Transcript Towards a Comprehensive Accounting of the Knowledge Economy

Innovation, Intangibles and Economic Growth:
Towards a Comprehensive Accounting
of the Knowledge Economy
Bart van Ark*
University of Groningen
and The Conference Board
Seminar on Measurement of Capital –
Beyond the Traditional Measures
Conference of European Statisticians
12 June 2007, Geneva
* In co-operation with Charles Hulten, University of Maryland
and The Conference Board
1
Grab your iPod, flip it over, and read
the script at the bottom. It says:
"Designed by Apple in California.
Assembled in China." Where the
gizmo is made is immaterial to its
popularity. It is great design, technical
innovation, and savvy marketing that
have helped Apple Computer sell
more than 40 million iPods. Yet the
folks at the BEA don't count what
Apple spends on R&D and brand
development, which totaled at least
$800 million in 2005. Rather, they
count each iPod twice: when it arrives
from China, and when it sells. That, in
effect, reduces Apple -- one of the
world's greatest innovators -- to a
reseller of imported goods.
2
Do the Official Statistics Capture the
Knowledge Economy?
 How do we measure innovation?


New ways of making current products better, faster,
cheaper
Creating something not previously created
 Is the impact of innovations accurately represented
in official statistics on national income and
productivity? NO !!




“You see the computers everywhere except in the
productivity data.” Solow (1987)
Official statistics “miss the most important technological
revolutions in history.” Nordhaus (1997)
Hence: “Whilst the knowledge economy is
all around us, it is still hard to see it in the official statistics”
Skepticism about GDP, CPI and productivity statistics
capturing key components of the knowledge economy3
National Accounts Framework Provides
Adequate Point of Departure
 Basic solutions require rethinking of:





The scope of national accounts on output side
The measurement of output in particular in services
The measurement of prices to capture product quality
Consistency of output and input measures to provide
productivity
The treatment of inputs in terms of expenditure versus
investment
 Development of growth accounts in conjunction
to national accounts is key


KLEMS framework
Measurement of intangibles
4
Basic Determinants of Sources of Growth/
Growth Accounts Model
 Output is key measure of standard of living
 Output is driven by




capital (K)
labor (L)
intermediate inputs (E, M, S)
productivity (LP, MFP)
 Capital and labor can by divided into


quantity
quality/composition
 Output increases that cannot be explained by
these “inputs” are attributed to multifactor
productivity (MFP)
5
EU KLEMS Growth and Productivity
Accounts Complements National Accounts
 EU KLEMS is analytical research database,
based on national accounts and complementary
official sources (LFS and production statistics)
 Long time coverage 1970-2004, with greatest
detail for post-1995
 Harmonized industry classification, capital and
labour input, deflation and industry aggregations
(e.g. market economy, market services)
 Decomposition of capital and labour input:


Capital assets in 7 asset types
Labour input in 18 categories (3 x skill; 3 x age; gender)
 Broad coverage of EU countries and
comparisons with U.S. and Japan
 Public database: www.euklems.net
6
Growth in Quantity of Standard Factor Inputs Accounts for
only Small Part of Real GDP Growth in Market Economy
4.0
1980-1995
1995-2004
3.0
2.0
1.0
0.0
EU15ex excludes Portugal, Luxembourg,
Ireland, Sweden and Greece
-1.0
EU15ex
USA-SIC
Japan
EU15ex
Hours worked
USA-SIC
Japan
Growth in Quantity of Standard Factor Inputs Accounts for
only Small Part of Real GDP Growth in Market Economy
4.0
1980-1995
1995-2004
3.0
2.0
1.0
0.0
-1.0
EU15ex
USA-SIC
Japan
Hours worked
EU15ex
USA-SIC
Non-ICT Capital
Japan
Even in the Traditional Growth Accounts System the
Knowledge Economy Features Strongly but Unequal
4.0
1980-1995
1995-2004
3.0
2.0
1.0
0.0
-1.0
EU15ex
USA-SIC
Japan
Hours worked
EU15ex
Non-ICT Capital
USA-SIC
Japan
ICT Capital
Even in the Traditional Growth Accounts System the
Knowledge Economy Features Strongly but Unequal
4.0
1980-1995
1995-2004
3.0
2.0
1.0
0.0
-1.0
EU15ex
USA-SIC
Japan
EU15ex
USA-SIC
Japan
Hours worked
Non-ICT Capital
ICT Capital
Labour Composition
Even in the Traditional Growth Accounts System the
Knowledge Economy Features Strongly but Unequal
4.0
1980-1995
1995-2004
3.0
The
knowledge
economy
2.0
1.0
0.0
-1.0
EU15ex
USA-SIC
Japan
Hours worked
ICT Capital
MFP
EU15ex
USA-SIC
Japan
Non-ICT Capital
Labour Composition
Fast Growing Countries Get Bigger Bang from
Knowledge Economy, Notably MFP
5.0
4.0
3.0
2.0
1.0
0.0
-1.0
GER
ITA
LUX
DNK
BEL
Hours worked
ICT Capital
MFP
FRA
AUT
NLD
SWE
UK
ESP
USA
Non-ICT Capital
Labour Composition
FIN
Unmeasured intangible capital is hidden in MFP
a) Physical Capital
a1) ICT capital (IT hardware, communications equipment)
a2) Other capital (plant, machinery, buildings)
b) Human Capital
b1) Formal Education
b2) Company training
c)
Knowledge Capital
c1) Research and Development
c2) Patents
Asbrands,
long as
intangibles
c3) Licenses,
copyrights
c3) Other technological
innovations, not related to b1) to b3)
capital remain
[c4) Software]*
unmeasured, its
c5) Mineral Exploration
productivity effects
c6) Experience
are hidden in MFP
d) Process Capital
d1) Engineering design
d2) Organisation design
d3) Construction and use of data bases
d4) Remuneration of innovative ideas
e)
Customer Capital
e1) Marketing of new products
f)
Multi Factor Productivity (residual)
Corrado, Hulten and Sichel (CHS, 2005)
calculated intangible investment in U.S.
 Inherent measurement difficulties of intangible
capital going beyond those of tangible capital as
follows:
The knowledge-input problem
 The knowledge-investment problem
 The quality improvement problem
 The obsolescence problem
(Howell, 1996)

 But no clearcut distinction between tangibles and
intangibles that justify a distinction between
capitalizing and expensing
 “Any outlay than is intended to increase future
rather than current consumption is treated as a
capital investment”
14
Source: Corrado, Hulten and Sichel (2005)
CHS (2005) measure $1.2 trillion of
intangible investment in U.S. nonfarm
business, 1998-2000
 COMPUTERIZED INFORMATION ($154,$154)
 COMPUTER SOFT WARE ($151)
 COMPUTERIZED DATABASES ($3)
 SCIENTIFIC AND CREATIVE PROPERTY ($424,$424)
 SCIENTIFIC R&D ($184)
 MINERAL EXPLORATION ($18)
 COPYRIGHT AND LICENCE COSTS ($75)
 OTHER PRODUCT DEVELOPMENT (FINANCE,
ARCHIT.) ($149)
 ECONOMIC COMPETENCIES ($642,$505)
 BRAND EQUITY (ADVERTISING) ($236)
 FIRM-SPECIFIC HUMAN CAPITAL (TRAINING) ($116)
 ORGANIZATIONAL STRUCTURE MANANGEMENT
CONSULTING, PLANNING ETC.) ($291)
16
Source: Corrado, Hulten and Sichel (2005)
When including intangibles, investment
share in U.S. GDP holds up
17
Source: Corrado, Hulten and Sichel (2005)
… but the difference in growth rates arises
from firm specific resources, not R&D
18
Source: Corrado, Hulten and Sichel (2005)
Results for U.S. Economy, 2000-03
(70.4%)
(60%)
(29.6%)
(25%)
(15%)
19
Source: Corrado, Hulten and Sichel (2005)
When including intangibles in growth
accounts the weight on capital raises
20
Source: Corrado, Hulten and Sichel (2006)
SOURCES OF GROWTH IN OUTPUT PER HOUR
NFB 1995-2003
3.09%
2.78%
25%
11%
14%
50%
INTANG
27%
IT CAP
19%
L QUAL
8%
11%
MFP
35%
IT CAP
L QUAL
MFP
WITHOUT
INTANGIBLES
WITH
INTANGIBLES
21
Source: Corrado, Hulten and Sichel (2006)
SOURCES OF GROWTH IN OUTPUT PER HOUR
NON-FARM BUSINESS SECTOR
3.09%
1.63%
26%
INTANG
18%
15%
15%
25%
IT CAP
L QUAL
MFP
MFP
1973-1995
INTANG
27%
IT CAP
19%
L QUAL
8%
11%
MFP
1995-2003
35%
22
Source: Corrado, Hulten and Sichel (2006)
CONTRIBUTION OF DIFFERENT INTANGIBLES
TO ANNUAL CHANGE IN LABOR PRODUCTIVITY
0.84%
SOFTWARE 32%
0.43%
28% SOFTWARE
12% SCI R&D
19% N SCI R&D
9%
L QUAL
BRAND
MFP
30% FIRM
MFP
SPEC
1973-1995
SCI R&D
10%
N SCI R&D
17%
BRAND
10%
FIRM SPEC 32%
1995-2003
23
Source: Corrado, Hulten and Sichel (2006)
The Research Program on Intangibles
 Extend studies to more countries



UK: Haskell and Marrano
Japan: Fukao et al.
Ongoing work for Finland, France and Netherlands
 Price and output statistics that explicitly recognize
product innovations (“quality” change).
 Uncover the subtleties of interaction between
tangibles/intangibles and innovation/productivity
growth
 A stronger link in the data between human capital
and worker competencies
 More detailed study at industry level (mnf/services)
 Intangibles need more accurately represented in
the financial data of the innovators themselves
24
The Role of National Statistical Institutes
 Intangible accounting is in early days, but
statistical description of knowledge economy is too
far removed from reality to ignore
 Official statistics need to be as precise as possible
 … but it is sometimes better to be “imprecisely
right than precisely wrong” (Keynes)
 … requiring a balance between research and
official statistics
 NSI’s could help to:




develop concepts of intangibles
set (international) standards for measurement
provide value metrics of intangibles, e.g., survey metrics
on innovation with quantitative magnitudes.
Transfer experimental and research based measures into
national accounts satellites that can help to move the
25
measurement agenda forward