Linking Government R&D Subsidies and Innovation Performance:
A Chain-mediating Role of R&D Investment and Technological
Collaboration
Xingxiu Wang
a
and Huiying Jiao
School of Management, Changchun University, Changchun, China
Keywords: Innovation Performance, Government R&D Subsidies, Technological Collaboration, R&D Investment,
Chain-Mediating Role.
Abstract: This paper aims to probe how government R&D subsidies relate to innovation performance. The authors
examined R&D investment and technological collaboration as mediators of relationship between government
R&D subsidies and innovation performance. The paper opted for a time-lagged research design to test
hypotheses with data covering 483 high-tech listed firms’ data in China from 2007 to 2019. STATA and the
PROCESS macro in SPSS are used in regression analysis. The results show that R&D subsidies are positively
related to firms’ innovation performance. The relationship is mediated by R&D investment and technological
collaboration. Furthermore, R&D investment is positively related to technological collaboration, there is a
chain-mediating relationship among R&D subsidies, R&D investment, technological collaboration and
innovation performance. This paper constructs a theoretical framework to specifies the process through which
R&D subsidies affects firms’ innovation performance to expand understandings of R&D subsidies, which
further provides practical value to administrative staffs and policymakers for formulating innovation strategies
and R&D subsidies decisions more effectively.
1 INTRODUCTION
The fast growth model of China’s economy has been
replaced by a high-quality development one recently.
It has been generally acknowledged that
technological innovation exerts a major function on
keeping firms’ sustainable development and is the
engine of high-quality economic development. In
order to raise firms’ enthusiasm for technological
innovation, China’s government subsidizes their
research and development (R&D) programs by
increasing its intensity of funding continuously (Liu,
et al., 2021).
A considerable number of studies tested the
associations between R&D subsidies and firms’
innovation performance (Yi, et al., 2021; Gao, et al.,
2021). Some studies found evidence indicating
positive innovation performance effects linked to
government R&D subsidies (Wu, et al., 2020, Xu, et
al., 2021). Other studies reported that government
R&D subsidies distorted factors’ price in the process
a
https://orcid.org/0000-0003-3319-7850
of innovation, resulting in rent-seeking (Gao, et al.,
2021, Zhang 2019). In order to obtain public
subsidies, some companies may ignore the actual and
emerging needs of innovation, which has a crowding
out effect on private R&D capital contribution
(Zhang, 2019, Yu, et al., 2016). In recent studies,
from a perspective of contingency, researchers found
the underlying value to discover the factors
influencing the link between R&D funding and firms’
innovation performance (Gao, et al., 2021). The local
R&D financial assistance and specialized industrial
agglomeration have been regarded as potentially
crucial elements in mitigating the influence of R&D
subsidies for innovation performance (Gao, et al.,
2021).
Previous literature postulated that a direct link
exists between government R&D subsidies and
innovation performance. However, few studies have
explored how government R&D subsidies relate to
innovation performance. In general, R&D investment
can be stimulated by government R&D grants, which
Wang, X. and Jiao, H.
Linking Government RD Subsidies and Innovation Performance: A Chain-mediating Role of RD Investment and Technological Collaboration.
DOI: 10.5220/0011287600003440
In Proceedings of the International Conference on Big Data Economy and Digital Management (BDEDM 2022), pages 749-756
ISBN: 978-989-758-593-7
Copyright
c
2022 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved
749
also positively influence innovation performance,
thus some empirical studies found that R&D
investment is the link between R&D funding and
innovation performance (Liu, et al., 2021, Xu, et al.,
2021, Cerulli, et al., 2015). But existing researches
chiefly direct attention to the incentive mechanism of
R&D financial support. In fact, R&D subsidies can
be regarded as dual signals during the innovation
process (Bianchi, et al., 2019). Therefore, numerous
investigations and studies are needed to reveal the
mechanism of how government R&D subsidies affect
firms’ innovation performance.
We expand the work on the inner influencing
mechanism of R&D subsidies on innovation
performance in this article. We suggest that R&D
subsidies have a positive effect on innovation
performance. Furthermore, we expect that R&D
investment and technological collaboration are
potential mechanisms for explaining the association
between government R&D subsidies and innovation
performance. R&D investment is more likely to
improve technological collaboration of firms, thereby
R&D financial allowances will relate indirectly and
effectively to innovation performance via the chain-
mediating role of R&D investment and technological
collaboration. Using panel data on China’s 483 high-
tech listed firms from 2007 to 2019, we demonstrate
the influencing mechanism of R&D subsidies on
innovation performance.
Our research extends the previous literature in
two aspects. Based on what we have learned, this is
the first research that combines R&D investment and
technological collaboration to explain how
government R&D subsidies create value for firms’
innovation. Compared to other studies on government
R&D subsidies (Gao, et al., 2021, Wu, et al., 2020,
Yu, et al., 2016), our research demonstrates that R&D
investment and technological collaboration are
essential factors in enabling companies to reap the
benefits of R&D subsidies. Several researches have
started to take the mediating role of R&D investment
into account (Xu, et al., 2021). Nevertheless, they
ignore the signalling of R&D subsidies will
encourage technological collaboration (Chapman, et
al., 2018, Kim, et al., 2021), which may play the
potential mediating role. In addition, new evidence
has been provided to indicate that the effects of R&D
subsidies on corporate innovation performance are
positive in this paper. The main framework of our
study is arranged as follows. The very next part
shows the theoretical foundation for probing the
connections among R&D subsidies, R&D
investment, technological collaboration and
innovation performance. Then, data, methods,
consequences of the study are described in detail. At
last, the conclusion part is given.
2 THEORY AND HYPOTHESES
2.1 Government R&D Subsidies and
Innovation Performance
As competition between countries becomes fiercer,
the importance of technological innovation becomes
more and more prominent. Technological innovation
behaviours have been strongly supported by
governments in most countries, and the relevant
policies are tilted towards innovative firms, and the
most important is the subsidy of R&D activities.
Government R&D assistance exerts the following
multiple influences on firms’ technological
innovation. One is to provide direct financial support
to reduce R&D costs. Public R&D funding can be
regarded as an incentive policy, which gives free
financial support to firms’ technological innovation
activities (Chapman, et al., 2018). Bérubé and
Mohnen (2009) showed that R&D subsidies motivate
corporations to introduce more new products. The
second is to transmit signals and improve innovation
success rates. Acting as a “stamp of approval”, the
award of Government R&D subsidies is a signal to
distinguish firms from their competitors (Bianchi, et
al., 2019). R&D financial support will enhance the
attractiveness of enterprises’ innovation projects,
appeal outstanding technical talents to join in and
provide access to other innovative factors, so as to
improve the success probability of innovation. Third,
it will ease the financing constraints. Yang et al.
(2021) pointed out that if engaging in R&D activities
actively, firms face lower financing costs in the bond
market. Government R&D subsidies provide
important signals to financial institutions to identify
firms’ technological innovation ability, thus reducing
firms’ financing cost via financial support for
technological innovation. In addition, R&D subsidies
are also one of the signals for consumers to measure
product quality in product markets, improving the
competitiveness of firms in the market and forming a
virtuous circle. There is no denying that R&D
subsidies may lead to adverse selection effect due to
rent-seeking, encroachment and other issues.
However, in general, the government R&D subsidies
have clear categories of subsidies and higher
application thresholds, so the probability of reverse
selection effect is low on the whole. Therefore, the
public R&D funding can promote the efficiency of
corporate innovation programs. Using data from
BDEDM 2022 - The International Conference on Big Data Economy and Digital Management
750
Germany, Plank and Dubliner (2018) found public
R&D subsidies enhance firms’ innovation. Thereby
we put forward the following hypothesis: H1.
Government R&D subsidies have a positive impact
on innovation performance.
2.2 The Mediating Role of R&D
Investment
As special assets accumulate, firms’ private
investment in R&D is conducive to the creation of
knowledge, thus improving their competitive
advantage in the market. Firms’ R&D investment is
influenced by government R&D subsidies through
the following means. First, it has competitive effects
on firms’ innovation activities. To apply for
government R&D subsidies, enterprises must meet
the threshold, so they need to improve their
competitiveness, and the simple and direct way is to
raise the amount of R&D capital. In addition, the
awarding of public R&D grants ought to be more
observable to outsiders, so as to avoid information
asymmetry. Public R&D subsidies draw the
government’s attention to key industries and
technical areas, which will induce firms to increase
investment in R&D projects and make the market
participants to form a positive expectation, changing
the expected return of firms in related fields. Third,
R&D subsidies are also an important way to help
firms share the risk of technological innovation, thus
increasing firms’ enthusiasm to provide capital to
R&D projects (Cerulli, et al., 2015). The
improvement of R&D investment gives fiscal
guarantee to technological innovation. Chen (2021)
also confirmed that firms’ R&D investment is a
crucial factor to improve the performance of
technological innovation. Some previous studies
have indicated that firms’ R&D investment is a
critical pathway of public R&D funding on
innovation performance (Xu, et al., 2021, Cerulli, et
al., 2015). For the above reasons, we come to the next
hypothesis: H2. Firms’ R&D investment mediates the
relationship between government R&D subsidies and
innovation performance.
2.3 The Mediating Role of
Technological Collaboration
As far as the firm itself is concerned, the
organizational boundary is becoming more and more
blurred, the knowledge flow is more frequent, and it
becomes a common phenomenon to establish
technological collaboration with partners and
enhance innovation capacity using external
intellectual capital due to increasing technical
complexity. In addition to R&D investment, public
R&D funds will also affect technological innovation
by promoting technical collaboration. On the one
hand, public R&D funds often favour unconventional
or challenging innovation projects, which will urge
firms to search and acquire knowledge in multiple
technology areas and increase the diversity of
technical knowledge (Chapman, et al., 2018). In order
to improve the probability of success, firms not only
need to cooperate with different types of partners, but
also should communicate effectively to reduce the
costs of collaboration. On the other hand, R&D
subsidies provide funding and other potential
resources to support various technological
collaboration activities. Bianchi et al. (2019)
proposed the twofold signalling effect of public R&D
funds, which provides correlative personal
information about firms’ quality and innovation
potentiality. Therefore, public R&D funding provides
more opportunities for enterprises to obtain external
financing and work with high-quality partners
(Chapman et al. 2018, Mo et al. 2020). In addition,
the greater the intensity of R&D subsidies, the greater
the importance or quantity of projects financed. This
will facilitate the identification, absorption and
application of external knowledge in related technical
fields and enhance the intensity of technological
collaboration. The rise in the intensity of
technological collaboration will further improve
firms’ innovation performance (Kim, et al., 2021).
Accordingly, we put forward the third hypothesis:
H3. Technological collaboration mediates the
relationship between government R&D subsidies and
innovation performance.
2.4 The Chain-Mediating Role of R&D
Investment and Technological
Collaboration
Literature on technological collaboration
demonstrated that firms’ R&D investment improves
technological collaboration from many aspects
(Cerulli, et al., 2015). First, it avoids information
asymmetry. There is a widespread problem of
information asymmetry during the process of
technological collaboration, because firms know
more about their own resources, information, and
capabilities than their partners. At this point, firms
with higher quality can signal that they are better than
their competitors by increasing R&D investment. By
observing firms’ innovation capabilities, partners can
identify firms’ quality and increase the possibility of
collaboration. Second, it enhances the confidence in
Linking Government RD Subsidies and Innovation Performance: A Chain-mediating Role of RD Investment and Technological
Collaboration
751
successful technological collaboration and boosts
firms’ attractiveness. Innovation is the source of
firms’ sustainable development, big R&D
investments show their confidence to promote
innovation vigorously, thereby enhancing partners’
enthusiasm in technical collaboration. Third, it
improves firms’ absorption capacity. The intensity of
technical collaboration depends on the individual
absorption capacity of the members of the firm
(Laursen, Salter, 2014). Absorption capacity is a by-
product of previous innovation activities and problem
solving; stronger innovation input means that
enterprise innovation activities are more active and
experienced. This helps to recognize and acquire
external knowledge that is valuable to technological
innovation as well as further improve firms’ practices
and processes to analyse and interpret external
information, which will exert positive impact on
technical collaboration and improve innovation
performance. Accordingly, we come up with the
subsequent hypothesis: H4. R&D investment and
technological collaboration will play a chain-
mediating role in the relationship between
government R&D subsidies and innovation
performance.
3 DATA AND MEASURES
3.1 Data
High-tech firms have strong willingness to innovate
and participate in innovation activities frequently,
which is the focus of public R&D funds in China, so
we collect the firm-level data set of the study of listed
high-tech companies on the Shanghai and Shenzhen
stock exchanges. Among all listed firms, a total of
3,083 firms were identified as high-tech enterprises
covering the period of 2001-2019. In order to avoid
common research bias, we clean the data by
following steps. First, the study excludes high-tech
firms whose information disclosure is incomplete.
Second, we rule out firms that are treated by ST and
*ST. Third, listed firms in the financial insurance
category are precluded. Finally, we also exclude
firms with a large number of missing observations
and outliers, i.e. firms with an asset-liability ratio
greater than or equal to 1. After data cleaning, the data
of 483 high-tech listed firms are obtained between
2007 and 2019. These firms are distributed in 16
industries, including the computer and electronic
product manufacturing industry, electrical equipment
manufacturing industry, etc. Research data consist of
firms’ basic data and technological collaboration
data. Basic data are related to R&D expenses, R&D
subsidies, asset-liability ratio, etc. over the years,
which is collected and organized through the
CSMAR database. Government R&D subsidies come
from details of government subsidies in financial
statements and are collected manually. According to
the research of Gong and Zhu (2021), if the title of a
subsidies project contains any of the following words,
namely, “research and development”, “patents”,
“technological innovation”, “technological
transformation”, “independent innovation”,
“copyright”, “research”, “new products”, “science
and technology”, “industrial innovation”, “industrial
upgrading”, “knowledge copyright”, “technical
standards”, “design specifications”, “development”,
“high-tech”, “gazelle”, “Ph.D”, the project is
considered to be awarded government R&D
subsidies. Technological collaboration data relevant
to the intensity of technological collaboration are
collected and calculated manually, mainly through
the patent search and analysis system, which belongs
to the State Intellectual Property Office in China
(CNIPA). First, the sample firms’ patent application
data are retrieved by regular means from 1 January
2001 to 31 December 2020, of which a total of 33764
co-patent applications are collected, and the number
of firms’ partners is counted to calculate the intensity
of technological collaboration. In addition,
continuous variables are winsorized at quantiles of
1% and 99% to avoid the effects of extreme values.
3.2 Measures
Dependent variable: innovation performance (Inno).
Patents are the main objective index of technological
innovation output. They are classified into
three kinds,
i.e. design, utility model and invention in China. The
application time of invention is long due to the stage
of substantive examination; thus its protection time is
longer than that of others. Correspondingly, the
annual fee and agency cost are high. Therefore,
consistent with the existing research (Zhang 2019),
the study treats the number of invention patent
applications as the variable representing innovation
performance.
Independent variable: Government R&D
subsidies (RD_G). According to Bianchi et al.
(2019), it can be measured by two methods. One is a
dummy variable; if a firm is awarded government
R&D financial assistance, the dummy variable equals
1, otherwise 0. The other one is the logarithms of one
plus total amount increased through R&D funds; the
greater the value, the more R&D subsidies firms
receive. The study mainly adopts the second method.
BDEDM 2022 - The International Conference on Big Data Economy and Digital Management
752
Mediator variables: R&D investment (RD_F) and
technological collaboration (Depth). Following the
practice of existing research (Xu, et al., 2021), R&D
investment is evaluated by the logarithms of one plus
total amount of enterprise’s R&D expenses, which is
R&D intensity essentially. Following Yang et al.
(2019), we adopt technological collaboration depth as
a proxy for technological collaboration, which is
evaluated by the average number of co-patent
applications.
Consistent with previous literature (Liu, et al.,
2021, Xu, et al., 2021, Bianchi, et al., 2019, Yang et
al. 2019), this study chooses 11 control variables, i.e.
firm age (Age), firm size (Size), etc., which are
demonstrated in Table 1.
Table 1: Variables and Their Measurements
Variable Name Symbol Measurable Indicato
r
Innovation
performance
Inno Number of patent
applications for inventions
Government
R&D subsidies
RD_G Log of ( 1+ total amount
raised through R&D
subsidies)
R&D
investmen
t
RD_F Log of ( 1+ total R&D
expenses)
Technological
collaboration
Depth Average number of co-
patent applications
Firm size SIZE Log of the total asse
t
Firm age AGE Years since a firm was
funde
d
Leverage LEV Ratio of total debt to total
assets
Return on total
assets
ROA Ratio of net profit to total
average assets
Export Export The dummy variable equals
1 if the product of firm is
exported abroad, otherwise
0
Firm group Group The dummy variable equals
1 if the company belongs to
a firm group, otherwise 0
Institutional
environmen
t
Institution Market-oriented total index
score (Fan et al., 2018)
Market
concentration
HHI Heffendahl Hirschman
index
State-owned
enterprises
SOE The dummy variable equals
1 if the company is state-
owned, otherwise 0
Industry Industry According to the
technology intensity classes
of OECD, there are six
industry dummies, i.e. high-
tech manufacturing, high
medium-tech
manufacturing, etc.
(Herstad et al., 2015)
Year Year Dummy variables for the
years 2008
–
2019
3.3 Empirical Results
The descriptive statistics of all variables are shown in
Table 2. The mean value of innovation performance
is 21.547, its standard deviation value is 86.622,
demonstrating that the innovation output of Chinese
high-tech firms is in its infancy with a big gap
between high and low. The mean value of R&D
subsidies is 10.382 with a maximum of 18.390 and a
minimum of 0, which suggests that government R&D
subsidies is at a relatively top-level stage, but there is
also a big gap among firms. The average R&D
investment is 18.037 with a minimum of 15.251 and
a maximum of 21.512, displaying that the intensity of
R&D is comparatively balanced. The mean value of
technological collaboration is 1.329 with a maximum
of 26.125 and a minimum of 0, indicating that
technological collaboration depth is at a low level
with a big gap between high and low.
Table 2: Descriptive Statistics.
Variable Obs. Mean SD Min Max
Inno 5 205 21.547 86.622 0.000 1 919.00
RD_G 4 757 10.382 6.742 0.000 18.390
RD_F 3 722 18.037 1.253 15.251 21.512
Depth
5 206 1.329 3.689 0.000 26.125
SIZE 4 722 21.973 1.106 19.902 25.093
AGE 5 206 17.269 5.468 7.000 32.000
LEV
4 722 40.731 18.997 4.587 81.856
ROA 4 724 0.046 0.047 -0.114 0.194
Expor
t
4 136 0.582 0.493 0.000 1.000
Group 4 721 0.962 0.191 0.000 1.000
Institution
4 724 7.133 3.014 0.000 10.780
HHI 4 723 0.099 0.100 0.015 0.651
SOE 4 724 0.365 0.481 0.000 1.000
Note: Obs. denotes number of countries in the baseline model. SD denotes standard
deviation.
Table 3 reports the regression analysis results
through the causal-step method (Model 1 to 5). The
coefficient of RD_G shows that R&D subsidies can
improve firms’ innovation performance in Model 1
(b=0.009, p<0.01), which provides support for
Hypotheses 1. Model 2 indicates that the R&D
subsidies increase the R&D investment (b=0.004,
p<0.05). Meanwhile, the coefficient of RD_F is 0.080
(p<0.5) in Model 5, indicating that R&D investment
promotes innovation performance. Thus, Hypotheses
2 is confirmed. Model 3 implicitly assumes that R&D
subsidies have a constructive effect on technological
collaboration (b=0.014, p<0.01). The coefficient of
Depth in Model 5 suggests that technological
collaboration benefits innovation performance
(b=0.059, p<0.01). Therefore, technological
collaboration is an important pathway for R&D
subsidies to influence innovation performance,
Linking Government RD Subsidies and Innovation Performance: A Chain-mediating Role of RD Investment and Technological
Collaboration
753
providing, Hypotheses 3 is confirmed. Model 4
demonstrates that R&D investment undoubtedly
promotes the depth of technological collaboration
(b=0.132, p<0.05). The above regression results
together display that R&D investment and
technological collaboration play a chain-mediating
role in the links between government R&D subsidies
and firm innovation performance. Therefore,
Hypotheses 4 is confirmed.
Table 3: The Mediating Role of R&D Investment and
Technological Collaboration
Variable Inno RD
_
F De
p
th Inno
Model 1 Model 2 Model 3 Model 4 Model 5
RD_G
0.009
***
(3.04)
0.004
**
(2.28)
0.014
***
(2.94)
0.006
(1.06)
0.004
(1.22)
RD
_
F
0.132
**
(
2.10
)
0.080
**
(
2.41
)
De
p
th
0.059
***
(
16.96
)
Constant -0.364
(-0.50)
4.715
***
(6.16)
-11.128
***
(-10.36)
-10.064
***
(-7.33)
-0.742
(-0.99)
SIZE
t-1
-0.035
(-1.05)
0.590
***
（19.95
）
0.400
***
(8.08)
0.274
***
(3.34)
-0.098
**
(-2.29)
AGE
t-1
-0.011
(-1.30)
-0.035
（-0.85）
0.008
(0.71)
-0.000
(-0.00)
-0.006
(-0.85)
LEV
t-1
0.002
(
1.19
)
0.003
**
(
2.23
)
0.002
(
0.67
)
0.002
(
0.54
)
0.004
**
(
1.99
)
ROA
t-1
3.075
***
(
6.53
)
3.284
***
(
11.56
)
4.207
***
(
5.00
)
3.721
***
(
3.62
)
2.067
***
(
4.08
)
Ex
p
ort
t-1
-0.086
(
-1.64
)
-0.087
**
(
-2.28
)
0.114
(
1.34
)
-0.026
(
-0.25
)
-0.085
(
-1.57
)
Grou
p
t-1
0.105
(
0.91
)
-0.021
(
-0.29
)
0.098
(
0.37
)
0.023
(
0.07
)
0.137
(
1.08
)
Institution
t-1
0.013
(0.57)
-0.021
(-0.78)
0.071
**
(2.22)
0.083
**
(1.98)
0.050
**
(2.33)
HHI
t-1
-0.486
(
-1.59
)
0.592
***
(
3.11
)
-0.591
(
-1.23
)
-0.207
(
-0.32
)
-0.301
(
-0.087
)
SOE
t-1
0.295
***
(
4.19
)
0.183
*
(
1.72
)
0.471
***
(
4.43
)
0.262
*
(
1.88
)
0.222
***
(
3.03
)
Yea
r
Yes Yes Yes Yes Yes
Industr
y
Yes Yes Yes Yes Yes
Obs. 3 584 2 985 3 693 2 077 2 985
F - 117.01
***
- - -
Wald 387.44
***
- 240.76
***
127.91
***
594.33
***
Note: (i) The values in parentheses are the p-values. ***, ** and * display significance at
the level of 1%, 5% and 10%, respectively, (ii) Model 2 was estimated by a regression
model for panel data, the other models were estimated by the negative binomial models
for panel data, whether to choose a fixed effect model or a random effect model were
determined by the Haussmann test.
Table 4 depicts the results from the bootstrap test
using Process (Hayes) for SPSS with 5000 samples
and a 95% confidence interval. The direct effects of
R&D subsidies fail to be statistically significant and
is reported in Table 4. R&D investment and
technological collaboration appears as valid
mediation mechanisms between R&D subsidies and
innovation performance, consistent with the
conclusion of the causal step method. These findings
strongly support that public R&D subsidies have
indirect influence on innovation performance through
increases in R&D investment and technological
collaboration. In addition, Table 4 confirms that R&D
subsidies promote R&D investment, R&D
investment has a progressive influence on
technological collaboration, thus increasing firms’
innovation performance. Consequently, we find
support for H4, which predicts the chain-mediating
role of R&D investment and technological
collaboration.
Table 4: Bootstrap Test Results.
The mediation
path
Indirect
effects
Confidence
interval (95%)
Direct
effect
Lower
limit
Upper
limit
RD_G—RD_F—
Inno
0.229 0.124 0.361
0.305
(1.24)
RD_G—RD_F—
Depth
—
Inno
0.046 0.022 0.085
RD_G—Depth—
Inno
0.165 0.007 0.373
Total mediation
effect
0.440 0.234 0.698
Note: The values in parentheses are the p-values.
3.4 Robustness Checks
Three important robustness checks were conducted,
the results for which are shown in Table 5. First, we
tested the endogenous problems that may exist
between variables using system GMM estimation for
dynamic panel data. With leverage, ROA, R&D
investment at time t to t-2 as instrumental variables,
the coefficient of RD_G
t-1
is positive and significance
(b=1.118, p<0.05), passing the AR (2) test and
Hansen test in Model 6. Second, we checked the
sensitivity of key variables. With dummy variable as
a substitute measure of R&D subsidies, the
consequences show that the coefficient of RD_G is
positive in Model 7(b=0.131, p<0.01). With the count
of licensed patents instead of patent application
quantities for inventions in regression, Model 8
shows that the coefficient of RD_G is 0.009(p<0.05).
Finally, we replace the negative binomial models
with a panel Tobit model, which shows that the
coefficient of RD_G is 0.888 in Model 9(p<0.01). In
these settings, we obtain the same result of those
presented above.
Table 5: Robustness Checks Results.
Variable Inno
Model 6 Model 7 Model 8 Model 9
RD_G
t-1
1.118
**
(2.37)
0.009
**
(2.44)
RD_G
0.131
***
(2.96)
0.888
***
(3.35)
Inno
t-1
0.774
***
(
63.02
)
Constant 261.148
（0.91）
-0.460
(-0.63)
-1.179
***
(-1.25)
-490.000
***
(-7.26)
BDEDM 2022 - The International Conference on Big Data Economy and Digital Management
754
Control
Variables
Yes Yes Yes Yes
Year Yes Yes Yes Yes
Industry Yes Yes Yes Yes
Obs. 2 247 3 376 3 362 3 693
AR(2) 0.418 - - -
Sargan 0.000 - - -
Hansen 0.273 - - -
Wald 22 879.6
***
386.56
**
*
649.97
**
*
267.28
***
Note: The values in parentheses are the p-values. ***, ** and * indicate significance at
the level of 1%, 5% and 10% respectively. Due to limited layout, coefficients of control
variables are not listed here; if you’re interested, please contact the corresponding author.
4 CONCLUSIONS
Technological innovation is a fundamental factor to
boost the sustainable development of firms. As
important public policies, government R&D
subsidies support and stimulate firms to innovate
continuously, thus raise innovation performance. Our
intention in this paper has been to offer an explication
on how government R&D subsidies shape firms’
likelihood to implement technological innovation.
Using panel data from 483 Chinese high-tech listed
firms during the period 2007 and 2019, this research
estimates a chain-mediated model to investigate the
causal connection among R&D subsidies, R&D
investment, technological collaboration and
innovation performance.
Our findings provide strong support for the
assumption that R&D subsidies are conducive to
promoting innovation performance, proving the
effectiveness of public R&D funding in China. More
importantly, this research demonstrates that R&D
investment and technological collaboration illustrate
part of the process through which enterprises convert
the benefits of public R&D funds into enhanced
innovation performance. Particularly, the research
elaborates three substitute methods that allow
businesses to create value from government R&D
subsidies. For instance, consistent with previous
studies, our results demonstrate that R&D subsidies
enhance firms’ creativity by affecting their private
investment in R&D. Furthermore, technological
collaboration plays an important mediating role in the
association between government R&D subsidies and
innovation performance. In addition, our findings
also display that R&D investment and technological
collaboration play chain-mediating role through
which R&D subsidies have indirect influence on
innovation performance. Generally speaking, these
conclusions discover new methods through which
R&D subsidies drive innovation performance of
corporations, which further has implications for the
corporation innovation literature from the perspective
of signalling theory.
Our results have also implications for
practitioners. First, our results point out that public
R&D subsidies enhance innovation performance.
Therefore, the policy makers have an obligation to
keep on providing more R&D funding, firms should
actively apply for public R&D subsidies and enhance
their utilization efficiency, so as to jointly promote
innovation performance. What’s more, according to
the reported effects of R&D subsidies, it is obvious
that firms with reward of state R&D funds should
promote the intensity of R&D investment because it
not only boosts innovation performance directly, but
also plays an indirect role in strengthening
technology innovation by technological
collaboration. Third, firms should also establish open
innovation strategies, make good use of existing
R&D resources, increase the enthusiasm of partners,
and work together for more challenging technological
innovation activities. Finally, drawing on the findings
on multiple mediation roles of R&D investment
during technological innovation process, for one
thing, firms can choose different ways to achieve the
goal of improving innovation performance; for
another, policy-makers should take notice of the
signal transmitted by government R&D funds, and
manage them in a targeted manner, so as to increase
the effectiveness of public R&D funding.
However, this investigation is not without
limitations, and future work may explore the
following issues. Firstly, because the data of R&D
subsidies are collected manually, the study only
selected the effect of the amount of R&D subsidies
on innovation performance, but did not classify the
specific content of subsidies or funding agencies to
explore its impact on innovation; more research is
needed in this field. Secondly, considering the
availability of data, the research selected high-tech
listed companies as research samples, not non-listed
enterprises, a deeper analysis with a wider sample is
needed too. In addition, this paper only tests the
impact of R&D subsidies, R&D investment,
technological collaboration on innovation
performance, it could be better to recognize other
factors with potential effects on how R&D subsidies
create value for firms, such as technical cooperation
governance, science cooperation and so on.
ACKNOWLEDGEMENTS
This study was funded by the Education Department
in Jilin Province and Jilin Provincial Key Laboratory
Linking Government RD Subsidies and Innovation Performance: A Chain-mediating Role of RD Investment and Technological
Collaboration
755
of Human Health Status Identification and Function
Enhancement, Grant Number are JJKH20200596SK
and 20200601004JC.We also gratefully
acknowledges the support of Xiaoli Zhong.
REFERENCES
Bérubé, C., Mohnen, P. (2009) Are firms that receive R&D
subsidies more innovative? Canadian Journal of
Economics, 42 (1),206–225.
Bianchi, M., Murtinu, S., Scalera, V. G. (2019) R&D
subsidies as dual signals in technological
collaborations, Research Policy, 48(09), 1-19.
Cerulli, G., Gabriele, R., Poti', B. (2015), The role of firm
R&D effort and collaboration as mediating drivers of
innovation policy effectiveness, Industry and
Innovation, 23(5), 1-22.
Chapman, G., Lucena, A., Afcha, S. (2018) R&D subsidies
& external collaborative breadth: differential gains and
the role of collaboration experience, Research Policy,
47(3), 623-636.
Chen, J. (2021) Research on the relationship among
government support, enterprise R&D investment and
technological innovation performance, Forecasting, 40
(02), 40-46.
Fan, G., Wang, X.L., Hu, L.P. (2018), Marketization index
of China's provinces: neri report 2018, Social Sciences
Academic Press, Beijing.
Gao, Y., Hu, Y., Liu, X., et al. (2021) Can public R&D
subsidy facilitate firms' exploratory innovation? The
heterogeneous effects between central and local
subsidy program, Research Policy, 50(04), 104-221.
Gong, H., Zhu, L.X. (2021) The ‘halo effect’ of
government R&D and Non- R&D subsidies on the
external financing of enterprises- based on the analysis
of the new energy enterprises, Science & Technology
Progress and Policy, 38(04),70-77.
Herstad, J., Sandven, T., Ebersberger, B., et al. (2015)
Recruitment, knowledge integration and modes of
innovation, Research Policy, 44(01), 138-153.
Kim, K, Sang, O. C., Lee, S. (2021) The effect of a financial
support on firm innovation collaboration and output:
Does policy work on the diverse nature of firm
innovation? Journal of the Knowledge Economy, 12
(2), 645-675.
Laursen, K., Salter, A.J. (2014) The paradox of openness:
appropriability, external search and collaboration,
Research Policy,43(05),867-878.
Liu, S., Du, J., Zhang, W. et al. (2021) Opening the box of
subsidies: which is more effective for innovation?
Eurasian Economic Review, Vol.11,pp. 421-449.
Mo,J., Lee, W., Mortara L . (2020), Do government R&D
subsidies stimulate collaboration initiatives in private
firms? Technological Forecasting and Social Change,
151, 1-14.
Plank, J., Doblinger, C. (2018) The firm-level innovation
impact of public R&D funding: evidence from the
German renewable energy sector, Energy Policy, 113,
430–438.
Wu, R. R., Liu, Z. Y, Ma, C. L., Chen, X. F. (2020) Effect
of government R&D subsidies on firms’ innovation in
China, Asian Journal of Technology Innovation,
28(01), 42–59.
Xu, J., Wang, X. H., Liu, F. (2021) Government subsidies,
R&D investment and innovation performance: analysis
from pharmaceutical sector in China, Technology
Analysis & Strategic Management, 33(5), 535-553.
Yang, L., Xu, C., Wan, G. (2019) Exploring the impact of
TMTs' overseas experiences on innovation
performance of Chinese enterprises: The mediating
effects of R&D strategic decision-making, Chinese
Management Studies, 13 (4),1044-1085.
Yang, Z.Q., Yuan, M., Zhang, Y. T. (2021) Enterprise
R&D Innovation and bond credit spreads: analysis
based on the signal transmission theory, Journal of
Shanghai University of Finance and Economics, 23
(01), 42-60.
Yi, J., Murphree, M., Meng, S. et al. (2021) The more the
merrier? Chinese government R&D subsidies,
dependence and firm innovation performance, Journal
of Product Innovation Management, 38 (2), 289-310.
Yu, F. F., Guo, Y., Le-Nguyen, K., Barnes, S. J., Zhang, W.
T. (2016), The impact of government subsidies and
enterprises’ R&D investment: a panel data study from
renewable energy in China, Energy Policy, 89, 106–
113.
Zhang, X. (2019) The impact of government R&D
subsidies on enterprise technology innovation—Based
on evidence from Chinese listed companies, American
Journal of Industrial and Business Management, 9, 720-
742.
BDEDM 2022 - The International Conference on Big Data Economy and Digital Management
756
