Transcript Do examiners* workloads raise grant and invalidation of patent

Technological & Scientific Knowledge and
Economic Growth:
Implication for S&T Policy
of East Asia and Latin America
Yee Kyoung Kim
Keun Lee
Associate Research Fellow
KISTEP
Professor
Seoul National Univ.
Broad context of the Research:
1) Various forms of Knowledge on Growth
2) E Asia vs. Latin America
1) Patents = tech. knowledge
2) Articles = scientific knowledge
3) Trade marks, designs
4) Petit Patents (utility models)
5) Knowledge embodied in machinery
=> How E Asia and L America different?
Trend of the Income Levels as % of Japan:
Middle income Trap?
3
:
patents -> growth in middle or higher countries
Lee, Keun & B. Kim (2009, World Development)
4

Confirms importance of Innovation and high education
for middle and higher income countries;
cf) Institution and basic human capital matter
for low and lower middle C’s

Next, beyond just patent counts (innovation measure)
=> more details of the NIS (national innovation systems)
by other variables:
Eg) cycle time of tech.
localization of knowledge creation,
Innovation
system at
3levels:
firm, sector,
& country
Four Key Variables and Hypotheses
at three Levels of country, sector & Firms
4 Hypotheses : Growth strategies
short vs. long cycle-time knowledge
high vs. low originality knowledge
Localization of knowledge creation & diffusion
(vs. reliance on foreign sources)
Balanced vs. Concentration of knowledge creation
6
Originality
0.35
0.3
0.25
0.2
0.15
0.1
1975
1980
High Income countries
Korea and Taiwan
7
1985
1990
Middle Income countries
Brazil and Argentina
1995
Intra-national Citation in Patents (national self-citation)
-> indigenous corporations cf) MNCs (R&D at homes)
Localization of Knowledge creation & diffusion
0.12
0.1
0.08
0.06
0.04
0.02
0
1975
8
1980
1985
1990
High Income countries
Middle Income countries
Korea and Taiwan
Brazil and Argentina
1995
Short cycles: Less reliance on existing knowledge held by incumbent c’s
Cycle Time of Technologies
12
11
10
9
8
7
6
1975
9
1980
1985
1990
High Income countries
Middle Income countries
Korea and Taiwan
Brazil and Argentina
1995
Now, comparing Tech. vs. Scientific
knowledge for growth
1) Which one more important ? And Why?
2) If patent important, how to generate them?
2a) By encouraging scientific research which hopefully lead to tech.
knowledge (linear model)?
or by directly encouraging corporate R&D activity?
3) Scientific knowledge important? Can they be commercialized so
that they may be used in private sector?
Or are they just a symbol of academic achievement?
Our conclusion: not articles but patents matter?
Or right sequence: patents -> articles
cf) Argentina 4 Nobel prizes vs Korea 0 prizes
-> However, many latecomer countries emphasize articles
Before 90s: Higher article intensity in L America > E Asia
1990
Corporate patent intensity in E Asia vs L America
Early 1980s
Asia caught up with LA first in Patents and then later articles
Science & Tech. Policy in East Asia and Latin America
East Asia




the follower strategy by
importing foreign technology
But promoted indigenous firms
and their tech. capability
They increased in-house R&D
which lead to patents
Later; they also demanded
scientific knowledge
Latin America





same follower strategy
Relied on MNCS without
domestic technological
development
R&D activity conducted in
their home countries
National policy to support the
generation of scientific
knowledge
no sustained demand from
private sector
[R&D ratio to GDP in East Asia and Latin America
Main hypothesis
Economic
Economic
Growth
Growth
Patents
R&D
Papers
III. Empirical Framework and Evidence


DATA
 a panel data of countries from 1960-2005 (5-year average)
 two main groups of the countries: the countries at the later
development stage and the countries at the early development
stage
 GDP per capita in constant 2000 US dollars is used to classify
the level of economic development (10,000 dollars)
Cross- country Analysis
 Economic Growth Equation
 Technology Production Equation
 the impact of scientific knowledge on technological innovation
III. Empirical Framework and Evidence (2)


Growth rates ;higher in high income ; same in investment rates, secondary
enrollment rate, and per capita GDP
Also, corporate patents intensity and SCI journal articles intensity; higher in high i.
countries.
III. Empirical Framework and Evidence (3)
1. Economic Growth Equation
First Results: patents vs articles separately on growth
VARIABLES (lags)
GDP per capita
(1)
OLS
-0.0109***
(2)
FE
-0.0374***
(3)
GMM
-0.0158
(13)
OLS
-0.0048
(14)
FE
-0.0539***
(15)
GMM
0.00923
population growth
(-2.832)
-0.00305
(-2.674)
-0.000484
(-1.542)
-0.00101
(-1.485)
-0.00238
(-3.597)
0.00565
(1.074)
-0.00447
investment)t
(-1.360)
0.0627***
(-0.119)
0.0649***
(-0.155)
0.132***
(-0.879)
0.0351***
(1.109)
0.0429***
(-0.894)
0.0447*
secondary ed. enrollment
(8.892)
-0.000324
(-0.0778)
(7.312)
-0.000382
(-0.0820)
(6.787)
-0.0114
(-0.810)
(4.806)
0.00147
(0.311)
(3.175)
1.16E-05
(0.001)
(1.832)
-0.0230*
(-1.722)
corporate patents intensity
0.00383**
0.00740***
0.00600*
(2.480)
(2.697)
(1.698)
0.00379**
0.00418
0.000375
(2.028)
-0.0894***
(-3.371)
591
0.091
(0.987)
0.315**
(2.324)
591
0.116
(0.063)
-0.109
(-1.312)
591
article intensity
Constant
Observations
R-squared
Hansen
AR2
-0.0831***
(-2.659)
449
0.237
0.126
(1.049)
449
0.244
-0.197**
(-2.556)
449
0.413
0.661
0.199
0.481
Full Results: patents vs articles together on growth
Variables in logs
GDP per capita)
population growth
investment
Second. enroll
Corp. patents inten.
(1)
OLS
-0.0136**
(-2.573)
-0.00399
(-1.439)
0.0643***
(6.579)
0.00339
(0.422)
0.00456**
(2.344)
(2)
FE
-0.0626**
(-2.491)
0.00281
(0.725)
0.0622***
(2.918)
-0.00379
(-0.407)
0.0140**
(2.594)
patents intensity
article intensity
Constant; Time D
Observations
R-squared
Hansen
AR2
0.00137
(0.553)
YES
293
0.226
0.00353
(0.410)
YES
293
0.227
(3)
(4)
GMM
OLS
-0.0252* -0.0137**
(-1.721)
(-2.586)
0.00218
-0.00379
(0.360)
(-1.371)
0.0581*** 0.0642***
(2.793)
(6.690)
0.014
-0.00149
(1.282)
(-0.173)
0.00773*
(1.781)
0.00489**
(2.419)
0.00757
0.0017
(0.673)
(0.695)
YES
YES
293
301
0.21
0.847
0.232
(5)
FE
-0.0460*
(-1.928)
0.00388
(0.975)
0.0659***
(3.133)
-0.00744
(-0.643)
(6)
GMM
-0.0112
(-0.871)
5.91E-05
(0.011)
0.0732***
(3.991)
-0.0063
(-0.401)
0.0123**
(2.556)
-0.00242
(-0.267)
YES
301
0.196
0.00359
(0.927)
0.00675
(0.719)
YES
301
0.453
0.514
III. Empirical Framework and Evidence (5)
Empirical Results of Economic Equation



Technology and science knowledge on Growth
GMM model to control endogeneity
Corporate patent intensity vs. SCI journal article intensity
(corporate patent intensity) : +* in all OLS, fixed & GMM
(SCI journal article intensity) : + in many
Knowledge Production Equation:

To test linear model of innovation (Velho, 2004, Cimoli, 2000) modifying
Griliches’s knowledge production model
Results of Technological Knowledge Production Equation
Results of Scientific Knowledge Production Equation
Summary: Knowledge Production Equation
1) On Tech Knowledge (patents)


Increased scientific knowledge has no significant impact in
either High or middle income countries
Instead, R&D leads to patents
2) On Scientific Knowledge (articles)
 Increased patenting -> articles not in high but in middle Inc
countries;
 R&D has no impacts on articles
Verified in conclusion
Economic
Economic
Growth
Growth
Patents
R&D
Papers
IV. Implications





We investigated why Latin American and East Asian countries have
shown such diverging paths in economic growth, with focus on the
role of scientific and technological knowledge
Linear model of innovation emphasizing cientific knowledge as an
input for innovation is not supported. => mistakes in LA
Better: technology policy first to invigorate the industrial sector,
which lead R&D and then demand for scientific knowledge => EA
Without the anchor of private sector, academia tend to remain as
an ivory tower; Isolated
Divergence bt. East Asia and Latin America: explained by the
different priority on tech. and scientific knowledge and their
sequence (patents -> articles)
28
Thank you!
謝謝大家
ありがとう!
Gracias!
Meu Amigo! Obrigado!
감사합니다