CEO characteristics and bank performance: Case of Vietnamese commercial banks

Abstract

The article was conducted to evaluate the impact of CEO characteristics on the performance of Vietnamese commercial banks. Variables representing the performance of commercial banks are return on total assets (ROA) and return on equity (ROE). Data was collected from 30 Vietnamese commercial banks from 2012–2020. By the regression methods of Pooled OLS, Fixed Effects Model (FE), Random Effects Model (RE), Feasible Generalized Least Squares (FGLS), the research results show that CEO duality and CEO age positively impact the performance of Vietnamese commercial banks. In contrast, the female CEO factor has a negative impact on ROE. In addition, the study did not find evidence of the impact of the CEO’s financial background and the CEO tenure on the performance of Vietnamese commercial banks.

Keywords:

• CEO characteristics
• Bank performance
• Agency theory
• Stewardship theory
• Resource dependency theory
• Vietnam

JEL Classification:

• G20
• G21
• C23

PUBLIC INTEREST STATEMENT

For any business, the role of the CEO is significant. The CEO is the head of the business, implementing the business development strategy and running the business’s day-to-day operations. The success or failure of a business depends significantly on the CEO. In the banking sector, which is essential to all other sectors, the role of the CEO is even more critical because of its far-reaching influence. By studying the impact of CEO characteristics on the performance of Vietnamese commercial banks, the study will provide a guideline that banks to choose the right CEO to boost their performance.

1. Introduction

In corporate governance, one of the crucial tasks of the board of directors is to find and recruit chief executive officers (CEOs) with outstanding qualities and capabilities (Hambrick, 1991). However, identifying and measuring CEO capabilities is a difficult, inaccurate, and expensive process, as evidenced by the huge demand for CEO recruitment and the considerable resources used to look for a CEO. Despite this costly searching process, newly hired CEOs are sometimes fired shortly after they are hired (Bhagat et al., 2010). For the banking sector, which is a particular field and requires many conditions, the CEO position becomes even more critical. It needs higher qualities to ensure the development and soundness of the bank (Elsharkawy et al., 2018). The CEO of a bank needs not only highly specialized knowledge and experience but also understands many other areas such as organizational culture, customer and employee psychology, and legal regulations in the field of the banking sector to solve the difficult and complex problems that the job requires.

Due to the important role of the CEO in the overall performance of the business, many researchers have conducted studies to understand the CEO’s characteristics to the business’s performance. One research direction is the influence of the CEO’s power on the performance of the business. Some researchers argue that when the CEO concurrently holds the chairman of the board of directors, it will cause conflicts of interest and harm the performance of the business (Gillan, 2006; Jensen, 1993; Rechner & Dalton, 1991). The basis of this argument is based on the idea of agency theory, according to which the CEO is the person who, through the board of directors, runs the business intending to maximize the value of the company and the interests of the shareholders. However, the CEO may pursue personal interests, even engaging in many fraudulent practices, and is contrary to the interests of shareholders.

On the contrary, some other researchers argue that when the power concentrated on the CEO is large. It will lead to better performance of the company because the CEO will have the right to self-determination as well as make decisions faster. The CEO will also be more cautious in decisions to protect reputation and work, thereby increasing the company’s performance and ensuring the interests of shareholders (Brickley et al., 1997; Duru et al., 2016; Liang et al., 2013). Another research direction is on how the personal characteristics of CEOs affect the performance of enterprises (Jadiyappa et al., 2019; Johan & Sari, 2020; Kaur & Singh, 2018; Pena et al., 2021). Those characteristics include gender, education level, age, and work experience. The third research direction is the remuneration CEOs are entitled to undertake (Akhigbe et al., 1997; Hambrick & Quigley, 2014; Matousek & Tzeremes, 2016). Jensen and Meckling (1976) suggested that to reduce the conflict of interests between shareholders and CEOs, besides the mechanisms to monitor the CEO’s activities, it is necessary to have an appropriate remuneration mechanism for the CEO and other senior managers. The results of previous studies show different trends in the influence of CEO characteristics on business performance. That difference may be due to differences in geography, time, methods, and research samples.

In Vietnam, according to the author’s review, only the study of Phan and Tran (2017) and the study of Vo et al. (2020) are relevant to this topic. However, research on Vietnamese commercial banks does not currently have any works. Therefore, to fill this gap, the author studies the impact of CEO characteristics on the performance of Vietnamese commercial banks. This study has several significant contributions. First, the study provides direct evidence on the influence of CEO characteristics on the performance of Vietnamese commercial banks; Second, the study provides more detailed information regarding the CEO qualities required and the motivation from CEO compensation to bank performance. To the authors’ knowledge, this study is the first to examine the influence of CEO characteristics on the performance of Vietnamese commercial banks.

2. Literature review

Agency theory (Jensen & Meckling, 1976) is considered an important foundation for explaining the relationship between the principal (shareholders) and the agent (CEO, members of the Board of Directors). According to agency theory, a conflict of interest exists between the principal and the agent because the agent tries to achieve his interests instead of the interests of the shareholders. This conflict shows the need for strong mechanisms to control and manage company activities (Wagana & Nzulwa, 2016). It is necessary to select individuals with diverse skills and work experience to promote firm performance (Carter et al., 2003). It is necessary to prevent managers from engaging in actions that pursue their personal goals and cause damage to economic aspects for shareholders (Valls & Rambaud, 2019). Therefore, to effectively supervise the CEO, it is necessary to separate the roles of the CEO and the board’s Chairman to avoid conflicts of interest and promote company performance (Jensen, 1993).

In contrast, the Stewardship theory argues that the CEO needs more power to perform his role in the best way. Therefore, the CEO concurrently serving as the Chairman of the Board of Directors will help the Board of Directors to grasp timely information and, at the same time, create favorable conditions for the CEO to make decisions quickly and with high self-determination, thereby improving the performance of the business (Donaldson & Davis, 1991; Miller & Sardais, 2011). In addition, Stewardship theory expects managers to be a reliable representatives and responsible for the organization’s assets in order to gain appreciation from colleagues and higher authority (Donaldson & Davis, 1991). In addition, resource dependency theory emphasizes that managers are hired based on excellent professional skills and experience. Therefore, managers’ ability to handle complex situations will be better (Jackling & Johl, 2009; Murray, 1989; Terjesen & Singh, 2009). Furthermore, companies in different contexts employ CEOs with qualities that match the company’s background (Salancik & Pfeffer, 1978).

2.1. Empirical studies

Gupta and Mahakud (2020) evaluate the personal characteristics of CEOs on the performance of commercial banks in India. Using the data set of 36 commercial banks from 2009–2017 and the FEM (Fixed Effects Model) method, the research results show that the CEO has the financial expertise to help improve the bank’s performance. Besides CEO duality, male CEO working experience of CEO also has a positive influence on the bank’s performance.

Satriyo and Harymawan (2018) studied the role of female CEOs on the performance of companies listed on the Indonesian stock market in 2014–2015. They divide the research sample into small-sized and large-sized companies. The results showed evidence of a negative effect of female CEOs on the performance of small firms and found no evidence of such a relationship for large firms.

In a similar study, Jadiyappa et al. (2019) found evidence of a negative influence of female CEOs on the performance of 100 companies in India. In addition, they believe that this negative effect is related to the increase in agency costs after appointing a female CEO.

Johan and Sari (2020) study the impact of CEO characteristics on the performance of commercial banks in Indonesia the period 2014–2018. The seven CEO characteristics are age, education level, appointment term, experience, previous deputy director position, expert role, or internal promotion and associated with major shareholders. The results show that CEO age positively affects the performance of banks.

Lam et al. (2013) studied the impact of CEO characteristics on the performance of companies listed on the Chinese stock market in the period 2000–2008. The results of the study showed that CEO gender and CEOs concurrently do not affect the performance of companies. Meanwhile, the CEO’s tenure positively affects the performance of companies.

In Vietnam, a study by Vo et al. (2020) and a study by Phan and Tran (2017) are related to the research. In the study of Vo et al. (2020), they examine the influence of the CEO’s gender on the performance and risk of listed companies in Vietnam. The authors find that companies with female CEOs generate higher profits than those with male CEOs. In addition, companies led by female CEOs have less systematic and specific risk and lower volatility in firms’ return on assets.

Meanwhile, Phan and Tran (2017) conducted a study to measure the impact of CEO characteristics on firm performance in Vietnam. Based on a data sample of 120 companies listed on the Ho Chi Minh City Stock Exchange from 2009–2015, the research results show that the impact of the CEO’s age and ownership ratio on firm performance is nonlinear. In addition, companies with a CEO concurrently serving as chairman of the board (CEO duality) perform better than companies that do not maintain this structure.

2.2. Research hypothesis

2.2.1. CEO duality

In the organizational structure of companies, the CEO and the chairman of the board of directors are two different individuals, or the CEO is also the board chairman. According to agency theory (Jensen & Meckling, 1976), it is necessary to separate the roles of the CEO and the chairman of the board to resolve the issue of conflict of interest. However, according to the Stewardship theory, it is necessary to give the CEO many powers and greater self-determination ability (Donaldson & Davis, 1991; Miller & Sardais, 2011). The company’s organizational structure in the form of duality will give the CEO a high degree of self-determination, forming a leadership style, and thus create value and better corporate performance (Phan & Tran, 2017). Gupta and Mahakud (2020) argue that the person who holds both positions of CEO and Chairman can have more extensive knowledge of the business environment and make strategic decisions more successfully; CEO duality also can weaken the influence of interest groups and increase the ability to react to changes. Therefore, the author proposes the hypothesis:

Hypothesis H1: CEO duality has a significant positive impact on the performance of commercial banks.

2.2.2. Female CEO

Stewardship theory argues that the CEO is the person who acts on behalf of the owners, and the CEO is responsible for the critical tasks of the company. In particular, female CEOs always want to perform the best in all essential tasks (Donaldson & Davis, 1991; Zhu et al., 2022). However, female CEOs are often risk-averse, more cautious in using corporate capital, and often less self-interested than men (Khan & Vieito, 2013). However, Gupta and Mahakud (2020) argue that male CEOs are risk-loving and, therefore, bring higher performance because of the positive relationship between risk and return. Therefore, the author proposes the hypothesis:

Hypothesis H2: Female CEO has a significant negative impact on the performance of commercial banks.

2.2.3. The CEO’s financial background

Resource dependency theory states that CEOs need to have good technical skills and are suitable for the background of the business. Therefore, CEOs with a background in finance will have specialized skills in finance and are suitable for the banking sector. A CEO with a financial background helps the bank to run smoothly and efficiently and has a good understanding of financial problems in the financial markets (Gupta & Mahakud, 2020; Lusardi & Mitchell, 2006). In addition, a good financial background helps the CEO better communicate with external investors, execute financial policies more dynamically, and adapt well to the rapidly changing external environment (Custódio & Metzger, 2014). Therefore, the author proposes the hypothesis:

Hypothesis H3: CEO with a background in finance has a significant positive impact on the performance of commercial banks.

2.2.4. CEO tenure

Long-serving CEOs can form a management team that can collaborate effectively and thus improves work performance (Gupta & Mahakud, 2020). Longer-term CEOs are an important asset to banks as they develop and enhance their ability to learn and understand bank-specific issues (Finkelstein, 1992). Resource dependency theory suggests that CEOs with long tenure and experience handling complex situations promote organizational effectiveness (Jackling & Johl, 2009; Murray, 1989; Terjesen & Singh, 2009). Therefore, the author proposes the hypothesis:

Hypothesis H4: CEO tenure positively significant impacts the performance of commercial banks.

2.2.5. CEO age

Resource dependency theory suggests that CEOs with work experience will boost firm performance (Salancik & Pfeffer, 1978). However, the older the CEO age, the lower the company’s performance because the elderly CEO often values both career security and financial security; at the same time, they tend to be conservative, risk-averse, less agile, and creative while the market competition is getting stronger (Phan & Tran, 2017). Older CEOs may be more inclined to promote their interests and goals and enjoy a peaceful life, which can lead to a decline in the performance of companies headed by older CEOs. (Bertrand & Mullainathan, 2003; Gupta & Mahakud, 2020). Therefore, the author proposes the hypothesis:

Hypothesis H5: CEO age has a significant negative impact on the performance of commercial banks.

3. Research model and methodolody

3.1. Data

The author collects data from audited annual reports and consolidated financial statements of 30 Vietnamese commercial banks from 2012–2020. As of 31 December 2020, there are 35 commercial banks in Vietnam, of which there are four commercial banks with 100% state-owned. The author excludes 100% foreign-owned and 100% state-owned banks in the sample because there is no data. The dependent variable in the research model is the performance of commercial banks, including ROA and ROE (Gupta & Mahakud, 2020; Lam et al., 2013; Vo et al., 2020). ROA is a profitability ratio that measures a bank’s ability to generate profits using all its resources or assets. ROE measures a bank’s ability to generate profits from shareholders’ capital in the bank.

Independent variables in the research model include CEO duality, female CEO, CEO financial background, CEO tenure, and CEO age. CEO dual (CDUAL) is a dummy variable, taking the value one if the CEO is also the chairman of the board and 0 if not. Female CEO (CFEMALE) is a dummy variable, taking the value one if the CEO is female, and taking the value 0 if male. CEO Financial Background (CFIN) is a dummy variable that takes the value of 1 if the CEO has a financial background and the value of 0 if there is no financial background. CEO tenure (CTENURE) is the number of years in the CEO’s office. CEO age (CAGE) is the CEO’s age in years.

Control variables include bank size (SIZE), bank capital (CAP), GDP growth (GGDP). In which bank size (SIZE) is calculated by the logarithm of total assets, bank capital is calculated as total equity divided by total assets, and GDP growth (GGDP) is annual economic growth.

3.2. Research model and methodology

The research model is built from relevant theories and empirical studies of Khan and Vieito (2013), Gupta and Mahakud (2020), and Vo et al. (2020). The specific research model is as follows:

$\begin{array}{rl}& RO{A}_{it}={\propto }_0+{\propto }_{1}CDUA{L}_{it}+{\propto }_{2}CFEMAL{E}_{it}+{\propto }_{3}CFI{N}_{it}+{\propto }_{4}CTENUR{E}_{it}+{\propto }_{5}CAG{E}_{it}+{\propto }_{6}SIZ{E}_{it}\\ & +{\propto }_{7}CA{P}_{it}+{\propto }_{8}GGD{P}_t+\epsilon \end{array}$

$\begin{array}{rl}& \mathrm{R}\mathrm{O}{\mathrm{E}}_{\mathrm{i}\mathrm{t}}={\beta }_0+{\beta }_1\mathrm{C}\mathrm{D}\mathrm{U}\mathrm{A}{\mathrm{L}}_{\mathrm{i}\mathrm{t}}+{\beta }_2\mathrm{C}\mathrm{F}\mathrm{E}\mathrm{M}\mathrm{A}\mathrm{L}{\mathrm{E}}_{\mathrm{i}\mathrm{t}}+{\beta }_3\mathrm{C}\mathrm{F}\mathrm{I}{\mathrm{N}}_{\mathrm{i}\mathrm{t}}+{\beta }_4\mathrm{C}\mathrm{T}\mathrm{E}\mathrm{N}\mathrm{U}\mathrm{R}{\mathrm{E}}_{\mathrm{i}\mathrm{t}}+{\beta }_5\mathrm{C}\mathrm{A}\mathrm{G}{\mathrm{E}}_{\mathrm{i}\mathrm{t}}+{\beta }_6\mathrm{S}\mathrm{I}\mathrm{Z}{\mathrm{E}}_{\mathrm{i}\mathrm{t}}\\ & +{\beta }_7\mathrm{C}\mathrm{A}{\mathrm{P}}_{\mathrm{i}\mathrm{t}}+{\beta }_8\mathrm{G}\mathrm{G}\mathrm{D}{\mathrm{P}}_{\mathrm{t}}+u\end{array}$

where ∝₀, β₀ are constant terms; ∝_i (i = 1,8); β_i (i = 1,8) are coefficients of variables; ε, $u$ are error terms.

SIZE, CAP, GGDP are control variables.

There are different methods for estimating a research model with panel data. In general, panel data regression can be estimated using Pooled OLS model, fixed effects model (FE), and random effects model (RE). Because each estimation method is valid, the author uses different tests to choose the estimation method that best fits the research data: F test to choose between Pooled OLS and FE; and Hausman test to choose between FE and RE. Besides, Modified Wald tests and Wooldridge tests will be performed to consider the model’s defects, such as heteroskedasticity and autocorrelation. If the model has defects, the author proceeds to overcome the defects with the FGLS method.

4. Research results and discussion

4.1. Descriptive statistics and correlation analysis

The statistics on CEO characteristics and performance of commercial banks along with the control variables are described in Table 1, including mean, maximum and minimum values, and standard deviation.

Table 1. Variable definitions

Variables		Notation	Previous studies	Expected results	Proxy variables
Dependent variable	Bank performance	Return on assets (ROA)	Lam et al. (2013), Gupta and Mahakud (2020), Vo et al. (2020)		Profit after tax /Total assets
		Return on equity (ROE)			Profit after tax /Equity
Independent variables	CEO characterristics	CEO duality (CDUAL)	Phan and Tran (2017), Gupta and Mahakud (2020)	+	Value 1 if the CEO is also the chairman of the board, value of 0 if not
		Female CEO (CFEMALE)	Khan and Vieito (2013), Satriyo and Harymawan (2018), Jadiyappa et al. (2019), Gupta and Mahakud (2020), Vo et al. (2020)	-	value 1 if the CEO is female, value 0 if male
		CEO’s financial background (CFIN)	Lusardi and Mitchell (2006), Gupta and Mahakud (2020)	+	Value of 1 if the CEO has a financial background, value of 0 if not
		CEO tenure (CTENURE))	Lam et al. (2013), Gupta and Mahakud (2020)	+	The number of years in office of the CEO
		CEO age (CAGE)	Phan and Tran (2017), Gupta and Mahakud (2020)	-	Age of CEO, in years

The results from Table 2 show that the ROA of banks has a mean value of 0.71%, with a standard deviation of 0.58%. The lowest ROA is 0.01%, and the highest is 2.80%. The ROE of banks has a mean of 8.30%, with a standard deviation of 6.36%. The lowest ROE is 0.02%, and the highest is 25.82%.

Table 2. Descriptive statistics of the bank performance and CEO characteristics

Variable	Obs.	Mean	Std.dev	Min	Max
Bank performance
ROA	248	0.0071	0.0058	0.0001	0.0280
ROE	248	0.0830	0.0636	0.0002	0.2582
CEO characteristics
CDUAL	260	0.6307	0.4835	0.0000	1.0000
CFEMALE	239	0.1171	0.3221	0.0000	1.0000
CFIN	239	0.0041	0.0646	0.0000	1.0000
CTENURE	172	13.4651	7.6160	1.0000	29.0000
CAGE	210	47.9476	6.5213	34.0000	64.0000
Control variables
SIZE	248	32.4213	1.1431	29.4931	34.9553
CAP	248	0.0937	0.0621	0.0262	0.6140
GGDP	9	0.0592	0.0123	0.0291	0.0708

Regarding CEO characteristics, the average value of the CDUAL variable is 63.07%, showing that a high percentage of CEOs of Vietnamese commercial banks hold the position of Chairman of the Board of Directors simultaneously. In contrast, the mean values of the variables CFEMALE and CFIN are pretty low, reaching values of 11.71% and 0.41%, respectively, indicating that most CEOs are male and have no background in finance and banking.

The mean of the CTENURE variable is 13.46, and the standard deviation is 7.6, indicating a relatively large degree of volatility of the CTENURE variable. Similarly, the CAGE variable has a mean of 47.94 with a standard deviation of 6.52. In addition, the CEO’s youngest age is 34, and the oldest is 64.

For control variables, bank size (SIZE) has a mean value of 32.43, bank capital (CAP) has a mean value of 9.37%, and average annual GDP growth (GGDP) is 5.92%.

5. Correlation analysis

The correlation matrix between the independent and control variables in the sample is shown in Table 3. The correlation coefficients are all less than 0.8 (the largest is 0.28). Therefore, it can be assumed that the phenomenon of multicollinearity is not serious.

Table 3. Correlation coefficients between measures of CEO characteristics and bank performance

	ROA	ROE	CDUAL	CFEMALE	CFIN	CTENURE	CAGE	SIZE	CAP	GGDP
ROA	1.0000
ROE	0.8404*	1.0000
CDUAL	0.0737	0.0986	1.0000
CFEMALE	−0.1235*	−0.1673*	0.0781	1.0000
CFIN	0.0253	0.0411	0.0438	−0.0236	1.0000
CTENURE	−0.0036	0.1259*	0.0608	0.2118*	−0.1356*	1.0000
CAGE	0.2871*	0.2243*	0.0007	0.2697*	−0.0402	0.0119	1.0000
SIZE	0.2291*	0.5332*	−0.0778	−0.2040*	−0.0021	0.2713*	−0.0865*	1.0000
CAP	0.2433*	−0.1633*	0.0533	0.0485	−0.0185	−0.1846*	0.0475	−0.5579	1.0000
GGDP	−0.1067*	−0.0352	0.0208	0.0028	−0.1595*	−0.0496	0.0076	−0.0081	−0.0777	1.0000

* indicate significance at the 10% level.

5.1. Pooled OLS regression

Table 4 presents the regression results using the Pooled OLS method, with ROA and ROE as the dependent variables.

Table 4. The effect of CEO characteristics on bank performance—Pooled OLS regression

	Dependent ROA	Dependent ROE
CDUAL	0.0020** (0.0007)	0.0264*** (0.0081)
CFEMALE	−0.0012 (0.0011)	−0.0252** (0.0029)
CFIN	0.0039 (0.0043)	0.0376 (0.0490)
CTENURE	0.0001 (0.0000)	−0.0021 (0.0005)
CAGE	0.0002* (0.0000)	0.0029*** (0.0006)
SIZE	0.2125*** (0.0032)	0.0406*** (0.0043)
CAP	0.1196*** (0.0132)	0.0390*** (0.1502)
GGDP	−0.0065 (0.0252)	−0.0607 (0.2873)
_cons	−0.1387** (0.0138)	−1.4124*** (0.1518)
Observations	167	167
Adj R–squared	0.5189	0.4733
Prob > F	0.0000	0.0000

Standard error in parentheses; *significant at the 10% level; **significant at the 5% level; ***significant at the 1% level.

The results from Table 4 show that the CEO characteristics that impact the performance of commercial banks include CDUAL, CFEMALE, CAGE. However, the limitation of the Pooled OLS method is that it does not control for individual characteristics of each bank in the sample. Therefore, the author continues to perform regression according to FE and RE methods to overcome the above limitation.

5.2. Fixed and random effects models

The results from Table 5 show the impact of CEO characteristics on the performance of Vietnamese commercial banks by using FE and RE models. The results show that Prob > F = 0.0000, so the FE model is more suitable than the Pooled OLS model. Hausman test is performed to choose between FE and RE models. With the dependent variable ROA, the results show that Prob>chi2 = 0.0003 < 5%, so the FE model is suitable. The results of the dependent variable ROE show that Prob>chi2 = 0.0026 < 5%, so the FE model is suitable.

Table 5. The effect of CEO characteristics on bank performance—FE and RE models

	Dependent ROA		Dependent ROE
	FE	RE	FE	RE
CDUAL	0.0001 (0.0006)	0.0007 (0.0006)	0.0081 (0.0046)	0.0155** (0.0079)
CFEMALE	−0.0036*** (0.0014)	−0.0017 (0.0013)	−0.0402*** (0.0182)	−0.0208 (0.0152)
CFIN	0.0015 (0.0034)	0.0029 (0.0047)	0.0034 (0.0011)	0.0396 (0.0506)
CTENURE	0.0001 (0.0003)	0.0001 (0.0001)	0.0013 (0.0008)	0.0002 (0.0006)
CAGE	−0.0001 (0.0032)	0.0002** (0.0776)	0.0001 (0.0045)	0.0016** (0.0007)
SIZE	0.0069*** (0.0007)	0.0053*** (0.0005)	0.0713*** (0.0091)	0.0503*** (0.0060)
CAP	0.1185*** (0.0425)	0.1137*** (0.0121)	0.3927*** (0.1550)	0.3406** (0.1345)
GGDP	0.0030 (0.0199)	−0.0149 (0.0202)	0.0067 (0.2465)	−0.1296 (0.2425)
_cons	−0.2297*** (0.0240)	−0.1736*** (0.0184)	−0.2873*** (0.2974)	−1.6547*** (0.2022)
Observations	167	167	167	167
Adj R–squared	0.3098	0.4182	0.3635	0.4373
Prob > F	0.0000	0.0000	0.0000	0.0000

Standard error in parentheses; *significant at the 10% level; **significant at the 5% level; ***significant at the 1% level.

To assess whether the two models have autocorrelation, the author conducts the Wooldridge test for both models with the dependent variable ROA and ROE. The results show that Prob > chi2 = 0.0000 < 5% in the two models, so it is concluded that both models exist autocorrelation.

To evaluate whether the two models have heteroskedasticity, the author conducts the Modified Wald test. The results show that the Prob value > chi2 = 0.0000 < 5% in the two models, so it is concluded that both models exist the phenomenon of heteroskedasticity.

5.3. Feasible Generalized Least Square method

The results from Table 6 show that the CEO characteristics that impact the performance of Vietnamese commercial banks are CEO duality, female CEO, and CEO age.

Table 6. The effect of CEO characteristics on bank performance—FGLS regression

	Dependent ROA	Dependent ROE
CDUAL	0.0020** (0.0006)	0.0264*** (0.0079)
CFEMALE	−0.0012 (0.0011)	−0.0252** (0.0126)
CFIN	0.0029 (0.0041)	0.0376 (0.0477)
CTENURE	0.0001 (0.0000)	−0.0021 (0.0005)
CAGE	0.0002*** (0.0000)	0.0029*** (0.0006)
SIZE	0.0037*** (0.0003)	0.0406*** (0.0041)
CAP	0.1196*** (0.0128)	0.3690*** (0.1461)
GGDP	−0.0065 (0.0246)	−0.0607 (0.2795)
_cons	−0.1387** (0.1387)	−1.4124*** (0.1476)
Observations	167	167
Prob > F	0.0000	0.0000

Standard error in parentheses; *significant at the 10% level; **significant at the 5% level; ***significant at the 1% level.

The factor CEO duality (CDUAL) has a positive impact on the performance of commercial banks, consistent with Stewardship theory, research hypothesis, and studies of Phan and Tran (2017), Gupta and Mahakud (2020). The banking sector is a special sector of the economy subject to rapid change and sensitivity to the central bank’s monetary policy. Therefore, the CEO duality will be given more power to make quick and timely decisions with the fluctuations of the macroeconomy. In addition, the CEO duality with a deep understanding and rich experience in the banking sector also promotes more efficient banking operations.

Female CEO (CFEMALE) have a negative impact on ROE of commercial banks, consistent with the research hypothesis, studies by Satriyo and Harymawan (2018), Jadiyappa et al. (2019), and Gupta and Mahakud (2020), but contrary to the research results of Khan and Vieito (2013), Vo et al. (2020). The female CEO is a cautious person who always puts the safety of the bank’s operations on top. Therefore, female CEO will not take risks in operating in order to achieve the highest profit and have a negative impact on the performance of commercial banks. In addition, female CEO is often more emotional, less assertive at work, and more limited in creating and maintaining relationships than male CEOs. That also has a negative impact on the performance of commercial banks.

CEO age (CAGE) has a positive impact on the performance of commercial banks, consistent with the research hypothesis, resource dependency theory, and studies of Lam et al. (2013), Gupta and Mahakud (2020). The CEO with a longer working time will have a deep understanding of the bank’s operations, making quicker and more accurate decisions in banking operations. In addition, the longer working time of the CEO also allows them to establish a reliable and efficient working group, thereby promoting the bank’s better performance.

6. Conclusion

The study was conducted to evaluate CEO characteristics on the performance of 30 Vietnamese commercial banks in the period 2012–2020. In this study, the performance of commercial banks is measured by ROA and ROE indicators. By quantitative method through Pooled OLS, FE, RE, FGLS methods, the regression results show that the factor CEO duality (CDUAL) has a positive impact on ROA and ROE, female CEO (CFEMALE) has a negative effect on ROE but has no effect on ROA, CEO age (CAGE) has a positive effect on ROA and ROE. In addition, the financial background of CEO (CFIN) and CEO tenure (CTENURE) have no impact on the performance of Vietnamese commercial banks. The above research results provide helpful information for the governance of Vietnamese commercial banks, thereby improving the performance of banks.

Firstly, the CEO needs to be given more power to promote the performance of commercial banks. The banking sector is subject to rapid and complex changes. Therefore, the CEO needs more power to quickly make decisions in line with market movements. If decisions are made late due to the CEO’s limited power, negative impact on banking operations will occur and reduce the bank’s performance. However, as the CEO is given more power, a monitoring mechanism is needed to prevent the CEO’s abuse of power.

Second, female CEO has a negative impact on ROE of Vietnamese commercial banks. Therefore, it is necessary for banks with female CEOs to build stronger teams to support CEO decision-making. For banks pursuing profit goals, male CEOs will be more suitable. On the contrary, female CEOs are the right choice for banks that need safety.

Banks should choose CEOs who are older and have more work experience to improve the bank’s operational efficiency. The CEO of a bank with long life will have the ability to accurately assess the movements of the market, thereby making accurate decisions.

The study has provided evidence showing the influence of CEO characteristics on the performance of Vietnamese commercial banks. However, the study has some limitations, such as not including in the model some other factors that are likely to influence, such as CEO nationality, CEO remuneration, or consider the maximum age of CEOs who can run a bank effectively. Therefore, in future studies, the author will consider and supplement these factors to better assess the influence of CEO characteristics on the performance of Vietnamese commercial banks.

Disclosure statement

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

Additional information

Funding

The authors have no funding to report.

Notes on contributors

Nam Hai Pham

Dr. Nam Hai Pham is working at Faculty of Banking, Ho Chi Minh University of Banking. His research interests are bank performance, bank risk, and corporate governance in commercial bank. His research papers have been published in prestigious journals and book series, such as Cogent Business and Management; Studies in Computational Intelligent; Studies in Systems, Decision, and Control; Journal of Asian Business and Economic Studies, and Asian Journal of Economics and Banking.