The Motivation and Consequences of Golden Parachute Provisions: A Case Study of TBEA Co., Ltd.

ABSTRACT The paper examines the motivation and consequences of Golden Parachute (GP) contracts in the context of TBEA, a Chinese company whose GP payment was 1,000 times executive annual salary and which rescinded its GP provision in 2019. We find that for TBEA, whose ownership is dispersed, anti-takeover was the main motivation for the adoption of GPs, and that managerial power was the key factor in designing GPs with payment of high monetary value. We also find that such GPs may induce higher excess executive compensation, lower shareholder participation, and reduce firm value, and that market reaction to the rescinding of GPs is positive. These results show that emerging capital markets should beware of the negative effect of GPs on firm value.


Introduction
Some listed companies in China have adopted golden parachutes (GPs), and these have attracted substantial attention from investors and public officials. GPs are contracts with key executives such that if the company is acquired, its directors and executives can receive a large lump-sum payment on their departure if their term has not expired. The lump-sum payment can be as high as three times the total annual salary of a director or executive (as in the case of Shanghai Kehua Bio-Engineering Co., Ltd., stock code 002022) or even 10 times salary (as with China Baoan Group Co., Ltd., stock code 000009).
There is a substantial empirical literature examining the motivation and consequence of GPs in the United States. Optimal contracting theory holds that GPs have a positive impact and that shareholders use GPs to reduce conflict between shareholders and managers. Especially during a period of merger and acquisition (M&A), GPs can increase the likelihood of acquisition success and an acquisition premium, which enhances the wealth of shareholders. In contrast, managerial entrenchment theory holds that GPs are adopted by management to protect their own interests by weakening the force of the market for control, thereby making managers less fearful of acquisitions and increasing managerial slack. As the agency problem can be more serious in such companies, GPs ultimately damage their market value.
Almost all studies of GPs have focused on the US capital market, and very few have focused on China. However, China's capital market environment and corporate governance structure are quite different from those of the US, and the motivation and consequences of GPs in the US will not be straightforwardly generalisable to China's market. Since only a limited number of companies in China have adopted GPs, it is not feasible to conduct large-sample empirical research in that context. Therefore, this paper adopts a case study approach to examine the adoption of GPs in listed companies in China, at the same time taking advantage of the opportunity that case studies afford to analyse causality in a more comprehensive way.
Reviewing the GP provisions of China's listed companies, we find that TBEA Co., Ltd. added GP provisions to its Articles of Association as far back as 2006: If the company is taken over by another firm and directors, supervisors, executives, core management team (i.e., the management team of the company's controlling subsidiaries and public branches) and other core backbones are terminated as a result of the merger or takeover, which must be identified by themselves, the company must pay a lump-sum economic compensation equivalent to 1000 times their total annual salary.
Of the companies that adopted GPs, TBEA gave the highest payment, which therefore had the most significant incentive effect; for research purposes, this characteristic is helpful for identifying the consequence of GPs. What is unique and even more valuable for research purposes is that in 2019 TBEA called a board meeting to rescind its GP provision from the Articles of Association. As these circumstances are very helpful in verifying the value effect of GPs, we take TBEA as our case study.
The results show that anti-takeover is the motivation for GP provision, especially for a company with a dispersed ownership structure, and that managerial power is the key factor in the adoption of GPs with high payments. We also find that GPs with high payments can induce greater excess executive compensation, lower shareholder participation, and a reduction in market value. The rescinding of GPs, on the other hand, meets with a positive market reaction. Therefore, emerging capital markets like China should beware of the negative effect of GPs on a firm's value.
The principal contributions of the paper are as follows. First, we extend the literature on GPs, most of which has focused on the US market and, given the mixed evidence it has provided, has failed to resolve the issue of the value effect of GPs. In a market like that of the US, where a company's ownership is dispersed, the agency problem mainly focuses on conflict between shareholders and managers. From this perspective, some researchers have argued that GPs have a positive effect (Berkovitch & Khanna, 1991;Harris, 1990;Lambert & Larcker, 1985;Small et al., 2007). However, a substantial body of work supports the view that GPs have a negative effect (Cochran et al., 1985;Singh & Harianto, 1989;Wade et al., 1990). In this paper, we examine the effect of GPs on a listed company in the transition economy of China, where the agency problem between controlling shareholders and minority shareholders coexists with the agency problem between shareholders and managers. The paper thus provides a better understanding of the adoption of GPs under different institutional environments, while also contributing to the literature on anti-takeover measures.
The conclusions are also valuable to regulators and investors. Regulators in China are already aware of the potential negative effects of GPs. For example, in 2016, China Baoan Group (stock code: 000009) and Shandong Jintai Group (stock code: 600385) added GP provisions to their Articles of Association. In the same year, the Shenzhen Stock Exchange and the Shanghai Stock Exchange sent a supervision attention letter 1 and an inquiry letter, 2 respectively, to those companies concerning their GPs. Although the companies explained their rationale for their adoption of GPs, so far there has been no empirical evidence to support their claims. Thus, the findings of this paper provide regulators with quantitative evidence to help formulate appropriate policies. For investors, the results also show that more attention should be paid to the adoption of GPs in a listed company. Although GPs may to some degree contribute to the stability of a company, in the long run they have a negative effect on firm value.

Literature review
In the early 1980s, GPs originated from a wave of takeover activity in the US. A GP is a kind of contract signed between the management and board of a company such that if control of the company changes hands, then the managers receive a certain degree of compensation if they leave their current position, either actively or passively. The amount of compensation is determined according to the agreed multiple in combination with the historical salary level of the manager concerned. The compensation can take the form of cash, stock options, or other long-term compensation schemes. Unlike a general severance payment contract, a GP can be triggered only when the company control rights change. The earliest example of a GP dates to 1982, when William Agee, the CEO of Bendix before it was taken over by Allied Corporation, made the board sign an economic compensation contract with the managers (including himself). According to the contract, if Bendix was acquired and the managers left their current positions, actively or passively, they would receive as compensation a lump-sum payment equivalent to five times their annual salary. The acquisition went ahead; Agee actively resigned and received $4 million compensation. Subsequently, there was a surge in the number of the companies adopting GPs. By 1986, one-third of the 250 largest companies in the US had adopted them. By 1990, more than 300 Fortune 500 companies had done so.
Previous theoretical work on GPs has focused on optimal contracting theory and managerial entrenchment theory, both of which seek to explain the emergence of GPs in terms of a conflict of interest between shareholders and managers. The main difference between the two theories lies in the identity of the implementer. Optimal contracting 1 On 7 June 2016, China Baoan Group Co. Ltd received a supervision attention letter from the Shenzhen Stock Exchange which asked, 'Does the setting of the economic compensation clause violate the eighth rule of the Managing Method of Listed Company Acquisition? What is the impact of the clause on the company? Will it be under suspicion of tunnelling? And will it violate the duty of loyalty?' 2 In August 2016, Shandong Jintai Group Co. Ltd received an inquiry letter from the Shanghai Stock Exchange about amendments to its Articles of Association. The group was required to answer certain questions about the economic compensation in its GPs: (i) What are the legislative authority and rationale for the compensation payment standard? (ii) As the compensated person is the director or senior manager of the company, does the compensation payment constitute a related party transaction, and has the decision-making procedure been adhered to? (iii) What are the legislative authority and rationale for paying a lump-sum payment for a 'hostile takeover'? Does it encroach on the interest of interests of the company and its shareholders? Will the company be under suspicion of tunnelling? (iv) What will be the impact of the compensation payment on the company's performance and what, in detail, are the relevant risks?
theory holds that GPs are adopted by shareholders to alleviate agency problems, whereas managerial entrenchment theory holds that GPs are a measure of entrenchment adopted by managers to safeguard their own interests.

GPs from the perspective of optimal contracting theory
Optimal contracting theory holds that GPs are adopted by shareholders to reduce conflict between shareholders and managers during a period of acquisition. Acquisition is an event that is usually accompanied by premiums that can enhance shareholder wealth. However, when an acquisition is completed, the company management are highly likely to be replaced, and the current managers will safeguard their own interests by resisting the acquisition. In this context, GPs, which are a kind of severance compensation, may be used to make up for the management's risk of dismissal in a transfer of control rights. GPs can thus reduce the management's potential loss, thereby reducing their resistance to and increasing the likelihood of the acquisition (Berkovitch & Khanna, 1991;Harris, 1990;Lambert & Larcker, 1985) and contributing to shareholder wealth. Research has shown that GPs increase not only acquisition likelihood but also acquisition premiums. Generally speaking, compared with M&A, tender offers involve higher payments and are associated with higher acquisition premiums (Berkovitch & Khanna, 1991). However, following a tender offer, the management are usually replaced. Thus, to safeguard their own interests, management are likely to resist tender offers. From the perspective of optimal contracting theory, the adoption of GPs can reduce management resistance to tender offers, which can increase shareholder wealth; to this extent, GPs are valuable to shareholders.
Apart from increasing the likelihood of acquisition, GPs can enhance shareholder wealth in other ways. GPs can protect shareholder wealth by encouraging the use of M&A instead of bankruptcy (Evans & Hefner, 2009). Similarly, when a company is in financial distress, GPs can help to recruit managers with better reputations, which helps the company handle its difficulties better (Maskara & Miller, 2018). Studies have also shown that companies with GPs have younger management, which indicates that shareholders tend to establish a stable long-term relationship with young CEOs by means of GPs (Lefanowicz et al., 2000); this is the embodiment of optimal contracting theory. In this connection, Falaschetti (2002) observed that GPs are more likely to be adopted in companies with more concentrated ownership, which suggests that GPs are a measure used by shareholders to reduce conflict between shareholders and managers, particularly during a period of acquisition.
From the optimal contracting theory perspective, GPs are a kind of corporate governance that shareholders actively adopt to facilitate the success of M&A. When there is a chance of external acquisition, the interests of management are usually damaged (in terms of lower pay and loss of position within the company) because of the change of control. Management may therefore resist external acquisition for personal interest, and their resistance will reduce the chances of an acquisition being successful. Nevertheless, acquisition is an important way to enhance shareholder wealth, and shareholders often take their own steps to increase the likelihood of successful M&A. By protecting the interests of management during an acquisition, GPs can therefore reduce the agency problem between shareholders and managers and protect the company's interests overall. Thus, optimal contracting theory holds that GPs are an important tool used by shareholders to reduce the agency problem between shareholders and managers, and that GPs are positively associated with both acquisition likelihood and premiums.

GPs from the perspective of managerial entrenchment theory
Managerial entrenchment theory holds that GPs are used by management when they feel threatened from M&A (Chakraborty, 2008). To protect their interests, managers demand GPs from the board (Cochran et al., 1985;Singh & Harianto, 1989;Wade et al., 1990). A company CEO will evaluate the risks, and the need for self-defence, according to the characteristics of the company. Company size is inversely correlated with the likelihood of being acquired; the larger the company, the less likely it is to be acquired and the lower the likelihood of management demanding GPs (Cochran et al., 1985).
Although research has confirmed that risk can enhance managerial entrenchment motivation, the influence that management can exert on the board is key to the adoption of GPs. The management's influence on the board depends on factors including management tenure (whether their term is absolute or relative to the board term), ownership dispersion, and the number of outside directors appointed by CEOs. Management from companies with longer CEO terms, greater ownership dispersion, or a higher proportion of outside directors appointed by CEOs can exert greater influence on the board, and this kind of company is more likely to adopt GPs (Cochran et al., 1985;Singh & Harianto, 1989). A company in which the management have great power often has cash-flow problems, as the management are in a position to spend cash, which may lead to overinvestment. Under such circumstances, management feel the need to seek protection, and this may take the form of demanding GPs (Subramaniam & Daley, 2000).
From the managerial entrenchment perspective, GPs reduce shareholder wealth. In the US, GPs are a kind of contract between shareholders and managers for which approval at a meeting of shareholders is not necessary. This leaves open the possibility that management can use their influence with the board of directors to damage the interests of shareholders. The adoption of GPs will make management less fearful of M&A and increase managerial slack (L. Bebchuk et al., 2014;Morck et al., 1989), and the existence of GPs can weaken the effect of external market supervision on the company. It follows from this that, rather than being a tool for solving the agency problem, GPs themselves are part of the agency problem.
In terms of managerial entrenchment, GPs are an anti-takeover measure. Managers use GPs to defend themselves against the threat of acquisition, which can strengthen their power (Chakraborty, 2008). This can weaken the governance effect of the market for control based on external acquisition, which is regarded as an important governance mechanism, as the external market's ability to compete with existing management plays a role in ensuring that managers are diligent and perform their duties appropriately (Jensen & Ruback, 1983). However, managerial entrenchment theory holds that the adoption of GPs with high compensation can stop an acquisition and then weaken market discipline in terms of corporate control. Therefore, GPs can reduce the power of the market for control, which weakens the monitoring role of the external market over management and increases managerial slack. As a measure of management entrenchment, then, GPs can exacerbate the agency problem between shareholders and managers, which will damage the development of the company and decrease its value.

Summary
From this review of the literature, we can see that research on GPs has mainly focused on the US market, where corporate governance is based on the agency problem between shareholders and managers (henceforth, the first agency problem). Optimal contracting and managerial entrenchment theories are both centred on the relationship between shareholders and managers, which is not particularly relevant to China's capital market. In emerging markets such as China, the ownership of a listed company is relatively concentrated, and the management are usually appointed by the controlling shareholders. As a result, conflict between shareholders and managers is not the only or most important source of agency problems, and conflict between controlling and minority shareholders constitutes another important kind of agency problem (Claessens et al., 2002;La Porta et al., 1999;Tang et al., 2012). La Porta et al. (1998Porta et al. ( ), (1999) observed that ownership of listed companies is dispersive in only a few countries, such as the US and the United Kingdom. In most countries, especially in emerging markets, ownership is highly concentrated. Therefore, research on this second kind of agency problem (the kind generated by conflict among shareholders) is more relevant for emerging markets such as China. This constitutes a very important difference between this paper and research on GPs in the US market.
As mentioned above, GPs in China are usually a product of the relationship between controlling and minority shareholders. In China's listed companies, management are unlikely to ask the board for GPs, and controlling shareholders are unlikely to provide management with GPs as a way of improving their relationship with management; the ownership of listed companies in China is concentrated, and management are usually appointed by the controlling shareholders. Thus, management and controlling shareholders share the same interests. As a result, there is little incentive to improve their relationship, and controlling shareholders are unlikely to adopt GPs due to pressure from management. Therefore, theories based on the relationship between shareholders and managers are not applicable to the interpretation of GPs in China's listed companies. Instead, the analytic framework based on the agency problem between controlling shareholders and minority shareholders is more appropriate for analysis of the adoption of GPs in the context of listed companies in China.
The agency problem between controlling and minority shareholders is the key element in the effectiveness of corporate governance in China (Tang et al., 2012). The main cause of the second agency problem is the separation of control rights from cash-flow rights, which can lead to negative behaviour, such as tunnelling by controlling shareholders (Bebchuk et al., 2000;Bertrand et al., 2002;Wang & Zhou, 2006). In a company with serious agency problems, governance problems are more likely, including shareholder embezzlement. No existing research has examined GPs from the perspective of the second agency problem. However, for the development of China's capital market and to safeguard the interests of minority shareholders, it is important to reduce the agency problem between controlling and minority shareholders. Therefore, this paper examines the adoption of GPs in listed companies in the context of the agency problem between controlling and minor shareholders.
As a kind of important executive compensation contract, GPs are closely associated with the agency problem and corporate governance (Berkovitch & Khanna, 1991;Chakraborty, 2008). From the optimal contracting perspective, a listed company uses GPs to provide a stable developing environment in the company, which can guarantee the effective and consistent implementation of the company's strategy. Under such circumstances, GPs can have a positive effect on corporate value. However, GPs can also increase the difficulty of acquisition, which weakens the force of the market for control. Manne (1965) argued that the market for control is a kind of important external corporate governance mechanism and that it can play a disciplinary role. Specifically, the controller of a company faces the threat of losing control, which places restrictions on the controller's actions, thereby reducing the agency problem in the company (Bertrand & Mullainathan, 2003;Jensen & Ruback, 1983).
In a capital market characterised by concentrated ownership, as in China, the external capital market can restrain tunnelling behaviour by large shareholders (Xu et al., 2018). Avoiding external acquisition can create a relatively stable environment for large shareholders and managers. However, it can also weaken the monitoring role of the market for control, and this weakening can damage the company's market value. Gompers et al. (2003) found that the weaker the force of the market, the lower the stock returns of the company. Thus, from the perspective of external governance, GPs have a negative effect on the market value of listed companies, and a better understanding of the causes and consequences of GPs in China is required. Although GPs in China are a relatively recent phenomenon, it is meaningful to use traditional academic research methods to examine their causes and consequences in the emerging market. The findings can serve a warning function concerning the fall-out risks and disadvantages of GPs, which will help to improve the emerging capital market. Table 1 summarises the listed companies that have adopted GPs, and the following characteristics are evident. First, the shareholding proportion of the largest shareholder is relatively low. From 2003 to 2018, the average shareholding proportion of the largest shareholders in China's listed companies was 36.46%, and the median was 34.41%. In contrast, the average shareholding proportion of the largest shareholders of the 20 companies with GPs was 21.85%, and the median was 20.33%. Second, the ownership concentration is relatively low. From 2003 to 2018, the average Herfindahl Index_10 3 for China's listed companies was 0.1740, and the median was 0.1431. However, the average for the Herfindahl Index_10 Herfindahl Index_10 was 0.0747, and the median was 0.1431. Third, the CEOs of companies that adopted GPs were appointed by the largest shareholder; the actual controller of the company was thus the largest shareholder (or the largest shareholder's controller). Given the consistent association between the largest shareholder and the actual controller, it is clear that the management were appointed by the largest shareholder and the actual controller of the listed company, and therefore that the management and largest shareholders were likely to share the same fundamental interests. As a result, the agency problem between controlling and minority shareholders Yes was more pressing than the agency problem between shareholders and managers. This makes it appropriate for an analysis of GPs in China to integrate optimal contract theory and managerial entrenchment theory. When the shareholding proportion of the largest shareholder is low, the probability of a change to the company's control rights is high. At the same time, the control rights of the listed company are of high value because of the approval system 4 in the China stock market. On the basis of this analysis, the adoption of GPs in China is likely to reflect the joint entrenchment of large shareholders and managers. The uniqueness of China's institutional environment therefore creates a need to examine GPs in the context of China's listed companies. Table 2 gives an overview of the GP situation, showing that all the listed companies set target GPs for directors, supervisors, and executives, with the exception of TBEA. Since TBEA is unique in this respect, we investigated this point further and found that it is common for TBEA management to hold important positions in its subsidiaries. The logic of broader coverage is the same as that of monetary value, which greatly increases the potential cost of acquisition and is a reflection of excessive management security. However, this broader coverage not only provides management with greater protection but also greatly increases the difficulty of acquisition. Therefore, compared with the GPs adopted by other companies, the TBEA GP has broader coverage, and this is one of the reasons why this study takes TBEA as its case.  Table 3. In terms of size, TBEA has grown rapidly, from about 900 million yuan in 1998 to 83.6 billion yuan in 2018. Its profitability is nothing unusual, with an average ROA of 4.4% and an average ROE of 11.1% over the past 20 years. Although its average debt ratio of 61.5% seems high, the average debt with interest ratio for which regular interest payments are required is only 22.3%. Thus, the pressure for TBEA to repay its debts is not high and, given the average commercial credit ratio of 20.3%, it is in a strong position as regards accounts payable and advances from suppliers and customers in the supply chain.

Corporate governance
TBEA's predecessor was the Changji transformer factory, where, in 1988, Zhang Xin, the current controller of TBEA, took over as director. At that time, the outlook for the factory was not optimistic, as its debts exceeded 700,000 yuan and bankruptcy liquidation was anticipated for the following year. However, under the leadership of Zhang Xin, the factory soon turned a profit and was listed on the mainland A-share market. This made TBEA the first listed company in the transformer industry. At that time, the largest shareholder of TBEA was the Xinjiang Changji special transformer factory, which had a collective holding of 16.18% of TBEA shares. Although Zhang Xin held control of TBEA through the Xinjiang Changji factory, the collective nature of the factory meant that his control was not permanent. In the early 2000s, China's collectively owned enterprises ushered in a wave of restructuring. In 2003, the Changji factory was restructured into Xinjiang Tianshan Investment Co., Ltd., making Zhang Xin the largest shareholder, with 33.82% of the shares of the Xinjiang Tianshan Investment Co., Ltd. In 2002, Shanghai Honglian Venture Capital became the second-largest shareholder of TBEA, holding 13.92% of shares. The investors of Shanghai Honglian were employees and management of TBEA acting in concert with the Xinjiang Tianshan group. Thus, Zhang Xin, the controlling shareholder of TBEA, became the actual controller of TBEA. According to the latest information disclosed by TBEA, the Xinjiang Tianshan group has changed its name to TBEA Group Co., Ltd. and holds 12.02% of TBEA shares. At the same time, Shanghai Honglian Co., Ltd. changed its name to Xinjiang Honglian Venture Capital Co., Ltd. and acquired 6.83% of TBEA shares. The shareholding proportion of the Xinjiang TBEA group controlled by Zhang Xin thus increased to 40.08%. The shareholders of Xinjiang Honglian Venture Capital Co., Ltd. are still management and employees of TBEA in 2002, but, as Figure 1 shows, Zhang Xin's control over TBEA had decreased from 30.10% in 2002 to 18.85%.
The TBEA 2016 annual report shows that the total shareholding proportion of TBEA held by the Xinjiang TBEA group and Xinjiang Honglian Venture Capital Co., Ltd. Xinjiang Honglian Venture Capital Co., Ltd., acting in concert, is 18.42%. The shares held by the founding team represented by Zhang Xin are not enough to enjoy absolute control over the listed company. TBEA's third-to tenth-biggest shareholders hold 16.88% of the total shares; this includes 10 fund companies, ranked sixth, that hold the same proportion of shares, at 1.07%. The proportion of the largest shareholder is therefore low, and 10.70% of the shares held by 10 funds are of high liquidity. As a result, TBEA is at high risk of acquisition. Since 1988, Zhang Xin has been the director of the Changji transformer factory and then, following the joint stock reform, the chairman of TBEA. As the controller of TBEA, Zhang Xin has been in the top position for 30 years and has absolute power. The  general manager is Huang Hanjie, who previously served as finance manager and then CFO. Overall, then, TBEA's corporate is characterised by ownership decentralisation and strong management power.

Motivations for adopting GPs
As observed in Section 2, GPs in China's capital market may reflect the joint entrenchment of the controlling shareholder and management. We examine this hypothesis from the perspective of ownership structure and managerial power.

Ownership structure
As shown in Section 3, the ownership concentration of TBEA is low. In this section, we further analyse TBEA's ownership structure in the context of the electrical component and equipment industry (fourth-level industry classification in the Wind database). Figure 2 shows that the sum of the squares of the top 10 shareholders' shareholding proportions of TBEA is far lower than the average and the median for the industry. The ownership concentration of TBEA is far lower than the industry level, which indicates that the actual controller's control of TBEA is not solid and that TBEA is at threat of external acquisition. To resist external acquisition and weaken the force of the market for control, the management of TBEA have a strong incentive to demand anti-takeover measures such as GPs. Therefore, ownership dispersion may be an important reason for the adoption of GPs in TBEA, which is consistent with the conclusion of Cochran et al. (1985) that the risk of acquisition triggers the defence motivation. Companies that adopt GPs generally have a concentrated ownership structure, and TBEA is not a special case. To date, 20 listed companies in China have adopted GPs, and the ownership concentration of these companies is shown in Table 1. With the exception of Guangdong Highsun, the concentration is lower than the average for all listed companies. In general, the higher the ownership concentration, the more stable the control of the major shareholders and the lower the threat of M&A from external investors. In the case of these 20 companies, the ownership concentration is low and the control of the large shareholders is not secure, and thus the threat of external M&A is high. Figure 3 shows the overall ownership concentration of companies with GPs, calculated as the average of the difference between the ownership concentration of the company that adopted GPs and the industry level of the company each year. From this we can see that the ownership concentration of companies with GPs is significantly lower than the industry level, which indicates that companies adopting GPs have decentralised ownership. It is thus quite possible that these companies are under threat of external acquisition.
Information for the listed companies in China that have adopted GPs confirms that one of the main purposes is to resist external acquisitions. For example, Hifuture Technology adopted GPs to resist the threat of 'Qihoo 360 backdoor listing'. Within one month of the first backdoor rumours, the management of Hifuture Technology had submitted to the board and shareholder meeting a proposal to add GPs into its Articles of Association. Similarly, following the reform of non-tradable shares in 2006, Guangdong Highsun, a listed company with vast mineral wealth, adopted GPs as protection against hostile takeover. After witnessing the fight for control between Vanke and Baoneng Holdings, 5 China Baoan was quick to  provide its management with GPs; after the lifting of share restrictions, Shanghai Kehua Bio-Engineering did the same. League Agent (HK) Ltd, an institutional investor that was the largest shareholder of Shanghai Kehua Bio-Engineering, held 95,863,038 shares, of which 40,489,116 were lifted on 20 June 2017. From the above data and analysis, it can be inferred that ownership dispersion and increase of stock liquidity are important reasons for the adoption of GPs in China's listed companies. In the case of decentralised ownership or enhanced stock liquidity, management are more likely to be threatened by the market for control, and the adoption of GPs can be used as an anti-takeover measure to consolidate their position.
The separation of control and cash-flow rights is another important feature of ownership structure and a significant source of the second agency problem, namely the problem between controlling and minor shareholders (Johnson et al., 2000). This paper analyses the second agency problem from the perspective of the separation of these rights. As Figure 4 shows, in the case of TBEA there is a serious separation of the two rights of the kind that may reflect the tunnelling of the large shareholder. This shows that TBEA may suffer from the second agency problem in a serious form, a hypothesis for which further evidence is presented in Section 5 of this paper.

Managerial power
The key to the adoption of GPs is the ability of management to influence the board of directors and shareholder meetings. Existing research shows that GPs are related to managerial power (Cochran et al., 1985;Singh & Harianto, 1989;Wade et al., 1990), and strong managerial power increases the likelihood that GPs will be approved. Therefore, this paper analyses managerial power in terms of tenure, position, and shareholding.

Management tenure
The core members of TBEA have a long term of service. As mentioned above, Zhang Xin, the actual controller, has been the main leader of TBEA for nearly 30 years, and thus has absolute power in TBEA. You Zhicai, another core member, has been the CFO of TBEA for nearly 30 years; he retired in March 2015 for age reasons. Ye Jun has been the director and deputy general manager of TBEA since the company was listed. In 2002, he took over from Zhang Xin as general manager and stayed in that capacity until 2015. Chen Weilin, the supervisor of TBEA, has been the chairman of Xinjiang TBEA group, the company's largest shareholder. Thus, it is clear that Zhang Xin, You Zhicai, Ye Jun, and Chen Weilin have played important roles in the company over a long period. Zhang Xin, the soul of TBEA, has been firmly in control of TBEA for more than 20 years, and the other three have also held key positions in the company for a long time.

Position of senior management
Zhang Xin, You Zhicai, and Ye Jun, the core management of TBEA, have held important positions in the corporate governance layer and shareholding entity. The 2001 annual report shows that Zhang Xin was both chairman and general manager of TBEA, and that Ye Jun was the director and vice general manager. Meanwhile, Zhang Xin and Ye Jun were also directors of Xinjiang Tianshan Electric Co., Ltd., TBEA's largest shareholder. The TBEA supervisor Chen Weilin has been the chairman of the Xinjiang TBEA group, and CFO You Zhicai is the chairman of Shanghai Honglian Venture Capital Co., Ltd., TBEA's secondlargest shareholder. As a related party of the largest shareholder, Shanghai Honglian Venture Capital Co., Ltd. holds more than 10% of TBEA shares. Thus, the core management not only hold key managerial positions but also play an important role in the governance layer, having significant influence as directors in the large shareholder entities. Therefore, the management, represented by Zhang Xin, have a great influence on the company from the position perspective.

Management shareholdings
Initially, Zhang Xin, Ye Jun, and Chen Weilin together held 83.42% of the shares of Xinjiang TBEA group, TBEA's largest shareholder. By 31 December 2015, they held 33.61% of the shares of Xinjiang TBEA group and 12.09% of the shares of Shanghai Honglian Venture Capital Co. Ltd., TBEA's second-largest shareholder. The remaining 87.91% of the shares of Shanghai Honglian Venture Capital Co., Ltd. are held by the middle and senior management of TBEA. These shareholding proportions indicate that the management of TBEA have great power within the company. Thus, from the shareholding perspective, the management represented by Zhang Xin wield a great influence on the board of directors and the shareholder meeting.
The analysis above shows that the management, represented by Zhang Xin, have the ability to influence the board of directors and the shareholder meeting to pass a resolution to adopt GPs. It could even be said that they are the actual controller of the company. Thus, the agency problem between shareholders and managers is not serious. Since optimal contracting theory focuses on reducing the agency problem between management and large shareholders, it is clearly not applicable to explaining the adoption of GPs in the current context, where managerial power is the key factor in designing GPs with great monetary value. The ownership of TBEA is dispersed, and the shareholding proportion of the actual controller is relatively low, which indicates that the risk of acquisition remains for TBEA. In such a situation, managers will tend to make use of their power to ensure that a shareholder meeting approves the adoption of GPs, and this may be an important factor in TBEA's economic compensation being as high as 1,000 times annual salary.

Consequences of adopting and cancelling GPs
Existing research has made use of market reaction to examine the effect of GPs. For example, Lambert and Larcker (1985) showed that the adoption of GPs is associated with a positive stock market reaction. However, other research has suggested that the adoption of GPs has a negative effect on stock market returns (L. Bebchuk et al., 2014;Hall & Anderson, 1997), and some research has found no effect of GPs on the stock market (Born et al., 1993;Davidson et al., 1998). Here, we note that the act of examining market reaction can be enough to erase the information behind the market reaction. Therefore, we analyse the economic consequences of GPs from the perspectives of executive compensation, shareholder activism in corporate governance, and corporate performance. The case of TBEA is uniquely valuable for our purposes, because in 2019 it rescinded GPs from its Articles of Association. Accordingly, this paper makes use of the rescinding behaviour to examine the positive and negative value of GPs and to derive conclusions that are robust and reliable.

Levels of executive compensation
GPs can strengthen managerial power and weaken the force of the market for control, which reduces the external supervision of management (Bebchuk et al., 2014;Morck et al., 1989). When the market force is weakened, the managers have greater power and can design a more favourable compensation plan to raise their income levels. The management of TBEA enjoyed GPs that provided economic compensation as high as 1,000 times salary. Such sky-high compensation largely weakens the force of the market for control, which can make executive compensation incompatible with company performance.
To better examine the effect of GPs, this paper takes as its baseline the listed companies in the same industry as TBEA. Figure 5 shows that, in the early stages, TBEA's executive compensation was lower than the industry level. In 2007, the total compensation of TBEA's top three executives began to exceed the industry level, and it stayed consistently high thereafter. In 2017, the total compensation of TBEA's top three executives amounted to 5.1 million yuan, which was more than twice the industry level. Thus, the level of TBEA's executive compensation was far higher than the industry level.
An optimal managerial compensation system has a great positive effect in reducing the agency problem between shareholders and management. Thus, the system should be highly sensitive to the company's performance in order to align the interests of shareholders and those of managers. Figure 6 shows that TBEA's ROA rose consistently, reaching 8.37% in 2009. However, after 2009, its ROA began to decline, reaching a low point of 2.22% in 2012. This is far lower than the industry level for the same period (4.54%). From  2012 to 2018, TBEA's ROA was maintained at 3%, still below the industry level. Therefore, despite an ROA that has been lower than the industry level since 2011, TBEA's executive compensation has been higher than the industry level since 2007.
The above analysis indicates that the profitability of TBEA is lower than the industry level, but that its executive compensation has always been higher. Thus, following the adoption of GPs, there was a large degree of separation between executive compensation and the performance of TBEA. Separation to that extent is usually attributable to managerial power, that is, to the management's control of the company (Bebchuk et al., 2002;Lu, 2008). The stronger the managerial power, the more likely management are to use that power to design compensation terms in their own favour, and this reduces the sensitivity of management pay to company performance. Furthermore, good corporate governance can inhibit management's self-interested behaviour in compensation incentives (Wu & Wu, 2010). GPs with high economic compensation can weaken the force of the capital market for control in corporate governance and reduce its constraining effect on selfinterested behaviours in relation to executive compensation. Thus, GPs can increase the separation between executive compensation and company performance.
To enhance the generalisability of these findings, this paper goes on to compare the executive compensation and ROA of companies that adopted GPs with their own industry levels. From the results in Figure 7, we see that the compensation of the top three executives of companies with GPs is significantly higher than that of the industry generally. However, the profitability of companies with GPs does not match this level of compensation; in most years, the ROA of these companies has been lower than the industry level, and, since 2015, the ROA has shown a clear deviation from the executive compensation. This further strengthens the conclusion that GPs weaken the positive effect of the market for control in corporate governance and reduce the constraining effect on self-interested behaviours in relation to executive compensation. Therefore, GPs can increase the mismatch between executive compensation and company performance.
Next, the paper calculates TBEA's excess executive compensation to provide a better understanding of the separation of executive compensation and company performance. Excess executive compensation involves managers using their power and influence to pursue rent-seeking behaviour and to secure more income than they would obtain from fair negotiation (Bebchuk & Fried, 2003). Following Core et al. (1999), Fang (2012, and Yang and Zhao (2012), this paper uses the following model to calculate executive excess compensation: Ln(PAY) is the natural logarithm of the top three executives' compensation. SIZE is measured by the natural logarithm of total assets. LEV is total liabilities divided by total assets. BM is the book market value ratio, that is, the ratio of total assets to total market value. ROA is the ratio of net profit to total assets at the beginning of the period. DUAL is a dummy that equals 1 when the chairman and the general manager are the same person, and 0 otherwise. BOARDSIZE is the size of the board of directors, measured as the natural logarithm of the number of directors. MNS measures the shareholding proportion of management. SOE is a dummy that equals 1 when the actual controller is state-owned, and 0 otherwise. The paper conducts regression analysis according to the model by year and industry. The residual is used to measure the excess executive compensation, which is a relative value used to compare the compensation levels of different companies after controlling for company performance. Taking companies in the same industry as TBEA as the sample, Figure 8 gives the residual of TBEA in the regression, which reflects the excess executive compensation of TBEA. We can see a clear trend in the growth of TBEA's excess executive compensation since the adoption of GPs. Although the excess decreased briefly in 2010, since 2013 it has been increasing, reaching a high point in 2018. Overall, the excess executive compensation was positive in most years and shows a clear upward trend.
In Figure 9, we summarise the excess executive compensation of companies that have adopted GPs. Their overall level of excess pay has risen consistently. Moreover, the overall level of managerial excess compensation of these companies grew rapidly, exceeding the industry level, after 2015, the time when most of these companies adopted GPs. These findings provide further evidence of a causal link between GPs and excess executive compensation.
These findings in relation to excess executive compensation confirm that, as a kind of management entrenchment measure, GPs weaken the force of the market for control. The increase of managerial power due to GPs provides conditions in which management can seek higher payment that is not sensitive to performance.

Shareholder activism
Shareholder activism reflects the belief that external shareholders can play an effective supervisory role and can help to improve the corporate governance structure (Smith, 1996). The shareholder meeting is an important corporate governance institution that provides minor shareholders with an opportunity to voice their opinions about the development of the company. However, in the case of TBEA, the adoption of GPs further strengthened the control of management, a situation that can aggravate the agency problem between controlling and minor shareholders. When minor shareholders realise that their interests have been damaged but can do nothing to change this, more and more of them will give up their rights (i.e. vote with their feet). The key indicator of shareholder activism is the proportion of those with voting rights who attend shareholder meetings. Table 4 shows that the proportion of shareholders attending annual shareholder meetings has decreased over time, and in the past decade was only a little higher than 20%. Acting in concert, Xinjiang TBEA group and Xinjiang Honglian Venture Capital Co., Ltd. together hold 18.42% of the shares. The remaining shareholders attending the shareholder annual meeting hold only 3% of the shares. Thus, any special resolution with the support of the actual controller can be passed, and the shareholder meeting cannot play its role in corporate governance.
Research has shown that institutional investors can improve corporate governance structure (Cheng, 2006). However, in this case we find that the 10 fund companies that together hold 10.7% of the shares did not participate in the shareholder meeting, which means that the institutional investors of TBEA are unlikely to be effective. The GP system, which is likely to have strengthened the entrenchment of the controlling shareholders, not only reduced the participation of minor shareholders in corporate governance but also decreased the motivation of institutional investors to play a role in corporate governance.
Overall, GPs further consolidate the control of the controlling shareholders and the managers, who are the same group. This can reduce the participation of external shareholders in corporate governance. However, a decrease in the activism of external shareholders can harm the improvement of corporate governance and the development of the company.

Market value of the company
The external market can have a positive effect on corporate governance. Through the disciplinary effect of the market for control, management can be supervised and constrained, and this has a positive impact on the company value (Jensen & Ruback, 1983). However, the high compensation involved in TBEA's GPs greatly increases the difficulty of external acquisition, which weakens the force of the market for control. This can damage the market value of the company, as high compensation can result in lower levels of P/E ratio and Tobin's Q, which are common measures of company value.
The valuation level of TBEA has decreased over years. Table 5 shows a clear downward trend in the P/E ratio and Tobin's Q of TBEA, which in 2007 reached their highest levels (45.22 and 2.75, respectively). Both indicators subsequently decreased; in 2018, they had fallen to 9.95 and 0.27, respectively. TBEA's valuation level is also low relative to the industry, and its P/E ratio and Tobin's Q have always been lower than the industry level. This gap between TBEA and the rest of the industry has gradually increased. In 2007, the P/E ratio of TBEA reached its highest-ever level, at 78.25% of the industry median. However, in 2018 it was only 39.70% of the industry median. As for Tobin's Q, TBEA exceeded the industry median of industry once only, in 2009. In 2008, its Tobin's Q was only 24.55% of the industry median. This provides further evidence that GPs with sky-high compensation greatly increase the difficulty of external acquisition and weaken the force of the market for control, which damages the market value of the listed company.

Market reaction to cancellation of GPs
It is clear from the above considerations that the adoption of GPs can damage the market value of a company. However, if there is causality from GPs to firm value, the cancellation of GPs can have a positive effect on firm value. On 12 April 2019, TBEA held a board meeting to review and approve a resolution to rescind the provision of GPs in its Articles of Association. Although TBEA did not announce a reason for cancelling its GPs, the cancelling behaviour itself reflects the fact that the GPs had weakened the force of corporate governance and had a negative effect on the company's performance. Accordingly, it is very likely that the behaviour was due to pressure from the external market. This paper uses the event study method to study the market reaction and to clarify the economic effect of GPs. Taking the announcement date of the board meeting cancelling GPs as the event day, (−300, −30) as the estimation window, and (−1, 0) as the event window, we use the industry returns to estimate normal performance and adopt a market adjustment method to determine abnormal returns. The results show that during the event window the cumulative abnormal return was 2.32%, which indicates a positive reaction by the capital market to TBEA's cancellation of GPs. Therefore, the market holds a negative attitude towards GPs, regarding them as not conducive to the development of a company and as damaging to its value. The cancellation of GPs means that large shareholders and management can be supervised better by the external market. As a result, they can better fulfil their duty of diligence, and the agency problem is reduced. These results provide further evidence that the joint behaviour of controlling shareholders and management are the most relevant factors for understanding the GPs of listed companies in China.

Conclusions
This paper has shown that entrenchment theory is more applicable than optimal contracting theory in explaining the adoption of GPs in China's listed companies. In the US, the ownership of listed companies is usually dispersed, and the entrenchment effect of GPs derives mainly from management. However, in China, the situation is totally different, as there is a significant possibility that the adoption of GPs in listed companies is an antitakeover measure and a joint action by large shareholders and managers. According to the present results, under the joint action of the larger shareholders and managers, the adoption of GPs can induce higher excess executive compensation, lead to a lower degree of minority shareholder participation in corporate governance, and reduce the market value of the company. Conversely, the cancellation of GPs can improve corporate performance and enhance the firm's value. Therefore, high compensation in the form of GPs is most plausibly regarded as a kind of agency problem that damages shareholders' interests and has a negative effect on the long-term development of companies.
A number of recommendations follow from this conclusion. First, institutional investors and minority shareholders should participate actively in reviewing any resolution to adopt GPs and should exercise their right to vote to safeguard their interests. However, if a resolution to adopt GPs with very high compensation has already been passed, it may be more appropriate for investors to vote with their feet rather than continue to hold shares. Second, regulators can issue inquiry letters to listed companies to express their concern about the unreasonably high compensation of GPs and to limit the compensation of GPs when necessary. If the adoption of GPs is an anti-takeover measure, shareholders and regulators should scrutinise its motivation and evaluate its impact, resisting unreasonable provisions in order to prevent the anti-takeover measure from becoming a managerial entrenchment tool. If they do not do this, the adoption of GPs may simply help management to safeguard their power and avoid market supervision. These implications are crucial for the improvement of corporate governance and the healthy development of the capital market in China.
This paper has also shown that the adoption of GPs is very likely to damage the value of listed companies, indicating that entrenchment theory may be better equipped to explain the adoption of GPs. However, there remains a possibility that the purpose of the adoption of GPs is to maintain a stable developing environment for the company. For listed companies, although the market for control can play a role in corporate governance, a stable environment is also important for healthy development, and listed companies might adopt GPs with this in mind. From this perspective, GPs may have a positive effect on the firm's value. Therefore, further research should continue to examine the effect of GPs.
As few listed companies in China have adopted GPs, the sample size of this study is not large enough for the purposes of archival research. Therefore, we have used the case study method, selecting a case in which GPs were cancelled a few years after their adoption, as this provides an appropriate context in which to examine their effects. However, given the disadvantages inherent in any case study in terms of determining causality, when more listed companies in China have adopted GPs, future studies should take the opportunity to conduct archival research to examine further the motivations for and economic consequences of this form of compensation.

Disclosure statement
No potential conflict of interest was reported by the author(s).