Top Management Team Stability and Corporate Innovation Sustainability

Abstract

In recent years, there has been growing recognition that the stability of the top management team (TMT) significantly impacts the operation and management of companies. However, few studies have focused on the impact of TMT stability on innovation sustainability. Therefore, based on the upper echelon theory and the faultline theory, this paper takes China’s A-share listed companies from 2010 to 2022 as a sample to explore the impact of TMT stability on corporate innovation sustainability, as well as the moderating effect of executive faultlines on this impact. The results indicate that TMT stability is positively correlated with corporate innovation sustainability, whereas the executive faultlines significantly weaken this correlation. The mechanism test reveals that a stable senior management team can reduce an enterprise’s operational risk through the management functions of executives, alleviate the financing constraints serving as a stability signal sent by the company to investors and creditors, and thus promote the sustainability of innovation. Heterogeneity analysis demonstrates that the influence of TMT stability on corporate innovation sustainability is more pronounced in companies with a high percentage of executive shareholdings, non-state ownership, and CEOs possessing technical expertise. This paper combines the overall stability of the executive team with the differentiation of its internal subgroups, broadens the research perspective of the upper echelon theory, and serves as a valuable reference for the development of corporate executive teams.

1. Introduction

In a highly competitive market, enterprises rely heavily on innovation to gain and sustain a competitive advantage. However, if an enterprise successfully innovates at a particular moment, it merely obtains short-term excess profits. Only persistent innovation can help enterprises to achieve sustainable development. Furthermore, due to the irreversibility, uncertainty, and extended return cycle of innovative operations, they will incur substantial losses to the firm if they are discontinued or resumed [1]. Therefore, ensuring innovation sustainability is crucial for companies to acquire a sustained competitive advantage in a dynamic environment.

The existing research investigates the internal and external influencing factors that affect corporate innovation sustainability and concludes that corporate social responsibility disclosure [2], digital technology [3], external trade policies [4], and industrial agglomeration [5] all have a significant impact on enterprises’ sustainable innovation ability. However, the impact of dynamic changes in TMT on corporate innovation sustainability has not been adequately discussed.

As the decision makers and implementers of corporate strategies, executives play a vital role in resource allocation and can determine the direction and intensity of corporate resource investment [6,7]. Only by achieving an optimized combination of human and financial capital can firms’ innovation capability be constantly improved [8]. Scholars have highlighted that the enterprises’ innovation performance is affected by core executives’ overconfidence [9], overseas experience [10], financial background [11], technical experience [12], academic background [13], educational experience [14], diversity of entrepreneurial passion [15], gender diversity [16], and cultural background [17]. However, the preceding studies solely examine the impact of executive demographic features on corporate innovation performance in a static manner, neglecting enterprises’ sustainable innovation ability. Meanwhile, they also lack the exploration of how the dynamic changes in the TMT, i.e., the TMT stability, affect corporate innovation sustainability.

In actuality, a single executive has less influence over the firm than the entire executive team [18]. The quality of the executive team’s decision making is also primarily determined by the team’s stability [19]. Frequent changes in senior management will significantly damage the continuity and stability of business activities, which is detrimental to corporate continuous innovation activities. However, stable executive team members have a strong sense of belonging and cohesion [20,21], which is conducive to improving the quality and efficiency of corporate decision making [22], ensuring that corporate innovation activities proceed without interruption [23]. The current literature indicates that a stable TMT can improve corporate performance and competitive advantage [19,24,25], adhere to long-term value orientation [19], and actively assume social responsibility [26]. Furthermore, a stable TMT will lower firms’ default risk and debt concentration [27,28], allowing them to cope with external uncertainties and receive more funding [18]. Therefore, it is necessary to advance the existing literature and explore the relationship between TMT stability and corporate innovation sustainability, as well as the corresponding influencing mechanisms.

In addition, according to the faultline theory [29], members of the executive team will be notably varied in terms of age, tenure, gender, and work experience, therefore, generating various subgroups. There are faultlines, or separating lines, between the subgroups. Members of the same subgroup identify with each other and communicate regularly, whereas members of separate subgroups differ more and interact less. Subgroups have diverse profit-seeking goals and variances in management conceptions, practices, and cognition levels, posing obstacles to the senior management team’s communication and collaboration. The greater the intensity of the executive faultlines, the more significant the differences between the subgroups and the more serious the differentiation within the executive team, which will seriously impact the executive team’s performance [30]. Based on this, this paper incorporates executive faultlines as a moderating variable and investigates their moderating effect on the relationship between TMT stability and innovation sustainability.

Based on the upper echelon theory and the faultline theory, and using China’s A-share listed companies from 2010 to 2022 as a sample, this paper examines the relationship between TMT stability and corporate innovation sustainability, as well as the moderation effect of executive faultlines. According to the empirical results, TMT stability is positively related to corporate innovation sustainability, while executive faultlines weaken this relationship. This conclusion is still valid after a series of robustness tests. The mechanism test reveals that stable TMT promotes corporate innovation sustainability by reducing operational risk and alleviating financing constraints. Heterogeneity analysis indicates that the impact of TMT stability on innovation sustainability is more pronounced in firms with a high proportion of executive shareholdings, non-state ownership rights, and CEOs’ possessing technical expertise.

The potential contributions of this study are as follows: First, this paper examines the impact of TMT stability on corporate innovation sustainability from a dynamic perspective, enriching the literature on upper echelon theory, which focuses primarily on the influence of executives’ demographic characteristics [31,32,33,34,35] and pays insufficient attention to dynamic changes in the executive team. Second, the existing research focuses on the impact of TMT changes on corporate performance, social responsibility, and default risk [19,26,27] and only addresses team personnel changes, mostly disregarding the differentiation of subgroups within the team. This study compensates for previous studies’ limitations by combining entire TMT changes to internal faultlines, contributing to a deeper understanding of the executive team characteristics. Third, it examines the mechanism of the TMT stability’s impact on innovation sustainability, offering a probable mechanism explanation for the impact of TMT stability on corporate innovation sustainability.

2. Literature Review and Research Hypothesis

2.1. Literature Review

2.1.1. The Consequences of TMT Stability

The upper echelon theory highlights the importance of TMT characteristics in shaping corporate behavior and development [36]. TMT stability represents the cohesion and cooperation of team members, as well as the steadiness of firm operations [20,21]. Current research has produced two opposing opinions on the consequences of TMT stability. The optimistic viewpoint holds that an unstable TMT impedes enterprise development, whereas a stable TMT fosters enterprise growth. Positive perspectives discuss the impacts of key executives and entire team changes, respectively. From the perspective of key executive changes, Fabrizi [37] found that CEO turnover will make enterprises disclose less social responsibility information. Barron et al. [38] and Schepker et al. [39] found that, after the resignation of a CEO, a new successor’s arrival may cause cognitive conflicts among the TMT members, which is not conducive to building team cohesion and thus harms corporate development. New entrants are easily opposed by the firm’s “veterans”, increasing the uncertainty of the firm’s future operations and negatively impacting the firm’s performance [24,25]. Bills et al. [40] stated that the uncertainty generated by CEO departures greatly increases the enterprise audit fees. Disorderly CEO changes will cause investor concerns [41]. From the perspective of the stability of the entire executive team, Crutchley et al. [24] demonstrated that keeping the board stable after listing can significantly improve the enterprise’s success. Agarwal et al. [19] stated that a stable executive team can help the firm to improve its long-term performance and create competitiveness. Zheng and Lin [42] and Huang et al. [26] pointed out that maintaining the TMT’s stability can help to promote corporate social responsibility performance. Liao et al. [27] stated that a steady leadership team can lower the firm’s default risk. Yao et al. [28] found that a stable executive team is conducive to improving corporate debt structure and expanding debt financing options.

The opposing viewpoints argue that maintaining long-term stability in the TMT is detrimental to the enterprise, and replacing senior managers is advantageous to the firm’s growth. The new leaders will bring fresh ideas and experience to the TMT, benefiting the enterprise [43]. For example, Liu [44] found that new CEO succession can improve corporate social responsibility performance. The turnover of key leaders promotes strategic reform in the firm [45]. An overly stable executive team may allow executives to exploit nepotism and accumulated power for their own interests [46].

2.1.2. Factors Influencing Innovation Sustainability

According to the existing research on the factors impacting innovation sustainability, prior R&D investments have a significant impact on both present and future R&D investments [47]. Enterprise strategic orientation has a strong influence on the sustainability of technological innovation [48], and enterprise operational capital and scale are essential to sustainable innovation endeavors [49]. Mukherjee et al. [50] indicated that the innovation sustainability of firms can be considerably influenced by corporation tax reform. Hu et al. [2] stated that corporate social responsibility disclosure will generate information effects and thus foster corporate sustainable innovation. Li et al. [4] stated that firms’ sustainable innovation capacity has always benefited from free trade’s “outbound” effect. Guo et al. [5] discovered that industrial agglomeration negatively influences a firm’s ability to sustain innovation.

Taking an overview of the preceding literature, discussions on the economic consequences of TMT stability mainly focus on corporate performance, with little attention being paid to innovation sustainability. Furthermore, researchers have investigated influencing factors of innovation sustainability from internal and external perspectives but have disregarded the impact of the executive team’s overall characteristics. Thus, there is a gap in the literature examining the connection between TMT and innovation sustainability, allowing for further investigation.

2.2. Research Hypothesis

2.2.1. TMT Stability and Innovation Sustainability

The TMT has a significant impact on the company’s innovation activities. When the firm’s senior leaders undergo significant changes, the strategy and operating attitude may shift, potentially jeopardizing the firm’s previous innovative experience and resources. For the successors, acclimating to the firm’s atmosphere and developing relationships with fellow team members will take time, resulting in insufficient team cohesion and collaboration, thereby hindering the advancement of innovative projects. On the contrary, a stable TMT achieves an organic combination of human and financial capital [8], ensuring that the firm’s innovation activities continue to advance [23].

In general, within a stable executive team, the individual pursuits of the members align closely with the team’s collective interests. Therefore, stable executive teams are more inclined to consider the future when making investment decisions, to commit to the firm’s long-term development [19], and to support enterprises in persisting research and development [51]. First, stable executive team members exhibit high levels of trust, solidarity, and tacit understanding [20,21,22,23,24]. The likelihood of conflicts and disagreements among team members is relatively small, and they can reach a consensus on innovation decisions and put them into practice. Second, a stable executive team can share information, communicate, and work together effectively, which lowers the costs of coordination within the company and enhances the quality and efficiency of decision making on innovation activities [19,22]. This prevents the innovation activities from going blind and lowers the risk of innovation being discontinued. Finally, due to the lagging nature of rewards for innovative activities, only a stable executive team can reap returns on lagging innovation output. Thus, it is difficult for frequently changing management teams to focus on the enterprise’s long-term goals [52], and they turn to pursue personal interests or short-term gains [53,54], resulting in damage to the enterprise’s sustainable innovation ability.

Specifically, a stable TMT can not only manage and control business risks, but also serve as a signal of corporate stability to the outside world. Thus, the TMT stability can enhance the innovation sustainability of the enterprise through “functional mechanism” and “signal mechanism”.

Based on the function mechanism, a stable TMT can mitigate the adverse impact of corporate operational risk on sustainable innovation activities and enhance innovation sustainability. Continuous innovation necessitates that enterprises commit resources and manage risks on an ongoing basis. However, an increase in operational risk may result in a reduction or even discontinuation of innovation investment, threatening the sustainability of corporate innovation. First, an unstable executive team increases the potential risks of enterprise strategic changes and the disruption of current business operations [38], creates uncertainties, and prevents the enterprise from obtaining sufficient financial resources to continue innovative activities. Second, there is insufficient integration among individuals in the unstable executive team. Members share information inefficiently, and the departments or businesses they manage may be incompatible. Therefore, the firm’s decision-making quality and operational efficiency drop considerably [25], exacerbating the firm’s operational risk and performance fluctuations and inhibiting or disrupting corporate innovation initiatives. On the contrary, stable executive team members form specialized human capital due to long-term collaboration and integration. Smooth team communication and high decision-making efficiency can reduce the risks of corporate performance fluctuations and cash flow disruptions, thereby mitigating the risk of interruption of corporate innovation activities internally. In addition, stable executive teams have more stable leadership styles [19]. Executives can capitalize on industry experience and insights to respond to external shocks and resolve operational risk, enhancing the sustainability of corporate innovation.

Based on the signaling mechanism, a stable TMT is a signal that the interests of external investors and creditors can be guaranteed, which can ease corporate financing constraints and promote innovation sustainability. According to recent research, external investors prioritize the size and cohesion of the entire executive team over individual executives [18,55]. A stable TMT signals to the capital market a firm’s stable operations and promising business prospects, thus strengthening its financial reputation [56]. At the same time, this signal also alleviates the information asymmetry between enterprises and financial institutions, lowering enterprise financing costs [28]. Furthermore, a stable TMT can boost the firm’s social capital [18], expand its financing channels, and lessen its financing concentration [28]. However, an unstable TMT increases the uncertainty of the business operations [38], causing the firm to lose the guarantee of future payment to creditors [57], as well as making it harder to pay dividends to shareholders. Frequent changes in corporate executives signal to creditors that the business is unstable and risky, prompting creditors to demand higher debt financing requirements to compensate for the additional risk [58]. Simultaneously, the frequent turnover of executives will enhance investors’ risk perception while decreasing the company’s stock liquidity [59]. Corporate financing constraints will worsen if the finance costs rise substantially and stock liquidity falls. Taken together, a stable TMT signals to creditors and investors that the enterprise is operating stably and in good condition, reducing their risk perception. External fund providers are also more likely to contribute financing to firms with relatively stable and cohesive executive teams [18,60]. Ultimately, the alleviation of financing constraints enables enterprises to invest more funds in innovation activities and enhance the sustainability of innovation.

Based on the above analysis, this paper puts forward the following hypotheses:

Hypothesis 1 (H1):

TMT stability significantly promotes corporate innovation sustainability.

Hypothesis 1a (H1a):

TMT stability promotes corporate innovation sustainability by reducing operational risk.

Hypothesis 1b (H1b):

TMT stability promotes corporate innovation sustainability by alleviating financing constraints.

2.2.2. The Moderating Effect of Executive Faultlines

Executive faultlines imply differentiation and distinctions within the executive team, which may damage its stability and impede its performance.

On the one hand, the executive faultlines undermine the executive team’s stability [61]. The team members will differentiate between people with similar characteristics for interaction and communication, according to social identity and self-categorization. Because of faultlines, team members identify more with their subgroup than other subgroups or the entire team [62]. Therefore, the stronger the executive faultlines are, the clearer the internal divides and group boundaries. Team members from distinct subgroups may be mutually exclusive and find it difficult to collaborate. This situation will dissolve team cohesion, exacerbate interpersonal conflicts, and weaken trust and respect among the members [63,64,65], causing the executive team to split [66,67].

On the other hand, executive faultlines hinder the adequate performance of executive functions. The presence of executive faultlines will limit executive members’ communication and information sharing within each subgroup, impede the necessary exchange of knowledge and information between subgroups, reduce the efficiency of information utilization, and make it impossible for executives to fully grasp the enterprise’s relevant information [64,68]. Furthermore, the stronger the faultlines are, the more difficult it is for executives to establish a consensus on investment projects [69]. Executives from different subgroups form “insiders” and “outsiders”, hindering the coordination and cooperation of team members, leading to cognitive conflicts in the management team and low efficiency in decision making and implementation [67,68,70,71]. Consequently, the faultlines undermine the solidarity and collaboration of executives, which may eventually result in the discontinuation of corporate innovation activities and weaken innovation sustainability. Based on above analysis, this paper puts forward the following hypothesis:

Hypothesis 2 (H2):

Executive faultlines will weaken the relationship between TMT stability and innovation sustainability.

3. Materials and Methods

3.1. Data and Sample

China’s A-share listed firms from 2010 to 2022 were selected as the initial research sample. The initial data used for the computation of TMT stability and executive faultlines come from executive resumes disclosed in the China Stock Market and Accounting Research Database (CSMAR), and the data used for the assessment of innovation sustainability are derived from the WIND database. The remaining control variables are sourced from CSMAR. To mitigate the influence of abnormal data and enhance the comparability of samples, this study processes the data in the following ways: (1) deleting financial firms; (2) removing specially treated firms (firms labeled ST or *ST); and (3) deleting samples with a substantial number of missing values. All continuous variables were winsorized at the upper and lower 1% quantiles to eliminate the influence of extreme values. Finally, a dataset of 30,952 observations was obtained.

3.2. Definition of Key Variables

3.2.1. Independent Variable

Top management team stability (STATMT): Based on the current scenario in China, the members of the executive team in this paper comprise all senior managers, excluding independent directors and supervisors, including the chairman, executive director, CEO, CFO, and other leaders. Drawing on the research of Laakso and Taagepera [72] and Zhu and Zhang [73], TMT stability is measured by the changes in the management team members over the last three years. The reason for choosing three years as the observation window is that the term of senior executives is three years. This paper determines the on-the-job time of each executive in the three-year window period according to the personnel ID and on-the-job status in the executive resume. Then, Formula (1) is used to calculate the stability factor of each executive within the three years, and Formula (2) is used to calculate the STATMT. The method of computation is shown in Formulas (1) and (2), as follows:

$$Z_{i,k}=P{W}_{i,k}\times Y_k\times Pwad{j}_{i,k}=\frac{n_{i,k}}{{\displaystyle {\sum }_{i=1}^m{n}_{i,k}}}\times \frac{y_k}{{\displaystyle {\sum }_{j=1}^p{y}_k}}\times \sqrt{\frac{1}{{\displaystyle {\sum }_{i=1}^{m}P{W_{i,k}}^2}}}$$

(1)

$$STATMT={{\displaystyle \sum Z_i}}^2={\displaystyle \sum ({\displaystyle \sum Z_{i,k}}}{)}^2$$

(2)

In the above formulas, Z_i,k denotes the stability factor of executive i in year k, and PW_i,k is the position weight of executive i in year k. n_i,k is the position assigned to executive i in year k. The position assigned to the chairman or CEO is 2 (n_i,k = 2), and 1 is assigned to other ordinary executives (n_i,k = 1). Y_k is the time weight of the year k, and y_k is the time assignment of the year k. The impact of executive changes on innovation sustainability increases closer to the observation period, so the values assigned to y_k from the current year back to the past two years are 3, 2, and 1, respectively. m is the total number of senior management team members in year k. Pwadj_i,k is the position adjustment value in year k used to eliminate the size impact of the executive team in the different years. The entire TMT stability index (STATMT) is calculated based on the stability factor of each executive using Formula (2), with a value range of (0, 1]. The closer this index is to 1, the higher the level of executive team stability.

3.2.2. Dependent Variable

Innovation sustainability (INNOVS): Since corporate innovation activities include not only R&D expenditures but also human capital development, new technology creation, and transformation, which are recorded as intangible assets in financial statements [74], intangible assets can more comprehensively reflect the enterprise’s innovation activities and innovation capability [75,76] and serve as a proxy for corporate innovation [77,78]. The increase in intangible assets is primarily the outcome of corporate innovation investments. Therefore, referring to the methodology of Ju et al. [1] and Hu et al. [2], the increase in firm intangible assets is utilized to quantify corporate innovation sustainability.

3.2.3. Moderator Variable

Drawing on Van Peteghem et al. [79], this paper selects the following eight items of executive characteristics: director executives (executives who are also directors), new executives, financial experience, gender, age, education, tenure, and shareholding amount as the basis for dividing the faultlines of executives. Subsequently, a cluster analysis was performed on the company’s annual executives based on these eight characteristics to determine whether the executive faultlines existed, and three optimal clustering groups were obtained. Referring to Van Peteghem et al. [79], executive faultlines are measured using faultline strength (FS), faultline distance (FD), and faultline total index (FTI), as indicated in Formulas (3)–(5), as follows:

$$FS=\frac{{\displaystyle \sum _{j=1}^p{\displaystyle \sum _{k=1}^q{N}_{g,k}{(m{x}_{k,j}-m{x}_j)}^2}}}{{\displaystyle \sum _{j=1}^p{\displaystyle \sum _{k=1}^q{\displaystyle \sum _{i=1}^{N_{g,k}}{(m{x}_{i,k,j}-m{x}_j)}^2}}}}$$

(3)

$$FD=\sqrt{{\displaystyle {\sum }_{j=1}^p{(m{x}_{j,1}-m{x}_{j,2})}^2}}$$

(4)

$$FTI=FS\times FD$$

(5)

In Formula (3), g represents a particular grouping approach and equals 3 in this paper, j is the executive characteristic, and p is the number of characteristics used to calculate the faultlines, which is 8 in this paper. k represents a subgroup, q is the number of subgroups with the value of 3 in this paper, and i represents a subgroup member. mx_k,j represents the average value of members in subgroup k on characteristic j, and mx_j represents the average value of all group members on characteristic j. N_g,k represents the number of members in the subgroup k under g grouping methods. The value range of faultline strength (FS) is (0, 1). The larger the value, the greater the faultline strength. In Formula (4), mx_j,1 represents the average value of subgroup 1 members on characteristic j, and mx_j,2 represents the average value of subgroup 2 members on characteristic j. Formula (4) is used to compute the distances between pairs of the 3 groups in sequence, and the faultline distance (FD) is obtained. The larger the value, the greater the faultline distance (FD). Finally, using Equation (5), the faultline total index (FTI) is calculated to assess the overall differentiation among senior executives. The greater these indicators, the more serious the divisions and friction within the executive team.

3.2.4. Control Variables

Concerning existing research practices, we controlled the following factors that may affect corporate innovation sustainability: enterprise age (Age), enterprise size (Size), enterprise debt level (Lev), growth opportunity (Growth), cash flow (Cash), nature of ownership (Soe), board size (Board), number of independent directors (Indepe), duality (Dual), ownership concentration (Top), institutional shareholding (Inst), shareholding status of senior executives (Mshare), average age of senior executives (Mage), and proportion of female senior executives (Gender).

Table 1. Definitions of primary variables.

Variable Name	Symbol	Definition
Top management team stability	STATMT	The result calculated according to Eequation (2)
Innovation sustainability	INNOVS	Incremental value of intangible assets
Faultline strength	FS	The result calculated according to Equation (3)
Faultline distance	FD	The result calculated according to Equation (4)
Faultline total index	FTI	FTI = FS × FD
Enterprise age	Age	Natural logarithm of enterprise age
Enterprise size	Size	Natural logarithm of total assets
Debt level	Lev	Total debts/total assets
Profitability	Roe	The rate of return on net asset
Growth opportunity	Growth	Operating income growth rate
Cash flow	Cash	Net cash flow from operating activities
Nature of ownership	Soe	Take 1 for state-owned enterprises and 0 for non-state-owned enterprises
Board size	Board	Number of directors
Board size	Indepe	Number of independent directors
Duality	Dual	If the chairman concurrently serves as CEO, the value is 1, and 0 if otherwise
Ownership concentration	Top	Shareholding ratio of the first shareholder
Institutional shareholding	Inst	Institutional shareholder shareholding ratio
Shareholding status of senior executives	Mshare	If the senior management team holds shares, the value is 1, and 0 if otherwise
Average age of executives	Mage	Average age of executives
Proportion of female executives	Gender	Number of female executives/total number of executives

shows the definitions of primary variables.

3.3. Empirical Model

Models (6) and (7) were constructed to test Hypotheses H1 and H2, respectively.

$$INNOV{S}_{i,t}={\alpha }_0+{\alpha }_{1}STATM{T}_{i,t-1}+{\alpha }_{2}Contro{l}_{i,t-1}+{\displaystyle \sum Year+{\displaystyle \sum Indu}}+{\epsilon }_{i,t-1}$$

(6)

$$INNOV{S}_{i,t}={\beta }_0+{\beta }_{1}STATM{T}_{i,t-1}+{\beta }_{2}MODERA{T}_{i,t-1}+{\beta }_{3}STATM{T}_{i,t-1}\times MODERA{T}_{i,t-1}+{\beta }_4{\displaystyle \sum Contro{l}_{i,t-1}}+{\displaystyle \sum \mathrm{Year}+{\displaystyle \sum Indu}}+{\sigma }_{i,t-1}$$

(7)

In the above models, INNOVS represents innovation sustainability, and STATMT is TMT stability. MODERAT represents the moderator variables’ faultline strength (FS), faultline distance (FD), and faultline total index (FTI). Control is a set of control variables, Year is a year dummy variable, and Indu is an industry dummy variable. ε and σ are random disturbance terms.

4. Results

4.1. Descriptive Statistics

Table 2. Descriptive statistics.

Variables	N	Mean	S.D.	Min	Max
STATMT	30,952	0.8244	0.1868	0.1087	1.0000
INNOVS	30,952	14.171	3.140	4.104	22.649
FS	30,952	0.484	0.218	0.136	0.907
FD	30,952	0.131	1.427	1.214	1.935
FTI	30,952	0.621	0.231	0.246	0.897
Age	30,952	2.8081	0.4219	1.3863	3.4657
Size	30,952	22.2304	1.2980	19.8141	26.2401
Lev	30,952	0.4316	0.2056	0.0561	0.8998
Roe	30,952	0.059	0.136	−0.692	0.351
Growth	30,952	0.1786	0.4235	−0.5756	2.7267
Cash	30,952	0.0459	0.0697	−0.1639	0.2421
Soe	30,952	0.3541	0.4783	0.0000	1.0000
Board	30,952	2.1273	0.1990	1.6094	2.7081
Indepe	30,952	0.3761	0.0539	0.3333	0.5714
Dual	30,952	0.2713	0.4446	0.0000	1.0000
Top	30,952	3.7167	10.5703	0.0898	54.3000
Inst	30,952	4.9144	15.0726	0.0006	77.0087
Mshare	30,952	0.7630	0.4253	0.0000	1.0000
Mage	30,952	49.264	3.206	35.600	62.880
Gender	30,952	18.965	11.288	0.000	72.220

displays the descriptive statistics of the main variables. TMT stability (STATMT) has a maximum value of 1, a minimum value of 0.405, and an average value of 0.830, indicating that the company’s management team is generally relatively stable, but some companies have experienced substantial changes in senior management. The discrepancy between the maximum and minimum value of innovation sustainability (INNOVS) is rather considerable, showing that certain companies have maintained high innovation sustainability. The maximum and minimum values of executive faultline strength (FS), faultline distance (FD), and faultline total index (FTI) are quite different, indicating that the internal differentiation in the executive teams of some companies is profound. The values of the other variables are within a reasonable range and meet further research needs.

4.2. Benchmark Regression Results

Table 3. Baseline regression results.

Model	(1)	(2)	(3)	(4)
Variables	INNOVS	INNOVS	INNOVS	INNOVS
STATMT	0.468 **	0.431 **	0.425 **	0.392 **
	(2.466)	(2.312)	(2.285)	(2.101)
Age		−3.423 ***	−3.640 ***	−3.651 ***
		(−20.237)	(−20.921)	(−20.976)
Size		1.750 ***	1.740 ***	1.723 ***
		(21.662)	(21.371)	(21.076)
Lev		1.867 ***	1.960 ***	1.973 ***
		(5.767)	(6.054)	(6.091)
Roe		−0.236	−0.255	−0.255
		(−0.815)	(−0.880)	(−0.879)
Growth		−0.107	−0.122	−0.116
		(−1.382)	(−1.573)	(−1.502)
Cash		2.835 ***	2.782 ***	2.785 ***
		(5.201)	(5.111)	(5.116)
Soe		−0.064	−0.001	−0.005
		(−0.296)	(−0.006)	(−0.023)
Board			−0.671 *	−0.695 **
			(−1.914)	(−1.983)
Indepe			1.671	1.601
			(1.573)	(1.506)
Dual			−0.332 ***	−0.329 ***
			(−3.139)	(−3.102)
Top			0.021 **	0.020 **
			(2.401)	(2.381)
Inst			0.026 ***	0.026 ***
			(5.297)	(5.316)
Mshare			0.103	0.104
			(1.293)	(1.303)
Mage				0.041 **
				(2.029)
Gender				−0.003
				(−0.484)
Constant	4.466 ***	−25.650 ***	−24.086 ***	−25.475 ***
	(4.123)	(−12.461)	(−10.716)	(−10.725)
Observations	30,952	30,952	30,952	30,952
Industry FE	YES	YES	YES	YES
Year FE	YES	YES	YES	YES
Within R2	0.276	0.304	0.306	0.306
F	318.68	315.49	274.11	262.05

Note: t-value in parentheses, *** p < 0.01, ** p < 0.05, * p < 0.1.

reports the test results of the relationship between TMT stability and corporate innovation sustainability. The fixed effect (FE) model, which considers year fixed effects, industry fixed effects, and company fixed effects, is utilized for regression analysis. As shown in Table 3, after gradually introducing the control variables, there is a significant positive correlation between TMT stability and innovation sustainability at a level of 5%, with a correlation coefficient of 0.392. The results demonstrate that TMT stability significantly promotes innovation sustainability, and Hypothesis H1 is supported. Simultaneously, the results in Table 3 reveal that the majority of the control variables are significantly correlated with INNOVS, demonstrating that this paper takes control variables into account adequately. For example, there is a significant negative correlation between enterprise age (Age) and INNOVS, illustrating that the longer an enterprise has been in operation, the lower its innovation potential and the more difficult it is to conduct continuous innovation activities. Enterprise size (Size), debt level (Lev), cash flow (Cash), institutional shareholding (Inst), and INNOVS show a significant positive correlation, indicating that the greater the enterprise size and the more financial capital, the more sustainable innovation activities may be carried out. Although the correlation coefficient between Roe, Growth, and INNOVS is negative, it does not show statistical significance. There is a significant negative relationship between board size (Board) and INNOVS, indicating that a larger board of directors is less conducive to the sustainability of corporate innovation. This may be because a large board will reduce the efficiency and quality of decision making, hindering continuous innovation activities.

4.3. Moderation Effect Test

The cross-terms, STATMT × FS, STATMT × FD, and STATMT × FTI, are generated by multiplying the TMT stability (STATMT) and executive faultlines first and subsequently incorporating them into the regression model for analysis. The results are presented in

Table 4. Moderation effect test results.

Model	(1)	(2)	(3)
Variables	INNOVS	INNOVS	INNOVS
STATMT	0.312 ***	0.049	0.132
	(3.873)	(0.180)	(0.518)
STATMT × FS	−0.202 ***
	(−3.245)
FS	0.453
	(−0.610)
STATMT × FD		−0.337 **
		(−2.437)
FD		−0.530
		(−0.450)
STATMT × FTI			−0.356 *
			(−1.686)
FTI			−0.511
			(−0.280)
Age	−3.638 ***	−3.675 ***	−3.666 ***
	(−20.901)	(−21.122)	(−21.070)
Size	1.721 ***	1.725 ***	1.721 ***
	(21.046)	(21.111)	(21.058)
Lev	1.971 ***	2.025 ***	2.006 ***
	(6.084)	(6.251)	(6.192)
Roe	−0.269	−0.247	−0.251
	(−0.929)	(−0.854)	(−0.866)
Growth	−0.117	−0.121	−0.121
	(−1.510)	(−1.563)	(−1.571)
Cash	2.794 ***	2.770 ***	2.772 ***
	(5.133)	(5.093)	(5.093)
Soe	0.011	−0.007	0.002
	(0.052)	(−0.033)	(0.010)
Board	−0.680 *	−0.734 **	−0.723 **
	(−1.940)	(−2.096)	(−2.063)
Indepe	1.560	1.640	1.658
	(1.468)	(1.544)	(1.560)
Dual	−0.331 ***	−0.343 ***	−0.338 ***
	(−3.128)	(−3.240)	(−3.192)
Top	0.020 **	0.019 **	0.020 **
	(2.355)	(2.276)	(2.344)
Inst	0.026 ***	0.025 ***	0.025 ***
	(5.342)	(5.125)	(5.173)
Mshare	0.105	0.098	0.100
	(1.318)	(1.231)	(1.252)
Mage	0.038 *	0.037 *	0.039 *
	(1.864)	(1.831)	(1.954)
Gender	−0.002	−0.005	−0.004
	(−0.468)	(−0.860)	(−0.696)
Constant	−26.014 ***	−25.254 ***	−25.322 ***
	(−10.910)	(−10.615)	(−10.647)
Observations	30,952	30,952	30,952
Industry FE	YES	YES	YES
Year FE	YES	YES	YES
Within R2	0.306	0.307	0.307
F	251.23	237.87	245.48

Note: t-value in parentheses, *** p < 0.01, ** p < 0.05, * p < 0.1.

. Column (1) reports the moderating effect of executive faultline strength on the impact of TMT stability. The cross-term is significantly negative at a level of 1%, indicating that faultline strength weakens the positive relationship between TMT stability and innovation sustainability. Column (2) reports the moderating effect of faultline distance on the relationship between TMT stability and innovation sustainability. The cross-term is significantly negative at a level of 5%, indicating that faultline distance attenuates the positive correlation between TMT stability and innovation sustainability. Column (3) reports the moderating effect of the faultline total index on the impact of TMT stability. The cross-term is significantly negative at the 10% level, indicating that the faultline total index weakens the positive impact of TMT stability on innovation sustainability. The above results indicate that differentiation within the TMT will diminish the promotion effect of TMT stability on innovation sustainability, supporting Hypothesis H2.

4.4. Endogeneity Issues

Considering potential endogeneity issues, this paper employs the Heckman two-stage test method, the propensity matching score method (PSM), and the instrumental variable method to alleviate the endogeneity.

4.4.1. Heckman Two-Stage Regression

To address the issue of sample self-selection, a robustness test on the main effects and moderating effects was performed using the Heckman two-stage test method. In the first phase, a dummy variable of whether the company’s intangible assets increased was employed as the dependent variable, inspired by Lai [80]. The explanatory variable is based on the original model’s control variables plus the infrastructure capital stock (INFRA), which is equal to the natural logarithm of the total scale of the fixed asset investment in the province where the firm is located. Probit regression is performed in this context to obtain the inverse Mills ratio (IMR). In the second phase, the inverse Mills ratio (IMR) is added as a control variable to the original regression model for analysis. The final results reported in

Table 5. Heckman two-stage test.

	(1)	(2)	(3)	(4)
Variables	INNOVS	INNOVS	INNOVS	INNOVS
STATMT	0.419 **	0.333 ***	0.122 *	0.160 *
	(2.235)	(3.932)	(1.780)	(1.689)
STATMT × FS		−0.201 ***
		(−3.227)
FS		−0.145 ***
		(2.605)
STATMT × FD			−0.341 **
			(−2.471)
FD			−0.531
			(−0.451)
STATMT × FTI				−0.360 *
				(−1.774)
FTI				−0.497
				(−0.272)
Age	−3.678 ***	−3.664 ***	−3.709 ***	−3.699 ***
	(−21.017)	(−20.936)	(−21.198)	(−21.135)
Size	1.737 ***	1.735 ***	1.742 ***	1.737 ***
	(21.111)	(21.075)	(21.184)	(21.120)
Lev	1.929 ***	1.929 ***	1.972 ***	1.956 ***
	(5.932)	(5.931)	(6.065)	(6.015)
Roe	−0.264	−0.278	−0.258	−0.261
	(−0.911)	(−0.959)	(−0.893)	(−0.903)
Growth	−0.123	−0.124	−0.130 *	−0.130 *
	(−1.591)	(−1.595)	(−1.673)	(−1.675)
Cash	2.812 ***	2.820 ***	2.804 ***	2.803 ***
	(5.164)	(5.178)	(5.151)	(5.148)
Soe	−0.022	−0.005	−0.027	−0.017
	(−0.099)	(−0.021)	(−0.125)	(−0.076)
Board	−0.672 *	−0.658 *	−0.707 **	−0.697 **
	(−1.915)	(−1.875)	(−2.014)	(−1.986)
Indepe	1.626	1.585	1.672	1.688
	(1.530)	(1.491)	(1.574)	(1.588)
Dual	−0.333 ***	−0.335 ***	−0.348 ***	−0.343 ***
	(−3.141)	(−3.165)	(−3.290)	(−3.239)
Top	0.021 **	0.021 **	0.020 **	0.021 **
	(2.430)	(2.403)	(2.335)	(2.400)
Inst	0.026 ***	0.026 ***	0.025 ***	0.026 ***
	(5.337)	(5.363)	(5.149)	(5.196)
Mshare	0.114	0.115	0.111	0.112
	(1.427)	(1.437)	(1.384)	(1.396)
Mage	0.040 **	0.037 *	0.036 *	0.039 *
	(2.007)	(1.843)	(1.800)	(1.926)
Gender	−0.003	−0.002	−0.005	−0.004
	(−0.481)	(−0.465)	(−0.862)	(−0.696)
IMR	−0.401 *	−0.258 *	−0.330 *	−0.311 *
	(−1.741)	(−1.857)	(−1.832)	(−1.727)
Constant	−26.086 ***	−26.602 ***	−26.003 ***	−26.029 ***
	(−10.823)	(−10.996)	(−10.772)	(−10.787)
Observations	30,952	30,952	30,952	30,952
Industry FE	YES	YES	YES	YES
Year FE	YES	YES	YES	YES
Within R2	0.306	0.306	0.307	0.307
F	237.56	249.14	225.33	296.78

Note: t-value in parentheses, *** p < 0.01, ** p < 0.05, * p < 0.1.

are consistent with the previous results, suggesting that the conclusions of this paper remain valid even after taking sample self-selection into account.

4.4.2. Propensity Matching Score Method (PSM)

First, a dummy variable TMT10 is generated, which takes 1 when the company’s senior management team’s stability exceeds the industry median value and 0 otherwise. The propensity matching score is calculated using the control variables, and the logit model is then applied for 1:1 neighbor matching. Finally, the matching data were re-regressed.

Table 6. Regression results of PSM.

Model	(1)	(2)	(3)	(4)
Variables	INNOVS	INNOVS	INNOVS	INNOVS
STATMT	0.730 ***	0.693 ***	0.695 **	0.798 **
	(3.128)	(3.968)	(2.036)	(2.496)
STATMT × FS		−0.210 ***
		(−2.693)
FS		−0.172 **
		(−2.410)
STATMT × FD			−0.203 **
			(−2.352)
FD			−0.133
			(−0.532)
STATMT × FTI				−0.195 *
				(−1.773)
FTI				−0.307
				(−1.302)
Age	−3.568 ***	−3.556 ***	−3.585 ***	−3.640 ***
	(−17.937)	(−17.873)	(−18.035)	(−18.198)
Size	1.674 ***	1.672 ***	1.677 ***	1.697 ***
	(17.368)	(17.341)	(17.410)	(17.519)
Lev	2.299 ***	2.301 ***	2.362 ***	2.250 ***
	(6.059)	(6.065)	(6.229)	(5.909)
Roe	−0.301	−0.315	−0.289	−0.321
	(−0.870)	(−0.910)	(−0.836)	(−0.929)
Growth	−0.093	−0.094	−0.102	−0.115
	(−0.986)	(−1.002)	(−1.084)	(−1.221)
Cash	2.285 ***	2.285 ***	2.263 ***	2.332 ***
	(3.595)	(3.594)	(3.563)	(3.668)
Soe	0.070	0.088	0.081	0.049
	(0.269)	(0.336)	(0.309)	(0.186)
Board	−0.789 *	−0.776 *	−0.843 **	−0.767 *
	(−1.913)	(−1.883)	(−2.045)	(−1.859)
Indepe	1.446	1.414	1.487	1.586
	(1.157)	(1.132)	(1.191)	(1.270)
Dual	−0.371 ***	−0.374 ***	−0.394 ***	−0.396 ***
	(−3.010)	(−3.030)	(−3.198)	(−3.210)
Top	0.018 *	0.018 *	0.016 *	0.018 *
	(1.806)	(1.810)	(1.658)	(1.833)
Inst	0.024 ***	0.024 ***	0.023 ***	0.023 ***
	(4.312)	(4.314)	(4.146)	(4.213)
Mshare	0.076	0.078	0.070	0.091
	(0.825)	(0.838)	(0.754)	(0.984)
Mage	0.046 *	0.043 *	0.039 *	0.042 *
	(1.958)	(1.814)	(1.674)	(1.772)
Gender	0.001	0.001	−0.001	−0.000
	(0.225)	(0.237)	(−0.208)	(−0.038)
Constant	−24.616 ***	−25.301 ***	−24.785 ***	−25.968 ***
	(−8.657)	(−8.855)	(−8.703)	(−8.999)
Observations	24,138	24,138	24,138	24,138
Industry FE	YES	YES	YES	YES
Year FE	YES	YES	YES	YES
Within R2	0.312	0.312	0.313	0.313
F	218.33	97.55	156.31	187.26

Note: t-value in parentheses, *** p < 0.01, ** p < 0.05, * p < 0.1.

displays the results. The conclusion of this paper remains robust.

4.4.3. Instrumental Variable Method

To address the reverse causality issue between TMT stability and innovation sustainability, we employed the two-stage least-squares method [81] to conduct instrumental variable tests. The industry mean of TMT stability (TMTIV) is used as an instrumental variable that satisfies exogenous requirement. Senior management teams in the same industry may show similar characteristics, so the stability of the senior management team in the same industry is related to the stability of a particular company’s senior management team. However, it is not directly related to the innovation sustainability of a specific company.

Table 7. The results of the instrumental variable method.

Phase	Phase 1	Phase 2
Variables	STATMT	INNOVS
TMTIV	0.980 ***
	(33.528)
STATMT		0.995 **
		(2.398)
Age	−0.0201 ***	0.092 ***
	(−6.8712)	(3.183)
Size	0.002 ***	0.700 ***
	(2.755)	(69.288)
Lev	−0.039 ***	0.018
	(−6.386)	(0.295)
Roe	−0.023 ***	0.015
	(−10.249)	(0.6070)
Growth	0.058 ***	0.095
	(3.733)	(0.625)
Cash	−0.007 ***	0.061 **
	(−2.785)	(2.47)
Soe	0.016 **	0.018
	(2.525)	(0.284)
Board	0.016	0.047
	(0.706)	(0.220)
Dual	0.009 ***	−0.033
	(3.8888)	(−1.402)
Top	−0.000	−0.006 ***
	(−0.671)	(−3.051)
Inst	−0.0000	−0.004 ***
	(−0.176)	(−2.803)
Mshare	−0.002	0.053 **
	(−0.890)	(2.243)
Mage	0.024 *	0.04 *
	(1.718)	(1.941)
Gender	0.004	0.007
	(0.245)	(0.747)
Constant	−0.4962	−194.8538 ***
	(−0.5932)	(−23.8909)
Observations	30,952	30,952
Company FE	YES	YES
Year FE	YES	YES
R2	0.056	0.358
First-stage F-value	121.68

Note: t-value in parentheses, *** p < 0.01, ** p < 0.05, * p < 0.1.

reports the results after introducing instrumental variables, showing that TMT stability is still significantly positively associated with innovation sustainability. There is no weak instrumental variable issue because the F-statistic in the first stage is greater than 10, indicating that the selected instrumental variable is valid.

4.5. Robustness Checks

4.5.1. Replacing the Dependent Variable

The incremental value of R&D expenditure is used as an alternative dependent variable (INNOVS1) to conduct a robustness test.

Table 8. Regression results of replacing the dependent variable.

Model	(1)	(2)	(3)	(4)
Variables	INNOVS1	INNOVS1	INNOVS1	INNOVS1
STATMT	0.303 ***	0.333 ***	0.121 *	0.028
	(4.152)	(3.713)	(1.683)	(1.560)
STATMT × FS		−0.438 ***
		(−2.655)
FS		−0.204
		(−1.379)
STATMT × FD			−0.117 **
			(−2.318)
FD			−0.243
			(−0.776)
STATMT × FTI				−0.014 **
				(−2.023)
FTI				−0.064
				(−0.383)
Age	−0.416 ***	−0.646 ***	−0.657 ***	−0.001
	(−11.781)	(−13.997)	(−14.234)	(−0.077)
Size	0.144 ***	0.676 ***	0.668 ***	0.143 ***
	(19.614)	(31.269)	(30.797)	(19.177)
Lev	−0.125 ***	0.644 ***	0.664 ***	−0.119 ***
	(−4.269)	(7.500)	(7.725)	(−4.022)
Roe	0.220 ***	0.136 *	0.139 *	0.220 ***
	(8.356)	(1.773)	(1.806)	(8.338)
Growth	0.108 ***	−0.039 *	−0.038 *	0.109 ***
	(15.409)	(−1.897)	(−1.828)	(15.438)
Cash	−0.108 **	0.288 **	0.291 **	−0.107 **
	(−2.184)	(1.991)	(2.016)	(−2.166)
Soe	−0.075 ***	0.021	0.017	−0.074***
	(−3.770)	(0.370)	(0.298)	(−3.739)
Board	0.002	−0.248 ***	−0.267 ***	0.002
	(0.057)	(−2.664)	(−2.872)	(0.055)
Indepe	−0.189 *	−0.716 **	−0.722 **	−0.171 *
	(−1.953)	(−2.542)	(−2.561)	(−1.769)
Dual	0.017 *	−0.012	−0.011	0.016 *
	(1.762)	(−0.409)	(−0.393)	(1.684)
Top	−0.000	0.007 ***	0.007 ***	−0.000
	(−0.062)	(3.209)	(3.166)	(−0.102)
Inst	0.001 **	0.004 ***	0.004 ***	0.001 **
	(2.535)	(2.987)	(2.866)	(2.470)
Mshare	−0.001	0.012	0.011	−0.002
	(−0.167)	(0.575)	(0.527)	(−0.222)
Mage	−0.005 ***	0.015 ***	0.016 ***	−0.005 ***
	(−2.746)	(2.886)	(2.904)	(−2.701)
Gender	−0.000	−0.002*	−0.003*	−0.000
	(−0.064)	(−1.645)	(−1.842)	(−0.014)
Constant	−2.223 ***	−11.251 ***	−11.195 ***	−2.246 ***
	(−11.533)	(−17.785)	(−17.726)	(−10.368)
Observations	30,952	30,952	30,952	30,952
Industry FE	YES	YES	YES	3705
Year FE	YES	YES	YES	YES
Within R2	0.049	0.224	0.224	0.050
F	49.85	55.41	52.69	58.44

Note: t-value in parentheses, *** p < 0.01, ** p < 0.05, * p < 0.1.

shows the regression results consistent with the prior conclusions. After replacing the dependent variable, there is still a significant positive relationship between TMT stability and innovation sustainability.

4.5.2. Replacing the Independent Variable

Referring to the research of Crutchley et al. [24] and Zheng and Lin [42], the replacement variable for the independent variable is computed using Equation (8). Members of the executive team remain the same as they were previously.

$$STATMT{1}_{t-1,t}=\frac{M{T}_{t-1}-D{E}_{t-1}}{M{T}_{t-1}}\times \frac{M{T}_t}{M{T}_{t-1}+M{T}_t}+\frac{M{T}_t-I{N}_t}{M{T}_t}\times \frac{M{T}_{t-1}}{M{T}_{t-1}+M{T}_t}$$

(8)

In Equation (8), STATMT1 represents the stability coefficient of the TMT at time t, with a value range of [0, 1]. A higher numerical value indicates a greater degree of stability for the TMT. DE_t−1 denotes the number of executives who were employed at time t−1 but resigned at time t; IN_t represents the number of executives who were not in the firm at time t−1 but newly joined at time t; MT_t−1 signifies the overall number of senior executives in the firm at time t−1; and MT_t is the total number of senior executives at time t. The regression results after replacing the independent variables are shown in

Table 9. Regression results of replacing the independent variable.

Model	(1)	(2)	(3)	(4)
Variables	INNOVS	INNOVS	INNOVS	INNOVS
STATMT1	0.280 ***	0.219 ***	0.132 **	0.132 *
	(3.665)	(2.970)	(2.438)	(1.918)
STATMT1 × FS		−0.020 ***
		(−3.641)
FS		−0.014 **
		(−2.181)
STATMT1 × FD			−0.308 **
			(−2.483)
FD			−0.248
			(−0.237)
STATMT1 × FTI				−0.356 *
				(1.675)
FTI				−0.511
				(−0.280)
Age	−3.646 ***	−3.675 ***	−3.711 ***	3.564 *
	(−20.881)	(−21.187)	(−21.405)	(1.655)
Size	1.659 ***	1.635 ***	1.640 ***	−3.666 ***
	(20.478)	(20.282)	(20.360)	(−21.070)
Lev	1.739 ***	2.121 ***	2.168 ***	1.721 ***
	(5.374)	(6.894)	(7.045)	(21.058)
Roe	−0.450	−0.144 *	−0.146 *	2.006 ***
	(−1.553)	(−1.926)	(−1.954)	(6.192)
Growth	−0.100	2.684 ***	2.663 ***	−0.251
	(−1.287)	(5.014)	(4.976)	(−0.866)
Cash	2.720 ***	−0.688 **	−0.738 **	−0.121
	(4.987)	(−1.967)	(−2.113)	(−1.571)
Soe	0.002	1.577	1.665	2.772 ***
	(0.010)	(1.489)	(1.572)	(5.093)
Board	−0.681 *	−0.336 ***	−0.347 ***	0.002
	(−1.937)	(−3.182)	(−3.288)	(0.010)
Indepe	1.275	0.041	0.023	−0.723 **
	(1.197)	(0.188)	(0.104)	(−2.063)
Dual	−0.336 ***	0.020 **	0.019 **	1.658
	(−3.161)	(2.358)	(2.277)	(1.560)
Top	0.020 **	0.025 ***	0.024 ***	−0.338 ***
	(2.293)	(5.013)	(4.797)	(−3.192)
Inst	0.026 ***	0.037 *	0.036 *	0.020 **
	(5.244)	(1.842)	(1.813)	(2.344)
Mshare	0.095	−0.003	−0.005	0.025 ***
	(1.186)	(−0.511)	(−0.897)	(5.173)
Mage	0.035 *	−3.675 ***	−3.711 ***	0.100
	(1.742)	(−21.187)	(−21.405)	(1.252)
Gender	−0.001	1.635 ***	1.640 ***	0.039 *
	(−0.223)	(20.282)	(20.360)	(1.954)
Constant	−21.812 ***	−23.925 ***	−23.207 ***	−25.322 ***
	(−10.272)	(−10.119)	(−9.834)	(−10.647)
Observations	30,952	30,952	30,952	30,952
Industry FE	YES	YES	YES	YES
Year FE	YES	YES	YES	YES
Within R2	0.301	0.303	0.303	0.307
F	417.69	251.46	252.00	245.48

Note: t-value in parentheses, *** p < 0.01, ** p < 0.05, * p < 0.1.

, which does not change the research conclusion of this paper.

4.6. Mechanism Test

According to the preceding analysis, a stable TMT can effectively reduce the operational risk, mitigate financing constraints, and thereby promote innovation sustainability. Operational risk (RISK) is measured by the firm’s return on total assets’ (ROA) volatility over the past three years [82]. Higher volatility suggests greater insecurity in the firm’s profitability and increased operational risk. The SA index’s absolute value is used to assess the corporate financing constraints. $\mathrm{S}\mathrm{A}=-0.737\times \mathrm{S}\mathrm{i}\mathrm{z}\mathrm{e}+0.043\times {\mathrm{S}\mathrm{i}\mathrm{z}\mathrm{e}}^2-0.04\times \mathrm{A}\mathrm{g}\mathrm{e}$. Size is the natural logarithm of total assets, and Age is the company’s listing time. The greater the absolute value of SA, the tighter the corporate financing constraints.

Table 10. Mechanism test results.

Model	(1)	(2)
Variables	INNOVS	INNOVS
RISK	−0.231 ***
	(−2.711)
SA		−2.479 ***
		(−4.474)
STATMT × RISK	0.026 ***
	(6.895)
STATMT × SA		0.011 ***
		(5.666)
STATMT	0.220 **	0.389 **
	(2.461)	(2.086)
Age	−0.323 **	−3.632 ***
	(−2.178)	(−20.875)
Size	1.127 ***	1.693 ***
	(16.275)	(20.700)
Lev	−1.669 ***	1.913 ***
	(−6.047)	(5.904)
Roe	−0.547 **	−0.274
	(−2.162)	(−0.946)
Growth	−0.300 ***	−0.139 *
	(−4.747)	(−1.800)
Cash	0.687	2.820 ***
	(1.550)	(5.186)
Soe	0.162	−0.704 **
	(0.574)	(−2.010)
Board	1.674 **	1.614
	(1.963)	(1.520)
Indepe	−0.030	−0.319 ***
	(−0.353)	(−3.009)
Dual	−0.278	−0.032
	(−1.586)	(−0.147)
Top	0.014 **	0.021 **
	(2.016)	(2.451)
Inst	0.025 ***	0.028 ***
	(6.212)	(5.706)
Mshare	−0.033 **	0.042 **
	(−1.995)	(2.066)
Mage	−0.022 ***	−0.003
	(−5.110)	(−0.545)
Gender	0.254 **	−0.093
	(2.152)	(−0.954)
Constant	−16.696 ***	−33.526 ***
	(−8.460)	(−11.078)
Observations	30,952	30,952
Industry FE	YES	YES
Year FE	YES	YES
Within R2	0.275	0.306
F	194.56	255.33

Note: t-value in parentheses, *** p < 0.01, ** p < 0.05, * p < 0.1.

displays the results of the mechanism tests. In Column (1), the operational risk is significantly negatively related to innovation sustainability, while the interaction term between TMT stability and operational risk is significantly positive, with a significance of 1%, indicating that TMT stability significantly weakens the negative impact of the operational risk on innovation sustainability. In Column (2), the financing constraints are significantly negatively related to innovation sustainability, while the interaction term between TMT stability and financing constraints is significantly positive, with a significance of 1%, indicating that TMT stability significantly weakens the negative impact of financing constraints on innovation sustainability. The above results indicate that the TMT stability is conducive to reducing the adverse effects of corporate operating risk and financing constraints on innovation sustainability.

4.7. Heterogeneity Analysis

The influence of TMT stability on innovation sustainability may differ due to differences in incentive mechanisms, property rights structures, and critical executive technical expertise. Therefore, this paper explores the heterogeneous impacts of TMT stability on innovation sustainability from executive equity incentives, the nature of enterprise ownership, and whether the CEO possesses technical expertise.

First, as an essential long-term incentive technique, equity incentives can closely tie managers’ interests to the enterprise’s long-term value [83] and significantly limit executives’ opportunistic conduct [84]. Equity incentives can effectively maintain the stability of the executive team and promote the company’s innovation performance [85]. Therefore, compared to companies with low executive ownership, senior management team stability can significantly increase corporate innovation sustainability in companies with high executive ownership. This paper classifies companies with executive shareholding ratios beyond the industry median as the high shareholding group, while the others are categorized as the low shareholding group. The regression analysis results on the two groups shown in Columns (1) and (2) of

Table 11. Heterogeneity test results.

Group	High Equity Incentive—Low Equity Incentive		State-Owned Firms—Non-State-Owned Firms		CEO with Technical Expertise—CEO without Technical Expertise
Model	(1)	(2)	(3)	(4)	(5)	(6)
Variables	INNOVS	INNOVS	INNOVS	INNOVS	INNOVS	INNOVS
STATMT	0.208 ***	0.073 *	0.047 *	0.392 **	0.393 ***	0.115 *
	(3.038)	(1.703)	(1.925)	(2.473)	(3.150)	(1.825)
Age	−5.007 ***	−2.538 ***	−2.112 ***	−4.087 ***	−4.726 ***	−3.374 ***
	(−17.945)	(−7.323)	(−5.558)	(−18.609)	(−11.380)	(−17.141)
Size	1.888 ***	1.422 ***	1.261 ***	1.853 ***	2.540 ***	1.696 ***
	(12.935)	(12.651)	(8.536)	(17.755)	(11.256)	(18.249)
Lev	4.290 ***	−0.104	−0.508	2.689 ***	0.230	1.863 ***
	(7.714)	(−0.234)	(−0.847)	(6.740)	(0.258)	(5.125)
Roe	0.175	−0.571	−1.266 **	0.168	−0.620	−0.351
	(0.369)	(−1.474)	(−2.482)	(0.469)	(−0.863)	(−1.087)
Growth	0.382 ***	−0.456 ***	−0.463 ***	0.047	0.234	−0.202 **
	(2.955)	(−4.566)	(−3.498)	(0.497)	(1.264)	(−2.338)
Cash	2.890 ***	1.942 ***	1.778 **	2.842 ***	3.611 ***	2.352 ***
	(3.275)	(2.695)	(1.962)	(4.157)	(2.694)	(3.892)
Soe	−0.959 **	0.297	/	/	0.441	−0.008
	(−2.528)	(0.999)	/	/	(0.801)	(−0.031)
Board	−0.815	−0.578	−0.949 *	−0.530	−1.853 **	−0.644
	(−1.403)	(−1.232)	(−1.675)	(−1.150)	(−2.138)	(−1.619)
Indepe	0.208	3.649 ***	3.212 **	0.502	−1.673	1.492
	(0.114)	(2.651)	(1.999)	(0.349)	(−0.633)	(1.244)
Dual	−0.359 **	−0.220	−0.183	−0.384 ***	−0.094	−0.371 ***
	(−2.309)	(−1.403)	(−0.859)	(−3.073)	(−0.361)	(−3.109)
Top	0.036 ***	−0.013	0.006	0.020 *	0.013	0.016 *
	(2.646)	(−0.821)	(0.354)	(1.884)	(0.626)	(1.705)
Inst	0.024 ***	0.072 ***	0.062 ***	0.024 ***	0.052 ***	0.022 ***
	(3.334)	(5.767)	(4.326)	(4.203)	(4.327)	(3.998)
Mshare	/	/	0.125	0.077	−0.130	0.130
	/	/	(0.982)	(0.763)	(−0.684)	(1.484)
Mage	−0.013	0.077 ***	0.030	0.040	0.088	0.025
	(−0.393)	(2.681)	(0.800)	(1.608)	(1.558)	(1.084)
Gender	0.022 ***	−0.030 ***	−0.026 ***	0.011 *	0.020	−0.005
	(2.670)	(−3.953)	(−2.691)	(1.675)	(1.438)	(−0.770)
Constant	−22.043 ***	−24.967 ***	−18.275 ***	−26.531 ***	−42.769 ***	−24.199 ***
	(−5.396)	(−7.351)	(−4.421)	(−8.606)	(−6.532)	(−9.046)
Observations	14,424	16,528	10,961	19,991	5538	25,414
Industry FE	YES	YES	YES	YES	YES	YES
Year FE	YES	YES	YES	YES	YES	YES
Within R2	0.319	0.303	0.308	0.311	0.303	0.303
p-Value	0.023 **		0.015 **		0.005 ***
F	132.77	158.41	158.49	167.53	119.22	254.66

Note: t-value in parentheses, *** p < 0.01, ** p < 0.05, * p < 0.1. The p-values represent the significance of the difference in regression coefficients between groups and were obtained by bootstrap sampling 1000 times based on Fisher’s permutation test.

demonstrate that the impact of TMT stability on innovation sustainability is stronger in companies with high executive ownership than in companies with low executive ownership.

Second, state-owned and non-state-owned firms have distinct institutional contexts and strategic priorities [86,87], which may lead to differences in their pursuit of innovation. In addition, state-owned firms have a competitive advantage in the market because they receive more government assistance, but non-state-owned firms face intense competition and must continue to push innovation activities to achieve a competitive advantage. Columns (3) and (4) of Table 11 present the group regression results of state-owned and non-state-owned firms, respectively, suggesting that TMT stability has a stronger positive effect on innovation sustainability in non-state-owned firms than in state-owned firms.

Finally, there is the CEO’s technical expertise scenario. The CEO is at the core of the executive team. CEOs with technical expertise are more likely to appreciate technological innovation, are more willing to implement innovation strategies, track cutting-edge technologies, and continue to expand investment in R&D. Furthermore, CEOs with technical expertise are more confident in the face of uncertainty in innovation, are more receptive to new technologies, new products, and new ideas, and have a higher tolerance for failure in innovation [88]. The sample is separated into two groups based on whether the CEO has technical expertise. Columns (5) and (6) of Table 11 show the group regression results. TMT stability has a stronger positive effect on innovation sustainability in companies where the CEO has technical expertise.

5. Discussion

Based on the upper echelon theory and the faultline theory and using China’s A-share listed companies as a sample from 2010 to 2021, this paper investigates the impact of TMT stability on corporate innovation sustainability from a dynamic perspective and analyzes the moderating effect of executive faultlines. Furthermore, this paper investigates the channel mechanism via which TMT stability influences innovation sustainability, as well as the heterogeneous effects of TMT stability on innovation sustainability. The empirical findings reveal a positive association between TMT stability and corporate innovation sustainability, while executive faultlines significantly inhibit the relationship. These findings demonstrate that a stable senior management team is more concerned with the firm’s long-term interests and is more driven to work for the firm’s innovation activities. However, the promotion effect of TMT stability is hampered by differentiation among subgroups within the management team. Mechanism testing demonstrates that a stable management team can reduce a firm’s operational risk, alleviate financing constraints, and thereby promote the sustainability of innovation. According to the findings of the heterogeneity analysis, the promotion effect of executive team stability on innovation sustainability is more substantial in enterprises with high executive shareholding percentages, non-state-owned property rights, and CEOs with technical expertise. The heterogeneity analysis results demonstrate that for a stable executive team to support a company’s sustainable innovation, suitable incentive mechanisms, a corporate platform, and the leadership of key executives with technical expertise are required.

This study identified a link between TMT stability and corporate innovation sustainability using theoretical analysis and empirical testing. A stable TMT promotes the sustainability of innovation activities and the long-term development of the firm, supporting the view that TMT stability has positive economic impacts [19,24,25]. The existing research has demonstrated that executive team stability has a significant impact on firm performance, social responsibility, default risk, and debt structure [19,26,27,28]. However, more research is needed on the impact of executive team stability on innovation sustainability. The current research primarily discusses the demographic characteristics of executives from a static perspective, such as overconfidence [9], technical experience [12], academic background [13], educational experience [14], gender diversity [16], and other factors on corporate innovation performance, and lack attention to the dynamic changes in the entire TMT.

In reality, compared with individual executives or static characteristics, the impact of the entire dynamic characteristics of the executive team on the enterprise cannot be ignored [18]. The executive team’s stability substantially impacts the quality of the firm’s decision making [19], affecting the firm’s innovation activities. This paper discusses the impact of the stability of TMT on the sustainability of enterprise innovation from a dynamic perspective, compensating for the deficiency that existing studies mainly focus on the influence of executives from a static perspective, such as the demographic characteristics [31,32,33]. Furthermore, the current research on TMT stability primarily considers the changes in team members and ignores the influence of subgroups within the management team [19,26,27]. This paper combines TMT stability with executive team subgroup differentiation and examines the moderating role of executive faultlines on the positive impact of TMT stability, which provides a more comprehensive focus on the impact of executive team characteristics on the enterprise and a deeper understanding of executive teams. In addition to examining the impact of TMT stability on innovation sustainability, we discovered the potential impact mechanisms, i.e., reducing operational risk and alleviating financing constraints. This fills in a gap in the existing literature by providing a mechanism-based explanation for how TMT stability affects innovation sustainability.

6. Conclusions

This study employs Chinese listed firms in a collectivistic cultural scenario as a sample to investigate the influence and mechanism of TMT stability on corporate innovation sustainability, as well as the moderating role of executive faultlines. Overall, the stability of TMT has a significant favorable impact on corporate innovation sustainability. Through the performance of the functions of the senior management team and the transmission of a signal to the outside world, the feasible mechanisms of the promoting effect of TMT stability on innovation sustainability include reducing corporate operational risk and alleviating financing constraints. However, the executive faultlines weaken this promoting effect. Executive team stability has a greater impact on innovation sustainability in enterprises with a higher proportion of executive shareholdings, non-state-owned property rights, and CEOs possessing technical expertise.

The conclusions of this research have significant theoretical and practical implications. Theoretically, the findings reveal that the stability of TMT can promote the enterprises’ sustainable innovation activities by reducing business risks and alleviating financing constraints, whereas the differentiation of subgroups within the executive team weakens this promoting effect. Moreover, the impact of TMT stability on corporate innovation sustainability varies across firms with different levels of managerial incentives, ownership structures, and key executive technical expertise. This work contributes to the studies of the economic consequences of TMT stability by connecting executives’ dynamic traits to the sustainability of enterprise innovation. By combining the analysis of TMT stability with executive faultlines, we gain a better understanding of the dynamic and static characteristics of TMTs, which is valuable in furthering the upper echelon theory.

In practice, this study has important implications for corporate management practice. First, this study demonstrates that a stable top management team enhances a firm’s efforts to sustain innovation. As a result, to accomplish lasting innovation, firms must develop executive teams with realistic structure, orderly transitions, and relative stability. The frequent turnover of senior management members makes it difficult for the team to develop tacit understanding, coordination, and cohesion, as well as to carry out continuous innovation activities. Second, the differentiation of subgroups within the executive team undermines the positive impact of TMT stability. Firms need to enhance communication, trust, and cohesion among management members in order to avoid the emergence of isolated subgroups. When there are several subgroups in an executive team, enterprises should pay attention to the features of each subgroup in order to avoid excessive divergence among the subgroups. Third, the influence of TMT stability on innovation sustainability varies among scenarios. The firms should focus on long-term incentives for executives, provide a platform for them, and acknowledge the value of key leaders with technical expertise. If enterprises want to carry out continuous innovation operations, they should provide longer-term incentives to the senior management team, such as equity incentives, decrease intervention, and fully leverage the leadership position of key executives with technical expertise.

Although this study has relatively important theoretical and practical implications, it still has certain limitations. First, the research sample for this paper is from China, which has a solid collectivistic cultural context. There is a favorable association between executive team stability and the sustainability of corporate innovation. However, executives from non-collectivist countries may prioritize individual value and development. Therefore, sample data from non-collectivist countries may yield different outcomes. Secondly, the impact mechanisms of TMT on innovation sustainability may extend beyond reducing the operational risk and alleviating financing constraints, and there may be other impact mechanisms that this paper overlooks. Despite its limitations, this study suggests several potential avenues for future research. Firstly, future research can explore the impact of the stability of TMTs in non-collectivist countries on the sustainability of enterprise innovation, which may lead to conclusions that are different from those drawn in this work. Comparing the impact of TMT stability across cultures will help to deepen our understanding of TMTs. Secondly, the impact mechanisms of TMTs on corporate innovation sustainability may be multiple, and future research can further investigate other impact mechanisms of stable TMTs on innovation sustainability. Thirdly, the relationship network and communication mechanisms among subgroups within the executive team may become a future research topic.