The Impact of R&D Subsidies and Non-R&D Subsidies on Innovation Output of Biopharmaceutical Firms

Abstract

Government subsidies are important to the improvement of innovation level and sustainable development of biopharmaceutical firms. This paper empirically analyses the impact of government R&D and non-R&D subsidies on the innovation output of biopharmaceutical firms. We establish static, and dynamic panel mediation effect models using 2013–2019 data from China’s listed biopharmaceutical companies. Additionally, we further explore the mediating role of R&D investment between R&D subsidies and innovation output. The findings show that, first, R&D subsidies contribute to the innovation output of biopharmaceutical firms, while the effect of non-R&D subsidies is insignificant. Second, the study of the mechanism shows that R&D subsidies can significantly promote firms’ innovation output through R&D investment in the static mediating model; but interestingly, the R&D subsidies fail to promote innovation output in the dynamic mediating model through R&D investment, perhaps due to the sustainable impact of innovation. Based on the above results, this paper provides suggestions and insights for the formulation of different types of subsidy policies and the improvement of the innovation level of biopharmaceutical firms.

Keywords

Subsidies innovation biopharmaceutical firms R&D investment dynamic mediation effect

Highlights

R&D subsidies contribute to the innovation output of biopharmaceutical firms, while the effect of non-R&D subsidies is not significant.

Static and dynamic models are established to compare and analyse the mediating role of R&D investment in the relationship between subsidies and innovation output.

Biopharmaceutical firms mainly rely on the sustainability of their intrinsic innovation capability to improve innovation level.

Introduction

With the development of biotechnology and the aggravation of population aging in recent years, the biopharmaceutical industry has gradually become a focus of global pharmaceutical industry development (Majidpour et al., 2021). The continuous development of people’s living environment and the change in health awareness has accelerated the development of the biopharmaceutical industry (Chakravarty, 2022). In addition, the global outbreak and spread of COVID-19, and the emergence of various new viruses have further stimulated the development of vaccines, blood products and related reagents and other drugs (Xu et al., 2020).

In China, classified as a high-tech industry (Mazzola et al., 2016), the state attaches importance to its development and has issued corresponding development plans (Hu & Chung, 2015). China is a big country of generic drugs and has long focused on the development of generic drugs (He et al., 2022). However, there is a large gap between the efficacy of generic drugs and the original drugs. Therefore, it is a necessary development direction for upgrading China’s biopharmaceutical industry to improve the level of drug technology innovation (Ye et al., 2022). In recent years, China has begun to actively promote the transformation and upgrading of the biopharmaceutical industry. Furthermore, in the context of the global epidemic, China’s pharmaceutical industry has begun to enter a new era of combining generic drugs with innovative drugs. At this stage, although an innovative R&D system for the pharmaceutical industry has been initially established, it is still unable to meet the growing demand for innovation. China lags far behind other countries in innovative drug development (Fu et al., 2019). In the global pharmaceutical market, the United States, Europe, and Japan have more than 80% of the pharmaceutical market share. Among them, the United States is a global leader in biopharmaceutical technology, with the number of drugs developed and market sales accounting for more than 35% of the world. Therefore, it is important to study the Chinese biopharmaceutical industry to improve innovation development, though the related research in the field of the biopharmaceutical industry is relatively rich for western countries (Desai et al., 2018; Piñeiro-Chousa et al., 2022). There are also a few kinds of literatures on China’s pharmaceutical industry, but it is mainly focused on pharmaceutical manufacturing (Cao & Lin, 2020), and the research has certain limitations. Therefore, this paper collects different types of pharmaceutical companies in China, including biopharmaceuticals, biological vaccines, intelligent healthcare, interconnected medicine and pharmaceutical R&D for a comprehensive examination. Besides, China is one of the world’s most populous country (Bu et al., 2022), with one-fifth of the world’s population. The pharmaceutical industry is closely related to the lives of people. It is a special industry for the people to prevent and treat diseases, rehabilitation and health care.

Unlike traditional industries, biopharmaceutical firms require clinical research and basic experiments, which are characterised by high innovation costs, high investment risks and long R&D cycles (Pal & Nandy, 2019). Therefore, the government needs to grant subsidies to firms to reduce R&D costs and risks, alleviate financing constraints, increase R&D investment and enhance the motivation of independent innovation.

Government subsidies cover both R&D and non-R&D subsidies (Hu and Liu, 2019). The implementation of subsidy policies may cause R&D manipulation and rent-seeking behaviours in some firms. Rent-seeking behaviours are a phenomenon that exists between government and enterprises (Krueger, 1974), and it is a wealth transfer activity that people seek government protection. From the perspective of the government, the government holds the right to disburse subsidies and has a large number of special funds for strategic emerging industries. Through the relevant subsidy criteria, the government grants corresponding subsidies to enterprises, which in turn stimulates innovative production. However, there is information asymmetry between the government and enterprises (Aney & Banerji, 2022). When allocating subsidies, it is difficult for the government to distinguish between enterprises that actually meet the relevant criteria for subsidies and enterprises that are in urgent need of subsidies (Zhao et al., 2020). Therefore, some enterprises will choose rent-seeking behaviours to obtain government transfer subsidies and obtain more funds. From the perspective of enterprises, the enterprises that have established relations with government departments through rent-seeking convey to the government the relevant conditions that they meet the subsidies, and can make effective use of the subsidies (Du & Mickiewicz, 2016). Thus, on the one hand, some enterprises may bribe officials who have the right to grant government subsidies through rent-seeking activities. On the other hand, through R&D manipulation, it will convey to the external world that it meets the relevant subsidy criteria and thus obtains the relevant subsidies (Yang & Zhang, 2022). Therefore, the different types of subsidies may have different impacts on a firm’s innovation activities.

This paper explores the impact of government R&D and non-R&D subsidies on firms’ innovation output and conducts an in-depth analysis of R&D investment as a mediating variable. Specific contributions are as follows: first, from the current research, studies on the relationship between government R&D subsidies, non-R&D subsidies, firm’s R&D investment and innovation output have made remarkable achievements in other fields such as new energy and high-tech industries. But there are relatively few studies that combine them and select Chinese biopharmaceutical firms as samples. In the context of the global COVID-19 pandemic, the biopharmaceutical industry, as a priority strategic emerging industry, is receiving more and more attention from different countries. Hence, this paper studies biopharmaceutical companies to investigate the impact of subsidies on innovation output. Second, when studying government subsidies, R&D investment and innovation output, most existing studies only constructed a static mediating effect model, but innovation output is a dynamic process with sustainable impact. Although a few studies applied the dynamic mediating effect mode (Wang et al., 2019; Yu et al., 2022), there are no studies on the dynamic mechanisms of biopharmaceutical firms to our knowledge. Therefore, based on existing studies, this paper examines the sustainable impact of innovation and analyses the role of R&D investment in the relationship between government subsidies and the innovation output of biopharmaceutical firms, comparing the results from static and dynamic panel mediation effect models.

Theory and Hypotheses

Government Subsidies and Innovation Output

When studying the impact of government subsidies on firms’ innovation outcomes, there are controversies among different scholars. Through meta-regression analysis, Dimos and Pugh (2016) found that government subsidies may have the following effects on firm innovation: stimulus effect (Bronzini & Piselli, 2016), inhibitory effect (Shu et al., 2015), insignificant effect (Klette & Møen, 2012) and nonlinear effect (Dai & Cheng, 2015). Santos (2019) believed that the impact of government subsidies on a firm’s innovation is closely related to the specific purpose of subsidies, and different purposes of subsidies will have different effects on innovation. Therefore, many researchers have distinguished between R&D and non-R&D subsidies to explore their impact on firm innovation. For example, Sung (2019) analysed the impact of Korean government subsidy policies on firm innovation and found that R&D subsidies significantly promote firm technological innovation, while the effect of non-R&D subsidies is insignificant. Thomson and Jensen (2013) argued that there is a lag of 1–2 years in the impact of government subsidies on firm innovation performance. Li et al. (2021) pointed out that a lagged effect of government subsidies and R&D investment on firms’ innovation output based on time series and dynamic panel data. In summary, R&D subsidies can reduce firms’ R&D costs and promote their innovative production (Shin et al., 2019). Non-R&D subsidies fail to act directly on firms’ R&D process and have no significant impact on firms’ innovation (Le & Jaffe, 2016). Moreover, there may be a lag in the transformation of subsidies to innovation results. Based on the above analysis, the following hypotheses are proposed (detailed explanations of the literature are provided in the supplementary information):

H₁: R&D subsidies in the lag period may promote the current innovation output of biopharmaceutical firms.

H₂: Non-R&D subsidies in the lag period may not significantly promote the current innovation output of biopharmaceutical firms.

Subsidies and R&D Investment

The current research on the impact of R&D subsidies on firms’ R&D investment has mainly the following three views: (1) Government R&D subsidies are negatively correlated with corporate R&D investment (Gelabert et al., 2009). (2) There may exist an ‘inverted U’ relationship between R&D subsidies and R&D investment (Wei & Mahnoor, 2020). (3) Government R&D subsidies are positively correlated with corporate R&D investment. Recent studies on government R&D subsidies and firms’ R&D investment have mainly concluded that R&D subsidies promote R&D investment, and there is a ‘crowding-in effect’ between the two (Becker, 2015). Yu et al. (2016) showed a continuous promotion between government subsidies and firms’ R&D investment, with a 1–2 period lag of subsidies positively promoting firm’s R&D investment. In summary, scholars have not reached the unified conclusion on the effect of R&D subsidies on a firm’s R&D investment, but studies in recent years have shown that receiving government R&D subsidies will stimulate a firm’s innovation vitality and increase R&D investment (Wu et al., 2021). Additionally, there may be a lag effect between R&D subsidies and R&D investment. Based on the above analysis, the following hypothesis is proposed:

H₃: R&D subsidies in the lag period may promote the current R&D investment of biopharmaceutical firms.

The Mediating Role of R&D Investment

From the theory of resource tendency mechanism, R&D subsidies can influence investment on R&D personnel and technology, and determine whether firms can better develop new products, technologies and new markets to improve innovation output (Qu et al., 2017). Yi et al. (2021) concluded that government R&D subsidies can diversify R&D risks, increase R&D investment and thus improve innovation level. From the theory of signaling mechanism, firms receiving government subsidies are generally considered as high-quality firms, with good development prospects and qualifications. Therefore, these firms release positive signals to external investors, thus raising more funds for the firm and indirectly promoting innovation output (Okamuro et al., 2019). Meuleman and Maesenerie (2012) argued that subsidies alleviate the information asymmetry faced by investment institutions and improve external financing. By analysing German firms that received government funding, Bianchi et al. (2019) confirmed that subsidies can bring more information and technology to firms, increase R&D investment, and promote the level of innovation. Based on the above analysis, the following hypothesis is proposed:

H₄: R&D investment may play a mediating role between government R&D subsidies and innovation output of biopharmaceutical firms.

Methodology

Data

This paper selects listed biopharmaceutical companies in Shanghai and Shenzhen A-share (2013–2019) as sample. In order to avoid the omission of biopharmaceutical companies, we summarise and organise biopharmaceutical companies from different conceptual sources including biopharmaceuticals, biological vaccines, intelligent healthcare, interconnected medicine and pharmaceutical R&D. The collection of listed biopharmaceutical companies is mainly from Sina Finance, Tonghuashun, Dongfang Fortune Network, and so on. The data were processed as follows: (1) Companies with special treatment (ST, ST*, which refers to stocks that are specially treated, and is also a warning of delisting risk.) were deleted from our sample. (2) Companies with missing key variables such as invention patents and R&D investment data were excluded. After screening, we obtained 1,267 observations from 181 biopharmaceutical companies. The data on government subsidies, R&D investment and basic characteristics of firms come from the China Stock Market & Accounting Research (CSMAR)¹ database. The patent data are obtained from the CSMAR and China National Knowledge Infrastructure (CNKI)² patent Retrieval and analysis systems.

Variables

Dependent Variable

Innovation output (Ipatent): Existing studies measuring innovation output mainly include the number of patent applications, the number of patents granted, new product sales revenue, and so on (Lin et al., 2011). Compared with indicators such as new product sales revenue, the number of patents better reflects the innovation achievements of firms. There is a certain period from patent application to patent grant (Li et al., 2020). To avoid the impact of approval time, this paper adopts the number of invention patent applications as a measure of innovation output. Considering that the number of patent applications of some firms is zero, the variable is taken as the natural logarithm after adding one with reference to the practice of general literature.

Explanatory Variables

Government R&D subsidies (Rsub): Since data on R&D subsidies received by listed companies are not formally disclosed, R&D and non-R&D subsidies are mainly divided by keyword searches in current academic research (Wu et al., 2020). When dividing R&D subsidies, we manually sorted out R&D subsidies with the following keywords in the subsidy projects: R&D, patent, technological innovation, independent innovation, technical transformation, new product, patent, technological achievements, intelligence introduction, etc. (Li et al., 2020). After this, the natural logarithm of the R&D subsidies is taken.

Government Non-R&D Subsidies (NRsub): Total government subsidies minus R&D subsidies, which mainly include non-R&D activities such as local government investment promotion, pollution control, financial contribution incentives, tax rebates, loan subsidies, demolition and relocation compensation. Then, we take the natural logarithm of non-R&D subsidies.

Mediation Variable

R&D investment (R&D): R&D intensity represents the ability of firms from the process of R&D investment to the final transformation into innovation results (Wang et al., 2020). This paper chooses R&D investment intensity to measure the level of R&D investment of firms, that is, the ratio of the R&D investment amount to operating revenue.

Control Variables

Firm age (Age): Firm age closely influences the innovation activities of firms. Normally, the older the firm is, the stronger the organisational cohesion and coordination of the firm, and the more experience it has to carry out innovative activities. However, old firms may also be dependent on capital accumulated in previous years due to long years on the market, which may hinder innovation activities. Referring to the practice of Qi and Yang (2019), the number of years a firm has been listed is defined as the age of the firm.

Firm size (Size): Firms with different scales have obvious differences in capital, risk-taking, technology, personnel, etc., and the emphasis on innovation is slightly different (Shin et al., 2019). We use the logarithm of the total assets of the firm to measure firm size.

Profitability (Roe): The profitability rate of firms reflects the operating condition. To consolidate and improve their market position, firms with higher profitability will further improve the level of technological innovation to maintain market competitiveness (Klingenberg et al., 2013). This paper uses the ratio of net profit to net assets as a measure of profitability with reference to existing literature.

Ownership concentration (Holder): In a firm with a moderate concentration of equity, the higher the shareholding of the majority shareholder, the stronger their ability to resist risks. Moreover, shareholders will consider their own development in the long term and pay more attention to innovative activities. Referring to the practice of Wang et al. (2021), the shareholding ratio of the largest shareholder is taken as a measure index of ownership concentration.

Independent directors (Indir): Independent directors come from high-end, intellectual, and other talents in various fields of society and have a strong knowledge background. Independent directors are relatively subjective when making decisions on major resolutions of the firm. Therefore, there may be some impact on the decision to innovate in the firm. In this paper, the proportion of independent directors on the board of directors is used as a control variable.

Capital intensity (Fasset): Capital intensive industries often use more advanced production techniques and equipment, objectively requiring highly skilled labour. In order to increase labour productivity, capital intensive industries will spend more time selecting and training employees, attracting a concentration of highly skilled labour to capital intensive industries, which will be in a position to conduct more R&D and innovation. Referring to Gao et al. (2021), this paper measures capital intensity as the ratio of the natural logarithm of total fixed assets and the number of employees in the firms.

Ownership ratio (Ower): Ownership equity reflects the number of shares acquired by the investors. Investors with different proportions of ownership equity represent different levels of control over the firm. Therefore, the scientific research support for the firm innovation may be different. We use the ratio of total shareholders’ equity to total assets to measure the ownership ratio.

Operating debt ratio (Oplr): The operating debt ratio, as a special type of liability linked to specific transaction behaviour, can constrain management behaviour. It can also bring about an information transfer effect, indirectly influencing firms’ innovation behaviour and external financing. The operating debt ratio examines the risk situation of the firm, and this paper introduces the operating debt ratio as a control variable. Table 1 summarizes the definitions of variables.

Table 1.

Definitions of Variables.

Category	Symbol	Definition
Dependent variable	Ipatent	The natural logarithm is taken after adding one to the number of firms’ invention patents
Explanatory variables	Rsub NRsub	The government R&D subsidies in the year are taken as the natural logarithm The government Non-R&D subsidies in the year are taken as the natural logarithm
Mediation variable	R&D	The ratio of R&D investment to operating revenue
Control variables	Age	Firms’ listed years
	Size	The natural logarithm is taken for firms’ total assets
	Roe	The ratio of net profit to net asset
	Holder Indir	The proportion of shareholding of the largest shareholder The ratio of independent directors to directors
	Fasset Ower Oplr	The natural logarithm of total fixed assets to the number of employees The ratio of total shareholders’ equity to total assets The ratio of long-term liabilities to owner’s equity

Source: CSMAR; Related Literatures.

Model Settings

Static Panel Mediation Effect Model

Based on the previous hypotheses, to test whether government subsidies can significantly contribute to the innovation and whether R&D investment plays a mediating role between subsidies and the innovation output of biopharmaceutical firms. Referring to the practice of Baron and Kenny (1986), the following econometric models are set up:

I p a t e n t_{i t} = α_{0} + α_{1} S u b_{i, t - 1} + σ_{1} C o n t r o l s_{i t} + λ_{i} + η_{t} + ε_{i t}

(1)

R & D_{i t} = β_{0} + β_{1} S u b_{i, t - 1} + σ_{2} C o n t r o l s_{i t} + λ_{i} + η_{t} + ε_{i t}

(2)

I p a t e n t_{i t} = λ_{0} + λ_{1} S u b_{i, t - 1} + λ_{2} R & D_{i t} + σ_{3} C o n t r o l s_{i t} + λ_{i} + η_{t} + ε_{i t}

(3)

where i represents the firm, t represents the year. Ipatent_it is the dependent variable. Sub_i_,t–1 is the core explanatory variable representing Rsub_i_,t–1 and NRsub_i_,t–1; R&D_it is the mediation variable. Controls_it include Age, Size, Roe, Holder, Indir, Fasset, Ower and Oplr. λ_i is the individual fixed effect, η_t is the time fixed effect and ε_it is the random perturbation terms. In the regression, the following strategy was performed: first, we consider that it takes some time for government subsidies affect the innovation output of firms. This paper treats subsidies with a lag, which can also moderate the endogeneity problem of reverse causality (Xu et al., 2020). Second, we use the classical ‘two-way fixed effects model’ to avoid the possible influence of time and individual on the firm innovation process. Robust standard errors are adopted by default in the regression tests.

Dynamic Panel Mediation Effect Model

On the one hand, the sustainable impact of innovation should be considered; on the other hand, there may be endogenous problems such as missing variables and bidirectional causality. Referring to the practice of Wang et al. (2019), this paper introduces the lagging term of the dependent variable as an explanatory variable to reduce the errors caused by unobservable factors. We construct the following econometric models:

\begin{array}{l} I p a t e n t_{i t} = a_{0} + a_{1} I p a t e n t_{i, t - 1} + I p a t e n t_{i, t - 2} + a_{2} R s u b_{i, t - 1} + \\ σ_{4} C o n t r o l s_{i t} + λ_{i} + η_{t} + ε_{i t} \end{array}

(4)

R & D_{i t} = b_{0} + b_{1} R D_{i . t - 1} + b_{2} S u b_{i, t - 1} + σ_{5} C o n t r o l s_{i t} + λ_{i} + η_{t} + ε_{i t}

(5)

\begin{array}{l} I p a t e n t_{it} = c_{0} + c_{1} I p a t e n t_{i, t - 1} + I p a t e n t_{i, t - 2} + c_{2} R s u b_{i, t - 1} + c_{3} R D_{i t} + \\ σ_{6} C o n t r o l s_{i t} + λ_{i} + η_{t} + ε_{i t} \end{array}

(6)

where Ipatent_i_,t–1, Ipatent_i_,t–2 and R&D_i_,t–1 are the explanatory variables introduced. Controls_it are control variables, and the definition of the variable is the same as above.

Empirical Results

Descriptive Statistics

From the results of descriptive statistics in Table 2, the mean value of innovation output is 2.47 and the maximum value is 6.66. In addition, we use ArcGIS to illustrate the differences in the average level of innovation among various regions in China. From Figure 1, most regions in China are below the average level of innovation of 2.47, and relatively few regions have a high level of innovation. It indicates that the overall level of innovation in China’s biopharmaceutical firms is weak, which is consistent with the study by Woolliscroft (2020). The mean value of R&D investment intensity is 5.76%, with a relatively large standard deviation, suggesting that the overall level of R&D investment in China’s biopharmaceutical firms is low. The standard deviation of government R&D subsidies is 4.54, that is, biopharmaceutical firms may get different R&D subsidies due to the difference of firm’s characteristics. The average value of non-R&D subsidies is higher than R&D subsidies.

Table 2.

Descriptive Statistics.

Variables	N	Mean	SD	Min	Max
Ipatent Rsub NRsub R&D Age Size	1267 1267 1267 1267 1267 1267	2.47 13.61 15.75 5.76 10.02 22.03	1.22 4.54 2.70 5.19 6.40 0.97	0.00 0.00 0.00 0.04 1.00 19.64	6.66 19.41 20.34 52.61 27.00 25.64
Roe Holder Indir Fasset Ower Oplr	1267 1267 1267 1267 1267 1267	0.07 0.31 0.38 0.01 0.65 0.58	0.18 0.13 0.06 0.01 0.19 0.25	−4.70 0.06 0.25 0.00 −1.13 0.04	0.68 0.72 0.67 0.28 0.98 1.00

Source: The original data comes from CSMAR, and the results are obtained with Stata 16.

Figure 1.

Source: Drawn with ArcGIS by the authors.

Correlation Analysis

From the results, the correlation between R&D subsidies, non-R&D subsidies, R&D investment among firms and innovation output is relatively high and positive. The size of biopharmaceutical firms has a high correlation with innovation output, non-R&D subsidies, and firms’ age, but a low correlation with other explanatory variables. Multicollinearity was tested by the Variance Inflation Factor (VIF) as shown in Table 3.

Table 3.

Correlation and VIF.

	Ipatent	RD	Rsub	NRsub	Age	Size	Roe	Holder	Indir	Fasset	Ower	Oplr
Ipatent	1
R&D	0.194***	1
Rsub	0.148***	0.083***	1
NRsub	0.227***	0.083***	−0.017	1
Age	0.193***	−0.213***	−0.011	0.215***	1
Size	0.433***	−0.062**	0.090**	0.404***	0.481***	1
Roe	0.049*	−0.057**	0.042	0.060**	0.029	0.070**	1
Holder	−0.033	−0.173***	0.066**	−0.047*	−0.045	0.011	0.144***	1
Indir	−0.076**	0.123***	−0.036	0.066**	−0.063**	−0.038	−0.075***	−0.032	1
Fasset	−0.275***	0.058**	−0.053*	−0.239***	−0.285***	−0.505***	−0.007	−0.068**	0.067**	1
Ower	−0.145***	0.155***	0.014	−0.170***	−0.319***	−0.370***	0.155***	0.112***	0.031	0.035	1
Oplr	−0.030	0.028	0.012	−0.039	0.039	−0.144***	0.123***	0.134***	−0.083***	−0.251***	0.390***	1
VIF			1.03	1.24	1.47	1.92	1.08	1.10	1.04	1.39	1.50	1.28
Tolerance			0.97	0.81	0.68	0.52	0.93	0.91	0.96	0.72	0.67	0.78

Source: The original data comes from CSMAR, and the results are obtained with Stata 16.

Notes: Significant at the *p<0.05, **p<0.01, ***p<0.001 level.

Regression Analysis

Basic Model Regression Results

This section analyses the impact of government R&D and non-R&D subsidies on the innovation output of biopharmaceutical firms. We establish a static panel fixed effect model and compared it with the dynamic panel System GMM. Table 4 shows the regression results of the basic model. In general, the regression results of the two models are similar. The static model regression results show that R&D subsidies with one period lag have a significantly positive effect on the innovation output of the current biopharmaceutical firms at the level of 5%. Non-R&D subsidies with one period lag cannot significantly promote the innovation output of firms, but the coefficient is positive. The results of the dynamic model regression show that the p-value of AR(2) is greater than 0.1, indicating that there is no autocorrelation of the perturbation term. The p-value corresponding to Hansen’s statistics are all higher than 0.1, showing that the null hypothesis of ‘all instrumental variables are effective’ cannot be rejected. R&D subsidies with one period lag significantly contribute to firms’ innovation output at the 5% level, while the impact of non-R&D subsidies is insignificant. The results are consistent with the static model, that is, hypotheses H1 and H2 are verified. Specifically, R&D subsidies provide biopharmaceutical firms with direct resources, including the financial form of R&D funds and highly skilled personnel. Secondly, R&D subsidies also indirectly release good signals and alleviate the financing difficulties in biopharmaceutical firms. Hence, R&D subsidies will promote the innovative production of firms to a certain extent. As for the non-R&D subsidies, firms can use them flexibly and conveniently according to their own development needs in an effective way. For example, firms can use the funds for non-R&D activities such as purchasing new equipment, pollution control, and other non-R&D activities. Hence, non-R&D subsidies may not significantly contribute to firms’ innovation output. However, it takes some time for firms to transform from government subsidies to final innovation results, so there is a lag in the impact of subsidies on innovation output. Since there is no significant contribution of non-R&D subsidies to innovation output of biopharmaceutical firms, the mediating effect model below focuses on the relationship between R&D subsidies, R&D investment and the innovation output.

Table 4.

Basic Model Regression Results.

	Fixed Effects		System GMM
	Ipatent	Ipatent	IPatent	Ipatent
L.Ipatent			0.264***	0.268***
L.Ipatent			(2.762)	(2.806)
L2.Ipatent			0.238***	0.232***
L2.Ipatent			(3.298)	(3.197)
Rsub_i _,t–1	0.019**		0.019**
Rsub_i _,t–1	(2.287)		(2.176)
NRsub_i _,t–1		0.012		0.013
NRsub_i _,t–1		(0.941)		(0.963)
Age	0.030	0.025	−0.004	−0.004
Age	(1.427)	(1.174)	(−0.419)	(−0.463)
Size	0.271**	0.265**	0.078	0.080
Size	(2.530)	(2.459)	(0.805)	(0.822)
Roe	−0.082	−0.094	1.043	0.991
Roe	(−0.981)	(−1.175)	(1.139)	(1.066)
Holder	−0.767	−0.575	−0.294	−0.238
Holder	(−1.309)	(−0.996)	(−0.676)	(−0.550)
Indir	−1.573**	−1.537**	−1.343**	−1.408**
Indir	(−2.296)	(−2.219)	(−2.192)	(−2.335)
Fasset	2.553	2.752	−11.905**	−11.528**
Fasset	(1.004)	(1.115)	(−2.049)	(−2.045)
Ower	0.044	0.043	−0.337	−0.352
Ower	(0.162)	(0.157)	(−0.908)	(−1.035)
Oplr	−0.386*	−0.373*	−0.266	−0.240
Oplr	(−1.793)	(−1.721)	(−0.486)	(−0.438)
Obervations	1086	1086	905	905
Hansen test			p = 0.311	p = 0.284
AR(1)			p = 0.000	p = 0.000
AR(2)			p = 0.345	p = 0.293

Source: The original data comes from CSMAR, and the results are obtained with Stata 16.

Notes: Significant at the *p <0.05, **p <0.01, ***p <0.001 level.

From the results of the control variables, firm size significantly contributes to the innovation output of biopharmaceutical firms at the 5% level in the static model; in the dynamic model, it is not significant for the innovation output of firms, but the coefficient is positive, indicating that larger firms have a relatively higher level of innovation output to some extent. In the static and dynamic models, the proportion of independent directors is significantly negatively related to the innovation output at the level of 5%. The high proportion of independent directors will dissatisfy with firms, which in turn will have a negative effect on corporate innovation. The operating debt ratio is negatively related to the innovation output of firms at the 5% and 10% levels in the static model, respectively. In the dynamic model, it is not significant at the 10% significance level, but the coefficient is negative. The operating debt ratio reflects the financial and risk status of firms. Higher debt ratio means that the financial structure of firms may have certain problems, which will be harmful to the firm’s innovation.

Mediating Effect of R&D Investment

Table 5 shows the regression results for R&D investment as a mediating variable. First of all, the static model regression results show that the estimated parameter of R&D subsidies with one period lag is significantly positive at the level of 10%. It indicates that the increase of R&D subsidies with one period lag will enhance the R&D investment of biopharmaceutical firms. Hypothesis H3 is verified. R&D subsidies have a positive significant contribution to innovation output at the 95% confidence level while the contribution of R&D investment is not significant, so we apply the Bootstrap test. From the results of the Fixed Effects in Table 6, the 95% confidence interval excludes zero, that is, the mediating effect of R&D investment exists. Hypothesis H4 is verified. Specifically, when R&D subsidies increase by one unit, the innovation output of firms will directly increase by 0.019 units. Meanwhile, the R&D investment will increase by 0.037 units, leading to the indirect growth of the firm’s innovation output by 0.000185 (0.037×0.005). The sum of the direct effect and mediating effect is 0.024 (0.019 + 0.005), and the mediating effect accounted for 7.8% of the total effect. There is a weak mediating effect. The results of dynamic model regression show that R&D subsidies with one period lag have no obvious promotion effect on the R&D investment of biopharmaceutical firms, so the Bootstrap test is required. From the results of the test in Table 6, the 95% confidence interval includes zero, that is, the mediating effect of R&D investment does not exist.

Table 5.

Regression Results of the Mediating Effect Model.

Fixed Effects			System GMM
R&D Ipatent			R&D Ipatent
L.Ipatent				0.289***
				(3.188)
L2.Ipatent				0.256***
				(3.699)
L.R&D			0.739***
			(7.951)
Rsubi,t–1	0.037* (1.775)	0.019** (2.258)	0.021 (1.265)	0.016* (1.887)
R&D		0.005		0.026*
		(0.620)		(1.762)
Age	0.200*	0.029	−0.014	0.001
	(1.833)	(1.377)	(−0.377)	(0.063)
Size	−0.585	0.274**	0.185	0.046
	(−1.107)	(2.857)	(0.866)	(0.493)
Roe	−1.984***	−0.071	−2.846	1.429
	(−3.285)	(−0.824)	(−0.464)	(1.475)
Holder	0.417	−0.769	−0.841	−0.205
	(0.188)	(−1.369)	(−0.701)	(−0.493)
Indir	1.685	−1.582**	1.979	−1.549**
	(0.430)	(−2.310)	(0.994)	(−2.465)
Fasset	6.180	2.520	−8.309	−10.482*
	(0.627)	(0.982)	(−1.184)	(−0.394)
Ower	2.784	0.029	3.394	−0.402
	(1.356)	(0.106)	(1.564)	(−1.102)
Oplr	−2.925***	−0.371*	−1.481	−0.167
	(−2.934)	(−1.704)	(−0.987)	(−0.295)
Observations	1086	1086	1086	905
Hansen test			p = 0.241	p = 0.347
AR(1)			p = 0.015	p = 0.000
AR(2)			p = 0.853	p = 0.351

Source: The original data comes from CSMAR, and the results are obtained with Stata 16.

Notes: Significant at the*p < 0.05, **p < 0.01, ***p < 0.001 level.

Table 6.

Effect Decomposition Table.

	Effect	Z-value	P > \|Z\|	95% Confidence Interval
Fixed Effects	Indirect effect Direct effect	3.52 4.61	0.000 0.000	[0.003, 0.012] [0.025, 0.062]
System GMM	Indirect effect Direct effect	1.24 1.90	0.216 0.057	[−0.000, 0.002] [−0.000, 0.030]

Source: The original data comes from CSMAR, and the results are obtained with Stata 16.

In the static model, biopharmaceutical firms only need to carry out innovation activities according to their actual capabilities. The development of vaccines, drugs, and related reagents requires high funding, and R&D subsidies can provide direct financial support to companies. Therefore, firms will continue to increase R&D investment, such as the introduction of highly skilled personnel and related technologies, to promote the improvement of firm’s innovation. In the dynamic model, biopharmaceutical firms will invest R&D amounts and carry out innovation activities based on previous years’ R&D investment, the sustainable impact of innovation output, and the specific situation of innovation activities carried out by firms. However, R&D subsidies cannot indirectly promote the innovation output of firms through R&D investment. The possible reason is that the innovation of biopharmaceutical firms is a process of sustainable behaviour. The regression results also show that the innovation output lagging in one and two periods positively affects the current innovation output of firms at the level of 1%. Biopharmaceutical firms will form an innovation mindset and long-term thinking in the process of clinical research and vaccine drug research and development, which will continue to influence the innovation of firms. Continuous innovation of firms will form innovative routines and simple rules, that is, innovation inertia. The existence of inertia and under the influence of long-term, the improvement of the innovation level of biopharmaceutical firms depends not only on the R&D investment of firms but also on the internal innovation ability formed by firms themselves. Furthermore, under the long-term dynamic environment, the role of intrinsic innovation capability is much greater than the impact of a firm’s current R&D investment. That is, biopharmaceutical firms mainly rely on the sustainable influence of their intrinsic innovation to improve innovation levels. Therefore, the mediating effect of R&D investment is not significant in the dynamic model. The qualitative results are shown in Figure 2.

Figure 2.

Source: The authors.

Robustness Checks

Replacing the Dependent Variable

This paper further replaces the dependent variable to examine the robustness of the baseline regression results. Referring to the practice of He and Tian (2013), we change the way to measure the innovation output of firms, and take the natural logarithm of the total number of invention utility model and design patent applications after adding 1 to represent the innovation output of firms. As shown in Table 7 (due to space constraints, robustness tests for the replacement variables are reported only for the core explanatory variables), there is still a significant positive effect of R&D subsidies on the innovation output in both static and dynamic models, while the effect of non-R&D is insignificant and the coefficient is positive. That is, the core findings do not change qualitatively by changing the measurement indicator of innovation output.

Table 7.

Baseline Model Robustness Test.

	Fixed Effects		System GMM
	Ipatent	Ipatent	Ipatent	Ipatent
L.Ipatent L2.Ipatent Rsub_i _,t–1	0.017** (2.319)		0.252* (2.498) 0.216* (2.837) 0.018** (1.757)	0.261* (2.623) 0.219* (2.860)
NRsub_i _,t–1		0.008 (0.689)		0.004 (0.275)
Hansen test			p = 0.514	p = 0.486
AR(1)			p = 0.000	p = 0.000
AR(2)			p = 0.838	p = 0.920

Source: The original data comes from CSMAR, and the results are obtained by the Stata16.

Notes: Significant at the *p < 0.05,**p < 0.01,***p < 0.001 level.

The Mediating Variable

Considering possible errors in the mediating variable, this paper further examines the regression results of the mediation model by referring to the study by Saidani et al. (2017). We use the proportion of R&D investment to total assets at the end of the period as a measure of R&D investment. Additionally, we use Bootstrap to test the mediation effect of R&D investment. Table 8 shows that, in the static model, R&D investment plays a partial intermediary role between R&D subsidies and the firm’s innovation output. In the dynamic model, government R&D subsidies are unable to promote a firm’s innovation output through R&D investment. Fundamentally, the significance and direction of core explanatory variables did not change (see Table S1 in supplemental material for detailed data).

Table 8.

Effect Decomposition Table.

	Effect	Z-value	P > \|Z\|	95% Confidence Interval
Fixed effects	Indirect effect Direct effect	3.23 4.46	0.001 0.000	[0.003, 0.016] [0.023, 0.060]
System GMM	Indirect effect Direct effect	1.06 1.89	0.288 0.059	[−0.000, 0.002] [−0.000, 0.029]

Source: The original data comes from CSMAR, and the results are obtained by the Stata16.

Conclusion and Discussion

In the past 2 years, with the outbreak and spread of COVID-19, the global research and development of vaccines and drugs have continuously intensified. The Biopharmaceutical industry significantly contributes to the development process. Clinical research and basic experiments in biopharmaceutical firms are inseparable from long-term and stable financial support. Moreover, the demand for drug development and innovation is much higher than in other industries. On the one hand, the implementation of a subsidy policy can guide a firm’s R&D investment; on the other hand, it will promote the innovation output of firms to some extent. From the existing studies, there is a relative lack of research on the relationship between government subsidies, R&D investment, and innovation output in biopharmaceutical firms. Therefore, this study establishes a static and dynamic panel mediation effect model based on the mediation mechanism of R&D investment; and further analyses the impact of government subsidies on innovation output in biopharmaceutical firms by using R&D investment as a threshold variable. The conclusions are as follows:

First, different types of government subsidies have different effects on the innovation of biopharmaceutical firms. R&D subsidies with a one-period lag can significantly promote the innovation output, while the effect of non-R&D subsidies is insignificant. This finding is concordant with the literature for general firms (e.g., Jiang et al., 2020; Chen et al., 2021). Moreover, there is a certain period between the government subsidies and the transformation of the final innovation results, i.e., there is a lag period for the impact of subsidies on innovation output.

Second, in the static model, R&D subsidies can indirectly promote innovation output of biopharmaceutical firms through R&D investment, while R&D investment cannot effectively play a mediating role in the dynamic model due to the influence of innovation inertia. This differs from the findings of a few studies using dynamic mediating effects models (Wang et al., 2019; Yu et al., 2022). Wang et al. (2019) empirically analysed the mechanism of the effect of regional financial autonomy on carbon economic performance from the perspective of industrial structure upgrading using a dynamic mediating effects model. The study found that regional financial autonomy can promote the growth of carbon economic performance through industrial structure upgrading, that is, there is a significant mediating effect of industrial upgrading. Yu et al. (2022) investigated the relationship between monetary policy, bank competitiveness, and risk-taking, and found that bank competitiveness plays mediating effect between monetary policy and risk-taking. Based on the failure of R&D investment in playing an effective mediating role in the dynamic model, this paper argues that under the sustainability of innovation, the innovation capacity developed by biopharmaceutical firms themselves is much greater than that brought by the R&D investment. Biopharmaceutical firms have already formed innovative thinking about the drugs and vaccines that have been developed, and this process takes a lot of time. R&D investment can bring direct funding for biopharmaceutical firms to carry out innovation activities. However, the implementation of subsidy policies may have a crowding-out effect on firms’ R&D investment. Therefore, government R&D subsidies policy cannot promote the innovation production of biopharmaceutical firms only by influencing firms’ R&D investment. Biopharmaceutical firms should focus on cultivating their intrinsic innovation capabilities and enhancing their internal core technology advantages to promote innovation performance in a long-term and sustainable manner.

Although China’s biopharmaceutical R&D investment and technological innovation capability have been significantly improved, there are still gaps such as the incomplete setting of subsidy forms, the lack of independent innovation capability, and insufficient supervision of subsidy policies.

Concerning policy implications, firstly, in response to the different roles played by subsidy policies on the innovation output of firms, the government should consider various types of subsidies. The government can continuously increase R&D subsidies to solve the dilemma of insufficient innovation of biopharmaceutical firms so that firms can obtain sustainable innovation resources. In addition, the government is also supposed to allocate non-R&D subsidies reasonably to help firms improve infrastructure and provide a favourable environment for their innovation. Meanwhile, other incentives such as tax preference can be provided to encourage biopharmaceutical firms to increase their R&D investment and improve their innovation performance.

Second, the government had better establish a supervisory and management mechanism for biopharmaceutical firms. It can effectively supervise the firms’ R&D investment and innovation results, avoiding malicious access to subsidies to maximise the impact of limited subsidy resources.

For biopharmaceutical firms, in response to the conclusion that R&D subsidies can promote firm innovation output through R&D investment, it is recommended that firms should reasonably increase firm R&D investment while receiving R&D subsidies, such as the introduction of new foreign technical personnel and related technologies. Thus, the firms can enhance the utilisation rate of subsidies and improve the level of innovation. On the other hand, in response to the inertia of corporate innovation, it is suggested that biopharmaceutical firms should not ignore the sustainable impact of innovation capability. Firms should pay more attention to the cultivation of independent innovation capability based on the current internal innovation model. For example, the firms can set up special research bases within the company and strengthen external cooperation to attract major innovation projects (Melnychuk et al., 2021). Thus, it gives full play to the subjective initiative of innovation and forms a long-term sustainable mechanism.

Our study must be considered in light of several limitations. This paper refers to the practice of related scholars to divide government R&D subsidies and non-R&D subsidies through the form of keyword searches. However, currently, there is no specific disclosure of detailed data about government R&D and non-R&D subsidies to firms. Moreover, the problem of keyword omission may occur, leading to errors in the results, so the classification method about subsidies is neither complete nor optimal. Future research should further improve the R&D and non-R&D subsidies measurement. Additionally, government R&D subsidies can affect the innovation output of firms through direct resource transmission path (firms themselves) and indirect signaling path (external investors). This paper examines the effect of government R&D subsidies on the innovation output of biopharmaceutical firms using R&D investment as a mediating variable. Importantly, the mediating role should be considered as partial association but not causality. R&D investment only serves as a resource transmission path, so a larger framework can be constructed in future research to conduct a more profound analysis of the internal and external resources.

Footnotes

Acknowledgements

The authors thank the autonomous region for funding in social science research, as well as Xinjiang University for the research cultivation project in social science.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The authors thank the autonomous region for funding in social science research, as well as Xinjiang University for the research cultivation project in social science.

Notes>

References

Aney

M. S.

, & Banerji

(2022). Political connections, informational asymmetry, and the efficient resolution of financial distress. Economic Modelling, 114, 105901.

Baron

R. M.

, & Kenny

D. A.

(1986). The moderator–mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51(6), 1173–1182.

Becker

(2015). Public R&D policies and private R&D investment: A survey of the empirical evidence. Journal of Economic Surveys, 29(5), 917–942.

Bianchi

, Murtinu

, & Scalera

V. G.

(2019). R&D subsidies as dual signals in technological collaborations. Research Policy, 48(9), 103821.

Bronzini

, & Piselli

(2016). The impact of R&D subsidies on firm innovation. Research Policy, 45(2), 442–457.

, Wang

, Qiu

, & Möst

(2022). Impact assessment of population migration on energy consumption and carbon emissions in China: A spatial econometric investigation. Environmental Impact Assessment Review, 93, 106744.

Cao

, & Lin

(2020). Empirical study on the interaction between Foreign economic trade and pharmaceutical manufacturing industry based on the annual data of 1998–2018 in China. China Pharmacy, 31(14), 1670–1676. (In Chinese)

Chakravarty

(2022). Resource constrained innovation in a technology intensive sector: Frugal medical devices from manufacturing firms in South Africa. Technovation, 112, 102397.

Chen

, & Li

(2021). Environmental regulation, subsidy, and underperforming firms’ R&D expenditure: evidence from Chinese listed companies. International Journal of Technology Management, 85(2/3/4), 190–211.

10.

Dai

, & Cheng

(2015). The effect of public subsidies on corporate R&D investment: An application of the generalized propensity score. Technological Forecasting and Social Change, 90, 410–419.

11.

Desai

K. G.

, Obayashi

, Colandene

J. D.

, and Nesta

D. P.

(2018). Japan-specific key regulatory aspects for development of new biopharmaceutical drug products. Journal of Pharmaceutical Sciences, 107(7), 1773–1786.

12.

Dimos

, and Pugh

(2016). The effectiveness of R&D subsidies: A meta-regression analysis of the evaluation literature. Research Policy, 45(4), 797–815.

13.

, & Mickiewicz

(2016). Subsidies, rent seeking and performance: Being young, small or private in China. Journal of Business Venturing, 31(1), 22–38.

14.

, Liu

, & Zhou

(2019). Can open innovation improve firm performance? An investigation of financial information in the biopharmaceutical industry. Technology Analysis & Strategic Management, 31(7), 776–790.

15.

Gao

, Hu

, Liu

, & Zhang

(2021). Can public R&D subsidy facilitate firms’ exploratory innovation? the heterogeneous effects between central and local subsidy programs. Research Policy, 50(4), 104221.

16.

Gelabert

, Fosfuri

, & Tribo

J. A.

(2009). Does the effect of public support for R&D depend on the degree of appropriability? Journal of Industrial Economics 57(4), 736–767.

17.

, & Tian

(2013). The dark side of analyst coverage: The Case of Innovation. Journal of Financial Economics, 109(3), 856–878.

18.

J. H.

Shang

D. W.

, Wang

Z. Z.

, Li

X. F.

, & Wen

Y. G.

(2022) Physicians’ perceptions of generic drugs in China. Health Policy Open, 3, 100067.

19.

, & Chung

C. C.

(2015). The innovation and modernisation of ‘herbal piece’ in China: System evolution and policy, transitions (1950s–2010s). European Journal of Integrative Medicine, 7(6), 645–649.

20.

, & Liu

(2019). Do the coupling effects of environmental regulation and R&D subsidies work in the development of green innovation? Empirical evidence from China. Clean Technologies and Environmental Policy, 21(9), 1739–1749.

21.

Jiang

, Zhang

, Zhao

, & Wu

(2020). ‘The impact of purchase subsidy on firms’ R&D efforts: evidence from china’s new energy vehicle industry. Sustainability, 12(3), 1105.

22.

Klette

T. J.

, & Møen

(2012). R&D investment responses to R&D subsidies: a theoretical analysis and a microeconometric study. World Review of Science, Technology and Sustainable Development, 9(2/3/4), 169–203.

23.

Klingenberg

, Timberlake

, Geurts

T. G.

, & Brown

R. J.

(2013). The relationship of operational innovation and financial performance—A critical perspective. International Journal of Production Economics, 142(2), 317–323.

24.

Krueger

A. O.

(1974). The Political Economy of the Rent-Seeking Society. The American Economic Review, 64(3), 291–303.

25.

, & Jaffe

A. B.

(2016). The impact of R&D subsidy on innovation: evidence from New Zealand firms. Economics of Innovation and New Technology, 26(5), 429–452.

26.

, Ru

, Shi

, & Zhang

(2021). The effect of government subsidy policy for Foreign capital R&D from the perspective of global supply chains: An empirical analysis with time lag and propensity score matching. Sustainability, 13(15), 8188.

27.

, Wang

, & Xiangli

(2020). Do government subsidies promote new-energy firms’ innovation? Evidence from dynamic and threshold models. Journal of Cleaner Production, 286(1), 124992.

28.

Lin

, Lin

, Song

F. M.

, & Li

(2011). Managerial incentives, CEO characteristics and corporate innovation in China’s private sector. Journal of Comparative Economics, 39(2), 176–190.

29.

Majidpour

, Saber

, Elahi

, Shayan

, & Khorasani

S. M. S.

(2021). Technological catch-up in the biopharmaceutical sector: evidence from Iran. Technology in Society, 67, 101695.

30.

Mazzola

, Bruccoleri

, & Perrone

(2016). Open innovation and firms performance: state of the art and empirical evidences from the bio-pharmaceutical industry. International Journal of Technology Management, 70, 109–134.

31.

Melnychuk

, Schultz

, & Wirsich

(2021). The effects of university–industry collaboration in preclinical research on pharmaceutical firms’ R&D performance: Absorptive capacity’s role. Journal of Product Innovation Management, 38(3), 355–378.

32.

Meuleman

, & Maesenerie

W. D.

(2012). Do R&D subsidies affect SMEs’ access to external financing? Research Policy, 41(3), 580–591.

33.

Okamuro

, Nishimura

, & Kitagawa

(2019). Multilevel policy governance and territorial adaptability: evidence from Japanese SME innovation programs. Regional Studies, 53(6), 803–814.

34.

Pal

, & Nandy

(2019). Innovation and business, sustainability (IBS): empirical evidence from Indian pharmaceutical industry (IPI). Artificial Intelligence for Engineering Design, Analysis and Manufacturing, 33(2), 117–128.

35.

Piñeiro-Chousa

, López-Cabarcos

M. Á.

, Quiñoá-Piñeiro

, & Pérez-Pico

A. M.

(2022). US biopharmaceutical companies’ stock market reaction to the COVID-19 pandemic. Understanding the concept of the ‘paradoxical spiral’ from a sustainability perspective. Technological Forecasting & Social Change, 175, 121365.

36.

Z. Y.

, & Yang

S. Y.

(2019). Can government subsidies promote the TFP of enterprises? The mediating effect of R&D decisions. Science, Technology and Society, 26(3), 392–412.

37.

, Cao

, Wang

, Tang

, & Bukenya

J. O.

(2017). Political connections, government subsidies and technical innovation of wind energy companies in China. Sustainability, 9(10), 1812.

38.

Saidani

, Msolli

, & Ajina

(2017). Research and development investment, and financing constraints: The case of Japan. Research in International Business and Finance, 42, 1336–1342.

39.

Santos

(2019). Do selected firms show higher performance? The case of Portugal’s innovation subsidy. Structural Change and Economic Dynamics, 50, 39–50.

40.

Shin

, Choy

, Lee

, & Park

(2019). Government R&D subsidy and additionality of biotechnology firms: the case of the South Korean biotechnology industry. Sustainability, 11(6), 1583.

41.

Shu

, Wang

, Gao

, & Liu

(2015). Firm patenting, innovations, and government institutional support as a Double-Edged Sword. Journal of Product Innovation Management, 32(2), 290–305.

42.

Sung

(2019). Do government subsidies promote firm-level innovation? Evidence from the Korean renewable energy technology industry. Energy Policy, 132, 1333–1344.

43.

Thomson

, & Jensen

(2013). The effects of government subsidies on business R&D employment: evidence from OECD countries. National Tax Journal, 66(2), 281–309.

44.

Wang

, Li

, & Sun

(2019). The transformation effect of R&D subsidies on firm performance: An empirical study based on signal financing and innovation incentives. Chinese Management Studies, 14(2), 373–390.

45.

Wang

, Zhang

, & Li

(2020). R&D investment and total factor productivity: An empirical study of the listed companies in the coastal regions of China. Journal of Coastal Research, 106, 13.

46.

Wang

, Niu

, & Liu

(2019). Financial autonomy and carbon economic performance: The intermediary effect of industrial structure upgrading. Chinese Journal of Environmental, 11(6), 75–81. (In Chinese)

47.

Wang

, Li

, Shaikh

, Ranjha

A. R.

, & Batala

L. K.

(2021). Do government subsidies promote financial performance? Fresh evidence from China’s new energy vehicle industry. Sustainable Production and Consumption, 28, 142–153. (In Chinese)

48.

Wei

, & Mahnoor

(2020). Corporate finance policies, subsidies, and R&D: Evidence from China. International Journal of Finance & Economics, 26(3), 3875–3891.

49.

Woolliscroft

J. O.

(2020). Innovation in response to the COVID-19 pandemic crisis. Academic Medicine, 95(8), 1140–1142.

50.

, Yang

, & Tan

(2020). Impacts of government R&D subsidies on venture capital and renewable energy investment – an empirical study in China. Resources Policy, 68, 101715.

51.

, Zhao

, & Li

(2021). Can government subsidy strategies and strategy combinations effectively stimulate firm innovation? theory and evidence. Economia Politica, 38(2), 423–446.

52.

, Wang

, & Liu

(2020). Government subsidies, R&D investment and innovation performance: analysis from pharmaceutical sector in China. Technology Analysis & Strategic Management, 33(5), 535–553.

53.

Yang

, & Zhang

(2022). Targeting more effective industrial policies: Evidence from massive media data on R&D manipulation. China Economic Quarterly International, 2(2), 138–150.

54.

, Wang

, & Wang

(2022). New era of drug innovation in China. Acta Pharmaceutica Sinica B, 9(5), 1084–1085.

55.

, Murphree

, Meng

, & Li

(2021). The more the merrier? Chinese government R&D subsidies, dependence, and firm innovation performance. Journal of Product Innovation Management, 38(2), 289–310.

56.

, Fu

, & Li

(2022). Monetary policy, bank competitiveness and bank risk taking-an empirical study based on a systematic GMM approach. Friends of Accounting, 9, 49–55. (In Chinese)

57.

, Guo

, Le-Nguyen

, Barnes

S. J.

, & Zhang

(2016). The impact of government subsidies and firms’ R&D investment: A panel data study from renewable energy in China. Energy Policy, 89, 106–113.

58.

Zhao

, Zeng

, Che

, Zhou

, & Hu

(2020). Research on the profit change of new energy vehicle closed-loop supply chain members based on government subsidies. Environmental Technology & Innovation, 19, 100937.