CEO busyness and firm performance: evidence from Indonesia.

This study investigates whether busy CEOs are associated with lower firm performance, and if this relationship is moderated by firm growth, CEO tenure and corporate governance practices in Indonesia. This study uses 876 firms-year observations from 268 firms listed on the Indonesia Stock Exchange (IDX) for the period spanning 2014 to 2017. We find that busy CEOs are associated with lower firm performance. This negative relationship is stronger in firms with high growth and when busy CEOs have shorter tenure. We also show that corporate governance practices have no impact on the negative relationship between CEO busyness and firm performance. For firms and shareholders, our findings indicate that it is not a good idea for CEOs to hold two or more outside directorships, particularly in the early years of taking up their CEO position. Our results suggest that restrictions on CEO busyness would be beneficial to shareholders.

Introduction

Chief Executive Officers (CEOs) hold the top position in corporate management and are responsible for company operations and performance. Companies hire CEOs based on their expertise, experience and ability to create value for shareholders. An experienced CEO that holds directorships in other companies may be beneficial to firms as their extended network can help to create new partnerships and expand firm opportunities. However, too many outside commitments can make a CEO too busy, causing a negative effect on firm operations and performance. This study examines these issues in the context of Indonesia, by relating the busyness of CEOs (called President Directors in Indonesia) to firm performance.

Prior studies of boards of directors show that there are both benefits and costs of directors holding outside directorships. Fama and Jensen (1983) suggest that additional board seats are a signal of quality. Hillman and Dalziel (2003) argue that board members with more relevant industry expertise, experience, knowledge and skills are able to make more informed decisions to solve problems and improve firm performance. Other studies argue that directors with outside directorships have extensive social and professional networks to forge partnerships and deal with existing regulations, have high integrity and a good reputation (Masulis and Mobbs, 2014; Field et al., 2013; Mendez et al., 2017).

However, when directors overextend themselves they become busier and unfocused, which leads their performance to be less than optimal, thus ultimately harming the firm. A number of studies document the negative effect of busy directors. Core et al. (1999) find that busy directors do not have enough time to carry out their duties. Jiraporn et al. (2009) indicate that busier directors tend to miss more board meetings. Fich and Shivdasani (2006) show that busy directors are associated with negative firm performance.

We apply these same arguments to CEO busyness in Indonesia, a market where many CEOs of listed companies hold large numbers of outside positions, but the busyness of corporate officers has not been previously investigated. Using a sample of 876 firm-year observations of companies listed on the Indonesia Stock Exchange from 2014 to 2017, we test for a negative relationship between CEO busyness and firm performance. We also examine whether this negative relationship is stronger in high growth firms and in the early years of CEO tenure, and if strong corporate governance practices help to alleviate the negative influence of CEO busyness.

We find that the majority of CEOs are classified as busy in Indonesia and that busy CEOs are associated with lower firm performance. This indicates that CEOs with multiple outside directorships are distracted and lose their focus, causing a damaging effect on firm performance. We find consistent evidence that this negative relationship is stronger when busy CEOs have shorter tenure and some evidence that it is stronger when firm growth is higher. In these situations, it is even more important for CEOs to be focused on firm operations and not be distracted by outside activities. Finally, we find that corporate governance practices have no impact on the negative relationship between CEO busyness and firm performance. This suggests that boards of directors are not implementing effective practices to limit the negative effect of busy CEOs on firm operations.

This research makes a number of contributions. It identifies an issue of current importance in Indonesia, in that some CEOs are too busy and not focusing appropriate attention to their firm's operations. Our findings indicate that it is not a good idea for CEOs to hold two or more outside directorships. For firms experiencing high growth and for firms that have recently hired a new CEO, this effect is even stronger, suggesting that particular attention needs to be paid to this issue. Thus, this study extends our knowledge of the characteristics of CEOs that are related to firm performance. When hiring new CEOs it is important for firms to understand the outside commitments of new CEOs and whether they can devote enough attention to their position.

The remainder of this paper is organized as follows. The next section outlines the relevant research and develops the hypotheses. Section 3 details the sample, variables and empirical models. Section 4 provides the empirical analysis and results. Section 5 outlines the conclusions and implications of the study.

Hypothesis

CEOs play an important role in guiding and driving the success of their companies. In this study, we specifically examine the influence a CEO with two or more outside directorships has on the performance of their company. The existing literature has highlighted two schools of thought with respect to CEOs holding additional positions in other companies.

This first line of thinking is based on CEO reputation, expertise and human capital. Fama and Jensen (1983) suggest that the number of outside board seats a corporate officer holds signals the quality of their reputation. CEOs with better reputations will be in higher demand and have greater opportunity to acquire additional positions in other companies. With respect to expertise, a number of studies show that corporate officers with more experience and expertise in corporate management are favoured by shareholders. For example, Fich (2005) shows that shareholders react positively to the appointment of outside directors with experience as CEOs of other companies, as the hiring firm is expected to benefit from the expertise and experience of the CEO. Other studies also argue that corporate officers with extensive social and professional networks help companies to forge partnerships and deal with existing and new regulations (Masulis and Mobbs, 2014; Field et al., 2013; Mendez et al., 2017). Based on these arguments, we would expect a positive relationship between CEOs with outside directorships and firm performance.

The second school of thought is that multiple outside directorships are a signal of busyness. Busyness causes the CEO to not have enough time and energy to focus on the main tasks in managing the company. As a result, firm activity will be disrupted and firm performance will decrease. This view is in line with existing research on busy directors (Fich and Shivdasani, 2006). For example, Ferris et al. (2003) show that busy directors impose excessive restrictions on themselves, thereby reducing the time and effort they devote to managing the company. Core et al. (1999) also find that busy directors do not have enough time to carry out their duties. Jiraporn et al. (2009) indicates that busier directors tend to miss more board meetings. Other studies have also shown that the director's busyness interfere with their effectiveness in managing the company (Cashman et al., 2012; Falato et al., 2014). These arguments propose that CEOs with outside directorships are associated with negative firm performance.

In the context of this study, we expect the negative effect of CEO busyness to overshadow any potential positive effects of CEO reputation, expertise and human capital. Thus, our first hypothesis is:

Busy CEOs are associated with lower firm performance.

This study also examines two specific situations where we expect the negative effects of CEO busyness on firm performance to be even more dramatic. These two situations are in high growth firms and in the early years of a CEO's tenure. We believe that both of these situations require the complete focus of the CEO and if the CEO is not fully focused on their work, we expect the negative consequences of CEO busyness to be greater.

In high growth firms, firm management needs to be continually up-to-date with market and industry conditions to ensure the company is pressing ahead with its strategy and investing its limited resources in the most beneficial areas. If the CEO of a firm is busy, due to many commitments outside of the firm, we expect this loss of attention to incur a greater cost on the performance of high growth firms, than in low growth firms. Thus, we expect the negative effects of CEO busyness to be greater in firms with high growth.

The negative relationship between busy CEOs and firm performance is stronger in high growth firms.

In addition, when a new CEO is hired by a company we expect that it takes a substantial period of time for the new CEO to fully understand the operations of the company and its competitive position in the market. If a CEO also holds outside directorships, they may be distracted by their other commitments and therefore not perform to their highest ability. We expect this to be most serious in the early years of a CEO's tenure. In the later years, a CEO will have already built up their knowledge and expertise of the company's operations and its competitive position. Thus, any distraction from outside directorships will have a smaller effect in later years of a CEO's tenure. This argument is supported by prior studies that show that directors perform better in their duties as their tenure increases (Shiah-Hou and Cheng, 2012; Barker III and Mueller, 2002; Kor and Sundaramurthy, 2009; Fama and Jensen, 1983).

The negative relationship between busy CEOs and firm performance is weaker in firms with long-tenured CEOs.

Since we predict a negative relationship between busy CEOs and firm performance, we would like to further investigate whether this relationship is moderated by the corporate governance practices of companies. Supervisory bodies, such as the board of directors, are expected to monitor the activities of the CEO and be aware if their CEO has many outside commitments. An effective board of directors would recommend practices or take measures to limit the influence of the CEO's outside activities in order to offset the negative effect of CEO busyness on firm operations and performance. Thus, our final hypothesis investigates the moderating effect of strong corporate governance practices (e.g. bigger and more independent boards) on the relationship between CEO busyness and firm performance.

The negative relationship between busy CEOs and firm performance is weaker in firms with stronger corporate governance practices.

Method

Data and sample

The initial sample consists of all companies listed on the Indonesia Stock Exchange (BEI) for the period 2014–2017. Sources of data in this study include company annual reports, ORBIS database and Bloomberg website. Financial data was obtained from the ORBIS database. Data on CEOs (called President Directors in Indonesia), including the number of positions and the length of tenure, was obtained from the Bloomberg website. Data for corporate governance variables was obtained from company annual reports. These datasets were merged and the following sample selection criteria were applied. First, all companies from the finance, insurance and real estate industries (SIC 6) were excluded because of the different nature of their financial statements. Second, any observations without complete data were excluded from the sample. After applying the sample selection criteria, the final sample includes 876 firm-year observations.

Variable definitions

The main variable of interest in this study is the busyness of the CEO (BUSYCEO). This is measured using a dummy variable, with CEOs (President Directors) categorized as busy when they hold two or more outside directorships (Core et al., 1999; Ferris et al., 2003; and Fich and Shivdasani, 2006). Firm performance (FP) is the dependent variable and is measured by return on assets (ROA), return on sales (ROS) and return on equity (ROE). ROA is measured as net income divided by total assets. ROS is measured as profit before interest expense and tax divided by total sales. ROE is net income divided by the book value of total equity. These three return variables are expressed as a percentage and are winsorized at the 1% and 99% levels.

Referring to previous research (Masulis and Mobbs, 2014; Mendez et al., 2017; Cashman et al., 2012; Field et al., 2013; Andres et al., 2013; Kusnadi et al., 2016; Charas, 2015; Bravo and Reguera-Alvarado, 2017; Habib et al., 2016; Harymawan and Nowland, 2016; Tao and Hutchinson, 2013) the control variables used in this study include: CEO tenure (TENURE), the existence of a nomination and remuneration committee (NRC), board size (BOARDSIZE), the percentage of independent commissioners (INDCOM), number of audit committee members (AUDCOM), the existence of Big4 auditors (BIG4), firm leverage (LEVERAGE), firm size (FIRMSIZE), firm age (FIRMAGE) and operating cash flow (OCF). Financial variables are winsorized at the 1% and 99% levels. See Table 1 for a summary of variable definitions.

Table 1

Variable definitions.

Variable	Definition	Source
Dependent:ROAROSROE	Net income divided by total assets (%).Earnings before interest and tax divided by total sales (%).Net income divided by book value of total equity (%).	ORBISORBISORBIS
Independent:BUSYCEO	Dummy variable, 1 for CEOs who hold two or more other directorships, and 0 for CEOs who hold less than two other directorships.	Bloomberg
Controls:
TENURE	Dummy variable, 1 for CEOs who have served their company more than five years, and 0 for CEOs who have served their company less than or equal to five years.	Bloomberg
NRC	Dummy variable, 1 if the company has a Nomination and Remuneration Committee, and 0 if the company does not have a Nomination and Remuneration Committee.	Financial Report
BOARDSIZE	Natural logarithm of the number of members of the board of directors and board of commissioners in the company.	Financial Report
INDCOM	Percentage of independent commissioners in the company.	Financial Report
AUDCOM	Number of members of the audit committee.	Financial Report
GROWTH	The difference between total sales minus lag total sales scaled by lag total sales.	ORBIS
LEVERAGE	Total debt divided by total assets.	ORBIS
FIRMSIZE	Natural logarithm of total assets.	ORBIS
FIRMAGEOCF	Natural logarithm of the number of years since the company was founded.Net cash flow from operating activities divided by total assets.	ORBISORBIS
BIG4	Dummy variable, 1 for companies audited by Big 4 auditors, and 0 for companies audited by other auditors.	Financial Report

Methodology

This study uses an OLS regression model with fixed year and industry effects, and clustered standard errors (Petersen, 2009). To test the first hypothesis in this study, the following regression model is used. Based on Hypothesis 1, we expect the coefficient on BUSYCEO to be negative.

To test the second hypothesis, model 2 is used. We expect the coefficient on BUSYCEO*GROWTH to be negative.

For the third hypothesis, we use model 3. We expect the coefficient on BUSYCEO*TENURE to be positive.

The fourth hypothesis is tested using model 4. We expect the coefficient on BUSYCEO*GOV to be positive, where GOV represents the corporate governance variables BOARSIZE, INDCOM and RNC.

4. RESULTS AND DISCUSSION

Descriptive statistics and univariate comparisons

Table 2 contains the sample distribution by industry sector in Panel A, and by year in Panel B. The sample firms are distributed across the eight industry sectors, with the highest number of firm-year observations coming from Construction Industries (247), Manufacturing (162) and Mining (140). There are 238 firm-year observations from 2014, 239 from 2015, 237 from 2016 and 162 from 2017. Of the total of 876 firm-year observations, 477 (54%) are from firms with busy CEOs. Thus, busy CEOs are present in the majority of companies.

Table 2

Sample distribution.

Panel A: Sample Distribution by Industry Sector
Industry	Busy CEOs		Non-Busy CEOs		Total
	N	%	N	%	N	%
(SIC 0) Agriculture, Forestry and Fisheries	27	62.79	16	37.21	43	100
(SIC 1) Mining	82	58.57	58	41.43	140	100
(SIC 2) Construction Industries	106	42.91	141	57.09	247	100
(SIC 3) Manufacturing	89	54.94	73	45.06	162	100
(SIC 4) Transportation, Communications and Utilities	87	70.73	36	29.27	123	100
(SIC 5) Wholesale & Retail Trade	46	54.76	38	45.24	84	100
(SIC 7) Service Industries	33	57.89	24	42.11	57	100
(SIC 8) Health, Legal, and Educational Services and Consulting	7	35.00	13	65.00	20	100
Total	477	54.45	399	45.55	876	100
Panel B: Sample Distribution by Year
Year	Busy CEOs		Non-Busy CEOs		Total
	N	%	N	%	N	%
2014	125	52.52	113	47.48	238	100
2015	131	54.81	108	45.19	239	100
2016	132	55.70	105	44.30	237	100
2017	89	54.94	73	45.06	162	100
Total	477	54.45	399	45.55	876	100

Notes: This table shows the sample distribution of companies that have busy and non-busy CEOs. The sample includes 876 firm-year observations from companies listed on the IDX during 2014–2017.

Table 3 shows descriptive statistics of companies with Busy CEOs (Panel A) and companies with Non-Busy CEOs (Panel B). In Panel A, companies with Busy CEOs have mean ROA, ROS and ROE of 3.569%, 3.189% and 5.490%. The average board size is 9.31 directors and commissioners, and 51.8% of companies have a long-tenured CEO (more than 5 years of service). The average percentage of independent commissioners is 37.665%. A total of 44.7% of companies have a nomination and remuneration committee, and 54.9% hire a Big4 auditor. The average company has total assets of IDR 13,640,000,000,000, leverage of 41.1%, growth of 11.1% and operating cash flow of 7.7% of total assets.

Table 3

Descriptive statistics.

Panel A: Companies with Busy CEOs (N = 477)
Variable	Mean	Median	Minimum	Maximum
ROA	3.569	2.730	-19.600	39.480
ROS	3.189	7.030	-237.860	84.179
ROE	5.490	5.460	-98.610	89.890
TENURE	0.518	1.000	0.000	1.000
RNC	0.447	0.000	0.000	1.000
BOARDSIZE	2.231	2.303	1.386	3.135
INDCOM	37.665	33.333	0.000	80.000
AUDCOM	3.023	3.000	0.000	6.000
BIG4	0.549	1.000	0.000	1.000
GROWTH	0.111	0.051	-0.689	4.369
LEVERAGE	0.411	0.435	0.000	0.919
FIRMSIZE (Rp)	13,640,000,000.000	5,286,000,000.000	28,065,313.000	97,090,000,000.000
FIRMAGE	3.332	3.367	1.609	4.745
OCF	0.077	0.063	-0.157	0.423
Panel B: Companies with Non-Busy CEOs (N = 399)
Variable	Mean	Median	Minimum	Maximum
ROA	4.052	3.170	-19.600	39.480
ROS	5.404	7.440	-237.860	84.179
ROE	5.683	7.180	-98.610	89.890
TENURE	0.444	0.000	0.000	1.000
RNC	0.261	0.000	0.000	1.000
BOARDSIZE	2.112	2.079	1.386	3.332
INDCOM	38.089	33.333	0.000	100.000
AUDCOM	2.880	3.000	0.000	5.000
BIG4	0.366	0.000	0.000	1.000
GROWTH	0.090	0.066	-0.689	4.369
LEVERAGE	0.413	0.435	0.000	0.919
FIRMSIZE (Rp)	6,702,000,000.000	1,700,000,000.000	28,065,313.000	97,090,000,000.000
FIRMAGE	3.440	3.497	1.792	4.745
OCF	0.070	0.059	-0.157	0.423

Notes: This table shows descriptive statistics for all of the variables used in this study. The sample includes 876 firm-year observations from companies listed on the IDX during 2014–2017.

In Panel B, companies with Non-Busy CEOs have mean ROA, ROS and ROE of 4.052%, 5.404% and 5.683%. The average board size is 8.26 directors and commissioners, and 44.4% of companies have a long-tenured CEO (more than 5 years of service). The average percentage of independent commissioners is 39.089%. A total of 26.1% of companies have a nomination and remuneration committee, and 36.6% hire a Big4 auditor. The average company has total assets of IDR 6,702,000,000,000, leverage of 41.3%, growth of 9.0% and operating cash flow of 7.0% of total assets.

Table 4 displays the Pearson correlations. The correlations between BUSYCEO and the firm performance measures, ROA, ROS and ROE, are in the predicted direction, but insignificant. Other correlations between independent variables are generally low and don't raise any multicollinearity issues for our subsequent analysis. Unreported variance inflation factors (VIFs) have an average of 1.40 and a high of 2.40.

Table 4

Pearson correlations.

Variables	[1]	[2]	[3]	[4]	[5]	[6]	[7]	[8]	[9]	[10]	[11]	[12]	[13]	[14]	[15]
[1] ROA	1.000
[2] ROS	0.472∗∗∗	1.000
	(0.000)
[3] ROE	0.849∗∗∗	0.446∗∗∗	1.000
	(0.000)	(0.000)
[4] BUSYCEO	-0.027	-0.028	-0.004	1.000
	(0.422)	(0.400)	(0.898)
[5] TENURE	-0.025	0.053	-0.002	0.074∗∗	1.000
	(0.459)	(0.115)	(0.956)	(0.029)
[6] RNC	0.168∗∗∗	0.105∗∗∗	0.124∗∗∗	0.193∗∗∗	-0.064∗	1.000
	(0.000)	(0.002)	(0.000)	(0.000)	(0.059)
[7] BOARDSIZE	0.233∗∗∗	0.187∗∗∗	0.240∗∗∗	0.167∗∗∗	-0.009	0.322∗∗∗	1.000
	(0.000)	(0.000)	(0.000)	(0.000)	(0.797)	(0.000)
[8] INDCOM	-0.055	-0.035	-0.054	-0.017	0.115∗∗∗	-0.010	0.058∗	1.000
	(0.104)	(0.300)	(0.112)	(0.617)	(0.001)	(0.765)	(0.086)
[9] AUDCOM	0.083∗∗	0.074∗∗	0.080∗∗	0.101∗∗∗	-0.004	0.146∗∗∗	0.236∗∗∗	0.057∗	1.000
	(0.014)	(0.029)	(0.018)	(0.003)	(0.917)	(0.000)	(0.000)	(0.092)
[10] BIG4	0.241∗∗∗	0.127∗∗∗	0.186∗∗∗	0.183∗∗∗	-0.080∗∗	0.216∗∗∗	0.377∗∗∗	-0.009	0.214∗∗∗	1.000
	(0.000)	(0.000)	(0.000)	(0.000)	(0.018)	(0.000)	(0.000)	(0.792)	(0.000)
[11] GROWTH	0.053	0.013	0.075∗∗	0.022	-0.068∗∗	-0.032	-0.041	0.004	0.032	-0.032	1.000
	(0.120)	(0.694)	(0.026)	(0.513)	(0.044)	(0.352)	(0.222)	(0.916)	(0.346)	(0.343)
[12] LEVERAGE	-0.239∗∗∗	0.007	-0.179∗∗∗	-0.004	-0.050	0.058∗	0.009	0.034	-0.035	-0.015	-0.069∗∗	1.000
	(0.000)	(0.841)	(0.000)	(0.912)	(0.142)	(0.084)	(0.797)	(0.308)	(0.297)	(0.655)	(0.041)
[13] FIRMSIZE	0.158∗∗∗	0.174∗∗∗	0.179∗∗∗	0.275∗∗∗	-0.080∗∗	0.375∗∗∗	0.668∗∗∗	0.099∗∗∗	0.247∗∗∗	0.416∗∗∗	-0.014	0.110∗∗∗	1.000
	(0.000)	(0.000)	(0.000)	(0.000)	(0.018)	(0.000)	(0.000)	(0.004)	(0.000)	(0.000)	(0.671)	(0.001)
[14] FIRMAGE	0.129∗∗∗	0.048	0.107∗∗∗	-0.099∗∗∗	0.034	0.008	0.166∗∗∗	-0.014	0.002	0.056∗	-0.074∗∗	-0.025	-0.022	1.000
	(0.000)	(0.158)	(0.001)	(0.003)	(0.319)	(0.813)	(0.000)	(0.689)	(0.955)	(0.099)	(0.029)	(0.460)	(0.514)
[15] OCF	0.547∗∗∗	0.230∗∗∗	0.433∗∗∗	0.039	-0.018	0.205∗∗∗	0.183∗∗∗	-0.054	0.108∗∗∗	0.275∗∗∗	-0.017	-0.072∗∗	0.122∗∗∗	0.008	1.000
	(0.000)	(0.000)	(0.000)	(0.248)	(0.595)	(0.000)	(0.000)	(0.110)	(0.001)	(0.000)	(0.618)	(0.034)	(0.000)	(0.805)

Notes: This table shows Pearson Correlation test results. The sample includes 876 firm-year observations from companies listed on the IDX during 2014–2017. Tests show: ∗t > 1,645, ∗∗t > 1,960, ∗∗∗t > 2,326, significance at 10%, 5% and 1%.

Table 5 shows the results of t-tests and Wilcoxon tests between firms with and without a BUSYCEO. These tests show that firm performance (ROA, ROS and ROE) is generally lower in firms with a busy CEO, however these differences are not statistically significant. The other results indicate that firms with busy CEOs are more likely to have bigger boards, a nomination and remuneration committee, a longer tenured CEO, a bigger audit committee, a Big4 auditor, and are bigger and younger firms. These differences in corporate governance variables are examined in Hypothesis 4.

Table 5

T-tests and Wilcoxon tests.

Variables	BUSYCEO	BUSYCEO	t-value	z-value
	= 1	= 0
	N = 477	N = 399
ROA	3.569	4.052	-0.804	-1.299
ROS	3.189	5.404	-0.842	-0.505
ROE	5.490	5.683	-0.129	-1.189
TENURE	0.518	0.444	2.192∗∗	2.188∗∗
RNC	0.447	0.261	5.804∗∗∗	5.699∗∗∗
BOARDSIZE	2.231	2.112	4.993∗∗∗	5.350∗∗∗
INDCOM	37.665	38.089	-0.501	-0.429
AUDCOM	3.023	2.880	3.002∗∗∗	3.134∗∗∗
BIG4	0.549	0.366	5.505∗∗∗	5.415∗∗∗
GROWTH	0.111	0.090	0.654	-1.087
LEVERAGE	0.411	0.413	-0.110	-0.022
FIRMSIZE	29.184	28.243	8.461∗∗∗	8.579∗∗∗
FIRMAGE	3.332	3.440	-2.953∗∗∗	-2.506∗∗
OCF	0.077	0.070	1.156	0.797

Notes: This table shows the characteristics of companies that have busy and non-busy CEOs. The sample includes 876 firm-year observations from companies listed on the IDX during 2014–2017. The t-test results are displayed with *t > 1,645, **t > 1,960, ***t > 2,326, significance at 10%, 5% and 1%. The Wilcoxon (z-test) result are displayed with *z > 1,640, **z > 1,960, ***z > 2,570, significance at 10%, 5% and 1%.

CEO busyness and firm performance

Table 6 shows the results of model 1. We hypothesize a negative relationship between busy CEOs and firm performance. In the first specification, we find a significant negative relationship between ROA and busy CEOs. The coefficient on BUSYCEO is -1.333 (t = -2.58) and is significant at the 5% level. In the second specification, we also find a significant negative relationship between ROS and busy CEOs. The coefficient is -6.810 (t = -2.38) and is significant at the 5% level. In the third specification, using ROE, the coefficient on BUSYCEO is negative, but not significant (-1.880, t = *1.31). These results for ROA and ROS are consistent with Hypothesis 1 and indicate that firm performance is lower in firms whose CEOs hold two or more outside directorships.

Table 6

CEO busyness and firm performance.

Variables	ROA	ROS	ROE
BUSYCEO	-1.333∗∗	-6.810∗∗	-1.880
	(-2.58)	(-2.38)	(-1.31)
TENURE	-0.121	6.015∗∗	0.604
	(-0.24)	(2.21)	(0.44)
RNC	1.280∗∗	2.484	1.819
	(2.24)	(0.77)	(1.13)
BOARDSIZE	1.048	7.443	4.684∗
	(1.02)	(1.22)	(1.68)
INDCOM	-0.022	-0.253∗∗∗	-0.088
	(-1.10)	(-2.85)	(-1.45)
AUDCOM	0.068	0.692	0.558
	(0.26)	(0.49)	(0.62)
BIG4	0.706	0.840	-0.451
	(1.24)	(0.32)	(-0.28)
GROWTH	1.014	2.112	3.349∗∗
	(1.50)	(0.34)	(2.08)
LEVERAGE	-9.174∗∗∗	4.953	-13.123∗∗∗
	(-6.49)	(0.66)	(-2.63)
FIRMSIZE	0.581∗∗	2.632∗	1.860∗∗
	(2.36)	(1.79)	(2.54)
FIRMAGE	1.325∗∗∗	2.196	2.433∗
	(2.91)	(0.88)	(1.84)
OCF	46.411∗∗∗	83.104∗∗∗	92.640∗∗∗
	(10.19)	(4.13)	(8.88)
CONSTANT	-17.789∗∗∗	-85.135∗∗	-67.782∗∗∗
	(-2.87)	(-2.42)	(-3.75)
Year Dummies	Included	Included	Included
Industry Dummies	Included	Included	Included
R-squared	0.416	0.139	0.297
N	876	876	876

Notes: This table shows the results of OLS regressions between CEO busyness and firm performance. The sample includes 876 firm-year observations from companies listed on the IDX during 2014–2017. Asterisks indicate: ∗t > 1,645, ∗∗t > 1,960, ∗∗∗t > 2,326, significance at 10%, 5% and 1%.

The results of our analysis are consistent with the literature on director busyness (Core et al., 1999; Ferris et al., 2003; Fich and Shivdasani, 2006). In particular, our results show that CEO busyness overshadows any human capital or networking benefits attained from holding directorship positions in other companies. While it is important for CEOs to gain additional expertise and experience from outside sources, and to foster connections with regulators, government officials and other executives (Masulis and Mobbs, 2014; Field et al., 2013; Mendez et al., 2017), this should not be done at the expense of their primary occupation. As CEOs of listed companies, their primary focus should be on the performance of their company. Our results are consistent with CEOs holding two or more other directorships as being distracted or overburdened and not focusing enough of their time and energy on their own company's performance.

Across the three specifications, the results for the control variables are not perfectly consistent, but provide some evidence that firm performance is higher in bigger and older firms, firms with higher operating cashflow and growth, and lower leverage. There is also evidence that firm performance is higher for longer-tenured CEOs, firms with nomination and remuneration committees, and when board size is bigger but the proportion of independent commissioners is lower.

CEO busyness, firm growth and tenure

Table 7 shows the results of models 2 and 3. We hypothesize a negative coefficient on BUSYCEO*GROWTH and a positive coefficient on BUSYCEO*TENURE. The first group of three specifications examine Hypothesis 2. We find that the coefficients on BUSYCEO*GROWTH are -1.532 (t = -1.06), -9.596 (-0.90) and -7.642 (t = -1.93), with the latter one significant at the 10% level. While the results for ROA and ROS are insignificant, the result for ROE provides some support for Hypothesis 2 and indicates that the negative relationship between busy CEOs and firm performance is more pronounced in firms with high growth. This suggests that companies growing faster need the immediate attention of their CEO in their daily operations, and CEOs that have too many outside directorships have a loss of attention that incurs a greater cost on the performance of high growth firms.

Table 7

CEO busyness, firm growth and tenure.

Variables	(1)			(2)
	ROA	ROS	ROE	ROA	ROS	ROE
BUSYCEO*GROWTH	-1.532	-9.596	-7.642∗
	(-1.06)	(-0.90)	(-1.93)
BUSYCEO*TENURE				1.722∗	9.098∗	4.977∗
				(1.79)	(1.90)	(1.96)
BUSYCEO	-1.196∗∗	-5.951∗	-1.196	-2.148∗∗∗	-11.118∗∗∗	-4.237∗∗
	(-2.30)	(-1.87)	(-0.80)	(-2.88)	(-2.65)	(-2.09)
TENURE	-0.098	6.159∗∗	0.719	-1.065	1.033	-2.122
	(-0.19)	(2.29)	(0.53)	(-1.47)	(0.28)	(-1.07)
RNC	1.301∗∗	2.620	1.928	1.324∗∗	2.718	1.947
	(2.30)	(0.84)	(1.20)	(2.34)	(0.85)	(1.22)
BOARDSIZE	0.978	7.006	4.336	1.119	7.819	4.890∗
	(0.96)	(1.15)	(1.57)	(1.09)	(1.28)	(1.75)
INDCOM	-0.020	-0.242∗∗∗	-0.080	-0.018	-0.232∗∗∗	-0.077
	(-1.01)	(-2.76)	(-1.30)	(-0.88)	(-2.66)	(-1.23)
AUDCOM	0.061	0.647	0.522	0.080	0.752	0.591
	(0.23)	(0.45)	(0.57)	(0.31)	(0.54)	(0.66)
BIG4	0.680	0.676	-0.582	0.668	0.635	-0.563
	(1.20)	(0.26)	(-0.36)	(1.16)	(0.24)	(-0.35)
GROWTH	2.097∗	8.892	8.750∗∗	1.005	2.061	3.322∗∗
	(1.75)	(1.20)	(2.39)	(1.50)	(0.34)	(2.07)
LEVERAGE	-9.238∗∗∗	4.554	-13.441∗∗∗	-9.317∗∗∗	4.196	-13.537∗∗∗
	(-6.54)	(0.61)	(-2.69)	(-6.58)	(0.56)	(-2.74)
FIRMSIZE	0.594∗∗	2.718∗	1.929∗∗∗	0.557∗∗	2.505∗	1.791∗∗
	(2.39)	(1.82)	(2.64)	(2.28)	(1.73)	(2.46)
FIRMAGE	1.333∗∗∗	2.246	2.473∗	1.337∗∗∗	2.259	2.468∗
	(2.94)	(0.91)	(1.89)	(2.93)	(0.91)	(1.87)
OCF	46.604∗∗∗	84.317∗∗∗	93.606∗∗∗	46.418∗∗∗	83.145∗∗∗	92.663∗∗∗
	(10.37)	(4.29)	(9.08)	(10.23)	(4.15)	(8.92)
CONSTANT	-18.214∗∗∗	-87.792∗∗	-69.898∗∗∗	-17.104∗∗∗	-81.516∗∗	-65.802∗∗∗
	(-2.90)	(-2.45)	(-3.87)	(-2.79)	(-2.37)	(-3.65)
Year Dummies	Included	Included	Included	Included	Included	Included
Industry Dummies	Included	Included	Included	Included	Included	Included
R-squared	0.417	0.142	0.302	0.418	0.142	0.299
N	876	876	876	876	876	876

Notes: This table shows the results of OLS regressions between CEO busyness and firm performance, with interactions with firm growth and CEO tenure. The sample includes 876 firm-year observations from companies listed on the IDX during 2014–2017. Asterisks indicate: ∗t > 1,645, ∗∗t > 1,960, ∗∗∗t > 2,326, significance at 10%, 5% and 1%.

The second group of specifications examine Hypothesis 3. The results show that the coefficients on BUSYCEO*TENURE are 1.722 (t = 1.79), 9.098 (t = 1.90) and 4.977 (t = 1.96) across the three specifications, all significant at the 10% level. This shows strong support for Hypothesis 3 and indicates that the negative relationship between busy CEOs and firm performance is strongest when CEOs have tenure of less than 5 years. This suggests that new and shorter-tenure CEOs are more easily distracted by their outside directorships, causing lower performance in the firm where they are CEO.

Interestingly, the results also show that the negative effect of busy CEOs on firm performance basically disappears for long-serving CEOs. When the coefficients on BUSYCEO and BUSYCEO*TENURE are added together, the net effect is close to zero in all three specifications. This indicates that long-serving CEOs are better at managing their outside commitments than more recently hired CEOs.

CEO busyness and corporate governance

Table 8 shows the results of model 4. We predict that stronger corporate governance practices will help to alleviate the negative effect of busy CEOs on firm performance, hence a positive coefficient is expected on the BUSYCEO*GOV interaction terms. The three groups of specifications show the results for the GOV variables of BOARDSIZE, INDCOM and RNC. Across the nine specifications, we find no significant positive coefficients. The only significant result is a negative coefficient on BUSYCEO*INDCOM in the ROS specification. This result indicates that a higher proportion of independent commissioners exacerbates the negative relationship between busy CEOs and firm performance.

Table 8

CEO busyness and corporate governance.

Variables	(1)			(2)			(3)
	ROA	ROS	ROE	ROA	ROS	ROE	ROA	ROS	ROE
BUSYCEO*BOARDSIZE	1.208	-0.621	1.664
	(0.81)	(-0.08)	(0.37)
BUSYCEO*INDCOM				-0.031	-0.416∗∗	0.077
				(-0.81)	(-2.33)	(0.63)
BUSYCEO*RNC							0.031	4.606	-3.542
							(0.03)	(0.92)	(-1.21)
BUSYCEO	-3.934	-5.473	-5.464	-0.159	8.915	-4.791	-1.343∗∗	-8.349∗∗	-0.697
	(-1.18)	(-0.28)	(-0.55)	(-0.10)	(1.29)	(-1.02)	(-2.24)	(-2.39)	(-0.43)
TENURE	-0.132	6.021∗∗	0.590	-0.143	5.732∗∗	0.657	-0.121	6.078∗∗	0.556
	(-0.26)	(2.23)	(0.43)	(-0.28)	(2.12)	(0.47)	(-0.24)	(2.24)	(0.41)
RNC	1.244∗∗	2.502	1.770	1.293∗∗	2.665	1.786	1.261	-0.259	3.929∗
	(2.14)	(0.78)	(1.08)	(2.25)	(0.82)	(1.11)	(1.62)	(-0.06)	(1.70)
BOARDSIZE	0.560	7.693	4.012	1.015	7.005	4.765∗	1.048	7.492	4.646∗
	(0.47)	(1.10)	(1.20)	(0.98)	(1.13)	(1.71)	(1.02)	(1.22)	(1.67)
INDCOM	-0.020	-0.254∗∗∗	-0.086	-0.007	-0.060	-0.124	-0.022	-0.253∗∗∗	-0.088
	(-1.02)	(-2.76)	(-1.40)	(-0.33)	(-0.58)	(-1.52)	(-1.10)	(-2.84)	(-1.45)
AUDCOM	0.072	0.690	0.564	0.072	0.743	0.548	0.068	0.696	0.555
	(0.28)	(0.49)	(0.62)	(0.28)	(0.52)	(0.60)	(0.26)	(0.49)	(0.62)
BIG4	0.667	0.860	-0.506	0.715	0.953	-0.472	0.707	0.976	-0.556
	(1.16)	(0.33)	(-0.31)	(1.25)	(0.37)	(-0.29)	(1.23)	(0.37)	(-0.34)
GROWTH	1.038	2.100	3.382∗∗	1.028	2.299	3.315∗∗	1.014	2.075	3.377∗∗
	(1.55)	(0.34)	(2.10)	(1.52)	(0.37)	(2.06)	(1.50)	(0.34)	(2.07)
LEVERAGE	-9.120∗∗∗	4.925	-13.048∗∗∗	-9.197∗∗∗	4.651	-13.067∗∗∗	-9.172∗∗∗	5.245	-13.347∗∗∗
	(-6.44)	(0.65)	(-2.61)	(-6.50)	(0.62)	(-2.63)	(-6.50)	(0.70)	(-2.67)
FIRMSIZE	0.564∗∗	2.640∗	1.838∗∗	0.575∗∗	2.554∗	1.875∗∗	0.580∗∗	2.574∗	1.905∗∗∗
	(2.31)	(1.81)	(2.48)	(2.32)	(1.74)	(2.56)	(2.34)	(1.76)	(2.59)
FIRMAGE	1.317∗∗∗	2.200	2.422∗	1.328∗∗∗	2.244	2.424∗	1.324∗∗∗	2.160	2.460∗
	(2.91)	(0.89)	(1.84)	(2.92)	(0.91)	(1.83)	(2.91)	(0.87)	(1.86)
OCF	46.517∗∗∗	83.050∗∗∗	92.786∗∗∗	46.235∗∗∗	80.751∗∗∗	93.076∗∗∗	46.407∗∗∗	82.567∗∗∗	93.053∗∗∗
	(10.25)	(4.13)	(8.86)	(10.19)	(4.01)	(8.95)	(10.19)	(4.10)	(8.90)
CONSTANT	-16.424∗∗∗	-85.837∗∗	-65.901∗∗∗	-18.111∗∗∗	-89.444∗∗	-66.985∗∗∗	-17.775∗∗∗	-83.085∗∗	-69.359∗∗∗
	(-2.69)	(-2.50)	(-3.45)	(-2.92)	(-2.53)	(-3.74)	(-2.84)	(-2.37)	(-3.78)
Year Dummies	Included	Included	Included	Included	Included	Included	Included	Included	Included
Industry Dummies	Included	Included	Included	Included	Included	Included	Included	Included	Included
R-squared	0.417	0.139	0.297	0.417	0.143	0.297	0.416	0.140	0.298
N	876	876	876	876	876	876	876	876	876

Notes:This table shows the results of OLS regressions between CEO busyness and firm performance, with interactions with corporate governance variables. The sample includes 876 firm-year observations from companies listed on the IDX during 2014–2017. Asterisks indicate: ∗t > 1,645, ∗∗t > 1,960, ∗∗∗t > 2,326, significance at 10%, 5% and 1%.

Overall, we find no support for Hypothesis 4, which indicates that corporate governance has no measurable impact on the negative influence of busy CEOs on firm performance in Indonesia. This result is somewhat disappointing as Indonesia has undertaken a program to improve the corporate governance practices of listed companies over the past decade. Our results suggest that further improvements are needed as the corporate governance variables we test have no measurable effect on the supervision of the CEO. An effective supervisory and enforcement mechanism would be desirable to reduce the ability of CEOs to take on too many responsibilities outside of the firm.

Conclusions

This study investigates whether CEOs with multiple outside directorships are busy and associated with lower firm performance. This is an important current issue in Indonesia as our analysis indicates that more than 50% of CEOs (President Directors) of listed companies on the Indonesian Stock Exchange have multiple appointments in other companies.

We find that busy CEOs are associated with lower firm performance in Indonesia. This result indicates that CEO busyness is a problem, in that CEOs are not effectively devoting sufficient time and energy to their firms. It seems that their focus is being divided across their different positions, resulting in a damaging effect on firm operations. We find some evidence that this is even more or a problem in high growth firms, where the changing conditions facing the firm require the constant and ongoing attention of the CEO.

It is also a more serious issue when a new CEO has been recently hired. We find that the negative effect of busy CEOs on firm performance is more pronounced for low-tenure CEOs. When a new CEO has been hired it takes time for the new CEO to fully understand the operations of the company and its competitive position in the market. If a CEO also holds outside directorships, they may be distracted by their other commitments and therefore not perform to their highest ability. However, this issue disappears as CEO tenure increases. In later years, a CEO will have already built up their knowledge and expertise of the company's operations and its competitive position. Thus, any distraction from outside directorships has a minimal impact on firm performance.

Together, these results indicate that CEO busyness should be restricted in Indonesia. In general, our results show that limiting the outside involvement of the CEO would have a positive impact on firm performance. It is only after CEOs have tenure of 5 years or more in the firm that there is minimal impact from holding multiple outside positions.

Finally, we find that stronger corporate governance practices do not help to alleviate the negative effect of CEO busyness on firm performance. This suggests that boards of directors are either not aware of this issue or have not been successful in implementing practices to offset the negative effect of CEO busyness. Further investigation of corporate governance practices that could effectively mitigate this issue is warranted and would be a good avenue for future research.

Declarations

Author contribution statement

Iman Harymawan, Melinda Ratri: Conceived and designed the experiments; Performed the experiments; Analyzed and interpreted the data; Contributed reagents, materials, analysis tools or data; Wrote the paper.

Mohammad Nasih: Conceived and designed the experiments; Contributed reagents, materials, analysis tools or data.

John Nowland: Analyzed and interpreted the data; Contributed reagents, materials, analysis tools or data; Wrote the paper.

Funding statement

This work was supported by the Universitas Airlangga, Indonesia under Riset Mandat scheme 2018.

Competing interest statement

The authors declare no conflict of interest.

Additional information

No additional information is available for this paper.