Transcript CERRANDO LA BRECHA EN EDUCACION Y TECNOLOGIA

IS INNOVATION IMPORTANT TO
PRODUCTIVITY GROWTH?
William F. Maloney
Office of the Chief Economist
Latin America and Caribbean Region
World Bank
ECA Productivity Workshop, October, ‘05
Half of growth is Total Factor Productivity
Annual Growth of TFP
3
2. 5
2
1. 5
1
0. 5
0
-0. 5
Chile
Mexico
Latin America
OECD
East Asia
-1
-1. 5
1970s
-2
Source: Calderon, Fajnzylber y Loayza (2002)
1980s
1990s
How do we know if a country has an
innovation problem?



Financial markets, Infra., SME
NZ- dual strategy of business
mentoring etc. and innovation
W/o.. Pushing on a string
Figure 1: R&D investment rates versus capital-output
ratios
5
4
R&D/GDP
Comparative Innovation
statistics?
Is (knowledge accumulation)
TFP below what would be
expected given the level of
accumulation of other factors?
In LAC, Yes
Lesson 1- Innovation cannot be
taken independently of general
accumulation issues-PSD
3
2
1
0
0
1
2
Composite capital-output ratio
Rodriguez and Maloney(2005)
3
How do we know innovation has
anything to do with it?
Levels of inefficiency required to generate
TFP shortfall in Chile too high
History “Missed historical opportunities”
Everybody else thinks so
What type of innovation?
Deviations from expected level
2
0.10
2
Finland
0.10
Korea
1
0.05
1
0.05
0
0.00
0
0.00
-1
-0.05
-1
-0.05
-2
-0.10
-2
-0.10
1972
1977
1982
FDI
Source: Maloney (2004)
1987
R&D
1992
1998
1972
Royalties
1977
1982
1987
1992
Growth per capita (rigth axis)
1998
“Latin” model: The opposite: low R&D,
High FDI,moderate licensing
2
0.10
2
Chile
0.10
Mexico
1
0.05
1
0.05
0
0.00
0
0.00
-1
-0.05
-1
-0.05
-0.10
-2
-2
1972
1977
1982
1987
1992
1997
2
-0.10
1972
0.10
1977
1982
1987
1992
1997
2
0.10
Spain
Italy
1
0.05
1
0.05
0
0.00
0
0.00
-1
-0.05
-1
-0.05
-0.10
-2
-2
1972
1977
FDI
Source: Maloney (2004)
1982
R&D
1987
1992
Royalties
1997
-0.10
1972
1977
1982
Growth per capita (right axis)
1987
1992
1997
Lesson 2: Lots of room for non-sexy types of innovation




Tech Institutes in Spain- focus on diffusion
Italy- highest TFP growth in post war-all SMEs
We need to simulate all the effects of becoming part of EU.
HOWEVER…
Lesson 3: Eventually, big advances in TFP will need
inventive capacity




Italy now stagnant- hit the invention wall
Spain- efforts to upgrade NIS- including TC
But these take place with a 20 year lag so must start now
LAC is lagging in effort, and in quality of effort
Mexican high tech is not so high tech.
Comparative Advantage in Innovation
Brazil and Mexico: IRCA in Aircraft and Computing Equipment
1.4
Brazil Aircrafts
Mexico Office Computing & Acctg.
Brazil: Airplanes
3.5!!!
RCA Index in SIC 372 (18-yr MA)
1.2
1
0.8
Taiwan: Computers
0.6
0.4
Mexico: Computers
0.2
0
1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000
Year
Textiles and Leather Products
.02
.025
Food, Beverages & Tobacco
.005
.015
Spain
Denmark
United Kingdom
0
Canada
Spain
.005
Canada
Netherlands
France
United States
Australia
.005
.01
Belgium
Australia
United States
United Kingdom
France
.01
.015 .02
1985
.025
.03
0
Finland
1995
Belgium
.005
.02
Belgium
Netherlands
United
States
Finland
France
DenmarkUnited Kingdom
.05
.01
Japan
Norway
.015
Japan
.1
.015
Australia
.01
1985
Chemical excluding Pharmaceuticals
Sweden
United States
.005
.15
.02
Wood, Cork, Pulp ,Paper & Publishing
1995
Japan
Sweden
Sweden
1995
1995
.015
Norway
Denmark
.01
.02
Netherlands
Finland
Japan
Norway
Finland
Belgium
Sweden
Canada
France
Spain
Netherlands
Denmark
United Kingdom
0
0
Spain
Canada
0
.005
.01
1985
.015
.02
0
.06
.09 .12
1985
.15
.18
.2
Rubber & Plasitc Products
.5
Pharmaceuticals
.03
Japan
.1
1995
Denmark
Finland
France
Netherlands
Belgium
Canada
United States
Japan
Norway
.05
Spain
.1
Finland
Belgium
France
Sweden
United
NorwayStates
Denmark
Netherlands
Australia
Spain
United
CanadaKingdom
0
.3
.2
.1
1995
.4
Sweden
.15
United Kingdom
.2
.3
1985
.4
.5
0
.05
.1
1985
.15
.2
There are a range of
R&D intensities to
produce any good
OECD- in most
countries increased
intensity as/more
important than shift to
high tech products
Rates of Return to R&D are lower
for LDCs. Why?
Return
to R&D
1.6
basic
initial
1.4
time
output
gap
dummies
1.2
1
0.8
0.6
0.4
0.2
GDP per
10000
Lederman, Maloney and Siga (2005)
20000
30000
40000
Capita
Not just how much, but how: efficiency of
converting R & D into patents,TFP
20.00%
Patents = B1I&D + Bp country*R&D
15.00%
10.00%
5.00%
-5.00%
-10.00%
Source: Bosch, Lederman, Maloney (2004)
NOR
FIN
ISR
KOR
TWN
VEN
URY
PER
MEX
CRI
COL
CHL
BRA
ARG
0.00%
A Simple NIS
Global Knowledge Economy
Human Capital
Think Tanks/
Antenna
University
HD
Firms
Innovation & TFP Growth
PSD:
Entrepreneurship mentoring
Infrastructure (ICT)
Credit/labor markets
Clustering
Subsidies/Tax incentives
Global Knowledge Economy
End