The impact of comment letters on forced CEO turnover

Xue Chang

Research Article: 2022 Vol: 26 Issue: 5

The impact of comment letters on forced CEO turnover

Xue Chang, University of Nottingham Ningbo China

Citation Information: Chang X. (2022). The Impact of Comment Letters on Forced CEO Turnover. International Journal of Entrepreneurship, 26(S5), 1-25.

Abstract

Comment letters are an important tool for stock exchanges to supervise listed companies. This paper using LPM method to analyze the relationship between comment letters and CEO turnover from 2015 to 2020. The results indicate that: (1) there is a significant positive correlation between the current year’s comment letter and forced CEO turnover. (2) The cumulative number of comment letters received in the current year is significantly positive correlated with the forced CEO turnover. (3) We also find that the CEO in state-owned enterprises receive comment letters is less likely to be replaced than that of non-state-owned enterprises.

Keywords

Comment Letters, CEO Turnover, Reputation Theory, Sec Supervisions.

Institutional Background

According to the Sarbanes-Oxley Act of 2002 (SOX) Section 408, the SEC review companies’ filings at least once every three years to ensure the disclosure follow SEC and Generally Accepted Accounting Principles (GAAP) disclosure requirements (Ryans et al., 2015). If the company's report violates the provisions of GAAP (generally accepted accounting standards), or the disclosure is defective and needs further explanation, the SEC will issue a comment letter to the company (Bozanic et al., 2017; Myers et al., 2013). The "dialogue" between the SEC and the company may be repeated many times until the problem is solved. The review may end with the company make a restatement financial report if the SEC’s concerns remain unsolved (Myers et al., 2013).

The comment letter supervision mechanism is similar to United States and starts late in China. The Shanghai Stock Exchange (SSE) and Shenzhen Stock Exchange (SZSE) implemented the “Express Train of Information Disclosure” for all listed companies in 2014 and 2015 respectively. Comment letters have been issued frequently and have gradually become an important tool for SEC supervision since 2015. Table 1 show that there are 9668 comment letters between 2015-2020 and presents an upward trend. Comment letters mainly focus on financial reports, related party transactions, asset restructuring, media reports, provision of guarantees and other matters.

The Chinese comment letter system has three features. First, annual reports are checked every year, compared with at least once every three years in the United States. Second, though the SSE and the SZSE are not administrative departments, they have close relationships with the CSRC and operate under its supervision. The stock exchanges are usually regarded as the extensions of the CSRC, and the Chinese comment letter system is designed to provide substantive supervision, compared with SEC procedural supervision. Finally, the Chinese comment letter system is substantive supervision and may be accompanied by follow-up investigations and penalties.

Literature Review

Research on Comment Letters

Corporate governance mechanism mainly includes law based mandatory external enforcement mechanism (referred to as legal mechanism or reputation mechanism) and internal enforcement mechanism (corporate governance) (Macaulay, 1963). As an external mechanism, the comment letter plays an irreplaceable role in supervising of the company. In recent years, issuing comment letters has become an important way of SEC supervision (Kubic, 2017). Related research on comment letters has focused on the determinants and consequences.

Many factors affect comment letters. Companies have lower profitability, higher business complexity, audited by small firms or weak internal governance are more likely to receive the comment letters from the SEC (Myers et al., 2013). The comment letter can point out the defects in the company's disclosure or financial statements (Bozanic et al., 2017; Myers et al., 2013). In general, receiving a comment letter means negative information disclosure (Gietzmann & Pettinicchio, 2013; Poe et al., 2013; Rosati et al., 2019). Correia (2014) shows that companies with political connections are unlikely to restate and have lower economic penalties when receiving the comment letter. While Heese et al. (2017) find that companies with political relations with the government or regulatory are more easily to obtain comment letters. Johnston and Petacchi (2017) find that companies with restated history, unstable business environment, non "big four" audit firms’ clients and abnormal P / E ratio are more likely to receive comment letters. Cassell et al. (2019) shows that if the readability of response is low, the SEC will have a longer processing period and the company is more likely to restate the report. Do and Zhang (2018) find that female staff in SEC is generally tougher reviewers.

The consequences of comment letters can be summarized as decreasing information asymmetry and improving information quality, cutting down managers' opportunistic activities and affecting other stakeholders (Cassell et al., 2019).

The public enforcement of law suggest that public enforcement in capital markets is necessary (OECD, 2014), which could reduce information asymmetries and improve information environment and quality (Duro et al., 2018). Comment letters provide more information to market participants, enhance the disclosure of potential adverse prospects, which reduce the information asymmetry and improve the quality of report (Cunningham et al., 2020). Lawrence et al. (2010) find that the CL process can identify material misstatements. Bens et al. (2016) find that after receiving the comment letters from the SEC, enterprises enhance the information transparency and decrease the uncertainty on estimation.

Prior literature document many effects of authority scrutiny on firm behavior, for example, reduction of tax avoidance activity after company receives a comment letter (Bozanic et al., 2017; Ryans et al., 2015). Johnston and Petacchi (2017) find that the comment letters reduce the stock bid ask spread. Cui and Zhang (2019) discover that there is a negative correlation between comment letters and future stock price crash risk. Cunningham et al. (2020) find that there are significant negative correlation between comment letters and corporate earnings management. Gietzmann et al. (2015) demonstrate that the CFO will change with the increase of the frequency of comment letters.

The comment letters also has influence on the behaviors of other stakeholders. M. B. Gietzmann and Isidro (2013) discover the institutional investors are sensitive to the CL received by the company. Auditors perceive higher audit risks when clients receive comment letters and hence increases audit fees, and the high fee for audit will continue for following years on reevaluate the clients’ reputation and litigation risk (Ballestero & Schmidt, 2019). The comment letters can improve the efficiency forecast by reducing forecast errors. Bozanic et al. (2017) demonstrate that the number of analysts increased after receive the comment letter. Banks charge higher loan spreads to corporations receiving comment letters. Because banks have an in-depth understanding of the cash flow, liquidity and other financial information after receive the comment letters, and thus weakening their trust in these companies (Myers et al., 2013). Gietzmann and Isidro (2013) find that institutional investors will cut down their equity holdings when the company receives the comment letter. Comment letters also have the "spillover effect" (Brown et al., 2018). Comment letters not only impact the firm’s disclosure quality but also that of industry peers (Kubick & Lockhart, 2016).

Literature Review on CEO Turnover

CEO turnover is an important area in corporate governance. The existing literature studies the influencing factors of CEO turnover from the perspectives of company performance, nature of property rights, political connection, information transparency, accounting fraud and violations. There is a conflict of interest from separation of ownership and control between managers and shareholders, which lead agency costs. CEO turnover is an important method to punish managers. This paper studies CEO turnover from the perspective of listed companies being questioned by SEC. A listed company's receipt of the comment letters means that the company has problems in information disclosure or operation and investment matters. Investors, auditors, media, regulators and other stakeholders pay more attention to the company and reduce their trust, resulting in more severe regulatory penalties. CEO as scapegoats to improve the company's reputation. Gietzmann et al. (2015) find that there is significant positive relationship between comment letters and CFO turnovers. In China, SEC sent a comment letter to ZHANGZIDAO GROUP Company because of financial fraud and failure to disclose information in time activities in 2019. The CEO in ZHANGZIDAO GROUP has been fired finally.

As an enterprise leader, CEO plays an important role in decision-making. It is generally accepted that forcing the CEO to leave is sign well-functioning corporate governance, which shows the board's monitoring ability. CEO turnover can be used to evaluate the effectiveness of the board (Huson et al., 2001). Combs et al. (2007) show that boards dominated by independent directors are more likely to dismiss the CEO due to poor firm performance. The stock prices react positively to forced CEO turnover announcements (Huson et al., 2004; Huson et al., 2001).CEO turnover as a specialized research field has attracted more and more attention. At present, there are many scholars studying CEO turnover, but they are relatively scattered and have not formed a unified theory. CEO turnover is a significant event, which has an important impact on the stock price and market performance. We make a literature review on CEO turnover from 2000 to 2020. We find that there are three areas relate with CEO turnover, which is company performance, board structure, and corporate governance. Shen and Cannella (2002) believe that CEO turnover will have a long-term and short-term impact on the company's performance, capital structure and future strategy.

Influence Factors of CEO Turnover Corporate Governance

The separation of ownership and control leads to the principal-agent problem. The effective corporate governance can evaluate the CEO’s ability effectively, eliminate low-quality CEO and recommend properly CEO. The external economic environment can also affect the fate of a company's executives (He et al., 2016).

Corporate Performance

Much literature has analyzed the relationship between CEO turnover and corporate performance (Jenter & Kanaan, 2015). Kato and Long (2006) discover that the negative correlation between CEO turnover and corporate performance in state-owned enterprises is weaker than that in private enterprises. When the CEO is one of the controlling shareholders, he unlikely to be dismissed because of performance. Rachpradit et al. (2012) find that when the company is family controlled, the probability of CEO turnover is low. Bernard et al. (2016) find that corporate social performance (CSP) improved after the CEO turnover.

Board Structure

The main responsibility of the board of directors is hire, dismiss, supervise and compensate the manager (Rutherford & Lozano, 2018). Brickley (2003) discover that corporate performance is not a decisive factor on CEO dismissal. The characters of directors play a significant role in CEO turnover (Wiersema & Zhang, 2013). Flickinger et al. (2016) find that the number of external directors plays an important role in the CEO dismissal decision. The board of directors also should consider the public opinion and media pressure (Stein & Zhao, 2019).

Previous studies have divided CEO turnover into two categories: voluntary and forced (Denis et al., 1997; Kang & Shivdasani, 1995; Maury, 2006). It is difficult to identify forced turnover, as they are often not described in press (Weisbach, 1988). Adams et al. (2010) define the voluntary turnover as death, illness or merger and takeover.

Consequences of CEO Turnover

CEO turnover has different impact on market reactions (Aupperle et al., 1985; Dahya et al., 2002; Denis et al., 1997; Shen et al., 2010; Weisbach, 1988). Huson et al. (2004) discover that companies with CEO experience face higher abnormal stock returns and increase profitability significant performance improvement before and after the announcement date of executive change. Huson et al. (2004) find that firms with CEO turnover face significant performance improvement, which suggest that shareholders view the CEO turnover as a good new. CEO under job pressure or personal reputation tends to reduce the opportunistic behavior and further improves the quality of financial report (Desai et al., 2006). CEO replacement along with impaired career opportunity are the most severe form of personal punishment (Desai et al., 2006; Fahlenbrach et al., 2008; Hillman et al., 2009; Hribar & Jenkins, 2004; Karpoff et al., 2008; Kravet & Shevlin, 2009; Palmrose et al., 2004; Sila et al., 2016).

Hypothesis Development Restatement

Research on restatements has grown significantly in recent years (Miller et al., 2008). In rare cases, SEC requires a restatement of the reviewed filing (Hennes et al., 2014). Restatements generate a large number of media coverage and investigate (Desai et al., 2007; Dyck et al., 2010; Gertsen et al., 2006). Firstly, restatement damage the reputation of the companies and managers reputations as reliable financial information providers, and causes negative market reaction (Anderson & Yohn, 2002; Gomulya & Boeker, 2014; Hennes et al., 2011; Irani & Xu, 2011). Companies punished by the SEC for financial misrepresentation will lose 25.5% of their market value due to reputation loss (Anderson & Yohn, 2002; Dechow et al., 1996; Karpoff et al., 2008; Palmrose et al., 2004). Secondly, the restatement of the financial report suggests that the company has misconduct behavior, resulting in significant loss of shareholder (Akhigbe et al., 2007 ; Wu & Hart, 2002), and increased information risk (Palmrose et al., 2004 ; Wilson, 2008), increased the cost of capital due to lose credibility (Chakravarthy et al., 2014; Farber, 2004; Hribar & Jenkins, 2004; Palmrose et al., 2004 ), as well as the increased cost of monitoring and future regulatory scrutiny (Palmrose et al., 2004). Thirdly, financial misrepresentation breaches the trust of stakeholders (Arthaud Day et al., 2006) and reflects corporate governance and internal control problems, which may suggest the CEO’s inability and integrity issues (Devers et al., 2009; Harris & Bromiley, 2007; Palmrose et al., 2004). Which lead investors lost confidence on financial reporting (Austin & Wilson, 2013; Harris & Bromiley, 2007) suggest that some restatements reflect earning management behavior.

Reputation and Legitimacy

Legitimacy is important for organizations because it provides access to key resources (Pfeffer, 1985; Suchman, 1995; Tang, 2017). Stakeholders may keep away from organizations suffering from loss of legitimacy to avoid the risk of "negative infection" and take away the financial, social and intellectual capital (Suchman, 1995). Studies have suggested that the sources of reputation damage are financial misconduct and misrepresentation (Hazarika et al., 2012; Zavyalova et al., 2016). Extensive literatures demonstrated that reputation provide many advantages, including increased financial support (Stuart et al., 2016), raised market share (Podolny & Phillips, 1996), increased growth and better financial performance (Rhee & Haunschild, 2006), support the legitimacy (Deephouse & Carter, 2005; Rao et al., 2018). Reputation also enhances management credibility, improves the reporting quality, and reduces information asymmetries (Hutton et al., 2009; Rindova et al., 2005). Comment letters presents that SEC has questions to companies’ annual report, which will weaken the legitimacy to some extent (Aldrich & Fiol, 1994; Healy & Wahlen, 1999; Ruef & Scott, 1998). Myers et al. (2013) discover that the companies with low profitability, high complexity, and weak governance are positively associated with the receipt of comment letters and the cost of remediation. Besides, companies receive a comment letter ate likely to be restatement and higher remediation costs, which need additional time and effort to resolve questions (Johnston & Petacchi, 2017; Liu & Moffitt, 2016; Shelton et al., 2009).

CEO and Comment Letters

CEO should take responsibility on the integrity and reliability of financial reports and safeguard the legitimacy (Connelly et al., 2016). When organizations thrive, CEOs’ ability will be reward. And when companies falter, executives as scapegoats to "fix" the problem and show the company is willing to accept external requirements (Pfeffer, 1985). CEO face discredited by association with the restatement (Cannella et al., 2002). The Sarbanes Oxley Act (SOX) of 2002 requires CEO take responsibility for financial supervision, highlighting the key role of CEOs in preventing financial misconduct.

The most important decisions for the board are appointing, compensating and, firing the CEO to meet the expectations of stakeholders and Society (Clayton et al., 2005; Fama, 1980). Reputation is important to both the CEO and the company. CEO's financial manipulation not only damages the company's reputation, but also damages the interests of all shareholders. This opportunistic behavior led to the breakdown of the implicit agreement on the appointment of the CEO and lost shareholders’ trust, resulting in the replacement of the CEO.

When company faces poor performance or financial distress, boards tend to change the CEO to restore investors' confidence and recover the firm's reputation (Li et al., 2010). In contrast to internal investigations or blaming external third parties, replacing the CEO is a straightforward signal that firm has intention to deal with the problems (Cannella et al., 2009; Devers et al., 2009; Gomulya & Boeker, 2014). Company placing blame on the CEO as a measurement to reestablish its reputation and legitimacy (Devers et al., 2009; Gilson & Vetsuypens, 1993; Suchman, 1995; Warner et al., 1988). Desai et al. (2006) find that 59% of restated earnings firms tend to replace the CEO or president within two years. As Pollock and Barnett (2012) notes, keeping a distance with the past mistakes of management in illegal activities is a direct way to restore the company's reputation. Corporations experience CEO turnover suggests that corporate commit to significant change commitment to change although the cost is expensive (Austin & Wilson, 2013; Gomulya & Boeker, 2014). Boards appoint the successor CEO immediately means that the company hopes to recover its reputation as soon as possible (Ballestero & Schmidt, 2019; Ranft et al., 2006; Rhee & Haunschild, 2006; Zhang & Wiersema, 2009). CEO turnover accelerates the restoration of financial reporting credibility after restatement. Besides, if the CEO is dismissed, the SEC is unlikely to take further enforcement action and may even reduce penalties (Denis et al., 1997).

Media Effect

The media is known as the "fourth power" in addition to legislation, administration and justice. As an independent external regulator, the media has a significant impact on the company through the dissemination of information to public (Cahan et al., 2015; Farrell & Whidbee, 2002). They have ability to influence the public perception through recognition or condemnation of corporate behavior (Cordella & Yeyati, 2003). Positive media coverage increases company's legitimacy and reputation (El Ghoul et al., 2016). The media plays a supervisory role in corporate behavior to ensure that enterprises maintain good corporate governance (Dyck et al., 2010). When a company is violated legal requirements or failed to meet standards or expectations, the media will focus on problematic corporate (Westphal & Deephouse, 2011). Therefore, firm have to take corrective action such as replace the CEO (Farrell & Whidbee, 2002).

The company's reputation and the CEO's personal reputation are negatively affected by the SEC's comment letter. In order to restore reputation and punish the CEOs, listed companies tend to fire the CEO after receiving the comment letter. Therefore, we put forward the first hypothesis

H₁: The comment letters will increase the forced CEO turnover.

Regulatory in China

China Securities Regulatory Commission (CSRC) as the main securities authority in China. In addition, CSRC also control the two stock exchanges (Shanghai Stock Exchange and Shenzhen Stock Exchange). The functions of the CSRC are similar to those of the US Securities and Exchange Commission (SEC). However, unlike the US SEC, the CSRC is not independent authority. It is directly subordinate to the State Council (China's highest decision-making institute) and has no judicial power. Firms in China can be grouped into SOEs and non-SOEs by controlling shareholder type. SOEs are naturally connected to the government by ownership ties (Calomiris, 2010). Their managers, appointed by government owners, have incentives to maintain the relationship with government. Due to the differences between Chinese and American regulatory agency and economic structure, CSRC may have different enforcement actions between SOEs and Non-SOEs. Since the SOEs under the government control and monitor, they may enjoy more preferential treatments compared with non-SOEs.

Political connection with government will affect the enforcement action, which is called capture theory by 1970s by the late George Stigler. He finds that SEC decrease the likelihood of enforcement actions and the magnitude of penalties in SOEs (Weingast, 1984). Empirical evidence shows a negative relation between firm with political connections and enforcement actions (Correia, 2014; Yu & Yu, 2011). The regulatory capture theory holds that political connections with government help enterprises obtain benefits from regulators (Stigler, 1971). Companies with political ties with the congress and the government, they can mediate with the SEC on behalf of the company. The political relations with enterprises affect the possibility of SEC law enforcement decisions (Correia, 2014; Yu & Yu, 2011). For example, after the bankruptcy of Enron, senior politicians from were criticized for loose regulation because of their close relationship with the company.

Other paper discovers the positive correlation between political connections and comment letters (Chen et al., 2018). Compared with the previous literature that the negative correlation between corporate PC and law enforcement, Heese et al. (2017) discover a positive correlation between corporate PC and SEC CL review. They believe that SEC views political connections as a distinct risk indicator and characteristic of financial-reporting risk. They also consider other possibility fraud-prone firms underreport political connections which lead positive relationship between PC and CL. Yu et al. (2020) find that compared with non-state-owned enterprises; CEOs of state-owned enterprises are more care about reputation issues and are more likely to bear the cost of reputation.

In the process of studying the influencing factors and economic consequences of administrative punishment, corporate ownership as an important factor should be considered separately. Compared with non-state-owned enterprises, SOE are associated with government departments and corresponding regulatory agencies. The existence of these associations may lead to their lobbying to the regulatory authorities to avoid supervision and punishment. Under the same circumstances, SOE is less likely to be punished by authorities. Therefore, we put forward the second hypothesis:

H_2a: The relationship between comment letter and CEO turnover is more significant in non- SOEs than in SOEs. Or
H_2b: The relationship between comment letter and CEO turnover is more significant in SOEs than in non-SOEs.

Sample, Variables, And Descriptive Statistics

Data Sources and Sample Selection

This paper takes Chinese enterprises listed in the Shanghai Stock Exchange (SHSE) and the Shenzhen Stock Exchange (SZSE) from 2015 to 2020 as initial samples. The data of comment letters are from "regulatory information disclosure" column of the Stock Exchange Committee and is obtained by manual sorting. Others are from China Stock Market and Accounting Research database (CSMAR), except for forced CEO turnover data, which are manually screened from their resume. According to the following criteria, we exclude (1) firms flagged ST or *ST, (2) financial services firms, because of their industry uniqueness, and (3) we also exclude the companies which CEO tenure is less than one year and incomplete information on the key variables. To avoid outliers, we wins rise at the 1% level in both tails. In addition, we control for year and industry fixed effects. Finally, we obtain 19,609 firm-year observations.

Model

We examine whether comment letters affect the likelihood of forced CEO turnover using the linear probability model (LPM). LPM simplifies the interpretation of coefficients, although it may produce fitted values outside 0 to 1 range (Wooldridge, 2002). According to Chyz and Gaertner (2017), LPM is suit for our test. In the robustness tests, we use a logit model showed in Table 9.

Forced CEO turnoveri,+1 = a0 + β1CLi,t + βkControlssi,t + Ij + Tt + εi,t (1)
Forced CEO turnoveri,+1 = a0 + β2ACC_CLi,t + βkControlssi,t + Ij + Tt + εi,t (2 )

All of the above variables are explained in Table 3. Industry (I) and year (T) fixed effects (industry and year dummies) are included in order to remove the effect of macroeconomic and cross-sectional effects. If the coefficients of β₁ and β₂ is positive and significant, hypothesis 1 and 2a are supported, which shows the probability of forced CEO turnover increase with the comment letters.

Key Variables

CEO turnover:

The dummy variable TO_FORCE is equal to 1 for turnover event in year t, and 0 otherwise (Cao et al., 2017). TO_FORCE is measured at t+1 period, while all other variables as measured at t period (Cao et al., 2017).

There are 2,532 CEO turnover events during the sample period. In Table 1, the reasons for CEO turnover are provided by CSMAR database. Change of job is taking up highest percentage, which accounting for 33.45% of total turnover. The second one is the completion of acting duties, which represent 26.42%, and the third is personal reasons (14.97%). Only 0.79% falls in the dismissal category. We reclassify reasons on job changes, resignations, personal reasons, and reasons not given (Firth et al., 2006). Other turnover is classified as normal with one exception: if the CEO is less than 60 years old and stated reason is the retirement, we classify this turnover as forced.

Table 1 Reasons For Ceo Turnovers Presented In Csmar Database
Reasons	Freq.	Percent	Cum
Change of job	847	33.45%	33.45%
Completion of acting duties	669	26.42%	59.87%
Contract expiration	137	5.41%	65.28%
Corporate governance reform	62	2.45%	67.73%
Dismissal	112	4.42%	72.16%
Health	34	1.34%	73.50%
Legal disputes	3	0.12%	73.62%
No reason given	197	7.78%	81.40%
Personal reasons	379	14.97%	96.37%
Resignation	20	0.79%	97.16%
Retirement	72	2.84%	100.00%
Total	2532	100.00%	100.00%

We can track the destinations of departing CEOs through the resume information provided by CSMAR. For example, the reason of a job change can be divided into force and voluntary turnover. If a departing CEO subsequently holds a position that is better than previous, then we classified it as non-forced (Huson et al., 2004). Table 2 summarizes the reasons for CEOs forced and normal turnover and the corresponding frequency. By reexamining 1535 cases through a search for CEO resume, 863 cases are not forced. We can see those 728 cases remaining as chairman or vice chairman and 135 cases are promoted (123 CEOs promote as chairman or vice chairman; 12 CEOs became government officials). We classify the remaining 672 cases as forced turnover. These included 176 CEOs who accepted new positions ranked lower than CEO position and 496 cases without any traceable destination information. In conclusion, 1827 normal turnover events, which accounted for 72.16% of the total and 705 cases as forced turnover (27.84%). Because we are investigating the impaction of risks on forced CEO turnover, we ignore voluntary turnovers (Atreya Chakraborty, 2007).

Table 2 Classify Forced And Voluntary Ceo Turnover
Reasons for turnover	Number of observations	Frequency (%)
1.Normal turnover	1827	72.16%
Retirement	59	2.33%
Contract expiration	137	5.41%
Health	34	1.34%
Corporate governance reform	62	2.45%
Legal disputes	3	0.12%
Completion of acting duties	669	26.42%
Remaining as board chairman or vice chairman	728	28.75%
Important government position	12	0.47%
Promoted to board chairman or vice chairman	123	4.86%
2. Forced turnover	705	27.84%
New position ranked lower than CEO position	176	6.95%
Retirement age less than 60	13	0.51%
Dismissed	20	0.79%
Information unavailable	496	19.59%
Total number of observations	2532	100.00%

Comment Letters Variables

Comment letter equals 1 if the firm received a SEC CL in year t, 0 otherwise. ACC_CL means number of CLs received up to year t, 0 otherwise.

Control variables:

We add the set of control variables included in the model used by Chyz and Gaertner (2017), which captures variables shown in the prior accounting research to influence executive turnover (Cao et al., 2017; Chyz & Gaertner, 2017; Deng et al., 2019; Guo & Masulis, 2015; Tran et al., 2016; Zhang, 2016). We control the following variables in the model: SIGMA, RET, MB, LEV, ROA, SIZE, DTURN, TENURE, STATE, FIRMAGE, DUAL (Myers et al., 2013), MA_SCORE. In addition, we also control in the industry and year dummy variables. Details of variables definition showed in Table 3.

Table 3 Variable Definitions
Variables	Notation	Definition
Forced CEO turnover	TO_FORCE	TO_FORCE is a dummy variable that equals 1 when a CEO has forced turnover and 0 otherwise.
Comment Letters Variables	CurrentCL	Comment letter equals 1 if the firm received a CL in year t, 0 otherwise.
Comment Letters Variables	AccumulateCL	ACC_CL means number of CLs received up to year t, 0 otherwise
Control variables	Size	The natural logarithm of total assets (VO & PHAN, 2013).
	Lev	Total liability scaled by total assets.
	ROA	Pretax income divided by total assets.
	BM	The book-to-market ratio equals firm's book value of equity divided by market value of equity.
	FirmAge	The natural logarithm of one plus years since incorporation.
	Tenure	The number of years in a CEO position.
	STATE	STATE is a dummy variable that equals 1 if the firm is a state-owned enterprise, and 0 otherwise.
	Board	The number of directors is taken as the natural logarithm.
	Dual	Coded “1” if Chairman also holds the position of CEO and “0” otherwise.
	Growth	Growth rate of operating revenue equals the Current year's operating income divided by previous year's operating income – 1.
	Tobin’s Q	We measure firm value using Tobin’s q, defined as the book value of assets minus the book value of equity plus the market value of equity, divided by the book value of assets.

Empirical Results

Descriptive Statistics

Table 4 presents the descriptive results, including mean, standard deviation, minimum, median and maximum dimensions. The mean and standard deviation of the TO_FORCE is 0.052 and 0.222. The mean of Current CL and Accumulate CL is 0.431 and 0.861.

Table 4 Descriptive Statistics
Variable	N	Mean	SD	Min	p50	Max
Board	14000	2.116	0.2	1.099	2.197	2.996
Growth	14000	0.455	16.425	-1.309	0.109	1878.372
Tobin’s Q	14000	2.209	2.425	0.694	1.671	122.189
Dual	14000	0.287	0.452	0	0	1
FirmAge	14000	2.929	0.297	1.609	2.944	3.97
BM	14000	1.027	1.306	0.011	0.63	20.965
ROA	14000	0.04	0.072	-1.859	0.038	0.675
Lev	14000	0.42	0.199	0.008	0.411	1.434
Size	14000	22.307	1.304	17.641	22.146	28.505
TO_FORCE	14000	0.052	0.222	0	0	1
CurrentCL	14000	0.247	0.431	0	0	1
AccumulateCL	14000	0.375	0.861	0	0	22

This table reports descriptive statistics for the full sample. Details of variables definition showed in Table 3.

Correlations

Table 5 presents Pearson correlation. Panel C shows the correlation results of the main variables. Consistent with H1, we find comment letters (both CurrnetCL and AccumulateCL) are positive and significantly correlated with forced CEO turnover. It is suggested that companies received comment letters are more likely lead CEOs leave.

Table 5 Correlations
Variables	Board	Growth	Tobin’s Q	Dual	FirmAge	BM	ROA	Lev	Size	TO FORCE	CurrentCL	AccumulateCL
Board	1
Growth	0.023***	1
Tobin’s Q	-0.101***	-0.006	1
Dual	-0.178***	0.015*	0.056***	1
FirmAge	0.101***	0.006	-0.035***	-0.116***	1
BM	0.150***	0.020**	-0.266***	-0.109***	0.199***	1
ROA	0.012	0.004	0.062***	0.027***	-0.039***	-0.156***	1
Lev	0.146***	0.027***	-0.168***	-0.113***	0.154***	0.533***	-0.298***	1
Size	0.271***	0.038***	-0.351***	-0.174***	0.136***	0.627***	0.008	0.523***	1
TO FORCE	0.023***	0	0.01	-0.077***	0.030***	0.047***	-0.099***	0.080***	0.043***	1
CurrentCL	-0.079***	0.005	0.030***	0.040***	0.026***	-0.012	-0.157***	0.064***	-0.065***	0.074***	1
AccumulateCL	-0.084***	0.005	0.027***	0.033***	0.031***	-0.016*	-0.202***	0.065***	-0.069***	0.085***	0.759***	1

This table reports Pearson correlation for the full sample, with p-values reported below each correlation. Details of variables definition showed in Table 3.

Regression Results

Table 6 column (1) and column (2) show the positive coefficient 0.0250 (0.0118) and significant at 1% between TO_FORCE and comment letters (CurrentCL & AccumulateCL), which support hypothesis 1. Comment letters will significantly increase the probability of CEO turnover. The more comment letters received and the more serious problems which company face, the greater the "cost" brought to the company and the higher the probability of CEO turnover.

Table 6 The Comment Letters Affect Forced Ceo Turnover
	(1)	(2)	(3)	(4)	(5)	(6)
	TO_FORCE	TO_FORCE	TO_FORCE	TO_FORCE	TO_FORCE	TO_FORCE
Board	-0.0141	-0.0137	-0.0054	-0.0058	-0.0142	-0.0139
	(-0.4420)	(-0.4317)	(-0.0770)	(-0.0834)	(-0.4345)	(-0.4239)
Growth	0	0	-0.0044	-0.0044	0	0
	(-0.7236)	(-0.7068)	(-1.3949)	(-1.3859)	(-0.4541)	(-0.4475)
Tobin’s Q	0.0015	0.0015	-0.0049	-0.005	0.0031*	0.0032*
	-0.8981	-0.9263	(-1.0397)	(-1.0502)	-1.7021	-1.7178
Dual	-0.0451***	-0.0450***	-0.0492**	-0.0494**	-0.0421***	-0.0419***
	(-4.2255)	(-4.2045)	(-2.3228)	(-2.3332)	(-3.3545)	(-3.3316)
FirmAge	0.1165	0.1094	0.1783	0.1752	0.0806	0.0753
	-1.4096	-1.3247	-0.9245	-0.9059	-0.9036	-0.8482
BM	-0.0049	-0.0047	0.002	0.0021	-0.0083	-0.0082
	(-1.1715)	(-1.1292)	-0.3114	-0.3346	(-1.4750)	(-1.4545)
ROA	-0.1794***	-0.1713***	-0.3015*	-0.3040*	-0.1495***	-0.1424***
	(-3.5587)	(-3.4065)	(-1.6613)	(-1.6704)	(-2.8695)	(-2.7394)
Lev	0.0441	0.0441	0.1429	0.1475	0.0204	0.0197
	-1.2289	-1.223	-1.4123	-1.4548	-0.5578	-0.533
Size	-0.0189	-0.0193	-0.0578*	-0.0611**	-0.0145	-0.014
	(-1.5350)	(-1.5612)	(-1.9068)	(-2.0110)	(-1.1441)	(-1.0999)
CurrentCL	0.0250***		0.0296*		0.0196***
	-4.146		-1.9547		-3.2074
ACCUMULATE_CL		0.0118***		0.0109		0.0101**
		-2.9683		-0.8525		-2.4381
_cons	-0.1858	-0.1377	0.8356	0.9153	-0.2031	-0.1818
	(-0.5467)	(-0.4059)	-0.9851	-1.0733	(-0.5576)	(-0.5005)
Industry	Yes	Yes	Yes	Yes	Yes	Yes
Year	Yes	Yes	Yes	Yes	Yes	Yes
N	13245	13245	4222	4222	9023	9023
adj. R²	0.012	0.011	0.016	0.015	0.015	0.015
t statistics in parentheses * p < 0.1, p < 0.05, * p < 0.01

In column (3) and (4) present the result in SOEs and in column (5) and (6) show the relationship between forced CEO turnover and comment letters in non-SOEs. We can see that comment letters have a more significant and positive relationship with forced CEO turnover in non-SOEs than SOEs, which is support the hypothesis 2a.

R-squared realizations are low in Table 6, which suggest limited explanatory power. However, according to Brickley (2003), low R is very common in the CEO turnover literature.

Table 7 Cl Changers After Ceo Turnover
	(1)	(2)	(3)	(4)	(5)	(6)
	CurrentCL	AccumulateCL	CurrentCL	AccumulateCL	CurrentCL	AccumulateCL
TO_FORCEt−1	-0.0187	-0.1099**	-0.0029	-0.0267	-0.0542*	-0.2397***
	(-0.9787)	(-2.4874)	(-0.1201)	(-0.7651)	(-1.7725)	(-2.7019)
Board	-0.0647	-0.1725	-0.1381*	-0.3107**	-0.0465	-0.0727
	(-1.2781)	(-1.3424)	(-1.7150)	(-2.2954)	(-0.7199)	(-0.4026)
Growth	0.0054***	0.0141***	0.0117**	0.0219**	0.0044**	0.0111*
	-2.759	-2.6357	-2.2842	-2.2744	-1.963	-1.8951
Tobin’s Q	0.0060**	0.0138	0.0140**	0.0284**	0.004	0.0099
	-2.0296	-1.3273	-1.9706	-2.1356	-1.2183	-0.7863
Dual	-0.0204	-0.0627	-0.0567	-0.0806	-0.0129	-0.0641
	(-1.1587)	(-1.4327)	(-1.5986)	(-1.5646)	(-0.6402)	(-1.1570)
FirmAge	0.8565***	2.5979***	-0.4437	-0.1782	0.9505***	2.4885***
	-4.2624	-5.5222	(-1.2291)	(-0.2994)	-3.8944	-4.0007
BM	0.0115*	0.019	0.0224**	0.0268**	0.0065	0.0249
	-1.7387	-1.2266	-2.3976	-1.9672	-0.6523	-0.8344
ROA	0.1102	-0.0489	0.1912	0.4731	0.1316*	0.0665
	-1.5405	(-0.2206)	-0.8047	-1.2116	-1.7202	-0.2709
Lev	0.4481***	1.4647***	0.4814***	0.7751***	0.4095***	1.5597***
	-7.3029	-8.2055	-3.5869	-3.4497	-6.0445	-7.2127
Size	-0.2022***	-0.5727***	-0.2020***	-0.2667***	-0.2096***	-0.6994***
	(-8.7583)	(-9.8324)	(-3.8300)	(-3.2465)	(-7.9869)	(-9.8565)
_cons	2.3376***	5.6081***	6.2383***	7.1262***	2.2523**	8.5393***
	-3.0916	-3.1487	-3.8805	-3.0001	-2.5221	-3.7835
Industry	Yes	Yes	Yes	Yes	Yes	Yes
Year	Yes	Yes	Yes	Yes	Yes	Yes
N	13390	13390	4256	4256	9134	9134
adj. R²	0.032	0.077	0.023	0.028	0.044	0.099
t statistics in parentheses * p < 0.1, p < 0.05, * p < 0.01

This table reports results for our main tests examining the comment letters on forced CEO turnover. Regression coefficients are reported above, while standard errors are reported below.

*, **, and *** denote statistical significance at the 10%, 5%, and 1% levels respectively for examining H1 & H2. This table uses OLS (Linear Probability Model) with robust standard errors controlling for firm-level clustering. Details of variables definition showed in Table 3.

Robustness

In this section, we undertake a series of additional tests to ensure the robustness of the results in Table 7, Table 8 and Table 9.

Table 8 Logit Model For Robustness Tests
	(1)	(2)	(3)	(4)	(5)	(6)
	TO_FORCE	TO_FORCE	TO_FORCE	TO_FORCE	TO_FORCE	TO_FORCE
Board	-0.1145	-0.13	0.1197	0.1233	-0.3353	-0.3786
Board	(-0.2488)	(-0.2834)	-0.1772	-0.1834	(-0.4872)	(-0.5525)
Growth	0.0003	0.0006	-0.0275	-0.0252	0.0063	0.0067
Growth	-0.0222	-0.043	(-0.6365)	(-0.5912)	-0.4417	-0.4473
Tobin’s Q	0.0201	0.0215	-0.0509	-0.0514	0.0907*	0.0947*
Tobin’s Q	-0.6585	-0.7008	(-0.8582)	(-0.8615)	-1.7918	-1.8555
Dual	-0.7083***	-0.7041***	-0.6876**	-0.7012**	-0.7217***	-0.7022***
Dual	(-4.1617)	(-4.1561)	(-2.1226)	(-2.1674)	(-3.4803)	(-3.4100)
FirmAge	2.3798	2.0627	2.8835	2.7534	1.269	1.0387
FirmAge	-1.1388	-0.9807	-0.7863	-0.7523	-0.4429	-0.3609
BM	-0.0277	-0.031	0.0768	0.0739	-0.1658	-0.1816
BM	(-0.3982)	(-0.4431)	-0.8723	-0.8442	(-1.1906)	(-1.3023)
ROA	-2.3042***	-2.2570***	-2.7760*	-2.9664*	-2.0796**	-1.9639**
ROA	(-3.1114)	(-3.0411)	(-1.6960)	(-1.8130)	(-2.4133)	(-2.2849)
Lev	0.7119	0.7485	0.8429	0.8895	0.4919	0.5561
Lev	-1.1024	-1.1582	-0.7076	-0.7453	-0.6047	-0.6838
Size	-0.3006*	-0.3239*	-0.5488*	-0.5912**	-0.1238	-0.1357
Size	(-1.7536)	(-1.8898)	(-1.8236)	(-1.9678)	(-0.5342)	(-0.5837)
CurrentCL	0.4845***		0.3458*		0.5297***
CurrentCL	-4.1236		-1.8836		-3.2445
ACCUMULATE_CL		0.1572***		0.098		0.1443**
ACCUMULATE_CL		-2.9911		-0.826		-2.3725
Year	Yes	Yes	Yes	Yes	Yes	Yes

Logit Model

We also test the relationship between comment letters and forced CEO turnover using logit model for robustness in Table 8, which is consistent with LPM.

Changes in Comment Letters Following Forced Turnover

According to Chyz and Gartner (2017), we also conduct analysis on the impact of forced CEO turnover on future comment letters in Table 7. If comment letters played an important role in the CEOs firing, we expect these trends to be reversed after turnover. Table 7 shows the results, a significant (5%) and negative correlation between forced CEO turnover and future comment letters, which suggest that these firms after experience CEO turnover appear to decrease comment letters.

Propensity Score Matching

Using OLS method cannot effectively solve the endogenous problem and sample selection error. In this paper, we use the propensity score matching (PSM) model to construct the non-issued CLs (control group) whose main characteristics are as similar as the issued CLs (treatment group). The steps are as follows: take CLs as the explained variable, the control variable, industry dummy variable, year dummy variable, and calculate the propensity score.

According to the propensity score value, the pairing is carried out according to the principle of nearest neighbor matching. The matched samples were analyzed by multiple regressions. The results are listed in the first and second columns of Table 9. It shows that there is no significant difference between the treatment group and the control group in matching propensity scores showed in figure 1, 2 & 3. In the matched sample regression, the estimated coefficient of CLs is still significantly positive at the level of 1%, and H₁ is verified.

Table 9 Psm And Tobit
		(1)	(2)	(3)	(4)
		TO_FORCE	TO_FORCE	TO_FORCE	TO_FORCE
CurrentCL		0.0368***		0.6953***
		-5.1178		-8.5581
ACCUMULATE_CL			0.0225***		0.3051***
			-7.2488		-9.0637
Industry		Yes	Yes	Yes	Yes
Year		Yes	Yes	Yes	Yes
N	5134		5134	13582	13582
adj. R²	0.012		0.017	0.012	0.017
F	3.4071		4.4733	3.4071	4.4733
t statistics in parentheses * p < 0.1, p < 0.05, * p < 0.01

International-Entrepreneurship-Propensity

Figure 1: Matching Propensity Scores.

Figure 2: Before Match.

Figure 3:After Match.

Tobit Model

Most CEO mandatory turnover is concentrated at 0. The sample in this paper can be applied to the truncated regression model (Tobit model). The results in the third and fourth columns of Table 9 show that the estimated coefficients of CLs and ACCUMULATECLs are significantly positive at the level of 1%, which proves that the result of H1 is robust.

Mediation Analysis

We exams the positive correlation between forced CEO turnover and comment letters. We will test the mediation analysis between them. In this paper, the mediation variables of the three mechanism paths are financial restatement (RESTATE), financial report transparency (TRANS) and earning management have influence on the comment letter to affect the CEO turnover. Since the forced CEO turnover is small and subject to some extent, we use CEO turnover as the dependent variable (Haider et al., 2020; Tofighi & MacKinnon, 2011; Uğural et al., 2020).

CEO Turnover, Statement Restatement and Comment Letters

As we analyzed earlier, if the company has serious problems and the response cannot reasonably dispel the SEC's questions, the company will be required to restate (Land, 2010). Restatement equals 1 if the SEC requires the company to restate, and 0 otherwise (Miller et al., 2008).

This paper constructs the following three models to verify this effect. Model (3) is used to test the impact of comment letters on CEO turnover. If coefficient α₁ significant, then model (4) can be used to test the influence of comment letters on mediation variable (RESTATE). The model (5) includes comment letters and intermediary variables (RESTATE) for analysis. If the coefficient μ₂is significant and μ₁ is not significant, it is a complete mediation analysis; If coefficient μ₂ and coefficient μ₁ are significant, it is a partial mediating effect; If coefficient μ₂ is not significant, the mediation analysis is not established.

Turnoveri, = α0 + α1 AccumulateCLi,t + α ControlSi,t + Σ Industryfe + Σ Yearfe + εi …… (3)
Restatei, = γ0 + γ1 AccumulateCLi,t + γ ControlSi,t + Σ Industryfe + Σ Yearfe + εi,t …..(4)
Turnoveri, = μ0 + μ1 AccumulateCLi,t + μ 2Restatei,t + μ ControlSi,t + Σ Industryfe + Σ Yearfe + εi, …….(5)

The results show in Table 9. In column (1), the coefficient of the comment letters is 0.0272, which is significantly positive at the level of 1%. Column (2) examines the impact of the comment letters on the mediation variable (RESTATE). The coefficient of the comment letter is about 0.0358 and is significant positive at the level of 1%, which indicates that company receive the comment letter are is easier to carry out financial restatement; In column (3), the comment letters and mediation variable (RESTATE) are included at the same time. The estimation coefficients of comment letters and financial restatement are significantly positive at the level of1%, and the coefficients of comment letters have decreased. Sobel test is also carried out in this paper. The statistic of Z is 2.689, which is significant at the level of 1%, This shows that financial restatement plays a partial mediation role between the comment letter and CEO turnover. The results of support the path of "comment letters financial restatement CEO turnover".

Comment Letters, Statement Transparency and CEO Turnover

As explained above, the comment letter issued by SEC can improve the information transparency and then affect the CEO turnover. We measure the transparency of the company (TRANS) from four dimensions: accrued earnings management (DD), information disclosure evaluation index (DSCORE), number of analysts (ANALYST), analyst earnings forecast (ACCURACY) and auditors (BIG4). With reference to Lang et al. (2012), we construct a transparent comprehensive index. DSCOR measure the information disclosure index from 1 to 4 points, which indicates the greater the score, the better information disclosure quality. ANALYST is the number of analysts covering a company. ACCURACY refers to the accuracy of analysts' earnings forecasts. BIG4 measure a firm’s auditor, which is audited by an affiliate of a global accounting firm.

We combine these five variables to create an aggregate measure, TRANS, by ranking each variable, summing the percentile ranks, and taking the average. If ACCURACY is not available, the remaining four variables (DD, BIG4, ANALYST and DSCORE) will be captured. This aggregate variable captures the combined effect.

The mediation effect test procedure is similar to the above and will not be repeated here. The regression results are shown in Table 10. In column (1), the estimated coefficient of the comment letter is about 0.0272, which is significant at the level of 1%. The estimated coefficient of the comment letter in column (2) is -0.0209, which is significantly negative at the level of 1%. This indicates that the comment letter can improve the company’s transparency. The estimated coefficient of the TRANS in column (3) is significantly negative at the level of 1%. The estimated coefficient of the comment letter is significantly positive at the level of 1% and the estimated coefficient has decreased. Sobel test shows the Z statistic is equal to 4.278 and significantly at the 1% level, which suggests that the TRANS is a partial mediation effect between the comment letters and CEO turnover. The results of support the way of "comment letters - improve transparency - CEO turnover" (Duro et al., 2018).

Table 10 Restate
	(1)	(2)	(3)
	Turnover	Restate	Turnover
ACCUMULATE_CL	0.0272***	0.0358***	0.0264***
	-6.5015	-7.9172	-6.2914
RESTATE			0.0230***
			-2.8593
Industry	Yes	Yes	Yes
Year	Yes	Yes	Yes
N	13245	13245	13245
adj. R²	0.022	0.272	0.022
F	9.7496	146.4824	9.7098
t statistics in parentheses * p < 0.1, p < 0.05, * p < 0.01

Comment Letter, Earnings Management and CEO Turnover

Many literatures discover comment letters are often sent to companies that adopt earnings management and further affect CEO turnover (Cassell et al., 2019). We use an aggregate REM measure (ABSREM) (Huang & Sun, 2017), which is the sum of abnormal operating activities cash flow, abnormal expenses and abnormal product cost. The higher the value of the aggregate measure, the more likely the firm engages in REM activities (Cohen et al., 2008).

The results are shown in Table 11. In column (1), the estimated coefficient of the comment letter is about 0.0272, which is significant at the level of 1%; the estimated coefficient of the comment letter in column (2) is significantly positive at the level of 1%, indicating that the comment letter finds the company has conducted earnings management.

Table 11 Trans
	(1)	(2)	(3)
	TURNOVER	TRANS	TURNOVER
ACCUMULATE_CL	0.0272***	-0.0209***	0.0250***
	-6.5015	(-13.7005)	-5.9275
TRANS			-0.1077***
			(-4.5036)
Industry	Yes	Yes	Yes
Year	Yes	Yes	Yes
N	13245	13245	13245
adj. R²	0.022	0.385	0.023
F	9.7496	244.8275	10.0643
t statistics in parentheses * p < 0.1, p < 0.05, * p < 0.01

The coefficient of earnings management in column (3) is significantly positive at the level of 1%. The estimated coefficient of the comment letters is significantly positive at the level of 1% and the estimated coefficient of the comment letters has decreased. The Sobel test shows the Z statistic is equal to 3.059 and significant at the 1% level, which suggests that earnings management is a partial mediation effect between the comment letter and CEO turnover. The results support the path of “comment letters-earning management-CEO turnover".

Table 12 Earning Management
	TURNOVER	ABSREM	TURNOVER
ACCUMULATE_CL	0.0272***	0.0050***	0.0253***
	-6.5015	-5.7029	-5.6065
ABSREM			0.1711***
			-3.6249
Industry	Yes	Yes	Yes
Year	Yes	Yes	Yes
N	13245	13245	13245
adj. R²	0.022	0.076	0.02
F	9.7496	32.1161	8.3168
t statistics in parentheses * p < 0.1, p < 0.05, * p < 0.01

Contribution

This paper studies the consequences of comment letters from the perspective of corporate governance, and find a significant positive correlation between comment letters and forced CEO turnover. This paper extending the research on comment letters to non-U.S. economies, this paper complements the literature on the consequences of CLs (Bozanic et al., 2017; Johnston & Petacchi, 2017; Myers et al., 2013; Ryans et al., 2015). First, we contribute to the literature investigating the determinants (Cassell et al., 2019; Myers et al., 2013) and consequences (Johnston & Petacchi, 2017) of SEC comment letters. Second, this paper has significant practical significance, which contributes to the literature on SEC scrutiny. Third, from the perspective of supervision, the existing literature studies the impact of CEO turnover from financial fraud, restatement and performance, but there is no literature on the impact of comment letter supervision on CEO turnover. Besides, the existing literature mainly takes American companies as samples, and there is little research evidence in China as an emerging market country.

Conclusions

This paper analyzes the influence of the comment letters supervision on forced CEO turnover. The empirical results show that the comment letters have a positive correlation with CEO turnover. The relation of comment letters on forced CEO turnover has different influence between SOEs and Non-SOEs. This paper has two suggestions: the supervision of comment letters is effective. In addition, listed companies should improve the level of corporate governance, so as to improve the supervision efficiency.

References

Adams, R.B., Hermalin, B.E., & Weisbach, M.S. (2010). The Role of Boards of Directors in Corporate Governance: A Conceptual Framework and Survey. Journal of Economic Literature, 48(1), 58-107.

Google Scholar, Cross Ref

Akhigbe, A., Kudla, R.J., & Madura, J. (2007). Why are some corporate earnings restatements more damaging? Applied Financial Economics, 15(5), 327-336.

International Journal of Entrepreneurship (Print ISSN: 1099-9264; Online ISSN: 1939-4675)

The impact of comment letters on forced CEO turnover

Abstract

Keywords

Institutional Background

Literature Review

Literature Review on CEO Turnover

Hypothesis Development Restatement

Sample, Variables, And Descriptive Statistics

Empirical Results

Conclusions

References