Американский Научный Журнал MEASURING THE EFFICIENCY OF TECHNOLOGICAL INNOVATION IN CHINA’S HIGH-TECH INDUSTRY BASED ON THE DEA-MALMQUIST INDEX METHOD (19-23)

American Scientific Journal № ( 42) / 2020 19

ЭКОНОМИЧЕСКИЕ НАУКИ

MEASURING THE EFFICIENCY OF TECHNOLOGICAL INNOVATION IN CHINA'S HIGH -TECH
INDUSTRY BASED ON THE DEA -MALMQUIST INDE X METHOD

Xu Xiaoyun
PhD graduate student, Economic faculty,
Belarusian State University,
Minsk
Korotkevich Alexey Ivanovich
PhD of Economic Sciences , Head of the Department of Banking,
Belarusian State University,
Minsk
DOI: 10.31618/asj.2707 -9864.2020.2.42.44
Abstract. This article uses DEA data envelopment analysis method, combined with Malmquist index,
through the analysis of China's high -tech industry scientific research input and outpu t data from 2008 to 2018,
obtained the Malmquist index of comprehensive innovation efficiency of China's high -tech industry in the past 10
years, then analyzes the efficiency of technological innovation in China's high -tech industry, and puts forward
optim ization and policy recommendations.
Keywords . DEA, Malmquist index, high technology , innovation

1 Introduction
Industry is the backbone of the social economy,
and high -tech industries are the industrial clusters that
produce high -tech products. The develop ment of high -
tech and basic industries is essential to promote the
upgrading of ind ustrial structure and increase labor
productivity and resource utilization.
China's high -tech industry started relatively late,
but after decades of rapid development, the l evel of
scientific research has approached or reached the
world's advanced level and has great development
potential. Technological innovation is the driving force
for the development of high -tech industries, and
innovation efficiency is related to the dev elopment
strategy, resource allocation, and structural
optimization of the entire industrial industry.
Therefore, studying the innovation efficiency of
China's high -tech industries is crucial to promoting the
construction of an innovative society.
This art icle studies the innovation efficiency of
China's high -tech industry, which is a typical efficiency
evaluation problem of multi -factor decision -making. At
the same time, because this article uses the panel data
of China's high -tech indust ry (5 industries) from 2008
to 2018, it is suitable to use DEA -based Malmquist
Index method is used to measure innovation efficiency.
By using the DEA -Malmquist index efficiency
evaluation method to measure the scientific research
activities of the five ty pes of high -tech i ndustries, to
study the changes in the efficiency of China's scientific
and technological innovation in the past 10 years, and
to propose optimization suggestions based on the
current development strategies of China's high -tech
industries and the internati onal situation.
2 DEA -Malmquist efficiency evaluation model
The Input -Output Method, founded by the famous
economist Wassily Leontief, is widely used to study
issues in the national economy and analyze the balance
between production input s and product dist ribution
among various departments. Data Envelopment
Analysis (DEA) based on the "input -output model" is a
non -parametric method for evaluating production
efficiency in operations research. It can be used to
evaluate multi -objective decis ion -making problem s,
and it can calculate efficiency while considering scale
Profit factor, get comprehensive efficiency evaluation
results with research significance.
The mathematical description of the DEA -
Malmquist index efficiency evaluation model is a s
follows:
For the period from t to t+1, the measurement of
innovation efficiency can be measured by the following
Malmquist index.
Among them: (�������+1) and (�������,�������) are the input and
output in perio ds t+1 and t respectively; ������0������ and ������0������+1
represent periods t and period t+1 refers to the distance
function of technological innovation ������������ in period t. If
the calculation result of this function is greater than 1,
it indicates that the rate of technological inno vation
increases from period t to period t+1. This paper uses
the linear programming method to calculate the
distance function related to innovation input and
innovation output to measure the Malmquist index to
illustrate the innovation efficiency of high -tech
industries.
3 Research data and evaluation indicators
3.1 Data selection
The time span of the research data in this paper is
10 years. In order to more accurately represent the
input -output relationship, this paper c hooses the lag
period of the input -output model to be 1 period [1], that
is, the innovation input indicator data is from 2008 to
2017, corresponding to the innovation output data is
from 2009 to 2018, and the "exponential smoothing ( ) ( )
( )
( )
( ) ()
1 0 1 1 0 1 0 1 1 1 00
,, , , , 1 ,,
tt t t t t t t t t tt t t t t
D x y D x y M x y x y D x y D x y
+ + + + ++ +
 = =  

20 American Scientific Journal № ( 42) / 2020
method" is used to predi ct some missing data, as shown
in F igure 1.

Figure 1 The input -output model with a lag of one period

The data used in this study are from the "China
High -tech Industry Statistical Ye arbook" published by
the National Bureau of Statistics of the correspond ing
year and the official website of the National Bureau of
Statistics [2].
3.2 Evaluation indicators of scientific and
technological innovation efficiency
According to the description in the latest Chinese
"Classification of National Economic Industries"
(GB/T4752 —2017), high -tech industries
(manufacturing industries) refer to manufacturing
industries with relatively high R&D investment
intensity in the national economic industries, in cluding
the following 6 categories: Pharmaceutical
manufacturing; aviati on, spacecraft and equipment
manufacturing; electronic and communication
equipment manufacturing; computer and office
equipment manufacturing; medical equipment and
instrumentation sys tems; information chemical
manufacturing (newly added). Due to the avail ability of
data, this paper selects only five categories for research,
namely: aerospace equipment manufacturing,
electronic and communication equipment
manufacturing, computer equipme nt manufacturing,
medical equipment and pharmaceutical manufacturing;
the information chemical manufacturing industry is
temporarily not studied.
According to the classification of the complex
giant system theory [3], technological innovation is
divided in to three categories: knowledge innovation,
technological innovation and management innovation.
Therefore, innovation input and output indicators will
be jointly explained by multidimensional data such as
knowledge, capital, technology, and markets. This
article chooses the following indicators to represent
innovation input and innovation output.
Innovation investment indicators include full -time
equivalent of R&D personnel (X1), R&D expenditure
(X2), and new product technology development
expenditure (X3). Innovation output indicators are
represented by the number of patent app lications (Y1)
and the sales revenue of new products (Y2).
The evaluation index information of innovation
efficiency is shown in Table 1.
Table 1
Evaluation indicators of scientific an d technological innovation efficiency
INDEX VARIABLE DESCRIPTION
Innov ation
input
Full -time equivalent of R&D
personnel X1 Full -time equivalent of scientific research
personnel during the period
R&D funding X2 Funds invested in scientific research during the
period
New product technology
development funding X3 New produc t technology development funding
during the period
Innovation
output
Number of patent applications Y1 Number of patent applications for scientific
research projects during the period
New product sales revenue Y2 New product market sales during the perio d
4 Empirical study of innovation efficiency based
on DEA -Malmquist index
This article uses DEAP 2.1 developed and
maintained by the University of Queensland as an
analysis tool. By inputting the R&D input and output
indicators of China's high -tech indust ry from 2008 to
2018, the Malmquist index of the comprehensive
innovation efficiency of China's high -tech industry in
the past 10 years is obtained, including annual
Comprehensive and detailed indicators such as EFFCH

American Scientific Journal № ( 42) / 2020 21

(Technology Change Rate Index), TECHCH
(Technology Up date Change Rate Index), PECH (Pure
Technology Update Change Rate Index), SECH (Scale
Change Rate Index), TFPCH (Comprehensive
Productivity Change Index), etc. [4], then analyze the
changes in China's high -tech industry innovation index,
the efficiency of technological innovation input and
output, and the reasons for the different levels of
innovation.
4.1 Analysis of the overall Malmquist innovation
efficiency change index of the high -tech industry
According to the system design, input the 1 0 years
of scie ntific and technological research and
development indicators, and obtain the comprehensive
innovation efficiency change index TFPCH (Total
Factor Productivity Change) and the index composition
of the five industries in the high -tech industry . The
summary r esults are shown in Table 2.
Table 2
Malmquist index summary of annual means
YEAR PERIOD EFFCH TECHCH PECH SECH TFPCH
2008 -2009 1 - - - - -
2009 -2010 2 0.802 0.863 0.909 0.882 0.692
2010 -2011 3 1.182 1.420 1.039 1.138 1.679
2011 -2012 4 1.115 0.802 1.004 1.111 0.901
2012 -2013 5 0.929 0.969 0.986 0.942 0.900
2013 -2014 6 1.010 1.063 0.962 1.050 1.074
2014 -2015 7 1.089 0.870 0.994 1.096 0.947
2015 -2016 8 0.991 1.080 0.995 0.996 1.070
2016 -2017 9 0.942 1.241 0.974 0.967 1.169
2017 -2018 10 1.130 0.906 1.026 1.101 1.024
Average Value 1.021 1.021 0.988 1.031 1.047
Figure 2 is drawn based on the TFP parameters in
the statistical results of Table 2 to show the changes in
the overall innovation efficiency of the high -tech
industry. Each year' s TFP parameter is the rate of
change of the innovation efficiency index of the current
year compared to the previous year.

Figure 2 The overall innovation efficiency change rate of the high -tech industry TFPCH

On the whole, the average TFP change rate o f the
comprehensive innovation index is 1.047, indicating
that the comprehensive innovation efficiency of China's
high -tech industry has been steadily improving in the
past 10 years , innovation activities have continued to
increase, and the level of scienc e and technology has
also been continuously improved.
From the perspective of annual changes, the
change rate of China's high -tech industry
comprehensive innovation index 2010 -2011 was
particularly different, reaching 1.677, indicating that
the current hig h-tech industry comprehensive
innovation TFP efficiency increased by 67.9%
compared with 2009 -2010, which was mainly due to
“The global financial crisis” in 2008 led to a decrease
in innovation input and output indicators for the current
period (2008, 2009 ). With the recovery of the economic
level and the increase in the scale of R&D investment
in 2010, it directly led to a substantial increase in the
innovation efficiency of high -tech industries in 2010 -
2011.
From 2010 to 2018, three of the annual innovati on
efficiency TFP indexes were between 0.9 and 1.0, and
the innovation efficiency index was relatively stable,

22 American Scientific Journal № ( 42) / 2020
mainly due to the weak technological change index of
the current scien tific research activities. In the last three
years, all were greater than 1 .0, indicating that China's
innovation efficiency has continued to improve, mainly
due to the increase in the level of scientific research
technology and the scale of scientific res earch
breakthroughs.
4.2 Malmquist innovation efficiency index
analysis of various industries in the high -tech industry
According to the analysis results of the DEAP
software, the comprehensive innovation index change
rate information table of each industr y in the high -tech
industry is compiled, as shown in Table 3.
Table 3
Malmq uist index summary of firm means
INDUSTRY FIR
M
EFFC
H
TECHC
H
PEC
H
SEC
H
TFPC
H
Electronics
and
communications
manufacturing
1 1.003 1.055 1.000 1.003 1.058
Computer
equipment
manufacturing
2 1.000 1.005 1.000 1.000 1.005
Aerospace
manufacturing 3 1.150 1.0 15 1.000 1.150 1.167
Medical
device
manufacturing
4 1.000 0.946 1.000 1.000 0.946
Pharmaceutica
l manufacturing 5 0.933 1.004 0.936 0.996 0.936
Among them, according to the TFPCH parameters
in Table 3, the innovation efficiency change rate of 5
industrie s in the high -tech industry in the past 10 years
is obtained, and an intuitive chart is drawn, as shown in
Figure 3.

Figure 3 The average value of TFPCH in the innovation efficiency change rate of each industry
in the high -tech industry

It can be seen in tuitively from Figure 3 that in the
pas t 10 years, China has made tremendous progress in
the communications equipment manufacturing,
computer equipment manufacturing, and aerospace
vehicle manufacturing industries. Innovation efficiency
has increased year by year, and the level of science and
technology has also been continuously improved.
Among them, the innovation efficiency of the
aerospace manufacturing industry has increased the
most. It can be seen from Table 3 that it is mainly
caused by SECH (R&D in vestment scale) and EFFCH
(renewal of s cientific research technology), indicating
that China has invested huge funds in the aerospace
field and technical support, and achieved outstanding
achievements.

American Scientific Journal № ( 42) / 2020 23

The TFPCH index of the medical device
manufacturing in dustry and the pharmaceutical
manufactu ring industry is below 1.0, indicating that
China's research and development foundation in this
field is relatively weak and the innovation efficiency is
not high in the past 10 years. These two fields have
always bee n weak industries in China’s industry a nd
require more capital investment and R&D
accumulation. According to the "Made in China 2025"
plan, "biomedicine and high -performance medical
devices" are priority areas of technology for
development. In the next 10 years, the comprehensive
innovation ind ex of China's pharmaceutical and
medical fields may increase rapidly.
5 Conclusions and recommendations
This paper adopts DEA data envelopment analysis
method, combined with Malmquist index, and DEAP
2.1 software. By analyzing the scientific research data
of China's high -tech industry from 2008 to 2018, the
Malmquist index of the comprehensive innovation
efficiency of China's high -tech industry in the past 10
years is obtained, including Annual comprehensive and
detail ed indicators such as EFFCH (technical change
rate), TECHCH (technical update rate of change),
PECH (pure technical efficiency change rate), SECH
(scale change rate), TFPCH (comprehensive
production change rate), etc., and then detailed It
analyzes the cha nges in the overall innovation index of
China's high -tech industry and the changes in the sub -
industry innovation index, and analyzes the specific
reasons for these changes.
The overall innovation efficiency of China's high -
tech industry has been increasin g year by year. The
electronics and com munications manufacturing,
aerospace, and computer equipment manufacturing
industries have developed well, and the innovation
index has increased rapidly. However, the
comprehensive innovation efficiency index of the
medical device manufacturing and pharma ceutical
manufacturing industries has not increased
significantly, and there is a problem of low innovation
efficiency. Based on the above research results, in order
to improve the innovation level of China's high -tec h
industry, the following optimization opinions and
policy recommendations for scientific research
activities are proposed:
First, increase scientific research investment. For
the electronics and communications manufacturing,
and computer equipment manufac turing industries,
investment in resear ch and development should be
increased, which will help enhance the industry's
innovation level. In the field of aerospace
manufacturing, technology introduction and talent
investment should be increased. While mainta ining
R&D funding, the level of innovat ion will be even
greater. For the fields of pharmaceutical manufacturing
and medical devices, it is necessary to increase
investment in basic research and development
facilities, promote talent training and introducti on, and
strengthen financial support, w hich will benefit the
level of technological innovation in these two fields and
enhance industry competitiveness.
Second, reasonable resource allocation. The
continuous improvement of China's technological level
is du e to the rapid development of China's e conomy,
which has invested a lot of capital and manpower in
high -tech fields, but excessive resource investment has
caused a waste of resources in some areas. Therefore,
under current conditions, improving resource
utilization and optimizing the industrial structure are
one of the urgent problems in China's industrial sector.
Third, take the path of sustainable development.
While vigorously developing industry, we must also
pay attention to the protection of the ecolog ical
environment. The development of hi gh-tech industries
has a stimulating effect on all industrial sectors and is a
key development area in China under the current severe
international environment. High -tech industries are
generally accompanied by proble ms such as high
pollution and high ener gy consumption. How to
develop green, sustainable, and environmentally
friendly production technologies on the premise of
improving the industrial level and improving industrial
competitiveness is related to the devel opment of the
entire society and China’ s future sustainable
competitiveness.

References
1. Dan Wang, Xinli Zhao (2016) "The Time Lags
Effects of Innovation Input on Output in National
Innovation Systems: The Case of China", Discrete
Dynamics in Nature and Society, 15.
2. National Bureau of Stat istics of China.
http://www.stats.gov.cn/ .
3. Qian Xuesen, Yu Jingyuan, Dai Ruwei (1990)
A new scientific field -open complex giant system and
its methodology. Journal of Nature 1: 3 -10.
4. Li Meijuan, Chen Guohong ( 2003) Research
and Application of Data Envelopment Analysis (DEA).
China Engineering Science 6: 88 -94.
5. Wu Qi, Wu Chunyou (2009) Research on
Energy Efficiency Evaluation Model Based on DEA.
Management Science 1:103 -112.
6. Yang Qingke, Duan Xuejun (2014) Research on
the Spatiotemporal Measurement and Provincial
Differences of High -tech Industry Development
Efficiency Based on DEA -Malmquist Model.
Economic Geography 7: 103 -110.
7. Wang Haifeng, Luo Yafei, Fan Xiaoyang
(2010) International comparison of R&D innovation
evaluation based on super -efficiency DEA and
Malmquist index. Science of Science and Management
of Science and Technology 4: 42 -49.