Application of System Dynamics on Policy Analysis of
Resource Allocation of Scientific Research
Xu Qingrui
Chen Jin
Wu Gang
Zhejing University, School of Management,
Hangzhou 310027, P.R.China
ABSTRACT
By the thought of coordinative development between
Science & Technology, economy, education and finance, this
paper first concerns the problems facing China on the
resource allocation of Scientific Research. A comparative
study on both developed and developing countries is made.
In the meantime, the mechanism of the coordinative
development between Science & Technology, economy, education
and finance, the coordinative development between
Scientific Research (Basic Research), Applied Research &
Development as well as the priority of Scientific
Research in different stages of social & economic
development, a system dynamic model is constructed,
focusing the” analysis of scale & speed of resource
allocation for Scientific Research in China.
INTRODUCTION
Over the past several decades many countries have come
increasingly to the realization of Basic Research may hold
the underpinning to their competitive advantages and
sustainable development. Advanced countries spend 12 - 30%
of their Research & Development (R&D) expenditure on Basic
Research by different patterns. Though China has made
marverous achievements in the field of Basic Research,
China’s spend only around 6.3 - 7.7% of R&D expenditure on
Basic Research do affect the stamina of science & technology
progress of China. Increasing the input on Basic Research is
emphasized by Chinese Government and many scholars.
As China is still one of the less developed economics, large
scale investment on Basic Research as developed countries is
impossible, but the low increasing rate of Basic Research
input also affects the catch up with the scientific &
technological more advantaged. As the prospects for
entering into modern industrialisation are not so obvious
nor can be taken for granted that late-comers have the
advantages, less development countries including China must
pay attention to the coordinative development so that the
development can take place through synchronization of
buildup. science & technology capacity and elimination of
SYSTEM DYNAMICS '93 593
buildup science & technology capacity.and elimination of
obstacles to development. .A distinguishing pattern on Basic
Research input may be used for China.
In order to deal with above-mentioned problems, a more
careful and systematical analysis aided by System Dynamics
is needed. During the model-building & simulation, the
thought of coordinative development is stressed with the
following contents:
» Coordinative development between science & technology,
economy, education and finance system;
» Coordinative development between Basic Research,
Applied Research and Experimental Development;
Coordinative with the Development Stage of one country.
Previously, the system analysis as well as the experience
diagnosis were taken as the main means to deal with the
resource allocation on Basic Research, in this paper the
policy analysis (including policy design and policy test)
will. be based.on the comparative study and the. mechanism
analysis (See Figure 1).
somparative mechaniam understanding
model co tection ——
verification
study
pattern
policy. design
reconization
policy test
policy suggestion
(strategy)
Figure 1. Research.Guideline & Methodology
SEEKING. PATTERNS ON RESOURCE. ALLOCATION. ON SCIENTIFIC
RESEARCH BY COMPARATIVE STUDY
The objectives of. the comparative study are following
issues:
. Finding out the patterns of. Basic Research input;
Analysing the relationship between resource allocation
504 SYSTEM DYNAMICS '93
patterns and performance of a country; =
Selecting the resource allocation strategies for
Basic Research input, which is necessary by policy
analysis.
Seven countries were chosen including (1) U.S.A.; (2) Japan;
(3) West Germany; (4) France; (5) U.K.; (6) Brazil; (7)
India. And four index are used to evaluate the performance
in these countries, these index are GNP (or GDP), balance of
high-technology trade, patents and R&D personnel. The social
& economic development stage is classified mainly according
to the GNP (or GDP) per capita, here stage I is featured as
300-2000 US$ ~while Stage. II and. III are 2000-4750 US$ and
more than 4750 US$ respectively.
Figure 2 illustrated the ratio of Basic Research input over
total R&D expenditure. The characteristics between Basic
Research over R&D in different development stages in these
seven countries is summarized as Table 1, which tells that
the proportion of Basic Research over R&D is increased
steadily in stage I and stage II, and declined a little bit
in stage ‘III.
Basie Research Input
. (s)
Total Research & Development Input
Country
Stage t Btage IT Stage III
ULB. AL a6 §.6 - 13.9 123.2 - 18.6
UK. 10.9 11.9 ~ 16.0 6.8
France 120 ~ 18.6 19.6 ~ 21 21.0
Japan 24.3 ~ 30 14.6 ~ 18.8 12. 9-14. 6
Weat Germany 18.5 a1 ~ 27 20. 6~21. 6
Brasil oe re es
Indie 10.27 195 fo -e--- J ween
In Average 10.~ 20 16. ~ 20 13 ~ 16
Table 1. Characteristics of Basic Research over R&D
‘ in 7. countries
SYSTEM DYNAMICS '93 595,
30.00 4
Japan —> West Germany
BR/R&D (m)
8
8
France
India
ea
= ULS.AL
10.00 4
Brazil
0.00 1
60.00 70.00 80.00 90.00
Year
Figure 2 Input Behavior of Basic Research in 7 Countries
According to Figure 2, there exist four patterns of
Basic Research over R&D ( S-Curve Pattern, S’-Curve Pattern
(S-Curve with saddle-form), Constant Growth Pattern and Low
Growth Pattern). The relationship between investment
patterns on Basic Research and the perfromances expressed
by above-mentioned index:.is summarized as Table 2.
S-Curve S’ -Curve | Constant Growth Low Growth
Pattern | Pattern Pattern Pattern
R&D
personnel + +t ++ -
gnp Z + + ++ -
high-tech. - ++ + it
patente ++ + + +
Table 2 The Relationship Between Basic Research Input and
Performance
(Notes: ++:excellence; +:good; -:ordanary)
Table 2 shows that S’-Curve Pattern (which is mainly adopted
by Japan ) and the Constant Growth Pattern (which is mainly
adopted by France and U.S.A.) have made better performance.
In order to make further analysis, correlation analysis
596 SYSTEM DYNAMICS '93
based on the Gray System Theory(GTS) is made, Table 3 shows
the caculated results. And Table 3 also tells that there
exists higher correlation between Basic Research and the
economic & social performance in France, U.S.A. and Japan.
BaD High-tech. Patent
GNP (GDP) | Personnel Trade
Pattern
Japan 0, 398 0. 609 | 0. 672 0. 426
UL 8. A. 0. 418 0. 639 0. 788 0. 844
West Germany 0. 429 0. 662 0. 641 0. 728
France 0. 717 0. 717 0. 809 0. 941
U.K. 0. 397 0. 463 0. 666 0. 146
Todle 0. 555 0. 647 0. 646 0. 664
Brazil 0. 588 0. 661 0.610 0. 674
Table 3 Correlation Analysis Results
MECHANISM ANALYSIS: WITH SYSTEMS THINKING
In our systems thinking, the input .of Basic Research must be
coordinated with .the development of economy, education,
finance as well as modification the internal structure of
science & technology per se. So there exists a large-scale
system with hirarchy structure (See Figure 3). According to
Figure. 3, there are 5 subsystems: economy subsystem,
population subsystem, education subsystem, finance subsystem
as well as science--technology subsystem. A detailed causal-
effect loops about science & technology with simplified
loops about the other 4 subsystems is shown as Figure 4. The
coming model simulation will seek the resource allocation
policy on Basic Research subjected to the requirement of
these two coordination.
The Basic Run of this modle is.shown as Figure 5. According
the Basic Run, the input on Basic Research increases to
around 18% (over total R&D) at 2016 in the form of
exponential form and decreases to 13% (over total R&D) in
the form of inverted s-curve), as a result, GNP and total
science & technology out increase with steady and
sustainable pace. Though the ratio between technology _
SYSTEM DYNAMICS '93 597
coordination
(higher level)
economy finance
sclence &
t
nology population
coordination
(lower Level)
basia research
development experiment.
Figure 3. Model Structure
LEVEL OF ECONOMIC
DEVELOPMENT
GNP
GNP
PER CAPITA.
™
BIRTHS. DEATHS
POPULATION
DEMAND
INPUT. ON INPUT ON
——
DEMAND FOR AR.
INPUT ON
EDUCATION
ra
INPUT ON
S&T
INPUT ON ED
INPUT ON AR
FOR ED
INPUT ed
FIXED CAPITAL
ALLOCATION 4 ‘\
meni LABOR PRODUCTIVITY
INPUT ON AR RESERVE
INPUT ON BR
/ourPUT OF AR OUTPUT OF iN
INTERNATIONAL ECONOMY
CIRCULATION
INFRASTRUCTURE,
SYSTEM
a, SN BR PERIOD: OF BR
ON
“lesion ebteton (7° 2 ff LA. BR RESULTS [/
EDUCATION Ee
OUTPUT we HIGH wae
, OF BR :
ot IGN BR
“Sar rach DEMAND FOR
BR. PERSON = RATE
ae 4 PI
BS oni oes
Va OF a FREE
Figure 4. Causal Relationship of The Total System
senor 7 BR
atar OF
598
SYSTEM DYNAMICS '93
acquisition expenditure over in-house R&D expenditure. is
high in. recent 10-15 years, the rely on the foreign
technology » will decrease in long range. operation... More and
more talent personnel (above master degree) could be trained
during the process of Basic Research..
15,04 acihca wacaidian tee “|
$0, input deer RED input] »4
7 7
t ° .
Teh ee ae
/ a
g neo" Pa
Toakur of =
Sauer tune People Tee | Ln
ben“) Le
wet bee TT eel toed thology Oat
19661976. 1986. 1996. a 2016. 2026. 2036. 2050.
Figure 5. Basic Run of Resource Allocation Model
POLICY TEST & ANALYSIS
Started at the time of 1992 when the ratio between Basic
Research over R&D was 7.2%, resource allocation patterns of
7 countries are simulated as Table 4 showed, Here four index
are used to evaluate the performance of development of China
which affected by the effort of Basic Research directly or
indirectly. These index are GNP (Unit: million US$), ratio
of technology acquisition expenditure over in-house R&D
expenditure (abbreviated as TA/RD), number of talent persons
(above master degree, unit: 10 thousand, abbreviated as
PSHEG), and the number of total science & technology output
(Unit: 10 thousand item, abbreviated as TTO). TTO is also
the index of the coordination among internal science &
technology system.
By the comprehensive analysis we could see that no country’s
pattern is so good to follow about the Basic Research input.
But the synergic pattern ( synergy of- France & Japan
Pattern, same as Basic Run) is the best pattern for its
SYSTEM DYNAMICS '93 599
better performance (the simulated results shows that the
annual growth rate of GNP could be 5.3% from 2000 to 2050,
and the corresponding growth rate for talent’ people and
total technology output are 3.7% and 7.7%,. and the ratio
between technology acquisition expenditure and total R&D
expenditure decreased from 0.28 to 0.15).
Basie Research Year
x Index
Input Pattern 2000 2030 3060
GNP 9480 82570 140900"
TA/RD » 281 . 319 . 200
U.S. Pattero PSHEG 8. 28 3.24 -9, 04
TTO 636.8 6681.7 16270*
GNP 9836 67440 $6510
Japan Pattern TA/RD 23 19 20
PSHEG 9. 87 9.26 6.40
TTO 366.1 1664.1 6887.1
GNP 9810 61600 144170
TA/BD -28 aes 18
France Pattern | PSHEG 9.26 12.0 39.8
TTO 336. 0 66a1 16200
GNP 93816 68000 60670
West gormany TA/RD 437 dt aL
Pattern PSHEG 9.29 86.1 1. 84
TTO 844.2 2886.6 9445.2
GNP 9279 40520 54480
TA/RD - 28 48 +88
U.K.Pattern PSHEG 9.18 19.1 2. 06
TTO 803.6 1196 8760
GNP 9810 46770 66760
TA/RD +26 3h AT
Indian pattern | PSHEG 9. 26 62. 12 -8. 44
TTO 334 1828 B11LS
GNP 9277 34700 39240
TA/RD = 38 86 36
Brasil Pattern PSHEG 9.18 BT. 68 4.58
TTO 804 633 1906
GNP gaa 6eaz0 129970"
TA/RD re) +17 .16 *
Synergy Pattern | PSHEG 9. 25 38. 79 67, 44%
TTO 362 6824 14500*
Table 4. The Results of Policy Tests on The Basic Research
Input
CONCLUSION & DISCUSSION
The conclusions by our research are:
600 SYSTEM DYNAMICS '93
1. The very first issue which arises in the course of
development of China is the dynamic choice of the resource
allocation pattern for Basic Research by the economy or
industrialization stage. Our comparative study and model
simulation shows during China’s development Stage II as we
defined previoursly, Resource allocation on Basic Research
might .as well. increse with expenontial growth form and
during Stage III, the iuput on Basic Research remains
steadily (a little bit decreasement).
2. An essential procondition for sustainable development of
China demands a coordination between science & technology,
education, economy, population and finance as well as the
coordination between science & technology. Priority must be
given to the first coordination. Though the simulation the
U.S.A. Pattern for the Basic Research input proved the best
choice of China’s coordination among internal science &
technology system (total technology output is around 162.7
million items), the France-Japan Synergy Pattern is advised
as the pattern for Basic Research input for its better
comprehensive performance.
3. Basic Research, the main source of productive activity,
by the very nature require a long period of time over which
benefit can be gained. According to the results of our
simulation, the performances resulted in different resource
allocation policy on Basic Research remain the same level in
recent 7-10 years (refer to Table 4). So the input on Basic
Research must be taken in a strategic view.
4. Concerning the problems of social, science & technology
system, system modeling and simulation must be based on
clear understanding of the potentials a country processes
and the contrains to which it is subjected. Some method,
such as comparative study is advised to be used to draw some
qualitative conclusion. The using of comparative study is
proved to enhance the effectiveness of policy test and
simulation.
5. During the process of system modeling on a large-scale
social & economic system, each subsystem should not be taken
as equal one, a hirarchy structure is advised to set up to
catch up the main subsystem, and coordinations at different
level required by the users. More mathemtical anslysis is
needed, and this is the very work of System Dynamics in 21th
century.
REFERENCES
Keith Pavitt, 1991. What Makes Basic Research Economically
Useful. Research Policy 20:109-119.
Xu Qingrui, Li Junjie, Jiang Shaozhong and Jiang Jiong,
1988. Science-Technology, Education and Economy System
Dynamic Model. Proceeding of ICSSE’88. Beijing:
SYSTEM DYNAMICS '93 601
International Academic Publishes.
Xu Qingrui, 1986. R&D Management. Beijing: High Education
Publishers,
Cheng: Zhengji, et al, 1992. ‘On’the Input of Science &
Technology. Beijing: Science & Technology Literature
Publishing House.
602 SYSTEM DYNAMICS '93
