CC BY-NC-ND
0
DOI: https://doLorg/10.17323/j.jcfr.2073-0438.18.L2024.20-36 JEL classification: G34, G38, H70, O32, O38
fee]
State Ownership Heterogeneity and Corporate Innovation: New Evidence from a Hierarchical Perspective
Tingqian Pu
PhD Candidate, School of Management, Universiti Sains Malaysia, Malaysia, tq_pu@outlook.com, ORCID
Abdul Hadi Zulkafli a
PhD, Associate Professor, School of Management, Universiti Sains Malaysia, Malaysia, hadi_zml@usm.my, ORCID
Unlike prior research, this study re-examines the relationship between state ownership and corporate innovation from a hierarchical perspective. Drawing upon institutional theory, our findings reveal the heterogeneous impact of state ownership, elucidating the positive role of central state ownership in fostering corporate innovation, while highlighting the inhibitory effect of local state ownership. This conclusion withstands rigorous scrutiny through a battery of robustness checks. Mechanism analysis indicates that central state-owned enterprises stimulate innovation by increasing innovation investment and enhancing efficiency, whereas local state-owned enterprises create obstacles for both innovation investment and efficiency. Our paper offers a hierarchical interpretation of the mixed evidence regarding the relationship between state ownership and corporate innovation. Whether state ownership serves as a facilitator or a hindrance to innovation depends on whether central or local state-owned enterprises dominate the national innovation process. Overall, this study offers new insights into the complex effects of state ownership heterogeneity on corporate innovation activities in emerging economies like China, advancing our understanding of the subtle relationship between corporate governance and innovation.
Keywords: state ownership heterogeneity, corporate innovation, central state ownership, local state ownership, institutional theory, China
For citation: Pu T., Zulkafli A. H. (2024) State Ownership Heterogeneity and Corporate Innovation: New Evidence from a Hierarchical Perspective. Journal of Corporate Finance Research. 18(1): 20-36. https://doi.org/10.17323Aj.jcfr.2073-0438.18.1.2024.20-36
The journal is an open access journal which means that everybody can read, download, copy, distribute, print, search, or link to the full texts of these articles in accordance with CC Licence type: Attribution 4.0 International (CC BY 4.0 http://creativecommons.org/licenses/by/4.0/).
Abstract
Introduction
Corporate innovation plays a key role in economic growth and enhances competitive advantage [1]. However, due to high institutional divergence, many emerging countries are not easily described by models established in developed countries [2]. For instance, state ownership in China is controlled by the central and local levels of government (Figure 1), so whether different hierarchies of state ownership have different impacts on corporate innovation remains an unsolved puzzle.
Existing literature suggests that state ownership may be a potential factor influencing a firm's innovation activities [2-4]. However, the understanding of how state ownership affects corporate innovation remains limited and marked by conflicting findings. For example, J. Yi et al. [5] assert that state-owned enterprises (SOEs) exhibit a proactive stance in innovation, insofar as governments, as shareholders, exert institutional pressures mandating compliance with regulations and alignment with government objectives [3; 6]. They argue that the efficiency and innovation potential of firms hinge significantly on the quality of the institutional framework [7]. In contrast, H. Kou and K. Kroll [8] establish a negative relationship between state ownership and corporate innovation, attributing this to self-interest-driven SOE managers pursuing goals misaligned with corporate performance [9]. Unless robust corporate governance mechanisms are in place, rent-seeking behaviours prevail in SOEs [10].
These mixed results may overlook the hierarchical dynamics within SOEs. Specifically, institutional pressures and innovation incentives vary between central and local SOEs. Central SOEs operate under the control of the central government [11]. The central government is responsible for setting and allocating tasks for central SOEs, with long-term sustainable economic growth strategy being a key objective [12]. In contrast, local governments have
shorter evaluation periods for performance and heavily rely on quantifiable performance indicators [8; 12]. Therefore, they often lean towards short-term economic growth strategies, while local SOE managers are more inclined to seek promotion rewards within the local political ladder and thus consistency with the local government.
Utilizing data from Chinese listed firms spanning the period 2012-2021, this study employs separate dummy variables to delineate the hierarchy of SOEs at the central and local levels. We provide compelling evidence that the inclination towards innovation is stratified by state ownership: central SOEs tend to foster innovation, whereas local SOEs tend to impede it. In the mechanism analysis, local SOEs hamper innovation by curtailing R&D investment and diminishing innovation efficiency, while central SOEs stimulate innovation primarily by facilitating increased R&D investment and improving innovation efficiency.
Our study makes several contributions. First, we trace how the different hierarchies of state ownership relate to corporate innovation, which goes beyond previous research that focused solely on the relationship between total state ownership and corporate innovation [2; 3; 5]. By examining the nonconformity between the influence of evaluation mechanisms and the hierarchical structure of state-owned enterprises, we can gain a better understanding of how state ownership heterogeneity influences innovation activities.
Second, our research contributes to the ongoing debate on the relationship between state ownership and corporate innovation. Previous studies have yielded mixed results on this issue, partly due to a lack of understanding of the heterogeneity of state ownership. By dividing state ownership into central and local categories, this study innovatively explains the discrepancies observed in prior studies, offering a comprehensive perspective from different levels of state ownership.
Figure 1. Percentage of state-controlled entities among Chinese A-share listed firms (2012-2021)
40
30
20
10
2012 2013 2014 2015 2016 2017 2018 2019 2020 2021
SO
SO L
SO C
Note: SO, SO_C, and SO_L refer, respectively, to the percentage of firms ultimately controlled by total government entities, central government entities, and local government entities. Source: prepared by authors.
Third, the design of government systems emerges as a pivotal factor for both transition countries and developing markets [13]. Previous research has seldom addressed the differences in the propensity to innovate between central and local SOEs from an institutional perspective. This study offers new insights in this domain, serving as an essential reference for developing countries aspiring to emulate the Chinese innovation system and transition towards an innovative economy.
The subsequent sections unfold as follows: the second part describes the theoretical foundations and hypothesis development; the third part sets out the study methodology; the fourth part analyses findings, makes robustness checks and further studies the influence mechanism; and the final part summarizes the conclusions.
Theoretical Foundations and Hypothesis Development
Institutional Theory
Institutional theory focuses on the interactions between institutions and organisations, emphasising that the behaviour of a firm is significantly shaped by the institutional environment in which it operates [14; 15]. Such institutions consist of societal, economic, and political organizations, as well as informal social norms and rules [5; 7]. Companies must adjust to diverse institutional constraints to obtain essential resources and support due to institutional pressures [16; 17].
Owing to path dependence, one of the noticeable aspects of Chinese SOEs is extensive government intervention, which comprises ownership control and personnel management systems [18; 19]. It is widely believed that SOEs enjoy privileges granted by the government and related agencies. According to Y. Liu et al. [20], SOEs in emerging markets often obtain financial support and other resources from the government. Nonetheless, the connection between government and state ownership results in institutional pressures that force SOEs to use resources in accordance with government-set public objectives, for instance, economic growth and national innovation strategies [17; 19].
Hypothesis Development
In emerging markets, distinctive institutional factors wield considerable influence over a firm's impetus and capacity for innovation investment [21; 22]. Within the framework of institutional theory, the role of ownership in corporate governance necessitates the consideration of institutional factors [9; 16]. The process of corporate innovation is perceived as the dynamic accumulation of learning and innovation, intricately entwined with the national economic structure and institutional milieu [23]. Within this trajectory, corporate conduct is frequently moulded by prevailing organisational norms and rules [14].
Diverging from most developed countries, the managerial cadre of Chinese state-owned enterprises (SOEs) typically comprises bureaucrats rather than entrepreneurs [24; 25].
This unique group bears a resemblance to formal government officials [5; 10; 16]. Significantly, within the Chinese institutional context, the evaluation mechanisms of central SOEs and local SOEs exhibit heterogeneity. Central government departments oversee central SOEs, whereas local governments, as the de facto controllers of local SOEs, dictate personnel decisions - such as appointments, transfers, and dismissals of top executives - bypassing market-oriented processes [5; 11].
Prior investigations have affirmed that managers of state-owned enterprises, serving as agents of government shareholders, are driven by political motivations to secure promotions to higher positions [26]. This political motivation transcends mere monetary compensation considerations [27]. However, the hierarchy of state ownership engenders substantial divergence in the political objectives of SOEs. The central government in China is inclined toward adopting long-term strategies to foster innovation and industrial upgrading, aiming to bolster the international competitiveness of Chinese firms. For instance, the 2006 "National Medium and Long-Term Program for Science and Technology Development (2006-2020)" outlined a 15-year government-led strategy for technological innovation, incorporating innovation subsidies, information, and technological support, as well as tax reductions and policy incentives linked to technology. Consequently, potential political motivations impel central SOEs to augment research and development (R&D) investments. Simultaneously, subsidies, tax reductions, and policy incentives hinge largely on firms' innovation achievements, intensifying the impetus for corporate innovation [28]. These advantages also streamline the firms' innovation processes, consequently enhancing innovation efficiency. Hence, central state ownership profoundly fosters corporate innovation, particularly through heightened inputs and efficiency. Conversely, the advancement of local government officials predominantly hinges on short-term economic growth within their regions and individual accomplishments. The divergence in political objectives underscores substantial disparities in the strategic approach to corporate innovation between central and local SOEs. Given the proclivity of local governments toward GDP-centric goals, they are more prone to steer state-owned enterprises toward investing in fixed assets, concurrently curbing long-term R&D investments fraught with greater uncertainty and higher failure rates [16; 25]. Guided by these policies, SOE managers are predisposed to adopt shorter-term investment strategies in the competitive landscape [8], thereby diminishing the impetus for corporate innovation and R&D expenditures and ultimately reducing the innovation efficiency of local SOEs. Consequently, the inhibitory impact of local state ownership on corporate innovation is more pronounced. Based on the foregoing arguments, we posit the following hypothesis:
H1. The impact of state ownership is hierarchical: central state ownership promotes corporate innovation, while local state ownership inhibits corporate innovation.
Methods
Sample and Data
This study collected ownership and financial data for all Chinese A-share listed firms from the CSMAR database spanning the years 2012-2021, while patent information was sourced from the CNRDS database. We meticulously cross-checked firm data with annual reports and official websites, adhering to the data pre-processing protocols articulated by R. Yuan and W. Wen [29]. First, financial firms (e.g., banks, insurance firms, and mutual funds) were excluded due to their distinctive governance and performance systems compared to non-financial Chinese firms. Second, "special treatment" firms - those experiencing continuous losses for two consecutive years and facing the risk of delisting - were omitted to mitigate the impact of abnormal financial conditions. Third, observations with missing information were discarded to minimize the influence of incomplete data on the results. Last, to further mitigate the impact of outliers, all continuous variables underwent winsorization at the 1st and 99th percentiles.
Variable Measurement and Model Specification
The dependent variables in this study measure corporate innovation (Patent_apply, Patent_grante, and Patent_cita-tion). The measurement method utilizes patent data provided by the CNRDS database. This database serves as a professional source for patent data analysis, covering multiple measurements and patent information, and has been widely accepted in the field. Following prior studies (e.g., R. Yuan, W. Wen; N. Ding et al. [29; 30]), the first measure, Patents_ apply, is the natural logarithm of a firm's total patent applications plus one, including invention patents, design patents, and utility patents. The second, Patents_ grante, is the natural logarithm of a firm's total granted patents plus one. The third, Patent_ citation, represents the natural logarithm of a firm's total patent citation counts plus one.
The independent variables used in this study are state ownership (SO) and its heterogeneous sub-variables - central state ownership (SO_C) and local state ownership (SO_L). In China, the prevalent phenomenon of cross-ownership and pyramidal control has been longstanding. The government often exercises indirect control over a specific enterprise by holding shares in other companies and implementing a hierarchical ownership structure within corporate groups. This intricate ownership framework complicates the calculation of the percentage of state ownership, as the extent of control may not be fully reflected in the direct shareholding percentage. Consequently, there is a risk of underestimating the control exerted by state-owned enterprises when computing ownership percentages, given that their influence may well extend beyond the direct ownership figures. To address this, inspired by P. Pessarossi and L. Weill [31] and N. Lin et al. [32], a dummy variable is employed to indicate state ownership (1 for state-owned entities and 0 otherwise). The two heterogeneous sub-var-
iables are central state ownership (SO_C) and local state ownership (SO_L). SO_C is a dummy variable equal to 1 for central state-owned entities and 0 otherwise, while SO_L is a dummy variable equal to 1 for local state-owned entities and 0 otherwise.
Furthermore, the study incorporates several control factors potentially affecting corporate innovation, aligning with prior research (e.g., R. McGuinness et al.; R. Yuan, W. Wen; N. Jia et al.; D. Kong et al.; N. Ding et al.; G. Liu, L. Lv [19; 29; 30; 33-35]). These include Firm Size (logarithm of total assets), Firm Age (natural logarithm of years since establishment plus one), Return on Assets (net income divided by total assets), Financial Leverage (total debts divided by total assets), Sale Growth (ratio of changed operating income to last year's operating income), Cash Ratio (cash holdings divided by total assets), Board Size (natural logarithm of total board directors), Ownership Concentration (percentage of shares owned by the largest shareholder), and Institutional Ownership (shares held by institutional investors divided by total shares). Refer to Table 1 for variable details and measurements.
To mitigate potential endogeneity, following previous studies (e.g., J. He, X. Tian; R. Yuan, W. Wen [29; 36]), we employ an OLS model and regress contemporaneous innovation measures on one-year lagged values of state ownership and other explanatory variables. The basic empirical model is as follows:
Patent _ applyi t+1 = a0 + alSOi t + a2FSi t + a3FAi t + +a4ROAit t + a5 LEVit t + a6 SGh t + a7CRit + a8 BSht + +a9OCi t + a10IOi t + Year + Industry + e, (1)
where a0 denotes the intercept, and a1 -a10 are the coefficients to be estimated. This study added dummy variables that control for year and industry fixed effects (Year and Industry); £ is the error term; i denotes the cross-sectional dimension for firms; and t denotes the time series dimension.
Table 1. Summary of Variable Descriptions and Measurements
Measurement
Panel A: Dependent Variables
Patent_ apply The natural logarithm of the firm's total patent applications plus one.
Patent_grante The natural logarithm of the firm's total granted patents plus one.
Patent_ citation The natural logarithm of the firm's total patent citation counts plus one.
Panel B: Independent Variables
State Ownership (SO) A dummy variable which equals 1 if the firm is a state-owned entity and 0 otherwise.
Measurement
Findings and Discussion
. j , , Descriptive Statistics and Correlation
Central State A dummy variable which equals 1 .
en ra a e if the firm is a central state-owned Matrix
entity and 0 otherwise. The descriptive statistics for the key variables in our study ......................................................................................................................................................................................................................................................................................................................are presented in Table 2, including the mean, standard deA dummy variable which equals viation, minimum, and maximum values. For Patent_ap-
L°cal State 1 if the firm is a local state-owned pty the mean and standard deviation are 2.622 and L721;
Ownership (SO_L) entity and 0 otherwise. for Patent_grante, 2.451 and 1.643; and, for Patent_citation,
........................................................................................................................................................... 1.959 and 1.803, respectively. These values reveal slight
variations in innovation measures among the sampled Panel C: Control Variables r „ m > .
firms. On average, 32.9% of firm-year observations pertain
to state-owned entities, with central state ownership (local Firm Size (FS) The logarithm of total assets. state ownership) accounting for 11.2% (21.7%). This con-
firms the existence of state ownership heterogeneity. The natural logarithm of the Regarding control variables, the sample firms exhibit an
Firm Age (FA) number of years since the firm's average Firm Size of 22.200, Firm Age of 2.908, ROA of
establishment plus one. 0.041, Financial Leverage of 0.412, Sale Growth of 0.169,
Cash Ratio of 0.049, Board Size of 2.120, Ownership Return on Assets The book value of net income Concentration of 34.383, and Institutional Ownership of (ROA) divided by total assets. 44.234. Additionally, R&D is 0.021, and Innovation Effi-........................................................................................................................................................... ciency is 0.140.
Financial Leverage The book value of total debts Table 3 displays the Pearson correlation matrix for the
(LEV) divided by total assets. major variables. The correlation coefficients between the
......................................................................................................................................................................................................................................................................................................................explanatory and control variables are mostly below 0.50.
The ratio of the changed Furthermore, we conducted a multicollinearity diagnostic
Sale Growth (SG) operating income to the operating test among the continuous variables. Each contro1 varia-income in the last year ble exhibits a low variance inflation factor (VIF) in the test
........................................................................................................................................................... (less than 2), indicating the absence of multicollinearity
issues in our model.
The book value of cash holdings
Cash Ratio (CR) divided by the book value of total Univariate Analysis assets
Table 4 presents the findings of univariate tests conducted on the dependent variable in our study. The mean of The natural logarithm of the total Patent_apply is 3.241 for firms classified as central stateBoard Size (BS) number of directors on the firm's owned entities and 2.544 for those not falling under central board. state ownership. These differences are statistically signifi-
cant at the 1% level, indicating that firms classified as cen-
Ownership r , . tral state-owned entities exhibit higher levels of innovation
The percentage of shares owned
Concentration output compared to their counterparts. by the largest shareholder.
(OC) Conversely, being categorized as a local state-owned entity significantly diminishes innovation output (Differences
Institutional The number of shares held by T-value = -0.477; P-values < 0.01). The negative t-statistics
institutional investors divided by for the mean differences, coupled with a 1% significance
the total shares. level, confirm the statistical significance of these variations
......................................................................................................................................................................................................................................................................................................................based on whether the firm is a local state-owned entity. In
Panel D: Other Variables summary, these outcomes lend initial support to Hypoth-
........................................................................................................................................................... esis 1, suggesting that central state ownership fosters cor-
Research and porate innovation, while local state ownership hampers it.
_ , The ratio of R&D expenditure to
^^mhture^R&D) total assets. Multivariate Results
The results of the OLS models are presented in Table 5, where the dependent variable is corporate innovation (PatInnovation Number of patent applications ent_apply). H1 is supported by the positive or negative co-Efficiency (IE) per unit of R&D input. efficients and significance level in the regressions of SO_C
........................................................................................................................................................... and SO_L. Specifically, the coefficient of SO_C in Column
Source: prepared by authors. (2) is 0.265, significant at the 1% level, indicating that
central state ownership promotes corporate innovation. However, the coefficient of SO_L in Column (3) is -0.072, significant at the 1% level, suggesting that local state ownership inhibits corporate innovation. Additionally, to examine whether the impact of total state ownership (SO) on corporate innovation is driven by central state ownership or local state ownership, Column (1) tests the relationship between SO and Patent_apply. The coefficient of SO is significant at the 1% level (a = 0.105), indicating that state ownership has a significantly positive overall effect on corporate innovation.
The aforementioned findings suggest that the relationship between state ownership and corporate innovation is hierarchical, with central state ownership promoting innovation and local state ownership inhibiting it. This provides a new explanation for the mixed evidence on the relationship
Table 2. Descriptive Statistics
Variables N Mean
between state ownership and corporate innovation (e.g., K. Kroll, H. Kroll; Y. Liu et al. [8; 20]), indicating that whether state ownership promotes or inhibits corporate innovation depends on whether central or local state ownership predominates in the innovation process.
Furthermore, the signs of the control variables are consistent with previous literature (e.g., Q. Hou et al.; K. Kroll, H. Kroll; R. Zhang et al. [8; 17; 37]). The results demonstrate that firm size, return on assets, and board size are positively and significantly related to Patent_apply in all columns, while firm age and financial leverage exhibit negative relationships with Patent_apply in all columns. Institutional ownership, however, is only negatively and significantly related to Patent_apply in Column (2). Sales growth, cash ratio, and ownership concentration are not significant with Patent_apply.
Std Min Max
Patent_apply 29 108 2.622 1.721 0.000 6.690
Patent_grante 29 108 2.451 1.643 0.000 6.409
Patent_citation 29 108 1.959 1.803 0.000 7.014
SO 29 108 0.329 0.470 0.000 1.000
SO_C 29 108 0.112 0.315 0.000 1.000
SO_L 29 108 0.217 0.412 0.000 1.000
FS 29 094 22.200 1.296 19.814 26.153
FA 29 094 2.908 0.325 1.609 3.497
ROA 27 239 0.041 0.063 -0.239 0.222
LEV 29 094 0.412 0.204 0.050 0.893
SG 27 234 0.169 0.390 -0.544 2.445
CR 29 094 0.049 0.067 -0.159 0.241
BS 29 053 2.120 0.197 1.609 2.708
OC 29 056 34.383 14.817 8.630 74.180
IO 29 025 44.234 25.232 0.321 94.529
R&D 29 108 0.021 0.020 0.000 0.101
IE 29 108 0.140 0.093 0.000 0.332
Table 3. Pearson Correlation
Patent_apply SO FS FA ROA LEV SG CR BS OC IO VIF
Patent_apply 1.000 -
SO -0.008 1.000 1.41
FS 0.293*** 0.385*** 1.000 1.78
FA -0.036*** 0.206*** 0.183*** 1.000 1.09
ROA 0.076*** -0.117*** -0.003 -0.080*** 1.000 1.55
LEV 0.081*** 0.309*** 0.533*** 0.178*** -0.358*** 1.000 1.71
SG 0.027*** -0.088*** 0.037*** -0.043*** 0.259*** 0.021*** 1.000 1.11
CR 0.052*** -0.017*** 0.064*** 0.004 0.411*** -0.169*** 0.026*** 1.000 1.24
BS 0.046*** 0.277*** 0.274*** 0.057*** -0.003 0.156*** -0.024*** 0.036*** 1.000 1.14
OC -0.006 0.210*** 0.186*** -0.084*** 0.127*** 0.051*** -0.010* 0.102*** 0.017*** 1.000 1.39
IO 0.045*** 0.415*** 0.440*** 0.053*** 0.102*** 0.208*** 0.028*** 0.119*** 0.230*** 0.485*** 1.000 1.78
Note. This table shows the correlation coefficients for the main variables defined in Table 1. The lower triangle in this table shows the Pearson correlation coefficients. VIF indicates the variance inflation factor. * p < 0.1, ** p < 0.05, *** p < 0.01, respectively. Source: calculated by authors.
Table 4. Univariate Analysis
Dummy (CSO) = 1 Dummy (CSO) = 0 Differences Dummy (LSO) = 1 Dummy (LSO)=0 Differences
N Mean N Mean T-value N Mean N Mean T-value
Patent_apply 3256 3.241 25852 2.544 0.697*** 6321 2.249 22787 2.725 -0.477***
Note. This table presents the results of univariate analysis on the mean difference of the corporate innovation indicator Patent_apply between "the firm is a central (local) state-owned entity" and "the firm is not a central (local) state-owned entity". The t-values for the mean differences are based on t-tests. ***denotes significance at the 1% level.
Source: calculated by authors.
Table 5. Multivariate Results
Patent_ aPPty (t+i)
(1) (2) (3)
SO 0.105***
(0.02)
SO_C 0.265***
(0.03)
SO_L -0.072***
(0.02)
FS 0.623*** 0.625*** 0.633***
(0.01) (0.01) (0.01)
FA -0.217*** -0.201*** -0.177***
(0.03) (0.03) (0.03)
ROA 2.113*** 2.172*** 2.104***
(0.18) (0.18) (0.18)
LEV -0.158*** -0.125** -0.106*
(0.06) (0.06) (0.06)
SG 0.001 -0.004 -0.015
(0.03) (0.03) (0.03)
CR 0.169 0.130 0.099
(0.15) (0.15) (0.15)
BS 0.206*** 0.198*** 0.231***
(0.05) (0.05) (0.05)
OC -0.001 -0.001 -0.000
(0.00) (0.00) (0.00)
IO -0.000 -0.001* -0.000
(0.00) (0.00) (0.00)
Cons -10.975* ** -11.034*** -11.368***
(0.23) (0.22) (0.22)
Year FE Yes Yes Yes
Industry FE Yes Yes Yes
N 22 935 22 935 22 935
Adj. R2 0.493 0.495 0.493
Note. This table presents the baseline result of the impact of state ownership heterogeneity on corporate innovation. The dependent variable is Patent_apply, while the independent variables are total state ownership (SO), central state ownership (SO_C) and local state ownership (SO_L). All regressions include year fixed effects and industry fixed effects. Parentheses show robust standard errors. * p < 0.1, ** p < 0.05, *** p < 0.01, respectively. All variables are defined in Table 1. Source: calculated by authors.
Robustness Check
Thus far, the estimations reveal a nuanced relationship between state ownership and corporate innovation. We employ various methods in this section to ensure the robustness of our results.
First, alternative dependent variables. To address concerns regarding potential measurement errors, following D. Kong et al. [33] and N. Ding et al. [30], we introduce two alternative dependent variables: the natural logarithm of the total number of patents granted to a firm plus one (Patent_grante) and the natural logarithm of the number of patent citations received by a firm plus one (Patent_ci-tation). Unlike the past reliance on the number of patent applications as an innovation indicator, the number of granted patents represents the actual quantity recognized and certified by governmental intellectual property agencies. Patent citations provide a quality-oriented perspective on innovation activity [30]. Panel A of Table 6 presents the robustness test results based on these two alternative dependent variables. The estimated coefficients of the primary variables exhibit similar magnitudes and directions as shown in Table 6, confirming the robustness of the baseline regression.
Second, two subsample tests. Addressing the argument by R. Zhang et al. [17] that firms in high-tech industries may have distinctive innovation needs, this study re-evaluates primary models using two different subsamples: one composed of high-tech industry firms and the other consisting of non-high-tech industry firms. Results in Panel B of Table 7 demonstrate that both SO and SO_C have a significantly positive influence on corporate innovation in both high-tech and non-high-tech industries. However, SO_L continues to exhibit a negative impact on corporate innovation at the 1% significance level within this subset. These results are in line with previous findings, confirming the consistency of our conclusions.
Third, alternative estimation methods. Considering the count nature of patents, R. Zhang et al. [17] suggest that fixed-effect model estimation might be misleading even with the logarithmic transformation of patent data. Therefore, this section employs the maximum likelihood method to estimate the Poisson regression model in Panel C of Table 8. Additionally, inspired by H. Kim et al. [38] to address truncation in patent data, the Tobit regression model is introduced. The results from both the Poisson model and the Tobit model (cf. Table 8) align with conclusions drawn in the previous main regression model.
Fourth, correcting for selection bias with the Heckman two-step selection model. Since the propensity of different SOEs to apply for patents may be non-random, causing self-selection bias, following R. Zhang et al. [16], the first stage estimates a probit model with a binary dummy (Dummy_Patent) as the dependent variable, equal to 1 if a firm has ever applied for a patent and 0 otherwise. The following probit model is used to estimate the probability of firms applying for patents:
Probit (Dummy _Patent). t = a0 +alFSi t +a2FAi t + +a3R°A1J +a4LEVht + a5SGit +a£RlS +a1BSlt + +a8OCi t + a9IOi t + Year + Industry + e. (2)
The inverse Mills ratio (IMR) is then obtained from this probability and included in the second stage regression. The results in Panel D of Table 9, after correcting for selection bias, indicate that the estimated coefficients of SO, SO_C, and SO_L consistently maintain the same signs as the previous ones and remain statistically significant. Thus, potential selection bias does not compromise our main findings.
Fifth, the application of a two-stage Data Envelopment Analysis model to control for managerial ability. Recognizing the pivotal role of managers in corporate innovation, differences in their ability levels can significantly impact innovation outputs. To address these differences, following R. Yuan and W. Wen [29], we employ a two-step procedure developed by P. Demerjian et al. [39] to estimate managerial ability.
In the first step, we assess the relative corporate efficiency of peer decision units using Data Envelopment Analysis (DEA). In the second step, we separate managerial contributions from corporate efficiency because the latter encompasses both corporate-level efficiency and manager-specific efficiency. This measurement criterion has been widely applied in accounting, finance, and management research (e.g., Z. Wang et al.; R. Yuan, W. Wen [15; 29]. In Equation (1), we introduce managerial ability (MA) as a new control variable and re-conduct a regression analysis.
The results in Panel E of Table 9 demonstrate that the signs and significance levels of all independent variables (SO, SO_C, and SO_L) remain consistent with our previous conclusions, suggesting that managerial ability is unlikely to drive our research findings.
Mechanism analysis
Research and development (R&D) expenditure and innovation efficiency (IE) are two pivotal determinants influencing corporate innovation [37; 40]. On the one hand, allocating funds and resources consistently to innovation activities enables firms to acquire new knowledge and technologies (referred to as the "input channel"). On the other hand, by enhancing innovation efficiency, organizations can bolster production efficiency, reduce costs, and consequently enhance innovation output (referred to as the "efficiency channel"). This section aims to investigate whether the heterogeneity in state ownership impacts these channels differently.
This research further adapts the baseline Equation (1) by replacing the dependent variable with R&D expenditure and innovation efficiency measured according to J. Lantz, J. Sahut [41] and A. Arundel, I. Kabla [42]. Table 10 presents regression results testing the impact on the input channel and the efficiency channel separately in panels F and G, respectively. These findings reveal a hierarchical influence of state ownership on innovation inputs and efficiency in which central state ownership (SO_C) notably amplifies firm R&D expenditure and innovation efficiency, while local state ownership (SO_L) inhibits both firm R&D expenditure and innovation efficiency.
Table 6. Robustness Check (1)
Panel A: Alternative Dependent Variables
Patent_grante Patent_ citation — (t+1)
(1) (2) (3) (4) (5) (6)
SO 0.096*** 0.187***
(0.02) (0.02)
SO_C 0.243*** 0.476***
(0.03) (0.03)
SO_L -0.066*** -0.136***
(0.02) (0.02)
FS 0.599*** 0.595*** 0.603*** 0.688*** 0.680*** 0.696***
(0.01) (0.01) (0.01) (0.01) (0.01) (0.01)
FA -0.220*** -0.204*** -0.183*** -0.089*** -0.057* -0.014
(0.03) (0.03) (0.03) (0.03) (0.03) (0.03)
ROA 1.412*** 1.416*** 1.353*** 1.078*** 1.084*** 0.960***
(0.17) (0.17) (0.17) (0.17) (0.17) (0.17)
LEV -0.103* -0.088 -0.070 -0.403*** -0.374*** -0.337***
(0.05) (0.05) (0.05) (0.06) (0.06) (0.06)
SG -0.021 -0.021 -0.031 -0.069*** -0.070*** -0.090***
(0.02) (0.02) (0.02) (0.02) (0.02) (0.02)
CR 0.278** 0.268* 0.239* 0.001 -0.018 -0.076
(0.14) (0.14) (0.14) (0.15) (0.15) (0.15)
BS 0.190*** 0.194*** 0.224*** 0.097** 0.105** 0.164***
(0.04) (0.04) (0.04) (0.05) (0.05) (0.05)
OC 0.000 0.000 0.000 -0.003*** -0.003*** -0.003***
(0.00) (0.00) (0.00) (0.00) (0.00) (0.00)
IO -0.001*** -0.001*** -0.001 -0.000 -0.001 0.001*
(0.00) (0.00) (0.00) (0.00) (0.00) (0.00)
Cons -10.576*** -10.535*** -10.841*** -12.915*** -12.834*** -13.441***
(0.21) (0.21) (0.21) (0.23) (0.22) (0.22)
Year FE Yes Yes Yes Yes Yes Yes
Industry FE Yes Yes Yes Yes Yes Yes
N 22 935 22 935 22 935 22 935 22935 22935
Adj. R2 0.513 0.514 0.512 0.534 0.539 0.533
Note. This table presents the results of the impact of state ownership heterogeneity on corporate innovation using alternative innovation measures. The dependent variables are Patent_grante and Patent_citation, and the independent variables are total state ownership (SO), central state ownership (SO_C), and local state ownership (SO_L). All regressions include year fixed effects and industry fixed effects. Parentheses show robust standard errors. * p < 0.1, ** p < 0.05, *** p < 0.01, respectively. All variables are defined in Table 1.
Table 7. Robustness Check (2)
Panel B: Alternative Samples
High-Tech Firms Non-High-Tech Firms
Patent_apply (M) Patent_apply (t+i)
(1) (2) (3) (4) (5) (6)
SO 0.094*** 0.068**
(0.03) (0.03)
SO_C 0.234*** 0.312***
(0.04) (0.05)
SO_L -0.095*** -0.104***
(0.03) (0.03)
FS 0.677*** 0.675*** 0.683*** 0.570*** 0.561*** 0.572***
(0.01) (0.01) (0.01) (0.02) (0.02) (0.02)
FA -0.160*** -0.154*** -0.119*** -0.334*** -0.307*** ■ -0.292***
(0.04) (0.04) (0.04) (0.05) (0.05) (0.05)
ROA 2.740*** 2.742*** 2.690*** 1.234*** 1.273*** 1.160***
(0.22) (0.22) (0.22) (0.30) (0.30) (0.30)
LEV 0.157** 0.183** 0.214*** -0.571*** -0.576*** -0.563***
(0.07) (0.07) (0.07) (0.10) (0.10) (0.10)
SG -0.017 -0.018 -0.029 0.019 0.023 0.010
(0.03) (0.03) (0.03) (0.04) (0.04) (0.04)
CR 0.446** 0.455** 0.404** -0.108 -0.140 -0.151
(0.20) (0.20) (0.20) (0.22) (0.22) (0.22)
BS 0.104* 0.109* 0.150*** 0.311*** 0.304*** 0.333***
(0.06) (0.06) (0.06) (0.08) (0.08) (0.08)
OC -0.003*** -0.004*** -0.003*** 0.003*** 0.003*** 0.004***
(0.00) (0.00) (0.00) (0.00) (0.00) (0.00)
IO -0.000 -0.001 0.000 -0.002** -0.002*** -0.001*
(0.00) (0.00) (0.00) (0.00) (0.00) (0.00)
Cons -11.567*** -11.548*** -11.911*** -10.360*** -10.200*** ■10.541***
(0.27) (0.26) (0.27) (0.37) (0.37) (0.37)
Year FE Yes Yes Yes Yes Yes Yes
Industry FE Yes Yes Yes Yes Yes Yes
N 13 337 13 337 13 337 9598 9598 9598
Adj. R2 0.428 0.430 0.428 0.447 0.449 0.447
Note. This table presents the results of the impact of state ownership heterogeneity on corporate innovation using alternative samples. The dependent variable is Patent_apply, and the independent variables are total state ownership (SO), central state ownership (SO_C), and local state ownership (SO_L). All regressions include year fixed effects and industry fixed effects. Parentheses show robust standard errors. * p < 0.1, ** p < 0.05, *** p < 0.01, respectively. All variables are defined in Table 1.
Table 8. Robustness Check (3)
Panel C: Alternative Estimation Methods
Poisson Method Tobit Method
Patent_apply(t+1) Patent_apply(t+1)
(1) (2) (3) (4) (5) (6)
SO 0.018** 0.105***
(0.01) (0.02)
SO_C 0.076*** 0.265***
(0.01) (0.03)
SO_L -0.044*** -0.072***
(0.01) (0.02)
FS 0.219*** 0.218*** 0.221*** 0.629*** 0.625*** 0.633***
(0.00) (0.00) (0.00) (0.01) (0.01) (0.01)
FA -0.088*** -0.086*** -0.074*** -0.218*** -0.201*** -0.177***
(0.01) (0.01) (0.01) (0.03) (0.03) (0.03)
ROA 0.899*** 0.911*** 0.880*** 2.169*** 2.172*** 2.104***
(0.07) (0.07) (0.07) (0.17) (0.17) (0.17)
LEV -0.044* -0.041* -0.030 -0.142** -0.125** -0.106*
(0.02) (0.02) (0.02) (0.06) (0.06) (0.06)
SG -0.001 0.000 -0.004 -0.003 -0.004 -0.015
2 (0.01) (0.01) (0.01) (0.02) (0.02) (0.02)
CR 0.005 0.008 -0.009 0.142 0.130 0.099
(0.06) (0.06) (0.06) (0.14) (0.14) (0.14)
BS 0.103*** 0.103*** 0.115*** 0.194*** 0.198*** 0.231***
(0.02) (0.02) (0.02) (0.05) (0.05) (0.05)
OC 0.000 0.000 0.000 -0.001 -0.001 -0.000
(0.00) (0.00) (0.00) (0.00) (0.00) (0.00)
IO -0.000*** -0.001*** -0.000** -0.001 -0.001* -0.000
(0.00) (0.00) (0.00) (0.00) (0.00) (0.00)
Cons -3.806*** -3.769*** -3.901*** -12.444*** -12.392*** -12.692***
(0.08) (0.08) (0.08) (0.24) (0.24) (0.24)
Year FE Yes Yes Yes Yes Yes Yes
Industry FE Yes Yes Yes Yes Yes Yes
N 22 935 22 935 22 935 22 935 22935 22935
Log-likelihood -38 795.19 -38 779.14 -38 789.57 -37 453.41 -37 419.07 -37 460.17
Wald/LR chi2 7554.85 7705.62 7783.11 15 685.10 15 753.80 15 671.58
Pseudo R2 0.150 0.150 0.149 0.173 0.174 0.173
Note. This table presents the results of the impact of state ownership heterogeneity on corporate innovation using alternative estimation methods. The dependent variable is Patent_apply, and the independent variables are total state ownership (SO), central state ownership (SO_C), and local state ownership (SO_L). All regressions include year fixed effects and industry fixed effects. Parentheses show robust standard errors. * p < 0.1, ** p < 0.05, *** p < 0.01, respectively. All variables are defined in Table 1.
Table 9. Robustness Check (4)
Panel D: Correcting Selection Bias Panel E: Controlling Managerial Ability
Patent_apply(ti) Patent_apply(ti)
(1) (2) (3) (4) (5) (6)
SO 0.087*** 0.122***
(0.02) (0.02)
SO_C 0.256*** 0.282***
(0.03) (0.03)
SO_L -0.087*** -0.057**
(0.02) (0.02)
FS 0.740*** 0.735*** 0.748*** 0.574*** 0.571*** 0.580***
(0.01) (0.01) (0.01) (0.01) (0.01) (0.01)
FA -0.290*** -0.277*** -0.254*** -0.170*** -0.149*** -0.127***
(0.03) (0.03) (0.03) (0.03) (0.03) (0.03)
ROA 2.319*** 2.329*** 2.265*** 3.441*** 3.435*** 3.367***
(0.18) (0.18) (0.18) (0.19) (0.19) (0.19)
LEV -0.243*** -0.230*** -0.212*** 0.129** 0.149** 0.162***
(0.06) (0.06) (0.06) (0.06) (0.06) (0.06)
SG -0.016 -0.016 -0.028 0.059** 0.058** 0.046*
(0.02) (0.02) (0.03) (0.03) (0.03) (0.03)
CR 0.057 0.049 0.012 0.261* 0.243 0.214
(0.15) (0.15) (0.15) (0.15) (0.15) (0.15)
BS 0.304*** 0.305*** 0.343*** 0.170*** 0.180*** 0.210***
(0.05) (0.05) (0.05) (0.05) (0.05) (0.05)
OC -0.001 -0.001 -0.000 -0.000 -0.000 0.000
(0.00) (0.00) (0.00) (0.00) (0.00) (0.00)
IO -0.002*** -0.002*** -0.001** -0.000 -0.001 0.000
(0.00) (0.00) (0.00) (0.00) (0.00) (0.00)
IMR 0.882*** 0.883*** 0.918***
(0.08) (0.08) (0.08)
MA -1.702*** -1.691*** -1.689***
(0.07) (0.07) (0.07)
Cons -13.763*** -13.694*** -14.140*** -10.143*** -10.137*** -10.473***
(0.30) (0.29) (0.29) (0.23) (0.23) (0.23)
Year FE Yes Yes Yes Yes Yes Yes
Industry FE Yes Yes Yes Yes Yes Yes
N 22 898 22 898 22 898 21 603 21603 21603
Adj. R2 0.496 0.498 0.496 0.504 0.506 0.504
Note. This table presents the results of the impact of state ownership heterogeneity on corporate innovation by correcting selection bias and controlling managerial ability. The dependent variable is Patent_apply, and the independent variables are total state ownership (SO), central state ownership (SO_C), and local state ownership (SO_L). IMR denotes the inverse Mills ratio. MA denotes managerial ability. All regressions include year fixed effects and industry fixed effects. Parentheses show robust standard errors. * p < 0.1, ** p < 0.05, *** p < 0.01, respectively. All variables are defined in Table 1.
Table 10. Mechanism Analysis
Panel F: Input Channel Panel G: Efficiency Channel
R&D Expenditure Innovation Efficiency
(1) (2) (3) (4)
SO_C 0.001*** 0.012***
(0.00) (0.00)
SO_L -0.001*** -0.003**
(0.00) (0.00)
FS -0.001*** -0.001*** 0.027*** 0.027***
(0.00) (0.00) (0.00) (0.00)
FA -0.003*** -0.003*** -0.013*** -0.012***
(0.00) (0.00) (0.00) (0.00)
ROA 0.031*** 0.030*** 0.101*** 0.098***
(0.00) (0.00) (0.01) (0.01)
LEV -0.002** -0.001* -0.007** -0.006*
(0.00) (0.00) (0.00) (0.00)
SG 0.000 0.000 0.000 -0.000
(0.00) (0.00) (0.00) (0.00)
CR 0.021*** 0.021*** -0.004 -0.005
(0.00) (0.00) (0.01) (0.01)
BS 0.001* 0.001** 0.012*** 0.014***
(0.00) (0.00) (0.00) (0.00)
OC -0.000*** -0.000*** -0.000 -0.000
(0.00) (0.00) (0.00) (0.00)
IO 0.000*** 0.000*** -0.000** -0.000
(0.00) (0.00) (0.00) (0.00)
Cons 0.046*** 0.044*** -0.442*** -0.457***
(0.00) (0.00) (0.01) (0.01)
Year FE Yes Yes Yes Yes
Industry FE Yes Yes Yes Yes
N 22 935 22 935 22 935 22935
Adj. R2 0.468 0.468 0.458 0.457
Note. This table shows the regression results of the two channels through which state ownership influences corporate innovation. The dependent variables are R&D Expenditure and Innovation Efficiency, and the independent variables are central state ownership (SO_C) and local state ownership (SO_L). All regressions include year fixed effects and industry fixed effects. Parentheses show robust standard errors. * p < 0.1, ** p < 0.05, *** p < 0.01, respectively. All variables are defined in Table 1.
Conclusion
Utilizing firm-level data spanning the period 2012-2021 in China, this study addressed the puzzle surrounding the impact of state ownership on corporate innovation, focusing on the context of Chinese SOEs. From a hierarchical perspective, our findings helped to resolve the inconsistency observed in previous research, which has been attributed to the hierarchical structure of state ownership and to differences in human resource control mechanisms within SOEs. Owing to dissimilarities in evaluation protocols and the hierarchical configuration of SOEs, the innovation orientation of state ownership revealed a hierarchical pattern: central state ownership tends to foster innovation, while local state ownership tends to impede it. The result was shown to be robust by a series of checks, including alternative dependent variables, subsample tests, the Poisson model, the Tobit model, the Heckman two-step sample selection model, and the application of a two-step Data Envelopment Analysis model to control for managerial ability.
Additionally, this paper substantiated the existence of two influential channels through which state ownership heterogeneity impacts corporate innovation - the input channel and the efficiency channel. The hierarchical structure extends its influence to these pivotal facets of corporate innovation, with central state ownership positively affecting both channels and local state ownership exerting a negative influence.
This study contributes to the literature on state ownership and corporate innovation within the framework of institutional theory. Prior research has underscored the pivotal role of political affiliations in overcoming institutional voids in emerging markets [15; 17]. This study suggests that state ownership serves as a crucial means for accessing scarce resources and addressing institutional voids. Simultaneously, the heterogeneous impact of central and local state ownership on corporate innovation indicates that state-owned enterprises may exhibit varying levels of innovation inputs, innovation efficiency, and innovation outputs due to distinct institutional pressures stemming from state ownership heterogeneity. These findings illuminate the intricate interplay between China's institutional landscape, state ownership, and corporate innovation, providing fresh insights into the ongoing development of institutional perspectives.
Furthermore, our findings have significant practical implications for emerging countries seeking to emulate the Chinese system of governance in their transition to an innovative economy. In such contexts, the government's control over corporate ownership is divided between the central and local governments, whose institutional frameworks may be said to consist of government entities rather than purely private enterprises. Our findings suggest that the personnel control systems of hierarchical state ownership take different approaches to the political promotion and incentives of managers of state-owned enterprises, making central state ownership more conducive
to corporate innovation. Additionally, the conclusions of this study imply that policymakers should recognize the nuanced relationship between state ownership and corporate innovation to take institutional differences into account for the purposes of creating appropriate innovation-oriented systems and avoiding a one-size-fits-all approach.
At the same time, this study has several limitations. First, it does not split local state ownership types into sub-levels. For example, the impact of local state ownership may change if one considers the provincial, city, and county levels separately. Second, the generalisability of the findings is limited by our use of a single country as the research context. Future work should address these limitations by considering a more fine-grained decomposition of the various forms of local state ownership, as well as other transitioning economies. These extensions would further develop our understanding of corporate innovation in a transition economy from a hierarchical state ownership perspective.
References
1. Jiang X., Yuan Q. Institutional investors' corporate site visits and corporate innovation. Journal of Corporate Finance. 2018;48:148-168. https://doi. org/10.1016/j.jcorpfin.2017.09.019
2. Choi S.B., Lee S.H., Williams C. Ownership and firm innovation in a transition economy: Evidence from China. Research Policy. 2011;40(3):441-452. https:// doi.org/10.1016/j.respol.2011.01.004
3. Tónurist P. Framework for analysing the role of state owned enterprises in innovation policy management: The case of energy technologies and Eesti Energia. Technovation. 2016;38:1-14. https://doi.org/10.1016/'. technovation.2014.08.001
4. Li J., Shan Y., Tian G. et al. Labor cost, government intervention, and corporate innovation: Evidence from China. Journal of Corporate Finance. 2020;64:101668. https://doi.org/10.1016/'. jcorpfin.2020.101668
5. Yi J., Hong J., Hsu W., et al. The role of state ownership and institutions in the innovation performance of emerging market enterprises: Evidence from China. Technovation. 2017;62-63:4-13. https://doi.org/10.1016/'. technovation.2017.04.002
6. Baum J. a. C., Oliver C. Institutional Linkages and Organizational Mortality. Administrative Science Quarterly. 1991;36(2):187-218. https://doi. org/10.2307/2393353
7. Tebaldi E., Elmslie B. Does institutional quality impact innovation? Evidence from crosscountry patent grant data. Applied Economics. 2011;45(7):887-900. https://doi.org/10.1080/0003684 6.2011.613777
8. Kroll H., Kou K. Innovation output and state ownership: empirical evidence from China's listed firms. Industry and Innovation. 2018;26(2):176-198. https://doi.org/10.1080/13662716.2018.1456323
9. Peng M.W., Bruton G.D., Stan C.V., et al. Theories of the (state-owned) firm. Asia Pacific Journal of Management. 2016;33(2):293-317. https://doi. org/10.1007/s10490-016-9462-3
10. Chen H.Y.H., Yoon S.S. Government efficiency and enterprise innovation - evidence from China. Asian Journal of Technology Innovation. 2019;27(3):280-300. https://doi.org/10.1080/19761597.2019.1678389
11. Ding M., Suardi S. Government ownership and stock liquidity: Evidence from China. Emerging Markets Review. 2019;40:100625. https://doi.org/10.1016/'. ememar.2019.100625
12. Zhang M., Yu F., Zhong C. How State Ownership Affects Firm Innovation Performance: Evidence from China. Emerging Markets Finance and Trade. 2022;59(5):1390-1407. https://doi.org/10.1080/15404 96x.2022.2137374
13. Blanchard O., Shleifer A. Federalism With and Without Political Centralization: China versus Russia. IMF Staff Papers. 2001;48(1):171-179. https://doi. org/10.2307/4621694
14. Scott W.R. Institutions and Organizations: Ideas, Interests, and Identities. SAGE Publications; 2013.
15. Wang Z., Chen M., Chin C., et al. Managerial ability, political connections, and fraudulent financial reporting in China. Journal of Accounting and Public Policy. 2017;36(2):141-162. https://doi.org/10.1016/'. jaccpubpol.2017.02.004
16. Dacin M.T., Oliver C., Roy J. The legitimacy of strategic alliances: an institutional perspective. Strategic Management Journal. 2006;28(2):169-187. https://doi.org/10.1002/smj.577
17. Zhang R., Xiong Z., Li H., et al. Political connection heterogeneity and corporate innovation. Journal of Innovation & Knowledge. 2022;7(3):100224. https:// doi.org/10.1016/j.jik.2022.100224
18. Chen S., Sun Z., Tang S., et al. Government intervention and investment efficiency: Evidence from China. Journal of Corporate Finance. 2011;17(2):259-271. https://doi.org/10.1016/j-. jcorpfin.2010.08.004
19. Liu G., Lv L. Government regulation on corporate compensation and innovation: Evidence from China's minimum wage policy. Finance Research Letters. 2022;50:103272. https://doi.org/10.1016/-'. frl.2022.103272
20. Liu Y., Zhang Z., Zhai H., et al. Does flattening the government hierarchy improve corporate innovation? Evidence from China. Regional Studies.
2023;57(8):1559-1577. https://doi.org/10.1080/00343 404.2022.2148644
21. Chen V.Z., Li J., Shapiro D., et al. Ownership structure and innovation: An emerging market perspective. Asia Pacific Journal of Management. 2014;31(1):1-24. https://doi.org/10.1007/s10490-013-9357-5
22. Yang X., Sun S.L., Yang H. Market-based reforms, synchronization and product innovation. Industrial Marketing Management. 2015;50:30-39. https://doi. org/10.1016/j.indmarman.2015.04.015
23. Lundvall B. Why study national systems and national styles of innovation? Technology Analysis & Strategic Management. 1998;10(4):403-422. https://doi. org/10.1080/09537329808524324
24. Xu E., Zhang, H. The impact of state shares on corporate innovation strategy and performance in China. Asia Pacific Journal of Management. 2008;25(3):473-487. https://doi.org/10.1007/s10490-008-9093-4
25. Lou X.Y., Qian A., Zhang C. Do CEO's political promotion incentives influence the value of cash holdings: Evidence from state-owned enterprises in China. Pacific-Basin Finance Journal. 2021;68:101617. https://doi.org/10.1016/j-.pacfin.2021.101617
26. Li H., Zhou L. Political turnover and economic performance: the incentive role of personnel control in China. Journal of Public Economics. 2005;89(9-10):1743-1762. https://doi.org/10.1016/-. jpubeco.2004.06.009
27. Cao X., Lemmon M.D., Pan X., et al. Political Promotion, CEO Incentives, and the Relationship Between Pay and Performance. Management Science. 2019;65(7):2947-2965. https://doi.org/10.1287/ mnsc.2017.2966
28. Bronzini R., Piselli P. The impact of R&D subsidies on firm innovation. Research Policy. 2016;45(2):442-457. https://doi.org/10.1016/j-.respol.2015.10.008
29. Yuan R., Wen W. Managerial foreign experience and corporate innovation. Journal of Corporate Finance. 2018;48:752-770. https://doi.org/10.1016/j-. jcorpfin.2017.12.015
30. Ding N., Gu L., Peng Y. Fintech, financial constraints and innovation: Evidence from China. Journal of Corporate Finance. 2022;73:102194. https://doi. org/10.1016/j.jcorpfin.2022.102194
31. Pessarossi P., Weill L. Choice of corporate debt in China: The role of state ownership. China Economic Review. 2013;26:1-16. https://doi.org/10.1016/j-. chieco.2013.03.005
32. Lin N., Chen H.Y.H., Zhang P., et al. Does gambling culture affect firms' investment efficiency? Finance Research Letters. 2022;49:103148. https://doi. org/10.1016/j.frl.2022.103148
33. Kong D., Wang Y., Zhang J. Efficiency wages as gift exchange: Evidence from corporate innovation in China. Journal of Corporate Finance. 2020;65;101725. https://doi.org/10.1016/j.jcorpfin.2020.101725
34. McGuinness P.B., Vieito J.P., Wang, M. The role of board gender and foreign ownership in the CSR performance of Chinese listed firms. Journal of Corporate Finance. 2017;42:75-99. https://doi. org/10.1016/j.jcorpfin.2016.11.001
35. Jia N., Mao X., Yuan R. Political connections and directors' and officers' liability insurance - Evidence from China. Journal of Corporate Finance. 2019;58:353-372. https://doi.org/10.1016/j. 'corpfin.2019.06.001
36. He J., Tian X. The dark side of analyst coverage: The case of innovation. Journal of Financial Economics. 2013;109(3):856-878. https://doi.org/10.1016/'. jfineco.2013.04.001
37. Hou Q., Hu M., Yuan Y. Corporate innovation and political connections in Chinese listed firms. Pacific-basin Finance Journal. 2017;46(Part A):158-176. https://doi.org/10.1016/jj.pacfin.2017.09.004
38. Kim H.D., Park K., Song, K. Do long-term institutional investors foster corporate innovation? Accounting and Finance. 2017;59(2):1163-1195. https://doi.org/10.1111/acfi.12284
39. Demerjian P.R., Lev B., McVay S. E. Quantifying Managerial Ability: A New Measure and Validity Tests. Management Science. 2012;58(7):1229-1248. https://doi.org/10.1287/mnsc.1110.1487
40. Pu T., Zulkafli A. H. Managerial ownership and corporate innovation: evidence of patenting activity from Chinese listed manufacturing firms. Cogent Business & Management. 2024;11(1):2289202. https:// doi.org/10.1080/23311975.2023.2289202
41. Lantz J., Sahut, J. R&D investment and the financial performance of technological firms. International Journal of Business. 2005;10(3):251.
42. Arundel A., Kabla I. What percentage of innovations are patented? empirical estimates for European firms. Research Policy. 1998;27(2):127-141. https://doi. org/10.1016/s0048-7333(98)00033-x
Contribution of the authors: the authors contributed equally to this article. The authors declare no conflicts of interests.
The article was submitted 06.01.2024; approved after reviewing 08.02.2024; accepted for publication 29.02.2024.
