Risk preference of top management team and the failure of technological innovation in firms–based on principal component analysis and probit regression

Abstract

As an important psychological feature that affects the strategic decision-making of top management team (TMT), risk preference exerts important impact on the failure of innovation. Taking Chinese pharmaceutical listed companies from 2011 to 2017 as research samples, this paper uses principal component analysis and probit regression model to empirically test the impact of TMT risk preference on technological innovation failure of firms from the perspective of government intervention. It is found that TMT’s risk preference has a significantly positive impact on the failure of firms’ technological innovation. The regulatory role of government intervention between TMT’s risk preference and failure of firms’ technological innovation is an inverted U-shaped relationship. Low or high level of government intervention will promote the positive impact of TMT’s risk preference on the failure of technological innovation, and increase the risk of technological innovation failure. The findings of this study contribute to: (1) broaden the research perspective of technological innovation failure. (2) The innovation failure rate can be reduced by constructing or optimizing the TMT and the government intervening in firm innovation through reasonable and perfect technology innovation guidance policies. (3) expand the research means of innovation failure, making the results more convincing.

Keywords

TMT Risk preference Failure of technological innovation Government intervention

1 Introduction

Innovation as a driving force for national economic growth and development and a decisive factor for long-term competitive advantage, innovation has become the direct embodiment of comprehensive national strength and core competitiveness, which is of strategic importance for promoting the healthy, stable and sustainable development of China’s economy. However, technological innovation failure exists objectively due to the influence of many factors. Specifically, technological innovation is influenced by various factors, such as market and institutional environment, innovation system and mechanism, decision-makers’ innovation tendency and leadership quality as well as the relationship between risk and benefit of technological innovation [26]. According to the statistics,the commercialization failure rate of technological innovation projects is as high as 70% among Advanced Micro Devices and other companies in the United States [12]. In China, 90% of the technological innovation projects carried out by 2130 firms in six major industries are terminated or failed [29]. According to Schumpeter’s definition of innovation, the core criterion for judging the success or failure of technological innovation lies in whether the innovation subject has achieved the expected innovation goal. That is the passive or active termination of innovation due to the influence of knowledge resources, technological capabilities, product and market positioning, interface efficiency and managerial characteristics [4]. From the perspective of innovation system theory, technological innovation failure is affected by the uncertainty of internal and external system. With the increasing complexity of the external environment, the limitations of individual decision-making by managers have become increasingly prominent. The top management team (TMT) has gradually replaced individual managers as the collective owners of the actual control power of the firm [32]. TMT’s risk perception, cognition and preference will have an important impact on firms’ innovation, and then affect the failure of firm technological innovation. On the other hand, TMT’s innovation behavior is often determined by the subjective consciousness of team members. Therefore, as an important psychological feature affecting managers’ decision-making, it has a more direct impact on corporate innovation failure than the demographic features such as top managers’ age, gender and tenure [30]. At present, China is in the critical period of economic transformation. The interaction between market mechanism and government’s intervention in capital market makes the government replace the allocation of factor resources through market mechanism to a certain extent [33]. Therefore, the impact of government intervention on the innovation activities of firms can not be ignored. When studying the relationship between TMT’s risk preference and firms’ technological innovation failure, it is necessary to consider whether government intervention has an impact between them.

However, given the “negative” nature of innovation failure, most people have the “anti-failure” bias [22]. At present, it was still lack of systematic research on the risk preference of TMT and the failure of firm technology innovation in practice and theory. Relevant researches are mainly carried out from the following aspects: firstly, the influence of individual characteristics or risk preference of TMT on firm innovation [3]. Secondly, only consider the impact of external environment such as financial development, legal environment and market on the technological innovation failure of firms [33]. Thirdly, from the perspective of failure learning, analyze the failure of firm innovation through cases or theories [17]. But these aspects of the study are worth further expansion. Firstly, from the perspective of the influence of the individual characteristics of the TMT or the risk preference on the firm innovation, the influence of the risk preference of the TMT on the firm technological innovation failure is not considered. Secondly, it only studies the relationship between technological innovation failure of firms from the perspective of external environment, ignoring the factors of innovation subjects themselves, especially the overall psychological risk preference of the firm’s TMT. Finally, from the perspective of failure learning, one view holds that technological innovation failure is beneficial. Through failure learning, managers can examine innovation issues more calmly and objectively [7]. On the other hand, failure is harmful but not beneficial. Failure of technological innovation will cause psychological trauma to managers, and the negative emotions of failure will prevent them from learning from failure [6]. The reason for the inconsistency of the above conclusions may be that the existing research on innovation failure is still theoretical analysis and lack of empirical research. Therefore, the research space is left for this paper.

The innovation of this paper may be reflected in: (1) broaden the research perspective. Different from the previous studies which study the innovation input or performance from the perspective of individual characteristics of TMT, this paper studies the impact of firm innovation failure on firm technological innovation failure from the perspective of risk preference of TMT; (2) comprehensively consider the impact of firm internal and external factors on technological innovation failure. While studying the relationship between TMT risk preference and technological innovation failure, we also study the impact of TMT risk preference and technological innovation failure from the perspective of government intervention; (3) innovation on sample data. In the selection of sample data, different from the previous research methods such as questionnaire survey or case analysis, this paper focuses on listed pharmaceutical manufacturing firms, and verifies the impact of TMT risk preference on technological innovation failure through empirical analysis, and the robustness and reliability of the results are also greatly improved.

The remaining structure of this paper is as follows. Section 2 is literature review and theoretical hypothesis. The research methods including the main variables, data sources and model building are introduced in Section 3. Section 4 is the analysis of empirical results. The conclusions and implications are summarized in Section 5.

2 Literature review and theoretical hypotheses

2.1 Impact of risk preference of TMT on failed technological innovation

According to the higher-order theory and prospect theory, managers’ innovation behavior when facing uncertainty is often affected by their own psychological cognition and values, and managers are not “completely rational”, due to TMT members’ different growth experience and education background. In the process of firm development, most of failures come from the complacency of senior managers on the firm development strategy [2]. Social environment risk, firms’ internal risk, and managers’ risk preference will lead to the failure of firm innovation [14]. The managers with stronger risk preference are more likely to make firms’ daily operation at high risk and further make firms fall into passivity and predicament. Naldi [18] studied Swedish family business and their conclusion from empirical analysis also supported this view. They believed that the higher managers’ risk preference is, the easier firms’ daily operation to get into trouble. However, firms have to take the risk to defeat competitors by upgrading technology and products or increasing market share so as to stand out in the fierce market competition. The level of TMT’s risk preference reflects their psychological attitude in the face of uncertainty. And there are obvious differences in the attitude of different managers to risk. Generally, risk preference managers tend to be more adventurous and pay more attention to the opportunities of reform or innovation. They will attach more importance to firms’ growth, and more keen to achieve firms’ innovation goals by increasing R & D investment and developing new products to seize the market. However, risk aversion managers tend to give up higher-risk investment projects that have extra benefits and are beneficial to shareholders, and reduce innovation activities rather than invest too much in new product development and new market development. As a result, risk aversion managers usually have lower risk tolerance and can only get lower returns. On the contrary, risk preference managers have stronger risk tolerance and higher returns [23]. To some extent, risk-averting managers can reduce the risks of technological innovation. Meanwhile, they can also inhibit firms’ growth [20].

The above analysis indicates that the increase of TMT’s risk preference level can subjectively mobilize firms’ innovation enthusiasm. However, high income also means high risk. Based on these, this paper proposes the following assumption:

Hypothesis 1: The risk preference of the TMT has a positive impact on failed technological innovation.

2.2 The regulating effect of government intervention

From the perspective of institutional theory and market transformation theory, institutional environment factors at the macro level may have a greatly important impact on firms’ innovation behavior in countries with emerging economies in transition [15]. Especially in the process of China’s market-oriented reform, the relationship between government and market often plays a decisive role in the flow and allocation of factor resources, and government intervention has a significant correlation with firm innovation [4]. The government influences the TMT’s strategic decision-making through a variety of interventions such as government finance, finance, economy and trade, thus affecting firms’ technological innovation activities [11], and to a certain extent, affecting the failure of the technological innovation of the firm. From the existing literature, the conclusions about the effect have two completely opposite views. Some scholars believe that the lack of financial policy, industrial policy and other resources support often leads to the passive interruption after technology diffusion or innovation lag, which makes the technology innovation of firm fail [24]. The government can decrease the financial pressure of firm technological innovation, reduce the managers’ risk perception of technological innovation, improve the enthusiasm of firm research and development, provide sufficient resource support for technological innovation [27], and increase the innovation results and success rate of innovation by intervening in firm technological innovation through policy tools such as tax incentives and government subsidies [16]. When the government implements the tax incentive policy for firm technological innovation, it will effectively promote the managers to invest in R & D, so that firms can produce higher innovation output [8]. On the contrary, when tax revenue increases, managers tend to reduce R & D investment and patent application, reduce new product access to the market, decrease innovation, or even non innovation [19]. Some studies in China also propose that the government’s innovation incentive policy can stimulate the R & D investment and share the risk of technological innovation failure. Government subsidies can effectively alleviate the external funding pressure of firm managers [25], make firms more motivated to invest in R & D, and their R & D is more frequent than that of firms without government subsidies. The scholars who hold the opposite view believe that the government’s financial support is essentially a substitute for the firm’s own R & D expenditure. The government subsidy does not promote the firm’s technological innovation, but instead squeezes the firm’s own R & D investment [1]. Especially in the case of imperfect relevant regulatory system, the firm managers may take advantage of the information asymmetry or rent-seeking activities with government officials to obtain government subsidies improperly and use government subsidies for other purposes, which will lead to the reduction of firms’ own necessary investment in technological innovation [5]. And government officials tend to invest more government subsidies in projects with high success rate and high short-term return rate for their own achievements and promotion. At the same time, too much government intervention will also cause firm managers to be more inclined to projects with short-term benefits [34]. To some extent, it will hinder or even damage the firm’s own technological innovation behavior, cause negative impact on the resource supply and smooth implementation of firm’s technological innovation activities [2], and increase the risk of firm’s technological innovation failure.

The above analysis shows that insufficient or excessive government intervention will lead to the failure of firm technological innovation. When the TMT makes the firm strategic plan, it will try to reconcile the administrative objectives of the government and the economic interests of the firm, so as to obtain the resources needed for the survival and development of the firm [9]. Based on this, this paper proposes the following assumption:

Hypothesis 2: Government intervention has a significant regulating effect on the risk preference of TMT and the failure of technological innovation. And the regulating effect shows an inverted U-shaped curve.

3 Methodology

3.1 Data

According to the Guidelines for Classification of Listed Firms (Revised in 2012) issued by China’s Securities Regulatory Commission, this study chooses listed pharmaceutical manufacturing firms in China A-share market from 2011 to 2017 as samples. The sample data are obtained and screened as follows:

First, excluding veterinary and agricultural drugs in the range of non-ADRs monitoring, drug manufacturers were further matched manually by using the information disclosed in the official websites of listed firms and annual reports of firms. A total of 398 subsidiary drug manufacturers included by listed pharmaceutical manufacturing firms are obtained. The information retrieval in the field of manufacturer is performed by using the national ADRs monitoring system, and the search results are used to manually judge and set the variable values. When the ADRs produced by a listed firm’s affiliated firms occur in a certain year, that is, if the sample firms fail to innovate in the same year, the value of innovfail is 1, and vice versa is 0.

Second, according to the definition of the major variables and the time interval of the sample, the missing data of relevant indicators and the sample of listed firms that do not meet the time interval of 2011–2017 are eliminated.

Third, about the moderator variable, the index of relationship between government and market is derived from the newly published NERI INDEX of Marketization of China’s Provinces 2018 Report [28]. Based on the above data, the sample data of government intervention is matched according to the registered location of listed firms.

Last, the data of listed firms in pharmaceutical industry are from China Stock Market and Accounting Research (CSMAR&WIND) database. According to the above principles, this study finally obtains 480 valid samples of 80 listed pharmaceutical manufacturing firms from 2011 to 2017.

3.2 Variable definition and calculation

3.2.1 Dependent variable

Failed technological innovation (innovfail): when a firm’s technological innovation fails, it is often because of the fear of the disclosure of its failure information that the outside world questions its technological innovation ability, but is reluctant to face up to the failure. Especially in China’s current economic context, with the government controlling most of the key domestic resources, companies are reluctant to talk about failure. Executives are afraid that failure will send a negative signal to the outside world, because positive innovation signals can enable them to get more government help. This paper draws on Maslach D’s [10] approach to the innovation failure variable, that is, when an adverse reaction occurs in a pharmaceutical product produced by a listed company in the pharmaceutical manufacturing industry, it indicates that its innovation has failed. Adverse drug reactions (adrs) refer to the harmful reactions that occur to the body of patients that have nothing to do with the purpose of medication and are not expected when the patients use the qualified drugs that have been approved by the state drug administration and other relevant departments. When the drug has adverse reactions that have nothing to do with the patient’s medication purpose, to a certain extent, it shows that the drug has not reached the expected goal, which is consistent with the core criteria of defining the connotation of innovation failure and judging innovation failure. Therefore, in this paper, technological innovation failure is set as a virtual variable. When adverse reactions occur in pharmaceutical products, it means innovation failure. The value is 1, otherwise it is 0.

3.2.2 Independent variable

Risk preference of TMT(mrp): as an important strategic management resource within a firm, the risk preference of the TMT will have an important impact on the future inno-vation activities of the firm. In existing studi-es, domestic and foreign scholars mainly ad-opt four methods:DARA&DRRA coefficient measurement, scale form measurement, exe-cutive personal characteristics measurement and custom variable measurement. It can be seen that the measurement of the risk preference of the TMT in the existing li-terature is not uniform. In view of the speci-al situation of China’s capital market, this paper chooses to use the method of Cain et al., [21] to build the TMT from the firm and individual the risk preference evaluation index system, and the time series principal component analysis is used to calculate the risk preference score of the TMT as the value of this variable.

According to the research of the predecessors’ [31], In combination with the actual situation in China, this paper carries out forward processing and standardization processing on the risk preference evaluation indexes of the TMT (Table 1), as well as KMO test and Bartlett’s spherical test to ensure that the selected indexes meet the basic requirements of time series principal component analysis, and then establishes the risk preference evaluation model of the TMT:

Table 1
Risk preference evaluation index system of TMT

Dimension Name Definition

Firm level quick ratio quick assets/current liabilities

ratio of short-term risk assets (trading financial assets + accounts receivable)/current assets

current liability ratio current liabilities/total liabilities

asset-liability ratio total liabilities/total assets

long term asset suitability rate (total owner’s equity + total long-term liabilities)/(total fixed assets + total long-term investments) *100%

core earnings ratio operating income/(operating income + fair value change profit and loss + investment income + non-operating income)

cash ratio (cash + trading financial assets)/current liabilities

rate of long-term use of short-term funds 1- (non-current liabilities + shareholders’ equity)/non-current assets

Individual level the proportion of male executives the dummy variable, male 1, female 0. Proportion of male executives to total executives

average age of top management team average of the sum of top executives’ ages

average education level of top management team technical secondary school and below = 1, junior college = 2, bachelor = 3, master = 4, master above = 5;Average of the sum of executive education

average tenure of top management team average of the total number of months of executive cumulative tenure

average total wealth of top management team the natural logarithm of the average value of the sum of stock market value and annual salary held by top executives at the end of the year

average number of shares held by top management team at the end of the year the natural logarithm of the average number of shares held by executives at the end of the year

average risk assets ratio of top management team average ratio of year-end stock market value to annual salary and year-end stock market value of executives

Dimension	Name	Definition
Firm level	quick ratio	quick assets/current liabilities
	ratio of short-term risk assets	(trading financial assets + accounts receivable)/current assets
	current liability ratio	current liabilities/total liabilities
	asset-liability ratio	total liabilities/total assets
	long term asset suitability rate	(total owner’s equity + total long-term liabilities)/(total fixed assets + total long-term investments) *100%
	core earnings ratio	operating income/(operating income + fair value change profit and loss + investment income + non-operating income)
	cash ratio	(cash + trading financial assets)/current liabilities
	rate of long-term use of short-term funds	1- (non-current liabilities + shareholders’ equity)/non-current assets
Individual level	the proportion of male executives	the dummy variable, male 1, female 0. Proportion of male executives to total executives
	average age of top management team	average of the sum of top executives’ ages
	average education level of top management team	technical secondary school and below = 1, junior college = 2, bachelor = 3, master = 4, master above = 5;Average of the sum of executive education
	average tenure of top management team	average of the total number of months of executive cumulative tenure
	average total wealth of top management team	the natural logarithm of the average value of the sum of stock market value and annual salary held by top executives at the end of the year
	average number of shares held by top management team at the end of the year	the natural logarithm of the average number of shares held by executives at the end of the year
	average risk assets ratio of top management team	average ratio of year-end stock market value to annual salary and year-end stock market value of executives

Note: The data are collated by the authors.

$mrp = \sum λ_{i} * f_{i}$ (1)

Among them, mrp represents the risk preference score of the TMT; λ_i indicates the variance contribution rate corresponding to each principal component; f_i represents the score of each principal component. Six principal components were extracted from 15 indicators, the scores of each principal component and the weight of the corresponding variance contribution rate were substituted into model (1), and the final score was calculated as the assessment value of the risk preference of the TMT in this paper.

3.2.3 Regulated variable

Government intervention (gover): in view of China’s unique national conditions, local governments in China have a large number of resources in the process of market-oriented reform, and the government plays an absolute leading role in resource allocation and infrastructure supply through planning. In order to obtain enough innovation resources, firms need to maintain a good relationship with the government in order to ensure the smooth implementation of the technological innovation process and reduce the innovation risk. However, excessive government intervention may cause the firm’s TMT to prefer short-term profit projects, reduce the government’s support for the exploratory innovation activities of firms, and increase the failure rate of technological innovation. In the “Government-Market Relations” index, secondary indicators such as the proportion of market allocation of economic resources, reduction of government intervention in firms, reduction of extra-tax burden on firms, and reduction of government size can better reflect resource allocation The role of government, the degree of government intervention in firms, and the degree of influence of government expansion on normal market activities. Therefore, this paper uses the “relationship between government and market” index to measure the level of government intervention, and the index is a reverse index. The larger the index, the lower the level of government intervention in the region, and vice versa.

3.2.4 Control variable

In the setting of control variables, this study mainly follows these considerations:

First of all, technical barriers are an important internal cause of firm innovation failure. Therefore, the technological resources and level of a firm have a great impact on the success or failure of technological innovation. On the one hand, when the firm has sufficient technical resources such as knowledge resources and human capital, it can provide resource guarantee for the implementation of technological innovation projects and reduce the risk of innovation failure. On the other hand, a deep professional level can help firms overcome technical obstacles in the process of technological innovation. Therefore, this paper uses the ratio of R&D expenditure to main business income for measurement. The larger the ratio is, the higher the technological innovation resources and level of the firm are.

Secondly, in view of China’s national conditions, state-owned firms have a greater advantage in acquiring external resources than non-state-owned firms, and state-owned firms tend to be larger in scale, so they will have more sufficient resources to ensure the smooth implementation of technological innovation projects. In this paper, the nature of firm ownership and firm scale are taken as control variables. When the firm is a state-owned firm, state = 1, otherwise state = 0. The firm scale adopts the natural logarithm of the total assets of the sample firm. The larger the value is, the larger the firm scale is.

In addition, according to the existing literature [10, 33]. This paper also controls the firm growth and returns on assets, whether the chairman and the manager are the same one, the difference between control and ownership and other variables. The definition of major variables are shown in Table 2.

Table 2
The definition of major variables

Type Name Definition

Dependent Variable innovfail virtual variables, occurrence of ADRs in pharmaceutical products indicate innovation failure, innovfail = 1, otherwise, innovfail = 0

Independent Variable mrp results of time series principal component analysis

Regulated Variable gover the higher the value of the index “relationship between government and market”, the lower the government intervention

Control Variables size the natural logarithm of total assets, the larger of the value, the larger the scale of firms

growth the growth rate of main business income, (main business income of this year –main business income at the beginning of this year) / main business income at the beginning of this year

state virtual variable, state = 1 for state-owned firms, state = 0 for non-state-owned firms

roe net profit/average total assets, average total assets = total end-of-period balance of assets

tech r&d expenditure density = r&d expenditure / main business income, the greater the value, the higher the level of technical resources

duality the chairman and the general manager are the same one, duality = 1, otherwise, duality = 0

seperation the difference between control and ownership

Type	Name	Definition
Dependent Variable	innovfail	virtual variables, occurrence of ADRs in pharmaceutical products indicate innovation failure, innovfail = 1, otherwise, innovfail = 0
Independent Variable	mrp	results of time series principal component analysis
Regulated Variable	gover	the higher the value of the index “relationship between government and market”, the lower the government intervention
Control Variables	size	the natural logarithm of total assets, the larger of the value, the larger the scale of firms
	growth	the growth rate of main business income, (main business income of this year –main business income at the beginning of this year) / main business income at the beginning of this year
	state	virtual variable, state = 1 for state-owned firms, state = 0 for non-state-owned firms
	roe	net profit/average total assets, average total assets = total end-of-period balance of assets
	tech	r&d expenditure density = r&d expenditure / main business income, the greater the value, the higher the level of technical resources
	duality	the chairman and the general manager are the same one, duality = 1, otherwise, duality = 0
	seperation	the difference between control and ownership

Note: The data are collated by the authors.

3.3 Modelling

As the implementation of innovation activities by TMT may take a long time to think and discuss, and government intervention also has certain periodicity, the impact of TMT risk preference and government intervention on technological innovation failure may be lagging behind, therefore, the independent variables in the model, risk preference of TMT and regulatory variables, government intervention, are lagging behind for one period to reduce the interference of less endogenous problems to the results. According to the previous hypothesis and Wen et al. [35] test method of regulatory effect, and referring to Ma et al. [13] treatment method of binary dependent variable, the probit regression models are constructed as follows: $\begin{matrix} {innovfail}_{(i, t)} = α + β_{1} {mrp}_{(i, t - 1)} + β_{2} {size}_{(i, t - 1)} + β_{3} {growth}_{(i, t - 1)} \\ + β_{4} {state}_{(i, t - 1)} + β_{5} {roe}_{(i, t - 1)} + β_{6} {tech}_{(i, t - 1)} \\ + β_{7} {duality}_{(i, t - 1)} + β_{8} {seperation}_{(i, t - 1)} + μ_{(i, t)} \end{matrix}$ (2) $\begin{matrix} {innovfail}_{(i, t)} = α + β_{1} {mrp}_{(i, t - 1)} + β_{2} {gover}_{(i, t - 1)} \\ + β_{3} mrp * {gover}_{(i, t - 1)} + β_{4} {gover}^{2}_{(i, t - 1)} \\ + β_{5} mrp * {gover}^{2}_{(i, t - 1)} + β_{6} {size}^{2}_{(i, t - 1)} \\ + β_{7} {growth}_{(i, t - 1)} + β_{8} {state}_{(i, t - 1)} + β_{9} {roe}_{(i, t - 1)} \\ + β_{10} {tech}_{(i, t - 1)} β_{11} {duality}_{(i, t - 1)} \\ + β_{12} {seperation}_{(i, t - 1)} + μ_{(i, t)} \end{matrix}$ (3)

Among them, innovfail_(i,t) is the innovation failure status of the i-th sample firm in period t. mrp_(i,t-1), gover_(i,t-1), ${gover}_{(i, t - 1)}^{2}$ , mrp * gover_(i,t-1) and $mrp * {gover}_{(i, t - 1)}^{2}$ indicate the TMT risk preference index, the government intervention index, the square of the government intervention index, and the state of the government intervention regulating effect of the i-th sample firm in period t–1,respectively. size_(i,t-1), growth_(i,t-1), state_(i,t-1), roe_(i,t-1), tech_(i,t-1), duality_(i,t-1) and seperation_(i,t-1) represent the i-th sample firm in period t-1’s firm scale, firm growth, firm nature, firm asset return rate and firm technology resource level, respectively, whether the chairman and the general manager are the same one as well as the difference between control and ownership. To reduce the multicollinearity in the interaction effect model, the interaction terms are centralized in the process of regression analysis.

4 Result analysis

4.1 Regression analysis

Table 3 shows that in model (1), the regression coefficient of risk preference of TMT is 0.053, which is significant at the 1% level, indicating that risk preference of TMT has a significant positive impact on innovation failure, supporting hypothesis1. In model (2), government intervention and the square term of government intervention are further added. The regression coefficient of government intervention is 10.302, and the regression coefficient of the square term is –2.614. And all of them are significant at the level of 5%, indicating that the impact of government intervention on innovation failure is “inverted U” relationship. Because the measurement index of government intervention is reverse index, the lower the intensity of government intervention is, the greater the probability of technological innovation failure is. However, when the intensity of government intervention is lower than a certain critical value, government intervention can reduce the probability of technological innovation failure. Similarly, when government intervention intensity is higher than a certain critical value, government intervention will increase the probability of technological innovation failure of firms. In the model (3), the product of risk preference of TMT with government intervention and the product of risk preference of TMT with the square of government intervention are added. The regression coefficients of the product are –2.849 and 0.749 respectively, And they were all significant at the level of 5%, The results show that the regulatory effect of government intervention on risk preference and innovation failure of TMT is inverted U-shaped, which supports the hypothesis 2.

Table 3
Results of regression estimates for probit model

Dependent Variable: innovfail

(1) (2) (3)

mrp 0.053 0.055 2.737

(0.026) (0.027) (1.267)

gover 10.302 28.612**

(4.753) (12.554)

mrp*gover –2.849

(1.369)

gover² –2.614 –7.425**

(1.292) (3.392)

mrpgover² 0.749

(0.374)

size 0.267 0.183 0.179

(0.069) (0.074) (0.075)

growth –0.231 –0.326 –0.313

(0.244) (0.253) (0.252)

roe 1.514 2.364* 2.327*

(0.702) (0.802) (0.797)

state 0.1 –0.064 –0.071

(0.145) (0.154) (0.154)

tech 0.131 0.719 0.727

(1.639) (1.76) (1.761)

duality –0.105 –0.037 –0.041

(0.139) (0.143) (0.143)

seperation –3.088* –2.998* –2.971*

(0.778) (0.855) (0.848)

_cons –6.226* –14.48* –31.647***

(1.521) (4.626) (11.759)

N 480 480 480

Pseudo R² 0.057 0.099 0.101

	Dependent Variable: innovfail
mrp	0.053**	0.055**	2.737**
	(0.026)	(0.027)	(1.267)
gover		10.302**	28.612**
		(4.753)	(12.554)
mrp*gover			–2.849**
			(1.369)
gover²		–2.614**	–7.425**
		(1.292)	(3.392)
mrp*gover²			0.749**
			(0.374)
size	0.267***	0.183**	0.179**
	(0.069)	(0.074)	(0.075)
growth	–0.231	–0.326	–0.313
	(0.244)	(0.253)	(0.252)
roe	1.514**	2.364***	2.327***
	(0.702)	(0.802)	(0.797)
state	0.1	–0.064	–0.071
	(0.145)	(0.154)	(0.154)
tech	0.131	0.719	0.727
	(1.639)	(1.76)	(1.761)
duality	–0.105	–0.037	–0.041
	(0.139)	(0.143)	(0.143)
seperation	–3.088***	–2.998***	–2.971***
	(0.778)	(0.855)	(0.848)
_cons	–6.226***	–14.48***	–31.647***
	(1.521)	(4.626)	(11.759)
N	480	480	480
Pseudo R²	0.057	0.099	0.101

Note: *, ** and *** indicate the significant level of 10%, 5% and 1% respectively, and the standard error SE in brackets.

In addition, in terms of control variables, firms size, return on assets and the difference between control and ownership have significant impact on innovation failure. Among them, the difference between control and ownership has a negative influence, that is, the increase of the difference between control and ownership can significantly reduce the probability of innovation failure of firms. However, firms scale and return on assets have a positive impact, that is, with the continuous expansion of firms scale and the increasing return on assets, the possibility of firms innovation failure will also increase. The reason for this result may be that with the expansion of firm scale and the increase of asset income, firms need to invest more operating costs in R & D, production, sales and other aspects, and with the expansion of personnel and organizational structure, firms may lose management efficiency, thus affecting the technological innovation activities of firms.

4.2 Robustness test

In order to verify the robustness of the regression results, this paper tests the relevant assumptions as follows:

Logit model is used as the change estimation method of robustness test. In Table 4, logit models (1)-(3) show the regression results of using logit model. The results show that the risk preference of TMT has a significant positive impact on innovation failure, and the cross product of risk preference of TMT, government intervention and government intervention square has a significant impact on innovation failure, which is consistent with the regression results of probit model, supporting the hypothesis 1 and 2.

To reduce the impact of firms ownership on the regression results, the samples were divided into state-owned and non-state-owned firms, and the Probit model was used for robustness test. In Table 4, according to the results of the non-state-owned firms Probit models (4)-(6), the risk preference of TMT has a significant positive impact on innovation failure, with a regression coefficient of 0.043 (P < 0.1). Moreover, the product of risk preference of TMT and government intervention, and the product of risk preference of TMT and square term of government intervention have significant impact on innovation failure, and the regression coefficients are –1.624 (P < 0.1) and 0.413 (P < 0.1), respectively. The results show that for non-state-owned firms, the probability of innovation failure will increase with the increase of risk preference of the TMT. And the regulatory role of government intervention between TMT’s risk preference and failure of firms’ technological innovation is an inverted U-shaped relationship. According to the results of state-owned firms Probit models (7)-(9), Although the symbols of each regression coefficient are consistent with the probit model before grouping, the risk preference of TMT has no significant effect on innovation failure. Moreover, the product of risk preference of TMT and government intervention, and the product of risk preference of TMT and square term of government intervention have no impact on innovation failure. In terms of control variables, it is basically consistent with probit model before grouping.

Table 4
Results of regression estimates for logit model with Probit regression results of different firms ownership properties

Dependent Variable: innovfail

Results of regression estimates for logit model Probit regression results of different firms ownership properties (4)-(6) are non-state-owned firms, (7)-(9) are state-owned firms

(1) (2) (3) (4) (5) (6) (7) (8) (9)

mrp 0.089* 0.092* 4.427 0.043* 0.042* 1.627 0.159 0.121 20.384

(0.048) (0.05) (3.105) (0.023) (0.024) (3.229) (0.097) (0.095) (28.347)

gover 16.57** 46.485* 18.647*** 28.966 7.711 134.199

(8.219) (26.461) (5.778) (22.115) (22.157) (179.708)

mrpgover –4.604 –1.624* –21.546

(3.318) (3.478) (28.214)

gover² –4.206* –12.067* –5.019*** –7.64 –1.124 –34.357

(2.221) (7.051) (1.584) (5.905) (5.611) (44.533)

mrp* gover² 1.211* 0.413* 5.662

(0.886) (0.927) (7.001)

size 0.435* 0.303 0.298 0.297* 0.239* 0.234* 0.095 0.13 0.097

(0.116) (0.122) (0.122) (0.075) (0.081) (0.082) (0.184) (0.176) (0.177)

growth –0.387 –0.537 –0.516 –0.43 –0.595* –0.577* 0.166 0.029 –0.038

(0.4) (0.414) (0.414) (0.285) (0.309) (0.309) (0.447) (0.456) (0.453)

roe 2.437 3.809* 3.751* 1.084 2.662 2.622** 2.365* 2.784* 2.695*

(1.175) (1.337) (1.332) (0.85) (1.027) (1.03) (1.474) (1.598) (1.547)

state 0.158 –0.099 –0.108 — — — — — —

(0.24) (0.259) (0.259)

tech 0.13 1.086 1.095 –1.121 –0.385 –0.388 3.785 2.722 2.095

(2.5) (2.631) (2.632) (1.63) (1.889) (1.891) (3.927) (4.093) (4.121)

duality –0.158 –0.044 –0.05 –0.168 –0.103 –0.101 0.169 0.3 0.327

(0.22) (0.226) (0.227) (0.161) (0.165) (0.166) (0.289) (0.315) (0.321)

seperation –5.022* –4.924* –4.89* –3.204* –3.346* –3.327* –2.711* –1.878* –1.804

(1.265) (1.348) (1.341) (0.894) (1.076) (1.078) (1.546) (1.58) (1.572)

_cons –10.142* –23.48* –51.555 –6.65* –22.597*** –32.589 –3.545 –15.133 –133.37

(2.51) (8.059) (24.899) (1.653) (5.596) (20.656) (3.852) (22.257) (180.556)

N 480 480 480 372 372 372 108 108 108

Pseudo R² 0.057 0.099 0.101 0.06 0.119 0.119 0.096 0.16 0.178

	Dependent Variable: innovfail
mrp	0.089*	0.092*	4.427	0.043*	0.042*	1.627	0.159	0.121	20.384
	(0.048)	(0.05)	(3.105)	(0.023)	(0.024)	(3.229)	(0.097)	(0.095)	(28.347)
gover		16.57**	46.485*		18.647***	28.966		7.711	134.199
		(8.219)	(26.461)		(5.778)	(22.115)		(22.157)	(179.708)
mrp*gover			–4.604*			–1.624*			–21.546
			(3.318)			(3.478)			(28.214)
gover²		–4.206*	–12.067*		–5.019***	–7.64		–1.124	–34.357
		(2.221)	(7.051)		(1.584)	(5.905)		(5.611)	(44.533)
mrp* gover²			1.211*			0.413*			5.662
			(0.886)			(0.927)			(7.001)
size	0.435***	0.303**	0.298**	0.297***	0.239***	0.234***	0.095	0.13	0.097
	(0.116)	(0.122)	(0.122)	(0.075)	(0.081)	(0.082)	(0.184)	(0.176)	(0.177)
growth	–0.387	–0.537	–0.516	–0.43	–0.595*	–0.577*	0.166	0.029	–0.038
	(0.4)	(0.414)	(0.414)	(0.285)	(0.309)	(0.309)	(0.447)	(0.456)	(0.453)
roe	2.437**	3.809***	3.751***	1.084	2.662**	2.622**	2.365*	2.784*	2.695*
	(1.175)	(1.337)	(1.332)	(0.85)	(1.027)	(1.03)	(1.474)	(1.598)	(1.547)
state	0.158	–0.099	–0.108	—	—	—	—	—	—
	(0.24)	(0.259)	(0.259)
tech	0.13	1.086	1.095	–1.121	–0.385	–0.388	3.785	2.722	2.095
	(2.5)	(2.631)	(2.632)	(1.63)	(1.889)	(1.891)	(3.927)	(4.093)	(4.121)
duality	–0.158	–0.044	–0.05	–0.168	–0.103	–0.101	0.169	0.3	0.327
	(0.22)	(0.226)	(0.227)	(0.161)	(0.165)	(0.166)	(0.289)	(0.315)	(0.321)
seperation	–5.022***	–4.924***	–4.89***	–3.204***	–3.346***	–3.327***	–2.711*	–1.878*	–1.804
	(1.265)	(1.348)	(1.341)	(0.894)	(1.076)	(1.078)	(1.546)	(1.58)	(1.572)
_cons	–10.142***	–23.48***	–51.555**	–6.65***	–22.597***	–32.589	–3.545	–15.133	–133.37
	(2.51)	(8.059)	(24.899)	(1.653)	(5.596)	(20.656)	(3.852)	(22.257)	(180.556)
N	480	480	480	372	372	372	108	108	108
Pseudo R²	0.057	0.099	0.101	0.06	0.119	0.119	0.096	0.16	0.178

Note: *, ** and *** indicate the significant level of 10%, 5% and 1% respectively, and the standard error SE in brackets.

To sum up, there is no substantial difference between the results of robustness test and the basic regression results, no matter whether the estimation method is changed or the samples are divided. Therefore, it can be considered that the result is robust.

4.3 Discussion

The results in Table 3 show that the increase of TMT’s risk preference can stimulate firms to carry out technological innovation activities to a certain extent. However, with the increase of technological innovation activities, the probability of technological innovation failure will increase. When the intensity of government intervention is lower, the government intervention has a positive regulatory effect on the positive impact of TMT’s risk preference on technological innovation failure. At this time, the positive impact of TMT’s risk preference on technological innovation failure is stronger. However, when the government intervention is lower than a certain critical value, the government intervention plays a negative role in regulating the positive impact of the TMT’s risk preference on the technological innovation failure. At this time, the positive impact of the TMT’s risk preference on the innovation failure becomes weaker. In the same way, when the intensity of government intervention is higher, government intervention plays a negative role in regulating the positive impact of TMT’s risk preference on technological innovation failure. At this time, the positive impact of TMT’s risk preference on technological innovation failure becomes weaker.However,when the government intervention is higher than a certain threshold, the government intervention plays a positive role in regulating the positive impact of TMT’s risk preference on the failure of technological innovation. At this time, the positive impact of TMT’s risk preference on the failure of technological innovation becomes stronger. It shows that in the current market reform process in China, the government can effectively reduce the failure probability of technological innovation due to TMT’s high risk preference and the lack of the firm’s own resources in the process of technological innovation, by means of incentive policies such as R & D fund subsidy, tax preference and financing support. However, government intervention may effectively reduce the probability of technological innovation failure in a short period of time. However, excessive government intervention will give the TMT a psychological illusion that there is a government guarantee in everything, so that government intervention increases the positive impact of TMT’s risk preference on the failure of technological innovation, and increases the risk of innovation failure. In addition, there is a certain degree of information asymmetry between the government and firms, so it is difficult for the government to control the degree of its intervention. Therefore, the government needs to be cautious about intervention in firms’ innovation behavior. On the other hand, as far as the firm itself is concerned, the preconditions to reduce the failure rate of technological innovation are building a TMT with the reasonable risk preference structure,improving firm’s technical level and strengthening firm’s technical resources.

Sample regression results in Table 4 show that the impact of TMT’s risk preference on technological innovation failure of firms and the regulatory role of government intervention are also different. The TMT’s risk preference has a significantly positive impact on technological innovation failure of non-state-owned firms, but the impact on technological innovation failure of state-owned firms is not significant. However, the regulatory effect of government intervention on the risk preference of non-state-owned firms’ TMT and technological innovation failure is inverted U-shaped, but no significant regulatory effect on state-owned firms. The reason for this may be that, compared with non-state-owned firms, state-owned firms tend to be larger in scale and more complex in organizational structure. In addition, due to China’s special national conditions, the internal and external environment that TMT face may be more complex than that of non-state-owned firms when making innovation decisions, so that TMT’s risk preference and technological innovation failure in state-owned firms is not significant. With more innate resource advantages, state-owned firms obtain more long-term and lasting national support. Therefore, the impact of government intervention on the risk preference of state-owned firms’ executive team and technological innovation failure is not significant. For the non-state-owned firms, due to the limited and periodic support from the government, the non-state-owned firms have to continue to carry out technological innovation so as to seek their own long-term development and maintain their long-term competitiveness in the market. When the TMT’s risk preference is higher, the more technological innovation activities the firm carries out, so failure rate of technological innovation. When the government intervenes in firms through R & D subsidies and tax incentives, it will also have an impact on the relationship between TMT’s risk preference and the failure of technological innovation.

5 Conclusions

5.1 Conclusions

In order to explore the impact of TMT’s risk preference on the failure of technological innovation, this paper makes an empirical study on the effect of government intervention being a moderating variable on the relationship between TMT’s risk preference and technological innovation failure. Taking Chinese pharmaceutical listed companies in 2011–2017 as the research sample, the following conclusions are drawn: (1) TMT’s risk preference has a significantly positive impact on the failure of technological innovation. With the increase of TMT’s risk preference, the risk of technological innovation failure will increase. (2) Government intervention has a significant regulatory role in the positive impact of TMT’s risk preference on technological innovation failure, and the regulatory role is inverted U-shaped relationship. Specifically, when government intervention is lower than a certain threshold, the regulatory role of government intervention in the positive impact of TMT’s risk preference on technological innovation failure will change from promotion to inhibition; when the intensity of government intervention is higher than a certain threshold, the regulatory role of government intervention in the positive impact of TMT’s risk preference on technological innovation failure will change from inhibition to promotion.

The contribution of this paper is mainly reflected in the following aspects: Firstly, it helps to grasp the risk preference of TMT, which is the key factor affecting the failure of firm technological innovation, provides a new perspective for the study of TMT and firm innovation, and expands the research perspective of TMT and innovation failure. Secondly, the pharmaceutical manufacturing industry is related to the national economy and people’s livelihood, and its innovation achievements are directly related to human survival and quality of life. According to China’s special national conditions, taking listed pharmaceutical manufacturing firms in China as samples, this paper studies the relationship between TMT risk preference, government intervention and technological innovation failure, which is helpful to help firms to build or improve the risk preference of TMT members Optimize the TMT and the government to intervene in firm innovation and reduce the failure rate of innovation through reasonable and perfect technology innovation guidance policies. Thirdly, in terms of sample data selection, different from the previous qualitative research methods such as questionnaire survey or case analysis, this paper focuses on pharmaceutical manufacturing listed firms,and uses Quantitative analysis is used to verify the relationship between TMT risk preference, government intervention and technological innovation failure. The results are more convincing than qualitative analysis, and expand the research on innovation failure Research means.

From the perspective of TMT’s risk preference and government intervention, the research results of this paper further deepen the understanding of the causes of firms’ technological innovation failure, and provide beneficial decision-making reference for improving firms’ innovation motivation and reducing the risk of technological innovation. However, the inadequacy of this study is that only pharmaceutical manufacturing firms are taken as samples, and the conclusions may be affected by the characteristics of the industry. Due to the limitation of data availability, the time span of sample data is only 7 years, which may also affect the stability of the conclusions. Therefore, in the future research, with the expansion of sample selection range and time, we will further analyze the internal relationship among TMT’s risk preference, government intervention and technological innovation failure of firms.

5.2 Implications

Based on the conclusion of this paper, we can draw the following management implications:

Firstly, the risk preference of TMT has a significant positive impact on the failure of technological innovation. Therefore, firms should reasonably construct and optimize the risk preference structure of their TMT members. In the process of selecting new executives or building a new TMT, we can make a basic judgment on their risk preference according to their personal age, gender, education background and personal risk assets. When building a new TMT, it is necessary to make reasonable collocation according to the difference of members’ risk preference. It is necessary to neutralize the gender and age of team members, and effectively combine the overall view of men and the care of women, as well as the creativity of young people and the rich experience of the elderly, so as to give full play to the keen observation and understanding ability of highly educated talents, set reasonable tenure, and give corresponding positions to TMT members with different risk preferences. In the process of firm innovation, we should strengthen the management and supervision of the risk preference of the members of the TMT, and strictly control the risk preference of the members of the TMT. For the existing team members in the firm, psychological tests and case scenario analysis can be held regularly to timely understand and control the subtle changes of their risk preference. If necessary, the existing team members with high risk preference can be adjusted to reduce the failure rate of firm technological innovation.

Secondly, government intervention has a significant inverted U-shaped moderating effect between TMT risk preference and technological innovation failure. Therefore, the government should establish a sound guidance policy for technological innovation. The government should strengthen the fund subsidy to the technological innovation projects of firms, reduce the proportion of subsidies for non innovation activities, increase the support of government special funds for technological innovation, strengthen the guiding role of the government, maximize the use of limited government funds, and strengthen the process supervision of subsidy funds, so as to prevent rent-seeking and moral style caused by asymmetric information in the use of funds risks and corruption. And according to different regions, different firms can formulate preferential tax policies, and firms can increase tax reduction or reduction according to the situation. Due to the stage and periodicity of the process of technological innovation, firms’ demand for funds varies greatly in different stages. Moreover, most of the technological innovation firms in China are small and medium-sized firms with relatively low credit guarantee level and high financing cost. The problems of financing constraints their technological innovation activities and increase the risk of technological innovation failure. Therefore, the government should increase the support of the existing policy banks for technological innovation firms, establish corresponding development banks, timely promote the establishment of information guarantee fund for small and medium-sized firms, improve the credit guarantee mechanism of firms, and provide financial services for technological innovation firms through the establishment of specialized scientific and technological financial institutions, and disperse firm technology through scientific and technological innovation insurance system In order to solve the problem of financing difficulties of small and medium-sized firms, the risk of innovation failure should be solved.

Footnotes

Acknowledgments

This research was supported by the National Social Science Fund of China (Grant No. 19BGL039).

References

Catozzella

and Vivarelli

, The possible adverse impact of innovation subsidies: some evidence from Italy, International Entrepreneurship & Management Journal 12(2) (2016), 1–18.

Goolsbee

, Does government R&D policy mainly benefit scientists and engineers?, American Economic Review 88(2) (1998), 298–302.

Abdel-Khalik

A.R.

, Prospect Theory predictions in the field: Risk seekers in settings of weak accounting controls, Journal of Accounting Literature 33(1-2) (2014), 58–84.

Mercedes

and Pablo

, The market failure and the systemic failure rationales in technological innovation systems, Research Policy 42(5) (2013), 1039–1052.

Antonelli

and Crespi

, Matthew effects and R&D subsidies: knowledge cumulability in high-tech and low-tech industries, Giornale degli Economisti e Annali di Economia 71(1) (2012), 5–31.

Mandl

, Berger

C.S.E.

and Kuckertz

, Do you plead guilty? Exploring entrepreneurs’ sense making-behavior link after business failure, Journal of Business Venturing Insights (5) (2016), 9–13.

Ucbasaran

, Lockett

, Wright

, et al., Entrepreneurial founder teams: factors associated with member entry and exit, Entrepreneurship Theory & Practice 28(2) (2010), 107–128.

Czarnitzki

, Hanel

and Rosa

J.M.

, Evaluating the impact of R&D tax credits on innovation: A microeconometric study on Canadian firms, Research Policy 40(2) (2011), 217–229.

D.W.

and Wu

, Top Management Team Heterogeneity, Government Intervention and Corporate Social Performance, Management Review 29(12) (2017), 84–93.

10.

Maslach

, Change and persistence with failed technological innovation, Strategic Management Journal 37 (2016), 714–723.

11.

Castellacci

and Lie

C.M.

, Do the effects of R&D tax credits vary across industries? A meta-regression analysis, Research Policy 44(4) (2015), 819–832.

12.

Athaide

G.A.

and Klink

R.R.

, Managing Seller-Buyer Relationships during New Product Development, Journal of Product Innovation Management T 26(5) (2010), 566–577.

13.

G.R.

and Li

L.X.

, Efficiency of financial contract, exit of firms and misplacement of resources, The Journal of World Economy 10 (2014), 77–103.

14.

Xiang

, Wu

and Yang

Y.J.

, Research on operation an made risk warning of firm’s sustainable innovation process, Science Research Management 31(3) (2010), 10–16.

15.

Zuo

J.J.

, Tang

Y.J.

and Ji

Z.C.

, Government Intervention, Marketization Reform and Corporate R&D Innovation, R&D Management 28(6) (2016), 80–90.

16.

Ian

, Peat

, Richard

, Jonathan

and Morgan

, Miles,Innovation system problems: Causal configurations of innovation failure, Journal of Business Research 69(11) (2016), 5408–5412.

17.

Cope

, Entrepreneurial learning from failure: An interpretative phenomenological analysis, Journal of Business Venturing 26(6) (2011), 604–623.

18.

Naldi

, Nordqvist

and Hellerstedt

, Entrepreneurial Orientation, Risk Taking, and Performance in Family Firms, Family Business Review 20(1) (2010), 33–47.

19.

Abhiroop

, Manpreet

and Almins

, Do corporate taxes hinder innovation?, Journal of Financial Economics 124(1) (2017), 195–221.

20.

Cucculelli

and Ermini

, Risk attitude, product innovation, and firm growth. Evidence from Italian manufacturing firms, Economics Letters 118(2) (2013), 275–279.

21.

Cain

M.D.

and McKeon

S.B.

, “CEO personal risk-taking and corporate policies”, Journal of Financial and Quantitative Analysis 51(1) (2013), 139–164.

22.

McGrath

R.G.

, Falling forward: Real options reasoning and entrepreneurial failure, Academy of Management Review 24(1) (1999), 13–30.

23.

Cressy

, Why do most firms die young?, Small Business Economics 26(2) (2006), 103–116.

24.

Lhuillery

and Pfister

, R&D cooperation and failures in innovation projects: Empirical evidence from French CIS data, Research Policy 38(1) (2009), 45–57.

25.

Ren

S.M.

and Lv

, The Financial Constraints, the Government Subsidies and the Factor Productivity:A Case Study on the Equipment-Manufacturing Firms in China, Management World 11 (2014), 10–23.

26.

Luo

X.F.

, An analysis of the factors influencing the decision-making of technological innovation in Firms, Science of Science and Management of S & T 20(6) (1999), 8–10.

27.

González

and Pazó

, Do public subsidies stimulate private R&D spending?, Research Policy 37(3) (2008), 371–389.

28.

Wang

X.L.

, Fan

and Yu

J.W.

, ERI INDEX of Marketization of China’s Provinces 2018 Report, Beijing: Social Sciences Academic Press (2019)

29.

Zhao

Y.J.

, Obstacles to Technological Innovation of Firms and Policy Suggestions, Contemporary Economics 17 (2007), 54–55.

30.

Bai

Y.T.

, Guo

J.E.

and Xi

Y.M.

, The effects of top management team’s risk preference diversity on strategic investment decision making: an experiment research, Nankai Business Review 10(12) (2007), 25–30+44.

31.

Belghitar

and Clark

E.A.

, The effect of CEO risk appetite on firm volatility: an empirical analysis of financial firms, Routledge (2012), 195–211.

32.

Zhang

Z.G.

, Cao

D.T.

and Zhang

, Does TMT Stability Affect Firm Technological Innovation Performance?-The Moderating Effect of Compensation Incentive and Human Relationship, Accounting Research 12 (2018), 107–128.

33.

Xiong

, Ye

J.M.

and Wang

P.J.

, Does the institutional environment affect the failed technological innovation in firms? Evidence from listed companies in China’s pharmaceutical manufacturing industry, Transformations in Business & Econmics 18(1) (2019), 60–80.

34.

and Liu

, The Effects of Government Intervention and Corporate Resources on Corporate Diversification, China Soft Science 1 (2010), 116–127.

35.

Wen

Z.L.

, Hou

J.T.

and Zhang

, A comparison of moderator and mediator and their applications, Acta Psychologica Sinica 37(2) (2005), 268–274.

	Dependent Variable: innovfail
	Results of regression estimates for logit model			Probit regression results of different firms ownership properties (4)-(6) are non-state-owned firms, (7)-(9) are state-owned firms
	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)
mrp	0.089*	0.092*	4.427	0.043*	0.042*	1.627	0.159	0.121	20.384
	(0.048)	(0.05)	(3.105)	(0.023)	(0.024)	(3.229)	(0.097)	(0.095)	(28.347)
gover		16.57**	46.485*		18.647***	28.966		7.711	134.199
		(8.219)	(26.461)		(5.778)	(22.115)		(22.157)	(179.708)
mrp*gover			–4.604*			–1.624*			–21.546
			(3.318)			(3.478)			(28.214)
gover²		–4.206*	–12.067*		–5.019***	–7.64		–1.124	–34.357
		(2.221)	(7.051)		(1.584)	(5.905)		(5.611)	(44.533)
mrp* gover²			1.211*			0.413*			5.662
			(0.886)			(0.927)			(7.001)
size	0.435***	0.303**	0.298**	0.297***	0.239***	0.234***	0.095	0.13	0.097
	(0.116)	(0.122)	(0.122)	(0.075)	(0.081)	(0.082)	(0.184)	(0.176)	(0.177)
growth	–0.387	–0.537	–0.516	–0.43	–0.595*	–0.577*	0.166	0.029	–0.038
	(0.4)	(0.414)	(0.414)	(0.285)	(0.309)	(0.309)	(0.447)	(0.456)	(0.453)
roe	2.437**	3.809***	3.751***	1.084	2.662**	2.622**	2.365*	2.784*	2.695*
	(1.175)	(1.337)	(1.332)	(0.85)	(1.027)	(1.03)	(1.474)	(1.598)	(1.547)
state	0.158	–0.099	–0.108	—	—	—	—	—	—
	(0.24)	(0.259)	(0.259)
tech	0.13	1.086	1.095	–1.121	–0.385	–0.388	3.785	2.722	2.095
	(2.5)	(2.631)	(2.632)	(1.63)	(1.889)	(1.891)	(3.927)	(4.093)	(4.121)
duality	–0.158	–0.044	–0.05	–0.168	–0.103	–0.101	0.169	0.3	0.327
	(0.22)	(0.226)	(0.227)	(0.161)	(0.165)	(0.166)	(0.289)	(0.315)	(0.321)
seperation	–5.022***	–4.924***	–4.89***	–3.204***	–3.346***	–3.327***	–2.711*	–1.878*	–1.804
	(1.265)	(1.348)	(1.341)	(0.894)	(1.076)	(1.078)	(1.546)	(1.58)	(1.572)
_cons	–10.142***	–23.48***	–51.555**	–6.65***	–22.597***	–32.589	–3.545	–15.133	–133.37
	(2.51)	(8.059)	(24.899)	(1.653)	(5.596)	(20.656)	(3.852)	(22.257)	(180.556)
N	480	480	480	372	372	372	108	108	108
Pseudo R²	0.057	0.099	0.101	0.06	0.119	0.119	0.096	0.16	0.178