The Influence Mechanism of Government Venture Capital on the Innovation of Specialized and Special New “Little Giant” Enterprises

Abstract

Specialized and special new “little giant” enterprises are characterized by specialization, refinement, uniqueness, and innovation. They have relatively strong innovation capabilities and enterprise vitality. However, they also face problems such as high innovation costs, long investment recovery cycles, and high risks of investment returns, which lead to information asymmetry and financing difficulties. Government venture capital is a policy fund provided by the government and established with the participation of local governments, financial institutions, and private capital. They can utilize fiscal policies to attract market funds and support the development of key industries. Therefore, in this study, the first through sixth batches of specialized and special new “little giant” enterprises listed on the A-share and New Third Board from 2013 to 2023 were taken as samples, and their investment behavior and investment effects were empirically studied using the multiple linear regression method. The investment behavior of government venture capital tends to target strategic emerging industries. The intervention of government venture capital can enhance the innovation of “little giant” enterprises and has an impact through the intermediary mechanism of R&D investment. This paper draws conclusions and puts forward relevant policy suggestions for supporting the development of “little giant” enterprises.

1. Introduction

Since the 18th National Congress of the Communist Party of China (CPC), we have upheld the central role of innovation in China’s overall modernization drive, emphasized that innovation is the primary driving force, fully implemented an innovation-driven development strategy, and strived to build a world-class scientific and technological power. In 2024, the “Implementation Opinions on Promoting the Innovation and Development of Future Industries”, jointly issued by seven departments, including the Ministry of Industry and Information Technology, proposed that major national science and technology projects and major science and technology research projects should be implemented, and 100 cutting-edge key core technologies should achieve breakthroughs to form 100 iconic products. On the one hand, innovation policies can inject new momentum into economic growth. Innovation can promote the transformation of traditional industries toward high-end, intelligent, and green development, improving the production and operational efficiency of enterprises. Moreover, it can promote the optimization and upgrading of the industrial structure. On the other hand, innovation is key to enhancing international competitiveness. We need to address key and core technologies and promote integrated innovation among midstream, downstream, and large-, small-, and medium-sized enterprises along the industrial chain. To support the innovation strategy, ensure the safety and integrity of the industrial chain, and enhance its resilience and risk resistance capacity, China has pledged to cultivate specialized and special new enterprises.

Specialized and special new enterprises refer to those with specialization, refinement, and novelty. Most of them focus on a specific segment of the industrial chain, with a clear business focus, relatively strong innovation ability, innovation vitality, and risk resilience. Specialized and special new “little giant” enterprises are selected leaders of small- and medium-sized enterprises (SMEs) [1], representing a higher-level category of specialized and special new enterprises. In 2023, the total revenue of state-level “little giant” listed enterprises reached CNY 1.26 trillion, accounting for approximately 1% of China’s GDP. “Little giant” enterprises have far-reaching significance in promoting China’s economic development and scientific and technological innovation. They are characterized by high innovation and R&D costs, long investment recovery cycles, and high risks of investment returns. They also face financing difficulties and need very patient capital investment. “Little giant” enterprises usually take human capital as their core asset, and many of their intangible assets cannot be easily measured. External financiers are often reluctant to provide financing to them because of the agency costs caused by information asymmetry and moral hazard. Relying only on the market for resource allocation will lead to a large credit rationing problem, making it difficult to support the innovation of “little giant” enterprises [2,3,4].

Government venture capital is a policy instrument used by the government to guide private capital investment in innovation and entrepreneurship, high and new technology, and other fields. It is a policy fund provided by the government and established with the participation of local governments, financial institutions, and private capital. Government venture capital uses fewer funds to leverage the influx of social capital, plays a financial leverage role, jointly helps the development of science and technology innovation SMEs with venture capital, and plays a powerful role in improving the supply of venture capital and overcoming market failures [5]. The exertion of this leverage effect enhances the attractiveness of the project by transmitting signals to the market. In addition, setting up a risk compensation mechanism can help reduce investor risk. For “little giant” enterprises, in addition to the credit endorsement of their own “Little Giant” recognition, the financing constraints they face also require government support to be addressed. Government venture capital can effectively encourage market investors to flow into these fields and promote the upgrading of the whole industry. Meanwhile, government venture capital, as a form of “patient capital”, supports long-term innovation. Compared to venture capital, which has a longer duration and weaker assessment of profitability, it is more suitable for the innovative development of “little giant” enterprises. However, government venture capital still faces a contradiction between government goals and market-oriented operations. Excessive government intervention in investment decisions and the setting of counter-investment ratios have led to their inability to effectively support the development of specialized, refined, distinctive, and innovative “little giants.” Therefore, their specific impacts still need further verification.

On the basis of the functions and objectives of government venture capital, this study focused on their investment behavior and effect, and multiple linear regression was used to conduct empirical research. First, this paper explores the investment effect of government venture capital and verifies whether its intervention can effectively improve the innovation of “little giant” enterprises. It also studies its mechanism of action, whether it influences the output of innovation results through the intermediary effect of R&D input, and discusses its regional heterogeneity and industry heterogeneity, as well as its synergistic effect with traditional policy measures such as government subsidies. Finally, this study proposes corresponding policy recommendations to improve the policy system to support the innovation and development of “little giants”. This article helps to clarify the mechanism through which government venture capital promotes the development of “little giant” enterprises and identifies the specific conditions for the exertion of its effect, such as industries and regions. It also provides valuable suggestions for state support of small- and medium-sized enterprises, which have both theoretical and practical significance.

2. Literature Review

Many scholars have confirmed the driving effect and policy guidance effect of government venture capital on the innovation of new technology enterprises. When assessing its policy guidance effect, the ability to attract finance is often used as a key metric. For instance, Mazhar Islam et al. [6] found that clean energy start-ups receiving GVC are significantly more likely to secure subsequent venture capital (VC) funding compared to those without such subsidies. Massimiliano Guerini and Anita Quas [7] demonstrated that GVC increases the possibility of obtaining private venture capital. Some scholars have conducted further research about the specific manifestations of this policy guidance effect. Minli Yang et al. [8], based on provincial-level venture capital data from 2000 to 2011, reported that in provinces with less-developed VC markets, GVC plays a guiding role in attracting social capital. Xiaomin Ni [9], focusing on the mechanisms of policy effects, found that subsidies provided by government venture capital substantially enhance investment returns and mitigate the risks for social capital investing in start-ups, thereby promoting follow-on financing.

Research on the impact of government venture capital on enterprises includes both financing improvements and the allocation of internal resources [10]. This is reflected in the subsequent governance of the enterprises receiving such funding. Yuchen Li [11] argued that GVC can attract more private and other public funds, consequently promoting green innovation within firms. Bertoni’s [12] research on joint venture investments revealed that while the innovation performance of firms receiving only GVC support does not show significant improvement, firms receiving both GVC and private VC exhibit superior innovation performance compared to firms receiving no investment. This highlights the conditional nature of GVC’s effectiveness. Yael Alperovitch et al. [13] explored factors influencing the achievement of innovation objectives by European GVC-backed firms, finding significant impacts from location selection, co-working arrangements, and joint investment. Research on specialized and special new “little giant” enterprises—a concept originating from Germany’s “Hidden Champions” and serving a crucial strategic role in “strengthening and supplementing industrial chains”—includes findings that the “Little Giant” certification itself stimulates enterprise innovation [14]. This certification promotes innovative performance by alleviating financing constraints [15]. Furthermore, scholars have combined government venture capital with “little giant” enterprises. Zihan Yang [16], employing a difference-in-differences regression model, indicated that participation by GVC stimulates innovation within “little giant” enterprises.

However, some scholars hold different views about the policy guidance and innovation-driving effects of government venture capital. Cumming and MacIntosh [17] believe that government venture capital is often motivated by non-market-oriented objectives, leading to deviations from the optimal allocation of market resources. Such distorted investment behavior exerts a “crowding-out effect” on social capital. Yuejia Zhang [18] found that start-ups receiving backing from hybrid syndicates (including government-backed entities) in their initial financing rounds are significantly less likely to secure follow-up financing in subsequent rounds compared to those funded solely by private VC syndicates. Discussions also address practical reasons for the emergence of such inhibiting effects. Yuchen Li [19] pointed out that government funds suffer from regional development imbalances, excessive government-led investment, and a failure to adhere to market-oriented operational principles. Additionally, government venture capital can restrict the role of strategic limited partners (LPs) on investment committees, thereby intervening in decision-making.

Compared to other investment funds, there is still some controversy in the existing research and discussion on whether government venture capital plays a positive role in guiding and promoting the innovation of new technology enterprises, and the actual effect of government venture capital on the innovation of new technology enterprises needs further investigation. Moreover, there are few studies on the investment strategy of government venture capital, and research on investment behavior and the investment effect has not been organically combined. The existing research literature from the perspective of specialized and special new “little giant” innovation ecology is also relatively insufficient, as is the research mechanism. Therefore, in this study, specialized and special new “little giant” enterprises were taken as samples, focusing on their investment behavior and effects, and an empirical study on the impact of government venture capital on the innovation of specialized and special new “little giant” enterprises was conducted to supplement the empirical evidence in this field.

3. Theoretical Basis and Hypothesis

For “little giant” enterprises, their relatively small scale, the inherent characteristics of their scientific and technological innovation activities, and the lack of traditional mortgages contribute to a serious problem of information asymmetry between these enterprises and market investors. This leads to corresponding market failures, resulting in enterprises falling into financing difficulties and limiting their further sustainable development [20]. Due to information asymmetry, enterprises cannot obtain sufficient external financing sources to support their professional and refined development path, and their production and operation are constrained by cash flow, which may lead to a return to traditional inefficient diversified operations or the crisis of bankruptcy. At the same time, the state has positioned “little giant” enterprises as the policy focus of supply-side structural reform, an important stabilizer of the new development pattern, and a new force to build an innovative country, which is necessary for macro-control. At present, “little giant” enterprises are generally highly concentrated in the main business, which is a double-edged sword that can not only concentrate the advantages of resource endowment but also fully realize the exchange of innovation resources, which greatly improves the possibility of breakthrough innovation. However, it can also increase the risk of failure, and early innovation investment may be wasted. In addition, in the principal–agent corporate structure, the information asymmetry between executives and shareholders affects the innovation activities of enterprises. The absence of a reasonable corporate governance mechanism is likely to lead to principal–agent problems, causing executives to ignore the long-term interests of enterprises and focus on the modification of financial indicators.

The theory of government intervention directly aims at the market failure caused by information asymmetry. It proposes exerting influence on resource allocation through government venture capital and other government intervention methods to encourage the sustainable development of “little giant” enterprises. According to signal transmission theory, government venture capital is a policy fund, which is funded by the government and endorsed by credit, and the development potential of the enterprise is screened by professional fund management institutions, which invest in specialized and special new “little giant” enterprises in the form of equity participation in sub-funds or follow-up investment. Based on the resource-based view, agency theory, and the institution-based view, by shedding light on how the different characteristics of government venture capital affect alternative energy production innovation, policymakers need to strike a balance between government intervention and market mechanisms [21]. Investment behavior serves as evidence that “little giant” enterprises have room for value-added profitability, moving beyond policy documents to leverage social capital. Government venture capital helps enterprises obtain credit financing, benefit from implicit credit endorsement, and receive policy guidance. It also plays a role in signal transmission, alleviates information asymmetry, and guides social capital investment to support “financing hematopoiesis,” thereby alleviating the practical dilemmas of financing difficulties and high costs faced by specialized and new “little giant” enterprises. Government venture capital can also play a positive role in helping investee companies improve their social identity and transform their reputation into a reduction in transaction costs and expansion of the sales market, which is conducive to alleviating the internal financing problem.

In addition, by relying on national policy guidance, government venture capital attracts technology, human resources, and other factors to invest in enterprises, thus promoting their spatial aggregation. In a highly uncertain market environment, the positive guiding effect of government venture capital on entrepreneurs’ mentality and behavior can help entrepreneurs cope with various uncertainties and reduce their cognitive pressure to some extent. Government venture capital can help enterprises establish good relationships with the government. A good government–enterprise relationship makes it convenient for enterprises to obtain the policy support they need to help them better realize cross-industry and cross-regional development of entrepreneurial projects and make diversified investments. Government venture capital can guide social capital to form a capital pool with high resilience to risk and rich factor endowment, and the nonfinancial resources formed have a certain positive pulling effect on financial support and transaction flow opportunities. The advantages of heterogeneous resources are that they can help enterprises carry out innovative reforms on productivity factors such as the labor force and labor objects, and they can improve the innovation of enterprises.

Therefore, this paper proposes Hypothesis 1.

H1. 

Government venture capital intervention can improve the innovation of “little giant” enterprises.

Compared to market-oriented venture capital funds, government venture capital is characterized by higher risk tolerance and weaker profit-target orientation. With lower short-term capital recovery requirements, it can tolerate enterprises conducting high-risk R&D projects. This fosters a corporate culture of innovation and experimentation, eases managers’ short-term performance pressure, and encourages long-term R&D investment.

Moreover, government venture capital assists “little giant” enterprises in establishing a scientific and comprehensive management model. This enables enterprises to allocate resources more efficiently and pay closer attention to the proportion of R&D expenditure during budget formulation and implementation. Aligned with the government’s macro-policy vision, the central financial fund supports “little giant” enterprises in increasing R&D investment centered on the “three innovations” (generating new drivers, tackling new technologies, and developing new products). Government venture capital, in turn, prompts enterprises to strengthen investment in these areas, ensuring policy implementation at the enterprise level.

Statistics show that the R&D intensity of “little giant” listed companies is 1.66 times that of the market average. Their innovation is manifested in product or service innovation, as well as in new technologies, processes, concepts, and models. “Little giant” enterprises either invest in preliminary basic research, focusing on cutting-edge theories to achieve scientific and technological breakthroughs and become “individual champions”, or pursue the transformation of scientific and technological achievements. They collaborate with universities, research institutions, and upstream and downstream enterprises in the industrial chain for joint R&D, jointly resolving bottlenecks in the industrial chain, and capitalizing on the rapid-iteration advantage of SMEs in new business forms.

By increasing R&D investment and allocating R&D expenditure to factors of production like labor, labor objects, and labor tools, these enterprises enhance their innovations through independent or joint innovation. This promotes the breakthrough of key bottlenecks in the industrial chain, achieving a healthy cycle and a strengthened chain.

Based on this, this paper proposes Hypothesis 2.

H2. 

Government venture capital will enhance the innovation of specialized and special new “little giant” enterprises by increasing R&D expenditure.

4. Materials and Methods

4.1. Data Sources and Processing

The first through sixth batches of specialized and special new “little giant” enterprises listed on the A-share and New Third Board from 2013 to 2023 were taken as samples, with a total of 13,205 sample data points. Among them, the investment events of government venture capital were sourced from the Zero2IPO private equity database, the relevant innovation data came from the China Research Data Service Platform (CNRDS), and other financial indicators were obtained from WIND or CSMAR. The following steps were taken to ensure the reliability of the research data: first, STs, *STs, and PT enterprises were excluded; second, samples with incomplete data or extreme outliers were eliminated; and finally, the continuous variables were winsorized by 1% on both sides.

4.2. Variable Selection

4.2.1. Selection of Explained Variables

Drawing on the practices of Ahuja [22], the number of patent applications of an enterprise in the current year was taken as the index to measure their innovation. Since it usually takes several years for an enterprise to apply for patents to be granted, this leads to time dislocation. Therefore, the number of patent applications can more accurately reflect the timeliness of innovation than can the number of patent grants.

4.2.2. Selection of Explanatory Variables

Government venture capital intervention. This variable is the interaction term of the time dummy (pre) and the policy variable (stock right). The policy variable is whether the enterprise belongs to the sample invested in by the guided fund. If it belongs to the sample, the value is 1; if it does not belong to the sample, the value is 0. The time dummy variable captures whether the observation occurs after the guided fund investment event, taking a value of 1 after the investment event and 0 before.

R&D investment. This variable is the ratio of a firm’s R&D expenditure to its operating revenue in the current year and measures the intensity of R&D investment.

4.2.3. Selection of Control Variables

This analysis also controlled for the total assets scale (Size), which is represented by the total assets of the enterprise; net profit margin of total assets (ROA), which is the ratio of net profit to total assets; asset–liability ratio (Lev), which is the ratio of liabilities to assets; profit on equity (ROE), which is the ratio of net profit to net assets; and company age (Age), which refers to the total number of years from the establishment of the enterprise to the data year. This analysis also controlled for time and individual fixed effects.

The main variables of this paper are shown in

Table 1. Definitions of the main variables.

Type of Variable	Variable	Variable Abbreviation	Variable Description	Data Source
Explained variable	Level of innovation	Pro	Number of patents filed during the year	CNRDS
Mediating variable	R&D expenditure	Tec	Tec A firm’s R&D expenditure as a percentage of revenue for the year	WIND
Explanatory variables	Government venture capital involvement	T	Interaction term of the time dummy (pre) and policy variable (stock right)	ZERO2IPO PRIVATE CONNECT
	Total assets	Size	Total assets of the enterprise at the end of the year	CSMAR DATABASE/WIND
Control variables	Net profit margin on total assets	ROA	Net profit/total assets	CSMAR DATABASE/WIND
	Asset–liability ratio	Lev	Fixed assets/total assets	CSMAR DATABASE/WIND
	Net asset profit margin	ROE	Net profit/net assets	CSMAR DATABASE/WIND
	Company age	Age	The time lag between the establishment of the business and the current year	CSMAR DATABASE/WIND
Dummy variables	Year dummy variable	YEAR	Year
	Individual dummy variables	ID	Individual code of each business

.

4.3. Model Design

To investigate the industry tendency of government venture capital investment and the influence of government venture capital intervention on the innovation of specialized and new “little giant” enterprises, the following model was constructed:

Model 1 Pro_it = C + αT_it + βX_it + y_i + λ_t + ε_it

(1)

Model 2 Pro_it = C + αT_it + βX_it + y_i + λ_t + ε_it

Tec_it = C + αT_it + βX_it + y_i + λ_t + ε_it

(2)

Both Models 1 and 2 include bidirectional fixed effects. The subscripts i and t, respectively, represent the t-th year of the i-th sample; T represents the intervention of the government-guided fund; Pro represents the innovation level or the invention level of the enterprise; Tec represents the innovation investment of the enterprise; X represents the control variable; y is the time fixed effect; λ is the individual fixed effect; ε is a random disturbance term; and C is a constant term.

5. Results

5.1. Descriptive Statistics

The results of the descriptive statistics are shown in the table below. As shown in

Table 2. Descriptive statistics.

Variable Name	Symbol	Observations	Mean	Standard Deviation	Maximum	Minimum
Level of innovation	Pro	13,205	17.45627	25.21033	620	1
Government venture capital steps in	T	13,205	0.1166225	0.3209821	1	0
Total assets	Size	13,205	1.60 × 109	4.53 × 1010	3.46 × 1012	1,967,227
Asset–liability ratio	Lev	13,205	0.3966455	1.330817	49.687	0
Net profit margin on total assets	ROA	13,205	0.2192176	15.33799	1760.994	−21.6976
Profit margin on net assets	ROE	13,205	0.3309636	24.55013	−79.40533	2815.416
Year of establishment	Age	13,205	16.78409	5.890615	44	1

, the mean value of innovation is 17.45627, the maximum value is 620, and the minimum value is 1, indicating large differences in the number of patent applications of the different sample enterprises. The enterprise with 620 patent applications is among the leading enterprises in the technology industry and demonstrates a high level of innovation. The mean value for government venture capital involvement is 0.1166225, indicating that only a few listed “little giant” enterprises have received such investment, while the majority have not. Therefore, government venture capital still requires follow-up participation from market investors. The standard deviation of total assets is 4.53 × 10¹⁰, indicating a large gap in asset scale. The average asset–liability ratio is 0.3966455, indicating that the overall debt level is not high and that the borrowing capacity of “little giant” enterprises needs to be improved. The highest asset–liability ratio is close to 50, and the lowest is 0, indicating that there are significant differences in the borrowing capacity and working capital management strategies among enterprises. There is a large gap between the maximum and minimum profit margins of total assets and net assets, and there are obvious differences in profitability among different enterprises. The mean value of the establishment year is 16.78409, the standard deviation is 5.890615, the maximum value is 44, and the minimum value is 1, indicating that the establishment time of the enterprise is relatively concentrated and strongly affected by the macro-economic environment.

5.2. Analysis of the Empirical Results

5.2.1. Regression Results of Model 1

Table 3. Regression results of Model 1.

		Pro
	(1)	(2)	(3)	(4)
T	8.9405 ***	4.3939 ***	8.9746 ***	4.4047 ***
	(13.0634)	(3.9526)	(13.0971)	(3.9613)
Size			0.0000	0.0000
			(0.3106)	(0.4659)
Lev			−0.0002	−0.0537
			(−0.0014)	(−0.1413)
ROA			−0.9830 ***	−0.3039
			(−3.1329)	(−1.1036)
ROE			0.6119 ***	0.1892
			(3.1206)	(1.0992)
Age			−0.0065	−5.2938 **
			(−1.1704)	(−2.0053)
ID	NO	YES	NO	YES
YEAR	NO	YES	NO	YES
_cons	16.4777 ***	17.0096 ***	16.6006 ***	110.0661 **
	(70.3910)	(86.2263)	(63.6481)	(2.3725)
N	13,205	13,205	13,205	13,205
adj.R2	0.013	0.551	0.013	0.551

p statistics in parentheses: ** p < 0.01, *** p < 0.001.

shows the regression results of innovation (Pro) on government venture capital involvement. The first and second columns display the regression results without control variables, while the third and fourth columns show the results after a series of control variables were added. The second and fourth columns control the two-way fixed-effect model, which includes time and individual fixed effects. The regression results indicate that government venture capital involvement has a positive effect on the enterprise’s innovation at the 0.01 significance level, and the results are stable to some extent. Therefore, Hypothesis 1 can be considered valid.

5.2.2. Regression Results of Model 2

Table 4. Regression results of Model 2.

	Pro	Tec
	(1)	(2)
T	4.4047 ***	0.6557 **
	(3.9613)	(2.0061)
Size	0.0000	−0.000 ***
	(0.4659)	(−4.2002)
Lev	−0.3039	0.0415
	(−1.1036)	(0.5769)
ROA	−0.0537	−0.5404 ***
	(−0.1413)	(−3.7787)
ROE	0.1892	−0.3312 ***
	(1.0992)	(−6.9571)
Age	−5.2938 **	−0.1418
	(−2.0053)	(−0.6600)
ID	YES	YES
YEAR	YES	YES
_cons	110.0661 **	17.7610 ***
	(2.3725)	(3.7922)
N	13,205	13,205
adj.R2	0.551	0.587

p statistics in parentheses: ** p < 0.01, *** p < 0.001.

shows the regression results of the mediating effect of R&D investment on the influence mechanism verified in model 1. The approach used in this study draws on the two-step method for mediating effects proposed by Jiang [23]. The first column shows the regression results of the innovation (Pro) on government venture capital involvement, and the second column shows the regression results of R&D investment (Tec) on government venture capital involvement. Since the effect mechanism of R&D investment on innovation in the two-step method was demonstrated in the previous hypothesis, no further regression verification was conducted. The results show that government venture capital has a positive effect on the innovation of specialized and innovative “little giant” enterprises at the 0.01 significance level, and government venture capital intervention increases the number of patent applications by 4.4047. At the 0.05 significance level, government venture capital intervention has a promoting effect on R&D investment. Therefore, Hypothesis 2 is verified.

5.3. Robustness Check

5.3.1. Parallel Trend Test

According to the basic assumptions of the DID estimation method, there should be no significant differences between the control group and the treatment group. Therefore, before policy intervention, the change trends in the two groups should be consistent. On this basis, six dummy variables were defined: pre3, pre2, pre1, current, post1, and post2, representing the three years before, the year of the intervention, the year after, and two years after the intervention event of the guided fund, respectively. The event study method was used to identify the change trends in the control and treatment groups. The specific model is as follows: Pro_it = C + ∑_mαD^m_it + βX_it + y_i + λ_t + ε_it, where D^m_it is a time dummy variable that indexes the change trends of the samples from three years before the intervention to two years after the intervention. According to the test results shown in Figure 1, there was no significant difference between the treatment and control groups before government venture capital intervention. Based on parallel trend testing, the DID estimation results are therefore reliable. In the year of the intervention, no significant differences were observed between the two groups. However, in the post1 and post2 phases, the treatment group exhibited significantly greater innovation compared to the control group. This may be because government venture capital targeting “little giant” specialization has a delayed effect on corporate governance. Its “guiding” role and the signal it transmits to the market may also take time to materialize. In addition, in some cases, government venture capital investments occur later, and the data are incomplete after the policy is implemented, which has a certain effect on the results of the parallel trend test.

5.3.2. Placebo Test

To exclude the influence of other policies or random factors on the innovation of enterprises during the sample period, enterprises were randomly selected from the sample, the time of their guided fund intervention was randomly generated, and then a randomized trial was conducted at the individual–time level to generate virtual policy variables. The above process was repeated 1000 times, and the distributions of the estimated coefficients of the virtual policy variables were drawn, as shown in the figure below. According to Figure 2, most of the estimated coefficients obtained under random treatment are distributed around the value of 0, and the p-value is mostly greater than 0.1, which is not significant at the level of 10%, meaning that the benchmark regression model did not miss enough important variables. In other words, the impact effect estimated by the benchmark regression model can be regarded as the result of the involvement of government venture capital, and the core conclusion is robust.

5.3.3. Replacing Explained Variables

At the level of industrial structure, “little giant” enterprises undertake the responsibilities of completing the industrial chain and promoting the technological breakthrough of the “stuck neck”. Exploratory innovation measures the active exploration of enterprises in new fields and technologies and shows the path of professional innovation. Following existing innovation-related research, a patent was identified as containing technical knowledge if the enterprise applied for a patent under a new IPC classification number (four). If the IPC classification number of a newly applied patent in the current year had not appeared in the enterprise’s patent applications over the past five years, we classified it as exploratory innovation (Pro2).

The patents of an enterprise were classified into three types: invention patents, design patents, and utility model patents. Among them, invention patents best represent the comprehensive innovation ability and technological level of an enterprise. When an enterprise focuses on technological research and development, the proportion of invention patents in the total number of patents increases. Therefore, this article also selected the proportion of invention (Pro3) patents in the total number of patents of an enterprise as an indicator to measure the innovation level.

To test the robustness of the empirical results, this study replaced the explained variable for exploratory innovation and the level of invention and verified the robustness of the empirical results. The empirical results are presented in

Table 5. Robustness checks.

		Pro2		Pro3		One-Period Lagged Pro
	(1)		(2)	(3)	(4)		(5)
T	2.6130 ***		2.6176 ***	0.0373 ***	0.0383 ***		2.3951 **
	(3.6052)		(3.6105)	(2.6209)	(2.6971)		(2.0166)
Size			0.0000		0.0000		0.0000 ***
			(0.2755)		(0.3930)		(9.2007)
Lev			−0.0754		−0.0085 *		−0.1081
			(−0.3042)		(−1.6645)		(−0.2769)
ROA			−0.3151 *		−0.1273 ***		8.8737
			(−1.7548)		(−6.9288)		(1.1915)
ROE			0.1975 *		0.0053 *		3.7312
			(1.7595)		(1.6788)		(0.8897)
Age			−1.9998		−0.0001		−4.3755 *
			(−1.1618)		(−0.4060)		(−1.7331)
ID	YES		YES	YES	YES		YES
YEAR	YES		YES	YES	YES		YES
_cons	9.7864 ***		44.9641	0.2851 ***	0.3292 ***		35.5850 *
	(76.0903)		(1.4865)	(40.6059)	(34.1230)		(1.9206)
N	13,061		13,061	13,061	13,061		9154
adj.R2	0.544		0.544	0.576	0.576		0.286

p statistics in parentheses: * p < 0.05, ** p < 0.01, *** p < 0.001.

.

5.3.4. Lagging the Explained Variable by One Period

Considering that the results of the guided fund policy may have a certain lag effect, the stability of the regression results was tested by lagging the innovation of the explained variable by one period. According to the regression results, the promotion effect of the guided fund intervention on the lagged innovation is significant at the 0.05 significance level. Therefore, it can be concluded that the guided fund intervention has a robust promoting effect on the innovation of “little giant” enterprises. The empirical results are presented in Table 5.

5.3.5. Endogeneity Issues

To address potential self-selection bias in our model examining the impact of government venture capital on corporate innovation, according to Jing [24] and Jing [25], we employed the Heckman two-stage model for correction. We believe that the endogeneity problem of mutual causality can be effectively solved through one-period lagging regression and the Heckman two-stage model. The first-stage selection model is specified as follows: prob{T_it=1} = C + C1Ind_it + C2Lev_it + ε_i, where T_it is a dummy variable indicating whether a firm received government-guided fund investment in year t (1 if yes, 0 otherwise). Given that government venture capital tends to focus on strategic emerging industries, we used whether a firm belongs to a strategic emerging industry (Ind_it) as a key determinant of Tit. At the firm level, higher leverage ratios make firms less attractive to investors, so we included the asset–liability ratio (Lev_it) as another determinant of investment decisions. The first-stage regression results, presented in Column (1) of

Table 6. The Heckman two-stage model.

	T	Pro	Pro
	(1)	(2)	(3)
T		4.3913 ***	4.4126 ***
		(3.9495)	(3.9637)
imr		−0.1474	−5.4768
		(−0.1342)	(−0.1482)
size			0.0000
			(0.4686)
ROA			−0.3052
			(−1.1077)
ROE			0.1900
			(1.1032)
AGE			−5.3079 **
			(−2.0092)
Lev	−0.3522 **		1.8398
	(−2.5495)		(0.1439)
Ind	0.4200 ***
	(7.5282)
ID		YES	YES
YEAR		YES	YES
_cons	−2.0542 ***	17.3679 ***	122.8621
	(−34.8093)	(6.4854)	(1.2535)
N	13,205	13,061	13,061
adj. R2		0.551	0.551

p statistics in parentheses: ** p < 0.01, *** p < 0.001.

, show that being in a strategic emerging industry significantly increases the probability of receiving government-guided fund investment, while higher leverage ratios significantly decrease this probability. In the second stage, we added the inverse Mills ratio (imr) obtained from the first stage as a control variable in our main regression model to test for potential sample selection bias. The results in Column (3) show that the coefficient on imr is statistically insignificant, indicating the absence of self-selection bias in our sample.

5.4. Further Discussion

5.4.1. Regression by Industry

According to legitimacy theory, the institutional environment causes compliance pressure on the survival of organizations, and the pursuit of legitimacy forces organizations to constantly adjust their internal systems to adapt to the external environment [26]. Government venture capital is faced with greater pressure for legitimacy, meaning its investment behaviors must comply with the external policy environment. Strategic emerging industries are the key development directions determined by the state, such as new-generation information technology, biology, high-end equipment manufacturing, and new energy. Government venture capital that focuses on these fields can effectively facilitate the adjustment, transformation, and upgrading of the national industrial structure; drive the coordinated development of upstream and downstream related industries; inject strong momentum into economic growth; help optimize the allocation of social resources; avoid the blind flow and waste of funds; and improve the economic efficiency of the whole society. From the perspective of investment efficiency, government venture capital that absorbs state-owned assets and raises social funds needs to consider the preservation of state-owned assets and returns on investment of market-oriented shareholders [27]. Therefore, when selecting appropriate investment projects, investment management agencies consider the growth and long-term profitability of enterprises, which aligns with the investment rationale of venture capital funds [26]. Strategic emerging industries have strong growth, are in line with policy guidance and the market outlet, have high development potential, and have sufficient space for value growth. Government venture capital should be a kind of patient capital that is able to withstand short-term fluctuations, bear short-term unstable business risks, support such SMEs through initial difficulties, and reap long-term investment returns. Therefore, from the perspective of policy and efficiency, government venture capital is motivated to invest in strategic emerging industries, which drives its investment behavior.

Therefore, we first examined whether the investment behavior of government venture capital has an investment orientation, that is, whether government venture capital tends to be invested in strategic emerging industries. Studying the preferences of government investment behavior is beneficial for identifying industry differences in the government’s provision of resources and governance to invested enterprises, thereby gaining a further understanding of the factors influencing innovation outcomes.

Whether an enterprise belongs to a strategic emerging industry is defined as follows. Industry information at levels 1–3 was extracted to form a text database, which was then searched for keywords such as energy conservation and environmental protection, information technology, biology, high-end equipment manufacturing, new energy, new materials, and new energy. A comprehensive judgment was made based on the presence of these keywords. The classification standard for strategic emerging industries was based on the decision of the State Council on Accelerating the Cultivation and Development of Strategic Emerging Industries.

Table 7. Industry tendency regression results.

Variables	T
	(1)	(2)
Ind	0.2283 ***	0.2147 ***
	(7.7044)	(7.1822)
Size		−0.0000
		(−0.1659)
Lev		−0.2589 ***
		(−4.0872)
ROA		−0.5330 ***
		(−3.8112)
ROE		0.0061
		(0.4211)
Age		0.0015 **
		(2.4207)
_cons	−1.2725 ***	−1.1714 ***
	(−70.7313)	(−32.7977)
N	13,205	13,205
adj.R2

p statistics in parentheses: ** p < 0.01, *** p < 0.001.

shows the regression results of guided fund involvement on whether an enterprise belongs to a strategic emerging industry.

According to the above division of the industries in which “little giant” enterprises are located, potential differences in the impact of government venture capital intervention on enterprises in the two types of industries were also explored. The innovation activities carried out by “little giant” enterprises have a serious information asymmetry. An enterprise’s expansion of research or application of basic theories can benefit the output of upstream and downstream enterprises, but it faces a greater risk of failure. In technology-intensive industries such as strategic new industries, the positive externality caused by information asymmetry is more serious. Many enterprises have high development potential and have carried out good research projects, but they do not receive the corresponding attention from investors. Therefore, guided fund intervention can play a more significant role in solving the problem of information asymmetry through a signal transmission mechanism, alleviating financing constraints and increasing R&D investment, and because of its long-term governance effect, it will be of great help to the basic innovation of strategic emerging industries. According to the regression results in

Table 8. Industry heterogeneity analysis.

	Non-Strategic Emerging Industries	Emerging Sectors of Strategic Importance	West		Middle	East	Synergistic Effect
				Pro
T	3.1301 **	5.5494 ***	0.8677		9.6053 ***	3.0019 **
	(2.2834)	(2.8738)	(0.2137)		(3.7597)	(2.3047)
T×Gov							0.0023 ***
							(2.7664)
Size	0.0000	0.0000 ***	−0.0000		0.0000 ***	0.0000	0.0000 *
	(0.4213)	(4.5532)	(−0.3079)		(6.0838)	(0.4116)	(1.8170)
Lev	0.3458	−0.1951	−25.7313 **		10.5238 *	−0.3138	−0.3771
	(0.5183)	(−0.3933)	(−2.1307)		(1.8656)	(−0.9189)	(−0.8757)
ROA	0.3467	−0.6080	0.0508		3.4801	−0.1838	−0.8069 *
	(0.4154)	(−0.7083)	(0.0665)		(1.0995)	(−0.4154)	(−1.7220)
ROE	−0.2180	0.2415	12.1444 **		−0.1107	0.1953	0.2330 **
	(−0.4175)	(1.2182)	(2.1405)		(−0.3265)	(0.9145)	(2.4027)
Age	−4.5913 *	0.0000	0.0000		0.0000	−5.2270 **	0.0000
	(−1.7375)	(.)	(.)		(.)	(−1.9763)	(.)
ID	YES	YES	YES		YES	YES	YES
YEAR	YES	YES	YES		YES	YES	YES
_cons	96.0276 **	15.7794 ***	15.7153 ***		11.4225 ***	111.8758 **	15.6420 ***
	(2.1034)	(26.3853)	(17.4464)		(8.5927)	(2.3473)	(67.1097)
N	8863	4191	1138		2355	9567	9513
adj.R2	0.558	0.542	0.616		0.533	0.551	0.572

p statistics in parentheses: * p < 0.05, ** p < 0.01, *** p < 0.001.

, guided fund intervention plays a more significant role in promoting the innovation of strategic emerging industries.

5.4.2. Regression by Region

Owing to the significant differences in economic level, concentration of innovation factors, and the policy environment in different regions, the promotion effect of guided fund intervention on the innovation of “little giant” enterprises also differs regionally. Therefore, according to the division method of the National Bureau of Statistics, this analysis divided the region into three types—east, middle, and west—and classified the invested enterprises into these three types of regions. According to the regression results in Table 8, the promotion effect of the western region is not significant, possibly because the venture capital market in the western region is not active enough. Therefore, guided fund intervention cannot fully exploit the leverage effect of funds, and given the lack of innovation factors, the overall innovation output is low. At the significance level of 0.01, the government venture capital in the central region has a promoting effect on the innovation of specialized and special “little giant” enterprises, and the coefficient is 9.6053, indicating that the industrial structure in the central region is relatively good. Therefore, the venture capital market has a certain degree of activity, and the allocation efficiency of innovation factors is good, leading to the government venture capital exerting its governance and leverage effects. The coefficient in the eastern region is relatively low, which may be due to the relatively developed economy in the eastern region and the frequent occurrence of investment behaviors in the venture capital market. The marginal leverage effect of the intervention of government venture capital on market funds is lower than that in the central region, so the promotion effect is weaker.

5.4.3. Synergies with Government Subsidies

Government subsidies are among the important tools of government macro-control, aiming to guide the rational allocation of resources and promote economic restructuring, industrial upgrading, social equity, and sustainable development. This traditional regulation method has been proven to play a role in stimulating and supporting the growth and development of enterprises, as well as R&D innovation, and even encourages venture capital funding similar to government venture capital [28]. Moreover, as a new regulatory tool, government venture capital needs to work synergistically with government subsidies in addition to playing a role in policy leveraging. Both government subsidies and government venture capital can create a macro-environment conducive to enterprise innovation activities through advanced support, follow-up supervision, and investment guidance strategies, thus promoting enterprise R&D innovation. There are many complementary factors between government subsidies and government venture capital in terms of intervention timing, innovation guidance, and coverage. The government may not be able to continuously meet the capital needs of enterprises in the later stage, but government-guided fund intervention can meet the differentiated capital needs of enterprises in different stages. Second, government subsidies pay more attention to the quantity of enterprise innovation, whereas government venture capital focuses on the sustainability and marketization of enterprise innovation. The combination of the two can better guide enterprises to realize the transformation of scientific and technological achievements. Finally, government subsidies are directly issued by government agencies, which cover a small area and are not universal. Guided fund intervention is relatively objective and has wide coverage. The two can compensate for each other’s policy deficiencies and play a synergistic role in promoting the innovation and development of enterprises.

Therefore, the interaction terms of guided fund intervention and government subsidies received in the current year were constructed, and two-way fixed-effect regression on innovation was conducted. The regression coefficient of the interaction term is 0.0023, which is significant at the 0.01 level, indicating that there is a certain synergistic effect between the two. The results are shown in Table 8.

5.4.4. Regression by Region and Industry

To verify the combined impact of the eastern, central, and western regions and their respective industries on government venture capital’s influence on specialized and special new “little giant” enterprises, we decided to further group the data based on these two dimensions and conduct regression, as shown in

Table 9. Industry and region heterogeneity analysis.

	(1)	(2)	(3)	(4)	(5)	(6)
	East-Emerging	West-Emerging	Middle-Emerging	East-Non-Emerging	West-Non-Emerging	Middle-Non-Emerging
t	7.4049 **	0.506	10.5222	1.0973	1.7958	12.9769 ***
	−2.4533	−0.0422	−1.4578	−0.5087	−0.3386	−3.1671
INsize	0.0000 ***	0	0.0000 **	0	0	0.0000 ***
	−3.1416	−0.2654	−2.3918	−0.1886	(−0.4657)	−4.7951
ROA	2.8203	−1.9 × 102 ***	9.3039	−5.5805	−28.3607	−0.5185
	−0.5495	(−2.8551)	−0.2041	(−0.5892)	(−1.6456)	(−0.0338)
GEAR	−1.7566	−16.0811	7.2222	−0.7773	−4.1349	2.8982
	(−0.4310)	(−0.9959)	−0.6672	(−0.3049)	(−0.6880)	−0.643
ROE	0.2952	125.2660 **	4.572	3.4891	6.1204	7.3211
	−1.0539	−2.3258	−0.1978	−0.589	−0.8345	−0.7027
AGE	0	0	0	0	0	0
	(.)	(.)	(.)	(.)	(.)	(.)
ID	YES	YES	YES	YES	YES	YES
YEAR	YES	YES	YES	YES	YES	YES
_cons	18.8908 ***	21.9769 ***	9.6704 **	19.2493 ***	19.7268 ***	12.7501 ***
	−11.1118	−3.8954	−2.2189	−19.7696	−7.8012	−6.7588
N	1917	278	444	4365	534	1184
adj. R2	0.541	0.573	0.494	0.545	0.646	0.566

p statistics in parentheses: ** p < 0.01, *** p < 0.001.

. According to the statistical results, only the strategic emerging industries in the east and the traditional industries in the central region have a significant promoting effect, while no significance was observed in other groups. The strategic emerging industries in the east have geographical advantages, and their industries have good development potential. There are a large number of high-quality projects. Therefore, after GVC provides financial support, they can effectively exert their potential and achieve innovative breakthroughs. Due to its geographical location, the industrial distribution in the central region is more traditional compared to that in the eastern region. It has a larger number of high-quality traditional enterprises with stable development. However, it also faces financing constraints. After the intervention of the government venture capital, it can provide more fresh heterogeneous resources, thereby achieving breakthroughs in innovation. The differences in empirical results reflect the differences in industry advantages and endowments in different regions.

6. Research Conclusions and Policy Recommendations

6.1. Research Conclusions

Empirical research was conducted using sample data from the first through sixth batches of specialized and special new “little giant” enterprises listed on A-share and NEEQ from 2013 to 2023, and the following conclusions were drawn.

Government venture capital tends to be invested in “little giants” specializing in strategic emerging industries. In allocating government venture capital, the government considers both policy orientation and investment efficiency in its investment decision-making, which is in line with the strategic goal of national industrial structure adjustment, transformation, and upgrading.

Government venture capital intervention can improve the innovation of “little giant” enterprises. It has a significant role in promoting enterprise innovation. Government venture capital can effectively alleviate the financing constraints of enterprises by means of signal transmission, resource aggregation, and policy support, and it can improve the ability and willingness of enterprises to improve their innovation. This is achieved through the mediating mechanism of R&D investment. Government venture capital intervention helps the invested “little giant” enterprises form a culture of innovation and trial and error, encouraging them to invest more resources in research and development activities and pay more attention to long-term input and output to achieve innovative breakthroughs.

The promotion effect of government venture capital on “little giant” enterprises is greatest in the central region, followed by the eastern region, and it is not significant in the western region. From the perspective of industry heterogeneity, this promotion effect is more notable in strategic emerging industries because enterprises in this industry face more serious information asymmetry problems, and government venture capital has a more significant mitigating effect on this effect. There is a synergistic effect between government venture capital and government subsidies. The combination of the two can play a positive role in the joint regulation of policies and can jointly promote the innovation and development of “little giant” enterprises.

6.2. Policy Suggestions

6.2.1. Strengthen Support for Strategic Emerging Industries

Continuously enhance support through government venture capital for strategic emerging industries. Strategic emerging industries are a vital cornerstone of China’s innovation drive, playing a crucial role in breaking through domestic circulation bottlenecks, ensuring the security of China’s industrial chain, and promoting industrial structure upgrading. Given the serious information asymmetry they face, further policy inclination is necessary.

Clearly define the investment direction of industrial investment funds. Focus on improving the modern industrial system, support the transformation and upgrading of traditional industries, cultivate and expand emerging industries, plan and construct future industries, concentrate on investing in key links of the industrial chain and projects for chain extension, supplementation, and strengthening, and boost the resilience and security levels of industrial and supply chains.

Regularly arrange for authoritative figures to communicate with general partners (GPs) on industrial trends and pain points, interpret the outline and goals of the national industrial development plan, and guide them to focus on specialized and special new enterprises and strategic emerging industries.

6.2.2. Enhance the Governance Role of Government Venture Capital

Strengthen the governance role of government venture capital for “little giant” enterprises and increase support through various resources. To start, ensure long-term investment in “little giant” enterprises. Reasonably determine the duration of government investment funds, give full play to the cross-cyclical and counter-cyclical adjustment functions of funds as long-term and patient capital, and actively guide the participation of long-term capital, such as the National Social Security Fund and insurance funds.

Optimize post-investment management. Provide professional non-capital value-added services, such as board supervision, joint investment, market connections, and core team training, to assist enterprises in improving corporate governance, obtaining long-term funding, unblocking market channels, and enhancing the team level.

6.2.3. Improve the Coordination Mechanism

Improve the coordination mechanism between government venture capital and government subsidies. The government should further refine the synergy mechanism between them and clarify their functional positioning and complementary relationships. Government venture capital should focus on supporting enterprises’ long-term innovation and market-oriented projects, while government subsidies should target enterprises’ short-term innovation needs.

Government venture capital should pay more attention to strategically important industries that lack government subsidy support to fill the policy gap. When selecting investment projects, guided fund management institutions should have some flexibility in the innovation achievement output requirements, attach more importance to the long-term sustainability and growth of innovation, as well as its macro-economic impact, and balance the two principles of policy objectives and market-oriented operation during investment.

6.3. Limitations and Future Research Directions

This study has several limitations that warrant further investigation in subsequent research.

First, while we focus on firm-level indicators, other factors—such as investment scale, financing rounds, and social networks—may also influence the innovation performance of specialized and sophisticated “little giant” enterprises supported by government venture capital. These variables were not included in the current analysis but should be explored in future work.

Second, due to data constraints in the Zero2IPO database, certain funds—such as those solely owned by local governments—may not be fully captured, even though their policy effects are similar. Given the challenges in obtaining comprehensive data, our sample was restricted to funds explicitly classified as “government-guided” in the database, which may have affected the robustness of our findings. Future studies should seek to incorporate a more representative sample.