Effects of CEO turnover on firm performance : a short-term and long-term analysis of listed companies in the Chinese market

(1)

Effects of CEO Turnover on Firm Performance

A short-term and long-term analysis of listed companies in the Chinese market

University of Amsterdam

Faculty of Economics and Business MSC Business Economics Finance Track Master Thesis Author: Haibin Xu Student number: 10874798 Supervisor: L. Zou Date: 05-07-2015

(2)

Statement of Originality

This document is written by Student Haibin Xu who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

ABSTRACT

In this study, I examine the short-term and long-term effects of CEO turnover on the firm performance for the Chinese listed companies from 2001 to 2010. The empirical results show that the CEO turnover tends to produce positive effects on the firm performance, both short-term and long-short-term, with the short-short-term effects more significant. I further study how the nature of succession affects the firm performance after a CEO turnover. The OLS regression model suggests that when a regular CEO turnover occurs, the internal succession is less efficient than external recruiting in terms of the improvement of firm’s return on equity.

(4)

Part 1: Introduction

1.1 Background

CEO turnover is an inevitable issue for company in the process of development. It could affect all aspects of a company. For internal organization, CEO turnover is regarded as the most important and the most sweeping change in management. For external organizations, such as shareholders, providers and so on, they may regard CEO turnover as the signal of the future development for company (Beatty & Zajac, 1987). Therefore, CEO turnover has a direct impact not only on the company's political and economic environment but also on the firm performance (Kesner & Sebora, 1994). CEO is one of the important human resources, and CEO who decides developing strategy and policy of company is also ultimately responsible for the performance of company. To some extent, a company's success or failure largely depends on the success or failure of the CEO (Kesner & Sebora, 1994). The ability of CEO often determines the company’s future developing prospects. Nevertheless, there is a proxy problem existing in modern enterprise which is caused by the separation of ownership and managerial, as well as the asymmetric information. This leads to a conflict of interests between CEO and shareholders. In order to ensure the interests of the shareholders, companies set restrictions on CEO’s right and a series of governance mechanisms to strengthen the supervision and constraint of CEO. When the company's performance is far below expectations of shareholders for executives’ personal reasons, the shareholders will consider firing and seeking the appropriate one. To sum up, CEO turnover is one of the major strategic decisions for the company's future development strategy, and would produce a profound influence on the firm performance. This leads to the following questions: whether CEO turnover would produce positive effects on the firm performance, both short-term and long-term. What are the main factors influencing the firm performance after CEO turnover? These are the questions this paper focuses on.

On this issue, foreign research starts early and has formed a large number of relatively mature and systematic results. And extensive research has been done on the impact of CEO turnover on firm performance in the western countries, which are based on the market economy. Nevertheless, fewer research is based on the transitional economic environment,

(6)

2

which is completely different from the market economy. After more than 20 years’ development, China is in the transition stage from planned economy to market economy. During this period, the forms of corporate ownership become complex and diverse. However, the emergence of CEO turnover events frequently happened in the Chinese listed companies in the process of rapid development since the Chinese securities market was established in the early 1990s. It is well known that executive has a pivotal position in the company. Hence, the research based on Chinese listed company is still of great significance. Because there are lots of differences between Chinese securities markets and foreign securities market. The foreign research results cannot be directly applied to the Chinese market. Additionally, few domestic scholars study in this respect, and research conclusion is not consistent. This paper does the research on the impact of CEO turnover on firm performance based on this background. Additionally, I further study how the nature of succession affects the firm performance after a CEO turnover, which is considered as the main related factor. I hope it could be helpful for the domestic scholars study in China.

This paper is divided into eight parts. The first part is introduction including the research background and research significance. The second part is related to predecessors’ research from home and abroad, and analyzes the research approaches, the innovation and shortage in this thesis. The third part introduces the foundations of turnover and firm performance theory. The fourth part expounds the classification of CEO turnover and the reasons for leaving. The fifth part explains the impact analysis of CEO turnover on firm performance. The sixth part sets the related hypotheses, and introduces the data and measurement that the paper uses in doing research about the effects of CEO turnover on the firm performance. The seventh part is about econometric specifications and results including the descriptive statistical analysis, the Logit regression to check the previous conclusion, and the OLS regression analysis to test empirically on the corporate performance change after CEO turnover. Lastly, the eighth Part is about some suggestions and prospects on this issue.

1.2 Research significance

1.2.1 Theoretical significance

(7)

3

not consistent. This paper summarizes related theory and highlights issues that are not yet fully understand.

 Provide new empirical evidence for CEO turnover study in Chinese listed companies. Because the research based on CEO turnover and firm performance in China started relatively late, there is no mature system. In contrast, foreign research has achieved fruitful results, so there is a large gap between the Chinese security market and foreign security market, and the foreign research results cannot be directly applied in China. Based on the actual situation of China, this paper uses the empirical method and tries to provide new empirical evidence.

1.2.2 Practical significance

 Provide some basis for the decision-making of CEO turnover in China. Because CEO turnover would not only affect the stability and solidarity of company organization, but also directly affect the company's future strategic decisions and the expectations of the investors, creditors, internal employees and so on. This paper could provide the owners in the company with some basis for decision-making of CEO turnover, and help to reduce the personnel changes’ shock and impact on the company.

 Provide some new insights into corporate governance in Chinese listed companies. Because CEO turnover is an important part of the corporate governance mechanism, research on this issue can analyze the existing problems in the corporate governance mechanism. It would be helpful for Chinese listed companies to establish modern enterprise system and improve the level of corporate governance.

Part2: Literature review

Prior literature could be divided into two directions because of the measurement of firm performance. One research aspect is about investigating and analyzing the stock market reaction to CEO turnover by using event study method. Warner et al. (1988) suggests that the CEO turnover cannot lead to a significant change in stock price. However Denis (1995), Weisbach (1988) and Huson et al. (2001) document that the announcement of CEO turnover could produce a positive reaction to the stock price. Hillier (2006) uses 703 UK listed companies from 1993 to 1998 as samples to study the company performance and stock price movements before and after CEO turnover and finds out that there is a decline in the

(8)

4

company's performance before compulsory CEO turnover, but a sharp increase after the change. Nevertheless, because there are many kinds of stock market in China, including state shares, legal person shares, the public social shares (A shares), and foreign shares (B shares and H shares) etc. The state shares and legal person shares cannot circulate on the market, mainly through the form of a transfer agreement, in which the price is lower than A share price. And the listed companies do not have the same price in the different markets (Du Ying and Liguo Liu, 2002). Hence, the stock price index is different in different Chinese market. Meanwhile, Due to the thick speculative atmosphere and high stock turnover rate, the stock price is seriously deviated from the firm performance. Hence, this paper will not focus on the stock price reaction.

The other research aspect is about analyzing the effects of CEO turnover on accounting performance, which is the main issue this paper will focus on. In terms of this issue, foreign scholars have done a lot of researches, but do not have any consistent conclusion. Theoretically, they could be divided into the following three kinds of different points of views: The first is the theory of "common sense". Since the poor performance of company leads to CEO turnover, the boards will choose a successor who is more capable and experienced. Thus the company's performance can be improved after CEO turnover in the theory of “common sense”. The second view is the "scapegoat" theory. The theory argues that the real reason for CEO turnover is not the poor firm performance, and executive is just a scapegoat and victim. Therefore, the replacement of CEO will not have any significant effects on the firm performance. The last point is a "vicious cycle theory". This view considers that CEO turnover will have a negative impact on corporate performance. Because CEO turnover will disturb the original network behavior patterns and relationships in the organization, and increase operating costs in the organization. Thus it will reduce the company’s performance, and poor performance will lead to another new CEO turnover.

2.1 Specific research conclusions

2.1.1 A review on foreign research

(9)

5

Denis (1995) collects 908 company samples of CEO turnover caused by merger, divides them based on turnover reasons into two groups, and examines the performance changes before and after three years. It documents that the company's performance does not fall before the regular change happening, but increases after CEO turnover. Furthermore, there is a sharp decline in the firm performance before the change, and a significant increase after the mandatory change. Kang and Shivdasani (1998) find out that CEO turnover is negatively related to the company's performance, and compulsory CEO turnover can significantly improve the firm performance on the study of the relationship between CEO turnover and corporate performance based on the Japanese listed companies. Leker and Safomo (2000) use structural equation model to analyze the different causes of CEO turnover and the relationship between the company's performance and CEO turnover. The study indicates that the reasons for CEO turnover will affect the company's performance after turnover, and there is an obvious improvement on the corporate performance after the mandatory replacement. Khorana (2001) uses 393 fund companies where there are CEO turnover as research samples, and concludes that the more poorly firm performed before CEO turnover, the higher improvement after the replacement there is. Kato and Long (2005) indicate that the CEO turnover will significantly improve the company's performance. And compared with the state-owned companies, there is a more obvious increase in the performance of non-state-state-owned companies after CEO turnover. Cristian (2006) collects data of 800 listed companies in the USA from 1998 to 2000 as sample and considers that mandatory CEO turnover can improve the firm performance in the short-term. Karlsson and Gary (2009) conclude that external succession after mandatory CEO turnover will improve the firm performance in the long-tern through the survey of CEO succession.

 CEO turnover could not significantly improve firm performance

Gamson and Scotch (1964) argue that it is understandable that the new CEO fails to improve the performance of a company after the change. Because the decline of firm performance is not necessarily due to the CEO, who is just a scapegoat. The replacement of CEO is just a strategy to placate shareholders. Zajac (1992) finds out that CEO turnover does not have obvious effects on the corporate performance based on the samples of 118 companies after CEO turnover. Firth, etc. (2002) study the relationship between the CEO turnover and corporate performance of listed companies, and suppose that the short-term effects of CEO

(10)

6

turnover are more significant than the long-term effects. Bushman (2010) studies that regular CEO turnover cannot change the company’s governance mechanism, which is the main reason why regular change could not affect firm performance.

From foreign literature review, it can be seen that the overseas research on CEO turnover starts early, and most scholars abroad have reached an agreement of the improvement on the company's financial performance after CEO turnover, especially in the case of compulsory CEO turnover. But there are also some scholars draw different conclusions. These studies have documented that CEO turnover cannot significantly improve the company's performance, or just have certain effects on the performance in the short-term. Kesner & Sebora (1994) consider that the diversity of different scholars' research conclusions are due to the different research method, the different sample selections, the different selection of performance indicators, and the different statistical methods. All of these bring about the inconsistent research conclusion.

2.1.2 A review on domestic research

 CEO turnover could improve firm performance significantly.

Xiaoxia Tian and Xiusheng Cheng (2006) study the present situation and change reasons in the private listed companies. This study documents that executive replacement is negatively related to the performance of private enterprises. Additionally, internal succession can improve the firm performance more significantly than externally succession. Chen Jian and Jia Juan (2006) analyze the impact of CEO turnover on corporate performance based on the listed companies whose control power transferred, and conclude that there is an obvious improvement in the company's performance in the listed companies with the control power transferred. Zhou Jian, Liu Xiaoyuan and Fang Gang (2009) detect that mandatory change improves firm performance more significantly than regular change based on the study of CEO turnover and the effectiveness of corporate governance. Jiang Rong and Liu Xing (2010) also conclude that mandatory change is more efficient than regular change based on the data of Chinese listed companies in 2008.

 CEO turnover could not improve firm performance significantly.

Zhang Dan and Chan Haixing (2005) use the data of listed companies from 1997 to 2003 in China as samples to study the relationship between CEO turnover and firm performance. The

(11)

7

results show that the change of top executive does not significantly improve the company's operating performance in the two years of observed time window. Ma Lei and Xin LiGuo (2008) study the relationship among the board characteristics, operating performance and CEO turnover based on the data from 1998 to 2006. The results indicate that the probability of CEO turnover in the listed company is improved with the poor firm performance, but the operating performance cannot be improved significantly after the replacement of executives. Zhao Zhenyu, Yang Zhishu and Bai Chongen (2007) conclude that CEO turnover is not the main reason for firm performance basing on the research of CEO turnover in China’s listed companies. Ran Min and Du Lanjun (2009) use the empirical tests on the CEO turnover in declining firms and find out that there is no significant relationship between CEO turnover and firm performance.

From the literature review, it is obvious that domestic scholars’ research on CEO turnover begins merely after 2000, and still stay in the exploratory stage. For the research conclusion, Chinese scholars studying in the executive corporate performance after CEO turnover still have not obtained a relatively consistent conclusion. Based on predecessors’ research, CEO turnover is mainly caused by two reasons: one is the replacement of the controlling shareholder led by the company, and the other is due to poor management. This helps to classify the reasons for CEO turnover into two categories: regular change and mandatory change. Furthermore, there are many drawbacks in most related literature based on the second research aspect. Most predecessor’s research focuses on the short-term effects and did not have uniform event windows for short-term and long-term effects. Some literatures even do not have suitable control variables and remove the effects of industry and year. These make their research have different results.

2.2 Research approach and methods

This paper uses the impact of CEO turnover on corporate performance as a foothold. Firstly this paper embarks from the theoretical basis, and expounds the relationship between CEO turnover and firm performance from the perspective of the principal-agent theory and corporate governance theory. The second part classifies and analyzes the reasons for CEO turnover and ways of succession. Then this paper also analyzes the impacts of the CEO

(12)

8

turnover on corporate performance based on the internal organization, the choice of successor and earnings management. Additionally, this paper collects data of listed companies in China from 2001 to 2010, and makes an empirical analysis and puts forward some countermeasures and suggestions according to the research conclusion and further research prospects.

2.3 Innovation and shortage

2.3.1 Innovation

 The selection of time span is large, which could make research results more scientific and comprehensive.

 On the selection of indicators, this paper uses two performance indicators to make the empirical results more scientific, not using a single indicator.

 This paper adds some important control variables such as company’s level, size and removes the effects of industry and year, which could reduce the impacts of other factors.  This paper studies not only the short-term effects but also the long-term effects of CEO

turnover.

2.3.2 Shortage

 Due to the limitation of my theoretical level, the research content of this paper may lack in depth and breadth, which it is still not perfect.

 Due to the limitation of empirical research itself, there are a lot of factors that influence corporate performance. In this paper, the research conclusion cannot fully reveal the comprehensive development of things, which may produce some impact of research results in the actual cases.

Part3: Foundations of turnover and firm performance theory

3.1 Definition of CEO

As the name implies, chief executive officer (CEO) is a senior manager who has the final decision-making, and he is ultimately responsible for enterprise operating performance. At present, CEO is a fuzzy concept in China that is close to the concept of general manager,

(13)

9

chairman and senior executive. These concepts have the similar meanings, but also have different connotations. In foreign study, the research is mainly based on the change in CEO. The concept of CEO is derived from the corporate governance structure in America. Nevertheless, there is no any concept of CEO in the company law in China. Generally speaking, CEO is the manager who masters the major business decision-making power. Hence, it could be simply interpreted as the combination of business leaders and managers.

In China, many limited liability companies are reformed by state-owned enterprises which carry out the system of overall responsibility by the factory manager. The factory manager is worthy of the name "head", who not only has the production and management decision-making, but also is a representative of the state-owned enterprises. After the enterprise is reformed, the original factory manager is often appointed as chairman and general manager of the company by the national government. According to the requirements of corporate governance, chairman and board of directors should be voted by the general meeting of shareholders, and general manager is appointed by the board of directors.

Under the complex governance structure, it is difficult to define clearly the real sense of the CEO, the chairman or general manager. Although the general manager has a production and management decision-making power. However, this power may not reach the level of CEO. Because regulation of Company Law said that chairman is the legal representative of the company and should be responsible for the company's production operation management comprehensively. Zhong Jiyin (2002) sets whether chairman works every day as standard, and reports three kinds of situations by surveying CEOs in Chinese listed companies:

(1) If the chairman holds a concurrent post of general manager, then he is the CEO.

(2) If the chairman does not hold a concurrent post of general manager and does not go to work every day in the company, then the general manager is the CEO.

(3) If the chairman does not hold a concurrent post of general manager, but goes to work every day in the company, then who is the CEO depends on who has more power.

In the practical situation, chairman and general manager tend to have the power of operation and management and only have difference in the specific scope and responsibilities. Hence, it is biased that if only the chairman of the board of directors or general manager is defined as the company's senior executive officer. CEO in this paper refers to the chairman, general manager and senior executive.

(14)

10

3.2 Principal-agent Theory

Principal-agent relationship is the product of highly developed commodity economy, and emerges with the socialization and the scale of production. On the one hand, productivity makes larger scale and more refining division of labor. Due to the lack of energy, knowledge and specialty, the owner of the enterprise may not be qualified for operation and management of company. On the other hand, specialized division could produce a large number of professional managers. In this case, the owners start to commission manager to manage company, and the enterprise ownership and right of management begin to separate. Then, there is a principal-agent relationship formed between owners and managers.

Principal-agent theory is an important part of the Contract Theory in economics. In 1932, Bede and Means published the book of in The Modern Corporation and Private Property and raised the concept of separation between ownership and control. They were keenly aware of the agency problem and found out that if control power is in the hand of managers, it would be harmful to the interest of the shareholders.

After that, some economists gradually developed the principal-agent theory based on the research of information asymmetry and incentive issues. In 1976, Jensen and Meckling (1976) made a great contribution and developed the principal-agent theory. They supposed that an agency relationship is one in which “one or more persons (the principal[s]) engage another person (the agent) to perform some service on their behalf that involves delegating some decision-making authority to the agent”. However, the issue about agency problem that arise from this, particularly the dilemma that the principal and agent, while nominally working toward a goal, may not always share the same interests.

In the modern enterprises, the company's shareholders and professional managers will sign the contract to guarantee normal operation of the company. Shareholders are not only the company's investors and owners, but also the principal. Professional managers are the agent who are commissioned and paid by shareholders to operate and manage company. Then two sides form a principal-agent relationship. Nevertheless, because both sides are the rational-economic man in this kind of principal-agent relationship, and their interests are different. The goal of shareholders is to maximize shareholder’s wealth, and managers want

(15)

11

to pursue self-interest maximization. Hence, managers may often operate and manage enterprise based on their own interests, such as getting additional compensation through on-the-job consumption, which would damage the interests of the shareholders. As a consequence, the conflict of different interests between shareholders and managers leads to the emergence of agency problem.

One of the most important causes of agency problem is information asymmetry between shareholders and managers. According to the information asymmetry problem, normally agency problems are divided into two categories: One is the adverse selection which occurs before establishing the principal-agent relationship. Shareholders may not know the information of managers’ background, ability and character and be unable to accurately identify the merits of the managers. This increases the possibility of hiring incompetent managers and causes adverse effects on the enterprise. Another one is moral hazard which occurs after establishing the principal-agent relationship. Managers engage in the specific operation and management of production, master and control the resource configuration, and are familiar with the operation of the entire enterprise. These lead managers to have more information and be in a dominant position for information. In this case, managers are likely to damage the interests of shareholders for their own sake.

In order to reduce the loss from agency problem and constrain managers' self-interest behavior, enterprise's shareholders would take incentive and constrained measures to limit the behavior of managers. However, it is impossible to completely eliminate the agency cost in the realistic society, because shareholders still need to hire a manager for their business. In this case, shareholders could set up some appropriate incentive mechanism, or take a certain supervision fees to constrain agents’ behavior, and then reduce differences in interests between the principal and agent. Additionally, principals could make a final decision to replace their managers and seek for other appropriate managers, if principals think the company's managers are unqualified.

3.3 Corporate governance theory

(16)

12

Corporate governance theory could be said as the extension and further development of Contract theory and Principal-agent theory. Williamson (1975) first theorizes the concept of corporate governance concept. In the early 80s of last century, lots of scholars started to study and developed corporate governance theory on the basis of the principal-agent theory. Oliver Hart (1995) puts forward the analysis framework of corporate governance theory in Corporate

Governance: Some Theory and Implications. He thinks that it will create corporate governance

problems if two below conditions occurred in an organization: the agency problem is the first condition. To be more precise, there are some conflicts of different interests among members of the organization (probably owners, workers and consumers). The second condition is the huge transaction costs that lead agency problem to be solved hardly by the contract. Miller (1995) supposes that mechanism of corporate governance should solve the following problems: "In the operation and management, which standards or guidelines managers should follow? Who will judge whether managers really successfully use company resources? Who is responsible for seeking for better managers to replace them if managers did not successfully use company resources?" Thus it can be seen that proxy agent and information asymmetry are the fundamental causes of corporate governance problem. And the corporate governance theory is to solve the efficiency loss and deviation caused by the differences in interests between shareholders and managers.

There are still no unified conclusion formed about the concept of corporate governance. Sheifer and Vishny (1997) consider that corporate governance is to make sure company's shareholders to get a return on investment. Colin Mayer (1999) defines the corporate governance as "an organization arrangement serve investors. There are still a lot of definitions about corporate governance, but this paper argues that the definition contains at least two following meanings: Firstly, corporate governance is an institutional arrangement involving structure and organization. Secondly, the purpose of corporate governance is to solve the principal-agent problem through measures such as motivator, power balance and supervision and constrain the behaviors of agent who would damage the interests of shareholders for their own interests.

(17)

13

The manifestation of corporate governance is the company's governance mechanism. Jensen (1993) divides corporate governance mechanism into four kinds of control power in The

Modern Industrial Revolution, Exit, and the Failure of Internal Control Systems: The capital

market and takeover market; legal, political and management system; product market and factor market, and the core of internal control system based on the board of directors. Among these, the first three are the company's external control mechanism. After that, many scholars study company governance mechanism and put forward his own understanding, which greatly enrich the content of the governance mechanism. Xiangdong Ning (2005) summarizes the governance system of American companies as "five lines of defense": executive compensation; the board of directors; the exercise of voting rights to ask for replacement or change the boards; merger and takeover activity; the supervision by public and securities regulatory institution. The five lines of defense essentially divide the company governance mechanism into two categories: internal and external interaction on managers, which ensure to manage enterprise based on maximizing the interests of shareholders. To make it better understand, this paper still uses the division of company internal and external governance mechanism. To be specific, the external control mechanism includes product and factor markets, capital and takeover market, manager market, and legal, political and management system, etc. Internal control mechanism includes the characteristics of the board of directors, the executive compensation and the ownership structure, etc. Corporate governance is a set of institutional arrangements to solve the principal-agent problem between shareholders and managers. Hence, the fundamental purpose is to prompt managers to better service for the company through motivator, supervision and constraint. The internal and external control mechanism mutual complete each other, and play a role in supervision and restraint of managers. And CEO turnover is the result of interaction of the internal and external control mechanism.

In the internal control mechanism, the board is the highest decision-making institution, which is selected by the shareholders to supervise company and makes approval decisions for the company's major business policy. The board chooses CEO to be the company's manager based on ability and level, establishes a principal-agent relationship, and uses a variety of measures to motivate, supervise and constrain the behaviors of CEO. Certainly, if CEO does not complete scheduled task, or even harm the interests of the company for his own self-interest, the board has the right to fire and seek suitable managers.

(18)

14

In the external control mechanism, product factor market, creditor governance and manager market would also restrict managers from different angles to avoid the short-term behavior. For instance, the supervision of the external creditors could prevent manager’s excessive investment behavior. The existence of manager market can make managers try to run a company to maintain his reputation. Because this would make managers not only face the fate of dismissal, but also affect the possibility of job and salary level when seeking for another job, if there is a downturn in the company's performance due to their own sake. Also, the external control mechanism is the last line of defense to constrain managers.

The first scholar mentioning the effect of control markets is Manne (1965). He regards control power as a kind of "expensive" assets, and supposes that there is no relationship among the value of the asset, economies of scale and monopoly profits. The market for corporate control is the role of equity market in facilitating corporate takeovers through buying a stake or proxy. There are three main ways of takeover: the proxy fight, acquisition and direct purchase shares. In an effective capital market, the company's performance will be reflected in the company's stock price, if there is poor management of the managers existing in an enterprise. Nevertheless, if the company's equity is more dispersed, and the potential acquirer investigates that the stock price is down because of the poor management of administrators which is not caused by the quality of the company, the potential acquirer will use low price to buy enough stocks to acquire the company, change managers after taking over the company, and reorganize enterprises to operate and obtain profits. However, especially in China, if a large portion of the company's equity is the current stock, most state-owned enterprises’ equity which is not state-state-owned shares and legal person share could not be circulated in the market. Even if acquirer purchases all outstanding shares, the control power still might not be changed. In other words, there is no market for corporate control in the external control mechanism for this enterprise. In this case, it is infeasible for acquirer to change the manager through the approach of taking over company.

(19)

15

Part4: Classification of the CEO turnover and the reason analysis

4.1 Classification of the CEO turnover

In the study of CEO turnover in the listed companies, one of the most important problem is the classification of CEO turnover, because there are many different reasons for CEO turnover. Previous literature usually divides the causes of CEO turnover into two major categories: regular change and mandatory change.

4.1.1 Regular change

Regular change is based on the personal reason which is not an external forcing replacement. Such as: retirement, healthy reason, normal personnel transfer or resign reasons. In a general way, this kind of change generally has nothing to do with the company's performance.

4.1.2 Mandatory change

Mandatory change is caused by external force and is not referred to the will of executive. In this case, this change is under the effect of the internal and external control mechanism in the board of directors, which is due to the company's poor performance.

China Securities Regulatory Commission requires that the listed companies must disclose the reasons for CEO turnover in the annual financial report. In general, the reasons disclosed in the financial report mainly include: the cause of illness or death, health reasons, expiration, retirement, transferred, resign, discharge and illegal reasons and so on. But in the actual situation, the information disclosure of CEO turnover is very fuzzy due to the various reasons in Chinese listed companies, some even do not disclose the reason. Even if company has revealed the reason, this reason is not necessarily the real reason for CEO turnover. Especially in the case of mandatory change, board may use reasons like: resign, healthy reason, and the expiration to hide the real reason behind the event for the sake of CEO’s face when the board decides to dismiss managers due to CEO’s mismanagement and poor firm performance. Therefore, it is very difficult to find the real reason based on the analysis of the information disclosure in the annual report.

(20)

16

4.2 The analysis of the reasons for CEO turnover

Because the reasons for CEO turnover disclosed in the annual report might not be the real reason. And CEO turnover is a part of the corporate governance mechanism and a result of the combined action of company’s internal and external governance mechanism. Hence, this paper would analyze the profound reasons behind the CEO turnover, including performance and corporate governance mechanism and so on.

4.2.1 Firm performance and CEO turnover

Most scholars have reached an agreement that inferior company performance could lead to the higher possibility of CEO turnover. In order to measure the efforts of the CEO’s level and ability, the board of company would use some accounting performance evaluation index for executive performance when hiring CEO to operate and manage enterprise. The company's operating performance is the most distinct symbol of the executive operating results, which embodies the CEO’s management level and the degree of efforts. Hence, the board will give CEO reward or punishment according to the company's operating performance. If the company's performance is very good, CEO will be awarded with many kinds of material or spiritual reward. And if the company's performance is very bad, CEO will also be punished correspondingly. Furthermore, if CEO abuses his power to seek personal gain, which causes a great loss to company, or the board thinks the CEO cannot be up to the management of company, the board will dismiss the CEO. This is the most extreme punishment and a kind of ultimate constraint for CEO. Therefore, in general, the better firm performance is, the smaller possibility of CEO turnover there is. Because it is unnecessary for board to fire an executive who can bring profits for the enterprise. Also, CEO turnover will disrupt the stability of the organization and personnel for company. Inversely, the worse business performance is, the greater likelihood of CEO turnover there is. Because the company needs a more capable manager to lead the company out of the woods. Hence, poor firm performance is the main and the most direct reason among the many factors for CEO turnover.

4.2.2 Ownership structure and CEO turnover

Ownership structure is defined by the distribution of equity concerning votes and capital and the identity of the equity owners. These structures are of major importance in corporate

(21)

17

governance because they determine the incentives of managers and the economic efficiency of the corporations. As a result, ownership structure would also affect CEO turnover. At present, the equity structure of listed companies in China generally exist the following two characteristics: One is the equity concentration. The second is the phenomenon of the single-large shareholder. Public ownership plays a dominant role in China, and most of the listed companies are from owned enterprise reform. This leads to higher proportion of state-owned shares in Chinese listed companies. In general, if the enterprise exists a dominant phenomenon, large shareholders have more incentive to replace the current executive when firm performance is poor. And no shareholders could compete with the first major shareholder when equity is relatively concentrated, then it is easier to fire executive. However, the body of the state-owned shares is country, and country does not directly enjoy the power of the shareholders due to the special nature of the state-owned enterprises in China. The control power is entrusted to a qualified agent with state-owned assets. As an agent of government officials, he does not enjoy the residual claim to assets. Hence, the agent lacks the motivation and enthusiasm of participating in the corporate governance. On the other hand, a lot of CEOs in state-owned enterprises are government officials, and their appointments are not based on the company's performance. Therefore, the CEO turnover is mostly due to job relocation in the state-owned enterprise, which is insensitive with firm performance. In addition to the nature of the control power, managerial ownership is an important factor of CEO turnover in the ownership structure, and is one of the incentive measures for CEO. Company could offer certain shares to CEOs and bring them into the company's ownership structure, which could connect the profitability of executive with the company's performance. This could reduce agency costs between shareholders and executives. Because the company's performance will be reflected in stock price, the price change will affect the executive's own income. Hence, executives will pay more effort for their own interests to run a company. Mikkelson and Parteh (1997), Denis (1997), and Tsai (2006) detect that when the proportion of managers’ shares is higher, it will reduce the possibility of CEO turnover. There are some reasons in the following several aspects: Firstly, because managerial ownership means that CEO becomes the company's shareholders, which objectively makes CEO and company share profits and risks. Hence, CEO will have a strong incentive to improve the company's operating performance, and the results would inevitably

(22)

18

reduce the possibility of CEO turnover. Secondly, managerial ownership enhances the sense of belonging and the master consciousness. When external markets have issued more generous terms to poach executives, the possibility of job-hopping will also be greatly reduced. Finally, if the proportion of manager’s shares is higher, the right to vote and voice will strengthen for them. CEO will resort to their equity to maintain their position and weaken the effectiveness of the company internal and external constraint mechanism. In this case, even if the company's performance is poor, it would be hard to replace CEO. Therefore, managerial ownership would reduce the possibility of CEO turnover.

4.2.3 External market for corporate control and CEO turnover

External market for corporate control is an important part of the external governance mechanism, and would have a significant impact on the CEO turnover in the listed companies. Since the first equity trading event in 1996, equity transfer has emerged in the Chinese listed company endlessly. Previous research has showed that the poorer performance, the greater the likelihood of CEO turnover in the presence of active control of external market. Because poor management of executive will be reflected in the company's stock price, while the external acquirers will use low price to buy the company's controlling stakes and replace executive. Therefore, in this case, the transfer of the control power is usually accompanied with CEO turnover. Nevertheless, the market for corporate control is not mature in China, and many listed company's shares are non-circulation shares. Even if acquirers buy all outstanding shares, it is still not enough to grasp control power of the company. Hence, the market for corporate control plays a very small role in China.

Part5: The impact analysis of CEO turnover on firm performance

We have analyzed the reasons for CEO turnover, and find out that the poor corporate performance is an important reason for CEO turnover. Most scholars also have reached an agreement on this point. CEO is the core and soul of a company, and plays a role of commander in chief in the company. Hence, CEO turnover will have a significant impact on corporate. In general, company’s CEO is not easily replaced regardless of special circumstances. And now that if the board replaces CEO for poor performance, this illustrates that the company wants to get rid of dilemma and improve the management. However,

(23)

19

whether new CEO can improve the company's operating performance after CEO turnover, and it is a problem that needed to be paid more attention.

5.1 Changes and adjustments in the internal organization

CEO turnover would bring out a major change in the organization strategy of the business. Because company has formed its own huge network, and the company's employees have been accustomed to his leadership style and formed a fixed pattern of behavior during the CEO’s job tenure. Once CEO turnover happens, it can disturb this stable relationship network and model, which will lead employees to adapt and be familiar with the new executive management style and increase the organization's operating costs. Certainly, if the new CEO starts to reform the company's organizational structure, abolish the backward enterprise system, and set up scientific management style, it will bring vitality to company. Although it will still increase the cost of organizational change, these actions still do more good than harm. Besides, CEO turnover can also affect the company staff's psychological state. As the leader of company, CEO has a very high prestige and influence on employees mind. Moreover, the company's senior management is usually a team, and team members will form a special kind of cohesive force after working together for a long period. Once the main leader of the team is replaced, the cohesion will be destroyed, which would produce harmful effects on the other team members’ loyalty and sense of belonging. Furthermore, it may lead other members of the managerial team to resign from the company after firing the leading executive. On the other hand, the new CEO is likely to eliminate managers who held dissatisfaction or negative attitudes about him in the company to avoid the implementation of the new policy, and cultivate and foster his new management team. Whether team personnel resign or are forced to leave, CEO turnover will affect the stability of the company management team and produce adverse impacts on future firm performance. Also, if the previous CEO has a high prestige in the staffs’ heart, this can also lead to the mood of discontent and low morale. But if previous CEO has poor management level and ability, and even abuses the power of the company, which would cause the decline in firm performance. In this case, the CEO turnover would become a symbol of reform, greatly inspired and elevate staffs’ spirits, and make employees restore feel significance for company's future.

(24)

20

5.2 The selection of successor

The selection of successor is an important problem for research on CEO turnover. The first important problem for company is to select a new executive to manage and operate, the selection of successor would have a direct impact on the business performance after CEO turnover. Therefore, how to choose a successor who could be qualified with management work is a major strategic issue faced by companies. In academia, research on the selection of successor is mainly based on the viewpoint of successor’s source. According to the source of successor, succession can be divided into external and internal selection. Internal succession refers that the successor is from company's internal staff, and external succession refers that the successor is outside and have never worked in company. In general, these two ways have their own advantages and disadvantages.

In the case of internal succession, the company is familiar with the candidate's ability and character, and can have its own destination to exercise and train candidates. This can effectively avoid the problem of information asymmetry. Secondly, internal succession is the promotion opportunity given by company, which can greatly inspire other employees’ enthusiasm and motivation. Furthermore, the internal successor is very familiar with operating process, and can quickly adapt to the new work environment. At the same time, internal successor can find company’s management loophole and insufficiency. Additionally, it is easier for internal successor to establish good relations with other employees and carry out company strategy. These would have very important significance for new CEO in the future work. Everything goes both ways. Company may not choose the most appropriate person to serve as executive due to the small range. And it can also bring the problem of "inbreeding".

In the case of external succession, on the one hand, external managers can solve the limitations of internal employee ability. On the other hand, it can also bring fresh blood for the company, improve enterprise’s innovation ability and energy, and avoid excessive competition behavior of internal employees. In addition, it could not only safeguard the staff’s cooperation spirit and reduce unnecessary losses, but also provide the necessary conditions for new managers to carry out management reform which could lead the company out of the woods. Certainly, it also has some shortcomings. While external succession will bring new

(25)

21

competitive power and advantage, it will also interrupt the company’s original operating strategy and internal structure, which would affect the stability of company and firm performance.

In a word, external and internal succession have their own advantages and disadvantages. In the actual situation, the company actually chooses the way of succession based on its demand. In general, when the company's operating performance is good or the board hopes to continue the company development strategy, the way of internal succession will be adapt frequently. And when the company's performance is very poor, it is an urgent need to reform and introduce external forces to change this situation and promote the development of the company. Even when the company is in rapid development, internal staffs do not have enough ability to be as a suitable candidate executive, the board will take external employment ways to select managers.

5.3 Earnings management

Earnings management in accounting meanings is the act of intentionally influencing the process of financial reporting to obtain some private gain. Earnings management involves the alteration of financial reports to mislead stakeholders about the organization's underlying performance, or to "influence contractual outcomes that depend on reported accounting numbers." The purpose of earnings management is to achieve the interests of the authorities, including the manager's private benefits or the interests of the board. In the modern company system, because there is information asymmetry existing the separation of the company's ownership and managerial, the company's managers would pursue their own interests and run against the interests of the company because of the risk of moral hazard and adverse selection. The company's operating performance is the direct reflection of CEO’s ability and level. In the present stage of executive incentive measures, it is very common to give the company stock dividend income as a part of the executive compensation or directly make CEO hold company's shares to encourage CEO. Hence, the CEO compensation is closely related to company's performance. In this case, whether the CEO is due to maintain their reputation, avoid dismissal or improve his salary pay, there will be a strong motivation to use earnings management to whitewash the company's performance and reach his own purpose. When

(26)

22

the company's performance is poor, CEO would be faced turnover, thus executive would use earnings management to whitewash company’s real performance and cover their mismanagement to maintain their position. Moreover, the new successor will also use means of earnings management to raise the company's performance, or lower former operating performance of the year to set up his good image. Especially when the CEO turnover is due to company's poor performance, and the new CEO is eager to improve the firm performance and lead the company out of the woods, he will use frequently earnings management to prove his ability level and prestige, and then further consolidate his position. Because of information asymmetry, it is difficult to find the behavior of earnings management. Therefore, in this case, the phenomenon of earnings management usually generates with CEO turnover. It could be said that CEO turnover is one of the most important factors that cause earnings management behavior. Nowadays, earnings management behavior caused by CEO turnover is very serious, and has raised concern in practice.

Part6: Data and measurement

The aim of this paper is mainly to investigate the effects of CEO turnover on the firm performance in the short-term and long-term. Due to the timespan of around 10 years, annual data will be obtained. This paper would compare the difference in ROA and ROE and set OLS model to find the linear relationship between CEO turnover and firm performance. Further, because the selection of successor is directly related to the firm performance after different types of CEO turnover. Hence, this paper also tries to analyze the effects of succession after different types of CEO turnover on the firm performance, which is the mainly reasons for the firm performance change.

6.1 Data

Base on all A-share listed companies as research samples from 2001 to 2010, this paper selects 1-year of observation time window and 3-year of observation time window to analyze the short-term and long-term effects of CEO turnover, hence the paper should use China Listed Company Database from 2000 to 2013.

(27)

23

The definition of A-Share: The official name of the A shares is RMB common stocks, which is issued by the company in mainland China. It is available for the domestic institutions, organizations or individuals (not including Taiwan, Hong Kong, Macao investors) subscription and trading in RMB. A shares are paperless electronic billing, which practice the “T+1”delivery system.

CSRC (China Securities Regulatory Commission) announces the industry classification standard. This offers us the industry standard and helps us to eliminate the industry fixed effects. Secondly, this paper collects CEO information (Reasons for CEO turnover) and Complexity of firms (industry, size, state, leverage) from the Taian CSMAR database. Additionally, Because ROA is an indicator of how profitable a company gets and gives an idea as to how efficient management is at using its assets to generate earnings. Return on equity (ROE) measures a corporation's profitability by revealing how much profit a company generates with the money shareholders have invested. I use ROA and ROE as the indicators of firm performance which are collected from RESSET database.

 Return on total assets.The indicators can reflect the relationship between the efficient use of assets and capital utilization effect.

ROA = net income average total assets⁄

average total assets = (initial total assets + final total assets) 2⁄ .

 Return on net assets. This indicator can reflect the shareholders' equity returns level and measure the efficiency of companies’ own capital.

ROE = net income net assets⁄ net assets = total assets − total liabilities

6.2 Methodology

6.2.1 Hypotheses

Most scholars have reached an agreement that inferior company performance is more likely to lead to the CEO turnover. Because the board of directors would use firm performance as an indicator to measure the efforts of the CEO’s level and ability when hiring CEO to operate and manage enterprise. Hence, I first check this hypothesis and set the Logit model:

(28)

24

H1: Poor firm performance would lead to the probability of the CEO turnover.

Logit(Turover) =𝛽₁ROA(ROE) + 𝛽₂𝑆𝑡𝑎𝑡𝑒 + 𝛽₃𝑆𝑖𝑧𝑒𝑡+ 𝛽₄𝐿𝑒𝑣𝑡+∑𝐼𝑛𝑑𝑢𝑠𝑡𝑟𝑦 +∑𝑌𝑒𝑎𝑟 + 𝜀 Secondly, most of the foreign scholars highly support that CEO turnover would produce positive impact on corporate performance of listed companies. Nevertheless, domestic scholars do not come to a conclusion on this issue. This paper argues that CEO turnover is a part of corporate governance mechanism, and reflects the determination to improve business performance. Hence, company would employ CEO with higher ability after CEO turnover, strengthen the supervision and constraint of executive, and enhance the level of corporate governance in the later business activities. Additionally, to prove their power and prestige, new CEO would pay more attention to manage enterprise and improve firm performance. Hence, this paper sets related assumption that CEO turnover would affect positively firm performance in the short-term and long-term. Thus I set the following assumption and OLS model:

H2: CEO turnover would produce positive effects on the accounting firm performance in the short-term and long-term.

∆𝑅𝑂𝐴(𝑅𝑂𝐸)𝑡 = 𝛽0+𝛽1∆𝑅𝑂𝐴(𝑅𝑂𝐸)0+ 𝛽2𝑇𝑢𝑟𝑛𝑜𝑣𝑒𝑟 + 𝛽3𝑆𝑡𝑎𝑡𝑒 + 𝛽4𝑇𝑢𝑟𝑛𝑜𝑣𝑒𝑟 × 𝑆𝑡𝑎𝑡𝑒

+ 𝛽5𝑆𝑖𝑧𝑒𝑡+ 𝛽6𝐿𝑒𝑣𝑡+ ∑ 𝐼𝑛𝑑𝑢𝑠𝑡𝑟𝑦 + ∑ 𝑌𝑒𝑎𝑟 + 𝜀

Lastly, this paper further analyzes the impacts of succession after regular change or mandatory change on the firm performance. This paper argues that the succession is the main reason for the improvement of firm performance after CEO turnover. Because most regular turnover is not caused by poor performance. It is unnecessary for company to change corporate governance mechanism and management mode. In the case of internal succession, the company is familiar with the candidate's ability and character, and could have its own destination to exercise and train candidates. Additionally, it is easier for internal successor to establish good relations with other employees and carry out company strategy. Hence this paper assumes that inside successor would be more suitable for regular CEO turnover. On the other hand, it could be assumed that external succession would be helpful for mandatory. Because in the case of external succession, external managers can solve the limitations of

(29)

25

internal employee ability and avoid excessive competition behavior of internal employees. Most mandatory change is caused by the poor performance, and board of directors would be eager to order new successor to improve firm performance by changing the company’s original operating strategy and internal structure. External successor could bring new blood to company and carry out reformation, which would be more suitable for mandatory change. In summary, this paper sets the third hypothesis based on the advantages and shortages of the two ways of successions:

H3: Internal succession would have the positive effects on firm performance after regular turnover, but would have negative effects after mandatory turnover in the short-term and long term.

∆𝑅𝑂𝐴(𝑅𝑂𝐸)𝑡= 𝛽0+𝛽1∆𝑅𝑂𝐴(𝑅𝑂𝐸)0+ 𝛽2𝐼𝑛𝑠𝑖𝑑𝑒 + 𝛽3𝑆𝑡𝑎𝑡𝑒 + 𝛽4𝐼𝑛𝑠𝑖𝑑𝑒 × 𝑆𝑡𝑎𝑡𝑒 + 𝛽5𝑆𝑖𝑧𝑒𝑡

+ 𝛽6𝐿𝑒𝑣𝑡+ ∑ 𝐼𝑛𝑑𝑢𝑠𝑡𝑟𝑦 + ∑ 𝑌𝑒𝑎𝑟 + 𝜀 6.2.2 The explanation of variables

Table I has been showed all the variables this paper would use. For the dependent variables, this paper would use the difference in the ∆ROA and ∆ROE as indicators, and define separately CEO turnover and way of succession (variable of inside) as independent variables for two OLS regressions to observe the effects. For the control variables, because there are lots of factors which are related to firm performance. For instance, different corporate structure, size, leverage, industry and whether it is state-owned or not may affect the result in different degree, hence this paper would select these control variables to ensure the accuracy of equation.

Part 7: Econometric Specifications and results

7.1 Descriptive statistical analysis

7.1.1 Industry distribution analysis of CEO turnover

In order to further analyze the current situation of CEO turnover, industry distribution analysis of CEO turnover would be described below. This paper refers to guideline issued by China Securities Regulatory Commission in 2012, which has classified industry into 19 categories: (1)

(30)

26

Agriculture, forestry, animal husbandry and fishery; (2) Extractive industry; (3) Manufacturing industry; (4) Electricity, gas and water production and supply industry; (5) Construction industry; (6) Wholesale and Retail Trade industry; (7) Transportation and warehousing industry;(8) Hotels and Catering Services industry;(9) Information technology industry;(10) Finance and Insurance industry;(11) Real estate industry;(12) Leasing and Business industry; (13) Scientific research and technical services industry;(14) Water, environment and public facilities management industry;(15) Residents service, repair and other services industry;(16) Education industry; (17) The department of health and social work industry; (18) Culture industry, sports and entertainment industry;(19) Comprehensive industry.

Specific industry distribution has been shown in Table II, we could see that the CEO turnover mainly occurs in the manufacturing industry in China, which makes up more than half. This is followed by Wholesale and Retail Trade industry, Real estate industry and Comprehensive industry, but they are all below 10%. The least amount of companies that CEO turnover happens is in the Education industry. There is no any events of CEO turnover during ten years. Because Education industry is the new industry divided by China Securities Regulatory Commission in 2012, and there are still a little amount number of education companies which are too young to be listed in China. Totally, there are approximate 1800 listed companies, and 3846 valid samples for events of CEO turnover happened from 2001 to 2010 in Chinese listed companies after dropping some missing data. Nevertheless, we could not conclude that the difference in industry is large because of the different numbers of companies in each industry. For example, the number of manufacturing companies in all listed companies is the largest base, thus CEO turnover would happened more frequently. The number of companies in Education industry is the least, and hence CEO turnover, of course, also will be less. Considering the number of companies in each industry is very different, the proportion of CEO turnover in each industry is equal distributed from the point of view of each industry.

7.1.2 The analysis of the reasons for CEO turnover

China Securities Regulatory Commission requires listed companies to disclose the information of CEO including name, term, and the reason for leaving. According to the ShenZhen taian CSMAR database, the reasons for CEO turnover mainly have the following categories: (1) Job

(31)

27

transfer: CEO is replaced because of job relocation; (2) Retirement: CEO arrives the retirement age and no longer continue to work; (3) Expiration: the term of expiration is arrived; (4) Controlling Changes in equity: CEO turnover refers to the control shift; (5) Resign: CEO resigns from the company; (6) Dismissal: company forces firing CEO; (7) Health reasons: CEO could not hold the position due to health reasons such as illness or death; (8) Personal reasons: CEO resigns from the post due to personal causes; (9) Corporate governance structure is improved: CEO turnover because of the adjustment in the corporate governance structure; (10) Illegal reason: CEO involves in the case and is fired by the company; (11) End of agent: CEO has this position due to the agent, and now it is at the end of the agent; (12)Other reasons.

This paper analyzes the distribution of the reasons for CEO turnover in China from 2001 to 2010, which has been shown in Table III. From Table III, we could see that the proportion of reason of job transfer (38.92%) is the highest in the disclosure for company’s CEO, which is followed by the reasons of dismissal and expiration, respectively 25.27% and 19.47%. Hence, job transfer, dismissal and expiration are the three main reasons for CEO turnover. The proportion of other reasons is relatively small and is below 10%. Some other companies do not announce CEO turnover reasons in the annual report disclosure, which accounts for 1.33% of the total samples. Additionally, it is clear that mandatory change happens more frequently in China and totally accounts for more than 50%. Surely, the reasons shown in the table are only the reasons disclosed in the annual report of listed companies, and it is unclear that whether they are the real reasons. In fact, many corporates will not disclose the real reason when CEO turnover happens. Usually, if there is a decline in the company's business performance, the company decides to fire CEO and often does not report the real reason for the sake of taking care of CEO’s face. Or when the company CEO turnover is due to some private reasons which are not convenient to disclose, companies will use other reasons to hide the facts. In a word, the reasons for CEO turnover in the report are not transparent. Hence, it is difficult to determine the real reason behind the change according to the disclosed information.

7.1.3 The distribution of CEO turnover based on the reasons and succession

The following Table IV is about the distribution of total samples based on the change reasons and the way of succession. And it shows that two main types of CEO changes and two ways

(32)

28

of succession after CEO turnover in different years. The information of CEO turnover and succession is collected from ShenZhen taian CAMAR database.

From Table IV, it reports that there are 815 samples of internal succession when CEO turnover is regular, which is higher than the number of external succession (562). Furthermore, the number of internal succession in the condition of mandatory changes which has reached to 1433 samples is still higher than the number of external succession from 2001 to 2010. Hence we could conclude that the internal succession happens more frequently in China when CEO turnover happens. This would be related to the reason for the different system in Chinese listed companies. Most Chinese companies would prefer internal successor who is more familiar with the institution and operational mode and do not want to take risk to select external successor even though there is a sharp decline in firm performance. Furthermore, most shareholders know that it is difficult to introduce new management system and change the old operating mode only by choosing a new external successor. Hence, unless they were dreadfully cornered, they may not choose external succession.

7.1.4 The analysis of firm performance comparison

In order to explore the comparison between companies with CEO turnover and companies with CEO unchanged and the difference in change, this paper compares the performance variables, as shown in Table V.

The values of ROA and ROE in the CEO changed companies are 0.0242016 and 0.0481369, and the values of ROA and ROE in the CEO unchanged companies are 0.388571 and 0.0729263 respectively. At the same time, T value shows that the firm performance of CEO unchanged companies is higher than CEO changed companies based on the 1% significance level. It could be proved that poor business performance can indeed cause CEO turnover. The means of the difference in ROA and ROE CEO in the changed companies are higher than the values in the unchanged companies. Additionally, T value could pass at 1% significance level. This suggests that the company's business performance would be improved after CEO turnover, and the improvement of performance is significantly better than CEO unchanged companies in the short-term and long-term.

Effects of CEO turnover on firm performance : a short-term and long-term analysis of listed companies in the Chinese market