample period from 2005 to 2012. Table 2 also reveals that promotion expectation firms tend to be located in petroleum, metal/non-metal, equipment, pharmaceutical products, retailing, or energy supply. Table 2. Sample distribution. Panel a: Distribution by year Obs. Obs. of PROM Obs. of PROM 1 3 2005 726 72 17 2006 718 109 22 2007 745 140 32 2008 787 172 43 2009 822 206 42 2010 841 219 42 2011 886 264 52 2012 950 280 57 t otal 6,475 1,462 307 Panel B: distribution by industry obs. obs. of PROM obs. of PROM 1 3 a griculture 121 27 5 Mining 272 69 15 f ood 287 44 14 t extile 140 17 8 Paper/printing 100 14 1 Petroleum 712 143 10 electronics 247 67 24 Metal/non-metal 625 162 46 equipment 1,060 252 53 Pharmaceutical products 355 93 4 o ther manufacturing 16 0 0 energy supply 405 91 17 Construction 147 40 7 t ransport 387 83 26 information technology 300 71 15 retailing 451 94 11 real estate 361 77 29 Services 206 61 7 Media 84 10 4 Multi-industry 199 47 11 t otal 6,475 1,462 307 t his table reports the sample distribution by year and industry between 2005 and 2012. Statistics are based on firm-year observations. t he 20 industries are based on the official industry classification of the China Securities regulatory Com- mission. PROM is gauged whether the firm’ executive has promotion expectation. PROM is gauged whether the firm’ 1 3 executive has political promotion expectation. CHINa JOuRNaL OF aCCOuNTING STuDIES 67 Table 3. Summary statistics. Variable Obs. Mean Min P25 P50 P75 Max SD COMP 6,475 30.93 3.80 14.41 24.39 39.05 149.61 24.91 WAGE 6,475 7.39 1.16 3.84 5.97 9.21 30.78 5.33 GAP 6,475 23.41 0.67 8.65 16.87 29.71 134.57 22.91 PROM 6,475 0.226 0 0 0 0 1 0.418 PROM 6,475 0.458 –1 0 0 1 2 0.879 PROM 6,475 0.047 0 0 0 0 1 0.213 PROM 307 3.433 1 3 3 4 5 0.819 PROM 307 4.954 1 4 5 6 7 1.557 expense ratio 6,475 0.083 0.008 0.040 0.066 0.103 0.403 0.065 asset utilisation ratio 6,475 0.740 0.073 0.381 0.618 0.931 2.785 0.517 BOARD 6,475 9.573 5 7 9 11 15 1.927 INDP 6,475 0.359 0.250 0.333 0.333 0.375 0.556 0.049 DUAL 6,475 0.097 0 0 0 0 1 0.295 TOP 6,475 39.565 10.160 27.190 39.040 51.110 75.400 15.507 ROA 6,475 0.033 –0.179 0.011 0.031 0.057 0.198 0.054 LEV 6,475 0.520 0.077 0.386 0.536 0.661 0.902 0.188 SIZE 6,475 22.010 19.702 21.109 21.823 22.723 25.888 1.264 C-SOE 6,475 0.320 0 0 0 1 1 0.467 t his table reports the summary statistics within the sample firms between 2005 and 2012. t he unit of COMP, WAGE and GAP is ten thousand yuan. in the summary statistics of ‘PROM ’ and ‘PROM ’, samples whose executives have political 4 5 promotion expectation are dropped. all variable definitions are in t able 1. 4. Empirical results 4.1. Descriptive statistics Table 3 presents descriptive statistics of all the variables. It shows that about 22.6% of the firms in our sample provide an expectation of opportunities, political promotion or intra- group promotion, but that only 4.7% of the firms in our sample provide an expectation of political promotion. Further, from the perspective of political promotion level, it is, more often than not, provincial and municipal promotions that are available for those executives qualified for political promotion, and the positions acquirable are mostly at the vice- provincial department level. During 2005–2012, the average executive compensation is CNY309,300 in Chinese listed SOEs. However, from the standard deviation, there is a larger difference among executive compensation. The highest compensation is CNY1.4961 million, while the lowest compen- sation is only CNY38,000. In contrast, the average employees’ wage is CNY73,900 in the same period, and the difference between executive compensation and employees’ wage reached CNY234,100. The average expense ratio of the sample is 0.083, and the maximum value is 0.403. The mean value of the asset utilisation ratio is 0.740, and the minimum value is 0.073. Compared with results reported in a ng, Cole, and Lin (2000), it shows that there is a higher agency cost but lower efficiency in Chinese listed SOEs than in their peers in developed economies. t he expense ratio is, in general, an indicator used to measure how effectively resources are utilised under the control of managers for accomplishing the same sales revenue. a higher expense ratio means that more resources consumed under the manager’s control are left available to the manager’s personal interests, hence a higher agency cost. g enerally, the asset utilisation ratio, measured as the sales revenue divided by total asset, is used as an indicator to capture the operating ec ffi iency of the firm. i t may be used to answer questions such as whether the management has made inappropriate and ineffective decisions on shareholders’ invests or whether management has behaved lazily and has not sought to maximum shareholders’ wealth. t he higher the expense ratio or the lower asset utilisation ratio, the higher the agency cost between shareholders and the manager. 68 D. Bu ET a L. 4.2. Results 4.2.1. Promotion expectation and pay gap The regression results of promotion expectation on pay gap are shown in Table 4, which reports a significantly negative relationship between promotion expectation (PROM and PROM ) and executive compensation at the 1% level of significance. This implies that exec - utives in Chinese listed SOEs are inclined to accept lower compensation when facing pro- motion expectation, which, in turn, suggests a substitution effect between implicit political promotion and explicit compensation. There is a positive but insignificant relation between promotion expectation and average employee wage. The promotion expectation is signifi- cantly negatively related to the pay gap at the 1% level, which captures the motivation of executives in SOEs of gaining more promotion competency by actively lowering compen- sation and narrowing the pay difference with employees. Such a promotion expectation helps to drive executives to narrow the pay gap. Control variables in the regression are consistent with the existing literature. Table 4. Promotion expectation and pay gap. (1) (2) (3) (4) (5) (6) COMP Wa GE GaP COMP Wa GE GaP Constant 5.892*** 7.267*** 4.374*** 5.899*** 7.275*** 4.381*** (41.07) (52.17) (20.19) (41.12) (52.24) (19.01) PROM –0.081*** 0.013 –0.114*** (–5.00) (0.82) (–4.62) PROM –0.036*** 0.011 –0.052*** (–4.68) (1.43) (–4.42) BOARD 0.007* 0.004 0.012** 0.007* 0.004 0.012** (1.81) (0.97) (2.12) (1.79) (0.96) (2.11) INDP 0.071 0.154 –0.117 0.068 0.159 –0.121 (0.49) (1.11) (–0.54) (0.48) (1.14) (–0.56) DUAL –0.019 –0.087*** 0.024 –0.018 –0.086*** 0.025 (–0.81) (–3.94) (0.69) (–0.77) (–3.91) (0.72) TOP –0.004*** 0.004*** –0.007*** –0.004*** 0.004*** –0.007*** (–9.22) (9.01) (–10.47) (–9.24) (8.98) (–10.49) ROA 3.418*** 1.566*** 4.164*** 3.419*** 1.565*** 4.166*** (24.03) (11.33) (19.39) (24.03) (11.33) (19.39) LEV –0.222*** –0.309*** –0.188*** –0.222*** –0.309*** –0.188*** (–5.02) (–7.20) (–2.82) (–5.02) (–7.21) (–2.82) SIZE 0.267*** 0.126*** 0.319*** 0.267*** 0.126*** 0.318*** (38.56) (18.75) (30.47) (38.52) (18.71) (30.43) C-SOE 0.114*** 0.045*** 0.139*** 0.115*** 0.045*** 0.140*** (7.62) (3.12) (6.15) (7.71) (3.10) (6.23) YEAR Yes Yes Yes Yes Yes Yes IND Yes Yes Yes Yes Yes Yes a djusted R 0.504 0.393 0.367 0.504 0.393 0.367 f statistic 189.03 120.84 108.08 188.85 120.91 108.00 no. of obs. 6,475 6,475 6,475 6,475 6,475 6,475 t his table reports the regression results of promotion expectation on pay gap. f rom columns 1 to 3, the independent varia- ble is PROM . f rom columns 3 to 6, the independent variable is PROM . all the variables are defined in t able 1. t -statistics 1 2 used to signal the robustness of standard errors clustered at the firm level are reported in parentheses. ***, **, and *, indicating statistical significance at 1%, 5%, and 10% levels respectively. CHINa JOuRNaL OF aCCOuNTING STuDIES 69 Table 5. Political promotion expectation and pay gap. Total Sample Political promotion expectation (1) (2) (3) (4) (5) (6) COMP Wa GE GaP COMP Wa GE GaP Constant 5.874*** 7.239*** 4.353*** 5.751*** 7.150*** 4.282*** (40.92) (51.97) (20.08) (36.19) (45.68) (17.78) PROM –0.062*** 0.043** –0.092*** (–3.48) (2.46) (–3.43) PROM *PROM –0.092*** –0.141*** –0.102* 1 3 (–2.67) (–4.20) (–1.95) PROM –0.152*** –0.047 –0.216*** (–4.16) (–1.31) (–3.91) BOARD 0.007* 0.004 0.012** 0.013*** 0.002 0.020*** (1.88) (1.09) (2.17) (3.02) (0.55) (3.20) INDP 0.092 0.187 –0.093 0.145 0.212 –0.035 (0.64) (1.34) (–0.43) (0.92) (1.36) (–0.14) DUAL –0.017 –0.085*** 0.026 –0.021 –0.083*** 0.016 (–0.74) (–3.82) (0.74) (–0.88) (–3.50) (0.44) TOP –0.004*** 0.004*** –0.007*** –0.005*** 0.004*** –0.008*** (–9.12) (9.16) (–10.40) (–8.97) (7.88) (–9.90) ROA 3.418*** 1.550*** 4.153*** 3.371*** 1.583*** 4.105*** (23.96) (11.23) (19.33) (22.06) (10.51) (17.72) LEV –0.224*** –0.313*** –0.191*** –0.174*** –0.285*** –0.142* (–5.08) (–7.29) (–2.86) (–3.59) (–5.98) (–1.94) SIZE 0.268*** 0.127*** 0.319*** 0.273*** 0.131*** 0.324*** (38.63) (18.86) (30.51) (35.59) (17.39) (27.89) C-SOE 0.112*** 0.042*** 0.136*** 0.087*** 0.056*** 0.096*** (7.48) (2.91) (6.05) (5.21) (3.36) (3.80) YEAR Yes Yes Yes Yes Yes Yes IND Yes Yes Yes Yes Yes Yes a djusted R 0.505 0.395 0.367 0.508 0.391 0.367 f statistic 184.15 118.28 105.23 155.33 96.94 87.75 no. of obs. 6,475 6,475 6,475 5,243 5,243 5,243 t his table reports the results of political promotion expectation on pay gap. f rom columns 1 to 3, the independent variable is PROM and the interaction term of PROM and PROM . f rom columns 3 to 6, the independent variable is PROM . all the 1 1 3 3 variables are defined in t able 1. t -statistics used to signal the robustness of standard errors clustered at the firm level are reported in parentheses. ***, **, and *, indicating statistical significance at 1%, 5%, and 10% levels respectively. 4.2.2. Political promotion expectation and pay gap For testing different effects of political promotion expectation and intra-group promotion expectation on compensation, we introduce an interaction term PROM *PROM in the regres- 1 3 sion. Table 5 shows that the coefficient of promotion expectation is significantly negative at the 1% level, and the coefficient of the interaction term is significantly negative at the 10% level, indicating that compared with intra-group promotion expectation, political pro- motion expectation has a more significant constraining effect on the pay gap. It also suggests that a scarce resource associated with political promotion expectation can also motivate executives to reduce their compensation and send a good signal to the media and public. This also indicates that the incentive effect of political promotion expectation on executives is better than that of the intra-group promotion expectation. 4.2.3. Political promotion space and pay gap To test the effect of political promotion space on pay gap, we use government level (PROM ) and position level (PROM ) as the measure of promotion space. Table 6 shows that whether 5 70 D. Bu ET a L. Table 6. Promotion expectation space and pay gap. (1) (2) (3) (4) (5) (6) COMP Wa GE GaP COMP Wa GE GaP Constant 5.828*** 7.213*** 4.295*** 5.842*** 7.210*** 4.313*** (40.28) (51.32) (19.65) (40.39) (51.34) (19.74) PROM –0.038*** –0.013* –0.050*** (–5.57) (–1.94) (–4.85) PROM –0.026*** –0.011** –0.034*** (–5.11) (–2.11) (–4.45) BOARD 0.007* 0.004 0.013** 0.007* 0.004 0.012** (1.90 (1.07) (2.20) (1.88) (1.07) (2.18) INDP 0.106 0.153 –0.068 0.109 0.154 –0.065 (0.74) (1.10) (–0.31) (0.76) (1.11) (–0.30) DUAL –0.015 –0.090*** 0.029 –0.015 –0.090*** 0.029 (–0.67) (–4.08) (0.83) (–0.64) (–4.08) (0.86) TOP –0.004*** 0.004*** –0.007*** –0.004*** 0.004*** –0.007*** (–9.10) (9.15) (–10.37) (–9.10) (9.17) (–10.38) ROA 3.402*** 1.562*** 4.143*** 3.404*** 1.561*** 4.144*** (23.93) (11.31) (19.29) (23.93) (11.30) (19.29) LEV –0.228*** –0.309*** –0.196*** –0.229*** –0.310*** –0.197*** (–5.16) (–7.21) (–2.94) (–5.17) (–7.22) (–2.96) SIZE 0.268*** 0.128*** 0.319*** 0.268*** 0.128*** 0.319*** (38.64) (19.04) (30.49) (38.58) (19.06) (30.45) C-SOE 0.112*** 0.044*** 0.137*** 0.113*** 0.044*** 0.137*** (7.53) (3.05) (6.07) (7.55) (3.05) (6.09) YEAR Yes Yes Yes Yes Yes Yes IND Yes Yes Yes Yes Yes Yes a djusted R 0.505 0.394 0.367 0.504 0.394 0.367 f statistic 189.38 120.99 108.18 189.09 121.02 108.01 no. of obs. 6,475 6,475 6,475 6,475 6,475 6,475 t his table reports the results of promotion expectation space on pay gap. f rom columns 1 to 3, the independent variable is PROM . f rom columns 3 to 6, the independent variable is PROM . all the variables are defined in t able 1. t -statistics used 4 5 to signal the robustness of standard errors clustered at the firm level are reported in parentheses. ***, **, and *, indicating statistical significance at 1%, 5%, and 10% levels respectively. using government level or position level, the political promotion space has a significant negative impact on the pay gap. The higher the political promotion space, the narrower will be the pay gap, indicating that when facing higher political promotion space, the executives have a stronger incentive to implement the laws to narrow the pay gap. 4.3. Robustness test To ensure a more robust conclusion, we use the following methods, namely, changing the measures of pay gap and promotion expectation. The regression results are consistent with the above findings, which indicates that the findings are robust. First and foremost, we use the executive compensation divided by employee wage as the measure of the absolute pay gap. We are aware that the regulators may allow a range for the pay gap. The relative pay gap may be reasonable for the gap. Our untabulated results show a significant and negative relation between promotion expectation and pay gap. It indicates that the conclusion is robust. Second, we use the method of Propensity Score Matching (PSM) to reduce the endoge- neity. The result shows that the pay gap in the promotion expectation firms is significantly lower than that of the non-promotion expectation firms using the measures of absolute compensation pay gap or relative compensation pay gap. CHINa JOuRNaL OF aCCOuNTING STuDIES 71 Third, to re-measure the promotion expectation, we select the samples with changes of chairman in order to conduct the regression for checking the robustness of our results, as most changes of general managers are to fill the vacant position after the chairman’s depar - ture, which are often considered as routine passive shifts (Yang et al., 2013). It is the chair- man’s voluntary departure that reflects the influence of the leadership authorities’ intention on executives. The regression result is consistent with the previous conclusion. Fourth, we further test some regressions concerning executive compensation for robust- ness analysis. Based on the above study of absolute compensation, we use the relative industrial executive compensation (executive compensation/the industrial executive com- pensation median) as a substitution for executive compensation. The result is consistent with the finding relating to the absolute pay gap. Finally, we re-measure the promotion expectation by taking the successor’s age into consideration. The results show a significant negative relationship between promotion and age. Then we take age 57 or 55 as a baseline because executives exceeding this age would often enter into a stage of waiting for retirement rather than hoping for promotion. Therefore, we define the promotion expectation of executives older than 57 or 55 as zero even if their predecessors got promoted either in a government department or firm. The result is con- sistent with the basic result. 5. Further analysis-the incentive effect of promotion expectation 5.1. Does promotion expectation mitigate agency cost? a ccording to the analysis above, the existence of promotion expectation drives executives to narrow the pay difference with employees, which, to some extent, helps the executives to build a good public reputation. But is the promotion expectation able to mitigate the agency cost? Is it possible for executives with promotion expectation to seek the rent of building a private empire and, thereby, hurt shareholders’ wealth? One explanation is that executives will enjoy the benefits of political promotion, and leave the risk and cost for shareholders. Thus, we further test whether the promotion expectation will mitigate the agency conflict between manager and shareholder. Following a ng et al. (2000), we use the expense ratio (operating expense scaled by sales) and asset utilisation ratio (the sales revenue scaled by total assets) to gauge the agency cost between shareholders and managers. We estimate the following regression equation. a ll the variables are defined in Table 1. Agencycost =  +  × PROM + Controls + (2) i,t i,t i,t Columns (1)–(4) in Table 7 show that promotion expectation measured either in PROM or PROM is not significantly associated with the management expense ratio and the asset utilisation ratio, which shows that promotion expectation not only fails to alleviate the agency cost between shareholders and management, it also does not improve agent efficiency. Columns (5)–(8) in Table 7 show that the coefficients of promotion expectation and the interaction term are positive but insignificant, suggesting that political promotion expec - tation is unable to reduce agency cost. Promotion expectation is only significantly and pos- itively related to asset utilisation ratio when the promotion expectation is political, implying that political promotion expectation could motivate executives to focus more on operating 72 D. Bu ET a L. Table 7. Promotion expectation and agency cost. (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) Expense a sset utilisa- Expense a sset utilisa- Expense a sset utilisa- Expense a sset utilisa- Expense a sset utilisa- Expense a sset utilisa- ratio tion ratio ratio tion ratio ratio tion ratio ratio tion ratio ratio tion ratio ratio tion ratio PROM 0.001 0.005 0.001 –0.018 (0.52) (0.37) (0.39) (–1.25) PROM –0.001 0.005 (–0.06) (0.78) PROM *PROM 0.001 0.110*** 1 3 (0.20) (3.88) PROM –0.001 0.150*** (–0.21) (4.95) PROM 0.001 0.017*** (0.61) (3.07) PROM 0.001 0.016*** (0.30) (3.84) Controls Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes YEAR Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes IND Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes a djusted R 0.252 0.290 0.252 0.290 0.252 0.292 0.262 0.282 0.252 0.291 0.252 0.292 f statistic 61.60 74.42 61.59 74.44 59.93 72.98 52.67 58.26 61.61 74.79 61.60 75.00 no. of obs. 6,475 6,475 6,475 6,475 6,475 6,475 5,243 5,243 6,475 6,475 6,475 6,475 t his table reports the results of promotion expectation and agency cost. f rom columns 1 to 2, the independent variable is PROM . f rom columns 3 to 4, the independent variable is PROM . f rom 1 2 columns 5 to 6, the independent variable is PROM and the interaction term of PROM and PROM . f rom columns 7 to 8, the independent variable is PROM . f rom columns 9 to 10, the independent 1 1 3 3 variable is PROM . f rom columns 11 to 12, the independent variable is PROM . t he dependent variables are ‘COST’ and ‘TURNOVER’. t he control variables are not tabulated in the table. all the varia- 4 5 bles are defined in t able 1. ***, **, and * indicate the statistical significance at 1%, 5%, and 10% levels respectively. CHINa JOuRNaL OF aCCOuNTING STuDIES 73 efficiency, reducing laziness and avoiding inappropriate investment. It further confirms that executives are concerned more about political promotion. Columns (9)–(12) in Table 7 report that both government level (PROM ) and position level (PROM ) are significantly and positively related only to the asset utilisation ratio, indicating that a higher promotion level motivates executives more strongly in improving operating efficiency and turnover ratio, when compared with a relatively lower promotion level. Overall, political promotion is ineffective in mitigating agency cost between shareholders and managers arising from the separation of ownership and operation rights. Moreover, political promotion, to some extent, motivates internal executives in SOEs to abuse accessible powers to achieve private interests at the expense of outside shareholders and the firm, hence that the promotion incentive bears implicitly negative effects. 5.2. Does promotion expectation improve performance? Theoretically, the promotion incentives weaken the compensation inclination of executives in SOEs. Given that promotion fails to mitigate agency problems, is it able to motivate exec- utives to work hard for the improvement in operation performance? We further test the incentive effect of promotion on performance. We choose the indicator of ROA , an account- ing performance measure used in assessing corporate profitability, as an alternative variable capturing the incentive effect. ROA =  +  × PROM + Controls + (3) i,t+1 1 i,t i,t i,t Table 8 reports the regression results of promotion expectation and performance. From Table 8 we can see that all other types of promotion are insignificantly associated with performance except the promotion type measured by PROM . The results, however, imply that promotion expectation is incapable of motivating executives to improve performance. Instead, the compensation behaves actively in driving executives to work hard to improve performance. But the positive inventiveness would be weakened when executives are prom- ised a promotion expectation. 5.4. Is the measurement of promotion expectation reasonable? a s above, we measure the promotion expectation using the predecessor promotion direc- tion. One problem is that if the definition is unreasonable, the conclusion is questionable. Therefore, whether the measurement of promotion expectation is reasonable need to be examined. We estimate the following logit model. If the coefficient of β is positive, it means the definition of promotion expectation is reasonable. Logit(PROM )=  +  × Exprom + Controls + (4) i,t i,t i,t i,t In Table 9, as we can see, the Ex prom is insignificantly positively associated with PROM. When we consider the types of promotion, the Ex prom being defined as political promotion is significant positively associated with the PROM. If the former executive is being promoted t he market performance indicator of t obin's Q is noisier than that in a developed market in that speculative activities are prevalent in the underdeveloped Chinese capital market. instead, the accounting performance could be more effective in reflecting executives’ efforts, so we choose the accounting performance rather than market performance as an alternative measuring the performance. 74 D. Bu ET a L. Table 8. Promotion expectation and performance. (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) ROA ROA ROA ROA ROA ROA ROA ROA ROA ROA t+1 t+1 t+1 t+1 t+1 t+1 t+1 t+1 t+1 t+1 PROM 0.004 0.081* (1.45) (1.88) PROM 0.002* 0.041** (1.76) (2.03) PROM 0.001 0.016 (0.27) (0.18) PROM 0.001 0.021 (1.02) (1.14) PROM 0.001 0.019 (1.05) (1.35) PAY 0.011*** 0.013*** 0.011*** 0.013*** 0.011*** 0.011*** 0.011*** 0.010*** 0.011*** 0.010*** (6.01) (6.26) (6.02) (6.32) (5.93) (5.88) (5.98) (4.85) (5.98) (5.18) PROM*PAY –0.006* –0.003* –0.001 –0.002 –0.002 (–1.80) (–1.92) (–0.17) (–1.08) (–1.30) Controls Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes YEAR Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes IND Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes a djusted R 0.242 0.242 0.242 0.243 0.242 0.242 0.242 0.242 0.242 0.242 f statistic 49.93 48.69 49.97 48.74 49.85 48.50 49.89 48.57 49.89 48.60 no. of obs. 5,517 5,517 5,517 5,517 5,517 5,517 5,517 5,517 5,517 5,517 t his table reports the results of promotion expectation and performance. f rom columns 1 to 2, 3 to 4, 5 to 6, 7 to 8, 9 to 10, the independent variables are PROM , PROM , PROM , PROM , PROM , 1 2 3 4 5 respectively. t he dependent variable is ‘ROA’. t he control variables are not tabulated in the t able. all the variables are defined in t able 1. t -statistics used to signal the robustness of standard errors clustered at the firm level are reported in parentheses. ***, **, and *, indicating statistical significance at 1%, 5%, and 10% levels respectively. CHINa JOuRNaL OF aCCOuNTING STuDIES 75 Table 9. association of the former executive’s promotion with that of the successor. (1) (2) (3) PROM Political PROM Intra PROM Constant –0.454 –4.716** –0.506 (–0.38) (–2.02) (–0.42) ex prom 0.062 0.641*** –0.101 (0.49) (2.98) (–0.78) PC 0.385*** 1.944*** –0.452*** (3.06) (9.06) (–3.06) EDUCATION 0.048 0.163 0.014 (0.69) (1.07) (0.19) AGE –0.117*** –0.064*** –0.114*** (–13.25) (–3.93) (–12.37) TENURE 0.080*** –0.024 0.097*** (3.99) (–0.64) (4.67) C-SOE –0.155 –0.775*** 0.016 (–1.39) (–3.20) (0.14) TOP 0.009** 0.012* 0.006 (2.35) (1.90) (1.63) LEV 0.160 –0.440 0.259 (0.46) (–0.69) (0.72) SIZE 0.204*** 0.202* 0.185*** (3.85) (1.93) (3.38) ROA –0.256 –4.750** 0.992 (–0.21) (–2.17) (0.77) YEAR Yes Yes Yes IND Yes Yes Yes Pseudo R 0.106 0.179 0.098 Wald chi 267.73 225.51 226.88 no. of obs. 2,151 2,151 2,151 t he table reports the regression result of the l ogit model. t he ‘PROM’ is a dummy variable equal to one if the executive is promoted to either a government position or a higher position within the company, and zero otherwise. t he ‘Political PROM’ is a dummy variable equal to one if the executive is promoted to a government position, and zero otherwise. t he ‘Intra PROM’ is a dummy variable equal to one if the executive is promoted to a higher position within the company and zero otherwise. t he Ex prom is a dummy variable which has the similar definition to PROM, Political PROM and Intra PROM, but is used to measure the former executive’s promotion. t -statistics used to signal the robustness of standard errors clustered at the firm level are reported in parentheses. ***, **, and *, indicating statistical significance at 1%, 5%, and 10% levels respectively. to government departments, the current executive may also have the chance to get pro- moted to the government departments. Given that our study mainly focuses on the type of political promotion, from the results of the logit model we can conclude that the definition of promotion expectation is reasonable. Moreover, we further analyse the executive status after their turnover during 2005 to 2012. In our sample, there are 129 executives being promoted to government departments and 549 executives being promoted to a higher position in the parent company. The pro- motion sample (678 cases) involves 440 listed firms. a mong them, there are 259 firms in which only one executive achieved promotion. The other firms have 181 executives being prompted, totalling 419 cases. It shows that there are two or more executives achieving promotion in the 181 firms. Overall, we can conclude that measuring the promotion status of the former senior exec- utive as the promotion expectation of current executives is reasonable. What makes this identification more convincing is that Gao, Luo, and Tang (2015) use the same method to test the influence of former executives on the executive compensation contract of the successor. 76 D. Bu ET a L. 6. Conclusion This paper tests the effects of executives’ promotion expectation on their compensation as well as the pay gap between executives and employees. We find that there is a significantly negative relationship between promotion expectation and pay gap, suggesting that senior executives who have a stronger promotion expectation are more willing to accept less com- pensation and therefore create a smaller pay gap. Furthermore, we find that different types of promotion expectation have different effects on senior executives’ perceptions. Compared with intra-group promotion, political promotion expectation has greater effects on down- sizing executive compensation and narrowing the pay gap. The higher the promotion ranks to which the former executives are promoted, the lower the compensation the current exec- utives receive and, thus, this narrows the pay gaps between executives and employees. However, we find that promotion expectation fails to mitigate agency conflict between shareholders and managers and further motivates the executives to maximise firm value. Our study provides an explanation to the ‘zero-paid’ phenomenon in China’s SOEs from the perspective of implicit incentives, and provides empirical evidence on incentive theory. In contrast to traditional compensation incentives – for example, options, salary and other explicit rewards – implicit incentives such as political promotion could also serve as a mech- anism to encourage executives to strive to maximise shareholders’ wealth. The restraining effect of political promotion on an excessive pay gap is stronger in an environment where the governance mechanism is relatively effective. However, the political promotion, while being effective in motivating executives in improving performance, is potentially harmful to the interests of firms and shareholders. Thus, in the current stage of China’s SOEs reform, the Sa Sa C should promote the establishment of a professional manager market and reduce government administrative intervention. Acknowledgements Danlu Bu acknowledges financial support from the National Natural Science Foundation of China (No. 71428008) and the Fundamental Research Funds for the Central universities (No. JBK160115). Disclosure statement No potential conflict of interest was reported by the authors. References a ng, J. S., Cole, R. a., & Lin, J. W. (2000). a gency costs and ownership structure. Journal of Finance, 55(1), 81–105. Baker, G. P., Jensen, M. C., & Murphy, K. J. (1998). Compensation and incentives: Practice vs. theory. Journal of Finance, 43, 593–616. Baker, G., Robert, G., & Kevin, J. M. (1994). Subjective performance measures in optimal incentive contracts. The Quarterly Journal of Economics, 109, 1125–1156. Bebchuk, L., Fried, J., & Walker, D. (2002). Managerial power and rent extraction in the design of executive compensation. The University of Chicago Law Review, 69, 751–846. Blanchard, O., & Shleifer, a. (2001). Federalism with and without political centralization: China vs. Russia. IMF Staff Papers, 48, 171–179. Campbell, D. (2008). Nonfinancial performance measures and promotion-based incentives. Journal of Accounting Research, 46, 297–332. CHINa JOuRNaL OF aCCOuNTING STuDIES 77 Cao, J., Michael, L., Pan, X. F., Qian, M. J., & Tian, G. (2011). Political promotion, CEO incentives and the relationship between pay and performance. Working paper. Chen, D. H., Chen, X. Y., & Wan, H. L. (2005). Regulation and non-pecuniary compensation in Chinese SOEs. Economic Research Journal, 40, 92–101. (In Chinese) Chen, X. Y., Chen, D. H., Wan, H. L., & Liang, S. K. (2009). The disparity in regions, the regulation of emolument, and the corruption of high-ranking managers. Management World, 11, 130–143. (In Chinese) Dai, Y. Y., Hong, Q., & Pan, Y. (2015). Official inspection, media coverage and government subsidies: Empirical evidence from Chinese listed companies. Economic Management Journal in China, 535, 13–25. (In Chinese) Eriksson, T. (1999). Executive compensation and tournament theory: empirical tests on Danish data. Journal of Labor Economics, 17, 262–280. Fan, J. P. H., Wong, T. J., & Zhang, T. Y. (2007). Politically connected CEOs, corporate governance, and Post-IPO performance of China's newly partially privatized firms. Journal of Financial Economics, 84, 330–357. Fang, J. X. (2009). Is top management compensation of Chinese public companies sticky? Economic Research Journal, 44, 110–124. (In Chinese) Fang, J. X. (2011). Managerial power and asymmetry of compensation change in China's public companies. Economic Research Journal, 46, 107–120. (In Chinese) Firth, M., Fung, P. Y., & Rui, O. M. (2006). Corporate performance and CEO compensation in China. Journal of Corporate Finance, 12, 693–714. Gao, H. S., Luo, J., & Tang, T. (2015). Effects of managerial labor market on executive compensation: evidence from job-hopping. Journal of Accounting and Economics, 59, 203–220. Gibbons, R., & Murphy, K. J. (1992). Optimal incentive contracts in the presence of career concerns: theory and evidence. Journal of Political Economy, 100, 468–505. Gibbs, M., Merchant, K. a., Van der Stede, W. a., & Vargus, M. E. (2004). Determinants and effects of subjectivity in incentives. The Accounting Review, 79, 409–436. Groves, T., Hong, Y. M., John, M. M., & Barry, N. (1995). China’s evolving managerial labor market. Journal of Political Economy, 103, 573–592. Huang, F. G., Li, G., & Li, X. W. (2011). Executive compensation, government-linked administrative rank and agency costs. Science of Science and Management of S. & T., 33, 171–179. (In Chinese) 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. Jenter, D., & Kanaan, F. (2015). CEO turnover and relative performance evaluation. Journal of Finance, 70, 2155–2184. Jia, R. X., & Nie, H. H. (2015). Decentralization, collusion and coalmine deaths. Review of Economics and Statistics, forthcoming. Leone, a. J., Wu, S. J., & Zimmerman, J. L. (2006). a symmetric sensitivity of CEO cash compensation to stock returns. Journal of Accounting and Economics, 42(1–2), 346–385. Li, H. B., & Zhou, L. a. (2005). Political turnover and economic performance: the incentive role of personnel control in China. Journal of Public Economics, 89, 1743–1762. Li, W. a., Qiu, a. C., & Gu, Z. H. (2010). Dual corporate governance environment, political connections preference and firm performance: Study on governance transition of China’s private listed firms. China Industrial Economics, 6, 8595. (In Chinese) Li, W. a., Qiu, a. C., Niu, J. B., & Xu, Y. K. (2010). Recent development of corporate governance: International trends and China’s mode. Nankai Business Review, 13, 13–24. (In Chinese) Li, W. J., & Hu, Y. M. (2012). Who is encouraged by pay dispersion in State-owned Enterprises? Economic Research Journal, 47(12), 125–136. (In Chinese) Li, Z. Q. (2000). Incentive mechanism and company performance: Empirical research based on China’s listed companies. Accounting Research, 1, 24–30. (In Chinese) Liang, S. K., Li, Z., Chen, D. H., & Chen, S. M. (2015). Political ranks, incentives and firm performance. China Journal of Accounting Studies, 3, 87–108. Lin, J. Y., & Liu, Z. (2000). Fiscal decentralization and economic growth in China. Economic Development and Cultural Change, 49(1), 1–21. 78 D. Bu ET a L. Lin, Y. F., Liu, M. X., & Zhang, Q. (2004). Political burden and company budget soft constraint: Empirical research from China. Management World, 8, 81–89. (In Chinese) Liu, C., & Sun, L. (2010). Pay gap and company performance: Empirical evidence from China’s listed SOEs. Nankai Business Review, 13, 30–39. (In Chinese) Liu, X. X. (2001). The development report of China SOEs: 1990-2000. Beijing: Social Sciences a cademic Press. (In Chinese) Lu, Z. F., Wang, X. Y., & Zhang, P. (2012). Do Chinese state-owned enterprises pay higher wage? Economic Research Journal, 47, 28–39. (In Chinese) Maslow, a. (1954). Maslow’s hierarchy of needs: Motivation and personality. New York, NY: Harper. Quan, X. F., Wu, S. N., & Wen, F. (2010). Managerial power, private income and compensation rigging. Economic Research Journal, 45(11), 73–87. (In Chinese) Siegel, P. a., & Hambrick, D. C. (2005). Pay dispersion within top management groups: Harmful effects on performance of high- technology firms. Organization Science, 16, 259–274. The Research Group of unirule Institute of Economics. (2011). Nature of state-owned enterprises, performance and reform. Working paper. (In Chinese) The World Bank. (2012). World Development Indicators 2012. Washington, DC: The World Bank. Wang, Z., Fu, G. Q., Huang, D. Y., & Wang, J. F. (2014). a study on the relationship between the ‘political rise’ of the CEO in State-owned firms and the consumption when they are at their post. Management World, 5, 157–171. (In Chinese) Xin, Q. Q., & Tan, W. Q. (2009). Market-oriented reform, firm performance and executive compensation in Chinese State-owned Enterprises. Economic Research Journal, 44, 68–81. (In Chinese) Yang, R. L., Wang, Y., & Nie, H. H. (2013). Promotion mechanism of quasi official: Evidence from China central enterprises. Management World, 3, 23–33. (In Chinese) Zhang, L. L., Liu, F., & Cai, G. L. (2015). Regulatory independence, market-oriented process and promotion mechanisms implementation of SOE’s executive. Management World, 10, 117–131. (In Chinese) Zhang, W. Y. (1999). The enterprise theory and Chinese SOE sector reform. Beijing: Peking u niversity Press. (In Chinese) Zhang, W. Y. (2014). SOE is impossible to establish effective corporate governance. Cai Zhi Dao, (198). Zheng, Z. G., Li, D. X., Xu, R., Lin, R. T., & Zhao, X. J. (2012). The political promotion and image establishment of SOE Executives: a case study based on province N, company a. Management World, 10, 146–156. (In Chinese) Zhou, L. a. (2002). Career Concerns, Incentive Contracts, and Contract Renegotiation in the Chinese Political Economy. PhD thesis, Stanford university. Zhou, L. a. (2004). Government officials’ incentive and cooperation in the promotion game: The problem of local protectionism in China and repeated construction. Economic Research Journal, 39(6), 33–40. (In Chinese) Zhou, L. a., Li, H. B., & Chen, Y. (2005). Relative performance evaluation: a n empirical study on the Chinese local officials’ promotion. China Journal of Economics, 1(1), 83–96. (In Chinese) Zhou, L. a. (2007). Study on the promotion tournament mode of China local officials. Economic Research Journal, 42(7), 36–50. (In Chinese) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png China Journal of Accounting Studies Taylor & Francis

Will political promotion expectation decrease the pay gap in state-owned enterprises in China?

Loading next page...
 
/lp/taylor-francis/will-political-promotion-expectation-decrease-the-pay-gap-in-state-sHMC0E8UgG

References (52)

Publisher
Taylor & Francis
Copyright
© 2016 Accounting Society of China
ISSN
2169-7221
eISSN
2169-7213
DOI
10.1080/21697213.2016.1144969
Publisher site
See Article on Publisher Site

Abstract

China Journal of aCC ounting StudieS , 2016 Vol . 4, no . 1, 53–78 http://dx.doi.org/10.1080/21697213.2016.1144969 Will political promotion expectation decrease the pay gap in state-owned enterprises in China?* a b c Danlu Bu , Chenyu Zhang and Teng Lin a b Southwestern university of f inance and economics, Chengdu, China; university of international Business and economics, Beijing, China; Shantou university, China KEYWORDS ABSTRACT a gency cost; pay gap; In this paper, we test the effects of executives’ promotion expectation political promotion; on compensation and the pay gap between executives and employees promotion expectation; in China. We measure the promotion expectation of current executives promotion rank by using their predecessors’ promotion ranks and direction. We find that there is a significantly negative relationship between promotion expectation and the pay gap, suggesting that senior executives with stronger expectation of political promotion are more willing to accept less compensation and therefore a smaller pay gap. Furthermore, we find that different types of promotion expectation have different effects on senior executives’ perceptions. Compared with intra- group promotion, political promotion expectation has greater effects on limiting executive compensation and narrowing the pay gap. However, we find that promotion expectation fails to mitigate agency conflicts between shareholders and managers and further motivates the executives to maximise firm value. 1. Introduction Company executives play an important role in wealth creation for shareholders. However, due to the agency problem, they do not necessarily align themselves fully with the share- holders. Thus, it is important to use various incentives to motivate managers in order to mitigate agency conflicts (Jensen & Meckling, 1976). There are several mechanisms that firms can apply to motivate executives. In addition to monetary compensation such as salaries, bonuses, stocks, options and perks, there is non-monetary compensation such as political promotion (Zhang, 1999). Within the special institutional background of China, the managers of state-owned enterprises (SOEs) in China all have political ranks. They have the same status as the government officials who have the corresponding political rank (Yang, Wang, & Nie,2013). The appointment and dismissal of SOE managers are the responsibility of the department of local government organisations. The nature of political promotion for the SOE manager gives us an opportunity to research the relationship between political pro- motion and compensation decision. CONTACT danlu Bu budanlu@swufe.edu.cn *Paper accept by heng Yue. © 2016 a ccounting Society of China 54 D. Bu ET a L. Since using monetary compensation may directly reduce the shareholders’ claim on the company’s wealth, agency problems still exist where managers cannot fully align their inter- ests with shareholders’ interests (Jensen & Meckling, 1976). In contrast to promotion accom- panied by monetary compensation, political promotion in China may not directly affect shareholders’ own wealth, but it could improve managers’ career development with the benefits of enjoying greater future economic and political power. Political promotion is often not explicitly based on official contracts. It not only links to performance (Cao, Michael, Pan, Qian, & Tian, 2011; Yang et al., 2013) but also has a relationship with some subjective and non-quantitative indicators, such as social welfare activities, employment opportunities, ecological construction, safe production and philanthropy (Campbell, 2008; Gibbs, Merchant, Van der Stede, & Vargus, 2004; Jia & Nie, 2015; Zheng, Li, Xu, Lin, & Zhao, 2012). Political promotion will not dilute shareholders’ wealth but it can bring benefits to managers. In China, political promotion does not lead directly to increased compensation. u sually, gov- ernors’ compensation is strictly regulated by government, therefore governors receive less remuneration than enterprises’ executives. We are not aware of any empirical evidence on the issue of whether the expectation of political promotion can influence executives’ demands for explicit compensation. Thus, this paper attempts to investigate the impact of political promotion expectation on executive compensation. a s disclosure of executives’ compensation has been mandated at an individual level since 2005 by the China Securities Regulatory Commission, the executives’ high compensation amount and the subsequent pay gap between them and the employees have attracted enormous public attention in China. In contrast to executives’ higher compensation, 448 chairmen in China received zero compensation in 2008, which accounts for approximately 37% of Chinese listed companies. From our analysis of the CSMaR database, during 2006– 2011, only about 21% of Chinese listed companies narrowed their pay gap in any year com- pared with the previous year. We also observed that in some firms the pay gap became smaller despite increasing firm performance. Theoretically, managers, including executives, have the power to influence compensation contracts (Bebchuk, Fried, & Walker, 2002; Quan, Wu, & Wen, 2011), which may increase the pay gap between executives and employees (Li & Hu, 2012). We investigate the motivation that drives managers to choose ‘Zero-payment’ and explain the phenomenon of the lower pay gap. We explain later in the paper that this is a specific feature of political promotion in China, whereby managers may choose to show their allegiance to political aims by accepting a lower level of remuneration in order to maintain a lower pay gap. We also ask whether there are other ways for an executive to pursue private benefits to complement insufficient compensation. Generally speaking, executives in China SOEs are appointed by the government and have certain official ranks, resembling those of government officials. Therefore, the SOE executives are entitled to similar political power, most of them expect to be government officials to gain governors’ power. That is because, according to the governors’ promotion route in China, government officials have greater scope for promotion and a longer political career (Yang et al., 2013). In 2006, the Central Committee of the Communist Party of China issued ‘the provisions on the official exchange between the enterprises and governments’, which clearly state the requirement of selecting outstanding SOE managers and promoting them into the government and the administrative system. Thus, the official exchange between SOEs and according to Sina f inance report, in 2008, 448 China listed company managers receive ‘zero compensation’, see: http://finance.sina.com.cn/roll/20090425/05032807627.shtml CHINa JOuRNaL OF aCCOuNTING STuDIES 55 government is widespread, and is mainly a unidirectional flow of executives in enterprises to be officials in the government (The Research Group of unirule Institute of Economics, 2011; Yang et al., 2013; Zheng et al., 2012). Therefore, the SOE executives have a strong incentive to advance their political careers (Yang et al., 2013; Zhang, 1999). Due to the political advantages, SOE executives pay more attention to political promotion. If executives’ political performance is outstanding, they can gain political promotion and hence enjoy more ben- efits (such as non-pecuniary compensation, Non-statutory income and political power) (Chen, Chen, Wan, & Liang, 2009). Meanwhile, the guiding principle of promoting equality and efficiency has shifted from efficiency to social equality since 2004. Thus, how to reverse the pay gap has become a primary concern of Chinese society. Therefore, the government initiated reforms in income allocation and issued regulations limiting pay to eliminate the pay gap. under the influence of ‘Compensation Cap’, senior executives accomplish their duty of closing the pay gap and send a positive signal to the public, which maintains a good image. Thus, this creates a research opportunity for investigating whether a promotion expectation would inspire executives, with the right to determine their own compensation, to voluntarily close the pay gap. Prior studies have focused on the incentive effect and economic consequences of the political promotion of government officials or SOE executives. For government officials, political promotion is more important than local fiscal revenue (Blanchard & Shleifer, 2001; Zhou, 2002). Moreover, this incentive brings growth of gross domestic product (GDP) to the province (Li & Zhou, 2005; Zhou, Li & Chen, 2005), and it will also lead to local protectionism and haphazard investment (Zhou, 2004). For SOE executives, there is a positive relationship between the incentive for political promotion and performance (Cao et al., 2011; Groves, Hong, John, & Barry, 1995; Jenter & Kanaan, 2015), but there are also aspects with a negative effect such as image project building and risk-taking behaviour (Zheng et al., 2012). However, the literature relating to executive political promotion generally tests the effect of political promotion on officials’ behaviour (Li & Zhou, 2005; Wang, Fu, Huang, & Wang, 2014; Yang et al., 2013). Such a method of testing may experience a problem of endogeneity. Commonly, if the former executive is being promoted to government departments, the current executive may also have the chance to be promoted to the government depart- ments. For example, Zongting Liu, the CEO of Fujian Mingdong Electric Power Limited Company (Stock: 000993), was promoted to the Deputy Secretary of Ningde Municipal Government in the year 2007. Bangheng He, the successor CEO after Liu, was promoted to the Forestry Bureau of Ningde in 2011. Lianxuan Zhang, the chairman of Tianjin Reality Development (Group) Co., Ltd (Stock: 600322), was promoted to the Deputy Director of Tianjin Municipal Construction and Management Commission in 2006. Yong Zhang, the t he unirule i nstitute of economics (unirule) is an independent, non-profit, non-governmental (ngo ) think tank, which was jointly initiated in July 1993 by five prominent economists, Professor Yushi Mao, Professor Shuguang Zhang, Professor hong Sheng, Professor g ang f an, and Professor Shouning t ang. unirule is dedicated to the open exchange of ideas in economics in general, with a particular focus on institutional economics, and maintains a highly prestigious status within academic circles. http://english.unirule.org.cn/ t he central government of China has issued some legislation on the compensation regulations, such as, ‘notice of the Ministry of f inance on the issues relevant to the management compensation in the State-o wned and State-Controlled financial enterprises’ (2009), ‘regulations on the executive compensation distribution in the State- o wned financial institutions’ (2009) ‘guiding opinions on further regulating executive compensation in the central-government controlled enterprises’(2009) and ‘notice of g eneral o ffice of the State Council on work division concerning further deepening reform of the income dis- tribution system’ (2013). t he results are given in t able 9. 56 D. Bu ET a L. successor chairman after Zhang, was promoted to the Deputy Secretary and mayor of Wuqing District, Tianjin in 2010. These examples are indications that both former and current executives who achieved promotion are present in our sample. Based on the above analysis, this paper measures the political promotion expectation of current executives by the political promotion status of former senior executives. If the former executives gain promotion to government or achieve a higher position in the group, we regard the current executives as having strong expectations of promotion. We use data from 2005 to 2012 in Chinese SOEs to test the effects of the executive’s promotion expectation on senior executives’ compensation as well as the pay gap between executives and employ- ees. We find that there is a significant negative relationship between promotion expectation and the pay gap. The senior executives who have stronger promotion expectations are more willing to accept less compensation and a lower pay gap. We further find that different types of promotion expectations have different effects on the behaviour of senior executives. Compared with intra-group promotion, political promotion expectation makes the pay gap smaller because the compensation of current executives is lower. The higher the official rank to which the former executives are promoted, the lower the compensation and the smaller the pay gap the current executives can accept. However, we find that the promotion expec - tation fails to mitigate the agency conflicts between the shareholders and managers and fails to motivate the executives to maximise firm value. Our study tries to provide an expla- nation for the ‘zero-payment’ phenomenon in China SOEs from the perspective of implicit incentives, and provides empirical evidence supporting incentive theory. Our results indicate that the executives tend to achieve promotion by using corporate resources for non-pro- ductive rent-seeking and they establish a personal social network to achieve their own personal interest, which intensifies the agency problem and weakens the profitability of the firm. Our paper contributes to prior studies in three aspects. First, we provide an explanation of the common ‘zero-payment’ phenomenon in China SOEs from the perspective of implicit incentives, and provide evidence for SOE regulators to set up management incentive con- tracts. a lthough there is prior literature testing compensation contracts based on optimal contract theory or management power theory (Bebchuk et al., 2002; Fang, 2009; Quan et al., 2011), few prior studies have investigated the ‘zero-payment’ phenomenon in China. Based on promotion expectation and implicit incentives, our paper provides new evidence and explanation for compensation contract theory. Second, compared with previous studies on the positive effect of political promotion (Cao et al., 2011; Wang et al., 2014; Yang et al., 2013), our paper provides evidence of negative effects of political promotion. This paper extends the framework of political promotion and provides evidence for regulators on how to design the executives’ incentive mechanism in SOEs. Third, our study systematically anal- yses types of executives’ promotion, distinguishing political promotion from intra-group promotion and classifying the promotion level of the executive. It helps us to understand the promotion route for China officials. Meanwhile, our setting avoids the endogeneity problem to some extent where promotion expectation is measured by the former execu- tives’ promotion. The rest of the paper is organised as follows. Section 2 discusses the insti- tutional environment that provides the setting for our analysis and the development of our hypotheses. Section 3 introduces our data sources and models. Section 4 presents our main empirical results and discussion. We run further analysis in Section 5, and conclude in Section 6. CHINa JOuRNaL OF aCCOuNTING STuDIES 57 2. Institutional background and hypothesis development 2.1. Institutional background The media coverage around executive compensation has attracted intensive public attention to the disclosure of the pay of individual executives in China since 2005. In addition, relatively transparent executive remuneration information reveals severe pay gap issues in a sensitive way. Examples like several millions executive compensation in financial institutions and monopoly industries have triggered heated public debates on executives’ pay. a ccording to the National Bureau of Statistics of PRC, China’s Gini coefficient reached 0.473 in 2013, which is larger than the internationally prescribed generally reasonable level and much larger than that of other countries at a similar level of development ( The World Bank, 2012). Thus, how to reverse the pay gap has become a primary concern of Chinese society. In 2004, the Fourth Plenary Session of the 16th Central Committee of the Communist Party of China first proposed the aim of building a ‘harmonious society’ and the determination to ‘take effective measures to solve the problem of an excessive pay gap’. In 2012, the Third Session of the 11th National People's Congress proposed ‘firmly reversing the widening trend of the income gap’ and the Fifth Session of the 11th National People’s Congress proposed ‘reversing the trend of the widening pay gap as soon as possible’. Meanwhile, in order to narrow the serious pay gap, the government began to promote reform targeting the rationalisation of income distribution with measures such as issuing a series of regulations restricting executive compensation and narrowing the pay gap. In 2009, the Ministry of Finance issued ‘Notice of the Ministry of Finance on the issues relevant to management compensation in the State-Owned and State-Controlled financial enterprises’, ‘Regulations on the executive compensation distribution in the State-Owned financial insti - tutions’, which clearly required that the executive compensation of the financial enterprises should be no more than CNY2.8 million, and that the executive compensation in 2008 should be no more than 90% of that of 2007. If the performance declines compared with 2007, the executive compensation should be further reduced by another 10% to 80% of that in 2007. In 2009, the Ministry of Human Resources and Social Security, the Ministry of Finance and another six ministries jointly issued ‘Guiding opinions on further regulating executive com- pensation in the central-government controlled enterprises’, which expresses a clear require- ment that SOE executives’ base salary should link to the employees’ average wages of the previous year. In 2013, ‘Notice of General Office of the State Council on work division con- cerning further deepening reform of the income distribution system’ proposed ‘to strengthen the management of executive compensation in SOEs. Moreover, executive compensation increases should be no higher than that of employee average increase’. Issued in a ugust 2014, under the leadership of Human Resources and Social Security of PRC and with partic- ipation by the Ministry of Finance as well as other ministries, another compensation regu- lation requests that compensation of executives in centrally controlled enterprises and Stated-owned finance firms should be decreased to about 30% of the present level and cannot exceed CNY 600,000. Moreover, the base salary of CEOs in central controlled http://news.xinhuanet.com/2014-01/20/c_119040570.htm t he gini coefficient was put forward by the i talian economist gini in 1922 and is an index that quantifies inequality among values of an income distribution. t he value of the gini coefficient is a number from 0 to 1. a value of zero means that all have the same income. a ccording to the international standard, if the coefficient is lower than 0.2, the disparity in income is low. if the coefficient is more than 0.5 then the disparities of income are relatively high. 58 D. Bu ET a L. enterprises should be twice the average annual pay of employees, the performance pay should be twice the base salary, and tenure incentives should be no more than 30% of the overall pay. a ll these restrictions signal that the government has been striving to narrow the increasingly widening pay gap. 2.2. Hypothesis development a compensation contract seeks to provide an optimal institutional arrangement to mitigate the conflicts between shareholders and agents arising from the separation of ownership and operation rights (Jensen & Meckling, 1976). With the gradual development and estab- lishment of China’s market economy system, aligned with further reform undertaken in SOEs, the government has issued a large number of laws and regulations on the enterprise com- pensation system, and a performance pay scheme also begins to take hold in China’s listed companies (Fang, 2009). However, SOEs still retain multiple non-profit goals in the process of market-oriented reformation. Specifically, the fiscal decentralisation reform left local gov - ernments with the right to gain more profit from enterprises located within the jurisdiction area, which would drive the local governments to retain influence over the enterprises within their jurisdiction area (Chen et al., 2009). a lthough a market economy mechanism and mod- ern corporate governance are gradually being established, the impact of administrative power on the resource redistribution still exists (Li, Qiu, & Gu, 2010). Due to the weak investor protection system and low efficiency of law enforcement, China is regarded as having an inefficient system of corporate governance (Firth, Fung, & Rui, 2006). The effect of dual governance environments (with both official governance and economic governance) (Li, Qiu, Niu, & Xu, 2010) on the incentive mechanism is seen in two ways. One is that although the official rank of executives is gradually being phased out, their appoint - ments, instead of being selected by the market, are still controlled by the government (Liu, 2001). The cross-border exchanges of SOE executives and government officials are wide - spread and many SOE executives are now, or have once been, government officials (Fan, Wong, & Zhang, 2007; Liang, Li, Chen, & Chen, 2015; Zhang, 1999). Hence, many of them would reasonably embrace expectations to be appointed as civil servants after accumulating experience in enterprises. Therefore, China’s SOE executives, who exhibit the characteristics of ‘half Officials half Entrepreneurs’, are more inclined to develop their political career rather than becoming professional managers (Xin & Tan, 2009). Meanwhile, the present overwhelm- ing political objectives of maintaining social fairness and stability, reducing local fiscal deficits but increasing fiscal surpluses, has led the compensation of SOE managers to be under strict government control (Chen, Chen, & Wan, 2005). a more diversified non-monetary compen- sation system, such as the economic incentive of perks and the administrative incentive of political promotion, may appear as a substitution for a direct monetary reward system when monetary pay-offs are restricted (Chen et al., 2009). Hence, managers’ motivation to pursue non-monetary interests such as political promotion would be strengthened when facing this unique institutional environment. Moreover, due to the impact of ‘official-oriented‘ ide - ology, political promotion is very attractive to executives, because it is accompanied by greater political power, more prestigious position, more perks and even rent-extraction (Chen et al., 2005), and governments (central and local) also tend to use political promotion as the primary way of motivating SOE executives (Zhang, 1999). Hence, political promotion CHINa JOuRNaL OF aCCOuNTING STuDIES 59 has become an important concern for SOE executives during the development of their professional careers (Yang et al., 2013). Official intervention and pressure from political goals have driven SOEs to target aims more than maximising profit (Lin, Liu, & Zhang, 2004). Consequently, there may be ambiguity in establishing a causal relationship between company performance and managers’ effort. The existence of multiple aims also weakens the effectiveness of a compensation contract based on performance. In this particular economic transition period, Chinese SOEs undertook multiple objectives, such as to obey the development strategies of the national economy, macroeconomic regulation, employment, and social stability as a result of the prevalent political intervention (Lin et al., 2004). under the institutional environment, where the pro - fessional destinations of executives in SOEs are mainly determined by the government, managers in SOEs would have to comply with the orders of government officials, who are influential and decisive in the selection, promotion and compensation setting of SOE man- agers, concerning the operation activities of the SOEs under their charge. Therefore, local government has the motivation and capacity to intervene in SOE production and business activities (Li & Zhou, 2005; Lin & Liu, 2000). The government intervention will often distort managers’ behaviour. What the managers care about is not company performance, but their own political careers. Theoretically, government officials’ assessment towards of SOE man- agers should focus on both economic and political performance (Zhang, 1999). However, sometimes, managers’ political performance will outweigh economic performance, that is, managers’ careers will continue to move upward rather than downward as long as they are politically doing well despite the general economic performance (Zhang, Liu, & Cai, 2015). This is because the multiple objectives of an SOE decrease the relevance of the manager’s effort and operating performance (Lin et al., 2004; Zhang, 1999). Thus, in order to ease the political burden of government at all levels, or to achieve the political promotion goal of government officials, or both, the government officials are more interested in the SOE man- ager’s political performance than economic performance (Zhang et al., 2015). In the case that government officials’ interests are more directly related to the political performance of SOE managers, the economic tournament incentive on executives will weaken sponta- neously (Zhang, 1999, 2014). under such an institutional background, government officials’ preference towards political gains will certainly lead enterprise managers to pay more atten- tion to political behaviour in that only SOE executives with a good political performance could be eligible for promotion in association with higher political rewards, more control power benets fi (such as in-oc ffi e consumption, hidden income), more political benefit (such as political promotion) and other forms of benefit (Chen et al., 2009). Since 2004, the guiding ideology around fairness and efficiency has changed from ‘giving priority to efficiency with consideration to fairness’ to ‘social fairness and efficiency are equally important’. With the purpose of narrowing the pay gap, the government began to reform income allocation by issuing a series of policies and regulations restricting excessive exec- utive compensation. under the background of a ‘compensation restriction order’, catering to the political goal of narrowing the pay gap can help corporations as well as executives to establish a good public image, to signal good news to outsiders vulnerable to information and to gain the public, media and government authority’s attention, which would in turn help executives’ personal interests, especially political promotions (Wang et al., 2014). Therefore, executives with promotion expectation are more willing to implement govern- ment policies and regulations. By doing so they can lay a solid foundation for future political 60 D. Bu ET a L. promotion and career development. Executives with an expectation of promotion are more inclined to narrow the pay gap in order to gain promotion advantages. Thus, based on the foregoing analysis, we propose our first hypothesis. H1: There is a negative correlation between promotion expectation and pay gap. a t present, the central government is vigorously promoting the reformation of separating enterprises from the government administration and conducting de-administration refor- mation in SOEs because executives in SOEs having government administrative ranks are still governed by the respective party committees. Specifically, most centrally controlled enter - prises are directly restructured from subordinate enterprises of the State Council’s professional ministries, and some of their top leaders are naturally shifted directly from the professional ministry leaders. a lthough central enterprise executives are not formal government officials/ civil servants they are managed and paid according to the corresponding governor scale ( The Research Group of unirule Institute of Economics, 2011; Yang et al., 2013). Obviously, all SOE executives are ‘quasi-officials’ being branded as administrative officials (Yang et al., 2013). Subsequently, these ‘quasi-officials’, similar to their government peers at the same level, receive similar political treatment and opportunities for political promotion (Yang et al., 2013). Most of the central enterprise executives personally expect to be promoted to the Party and to be government officials in power. Even if they achieve the status of ‘person in charge’ in the parent firm, their administrative level will be vice-ministerial level at most and there will be no further promotion thereafter (Yang et al., 2013; Zhang et al., 2015). Only by transferring from a ‘quasi-official’ to a ‘real official’ would it be possible to obtain a higher political promo - tion at administrative level by which to continue political life, such as being promoted to be the ‘person in charge’ in central ministries or local provincial government (ministerial level) after which an official would qualify to be a candidate for central leadership (Yang et al., 2013). In contrast to General Party and government cadres, the SOE executives equipped with a technical background and industrial experience are highly proficient in the market oper - ation and specialised in promoting the development of unique local industries. Therefore, the General Office of the Communist Party of China issued ‘Provisions on the exchange among the party and government cadres’ in 2006, clearly stating the requirement for select- ing outstanding SOE leaders to be appointed in the Party and government departments while recommending the Party and government cadres work in SOEs and similar institutions. This, however, provides ways for SOE executives to be promoted to the Party and government departments. For SOE executives there are two different types of promotion. One is to be promoted to an office in government. The other is to be promoted to a higher position in the firm. The differences are that being an official in government departments should be more beneficial than being an executive (Chen et al., 2009). On the one hand, it provides an opportunity for gaining higher political ranks as well as the continuation of political life. On the other hand, a government official will have easy access to scarce resources such as per - sonnel assignment rights and reallocation of other important materials (Yang et al., 2013). In contrast, intra-group promotion is a business position promotion. Even where an executive t he levels of civil servants regulated in China are nine in total, namely clerk, vice section, the section, vice department, the department, vice provincial department, the provincial department, vice-ministerial and ministerial in accordance with the civil servant rules. t he 53 top Central Soes in a list of central enterprise rankings issued by the SaSa C are ‘vice-ministerial central enterprises’. heads (chairman, general manager or secretary of the party committee) of these top 53 enterprises are appointed and managed jointly by the central organisation department and Bureau i Management managing corporation leaders subordinate to SaSa C in accordance with vice-ministerial official management codes. CHINa JOuRNaL OF aCCOuNTING STuDIES 61 is promoted to be the head of a central/parent enterprise, the corresponding administrative rank is vice-ministerial level. In many cases the executive is considered to have reached the ceiling of a business career. In that situation, their control of scarce resources is weaker rel- ative to that if politically promoted (Wang et al., 2014). The current executive appointment system in China SOEs implies a tournament among SOE executives who are motivated to strive for winning such a tournament (Li & Zhou, 2005; Zhou, 2004, 2007). The adoption of such a mode of appointment induces SOE executives to concentrate more on political promotion than economic ones, and in turn to obtain a higher likelihood of political promotion. Executives will often try their best to cater to political objectives, aiming at being more eligible for political promotion (Zhang et al., 2015). Such eo ff rts will, however, help to either relieve the political burden of local government, or make the political promotion goals of government officials come true, or both. This consideration, together with the background of a ‘compensation restriction order’, means it is reasonable for SOE executives to voluntarily reduce monetary compensation and eliminate pay differ - ences with employees. In doing so they may also build a favourable reputation that is helpful in their political promotion. Thus, compared with the expectation of intra-group promotion, the executive shows a stronger motivation to enter politics and is more likely to accept a lower monetary compensation and a smaller pay gap when facing the opportunity of polit- ical promotion. Thus, based on the analysis, we propose the second hypothesis. H2: Compared with internal/intra-group promotion expectation, the political promotion expec- tation has a larger negative effect on pay gap and makes the pay gap smaller. Gibbons and Murphy (1992) found that executives not only pay attention to the short-term benefits, but are also concerned about the performance that brings them competitive strength in the labour market. The career concerns can have important effects on incentives. The optimal compensation contracts neutralise career concern incentives by optimising the total incentives from the contract and from career concerns: explicit contractual incentives are high when implicit career concern incentives are low, and vice versa. The idea that optimal incentive contracts optimise total incentives can be applied to promotions (Baker, Jensen, & Murphy, 1998; Gibbons & Murphy, 1992). For Chinese SOEs, the higher the political rank of the company, the more positive the effect of career attraction for executives. So there is a substitute relationship for political ranks and compensation (Huang, Li, & Li, 2011). In China, there are two main factors determining the executive’s political promotion, one is the admin- istrative rank of the present ultimate controller of the SOE where the executive works, in which case the executive’s rank after promotion should be no higher than his peers whose ultimate controllers’ rank is in the same, and the other is the corresponding administrative rank of the executive’s specific position. The administrative level of the SOE’s ultimate controller is influ- ential over the executive’s political promotion space (Yang et al., 2013; Zhang, 1999). Since the administrative rank of the ultimate controller in the central enterprise is higher than that of the local enterprise, the executives in the central enterprise are endowed with easier access to government officials, therefore executives in the central SOEs focus more on political development and promotion (The Research Group of unirule Institute of Economics, 2011). a larger promotion space drives managers in centrally controlled enter - prises to obey the orders of the government at a higher level and restrain their own oppor- tunistic behaviour, compared with the managers in local government-controlled companies where executives are confined to a limited political career and a relatively smaller promotion 62 D. Bu ET a L. space (Yang et al., 2013). For a long time, state-controlled listed companies’ executives regarded the ‘Official Standard’ as the goal and the SOEs have implemented their compen- sation systems in such a way that the compensation has been significantly sensitive to the political rank, and rewarded executives differently according to their corresponding admin - istrative ranks (Zhang, 1999). In addition, executives at different administrative levels also receive different treatment in other aspects, such as in health care and welfare. It is obvious that a high administrative level itself and the opportunity of increasing administrative levels are incentives. Generally, the setting of executive compensation is based on the administra- tive level and average wages of workers in the same industry (Yang et al., 2013; Zhang, 2014; Zheng et al., 2012). It can be seen that there exists a different effect on the firm value of state-controlled companies, even of the same nature, as a result of their ultimate controllers’ executive ranks, which also exert different influences on executive compensation as well as agency cost. a ccording to the incentive theory (Maslow, 1954; Baker, Robert, & Kevin, 1994), the different political ranks will bring different benefits and generate different promotion motivations for an executive, and then the political ranks have different substitutive rela- tionships with compensation. However, the administrative ranks of executives in SOEs are not always the same. Take central SOEs for example. The top 53 central SOEs in a list of central enterprise rankings issued by the State-owned a ssets Supervision and a dministration Commission of the State Council (SaSa C) are ‘vice-ministerial central enterprises’, also known as the backbone of the central enterprises (Yang et al., 2013). Heads of these top 53 enterprises (chairman, general manager or secretary of the party committee) are appointed and managed jointly by the central organisation department and Bureau I Management of SaSa C in accordance with vice-ministerial official management codes. Separate from the top 53 firms, the other central enterprises in the list are the ‘provincial level central enterprises’, whose responsible persons are managed by the Bureau II Management of SaSa C according to the management of ‘provincial level’ government officials, which still needs formal approval from the central organisation department. SaSa C manages central enterprise leaders with reference to gov- ernment officials/civil servants who have the same political ranks (Yang et al., 2013). Since the establishment of SaSa C, the power of personnel appointment and dismissal of central enterprises has continued to shift from the central organisation department to the SaSa C, which forms a personnel management system with consolidated supervision on all leaders of central enterprises on the basis of SaSa C but supplemented by the central organisation department (Yang et al., 2013). In China, with the deepening of reformation of the govern- ment system, the central government has gradually delegated some power of administrative examination and approval to the local government, which enables the local government to exercise some powers (for example, approval authority and appointment power) (Zhang, 1999). Due to the difference in job position levels, government officials at different levels differ significantly in social connections, power and resources (Dai, Hong, & Pan, 2015). For example, compared with department-level officials, officials at the minister-level possess more relationship networks and power than the department-level officials. These differences, whether in power or resources, will cause different executive behaviour. The larger the polit - ical promotion space, the stronger the motivation to cater to the purpose of government t here are 112 central Soes who are supervised by SaSa C up to d ecember, 2015. in the list, the top 53 firms are ‘vice-minis- terial central enterprises’ according to management policy by SaSa C and the Communist Party Central Committee's organization d epartment. http://www.sasac.gov.cn/n1180/n1226/n2425/ CHINa JOuRNaL OF aCCOuNTING STuDIES 63 regulatory authorities and the incentives to implement corresponding policies ( Wang et al., 2014). Therefore, under the background of ‘compensation restriction requirement’, a broader political promotion space for executives serves as a stronger incentive for them to cater for political goals as well as accepting a smaller pay gap for the purpose of gaining political promotion capital. Thus, based on the foregoing analysis, our third hypothesis is stated as follows: H3: There is a negative relation between political promotion space and pay gap. 3. Research design 3.1. Variable definitions 3.1.1. Dependent variable: executive compensation (COMP), employee wage (WAGE) and pay gap (GAP) The executive compensation is mainly composed of monetary compensation and equity incentive. In the paper, we only investigate the monetary compensation of executives. Some executives receive a salary from the parent company and only receive an allowance (for expenses such as travel and conferences) from the listed firm. The independent directors only receive an allowance from the listed firm. We add up the executive compensation and exclude executives and independent directors who receive only an allowance, then we divide the result by the number of executives receiving compensation, and calculate the resulting average executive compensation. Following Li and Hu (2012) and Liu and Sun (2010) the employees’ average wage is equal to payments to employees, based on information in the Cash Flow Statement, and excluding the total compensation of directors, supervisors and executives, then scaled by the number of employees excluding the number of directors, supervisors and executives. The pay gap is the difference between the executives’ average compensation and employees’ average wage (Eriksson, 1999; Li & Hu, 2012; Siegel & Hambrick, 2005). We then take the natural logarithm of compensation to address the right skewness of the data. 3.1.2. Independent variable: promotion expectation The database, China Stock Market & a ccounting Research Database (CSMa R), does not track executive departure destinations after departure. We use the ‘Baidu’ search engine to find relevant information about the executives regarding their ultimate destination after depar- ture. We define the former executives as the chairman and general manager, and both of whom have great influence on strategic decisions and operating decisions, Moreover, their decision-making behaviour has a critical effect on performance (Li, 2000). Supervisors in Chinese companies are top managers, and they always receive higher compensation, therefore, we exclude them when we calculate employee payments. We follow prior Chinese literature about executive compensation to measure executive compensation as the average pay of directors, supervisors and executives. Such as the chairman Yanfeng Zhu of fa W Car (stock code: 000800), who worked as vice governor of Jilin Province after quitting the job in d ecember 2007. under the premise that we do not know anything about their destination, we enter ‘Zhu Yanfeng, fa W’ in the Baidu search engine, then the Baidu search engine will give us the relevant information about Zhu Yanfeng, and according to this, we can confirm that after Zhu Yanfeng left the job, he worked as vice governor of Jilin Province. 64 D. Bu ET a L. In the paper, we define promotion expectations as follows. (1) PROM . We consider departure from the company ( job-hopping) of a former exec- utive as a promotion if it happens to fall into at least one of the three following cases, (i) going forward to a government position; (ii) being the (vice) chairman, the general manager or the vice general manager (being only confined to the case where a general manager becomes the vice general manager of the parent firm while the transfer from chairman to general manager is treated as a non-promotion); (iii) becoming the chairman within the company (refers to change from general manager to chairman) or undertaking a similar position in companies at the same level. We use the promotion of the former executive as the proxy for the promotion expectation of the successor. When the executives have a promotion expectation, PROM is equal to one, otherwise it is zero. (2) PROM . We consider the destination of the former executive as the expectation of his successor. We give the value 2 to a promotion expectation, 1 to a lateral transfer, 0 to other and value –1 to a demotion. a ccordingly, PROM takes the value of 2, 1, 0, –1. (3) PROM . If the former executive acquires a government position after leaving the company, the PROM is equal to one, and zero otherwise. (4) PROM and PROM . PROM and PROM consider the levers of political promotion 4 5 4 5 expectation. PROM represents the executive levels of government department and PROM represents the executive levels of job ranks. Detailed explanations of the Promotion variables PROM and PROM are provided in Table 1. 4 5 3.1.3. Control variables Based on prior studies (Fang, 2009, 2011; Firth et al., 2006; Leone, Wu, & Zimmerman, 2006; Lu, Wang, & Zhang, 2012), we include the following control variables: (1) Ownership char- acteristics: Central SOE (C-SOE) equals one when a SOE’s ultimate controller is the SaS a C, the Treasury or other SOEs and zero otherwise; TOP is the stock proportion of the largest share- holder. (2) Board characteristic: DUAL equals one when the chairman also takes the position of the general manager, and zero otherwise; BOARD captures the size of the board; Independent Directors (INDP) is the number of independent directors on the board. (3) Financial characteristics: ROA is the ratio of total assets divided by net profit; SIZE is the natural logarithm of total assets; Leverage (LEV) is the ratio of total debt to total assets. (4) Industry (IND) and year dummy variables are included to control for the impact of industry and year. Detailed explanations of the variables in the paper are provided in Table 1. 3.2. Model We estimate the following equation to test the three hypotheses: GAP =  +  × PROM + Controls + (1) i,t i,t i,t i,t In model (1), the promotion variable is represented in turn by PROM , PROM , PROM , and 1 2 4 PROM . If the coefficient β of promotion is negative, it means that executives with a promo - tion expectation tend to narrow the pay gap, which is consistent with hypotheses 1 and 3. Whether the measurement of promotion expectation is reasonable or not should be examined. We test the rationality from two aspects which are given in 5.3. CHINa JOuRNaL OF aCCOuNTING STuDIES 65 Table 1. Variable definitions. Variable Definition COMP t he natural logarithm of executive average compensation WAGE employee average wage = t he natural logarithm of (cash paid to employees minus total compensation of directors, supervisors and executives)/(the number of employees minus the number of directors, supervisors and executives) GAP Pay gap = t he natural logarithm of (executives’ average compensation minus employees – average wage) PROM implies promotion type where former executives (either the chairman or the general manager) are promoted to (1) forward to a government position; (2) being the (vice) chairman, the general manager or the vice general manager (being only confined to the case where a general manager becomes the vice general manager of the parent firm while the transfer from chairman to general manager is treated as non-promotion); (3) becoming the chairman within the company (only refers to change from general manager to chairman) or undertaking a similar position in companies at the same level. We define the former executive’s promotion as a signal of promotion expectation for the successors, which is treated as a dummy variable (Promotion ) that is coded one if the executives have a promotion expectation, and zero otherwise. PROM We assign values 2, 1, 0, –1 to executive promotion expectation, lateral transfer, demotion and other respectively. PROM Coded one if the former executive (chairman or general manager) departed for a government department position, and zero otherwise. PROM l evels of government department where executive positions are are divided into central, province, city, county, town, within the group and other that are assigned the value 5, 4, 3, 2, 1, 0, –1 respectively, and the higher the value, the higher the level. PROM t he levels of civil servants regulated in China are nine in total, namely clerk, vice section, the section, vice department, the department, vice provincial department, the provincial department, vice-ministerial and ministerial in accordance to the civil servant rules, we assign the value 0 to promotions within the group and –1 to all other departures mentioned above. in light of such occasion, Promotion takes the value of –1, 0, 1, 2, 3, 4, 5, 6, 7, 8 correspondingly. expense ratio t he management expense scaled by sales asset utilisation ratio t he sales revenue scaled by total assets BOARD t he size of the board. INDP t he proportion of independent directors. DUAL When the chairman also takes a position of general manager, that is, a dual-position, the value is 1, otherwise 0. TOP t he largest shareholder’s share proportion. ROA t he total return on assets = net profit/ total assets LEV d ebt-to-asset ratio = total debt/total assets. SIZE t he natural logarithm of company total assets at the end of the year. C-SOE When the ultimate controllers are the SaSa C, the Ministry of f inance, other central agencies or the Soes directly report to central government, the value is 1, and otherwise 0. YEAR Year dummy variable. IND industry dummy variable. PC an indicator variable equal to one if the executive has past or concurrent work experience in the government or a political appointment, such as the People’s representative or Member of Chinese People’s Political Consultative Conference. EDUCATION an indicator variable equal to 5, 4, 3, 2, or 1 if the executive’ education is Phd , Master, Bachelor, College or senior middle school. AGE t he executive’s age in a given year. TENURE t he age of the executive being the chairman or general manager in a given year. To distinguish the effect of two types of promotion expectation, we include the interaction term of PROM1*PROM in model (1). The regression coefficient of political promotion expec - tation is (β +β ), and the intra-group promotion is β . a negative β means that the effect of 1 2 1 2 political promotion expectation on pay gap is more significant, which is consistent with Hypothesis 2. 66 D. Bu ET a L. 3.3. Data and sample Our sample consists of all the firms listed in the Shanghai and Shenzhen Stock Exchanges over the period of 2005–2012 where their ultimate controller is governmental or an SOE. a fter excluding financial firms and firms with missing financial information, our final sample consists of 6,475 firm-year observations. Information on promotion expectation is hand-collected. a ll other data are obtained from the China Stock Market and a ccounting Research (CSMaR) database. a ll continuous variables are winsorised at the 1% and 99% levels. Table 2 reports the sample distribution by year and industry. The annual trend in PROM and PROM suggests that the number of promotion expectations increased throughout our sample period from 2005 to 2012. Table 2 also reveals that promotion expectation firms tend to be located in petroleum, metal/non-metal, equipment, pharmaceutical products, retailing, or energy supply. Table 2. Sample distribution. Panel a: Distribution by year Obs. Obs. of PROM Obs. of PROM 1 3 2005 726 72 17 2006 718 109 22 2007 745 140 32 2008 787 172 43 2009 822 206 42 2010 841 219 42 2011 886 264 52 2012 950 280 57 t otal 6,475 1,462 307 Panel B: distribution by industry obs. obs. of PROM obs. of PROM 1 3 a griculture 121 27 5 Mining 272 69 15 f ood 287 44 14 t extile 140 17 8 Paper/printing 100 14 1 Petroleum 712 143 10 electronics 247 67 24 Metal/non-metal 625 162 46 equipment 1,060 252 53 Pharmaceutical products 355 93 4 o ther manufacturing 16 0 0 energy supply 405 91 17 Construction 147 40 7 t ransport 387 83 26 information technology 300 71 15 retailing 451 94 11 real estate 361 77 29 Services 206 61 7 Media 84 10 4 Multi-industry 199 47 11 t otal 6,475 1,462 307 t his table reports the sample distribution by year and industry between 2005 and 2012. Statistics are based on firm-year observations. t he 20 industries are based on the official industry classification of the China Securities regulatory Com- mission. PROM is gauged whether the firm’ executive has promotion expectation. PROM is gauged whether the firm’ 1 3 executive has political promotion expectation. CHINa JOuRNaL OF aCCOuNTING STuDIES 67 Table 3. Summary statistics. Variable Obs. Mean Min P25 P50 P75 Max SD COMP 6,475 30.93 3.80 14.41 24.39 39.05 149.61 24.91 WAGE 6,475 7.39 1.16 3.84 5.97 9.21 30.78 5.33 GAP 6,475 23.41 0.67 8.65 16.87 29.71 134.57 22.91 PROM 6,475 0.226 0 0 0 0 1 0.418 PROM 6,475 0.458 –1 0 0 1 2 0.879 PROM 6,475 0.047 0 0 0 0 1 0.213 PROM 307 3.433 1 3 3 4 5 0.819 PROM 307 4.954 1 4 5 6 7 1.557 expense ratio 6,475 0.083 0.008 0.040 0.066 0.103 0.403 0.065 asset utilisation ratio 6,475 0.740 0.073 0.381 0.618 0.931 2.785 0.517 BOARD 6,475 9.573 5 7 9 11 15 1.927 INDP 6,475 0.359 0.250 0.333 0.333 0.375 0.556 0.049 DUAL 6,475 0.097 0 0 0 0 1 0.295 TOP 6,475 39.565 10.160 27.190 39.040 51.110 75.400 15.507 ROA 6,475 0.033 –0.179 0.011 0.031 0.057 0.198 0.054 LEV 6,475 0.520 0.077 0.386 0.536 0.661 0.902 0.188 SIZE 6,475 22.010 19.702 21.109 21.823 22.723 25.888 1.264 C-SOE 6,475 0.320 0 0 0 1 1 0.467 t his table reports the summary statistics within the sample firms between 2005 and 2012. t he unit of COMP, WAGE and GAP is ten thousand yuan. in the summary statistics of ‘PROM ’ and ‘PROM ’, samples whose executives have political 4 5 promotion expectation are dropped. all variable definitions are in t able 1. 4. Empirical results 4.1. Descriptive statistics Table 3 presents descriptive statistics of all the variables. It shows that about 22.6% of the firms in our sample provide an expectation of opportunities, political promotion or intra- group promotion, but that only 4.7% of the firms in our sample provide an expectation of political promotion. Further, from the perspective of political promotion level, it is, more often than not, provincial and municipal promotions that are available for those executives qualified for political promotion, and the positions acquirable are mostly at the vice- provincial department level. During 2005–2012, the average executive compensation is CNY309,300 in Chinese listed SOEs. However, from the standard deviation, there is a larger difference among executive compensation. The highest compensation is CNY1.4961 million, while the lowest compen- sation is only CNY38,000. In contrast, the average employees’ wage is CNY73,900 in the same period, and the difference between executive compensation and employees’ wage reached CNY234,100. The average expense ratio of the sample is 0.083, and the maximum value is 0.403. The mean value of the asset utilisation ratio is 0.740, and the minimum value is 0.073. Compared with results reported in a ng, Cole, and Lin (2000), it shows that there is a higher agency cost but lower efficiency in Chinese listed SOEs than in their peers in developed economies. t he expense ratio is, in general, an indicator used to measure how effectively resources are utilised under the control of managers for accomplishing the same sales revenue. a higher expense ratio means that more resources consumed under the manager’s control are left available to the manager’s personal interests, hence a higher agency cost. g enerally, the asset utilisation ratio, measured as the sales revenue divided by total asset, is used as an indicator to capture the operating ec ffi iency of the firm. i t may be used to answer questions such as whether the management has made inappropriate and ineffective decisions on shareholders’ invests or whether management has behaved lazily and has not sought to maximum shareholders’ wealth. t he higher the expense ratio or the lower asset utilisation ratio, the higher the agency cost between shareholders and the manager. 68 D. Bu ET a L. 4.2. Results 4.2.1. Promotion expectation and pay gap The regression results of promotion expectation on pay gap are shown in Table 4, which reports a significantly negative relationship between promotion expectation (PROM and PROM ) and executive compensation at the 1% level of significance. This implies that exec - utives in Chinese listed SOEs are inclined to accept lower compensation when facing pro- motion expectation, which, in turn, suggests a substitution effect between implicit political promotion and explicit compensation. There is a positive but insignificant relation between promotion expectation and average employee wage. The promotion expectation is signifi- cantly negatively related to the pay gap at the 1% level, which captures the motivation of executives in SOEs of gaining more promotion competency by actively lowering compen- sation and narrowing the pay difference with employees. Such a promotion expectation helps to drive executives to narrow the pay gap. Control variables in the regression are consistent with the existing literature. Table 4. Promotion expectation and pay gap. (1) (2) (3) (4) (5) (6) COMP Wa GE GaP COMP Wa GE GaP Constant 5.892*** 7.267*** 4.374*** 5.899*** 7.275*** 4.381*** (41.07) (52.17) (20.19) (41.12) (52.24) (19.01) PROM –0.081*** 0.013 –0.114*** (–5.00) (0.82) (–4.62) PROM –0.036*** 0.011 –0.052*** (–4.68) (1.43) (–4.42) BOARD 0.007* 0.004 0.012** 0.007* 0.004 0.012** (1.81) (0.97) (2.12) (1.79) (0.96) (2.11) INDP 0.071 0.154 –0.117 0.068 0.159 –0.121 (0.49) (1.11) (–0.54) (0.48) (1.14) (–0.56) DUAL –0.019 –0.087*** 0.024 –0.018 –0.086*** 0.025 (–0.81) (–3.94) (0.69) (–0.77) (–3.91) (0.72) TOP –0.004*** 0.004*** –0.007*** –0.004*** 0.004*** –0.007*** (–9.22) (9.01) (–10.47) (–9.24) (8.98) (–10.49) ROA 3.418*** 1.566*** 4.164*** 3.419*** 1.565*** 4.166*** (24.03) (11.33) (19.39) (24.03) (11.33) (19.39) LEV –0.222*** –0.309*** –0.188*** –0.222*** –0.309*** –0.188*** (–5.02) (–7.20) (–2.82) (–5.02) (–7.21) (–2.82) SIZE 0.267*** 0.126*** 0.319*** 0.267*** 0.126*** 0.318*** (38.56) (18.75) (30.47) (38.52) (18.71) (30.43) C-SOE 0.114*** 0.045*** 0.139*** 0.115*** 0.045*** 0.140*** (7.62) (3.12) (6.15) (7.71) (3.10) (6.23) YEAR Yes Yes Yes Yes Yes Yes IND Yes Yes Yes Yes Yes Yes a djusted R 0.504 0.393 0.367 0.504 0.393 0.367 f statistic 189.03 120.84 108.08 188.85 120.91 108.00 no. of obs. 6,475 6,475 6,475 6,475 6,475 6,475 t his table reports the regression results of promotion expectation on pay gap. f rom columns 1 to 3, the independent varia- ble is PROM . f rom columns 3 to 6, the independent variable is PROM . all the variables are defined in t able 1. t -statistics 1 2 used to signal the robustness of standard errors clustered at the firm level are reported in parentheses. ***, **, and *, indicating statistical significance at 1%, 5%, and 10% levels respectively. CHINa JOuRNaL OF aCCOuNTING STuDIES 69 Table 5. Political promotion expectation and pay gap. Total Sample Political promotion expectation (1) (2) (3) (4) (5) (6) COMP Wa GE GaP COMP Wa GE GaP Constant 5.874*** 7.239*** 4.353*** 5.751*** 7.150*** 4.282*** (40.92) (51.97) (20.08) (36.19) (45.68) (17.78) PROM –0.062*** 0.043** –0.092*** (–3.48) (2.46) (–3.43) PROM *PROM –0.092*** –0.141*** –0.102* 1 3 (–2.67) (–4.20) (–1.95) PROM –0.152*** –0.047 –0.216*** (–4.16) (–1.31) (–3.91) BOARD 0.007* 0.004 0.012** 0.013*** 0.002 0.020*** (1.88) (1.09) (2.17) (3.02) (0.55) (3.20) INDP 0.092 0.187 –0.093 0.145 0.212 –0.035 (0.64) (1.34) (–0.43) (0.92) (1.36) (–0.14) DUAL –0.017 –0.085*** 0.026 –0.021 –0.083*** 0.016 (–0.74) (–3.82) (0.74) (–0.88) (–3.50) (0.44) TOP –0.004*** 0.004*** –0.007*** –0.005*** 0.004*** –0.008*** (–9.12) (9.16) (–10.40) (–8.97) (7.88) (–9.90) ROA 3.418*** 1.550*** 4.153*** 3.371*** 1.583*** 4.105*** (23.96) (11.23) (19.33) (22.06) (10.51) (17.72) LEV –0.224*** –0.313*** –0.191*** –0.174*** –0.285*** –0.142* (–5.08) (–7.29) (–2.86) (–3.59) (–5.98) (–1.94) SIZE 0.268*** 0.127*** 0.319*** 0.273*** 0.131*** 0.324*** (38.63) (18.86) (30.51) (35.59) (17.39) (27.89) C-SOE 0.112*** 0.042*** 0.136*** 0.087*** 0.056*** 0.096*** (7.48) (2.91) (6.05) (5.21) (3.36) (3.80) YEAR Yes Yes Yes Yes Yes Yes IND Yes Yes Yes Yes Yes Yes a djusted R 0.505 0.395 0.367 0.508 0.391 0.367 f statistic 184.15 118.28 105.23 155.33 96.94 87.75 no. of obs. 6,475 6,475 6,475 5,243 5,243 5,243 t his table reports the results of political promotion expectation on pay gap. f rom columns 1 to 3, the independent variable is PROM and the interaction term of PROM and PROM . f rom columns 3 to 6, the independent variable is PROM . all the 1 1 3 3 variables are defined in t able 1. t -statistics used to signal the robustness of standard errors clustered at the firm level are reported in parentheses. ***, **, and *, indicating statistical significance at 1%, 5%, and 10% levels respectively. 4.2.2. Political promotion expectation and pay gap For testing different effects of political promotion expectation and intra-group promotion expectation on compensation, we introduce an interaction term PROM *PROM in the regres- 1 3 sion. Table 5 shows that the coefficient of promotion expectation is significantly negative at the 1% level, and the coefficient of the interaction term is significantly negative at the 10% level, indicating that compared with intra-group promotion expectation, political pro- motion expectation has a more significant constraining effect on the pay gap. It also suggests that a scarce resource associated with political promotion expectation can also motivate executives to reduce their compensation and send a good signal to the media and public. This also indicates that the incentive effect of political promotion expectation on executives is better than that of the intra-group promotion expectation. 4.2.3. Political promotion space and pay gap To test the effect of political promotion space on pay gap, we use government level (PROM ) and position level (PROM ) as the measure of promotion space. Table 6 shows that whether 5 70 D. Bu ET a L. Table 6. Promotion expectation space and pay gap. (1) (2) (3) (4) (5) (6) COMP Wa GE GaP COMP Wa GE GaP Constant 5.828*** 7.213*** 4.295*** 5.842*** 7.210*** 4.313*** (40.28) (51.32) (19.65) (40.39) (51.34) (19.74) PROM –0.038*** –0.013* –0.050*** (–5.57) (–1.94) (–4.85) PROM –0.026*** –0.011** –0.034*** (–5.11) (–2.11) (–4.45) BOARD 0.007* 0.004 0.013** 0.007* 0.004 0.012** (1.90 (1.07) (2.20) (1.88) (1.07) (2.18) INDP 0.106 0.153 –0.068 0.109 0.154 –0.065 (0.74) (1.10) (–0.31) (0.76) (1.11) (–0.30) DUAL –0.015 –0.090*** 0.029 –0.015 –0.090*** 0.029 (–0.67) (–4.08) (0.83) (–0.64) (–4.08) (0.86) TOP –0.004*** 0.004*** –0.007*** –0.004*** 0.004*** –0.007*** (–9.10) (9.15) (–10.37) (–9.10) (9.17) (–10.38) ROA 3.402*** 1.562*** 4.143*** 3.404*** 1.561*** 4.144*** (23.93) (11.31) (19.29) (23.93) (11.30) (19.29) LEV –0.228*** –0.309*** –0.196*** –0.229*** –0.310*** –0.197*** (–5.16) (–7.21) (–2.94) (–5.17) (–7.22) (–2.96) SIZE 0.268*** 0.128*** 0.319*** 0.268*** 0.128*** 0.319*** (38.64) (19.04) (30.49) (38.58) (19.06) (30.45) C-SOE 0.112*** 0.044*** 0.137*** 0.113*** 0.044*** 0.137*** (7.53) (3.05) (6.07) (7.55) (3.05) (6.09) YEAR Yes Yes Yes Yes Yes Yes IND Yes Yes Yes Yes Yes Yes a djusted R 0.505 0.394 0.367 0.504 0.394 0.367 f statistic 189.38 120.99 108.18 189.09 121.02 108.01 no. of obs. 6,475 6,475 6,475 6,475 6,475 6,475 t his table reports the results of promotion expectation space on pay gap. f rom columns 1 to 3, the independent variable is PROM . f rom columns 3 to 6, the independent variable is PROM . all the variables are defined in t able 1. t -statistics used 4 5 to signal the robustness of standard errors clustered at the firm level are reported in parentheses. ***, **, and *, indicating statistical significance at 1%, 5%, and 10% levels respectively. using government level or position level, the political promotion space has a significant negative impact on the pay gap. The higher the political promotion space, the narrower will be the pay gap, indicating that when facing higher political promotion space, the executives have a stronger incentive to implement the laws to narrow the pay gap. 4.3. Robustness test To ensure a more robust conclusion, we use the following methods, namely, changing the measures of pay gap and promotion expectation. The regression results are consistent with the above findings, which indicates that the findings are robust. First and foremost, we use the executive compensation divided by employee wage as the measure of the absolute pay gap. We are aware that the regulators may allow a range for the pay gap. The relative pay gap may be reasonable for the gap. Our untabulated results show a significant and negative relation between promotion expectation and pay gap. It indicates that the conclusion is robust. Second, we use the method of Propensity Score Matching (PSM) to reduce the endoge- neity. The result shows that the pay gap in the promotion expectation firms is significantly lower than that of the non-promotion expectation firms using the measures of absolute compensation pay gap or relative compensation pay gap. CHINa JOuRNaL OF aCCOuNTING STuDIES 71 Third, to re-measure the promotion expectation, we select the samples with changes of chairman in order to conduct the regression for checking the robustness of our results, as most changes of general managers are to fill the vacant position after the chairman’s depar - ture, which are often considered as routine passive shifts (Yang et al., 2013). It is the chair- man’s voluntary departure that reflects the influence of the leadership authorities’ intention on executives. The regression result is consistent with the previous conclusion. Fourth, we further test some regressions concerning executive compensation for robust- ness analysis. Based on the above study of absolute compensation, we use the relative industrial executive compensation (executive compensation/the industrial executive com- pensation median) as a substitution for executive compensation. The result is consistent with the finding relating to the absolute pay gap. Finally, we re-measure the promotion expectation by taking the successor’s age into consideration. The results show a significant negative relationship between promotion and age. Then we take age 57 or 55 as a baseline because executives exceeding this age would often enter into a stage of waiting for retirement rather than hoping for promotion. Therefore, we define the promotion expectation of executives older than 57 or 55 as zero even if their predecessors got promoted either in a government department or firm. The result is con- sistent with the basic result. 5. Further analysis-the incentive effect of promotion expectation 5.1. Does promotion expectation mitigate agency cost? a ccording to the analysis above, the existence of promotion expectation drives executives to narrow the pay difference with employees, which, to some extent, helps the executives to build a good public reputation. But is the promotion expectation able to mitigate the agency cost? Is it possible for executives with promotion expectation to seek the rent of building a private empire and, thereby, hurt shareholders’ wealth? One explanation is that executives will enjoy the benefits of political promotion, and leave the risk and cost for shareholders. Thus, we further test whether the promotion expectation will mitigate the agency conflict between manager and shareholder. Following a ng et al. (2000), we use the expense ratio (operating expense scaled by sales) and asset utilisation ratio (the sales revenue scaled by total assets) to gauge the agency cost between shareholders and managers. We estimate the following regression equation. a ll the variables are defined in Table 1. Agencycost =  +  × PROM + Controls + (2) i,t i,t i,t Columns (1)–(4) in Table 7 show that promotion expectation measured either in PROM or PROM is not significantly associated with the management expense ratio and the asset utilisation ratio, which shows that promotion expectation not only fails to alleviate the agency cost between shareholders and management, it also does not improve agent efficiency. Columns (5)–(8) in Table 7 show that the coefficients of promotion expectation and the interaction term are positive but insignificant, suggesting that political promotion expec - tation is unable to reduce agency cost. Promotion expectation is only significantly and pos- itively related to asset utilisation ratio when the promotion expectation is political, implying that political promotion expectation could motivate executives to focus more on operating 72 D. Bu ET a L. Table 7. Promotion expectation and agency cost. (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) Expense a sset utilisa- Expense a sset utilisa- Expense a sset utilisa- Expense a sset utilisa- Expense a sset utilisa- Expense a sset utilisa- ratio tion ratio ratio tion ratio ratio tion ratio ratio tion ratio ratio tion ratio ratio tion ratio PROM 0.001 0.005 0.001 –0.018 (0.52) (0.37) (0.39) (–1.25) PROM –0.001 0.005 (–0.06) (0.78) PROM *PROM 0.001 0.110*** 1 3 (0.20) (3.88) PROM –0.001 0.150*** (–0.21) (4.95) PROM 0.001 0.017*** (0.61) (3.07) PROM 0.001 0.016*** (0.30) (3.84) Controls Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes YEAR Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes IND Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes a djusted R 0.252 0.290 0.252 0.290 0.252 0.292 0.262 0.282 0.252 0.291 0.252 0.292 f statistic 61.60 74.42 61.59 74.44 59.93 72.98 52.67 58.26 61.61 74.79 61.60 75.00 no. of obs. 6,475 6,475 6,475 6,475 6,475 6,475 5,243 5,243 6,475 6,475 6,475 6,475 t his table reports the results of promotion expectation and agency cost. f rom columns 1 to 2, the independent variable is PROM . f rom columns 3 to 4, the independent variable is PROM . f rom 1 2 columns 5 to 6, the independent variable is PROM and the interaction term of PROM and PROM . f rom columns 7 to 8, the independent variable is PROM . f rom columns 9 to 10, the independent 1 1 3 3 variable is PROM . f rom columns 11 to 12, the independent variable is PROM . t he dependent variables are ‘COST’ and ‘TURNOVER’. t he control variables are not tabulated in the table. all the varia- 4 5 bles are defined in t able 1. ***, **, and * indicate the statistical significance at 1%, 5%, and 10% levels respectively. CHINa JOuRNaL OF aCCOuNTING STuDIES 73 efficiency, reducing laziness and avoiding inappropriate investment. It further confirms that executives are concerned more about political promotion. Columns (9)–(12) in Table 7 report that both government level (PROM ) and position level (PROM ) are significantly and positively related only to the asset utilisation ratio, indicating that a higher promotion level motivates executives more strongly in improving operating efficiency and turnover ratio, when compared with a relatively lower promotion level. Overall, political promotion is ineffective in mitigating agency cost between shareholders and managers arising from the separation of ownership and operation rights. Moreover, political promotion, to some extent, motivates internal executives in SOEs to abuse accessible powers to achieve private interests at the expense of outside shareholders and the firm, hence that the promotion incentive bears implicitly negative effects. 5.2. Does promotion expectation improve performance? Theoretically, the promotion incentives weaken the compensation inclination of executives in SOEs. Given that promotion fails to mitigate agency problems, is it able to motivate exec- utives to work hard for the improvement in operation performance? We further test the incentive effect of promotion on performance. We choose the indicator of ROA , an account- ing performance measure used in assessing corporate profitability, as an alternative variable capturing the incentive effect. ROA =  +  × PROM + Controls + (3) i,t+1 1 i,t i,t i,t Table 8 reports the regression results of promotion expectation and performance. From Table 8 we can see that all other types of promotion are insignificantly associated with performance except the promotion type measured by PROM . The results, however, imply that promotion expectation is incapable of motivating executives to improve performance. Instead, the compensation behaves actively in driving executives to work hard to improve performance. But the positive inventiveness would be weakened when executives are prom- ised a promotion expectation. 5.4. Is the measurement of promotion expectation reasonable? a s above, we measure the promotion expectation using the predecessor promotion direc- tion. One problem is that if the definition is unreasonable, the conclusion is questionable. Therefore, whether the measurement of promotion expectation is reasonable need to be examined. We estimate the following logit model. If the coefficient of β is positive, it means the definition of promotion expectation is reasonable. Logit(PROM )=  +  × Exprom + Controls + (4) i,t i,t i,t i,t In Table 9, as we can see, the Ex prom is insignificantly positively associated with PROM. When we consider the types of promotion, the Ex prom being defined as political promotion is significant positively associated with the PROM. If the former executive is being promoted t he market performance indicator of t obin's Q is noisier than that in a developed market in that speculative activities are prevalent in the underdeveloped Chinese capital market. instead, the accounting performance could be more effective in reflecting executives’ efforts, so we choose the accounting performance rather than market performance as an alternative measuring the performance. 74 D. Bu ET a L. Table 8. Promotion expectation and performance. (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) ROA ROA ROA ROA ROA ROA ROA ROA ROA ROA t+1 t+1 t+1 t+1 t+1 t+1 t+1 t+1 t+1 t+1 PROM 0.004 0.081* (1.45) (1.88) PROM 0.002* 0.041** (1.76) (2.03) PROM 0.001 0.016 (0.27) (0.18) PROM 0.001 0.021 (1.02) (1.14) PROM 0.001 0.019 (1.05) (1.35) PAY 0.011*** 0.013*** 0.011*** 0.013*** 0.011*** 0.011*** 0.011*** 0.010*** 0.011*** 0.010*** (6.01) (6.26) (6.02) (6.32) (5.93) (5.88) (5.98) (4.85) (5.98) (5.18) PROM*PAY –0.006* –0.003* –0.001 –0.002 –0.002 (–1.80) (–1.92) (–0.17) (–1.08) (–1.30) Controls Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes YEAR Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes IND Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes a djusted R 0.242 0.242 0.242 0.243 0.242 0.242 0.242 0.242 0.242 0.242 f statistic 49.93 48.69 49.97 48.74 49.85 48.50 49.89 48.57 49.89 48.60 no. of obs. 5,517 5,517 5,517 5,517 5,517 5,517 5,517 5,517 5,517 5,517 t his table reports the results of promotion expectation and performance. f rom columns 1 to 2, 3 to 4, 5 to 6, 7 to 8, 9 to 10, the independent variables are PROM , PROM , PROM , PROM , PROM , 1 2 3 4 5 respectively. t he dependent variable is ‘ROA’. t he control variables are not tabulated in the t able. all the variables are defined in t able 1. t -statistics used to signal the robustness of standard errors clustered at the firm level are reported in parentheses. ***, **, and *, indicating statistical significance at 1%, 5%, and 10% levels respectively. CHINa JOuRNaL OF aCCOuNTING STuDIES 75 Table 9. association of the former executive’s promotion with that of the successor. (1) (2) (3) PROM Political PROM Intra PROM Constant –0.454 –4.716** –0.506 (–0.38) (–2.02) (–0.42) ex prom 0.062 0.641*** –0.101 (0.49) (2.98) (–0.78) PC 0.385*** 1.944*** –0.452*** (3.06) (9.06) (–3.06) EDUCATION 0.048 0.163 0.014 (0.69) (1.07) (0.19) AGE –0.117*** –0.064*** –0.114*** (–13.25) (–3.93) (–12.37) TENURE 0.080*** –0.024 0.097*** (3.99) (–0.64) (4.67) C-SOE –0.155 –0.775*** 0.016 (–1.39) (–3.20) (0.14) TOP 0.009** 0.012* 0.006 (2.35) (1.90) (1.63) LEV 0.160 –0.440 0.259 (0.46) (–0.69) (0.72) SIZE 0.204*** 0.202* 0.185*** (3.85) (1.93) (3.38) ROA –0.256 –4.750** 0.992 (–0.21) (–2.17) (0.77) YEAR Yes Yes Yes IND Yes Yes Yes Pseudo R 0.106 0.179 0.098 Wald chi 267.73 225.51 226.88 no. of obs. 2,151 2,151 2,151 t he table reports the regression result of the l ogit model. t he ‘PROM’ is a dummy variable equal to one if the executive is promoted to either a government position or a higher position within the company, and zero otherwise. t he ‘Political PROM’ is a dummy variable equal to one if the executive is promoted to a government position, and zero otherwise. t he ‘Intra PROM’ is a dummy variable equal to one if the executive is promoted to a higher position within the company and zero otherwise. t he Ex prom is a dummy variable which has the similar definition to PROM, Political PROM and Intra PROM, but is used to measure the former executive’s promotion. t -statistics used to signal the robustness of standard errors clustered at the firm level are reported in parentheses. ***, **, and *, indicating statistical significance at 1%, 5%, and 10% levels respectively. to government departments, the current executive may also have the chance to get pro- moted to the government departments. Given that our study mainly focuses on the type of political promotion, from the results of the logit model we can conclude that the definition of promotion expectation is reasonable. Moreover, we further analyse the executive status after their turnover during 2005 to 2012. In our sample, there are 129 executives being promoted to government departments and 549 executives being promoted to a higher position in the parent company. The pro- motion sample (678 cases) involves 440 listed firms. a mong them, there are 259 firms in which only one executive achieved promotion. The other firms have 181 executives being prompted, totalling 419 cases. It shows that there are two or more executives achieving promotion in the 181 firms. Overall, we can conclude that measuring the promotion status of the former senior exec- utive as the promotion expectation of current executives is reasonable. What makes this identification more convincing is that Gao, Luo, and Tang (2015) use the same method to test the influence of former executives on the executive compensation contract of the successor. 76 D. Bu ET a L. 6. Conclusion This paper tests the effects of executives’ promotion expectation on their compensation as well as the pay gap between executives and employees. We find that there is a significantly negative relationship between promotion expectation and pay gap, suggesting that senior executives who have a stronger promotion expectation are more willing to accept less com- pensation and therefore create a smaller pay gap. Furthermore, we find that different types of promotion expectation have different effects on senior executives’ perceptions. Compared with intra-group promotion, political promotion expectation has greater effects on down- sizing executive compensation and narrowing the pay gap. The higher the promotion ranks to which the former executives are promoted, the lower the compensation the current exec- utives receive and, thus, this narrows the pay gaps between executives and employees. However, we find that promotion expectation fails to mitigate agency conflict between shareholders and managers and further motivates the executives to maximise firm value. Our study provides an explanation to the ‘zero-paid’ phenomenon in China’s SOEs from the perspective of implicit incentives, and provides empirical evidence on incentive theory. In contrast to traditional compensation incentives – for example, options, salary and other explicit rewards – implicit incentives such as political promotion could also serve as a mech- anism to encourage executives to strive to maximise shareholders’ wealth. The restraining effect of political promotion on an excessive pay gap is stronger in an environment where the governance mechanism is relatively effective. However, the political promotion, while being effective in motivating executives in improving performance, is potentially harmful to the interests of firms and shareholders. Thus, in the current stage of China’s SOEs reform, the Sa Sa C should promote the establishment of a professional manager market and reduce government administrative intervention. Acknowledgements Danlu Bu acknowledges financial support from the National Natural Science Foundation of China (No. 71428008) and the Fundamental Research Funds for the Central universities (No. JBK160115). Disclosure statement No potential conflict of interest was reported by the authors. References a ng, J. S., Cole, R. a., & Lin, J. W. (2000). a gency costs and ownership structure. Journal of Finance, 55(1), 81–105. Baker, G. P., Jensen, M. C., & Murphy, K. J. (1998). Compensation and incentives: Practice vs. theory. Journal of Finance, 43, 593–616. Baker, G., Robert, G., & Kevin, J. M. (1994). Subjective performance measures in optimal incentive contracts. The Quarterly Journal of Economics, 109, 1125–1156. Bebchuk, L., Fried, J., & Walker, D. (2002). Managerial power and rent extraction in the design of executive compensation. The University of Chicago Law Review, 69, 751–846. Blanchard, O., & Shleifer, a. (2001). Federalism with and without political centralization: China vs. Russia. IMF Staff Papers, 48, 171–179. Campbell, D. (2008). Nonfinancial performance measures and promotion-based incentives. Journal of Accounting Research, 46, 297–332. CHINa JOuRNaL OF aCCOuNTING STuDIES 77 Cao, J., Michael, L., Pan, X. F., Qian, M. J., & Tian, G. (2011). Political promotion, CEO incentives and the relationship between pay and performance. Working paper. Chen, D. H., Chen, X. Y., & Wan, H. L. (2005). Regulation and non-pecuniary compensation in Chinese SOEs. Economic Research Journal, 40, 92–101. (In Chinese) Chen, X. Y., Chen, D. H., Wan, H. L., & Liang, S. K. (2009). The disparity in regions, the regulation of emolument, and the corruption of high-ranking managers. Management World, 11, 130–143. (In Chinese) Dai, Y. Y., Hong, Q., & Pan, Y. (2015). Official inspection, media coverage and government subsidies: Empirical evidence from Chinese listed companies. Economic Management Journal in China, 535, 13–25. (In Chinese) Eriksson, T. (1999). Executive compensation and tournament theory: empirical tests on Danish data. Journal of Labor Economics, 17, 262–280. Fan, J. P. H., Wong, T. J., & Zhang, T. Y. (2007). Politically connected CEOs, corporate governance, and Post-IPO performance of China's newly partially privatized firms. Journal of Financial Economics, 84, 330–357. Fang, J. X. (2009). Is top management compensation of Chinese public companies sticky? Economic Research Journal, 44, 110–124. (In Chinese) Fang, J. X. (2011). Managerial power and asymmetry of compensation change in China's public companies. Economic Research Journal, 46, 107–120. (In Chinese) Firth, M., Fung, P. Y., & Rui, O. M. (2006). Corporate performance and CEO compensation in China. Journal of Corporate Finance, 12, 693–714. Gao, H. S., Luo, J., & Tang, T. (2015). Effects of managerial labor market on executive compensation: evidence from job-hopping. Journal of Accounting and Economics, 59, 203–220. Gibbons, R., & Murphy, K. J. (1992). Optimal incentive contracts in the presence of career concerns: theory and evidence. Journal of Political Economy, 100, 468–505. Gibbs, M., Merchant, K. a., Van der Stede, W. a., & Vargus, M. E. (2004). Determinants and effects of subjectivity in incentives. The Accounting Review, 79, 409–436. Groves, T., Hong, Y. M., John, M. M., & Barry, N. (1995). China’s evolving managerial labor market. Journal of Political Economy, 103, 573–592. Huang, F. G., Li, G., & Li, X. W. (2011). Executive compensation, government-linked administrative rank and agency costs. Science of Science and Management of S. & T., 33, 171–179. (In Chinese) 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. Jenter, D., & Kanaan, F. (2015). CEO turnover and relative performance evaluation. Journal of Finance, 70, 2155–2184. Jia, R. X., & Nie, H. H. (2015). Decentralization, collusion and coalmine deaths. Review of Economics and Statistics, forthcoming. Leone, a. J., Wu, S. J., & Zimmerman, J. L. (2006). a symmetric sensitivity of CEO cash compensation to stock returns. Journal of Accounting and Economics, 42(1–2), 346–385. Li, H. B., & Zhou, L. a. (2005). Political turnover and economic performance: the incentive role of personnel control in China. Journal of Public Economics, 89, 1743–1762. Li, W. a., Qiu, a. C., & Gu, Z. H. (2010). Dual corporate governance environment, political connections preference and firm performance: Study on governance transition of China’s private listed firms. China Industrial Economics, 6, 8595. (In Chinese) Li, W. a., Qiu, a. C., Niu, J. B., & Xu, Y. K. (2010). Recent development of corporate governance: International trends and China’s mode. Nankai Business Review, 13, 13–24. (In Chinese) Li, W. J., & Hu, Y. M. (2012). Who is encouraged by pay dispersion in State-owned Enterprises? Economic Research Journal, 47(12), 125–136. (In Chinese) Li, Z. Q. (2000). Incentive mechanism and company performance: Empirical research based on China’s listed companies. Accounting Research, 1, 24–30. (In Chinese) Liang, S. K., Li, Z., Chen, D. H., & Chen, S. M. (2015). Political ranks, incentives and firm performance. China Journal of Accounting Studies, 3, 87–108. Lin, J. Y., & Liu, Z. (2000). Fiscal decentralization and economic growth in China. Economic Development and Cultural Change, 49(1), 1–21. 78 D. Bu ET a L. Lin, Y. F., Liu, M. X., & Zhang, Q. (2004). Political burden and company budget soft constraint: Empirical research from China. Management World, 8, 81–89. (In Chinese) Liu, C., & Sun, L. (2010). Pay gap and company performance: Empirical evidence from China’s listed SOEs. Nankai Business Review, 13, 30–39. (In Chinese) Liu, X. X. (2001). The development report of China SOEs: 1990-2000. Beijing: Social Sciences a cademic Press. (In Chinese) Lu, Z. F., Wang, X. Y., & Zhang, P. (2012). Do Chinese state-owned enterprises pay higher wage? Economic Research Journal, 47, 28–39. (In Chinese) Maslow, a. (1954). Maslow’s hierarchy of needs: Motivation and personality. New York, NY: Harper. Quan, X. F., Wu, S. N., & Wen, F. (2010). Managerial power, private income and compensation rigging. Economic Research Journal, 45(11), 73–87. (In Chinese) Siegel, P. a., & Hambrick, D. C. (2005). Pay dispersion within top management groups: Harmful effects on performance of high- technology firms. Organization Science, 16, 259–274. The Research Group of unirule Institute of Economics. (2011). Nature of state-owned enterprises, performance and reform. Working paper. (In Chinese) The World Bank. (2012). World Development Indicators 2012. Washington, DC: The World Bank. Wang, Z., Fu, G. Q., Huang, D. Y., & Wang, J. F. (2014). a study on the relationship between the ‘political rise’ of the CEO in State-owned firms and the consumption when they are at their post. Management World, 5, 157–171. (In Chinese) Xin, Q. Q., & Tan, W. Q. (2009). Market-oriented reform, firm performance and executive compensation in Chinese State-owned Enterprises. Economic Research Journal, 44, 68–81. (In Chinese) Yang, R. L., Wang, Y., & Nie, H. H. (2013). Promotion mechanism of quasi official: Evidence from China central enterprises. Management World, 3, 23–33. (In Chinese) Zhang, L. L., Liu, F., & Cai, G. L. (2015). Regulatory independence, market-oriented process and promotion mechanisms implementation of SOE’s executive. Management World, 10, 117–131. (In Chinese) Zhang, W. Y. (1999). The enterprise theory and Chinese SOE sector reform. Beijing: Peking u niversity Press. (In Chinese) Zhang, W. Y. (2014). SOE is impossible to establish effective corporate governance. Cai Zhi Dao, (198). Zheng, Z. G., Li, D. X., Xu, R., Lin, R. T., & Zhao, X. J. (2012). The political promotion and image establishment of SOE Executives: a case study based on province N, company a. Management World, 10, 146–156. (In Chinese) Zhou, L. a. (2002). Career Concerns, Incentive Contracts, and Contract Renegotiation in the Chinese Political Economy. PhD thesis, Stanford university. Zhou, L. a. (2004). Government officials’ incentive and cooperation in the promotion game: The problem of local protectionism in China and repeated construction. Economic Research Journal, 39(6), 33–40. (In Chinese) Zhou, L. a., Li, H. B., & Chen, Y. (2005). Relative performance evaluation: a n empirical study on the Chinese local officials’ promotion. China Journal of Economics, 1(1), 83–96. (In Chinese) Zhou, L. a. (2007). Study on the promotion tournament mode of China local officials. Economic Research Journal, 42(7), 36–50. (In Chinese)

Journal

China Journal of Accounting StudiesTaylor & Francis

Published: Jan 2, 2016

Keywords: Agency cost; pay gap; political promotion; promotion expectation; promotion rank

There are no references for this article.