Transcript Knowledge Spillovers

Economics and Business
Exchange
Supported by Deloitte.
1
Incentivising innovation
ippr economics and business exchange
13 November 2005
Chiara Criscuolo
Centre for Economic Performance, LSE
2
Why are we interested in
innovation?
• Technological knowledge is a driver of productivity
• Not the only one driver:
– Organisational knowledge
– Skills
– Physical Capital, ICT
• “Productivity isn’t everything, but in the long run it is
almost everything” (Paul Krugman)
• Productivity growth drives growth of real wages
• Productivity growth can be used to finance
government expenditure
3
A simplified scheme
Productivity
Knowledge
R&D
Modified
Figure 1
Griliches and Pakes NBER WP561 1980
4
A simplified scheme
Productivity
Selfreported
Innovation
Patents
Knowledge
Knowledge
Flows
competition
Basic Science
R&D
Knowledge base
Non-R&D
Expenditure
In knowledge
costs
Skills
5
How does the UK measure up?
UK Productivity Gap, 1990-2001,
Market Sector (UK=100)
160
140
120
output per hour
1990
output per hour
1995
output per hour
2001
100
80
60
40
20
0
France
Germany
Source: O’Mahony (2004)
US
6
140
UK Productivity Gap, 1999
(UK=100)
120
100
output per hour
(market sector)
80
60
TFP (capital and
skills adjusted,
market sector)
40
20
0
France
Germany
US
Source: O’Mahony and de Boer (2002)
7
INNOVATION OUTPUT: Patents
Share of countries in “triadic” patent families
Source: OECD
8
Other innovation measures
•
•
•
Alternative innovation measures.
Source: Community Innovation Survey, based on Oslo manual
Only comparable information for some EU countries
Source: IFS
9
UK R&D stagnant over medium run
R&D to GDP 1991-2003,G7
Intensity of gross expenditure in R&D
as a percentage of GDP
3.5
R&D intensity has stagnated since 1991 and fallen behind other countries
Not just de-industrialization: within sector
3.0
Japan
USA
Germany
2.5
France
UK
2.0
Canada
1.5
1.0
Italy
0.5
1991
1992
1993
1994
1995
1996
Source: SET, 2005;
http://www.ost.gov.uk/setstats/7.htm
1997
1998
1999
2000
2001
2002
2003
10
Broader measures of investment in knowledge (as % of GDP 2002)
R-D
Software
Higher education
%
Sweden
1.7
1.2
1.3
United States
Finland
1.0
Korea
1.8
Denmark
0.9
OECD
Japan
1.2
Canada
0.1
Australia
0.3
Germany
0.5
Belgium
0.0
EU
0.5
Netherlands
0.3
France
0.3
United Kingdom
Austria
0.2
1.2
Spain
0.7
New Zealand
0.0
Ireland
-0.2
Italy
Change in investment in
knowledge to GDP ratio
(1994-2002)
Greece
Portugal
0
2
4
6
0.3
0.8
0.5
8
11
A simplified scheme
Productivity
Selfreported
Innovation
Patents
Knowledge
Knowledge
Flows
competition
Basic Science
R&D
Knowledge base
Non-R&D
Expenditure
In knowledge
costs
Skills
12
Other innovation inputs: Knowledge Spillovers
• Hard to find cross-country evidence on knowledge
spillovers
• Possible proxies:
– FDI/Presence of foreign affiliates
– Trade openness of the country
– Patent citations
– Mobility of workers
– Cooperation
• Between Businesses
• Between Businesses and the science base
UK seems to do quite well on all of these proxies
13
Innovation Policies
• Correct for market failures (e.g. in financing R&D)
– R&D Tax credit system
– Loan guarantees
– Banking system/venture capital (?)
• Correct for externalities
– Knowledge Spillovers
• Role of foreign presence (embeddedness)
• Role of foreign sourcing (intra and extra firm)
• Absorptive capacity
• Make the country more conducive to innovation:
– Supporting basic scientific research
– commercialisation of inventions
– University-business links (Lambert)
– Increasing supply of skills
– Entry of new firm and growth of better firms
– Competition
14
R&D tax credits
• Many other countries with R&D tax credits…actually UK system
quite generous
• UK adopted in 2000, first for SMEs now for all firms
• Current cost about £430m p.a.
• Good econometric evidence that R&D does react to changes in its
tax-price (Hall and Van Reenen, 2000)
• But why no pick-up (even a fall in 2004)?
– Slow response; takes about 10 years for most of effect to be felt
– … still need to evaluate and do a proper cost benefit
• all R&D subsidies increases wages of (high income) R&D workers
(Goolsbee, 1999)
• the UK R&D intensity gap: smaller proportion of firms doing R&D at
all (firms that do are as R&D intensive as firms abroad)
15
Innovation support programs
Overcome credit constraint and high uncertainty (especially for SMEs)
• Loan guarantee
• SMART
• SPUR
Increase knowledge flows and collaboration
• LINK and Collaborative R&D (between businesses and between
businesses and research institutions)
• Knowledge Transfer Partnerships
• Faraday Partnerships and Knowledge Transfer networks
MANY others: business link webpage : 2667 listings in the support
directory for starting up and developing a business
16
Issues with Innovation Support
schemes
• Policy evaluation widespread in education or labour
market programs
• Existing evaluations of DTI schemes that support
innovation is very scarce and is lumped together with
other schemes (e.g. Regional Selective Assistance)
• Econometric Methodology and data linking allow
evaluation of innovation support programs
17
Role of foreign presence/sourcing
and global engagement
• Evidence that Multinationals and globally engaged firms are
more innovative
– Both because they invest more in R&D
– They learn from their headquarters/other firms in the group (Criscuolo
et al., 2005 and Brastetter, 2004)
• Role of International trade for learning
– Learning by importing (evidence MacGarvie, 2004)
– Learning by exporting (mixed evidence)
• Role of foreign affiliates as a source of knowledge spillovers
– Evidence shows that knowledge spillovers are along the value chain
(vertical)
• Role of domestic multinational as a source of knowledge
spillovers
– Reverse technology transfer/Technological sourcing (Griffith,
Harrison, Van Reenen, 2004)
18
Role of foreign affiliates
• Knowledge spillovers
• Increase competition
• Issues:
– They come but they need to stay: reasons for location
• Lower costs
• Acces to skilled labour/science base
• Proximity to market
– How much knowledge spillovers do UK firms get from foreign affiliates
(FA)?
• evidence that FA source more knowledge from abroad but not more
likely to transfer technology to the local economy as compared to
local firms (Cassiman and Veugelers, 2004) .
– How to increase “Embeddedness”?
– What is the role of UK multinationals?
• They are likely to be more embedded…(interesting issue for
research)
19
Role of foreign sourcing
• One important type of International technology transfer is
“technology sourcing” hypothesis
•
firms obtain international R&D spillovers by locating their
inventors/R&D labs close to leading edge research
• Question: Do UK firms with stronger inventor presence in the US
benefit
• disproportionately from US R&D efforts?
• UK behind the US (frontier) but a lot of historical linkages
• Answer: Griffith, Harrison and Van Reenen find that UK firms
benefit more from US R&D spillovers if they have strong US inventor
presence
• Policy relevance:
– EU policy makers attempted to “repatriate” R&D labs away from
US (Lisbon Agenda)
– More generally: is it nationality or location of R&D that matter? Is
there a trade-off between the two?
20
Other issues: Role of basic science
UK does well in elite science
**More papers per head and citations per head than main competitors
Papers and citations per head (UK=100), average over 1998-2003
Source: Evidence Ltd. Thomson ISI (2004)
21
• Why doesn’t this translate in innovation? Do universities have the
right incentives?
– E.g. evaluation and funding based on RAE performance
– Incentivising commercialisation of inventions by universities
(Higher education innovation fund) : trade-off: knowledge is a
public good…
– Collaboration with businesses
22
Other issues: Role of competition,
regulation and firm turnover
• Competition Important driver of innovation and
productivity (e.g. Blundell, Griffith, Van Reneen, 1999,
Bloom et al., 2004)
• UK position seems strong on OECD Indicators
• Schumpeterian theory of creative destruction stresses
the positive role of firms’ turnover and in particular of firm
entry
• Cross-country evidence shows that UK has quite high
turnover but that the rate of growth of high productivity
firms is low…
23
Policies to Raise Innovation and
Productivity
• Innovation inputs (e.g. R&D)
• Cooperation
• Human capital
– “policy complementarity”: effect on productivity and inequality
– technical Skills/Managerial skills
• Global engagement: trade/inward FDI/outward sourcing
• Competition/Regulation
• Much of policy framework in UK is “right” from economic
perspective: emphasis on skills, tough competition, openness to
trade and FDI, support for R&D
• Delivery. Need for rigorous evaluation (e.g. education reforms, R&D
policy)
24
What NOT to do
• Large state-directed “grand projet”
• Aggressive attempts to “repatriate” R&D from US (Griffith, Harrision,
Van Reenen, 2004)
• Create greater uncertainty through tinkering with small tax schemes.
Evidence that investment responses retarded by micro-uncertainty
(on the increase in US and UK as macro-uncertainty has declined).
Slows down reallocation which drives aggregate productivity
• Use R&D policy as a way of creating equality between regions of the
country: need to concentrate resources where they can be best
used (“pick winners”)
25
Conclusions
• UK continues to have a longstanding productivity gap with US (and
France/Germany)
• Some improvement in recent years at closing the gap with EU – and
has kept up with US even in the “miracle” period post 1995
• Policies to address the gap:
– We concentrates on R&D/innovation
– Skills
– Managerial practices
– Competition
– Infrastructure
• Need for evaluation
• …..although long-term nature of evaluation must be recognised
• Not a single answer
26
Thank you
For references and further reading
Contact:
[email protected]
27
Economics and Business
Exchange
Supported by Deloitte.
28