Third Workshop on Information Society Measurement in Latin
America and the Caribbean, Panama, 23 November 2006
THE IMPORTANCE OF MEASURING
BUSINESS USE OF ICT
Martin Schaaper
Economic Analysis and Statistics Division,
Directorate for Science, Technology and Industry
1 1
Why measuring business use of ICT?
Q
Assess take up and diffusion
– … and compare with other countries
Q
Q
How does the Internet change the way businesses
run their operations
Impacts on overall economic performance
(productivity, growth, etc.)
– … and compare with other countries
Policymakers need this information
to make sound decisions
2 2
The state of ICT diffusion
3 3
ICT diffusion among OECDand non-OECD countries
Q
Q
Q
Diffusion can be measured by activity and financial
statistics.
OECD has been collecting activity statistics on ICT
use and e-commerce from member countries for
several years and first published data in 2001
(“OECD Science, Technology and Industry
Scoreboard: Towards a Knowledge-based
economy”).
The most recent version of this publication has been
the 2005 edition.
4 4
%
Have Internet access
(1) All size classes
(2) 16+ employees
Ita
G ly
re
e
N ce
or
C
w
hi
ne P ay
H se o
on
la
g Tai nd
Ko pe
ng i (
(C 1)
hi
na
H
Si
un )
ng
ap P gar
or or y
e t
Sl (2 uga
ov 00 l
ak 3)
R (1)
e
Th pu
ai bli
la
c
nd
C (2)
hi
le
(1
)
(2
D 003
en )
m
a
Fi rk
nl
a
Sw nd
e
G den
er
m
an
Ko y
r
C ea
Au ana
d
s
M tra a
ex lia
ico (4
N (2 )
et 00
he 3
rla )
nd
s
Ja
pa
n
Business use of the Internet, 2004, Percentage
of businesses with 10 or more employees
Have own Web site
Broadband (2)
100
80
60
40
20
0
5 5
Internet selling and purchasing, 2004,
Percentage of businesses with 10 or more
employees
Purchasing
Selling
Thailand (2)
Chile (1)
Mexico (2003)
Chinese Taipei (1)
Korea (3)
Japan (2003)
Singapore (2003) (1)
Australia
Sw itzerland (2002)
Germany
United Kingdom
Hong Kong, China (1)
Canada
(1) All size classes
60
50
40
30
20
10
0
- 10
- 20
(2) 16+ employees
(3) Data refer to enterprises with 1 or more employees.
- 30
- 40
6 6
Total e-commerce transaction value (including
via the Internet), 2002 to 2004.
As a percentage of total enterprise turnover
%
25
2002
2003
2004
20
15
10
5
G
re
ec
e
Sp
ai
n
Po
la
nd
Ita
ly
Lu
xe
m
bo
ur
g
Au
Cz
st
ria
ec
h
Re
pu
bli
c
Po
rtu
Sl
ga
ov
ak
l
Re
pu
bl
ic
Ire
Un
lan
ite
d
d
Ki
ng
do
m
De
nm
ar
k
G
er
m
an
y
Fi
nl
an
d
No
rw
ay
Be
lg
iu
m
0
7 7
Business perceptions
of the benefits of
buying and selling
over the Internet
8 8
Business perceptions of the benefits of
buying and selling over the Internet
Q
Q
Australian Bureau of Statistics data show that
businesses which buy and sell over the Internet
perceive benefits from doing so.
The longer a business has been buying or selling
over the Internet, the more likely it is to report
benefits
– first mover advantage?
– survival of the fittest (those which are still selling are those
which have succeeded and therefore are more likely to see
benefits)?
– or just getting better at it over time?
9 9
Not applicable
No change
Increased
Bu
lity
of
cu
st
om
er
se
rv
sin
ic
Sa
e
es
le
s
s
co
or
st
cu
s
st
om
er
s
E
C
om ff ici
en
pe
cy
tit
iv
en
Tr
es
ad
s
in
g
ho
Tr
ur
ad
s
in
g
ar
ea
Decreased
Q
ua
% of businesses selling over the Internet
AUSTRALIA (ABS): BUSINESS PERCEPTIONS OF THE IMPACT
OF RECEIVING ORDERS (SELLING) VIA THE INTERNET, 2000–01
10 10
OECD case study
work on the impacts
of e-business
11 11
OECD case study work: Electronic Commerce
Business Impact Project (EBIP)
Q
Q
Q
Q
Q
During 2001-2002, the OECD co-ordinated a case study project
on the impacts of electronic commerce on business. The study
involved nearly 220 firms across 11 countries and a common
methodology was used across 14 broad sectors.
Participants: Canada, France, Italy, Korea, Mexico, the
Netherlands, Norway, Portugal, Spain, Sweden, the United
Kingdom.
The aim of the project was to improve understanding of impacts
of electronic commerce on business.
EBIP used common methodology for firm-level case studies to
improve cross-country and cross-sector comparability.
Previous case study information is anecdotal, fragmented, not
comparable across sectors or countries.
12 12
What is more important the ‘commerce’
factor or the ‘e’ factor?
Q
Successful e-commerce strategies led by commercial
considerations.
Q
E-commerce part of larger business and economic
transformations. Successful application and use are
embedded in broader business strategies with major
emphasis on both e-commerce and ICT skills.
Q
But e-commerce a major business innovation that
most firms will have to adopt.
13 13
Overall the Internet is having large impacts
on how firms conduct business
Expected/actual e-commerce impacts by business function
E x p e c t e d E ffe c t s
A c t u a l Im p a c t s
70%
60%
50%
40%
30%
20%
ar
ke
tD
ev
el
op
ly
s
na
tA
m
is
en
t
t
em
M
M
ar
M
n
io
at
C
In
fo
rm
ke
an
C
n
io
at
rm
fo
In
en
e
ap
iv
e
el
D
ag
tu
r
ry
e
nc
na
Fi
n g O rd
er
&
Pa i n g
ym
en
t
illi
B
at
al
og
ue
s
A
a n d ve
d rt i
s t si n
In
oc g
fo
rm
k
lis
at
ts
io
n
se
rv
ic
es
N
eg
ot
ia
t io
n
10%
0%
T RA NS A C T IO N
P REP A RA T IO N
T RA NS A C T IO N
C O M P L ET IO N
P RO DUC T IO N
S UP P O RT
14 14
.. Reshaping many business processes
Expected/actual e-com process impacts by business process
E xp e c te d E ffe c ts
A c tu a l Im p a c ts
50%
45%
40%
35%
30%
25%
20%
15%
10%
5%
P RO DUC T
IN N O V A T IO N
P R O C ES S
IN N O V A T IO N
t
ty
Lo
ya
l
us
Tr
Ex
pa
ns
Se
io
gm
n
en
ta
tio
n
Lo
gi
st
ic
s
Pr
od
uc
tio
C
on
or
di
na
tio
n
es
ig
n
D
Bu
nd
lin
g
D
iv
er
si
fic
at
io
D
n
iff
er
en
tia
tio
C
us
n
to
m
is
at
io
n
0%
R EL A T I O N A L
IN N O V A T IO N
15 15
.. With generally positive benefits:
one-third of firms had positive impacts on
turnover or profitability
Turnover
Increase
Decrease
Profitability
No Change
NA
Employment
0%
20%
40%
60%
80%
100%
16 16
Analysing the
impacts of ICT
17 17
OECD Growth Project
Q
Q
Q
A two year study on the reasons for differences in underlying
economic growth of OECD member nations during the 1990s.
Final report released in 2001 - “The New Economy: Beyond the
Hype”.
General findings were that:
– ICT is a key technology with the potential to transform economic
and social activity.
– But it is not the only factor, others are:
• quality of human capital
• providing more scope for entrepreneurs to explore business
opportunities
• support for innovation
• getting the fundamentals right – sound macro-economic
management, openness to trade and investment, efficient
markets, well functioning economic and social institutions.
18 18
OECD Growth Project (continued)
Q
ICT plays three roles
– Through capital deepening, as ICT is an important asset
in overall business investment.
– Through multi-factor productivity growth in the
production of ICT goods and services (e.g. technological
progress in semi-conductors).
– Through MFP growth thanks to the use of ICT, either
through efficiency gains in individual firms, or through
network/spillover effects from its use.
19 19
OECD Growth Project (continued)
Q
Findings in respect of ICT considered both ICT
production and consumption and were as follows:
– Productivity in the ICT sector can improve economic
productivity overall, but
– … successful economies were more likely to have rapid
diffusion of ICT, particularly in service industries.
– Deregulation of ICT industries encourages competition and
greater investment in ICT.
– Policies which build confidence – an appropriate regulatory
and legal environment and government leading by example –
are important.
20 20
OECD Growth Project (continued)
Q
In respect of ICT particularly, the policy
recommendation from the final report was that:
– “ICT is an enabling technology, that is transforming
economic activity.
– Governments should:
• focus policy efforts on increasing the use of new
technology
• increase competition and continue with regulatory reform
in the telecommunications industry to enhance the
uptake of ICT
• ensure sufficient competition in hardware and software to
lower costs
• build confidence in the use of ICT for business and
consumers; and, make e-government a priority.”
21 21
More recent
analytical work on
ICT by the OECD
and others
22 22
Much better evidence on the economic
impacts of ICT at three levels of analysis
Q
Q
Q
through macro-economic evidence on the role of
ICT investment in capital deepening
through sectoral analysis showing the contribution
of ICT-producing sectors and ICT-using services to
productivity growth
and through detailed firm-level analysis that has
demonstrated the wide-ranging impacts of ICT in
the economy, even in sectors where sectoral data
suggested that little was happening – this result is
thanks to work in 13 OECD countries.
23 23
24 24
The data on ICT have improved
Q
Early 1990s: mainly private data, sometimes with poor
coverage, sample bias, and unknown quality.
Q
Now:
– Surveys on ICT use by households and businesses in most OECD
countries, with considerable detail on uses and technologies.
– Official estimates of ICT investment, including software.
– Industry data for many countries.
– Growing comparability of data.
– Efforts to establish longitudinal (linked) databases in many
countries.
Q
Greater prospects for empirical research on ICT.
25 25
D
ec
-9
A 8
pr
-9
A 9
ug
-9
D 9
ec
-9
A 9
pr
-0
A 0
ug
-0
D 0
ec
-0
0
A
pr
-0
A 1
ug
-0
D 1
ec
-0
1
A
pr
-0
A 2
ug
-0
D 2
ec
-0
A 2
pr
-0
A 3
ug
-0
D 3
ec
-0
A 3
pr
-0
4
Prices have fallen…
(producer price index for PCs and workstations, 1998=100)
100
90
80
70
60
50
40
30
20
10
0
Source: BLS.
26 26
But countries have not invested to the same extent…
(ICT investment as % of non-residential investment)
35
30
25
1980
1990
2002*
20
15
10
5
Sp
a
P o in
rtu
ga
Ire l
la
n
Au d
st
ri
Fr a
an
ce
Ja
pa
n
I
G ta ly
e
Ne r ma
n
th
er y
la
nd
G s
re
e
Au ce
st
ra
Ca lia
n
De ada
nm
a
Un
Be rk
ite
d lgiu
Ki
m
ng
do
Fi m
nl
an
d
S
Un w
ite ed e
d
n
St
at
es
0
Source: OECD Database on Capital Services, March 2004.
27 27
Why the differences in growth from ICT?
Q
Returns (& investment) in ICT depend on other factors:
– Skills, innovation, organisational changes.
– Not all firms succeed – experimentation and selection play a
role – can successful firms gain market share?
– The scope for change depends on the business environment;
impacts in the US seem larger, perhaps because of greater
scope for innovation and organisational change.
Q
Competition matters:
– Competition forces firms to enhance efficiency.
– Firms that invest first/most in ICT are often new or foreign.
– Competition reduces the costs of ICT & fosters diffusion.
Q
Structural factors – share of services and of large firms.
28 28
Aggregate impacts of ICT investment
Q
Mainly analysed through standard growth accounting
approach, based on capital services:
– Requires long time series of ICT investment.
– For international comparisons – use of hedonic deflators for
all countries to enhance comparability of results.
Q
Q
Official data on ICT investment are now available for
15-20 OECD countries.
Still some measurement problems, e.g. as regards
software investment.
29 29
The US and small EU countries have had a large contribution of ICT
investment, France, Germany and Italy a small one
(contribution to GDP growth, in percentage points)
1.0%
0.9%
0.8%
90-95
95-2002*
0.7%
0.6%
0.5%
0.4%
0.3%
0.2%
0.1%
Fr
an
ce
G
er
m
an
y
Ita
ly
Sp
a
Po in
r tu
ga
l
Fi
nl
an
G d
re
ec
e
Ja
pa
n
C
an
ad
a
Ire
la
Au n d
st
N
et rali
a
h
U
ni erl a
te
nd
d
s
Ki
ng
do
Be m
lg
iu
m
D
e
nm
U
ni
ar
te
k
d
St
at
Sw e s
ed
en
0.0%
Source: OECD Productivity Database, May 2004.
30 30
Finding: ICT investment and diffusion
mainly differ because of lack of demand
Q
Q
Q
Q
Demand has been held back by a business environment that
was not sufficiently geared towards effective use of ICT – e.g.
regulatory barriers, lack of skills, difficulty to change
organisational set-ups, lack of innovation, etc.
Competition has not been equally strong in all countries – costs
of ICT and communications differ, as do pressures to improve
performance.
Demand has been held back by lack of security and trust.
Structural factors also matter – ICT is not suited to all markets,
sectors, or business models.
31 31
Sectoral approach to impacts of ICT
Q
Focus on ICT production – what role does it play, does
it explain US-EU differences:
– Requires definition of ICT production:
• Developed by OECD (both ICT manufacturing and ICT services)
• But not easily applicable for constant price series.
– Typically focus on some key components.
Q
Focus on ICT-using industries:
– Sectors that use ICT most (or have the greatest potential for
ICT use) might first observe productivity impacts – these are
mainly services industries.
– Definitions differ and are not used the same across studies.
32 32
ICT manufacturing is only important for some countries
(average annual contribution to labour productivity growth, in %)
1.1
1.0
1990-95
0.9
1996-2002*
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
M
ex
N
ic
et
o
he
rl a
nd
s
C
an
ad
G
a
er
m
an
y
D
U
e
ni
nm
te
ar
d
k
Ki
ng
Sw
d
i tz om
er
la
nd
Be
lg
iu
m
Fr
an
ce
Ja
U
ni
pa
te
n
d
St
at
Sw e s
ed
en
Fi
nl
an
d
Ire
la
nd
Ko
re
a
Ita
ly
Sp
ai
n
-0.1
33 33
Productivity growth in ICT services (telecom and
software) also plays a role
Annual average contribution to labour productivity growth, in %
0.8
1990-95
0.7
1996-2002*
0.6
0.5
0.4
0.3
0.2
0.1
Sw
i tz
er
la
n
Be d
lg
i
N um
or
w
D ay
en
m
a
Au rk
st
r
Fr ia
an
ce
U
ni Sp
te a
in
d
N Sta
et
he te s
rl a
nd
s
Ja
pa
M n
ex
ic
o
Ita
C ly
an
a
Sw da
U
ni
te ed
d
Ki en
ng
do
Ire m
la
nd
K
Lu
xe ore
m a
bo
u
Fi r g
nl
G and
er
m
an
y
0.0
34 34
Countries with less product market regulation have seen a
stronger pick-up in productivity in ICT-using services
Increase in productivity growth, ICT-using
industries, 1990-95 to 1996-2001 (% point)
1.5
Correlation = -0.71
T-statistics = -3.80
1.0
United States
Ireland
Australia
0.5
United Kingdom
0.0
Canada
Netherlands
Spain
Sweden
Finland
Denmark
France
GermanyAustria
Belgium
-0.5
Japan
Italy
-1.0
0.0
0.5
1.0
1.5
2.0
2.5
Index of Product Market Regulation, 1998
35 35
3.0
Countries with strict employment protection legislation and
product market regulation have invested less in ICT
30
ICT investment in 1998 (as % of GFCF)
United States
Correlation = -0.65
T-statistics = -3.46
25
United Kingdom
20
Netherlands
Sweden
Australia
Canada
Spain
Italy
BelgiumJapan
Denmark
15
Germany
Finland
Austria
France
Ireland
10
Greece
Portugal
5
0
0
1
2
Employment protection legislation index
3
4
36 36
Countries with strong innovation in applications
and software have invested more in ICT
ICT as a % of non-residential investment, 1998
30
Correlation = 0.59
T-statistic = 2.84
28
United States
26
24
United Kingdom
22
Netherlands
Sweden
20
Australia
Canada
18
Belgium Spain
Denmark
Germany
16
Italy
Japan
Greece
14
Ireland
Austria
Portugal
12
France
10
0
10
20
Share of ICT patents in all patents, 1998
30
40
37 37
Countries with an abundance of high-skilled
workers have invested more in ICT
6
Correlation = 0.68
T-statistic = 3.04
ICT investment (%GDP)
5
United States
Sweden
Netherlands
4
Italy
3
Spain
Greece
2
United Kingdom
Finland
Belgium
Denmark
Germany
Luxembourg*
Austria France
1
0
0
1
2
3
Share of high-skilled ICT workers in total occupations
4
38 38
And countries with a rapid increase in investment in
ICT have had more rapid MFP growth
Change in ICT investment as % of GFCF, 19902000
14
12
Sweden
Finland
10
United States
8
Japan
Denmark
Canada
Australia
Ireland
6
Netherlands
France
4
Austria
Germany
2
Italy
Spain
0
-1.5
-1.0
-0.5
0.0
Change in MFP growth from 1980-90 to 1990-2000
0.5
1.0
39 39
Problems with the industry approach
Q
Q
Q
Breakdown in ICT-producing and ICT-using industries
is not always very satisfactory.
Industry data are not always comparable, e.g. for ICT
production and certain services.
The impact of ICT on productivity is not formally
analysed, ideally:
– Regression analysis of impacts ICT use/ICT capital at the
industry level.
– Growth accounts at the industry level.
Q
Unfortunately, industry data on ICT investment are still
lacking for many countries (no time series).
40 40
A firm-level perspective on ICT
OECD project involving researchers and
statisticians in 13 countries
41 41
Large variety of methods and data
Q
Data based on combination (linking) of sources:
– Economic performance often from production surveys;
– ICT from investment surveys, specialised ICT and ecommerce surveys, innovation surveys (US from computer
network use survey).
– Additional data, e.g. on organisation, skills & innovation.
Q
Methods include:
– Labour productivity regressions.
– Estimates of production functions.
Q
Q
Variety of methods and data can help strengthen
evidence.
The studies point to many important interactions.
42 42
Example 1
Hollenstein: What determines ICT uptake by firms?
Q
Q
Study for Switzerland, shows that ICT is linked to
other firm-level factors.
Firms that adopt ICT:
– Anticipate benefits from improved customer-orientation
and lower costs.
– Have overcome problems in financing and skills.
– Have a high capacity to absorb new knowledge (linked to
human capital and innovation).
– Are faced with strong competition and in markets with a
high level of uptake of ICT.
– Also introduce new forms of work organisation.
43 43
Example 2
Gretton, et. al: The impacts of ICT in Australia
Q
Q
Q
Q
Q
Q
The use of ICT raised aggregate multi-factor productivity by
about 0.2% points – over and above impacts of ICT capital.
Impacts are largest in finance and insurance.
Impacts taper off over time – they are largest a few years after
adoption.
Complementary factors play a large role in explaining firm’s
success with ICT – human capital, experience with
innovation, use of advanced business practices and intensity
of organisational restructuring.
Early users of ICT are large firms with skilled managers.
Firms with high openness to trade use more ICT.
44 44
Estimated contribution of ICT to MFP growth in
Australia (% points)
0.35
Computer use & Internet alone
Computer use, Internet & computer use interactions
0.30
0.29
0.28
0.25
0.20
0.15
0.18
0.14 0.14
0.17
0.14
0.14
0.12
0.10
0.14
0.13
0.11
0.09
0.09
0.09
0.07
0.06
0.04
0.05
Group total
services
business
Property &
services
recreational
Cultural &
storage
Transport &
restuarants
Accom., cafes &
Retail trade
Wholesale trade
Construction
Manufacturing
0.00
45 45
Source: Gretton, et al. 2004.
Example 3
Hempell, et al. : The link between ICT and
innovation
Q
Q
Q
Q
Q
Comparative study with comparable data for Germany and the
Netherlands from innovation surveys.
Key question: do firms that engage in complementary
innovation benefit more from investment in ICT?
In both countries, returns to ICT investment are much higher
when accompanied by innovation.
Continuous innovation leads to higher returns than occasional
innovation efforts.
This is also the case in the services sector.
46 46
Example 4
Maliranta and Rouvinen: The impacts of ICT in
Finland
Q
Q
Q
Q
Q
The additional productivity of ICT-equipped workers ranges
from 8 to 18% (after controlling for many other factors).
This effect is higher in young (new) firms and in the ICTproducing sector.
Manufacturing firms benefit mostly from internal networks
(e.g. local area networks), whereas service firms benefit most
from external networks (e.g. the Internet).
Organisational factors are important, as suggested by the
greater productivity effects in new firms.
Selection also plays a role – not all firms succeed – having a
dynamic business sector is important.
47 47
Example 5
Clayton, et al.: Impacts of e-commerce in the UK
Q
Q
Q
Q
Computer networks for trading have positive impacts on
productivity.
E-buying has positive impacts on output growth, whereas Eselling has negative impacts.
This is likely due to pricing effects – e-buyers can source
more competitively, e-sellers may loose out.
In some industries, e-commerce seems to have contributed to
lower prices, perhaps due to greater price transparency and
more rapid reactions of suppliers.
48 48
UK: Productivity analysis and IT
Investment + Employees + Transactions + Communications
Purpose of analysis is to understand different effects
Labour Productivity: by Employee U se of IC T in
Manufacturing and Services
£thousa nd
50
40
30
20
10
0
Les s than 25 per
cent
25 per cent to 49 per 50 per cent to 74 per 75 per cent and over
cent
cent
Pe rce n tag e o f e mp lo ye e s u sin g co mp u te rs / in te rn e t
Using computers
Using Internet
49 49
UK: Productivity analysis and IT
Significant results – IT investment
– Hardware and software capital show significant productivity
gains, both levels and yearly changes
– Gains attributable to IT capital across service sectors higher
than those in manufacturing
– Returns to IT capital across high and low user sectors, and
for large and small firms
– IT capital reinforces productivity effects of other capital
investment, and of skills
– Returns to IT investment ‘explains’ productivity advantage
of US multinationals in UK, concentrated in high IT using
sectors
50 50
UK: Productivity analysis and IT
Significant results – IT use
– Employee use adds significantly to productivity, over and
above effects of hardware and software capital,
• after controlling for sector, size, ownership etc
– Manufacturing firms achieve an extra 2.2% output for each
additional 10% employees using computers
– For employees using internet the extra is 2.9%
– Productivity effect greater for more recently created firms
– Productivity effects associated with communications use
UK firm level productivity results can be found in full at
http://www.statistics.gov.uk/cci/article.asp?ID=1235 (Web link to ONS summary article)
and are broadly consistent with firm level results for Chinese manufacturing at
http//www.mo.rcast.u-tokyo.ac.jp/index-eng.html (analysis by Kazu Motohashi)
51 51
UK: Productivity and broadband
Significant results – broadband use
No clear link from broadband availability to productivity, but:
– Investment in IT higher in regions / sectors with broadband available,
especially for UK firms in non-IT intensive industries.
– Early adopters of broadband experienced high initial
telecommunication costs but these have declined over time
– Broadband users more likely to have multiple e-business links.
– Multiple links plus broadband improve labour productivity.
– Firms with a high broadband equipped labour share have significantly
higher productivity
Full results in OECD paper DSTI ICCP IIS(2006)9 Shikeb Farooqui and Raffaella Sadun
52 52
Example 6
Atrostic et al. : Impacts of computer networks in the
United States, Japan and Denmark
Q
Q
Q
Computer networks have positive impacts on productivity in
all three countries.
In the US manufacturing sector, plants using computer
networks have 5-11% higher productivity than plants not
using networks.
In Japan, both interfirm and intrafirm networks are linked to
strong firm performance.
53 53
The firm-level studies all show that ICT use
can contribute to improved firm performance
Q
Positive impacts of ICT on labour productivity (and
MFP, where measured) or market shares in all
countries:
– But these are conditional on other factors and firm
characteristics (skills, innovation, organisational change).
– Not all firms succeed – selection plays a role.
– ICT is only part of a broader strategy to improve firm
performance.
Q
Q
Networking technologies are particularly important.
The impacts of ICT are also found in the service
sector (despite lack of evidence at the industry level).
54 54
Why the difference between firm-level and aggregate
evidence on ICT’s impacts?
Q
Q
Q
Q
Q
Aggregate impacts ICT in some countries may still be
disguised by other factors.
Lags – the US invested earlier and more – aggregate
impacts in other countries might still come.
US firm-level impacts may be larger – because of scope
for reorganisation and complementary innovation.
Successful US firms may be able to gain more market
share (e.g. Walmart) – more re-allocation.
Spill-over effects? - Perhaps already some of this in the
US (and Australia), not much in other OECD countries.
55 55
Conclusions
Q
Q
Better data have enabled a wide range of empirical
research on ICT and its economic impacts.
This research has had important policy implications:
– It has shown that ICT plays an important role in economic
growth.
– ICT is no panacea, and linked to many other factors – only
focusing on ICT diffusion is not sufficient.
– Countries have not equally benefited from ICT – there is scope
for learning and improvements in the policy environment.
Q
There is scope for much more analysis, in particular
with firm-level data, and in the context of international
comparisons.
56 56
Policy implications from work on ICT and growth
Q
Fostering a business environment for effective use
of ICT
– Competition matters for ICT uptake and effective use.
– Regulatory burdens may limit investment and
experimentation.
– Enable firms to make the necessary organisational
changes.
– New skills are needed, requiring changes to education
and training systems.
– Need for coherent innovation policies.
– Management matters – ICT is no panacea.
57 57
Policy implications (continued)
Q
Competition in ICT goods and services
– Evidence of the benefits of competition is convincing and
liberalisation should be pursued.
– Crucial for broadband development.
– Regulators must be vigilant for anti-competitive
behaviour.
– Technology continues to develop – policy frameworks
and regulations should be technology neutral.
58 58
Policy implications (continued)
Q
Boost Security and Trust
– Uncertainty remains regarding the security of electronic
commerce.
– Attention is required to security of information systems
and networks, authentication, protection of privacy and
consumer protection.
– Cross-border issues and enforcement.
– E-government can help build confidence – and is
important to improve government efficiency and the
delivery of public services.
59 59
Policy implications (continued)
Q
Unleash growth in the services sector
– particularly important because of the potential
benefits
– policies to reduce regulation and reduce barriers
to entry
Q
Harness the potential of innovation and
technology diffusion
– ICT linked to, and assists, innovation
– ICT fosters networking and knowledge transfer.
60 60
Measurement issues
Q
Better measurement has clarified the picture
– Quality-adjusted price indices have helped
demonstrate the impact of ICT on growth.
– Software was included in 1993 SNA as an asset.
– Measures of R&D are being improved:
• New (2002) edition of OECD Frascati Manual.
• R&D may be included in new SNA.
– Measures of innovation are being improved.
– There is more attention at statistical offices for
analysis at the firm level -linking existing data
sources can help.
– Data on ICT demand are improving with adoption
of OECD standards.
61 61
However, some measurement problems remain
Q
Q
Q
Q
Q
Q
Skills data (apart from formal education statistics)
are poor.
Organisational factors are difficult to capture.
Software investment is not yet fully picked up in
national accounts.
Deflators for ICT production differ considerably.
Measurement of services output is poorly in some
areas e.g. finance, health, education.
Cross-country firm-level studies are still scarce,
although much work is underway. OECD is trying
to act as forum for exchange of best practices.
62 62
THANK YOU!
www.oecd.org/sti/measuring-infoeconomy
martin.schaaper@oecd.org
63 63