Does the Quality of Director Fusion Raise the Risk of Corporate Debt Default?

Abstract

This paper analyzes the impact of the instability brought about by the change of directors on the risk of corporate debt default from the perspective of the fusion of old and new directors. Combining Ab-sorptive Capacity Theory and Embeddedness Theory, on the one hand, analyzes the threshold effect of the hard integration of directors on corporate debt default risk from the proportion of new directors; on the other hand, through the proportion of the number of well-integrated people, and from the perspective of ability-based role matching and cultural-based group matching between new and old directors, it is judging the individual and interactive effects of director soft fusion quality on firm debt default risk. Through the above two perspectives, we comprehensively judge the independent and interactive effects of directors’ smooth integration quality on corporate debt default risk and consolidate. The study found that the proportion of new directors positively correlates with the increase in the risk of corporate debt default. The weakening of the threshold effect shows that the hard integration of the number of new directors alone will reduce instability due to the increase in the number of new directors, thereby reducing the risk of corporate debt default. Regarding the smooth integration of directors, the role matching between old and new directors has a rejuvenating contribution to corporate debt default risk and has a significant threshold effect. At the same time, group matching positively correlates with corporate debt default risk but has no threshold effect. After the interaction between the two, group matching contributes to debt default risk.

1. Introduction

In recent years, the risk of corporate debt default has increased due to the epidemic’s impact. On 20 December 2022, S&P reported that, as of 30 November, the number of global corporate defaults had risen to 77, five more than the total number of insolvencies in the same period in 2021. On March 14 of the same year, S&P Global Ratings believed that the default risk of Chinese companies would remain high in 2022. We can see that the ever-expanding scale of debt defaults has brought considerable challenges to resolving and preventing financial risks in China and the world.

As the “vertex subject” of the company’s internal governance system, the decision-making efficiency and functional stability of the board of directors have great importance to the good internal governance environment of the enterprise [1]. At the same time, the communication and coordination costs caused by the board members and the instability of the scale will affect the supervision and decision-making efficiency of the board of directors and aggravate the investment and financing costs and agency problems [2]. However, the management replacement and recruitment of enterprises are more designed to cope with the uncertainty of the business environment and adapt to the pace of industry development [3]. It is also widely accepted that regular membership of new directors is suitable for both the board and the company [4]. However, when directors are selected and replaced, it is often challenging to consider their ability and coordinate the relationship between them, which can easily lead to the brusque integration of new and old directors. We called it a “two-headed board of directors” problem. It leads to the instability of the internal governance structure of enterprises. For example, in July 2022, the Helen (300201.SZ) triggered a 65-trading-day range amplitude of 173%. Through the case, it is found that the “goose team” is better than the “eagle individual”; that is, the integration quality of new and old directors in the recruitment of board members is often more critical than individual ability. At the same time, according to the absorptive capacity theory and embeddedness theory, role matching based on the knowledgeable ability and group matching based on the social relationship of new and old directors primarily affect the overall rational judgment of the board of directors on corporate investment and financing [5].

To sum up, this paper aims to integrate new and old directors of enterprises and comprehensively judges the impact on corporate debt default risk from the dual perspectives of rigid proportion integration and soft resource integration. Additionally, the threshold level is introduced to fully reveal the “quantitative change” and “qualitative change” of the effect of the integration of new and old directors on the risk of debt default. To start with reducing the uncertainty of internal corporate governance, finding a way to ensure the quality of the integration of new and old directors, and minimizing the risk of debt default caused by the instability of the board of directors.

The main research contributions and innovative work of this paper are as follows: First, expand the uncertainty of corporate governance to the perspective of dynamic integration, and judge the threshold effect on corporate debt default risk from the quality of integration of new and old directors. It is found that relying on the complicated integration of the proportion of new directors can blur the boundaries between new and old directors, stabilizing corporate debt and reducing the risk of default. However, the weak threshold indicates that the effect of complex integration is limited and gradually declining; the impact on debt default risk is slowly weakening. Second, an in-depth and detailed analysis of the smooth integration quality of directors found not only the role matching and group matching between new and old directors but also the heterogeneity effect on corporate debt default risk alone. Additionally, through the interaction between the two, it is found that group matching is more dominant than individual ability roles. In addition, the threshold effect that only exists in role matching confirms the direct and core function of group matching.

The arrangement of the following text is as follows: The second part is a literature review; the third part is a theoretical analysis and research hypothesis; the fourth part expounds on the data source, variable selection, and the basis of the model setting; the fifth part is empirical analysis; and the sixth part is the research conclusion and related recommendations.

2. Literature Review

2.1. Corporate Debt Default

Current research focuses on finding the factors that affect corporate debt default from the external environment and the internal governance of the enterprise. The research on external environmental factors of enterprises mainly focuses on three aspects: First, the uncertainty of the external environment. Delays in the external environment will exacerbate the company’s information asymmetry and the difficulty of external agencies’ supervision of the company, reducing the investment efficiency of the company, which will lead to an increase in the risk of debt default [6]. At the same time, the effective fulfillment of corporate social responsibilities can, to a certain extent, resolve the negative impact of external environmental uncertainty [7]. The second is the uncertainty of external investors. Banks, the most common external investors, will have informal connections in frequent business transactions. Banks will lend a helping hand to closely related enterprises out of their reputation considerations [8]. The third is the uncertainty brought by other enterprises. A significant source of default clustering is companies’ joint forecasts of familiar or related risk factors such as interest rates, stock returns, and GDP growth [9].

The research on analyzing debt default from the perspective of internal corporate governance mainly focuses on two aspects: One is the aspect of corporate internal control. A company with an excellent internal control mechanism can effectively reduce excessive investment, thereby reducing the company’s default risk. For example, Switzer et al. compared and analyzed financial and non-financial companies believing that the ownership structure (institutional equity and internal equity) has a significant impact on the default risk of financial companies and the independence of the board of directors will also increase the risk of corporate debt default. Financial enterprises are just the opposite [2]. Ali et al. believe that companies with better governance among non-financial companies are closely related to lower levels of default risk, and this correlation is more robust among companies with more growth opportunities [10]. The second is in terms of management behavior. The existing literature believes that the social network of executives, as an informal system, can effectively reduce the risk of corporate default [11]. At the same time, the ratio of CEO cash compensation and opportunistic factors of managers will significantly reduce the risk of a corporate debt default [12,13].

2.2. Director Change and Diversity

Regarding management change, the existing literature focuses on successor-executive and outgoing-executive relationships. The increased risk of resignation brought about by the reorganization of the top management team will increase the salary burden of the company and reduce the salary gap within the top management team, which is not conducive to the company’s innovation activities [14]. At the same time, the research on routine and unconventional changes of top executives found that the unconventional changes of top executives will significantly increase the company’s cash holdings as a buffer mechanism for risk aversion, and this phenomenon will continue during the successor’s tenure [15].

With the gradual application of the high-level echelon theory, scholars have begun to focus on analyzing the structure and governance role of the board of directors from the perspective of diversity issues such as director heterogeneity, ability, and culture. The first is the heterogeneity of directors’ powers. Classified by relationship-oriented dimensions (i.e., gender, race, and age) and task-oriented dimensions (i.e., board performance in corporate investment oversight), We found that task-oriented director competency contributes to lower suboptimal investment efficiency [16]. In addition, Adams et al. also introduced the multidimensional concept of the heterogeneity of directors’ abilities, arguing that directors are a collection of multifaceted skills, and the effective integration of different capabilities among directors can help improve corporate performance [17]. The second is the issue of the cultural diversity of directors. For example, the hometown relationship between directors and CEOs will exacerbate corporate agency problems, and the resulting corporate violations will be more difficult to detect through external audits [18,19]. Moreover, the “family relationship” between the CEO and the directors will generate ingroup preference, reducing the communication cost and increasing the enterprise’s agency costs. Meanwhile, the “family relationship” between the CEO and the auditor will damage the independence of the audit and eventually increase financial errors—reported risk [20].

2.3. Nodule

To sum up, although the existing research is relatively wealthy, the following points still need to be improved: First, the current literature affirms the impact of director change on corporate governance. Nonetheless, it ignores the difficulty in matching members caused by the evolution of new and old directors, resulting in instability. It is even more impossible to determine the decision-making mechanism and logic of the influence of the integration quality of new and old directors on enterprise development. Second, according to the board capital theory, the governance role of directors depends on the adaptation of their abilities and functions, as well as the matching of individuals and teams. However, existing studies only statically analyze the role of board diversity and ignore the quality of dynamic integration in special situations, such as replacement, resulting in the neglect of the part of complicated integration of new and old directors in terms of personnel numbers and soft integration of resources in corporate debt governance.

3. Theoretical Analysis and Research Hypothesis

Under the framework of Resource Dependence Theory, an enterprise, as an organization, needs to absorb external resources to promote its development continuously. Enterprises reduce their dependence on the external environment by constantly internalizing favorable external factors, such as human resources and capital support [21]. Based on this, companies will seek new capabilities, culture, and other characteristics by recruiting new directors to maintain competitiveness in a dynamically changing environment. However, will attracting more elites be beneficial to the development of the enterprise? The employment paradox (Hiring Paradox) proposed by Wang & Zatzick provides a basis for this doubt. The employment paradox believes that the new capabilities brought by new employees will create a cognitive distance between new and old employees, causing old employees to resist new employees. Bring about changes to protect their job security and opportunities for advancement [22].

This paper uses the ratio of new directors to all directors to indicate the relative quantity of new features brought by new directors. To some extent, this indicator reflects the “hard integration” between old and new members. Different from the position hierarchy system of managers, the decision-making power of the board of directors needs to be exercised by the group, and the decision-making of the board of directors is the result of a compromise between different opinions among multiple directors [23]. The cognitive distance between the old and new directors will cause the old directors to doubt the business decisions made by the new directors. Old directors may think that the new director’s business decisions could be more conducive to business operations and thus resist the new director’s business suggestions. However, out of consideration for maintaining their status on the board of directors, the old directors may need to change the business suggestions put forward by the new directors. Please implement the great recommendations of the new directors to ensure the functioning of the board of directors is maintained. Under this kind of conflict between the old and the new, it isn’t easy to play the role of the board of directors. On the contrary, managers have the self-interested motivation to pursue their interests. At the same time, they will transfer the company’s wealth to themselves at the expense of shareholders’ interests, increasing business risk [24]. Therefore, the following assumptions are made:

Hypothesis 1 (H1). 

The proportion of new directors has a positive impact on the company’s debt default risk.

Based on the Board Capital Theory, this paper further divides the integration of new and old directors into two aspects: Role matching and group matching between new and old directors for analysis [25].

First, based on the Absorptive Capacity Theory, enterprises will rationally screen potential favorable external resources and judge whether they match and coordinate with the original resources of the enterprise to achieve the business goal of maximizing profits [26]. At the level of the board of directors, when a company recruits a new director, it will evaluate the director’s ability, consider whether the new director’s power is conducive to the company’s business development status and whether it can be integrated with the knowledge of the old director, that is, the director’s role fit of new director and the old director. The better the role fit, the more the new director candidate’s expertise, skills, and capabilities related to performing these functions are in the firm’s interest and can provide the firm with better monitoring or resources, thereby reducing the risk of debt default. At the same time, based on cognitive psychology, the higher the degree of role matching can reduce the mental distance between directors, thereby improving the efficiency of board decision-making and functional stability. Therefore, the following assumptions are made:

Hypothesis 2 (H2). 

The degree of role matching between new and old directors has a negative impact on the risk of corporate debt default.

Secondly, under the analytical framework of Embeddedness Theory, affected and restricted by their environment, the behavior of enterprises will not be entirely rational [27]. From a socioeconomic perspective, new directors may be selected at the board level for reasons such as director reputation and image management skills or career development and social cohesion. All the factors are based on the degree of social-environmental connection among directors [28], that is, the degree of group matching between old and new directors. The closer the social environment connection between old and new directors, that is, the higher the degree of group matching, there will be preferred within the group, which will lead to collusion between directors and damage the independence of directors, which will weaken the director’s ability to monitor the company, increase agency costs, and lead to the corporate value of future cash flow decreases, increasing the risk of a corporate debt default [29].

Hypothesis 3 (H3). 

The degree of group matching between old and new directors positive correlation corporate debt default risk.

At the same time, the higher degree of group matching will make the performance of new directors more controlled and restricted by old directors. Therefore, by weakening the role-matching degree of directors, companies can use the good resources brought by new directors, thereby affecting the negative effect of role-matching between new and old directors on corporate debt default risk. In other words, the more gregarious, the more the ability of the newcomer is suppressed by the old. Therefore, the following assumptions are made:

Hypothesis 4 (H4). 

The interaction term of group matching degree and role matching degree between new and old directors positively impacts corporate debt default risk.

4. Study Design

4.1. Data Source

The personal characteristics of directors and corporate financial data used in this paper are mainly from China Stock Market and Accounting Research Database (CSMAR). The data on corporate debt default risk comes from the CRI (Credit Research Initiative) database of the Risk Management Institute of the National University of Singapore. The specific variable selection process is as follows: (1) This paper selects A-share listed companies from 2010 to 2018 as samples; (2) eliminates financial industry, ST, and samples with missing information; (3) eliminates no changes in new and old directors Sample; (4) and winsorize the continuous variables at the 1% and 99% levels, and finally get 615 samples to form panel data.

4.2. Variable Design

• Corporate debt default risk (CDDR). The data on corporate debt default risk comes from the CRI (Credit Research Initiative) database of the Risk Management Institute of the National University of Singapore. The data on the expected corporate default rate of debt in the next year is selected to analyze the potential default risk of corporates from a prior perspective. The larger the indicator, the greater the risk of corporate debt default.
• Rate of new directors (RND). Learn from the method of Wang and Zatzick. First, the directors are divided into new and old directors using the data of the start and end date of the director’s tenure [22]. Taking 2010 as an example, if the term of office begins in 2010, it will be set as a new director; if the term of office starts before 2010, it will be recognized as an old director, and so on for other years. Then, calculate the ratio of new directors to all directors based on the number of new directors each year, combined with the total number of board members.

To more accurately describe the impact of the proportion of new directors on corporate debt default risk, in further analysis, this paper decomposes the balance of new directors into two levels: the degree of role matching between new and old directors, the degree of group matching between new and old directors.

(1) Degree of role matching (DRM). Based on Elms’ definition of role fit, combined with Absorptive Capacity Theory, this paper defines role fit as the degree to which new directors and old directors have professional knowledge, skills, and capabilities related to the execution of these skills. It is measured by the ability correlation between old and new directors [5].

Unlike the existing literature that focuses on the unilateral analysis of directors’ characteristics, tenure, functions, resources, and other capabilities, this paper regards directors’ capabilities as a multidimensional whole. Multidimensionality has two meanings. Executives, gender, educational background, age, tenure, number of part-time companies, academic experience, overseas background, professional knowledge background, occupational background (including financial background, government background), and other characteristics, the multidimensional capabilities of directors are integrated to study their corporate debt The impact of default risk [17]. Another aspect is to pay attention to the multidimensionality of directors’ professional knowledge background and professional background (service experience); directors have more than one professional knowledge background or service experience. If professional knowledge, environment, and service experience are counted, it will ignore differences between different environments.

Among them, referring to the calculation method of Jaffe, the directors’ tenure, education background, occupational background, and other characteristics are integrated to construct the director’s ability feature vector [30]. First, according to the formula (1), calculate the correlation coefficient of the individual ability feature vector between the new director and the old director, where $a_i=\left(a_{i1},a_{i2},\dots ,a_{i31}\right)$ is the ability feature vector of a single director, including a total of 31 items. For the specific construction method, see

Table 1. Directors’ personal competence characteristics vector.

ai	ai Method of Determining Numerical Values
Part-time Executives	1 = yes, 0 = no
Gender	1 = Male, 0 = Female
Academic qualifications	0 = unknown, 1 = Secondary school and below, 2 = College, 3 = Bachelor, 4 = Master, 5 = PhD, 6 = Other (degree published in other forms, such as honorary doctorate, correspondence, etc.), 7 = MBA/EMBA.
Age	Actual age of directors
Term of office	Counting of directors’ terms by month
Part Time Company	Number of directors holding part-time positions in other companies
Academic Background	Director’s experience in teaching at a university, serving at a research institution, or conducting research at an association. 0 = no academic experience, 1 = one of the above three experiences, 2 = two of the above three experiences, 3 = all three of the above experiences
Overseas Background	Director studied and worked overseas. 0 = no overseas experience, 1 = studied or worked overseas, 2 = studied and worked overseas
Philosophy	Determined according to the professional knowledge background of the director, the classification is based on the 13 majors identified in the Catalogue of Undergraduate Majors in General Higher Education Institutions (2012). Philosophy, economics, law, education, literature, history, science, engineering, agriculture, medicine, management, art, military science. 1 = Philosophy as director’s major, 0 = other as director’s major or unknown
Economics	1 = Director specializing in economics, 0 = Director specializing in other or unknown
Legal Studies	1 = Director’s major is law, 0 = Director’s major is other or unknown
Pedagogy	1 = Director’s major is education, 0 = Director’s major is other or unknown
Literature	1 = Director’s major is literature, 0 = Director’s major is other or unknown
Historiography	1 = Director’s major is history, 0 = Director’s major is other or unknown
Science	1 = Director’s major is science, 0 = Director’s major is other or unknown
Engineering	1 = Director’s major is engineering, 0 = Director’s major is other or unknown
Agronomy	1 = Director’s major is Agronomy, 0 = Director’s major is other or unknown
Medicine	1 = Director’s specialty is medicine, 0 = Director’s specialty is other or unknown
Management	1 = Director’s major is management, 0 = Director’s major is other or unknown
Artistic Studies	1 = Director’s major is art, 0 = Director’s major is other or unknown
Military Science Production	1 = Director’s major is military science, 0 = Director’s major is other or unknown According to CSMAR’s classification of directors’ professional backgrounds: production, R&D, design, human resources, management, marketing, finance, legal, government, finance. 1 = director has a professional background in production, 0 = director’s professional background is other or unknown
R&D	1 = Director has a professional background in R&D, 0 = Director’s professional background is other or unknown
Design	1 = Director has a professional background in design, 0 = Director’s professional background is other or unknown
Human Resources	1 = Director has a professional background in human resources, 0 = Director’s professional background is other or unknown
Management	1 = Director has a professional background in management, 0 = Director’s professional background is other or unknown
Market	1 = Director has a professional background in marketing, 0 = Director’s professional background is other or unknown
Finance	1 = Director has a professional background in finance, 0 = Director’s professional background is other or unknown
Legal	1 = Director has a legal background, 0 = Director has other or unknown professional background
Government Background	Determined by the category of the director’s government service, 0 = director’s professional background is other or unknown, 1 = director has held or is holding a government position in one category of organization, 2 = director has held or is holding a government position in two categories of organization, and so on
Financial Background	According to the classification of financial background by Guotaian (CSMAR): regulatory department, policy bank, commercial bank, insurance company, securities company, fund management company, securities registration and settlement company, futures company, investment bank, trust company, investment management company, exchange company. 0 = director has no financial background, 1 = director has one of the above financial backgrounds, 2 = director has two of the above financial backgrounds 2 = director has two of the above financial backgrounds, and so on.

. Then, the correlation coefficient is averaged to obtain the ability correlation between new and old directors of a specific company within a particular year.

$$DRM=\frac{a_i\times a_j{}^{\prime }}{\sqrt{\left(a_i\times a_i{}^{\prime }\right)\times \left(a_j\times a_j{}^{\prime }\right)}}$$

(1)

(2) Degree of group matching (DGM). Based on Elms’ (2015) definition of group fit, combined with Embeddedness Theory, this paper defines group fit as the closeness of the social environment in which new directors and old directors are connected [5]. The ratio of the same surname between new and old directors is used to measure the degree of group matching. That is, if the company has three new directors in a specific year, and two of them have old directors with the same surname, then divide two by the total number of directors of the company in that year to get the ratio of the same surname between the new and old directors.

3.

Control variables. Referring to the relevant literature the following control variables were introduced: (1) Return on assets (roa, net profit to total assets), (2) gearing ratio (lev, total liabilities to total assets), (3) operating efficiency (trover, ratio of operating income to total assets), and (4) cash flow ratio (cf, cash flow from operating activities to total assets) [20,31].

4.3. Model Setting

This paper establishes the following four regression models based on the above theoretical analysis. Model (2) analyzes the impact of the proportion of new directors on corporate debt default risk. Among them, model (3) focuses on the degree of role matching between new and old directors, that is, the impact of the ability correlation between new and old directors on the risk of corporate debt default; model (4) focuses on the degree of group matching between new and old directors, that is, the same surname between new and old directors The impact of the ratio on the risk of corporate debt default, model (5) focuses on the impact of the interaction term of role matching degree and group matching degree between new and old directors on corporate debt default risk. An individual-year fixed-effects model was used.

$$CDD{R}_{it}={\alpha }_0+{\alpha }_{1}RN{D}_{it}+{\alpha }_{2}le{v}_{it}+{\alpha }_{3}ro{a}_{it}+{\alpha }_{4}c{f}_{it}+{\alpha }_{5}trove{r}_{it}+v_i+{\gamma }_t+{\epsilon }_{it}$$

(2)

$$CDD{R}_{it}={\alpha }_0+{\alpha }_{1}DR{M}_{it}+{\alpha }_{2}le{v}_{it}+{\alpha }_{3}ro{a}_{it}+{\alpha }_{4}c{f}_{it}+{\alpha }_{5}trove{r}_{it}+v_i+{\gamma }_t+{\epsilon }_{it}$$

(3)

$$CDD{R}_{it}={\alpha }_0+{\alpha }_{1}DG{M}_{it}+{\alpha }_{2}le{v}_{it}+{\alpha }_{3}ro{a}_{it}+{\alpha }_{4}c{f}_{it}+{\alpha }_{5}trove{r}_{it}+v_i+{\gamma }_t+{\epsilon }_{it}$$

(4)

$$CDD{R}_{it}={\alpha }_0+{\alpha }_{1}DG{M}_{it}\times DR{M}_{it}+{\alpha }_{2}le{v}_{it}+{\alpha }_{3}ro{a}_{it}+{\alpha }_{4}c{f}_{it}+{\alpha }_{5}trove{r}_{it}+v_i+{\gamma }_t+{\epsilon }_{it}$$

(5)

5. Study Results

5.1. Descriptive Statistics

Table 2. Descriptive statistics.

	Mean	sd	CDDR	RND	DRM	DGM	lev	roa	cf	trover
CDDR	0.0037	0.0042	1
RND	0.0875	0.1140	0.0942 *	1
DRM	0.780	0.0971	−0.0514 *	−0.00140	1
DGM	0.0348	0.0586	0.0523 *	0.0828 *	0.0436 *	1
lev	0.2510	0.1640	0.6461 *	0.0574 *	−0.0857 *	0.00630	1
roa	0.0307	0.0342	−0.3726 *	−0.0775 *	0.0733 *	−0.0406 *	−0.2327 *	1
cf	0.0201	0.0728	−0.1727 *	−0.0269	−0.0226	−0.0286	−0.1609 *	0.2814 *	1
trover	0.3020	0.2540	0.0147	−0.0388 *	0.0505 *	−0.0275	0.0970 *	0.1770 *	0.0493 *	1

Note: * indicates significant at the 5% confidence level.

reports the descriptive statistical analysis results of the main variables’ mean, standard deviation, and correlation coefficient. Judging from the Person correlation coefficient among variables, each variable is correlated with the risk of corporate debt default.

Specifically, the average risk of corporate debt default $\mathrm{CDDR}$ is 0.37%, and the standard deviation is 0.0042, indicating that the expected level of corporate debt default in the next twelve months is generally low. The average value $\mathrm{RND}$ is 8.75%, and the standard deviation is 0.1140, indicating that the overall level of new directors in enterprises is not high, but there are significant differences among different samples. The average value $\mathrm{DRM}$ is 78%, and the standard deviation is 0.0971, indicating that the overall level of correlation between the ability of new directors hired by enterprises and the skills of old directors is relatively high, above 50%. The average value $\mathrm{DGM}$ is 3.48%, indicating that the overall level of the same surname ratio among new and old directors is low, maintaining at about 3%, and the group matching degree is average. This has much to do with the relatively low absolute number of new directors recruited by companies. At the same time, the ratio of new and old directors with the same surname differs significantly in different companies in different years, with a standard deviation of 0.0586 and a difference of about 20% between the maximum and minimum values.

At the same time, according to the correlation coefficient, it is judged that there is no multicollinearity problem among the main variables. Therefore, it can be further analyzed.

5.2. Threshold Regression Results of the Percentage of New Directors and Corporate Debt Default Risk

With the increase in the proportion of new directors, there will be a problem of blurring the boundary between new and old directors, leading to a change in the integration between new and old directors. This phenomenon led to a non-linear change in the impact of the proportion of new directors on the risk of corporate debt default. Therefore, this paper selects the proportion of new directors as the threshold variable to study the staged impact of the ratio of new directors on the enterprise’s debt default risk and uses the panel threshold model proposed by Hansento test the above nonlinear relationship [32]. The standard error of heteroscedasticity serial correlation cross-section correlation robustness is estimated. At the same time, based on model (6), with the proportion of new directors as the threshold variable, 2011–2018 is selected as the sample period to test the staged impact of the ratio of new directors on the enterprise’s debt default risk. The model (6) is as follows:

$$\begin{gathered} CDD{R}_{it}={\alpha }_0+{\alpha }_{1}RN{D}_{it}+{\alpha }_{2}le{v}_{it}+{\alpha }_{3}ro{a}_{it}+{\alpha }_{4}c{f}_{it}+{\alpha }_{5}trove{r}_{it}+ \\ {\alpha }_{6}RN{D}_{it}\times d_1\left(RN{D}_{it}<{\gamma }_1\right)+{\alpha }_{7}RN{D}_{it}\times d_2\left(RN{D}_{it}>{\gamma }_2\right)+v_i+{\gamma }_t+{\epsilon }_{it} \end{gathered}$$

(6)

In panel threshold model regression, the first step is to test whether there is a threshold effect, that is, the number of threshold values, the second step is to estimate the threshold values, and the last step is to evaluate the parameters.

Table 3. Threshold effect self-sampling test of the Percentage of New Directors.

Models	F-Value	p-Value	Number of BS	Threshold Value
				1%	5%	10%
Single Threshold	4.304 **	0.025	1000	6.367	3.428	2.535
Double Threshold	5.163 **	0.044	500	8.077	4.357	3.188
Three-fold threshold	0.000 ***	0.002	500	0.000	0.000	0.000

Note: ** and *** indicate significant at 5% and 1% confidence levels, respectively.

reports the results of the threshold self-sampling test. The results show three significant threshold values for the impact of the proportion of new directors on the enterprise’s debt default risk.

Table 4. Threshold estimates and confidence intervals of the Percentage of New Directors.

	Threshold Estimates	95% Confidence Interval
Single threshold model (g1)	0.083	(0.083, 0.364)
Double threshold model (g2)
Ito1 (g1)	0.231	(0.083, 0.364)
Ito1 (g2)	0.083	(0.083, 0.364)
Triple threshold model (g3)	0.097	(0.091, 0.200)

shows that, although there are three thresholds, the difference between the third threshold and the second threshold is minimal, and the 95% confidence interval of the triple threshold is within the 95% confidence interval of the double threshold, so two thresholds of 0.083 and 0.231 are finally obtained.

According to the above threshold value of the proportion of new directors, the ratio of new directors is divided into three ranges. See

Table 5. Number of companies with different new director share ranges in China, 2011–2018.

RND Interval	2011	2012	2013	2014	2015	2016	2017	2018
[0, 0.083)	136	155	146	96	109	123	121	136
[0.083, 0.231)	83	72	63	81	80	64	86	74
[0.231, 1)	14	6	24	56	44	46	26	23

for the distribution of the percentage of new directors from 2011 to 2018. It can be seen from the results in Table 5 that, between 2011 and 2018, about 50% of the new directors of the company accounted for the first range, and about 30% of the new directors of the company accounted for the second range. The difference between each year was minimal. This impact shows that the average annual recruitment of new directors in Chinese enterprises is low and has been relatively stable in the past eight years.

The threshold regression results are shown in

Table 6. Threshold regression results for the percentage of new directors and corporate debt default risk.

Variables	CDDR
	Threshold Return	FE
RND	0.00244 ***	0.00102 **
	(3.572)	(2.404)
lev	0.0126 ***	0.0132 ***
	(8.910)	(9.980)
roa	−0.0196 ***	−0.0188 ***
	(−6.487)	(−8.534)
cf	−0.00165 ***	−0.00154 **
	(−3.128)	(−2.198)
trover	−0.000369 ***	−0.000497 ***
	(−3.920)	(−3.042)
RND × d1	0.00933 ***
	(3.113)
RND × d2	−0.00151 *
	(−1.912)
Constant	0.000948 *	0.00110 ***
	(1.772)	(2.638)
Observations	1757	3473
Number of groups	233	615

Note: *, ** and *** indicate significant at 10%, 5% and 1% confidence levels, respectively, and values in parentheses are t-statistics.

, and the regression results of the fixed effect model are presented for comparative analysis. The threshold and fixed effect regression results show that the coefficient of the proportion of new directors is positive and significant, at least at 10%, assuming that H1 is valid. This indicates that from the perspective of “hard integration” of new and old directors, the increase in the proportion of new directors will lead to an increase in the risk of corporate debt default. That is to say, the practice of blind recruitment of new directors to enrich the human resources of their management to seek their development often makes it difficult to play its governance role immediately. Still, it will exacerbate the instability of the board of directors, leading to an increased risk of corporate debt default. In addition, threshold regression also shows the three staged effects of the proportion of new directors on corporate debt default, and the coefficients at each stage are 0.01177, 0.00244, and 0.00093, respectively. The coefficients of the three phases are significant, indicating that with the increase in the proportion of new directors, the positive impact of the board of directors on the risk of corporate debt default is significantly weakened due to the participation of new directors.

The positive impact of the proportion of new directors on the risk of corporate debt default is characterized by a gradual weakening. It shows that, when new directors join the enterprise, due to the cognitive distance between the new and old directors, the new directors do not know much about the production and operation status of the enterprise. This problem leads to the new directors’ ability not being fully utilized, and even differences of opinion between the new and old directors, that is, the integration of new and old directors, which will affect the decision-making efficiency of the board of directors. Because the board of directors is a crucial decision-making body for shareholders to prevent the moral hazard of managers, the reduction of the decision-making efficiency of the board of directors will, on the one hand, lead to the intensification of corporate agency problems, which will lead to the instability of corporate cash flow and increase the risk of corporate debt default. However, external investors (such as banks) are not optimistic about the development prospects of enterprises. When banks lend, they are susceptible to changes in risk. Once enterprises have signs of default, they will be constrained by bank loans and bear more significant risks [33]. The loan constraints of banks lead to the increase of external financing costs of enterprises, which further aggravates the risk of corporate debt default. However, with the increasing proportion of new directors, the updating speed of the board of directors has accelerated, the boundary between the new directors and the old directors of the board of directors has gradually blurred, and the integration between the new directors and the old directors has slightly weakened. Thus, the positive impact of the proportion of new directors on the enterprise’s debt default risk has weakened.

6. Further Research

6.1. Role Matching and Threshold Regression Results of Corporate Debt Default Risk

Role matching refers to the degree to which the new and old directors have professional knowledge, skills, and abilities related to implementing these skills. The higher the degree of role matching, the higher the degree of ability integration between new and old directors. On the one hand, the excellent role matching degree shows that the ability of new directors is more consistent with the business development status of the enterprise, can be better commercialized, achieve business value, bring more stable cash flow to the enterprise, and help reduce the risk of debt default of the enterprise. On the other hand, the higher the degree of role matching, the smaller the cognitive distance between the new and old directors, which is more conducive to communication and cooperation between the new directors and the old directors and can improve the decision-making efficiency and functional stability of the board of directors, which can enhance the external image of the enterprise, maintain the reasonable expectations of external investors on the enterprise, reduce the external financing costs of the enterprise, and help reduce the risk of debt default of the enterprise. Thus, will the improvement of communication and cooperation among directors magnify the business value of the new directors’ ability and further improve the role-matching degree to reduce the risk of corporate debt default? Is the impact of role matching on corporate debt default risk necessarily linear? Is there any difference in the stage effect?

The standard error of heteroscedasticity sequence correlation cross-section correlation robustness is estimated. Therefore, this part constructs a panel threshold regression model (7) to test whether the role matching has a nonlinear impact on the enterprise’s debt default risk. The model (7) is as follows:

$$\begin{gathered} CDD{R}_{it}={\alpha }_0+{\alpha }_{1}DR{M}_{it}+{\alpha }_{2}le{v}_{it}+{\alpha }_{3}ro{a}_{it}+{\alpha }_{4}c{f}_{it}+{\alpha }_{5}trove{r}_{it}+{\alpha }_{6}DR{M}_{it}\times d_1\left(DR{M}_{it}<{\gamma }_1\right) \\ +{\alpha }_{7}DR{M}_{it}\times d_2\left(DR{M}_{it}>{\gamma }_2\right)+v_i+{\gamma }_t+{\epsilon }_{it} \end{gathered}$$

(7)

The panel threshold model regression first tests whether the threshold effect exists.

Table 7. Threshold effect self-sampling test of Role Matching.

Models	F-Value	p-Value	Number of BS	Threshold Value
				1%	5%	10%
Single Threshold	0.000 ***	0.000	300	0.000	0.000	0.000
Double Threshold	27.281 ***	0.000	300	16.859	11.265	9.159
Three-fold threshold	0.000 ***	0.005	200	0.000	0.000	0.000

Note: ***indicates significant at the 1% confidence level.

reports the results of the threshold effect self-sampling test, which shows that there are triple thresholds for the impact of role matching on corporate debt default risk. Secondly, as shown in

Table 8. Threshold estimates and confidence intervals of Role Matching.

	Threshold Estimates	95% Confidence Interval
Single threshold model (g1)	0.986	(0.654, 0.986)
Double threshold model (g2)
Ito1 (g1)	0.785	(0.593, 0.986)
Ito1 (g2)	0.782	(0.781, 0.940)
Triple threshold model (g3)	0.785	(0.783, 0.785)

, the threshold values are estimated. The estimation results show that, although there are triple thresholds, the three thresholds are very close, and the 95% confidence interval overlaps. Therefore, the threshold value of 0.785 is finally retained.

Table 9. Threshold regression results of role matching on corporate debt default risk.

Variables	CDDR
	Threshold Return	FE
DRM	−0.00167 ***	−0.00451 ***
	(−3.035)	(−5.132)
lev	0.0122 ***	0.0123 ***
	(7.279)	(8.459)
roa	−0.0174 ***	−0.0178 ***
	(−6.596)	(−7.464)
cf	−0.000680	−0.00111
	(−0.663)	(−1.465)
trover	−0.000345 ***	−0.000489 ***
	(−7.088)	(−3.280)
DRM × d1	−0.00197 **
	(−2.572)
DRMl × d2	−0.00223 ***
	(−2.669)
Constant	0.00420 ***	0.00500 ***
	(6.607)	(7.205)
Observations	2143	2883
Number of groups	349	589

Note: ** and *** indicate significant at 5% and 1% confidence levels, respectively, and the values in parentheses are t-statistics.

reports the results of panel fixed effect and panel threshold regression, both of which show that the degree of role matching has a negative impact on corporate debt default risk, which verifies hypothesis H2. It shows that from the perspective of “soft integration” of new and old directors, considering the matching of the roles of new and old directors, the new board of directors recruited by enterprises can reduce their own debt default risk through the integration of individual capabilities and old directors.

In addition, threshold regression also shows the staged impact of role matching on corporate debt default. The coefficients at each stage are −0.00364 and −0.0039, respectively. The coefficients of the two steps are significant, indicating that, with the increase of role matching between new and old directors, the negative impact of role matching on the expected debt default risk of enterprises has slightly increased. It is widely believed in the industry that the new capabilities, new ideas, new external resources, and so forth, brought about by the new directors’ joining, will play a perfect role in promoting the operation and management of enterprises. At the same time, suitable replacement and integration between new and old directors can enable enterprises to maintain a sensitive response in the changing external environment [4]. That is, with the increasing role matching between new and old directors, the improved communication and cooperation between directors can further amplify the commercial value of the new directors’ ability, thereby further reducing the risk of corporate debt default.

6.2. Threshold Regression Results of Group Matching Corporate Debt Default Risk

Group matching analyzes the integration of new and old directors from a socio-economic perspective. Construct a panel threshold regression model (8) to test their relationship. The model (8) is as follows:

$$\begin{gathered} CDD{R}_{it}={\alpha }_0+{\alpha }_{1}DG{M}_{it}+{\alpha }_{2}le{v}_{it}+{\alpha }_{3}ro{a}_{it}+{\alpha }_{4}c{f}_{it}+{\alpha }_{5}trove{r}_{it}+{\alpha }_{6}DG{M}_{it}\times d_1\left(DG{M}_{it}<{\gamma }_1\right) \\ +{\alpha }_{7}DG{M}_{it}\times d_2\left(DG{M}_{it}>{\gamma }_2\right)+v_i+{\gamma }_t+{\epsilon }_{it} \end{gathered}$$

(8)

The results of the threshold self-sampling inspection are shown in

Table 10. Threshold effect self-sampling test of Group Matching.

Models	F-Value	p-Value	Number of BS	Threshold
				1%	5%	10%
Single Threshold	4.500	0.218	500	16.862	10.147	8.062
Double Threshold	2.909	0.176	500	9.306	5.190	3.982
Three-fold threshold	−3.464	0.612	500	10.541	5.456	3.633

below. The results show no threshold value, indicating that the group matching degree has a more direct and stable impact on corporate debt default. Therefore,

Table 11. Regression results of group matching on firms’ expected debt defaults.

CDDR
Variables	Group Matching
DGM	0.00205 ***
	(5.578)
DGM × DRM
lev	0.0131 ***
	(9.008)
roa	−0.0196 ***
	(−8.309)
cf	−0.00110
	(−1.445)
trover	−0.000545 ***
	(−3.339)
Constant	0.00125 ***
	(2.916)
Observations	2820
Number of groups	581

Note: *** indicate significant at 1% confidence levels, and the values in parentheses are t-statistics.

only reports the fixed effect regression results. The regression results show that the DGM coefficient is significantly positive at 1%, indicating that the impact of group matching on corporate debt default is not different in stages. It shows that the same surname, the “family relationship”, is easy to generate intra-group preference among directors, which will lead to collusion among directors, reduce the independence of directors, and increase the risk of debt default of enterprises. The higher the ratio of new and old directors with the same surname, the higher the group matches and the higher the risk of corporate debt default, which verifies the relevant conclusions of H3.

6.3. The Threshold Regression Results of the Interaction between Role Matching and Group Matching on the Default Risk of Corporate Debt

Based on the above analysis, the impact of role matching on corporate debt default changes periodically. Therefore, a panel threshold regression model (9) is constructed to analyze the effects of interaction items on corporate debt default. The threshold regression chooses 2013 to 2018 as the sample period to retain a sufficient sample size. The model (9) is as follows:

$$\begin{gathered} CDD{R}_{it}={\alpha }_0+{\alpha }_{1}DGM\times DR{M}_{it}+{\alpha }_{2}le{v}_{it}+{\alpha }_{3}ro{a}_{it}+{\alpha }_{4}c{f}_{it}+{\alpha }_{5}trove{r}_{it} \\ +{\alpha }_{6}DGM\times DR{M}_{it}\times d_1\left(DGM\times DR{M}_{it}<{\gamma }_1\right) \\ +{\alpha }_{7}DGM\times DR{M}_{it}\times d_2\left(DGM\times DR{M}_{it}>{\gamma }_2\right)+v_i+{\gamma }_t+{\epsilon }_{it} \end{gathered}$$

(9)

Through inspection, the interaction items shown in

Table 12. Threshold effect self-sampling test of the Interaction between Role Matching and Group Matching.

Models	F-Value	p-Value	Number of BS	Threshold Value
				1%	5%	10%
Single Threshold	3.628 *	0.088	2000	9.351	4.870	3.447
Double Threshold	9.819 **	0.019	1000	13.854	6.710	4.810
Three-fold threshold	−1.852	0.524	500	5.750	3.000	1.373

Note: * and ** indicate significant at 10% and 5% confidence levels, respectively.

and

Table 13. Threshold estimates and confidence intervals of the Interaction between Role Matching and Group Matching.

	Threshold Estimates	95% Confidence Interval
Single threshold model (g1)	0.126	(0.036, 0.188)
Double threshold model (g2)
Ito1 (g1)	0.083	(0.036, 0.188)
Ito1 (g2)	0.092	(0.036, 0.188)
Triple threshold model (g3)	0.126	(0.085, 0.188)

below have double thresholds, which are 0.083 and 0.092, respectively. The two thresholds are close, and the confidence intervals coincide.

Table 14. Number of companies in different interaction intervals, China, 2013–2018.

DGM × DRM Interval	2013	2014	2015	2016	2017	2018
[0, 0.083)	212	212	206	209	204	198
[0.083, 0.092)	6	4	3	3	5	11
[0.092, 1)	22	24	31	28	31	31

shows the number of enterprises in different interaction item ranges. It is found that about 90% of enterprises are in the first range. It shows that, although there are double thresholds for interaction terms, the two thresholds are almost the same, and most samples are in the first interval, so the impact of interaction terms on corporate debt default is linear.

Table 15. Regression results of role-matched and group-matched interaction terms on firms’ expected debt defaults.

Variables	CDDR
	Threshold Return	FE
DGM × DRM	−0.00927	0.00200 ***
	(−1.021)	(3.992)
lev	0.0128 ***	0.0129 ***
	(6.696)	(8.250)
roa	−0.0191 ***	−0.0193 ***
	(−12.46)	(−9.651)
cf	0.00149	−0.00104
	(0.951)	(−1.131)
trover	−0.000411 ***	−0.000470 ***
	(−4.425)	(−3.229)
DGMDRM × d1	0.00818
	(0.894)
DGMDRM × d1	0.0116
	(1.202)
Constant	0.00144 **	0.00133 ***
	(1.972)	(2.747)
Observations	1418	2704
Number of groups	240	577

Note: ** and *** indicate significant at 5% and 1% confidence levels, respectively, and the values in parentheses are t-statistics.

reports the threshold and fixed effect regression results of interaction items of role matching and group matching on corporate debt default risk. The threshold regression results show that the coefficients of interaction terms in different intervals are insignificant, which confirms the previous judgment that the impact of interaction terms on debt default is linear. The fixed effect regression results show that the coefficient is significantly positive at 1%, indicating that the higher the interaction between role matching and group matching between new and old directors, the higher the risk of corporate debt default, which verifies the relevant conclusions of H4. This shows that in the interaction items, the influence of group matching is the central aspect. The restriction and control of the social relationship between directors will limit the practical exertion of directors’ ability and will also cause the enterprise to have no more sensitive response to the complex and changing external business environment, thus increasing the risk of corporate debt violations.

7. Research Conclusions and Insights

7.1. Research Conclusions

Based on the situation that the scale of enterprise debt default is expanding and the amount of default is rising, this paper takes Chinese enterprises as a sample to analyze the impact of the quality of new and old directors’ integration on the debt default risk of Chinese enterprises from the dual perspectives of soft integration and complicated integration, and further consolidate and excavate the conclusions in family enterprises and non-family enterprises. The research finds that the proportion of new directors positively impacts the risk of corporate debt default and has a significantly weak threshold effect. With the increase in the proportion of new directors, their positive impact on corporate debt default risk gradually weakens. Further research shows that role matching between new and old directors has a negative impact on corporate debt default risk, while the threshold effect shows that with the strengthening of role matching, its negative impact on corporate debt default risk increases; group matching between new and old directors has a positive effect on corporate debt default risk, while role matching and group matching still have a positive impact on corporate debt default risk after an interaction, but there is no threshold effect.

In addition, we give a particular explanation for the conclusion of this paper. Role matching with a higher mean value and group matching with a lower mean value have opposite effects. (1) The ability of directors is an essential criterion for enterprises to select new directors; The ability of directors based on education is also the primary recruitment condition of most listed companies. Therefore, new directors of enterprises can often meet these basic ability requirements. This leads to similar results for new and old directors in capacity, and the average value is high. (2) While they have similar abilities, they will pay more attention to corporate governance and high-quality development with the passion and long-term pursuit of new directors. Therefore, when new directors with abilities increase, they will show a governance tendency toward the risk of corporate debt default. (3) If we increase the proportion of new directors without considering their ability, it may cause instability, which is the research content of this hypothesis H1. The weak threshold result of Hypothesis H1 further shows that this role is not static because, when the proportion of new directors continues to surge, its role in triggering debt default risk also weakens at any time. This result can be interpreted jointly with role matching. (4) While the average value of group matching is low, in China and even East Asia, where the relationship culture is more prominent, the concepts of seniority and inferiority will restrict the role of new directors, leading to the effect of “using small to expand”.

7.2. Management Enlightenment and Policy Suggestions

First, pay attention to the uncertainty caused by replacing new and old directors. The addition of new directors will, to a certain extent, increase the cost of communication and cooperation between directors, reduce the stability of the functions of the board of directors, and increase the risk of debt default of enterprises. Therefore, enterprises should thoroughly combine their development reality and characteristics, scientifically conduct the timing of director change, and reduce the possibility of increasing the risk of debt default due to the instability of directors’ decision-making.

Second, attach importance to matching directors’ ability and team background. When selecting appropriate new directors, enterprises should fully consider the integration between new directors and old directors. When selecting and replacing directors, it is necessary to give consideration to the personal ability of directors and the overall team’s cultural background, help balance and integrate the members of the board of directors, and create an environment in which both individual directors and the board of directors can effectively play their roles.