European Innovation Scoreboard 2007

8/3/2019 European Innovation Scoreboard 2007

1/60

EUROPEAN INNOVATION

SCOREBOARD 2007COMPARATIVE ANALYSIS OFINNOVATION PERFORMANCE

PRO INNO Europe paper N 6

Ee CmmDirECtoratE-GEnEral or EntErprisE anD inDustry


2/60


3/60

EUROPEAN INNOVATIONSCOREBOARD 2007

COMPARATIVE ANALYSIS OFINNOVATION PERFORMANCE

February 2008

PRO INNO Europe paper N 6


4/60

Europe Direct is a service to help you fnd answers

to your questions about the European Union

Freepone number (*):

00 800 6 7 8 9 10 11

(*) Certain mobile telephone operators do not allow access to 00 800 numbers or these calls may be billed.

More inormation on the European Union is aailable on the Internet (http://europa.eu).

Cataloguing data can be ound at the end o this publication.

Luembourg: Oce or Ocial Publications o the European Communities, 2008

ISBN 978-92-79-07319-9

European Communities, 2008

Reproduction is authorised proided the source is acknowledged.

Printed in Italy

Printedonwhitechlorine-freePaPer

The innoation policy initiatie PRO INNO Europe combines analysis and bencmarking

o national and regional innoation policy perormance with support or cooperation o national

and regional innoation programmes and incenties or innoation agencies and other

innoation stakeholders to implement joint actions. The initiatie aspires to become the main

European reerence or innoation policy analysis and deelopment throughout Europe

and brings together oer 200 innovation policy makers and stakeolders rom 33 countries.

Additional inormation on PRO INNO Europe is aailable on the Internet (www.proinno-europe.eu).

Disclaimer

The iews epressed in this report, as well as the inormation included in it, do not necessarily

refect the opinion or position o the European Commission and in no way commit the

institution.

This report has been prepared by the Maastricht Economic and social Research and training

centre on Innoation and Technology (UNU-MERIT) with the support o the Joint Research

Centre (Institute or the Protection and Security o the Citizen) o the European Commission.


5/60

03

TABLEOCONTENTSContents

1. Executive Summary 05

2. European Innovation Scoreboard: Base Findings 07

2.1. Summary Innoation Inde 07

2.2. Key dimensions o innoation perormance 08

3. Convergence in Innovation Perormance beteen EU Member States 011

3.1. Oerall process o conergence 011

3.2. Stable membership o country groups 011

3.3. Conergence between country groups 011

3.4. Epected time to conergence 012

4. Te EU Innovation Gap it te US and Japan 015

5. Tematics 019

5.1. Innoation in serices 019

5.2. Socio-economic and regulatory enironment 020

5.3. Innoation eciency: linking inputs to outputs 023

5.4. Non-R&D innoators 025

6. Future Callenges 030

7. Tecnical Annex: Coice o Indicators and Metodology 034

7.1. Indicators 034

7.2. Methodology o calculating the Summary Innoation Inde 034

7.3. Methodology o calculating the SII growth rate 036

7.4. Calculation o time to conergence 037

8. Annexes 038

Anne A: European Innoation Scoreboard 2007 Current perormance 039

Anne B: European Innoation Scoreboard 2007 Years used

or current perormance 041

Anne C: European Innoation Scoreboard 2007 Denitions

and interpretation 043

Anne D: European Innoation Scoreboard 2007 SII scores oer a 5 year

time period 051

Anne E: European Innoation Scoreboard 2007 Country abbreiations 052


6/60


7/60

05

1.ExE

CUTIvESUMMARY1. Executive Summary

This is the seenth edition o the European Innovation Scoreboard (EIS). The EIS is

the instrument deeloped at the initiatie o the European Commission, under the

Lisbon Strategy, to proide a comparatie assessment o the innoation perormanceo EU Member States. The EIS 2007 includes innoation indicators and trend

analyses or the EU27 Member States as well as or Croatia, Turkey, Iceland,

Norway, Switzerland, Japan, the US, Australia, Canada and Israel. Tables with

denitions as well as comprehensie data sheets or eery country are included in

the Annees. The EIS report and its Annees, accompanying thematic papers,

interactie tables to iew results and the indicators database are aailable at http://

www.proinno-europe.eu/metrics.

The methodology or the 2007 EIS remains largely the same as that used in 2006,

although a more robust analysis o country groupings has been added. or the rst

time, Australia, Canada and Israel hae been included as these countries proide

interesting comparisons to EU Member States. The thematic reports that

accompany this years Scoreboard are on innoation in serices, wider actorsinfuencing innoation perormance and on innoation eciency. In addition, the

2007 EIS refects on seen years eperience in comparing countries innoation

perormance and on where the main uture challenges lie.

Seden, Finland, Denmark, Germany and UK are te most innovativeEU countries and aead o te US (Section 2)

Based on their innoation perormance, the countries included in the EIS 2007 all

into the ollowing country groups:

The innovation leaders include Denmark, inland, Germany, Israel, Japan,

Sweden, Switzerland, the UK and the US. Sweden is the most innoatie

country, largely due to strong innoation inputs although it is less ecient than

some other countries in transorming these into innoation outputs. The innovation followers include Austria, Belgium, Canada, rance, Iceland,

Ireland, Luembourg and the Netherlands.

The moderate innovators include Australia, Cyprus, Czech Republic, Estonia,

Italy, Norway, Sloenia and Spain.

The catching-up countries include Bulgaria, Croatia, Greece, Hungary, Latia,

Lithuania, Malta, Poland, Portugal, Romania and Sloakia. Turkey currently

perorms below the other countries.

These country groups appear to hae been relatiely stable oer the last e years.

Within these groups, countries hae changed their relatie ranking but it is rare or

a country to hae moed between groups. Only Luembourg seems to be on the

erge o entering the group o innoation leaders.

Czec Republic, Estonia and Lituania are on track to reac te EUaverage itin a decade (Section 3)

Although there is relatie stability in the country groupings, oer a longer time

period there is a general process o conergence, with the countries showing below

aerage EU innoation perormance moing towards the EU aerage and closing

the gap with the innoation ollowers and leaders. Based on trends oer recent

years, it would take most moderate innoators and catching-up countries 20 or

more years to close the gap with the EU. Howeer Cyprus, Czech Republic, Estonia,

Lithuania and Sloenia seem to be in a position to close this gap in a shorter period

o time, and or the Czech Republic and Estonia and Lithuania this could occur

within 10 years.


8/60


9/60

07

2.EUROPEAN

INNOvATION

SCOREBOARD:BASEINDINGS2. European Innovation

Scoreboard: Base Findings

2.1. Summary Innovation Index

The Summary Innoation Inde (SII) gies an at a glance oeriew o aggregate

national innoation perormance. igure 1 shows the results or the 2007 SII. or

Australia, Canada, Croatia, Israel, Japan, Turkey and the US the SII is an estimate

based on a more limited set o indicators. The relatie position o these countries

in igure 1 should thus be interpreted with care1.

The SII is calculated using the most recent statistics rom Eurostat and other

internationally recognised sources as aailable at the time o analysis, as shown in

Anne A2. International sources hae been used whereer possible in order to

improe comparability between countries3. It is important to note that the data

relates to actual perormance in years preious to 2007 as indicated in Anne B 4.

As a consequence the 2007 SII does not capture the most recent changes in

innoation perormances, or the impacts o policies introduced in recent years

which may take some time to impact on innoation perormance.

Based on their SII scores the countries can be diided into the ollowing groups5.

This grouping also takes account o perormance oer a 5 year time period in order

to increase robustness.

Sweden,Switzerland,Finland,Israel,Denmark,Japan,Germany,theUKandthe

US are the innovation leaders, with SII scores well aboe that o the EU27 and

most other countries. Sweden has the highest SII o all countries, but its leading

position is mostly based on strong inputs.

Luxembourg, Iceland,Ireland,Austria, theNetherlands,France,Belgiumand

Canada are the innovation followers, with SII scores below those o the

innoation leaders but equal to or aboe that o the EU27.

1 The Technical Anne (section 7.2) proides more details.2 Data as aailable on 18 October 2007. More recent data which might hae become aailable ater 18 October 2007

could not be included due to the time constraint in the publication scheme o the EIS.3 The EU Member States, Iceland and Norway are ully coered by Eurostat. or these countries only data rom

international sources are used. or the other countries data rom other, sometimes national, sources are also used inorder to improe data aailability or these countries.

4 In the large majority o cases (almost 90%) data is rom 2004, 2005 or 2006.5 These country groups were determined using hierarchical clustering techniques (with between-groups linkage using

squared Euclidean distances as the clustering method) and SII scores or 5 years between 2003 and 2007.

igure 1: The 2007 Summary Innoation Inde (SII)

0.

08

0.

23

0.

23

0.

24

0.

25

0.

25

0.

26

0.

26

0.

27

0.

29

0.

31

0.

33

0.

33

0.

35

0.

36

0.

36

0.

36

0.

37

0.

44

0.

47

0.

47

0.

48

0.

48

0.

49

0.

50

0.

53

0.

55

0.

57

0.

59

0.

60

0.

61

0.

62

0.

64

0.

67

0.

73

0.

18

0.

19

0.

45

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

TR RO LV BG HR PL SK PT EL HU LT MTES CY IT SI CZ NOAU EE CA EU BE FR NL AT IE IS LU US UK DE JP DK IL FI CH SE


10/60

08

EUROPEAN

INNOvATION

SCOREBOARD2007

COMPARATIvEANALYSISOINNOvATION

PERORMANCE

Estonia,Australia,Norway,CzechRepublic,Slovenia,Italy,CyprusandSpainare

the moderate innovatorswith SII scores below that o the EU27.

Malta,Lithuania,Hungary,Greece,Portugal,Slovakia,Poland,Croatia,Bulgaria,

Latia and Romania are the catching-up countries. Although their SII scores are

signicantly below the EU aerage, these scores are increasing towards the EU

aerage oer time with the eception o Croatia and Greece. Turkey is currently

perorming below the other countries included in the EIS.

2.2. Key dimensions o innovation perormance

As in preious EIS reports, the 25 innoation indicators in the 2007 EIS hae been

classied into e dimensions to better capture the arious aspects o the innoation

process6. Innovation driversmeasure the structural conditions required or innoation

potential, Knowledge creation measures the inestments in R&D actiities, Innovation

& entrepreneurship measures the eorts towards innoation at the rm leel,

Applicationsmeasures the perormance epressed in terms o labour and business

actiities and their alue added in innoatie sectors, and Intellectual property

measures the achieed results in terms o successul know-how.

igure 2 shows the ranking o countries and or each o the 5 dimensions, romworst to best perormer. Countries and groups generally perorm at a comparable

leel in each o these dimensions but with some noteworthy eceptions.

The innovation leadersare among the best perormers in all 5 dimensions. Howeer,

Germany is perorming relatiely worse in Innoation driers, Denmark in

Knowledge creation and in Applications and the UK in Intellectual property.

Swedens oerall innoation leadership is based on its eceptional perormance in

the three dimensions capturing innoation inputs, but Swedens perormance in

the two dimensions capturing innoation outputs is not as good. O the newly

added countries, we obsere that Israel is a strong perormer in Innoation driers,

Knowledge creation and Applications, but that Intellectual property is a relatiely

weakness.

The innovation followersare aboe aerage perormers in almost all cases. Howeer,Luembourg is perorming relatiely worse in Innoation driers, the Netherlands

in Innoation & entrepreneurship and in Applications and Austria in Applications.

Iceland is perorming relatiely well in Knowledge creation and Luembourg in

Intellectual property.

The moderate innovators are close to or below aerage across the dimensions.

Howeer, Norway is perorming relatiely well in Innoation driers, Cyprus and

Estonia in Innoation & entrepreneurship and Czech Republic in Applications.

Perormance is relatiely worse or Italy in Innoation driers and Innoation &

entrepreneurship, Estonia in Knowledge creation and Cyprus in Applications. The

relatie gap between the moderate innoators and innoation leaders tends to be

greatest in Intellectual Property. O the newly added countries, Australia shows

relatiely strong perormance in Innoation driers and Innoation &entrepreneurship, but perormance in Knowledge creation and Intellectual property

is relatiely weak. or Canada only inormation or two o the dimensions is

aailable, showing about the same relatie moderate perormance.

The catching-up countriesare below EU aerage in all o the dimensions with the

noticeable eception on Applications where Malta has the highest ranking and

Sloakia ranks aboe some innoation leaders. In both cases these countries score

highly on sales o new to market products, which may be a refection o the

relatiely small markets that companies in these countries operate within. In both

cases the high ranking on Applications is also partly due to the structure o their

economies as Malta has high eports o high technology products and Sloakia a

high share o employment in medium-high and high tech manuacturing. Although

Turkeys oerall perormance is below that o EU Member States, it has a strongerperormance than some Member States on Knowledge creation7.

6 These dimensions were introduced in the EIS 2005. Details can be ound in the 2005 Methodology Report:http://www.proinno-europe.eu/etranet/admin/uploaded_documents/EIS_2005_Methodology_Report.pd

7 Turkeys perormance may not be accurately refected in the Intellectual property dimension as it does not hae thesame home adantage or EPO patents and Community designs and trademarks as the EU Member States hae.


11/60

09

2.EUROPEAN

INNOvATION

SCOREBOARD:BASEINDINGS

8 or Innoation driers CA is not ranked due to missing inormation. or Innoation & entrepreneurship CA, HR, IL, IS, JP,SI, TR and US are not ranked due to missing inormation. or Applications AU, CA, JP, TR and US are not ranked due tomissing inormation. See Anne A. or Intellectual property scores or RO and TR are too small to be shown in the gure.

Colour coding is conorm the groups o countries as identied in Section 2.1: bright green is Sweden,

green are the innoation leaders, yellow are the innoation ollowers, orange are the moderate innoators,

blue are the catching-up countries, dark blue is Turkey.

igure 2: Innoation perormance per innoation dimension8

Innovation drivers

0.00

0.20

0.40

0.60

0.80

1.00

CA TR MT PT RO HU EL IT SK CZHR LU PL BG LV CY DE ES AT EU SI LT EE BE IS FR NL IE US IL CH JP NOUK AU SE FI DK

Knowledge creation

RO SK MT LV EE PT LT TR PL CY BG EL ES AU HR NOCZ HUBE SI IT EU LU FR DKCA NL IE UK AT CH DE US JP FI IS IL SE

0.00

0.20

0.40

0.60

0.80

1.00

Innovation & entrepreneurship

CA HR IL IS JP SI TR US SK RO HU IT PL NOES BG MT NL LV LT EL CZ AT FR PT EU BE CH IE DE AU FI LU DK EE CY UK SE

0.00

0.20

0.40

0.60

0.80

1.00

Applications

AU CA JP TR US LV CY LT EL PT HR NOBG PL EE NLRO ES AT IT HU IS SI DK BE FR EU IE CH SK SE LU CZ UK FI IL DE MT

0.00

0.20

0.40

0.60

0.80

1.00

Intellectual property

RO TR BG HR LT LV EL SK HU PL EE CZ SI PT MT CY NOES AU IE IS UK IT CA FR EU BE IL US JP FI NL AT DK SE LU DE CH

0.00

0.20

0.40

0.60

0.80

1.00


12/60

010

EUROPEAN

INNOvATION

SCOREBOARD2007


PERORMANCE

An important result rom this analysis is that the innoation leaders and the

innoation ollowers hae a relatiely een and strong perormance across all e

dimensions o innoation9. This tends to indicate mature innoation systems,

although in all cases there are areas o relatie weakness that require attention. In

contrast, the moderate innoators and catching up countries tend to hae a less

een perormance across the e dimensions, indicating that these countries may

need to correct the imbalances in their innoation systems i they are to progress

to higher leels o perormance (igure 3).

9 As demonstrated in the EIS 2005 Thematic report on Strengths and Weaknesses, a well-rounded and equialentperormance on all dimensions increases oerall innoation perormance.

Current perormance as measured by

the SII is shown on the ertical ais.

Relatie to EU growth perormance o

the SII is shown on the horizontal ais.

This creates our quadrants: countries

aboe both the aerage EU trend and

the aerage EU SII are orging ahead

rom the EU, countries below the

aerage SII but with an aboe aerage

trend perormance are catching up,

countries with a below aerage SII and

a below aerage trend are alling

behind, and countries with an aboeaerage SII and a below aerage trend

maintain their lead but are growing at

a slower rate.

igure 3: Conergence in innoation perormance

AU

IL

SKEL LT

PT

LU

ATIE

FR

NLBE

UKUS DE

JP

CH

DK

HU

CYESMT

EE

ITCZ

SI

IS

CA

HR

PL

LV

RO

BG

SE

FI

NO

TR

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

-4.0 -3.0 -2.0 -1.0 0.0 1.0 2.0 3.0 4.0 5.0 6.0

Average growth rate of SII (2003-2007)

2007SummaryInnovationIndex

Sweden Innovation leaders Innovation followers

Moderate innovators Catching-up countries Turkey

Dotted lines show EU performance.


13/60

011

3.CONVERGENCEI

N

INNOVATION

PERFORMANCEBETwEENEU

MEMBERSTATES3. Convergence

in Innovation Perormance

beteen EU Member States3.1. Overall process o convergence

igure 3 shows current innoation perormance as measured by the SII on the

ertical ais against short-run trend perormance o the SII on the horizontal ais10.

There is a process of convergence in innoation perormance in Europe with most

Member States with below aerage perormance haing positie trends. Most o

the moderate innoators and catching-up countries are closing the gap with the

EU and the innoation leaders and ollowers. The innoation leaders and ollowers

are eperiencing a relatie decline in their innoation lead. Notable eceptions

include Luembourg which combines a moderate leel o perormance which a

high SII growth rate; Spain, Greece and Croatia which all hae relatiely low SIIgrowth rates; and Norway and Turkey which are eperiencing ery low SII growth

rates. The ollowing section will analyse in more detail i this oerall process o

conergence is taking place between and/or within the our identied country

groupings.

3.2. Stable membersip o country groups

As set out in Section 2.1, countries hae been classied into dierent innoation

groups based on their SII scores oer a 5-year period. Changes in group membership

within the 5-year period o time are shown in igure 4. Group memberships are

largely stable but we do see some changes:

Luxembourgisin theprocessofmovingfromtheinnovationfollowerstotheinnoation leaders;

CyprusandMaltahavemovedfromthecatching-upcountriestothemoderate

innoators;

LatviaandRomaniaarerstpartofaclusterwithTurkeyandthenmovedtothe

catching-up countries.

Cluster membership (igure 4) is more stable than the ranks o countries; ranks

within a cluster are ar rom stable, as shown by or eample Belgium in the cluster

o innoation ollowers and the US in the cluster o innoation leaders. These results

show that one should not ocus too much on changes in rank rom one year to the

net within the same cluster. It is better to ocus on cluster membership and the

countries within the same cluster and to identiy or each country peer countries.

This is consistent with the Strengths and Weaknesses report o 2005 where peer

countries were identied based on comparable relatie perormance leels.

3.3. Convergence beteen country groups

The preious section showed that despite the general process o conergence,

cluster membership is stable oer time. This suggests that the obsered conergence

is a general trend rather than the result o eceptional single countries

deelopments. This can be shown by plotting the eolution o the innoation

perormance o the dierent clusters (upper hal o igure 5. We obsere increasing

relatie perormance or the catching-up countries and the moderate innoators,

stable relatie perormance or the innoation ollowers and declining relatie

perormance or the innoation leaders. Conergence between the country groups

is shown in the lower hal o igure 5 where the dierences in the cluster SII scoreshae been plotted oer time. The results show a strong process o conergence

10 The Technical Anne (section 7.3) proides more details.


14/60

012

EUROPEAN

INNOvATION

SCOREBOARD2007


PERORMANCE

taking place between the innoation leaders, innoation ollowers and moderate

innoators. There is also some conergence between catching-up countries and

moderate innoators. We can estimate the theoretical time o conergence or each

o these processes using a simple linear approach which will be discussed in Section

3.4. On this simplied basis, it would take almost 30 years or the catching-up

countries to close the gap with the moderate perormers, and almost 40 years or

the latter to close the gap with the innoation ollowers and about 25 years or the

latter to close the gap with the innoation leaders. In conclusion one can see that

conergence between clusters is taking place, but it is likely to take many years

beore this conergence process is completed.

3.4. Expected time to convergence

Using a simple linear etrapolation o current perormance leels and growth

rates11, an estimate can be made or how many years it would take countries to

either catch up or decline to the EU aerage leel o perormance based on current

trends. igure 6 shows the estimated number o years to catch up to or decline to

the EU aerage or European countries only. or 4 o the moderate innoators and

catching-up countries a short-term conergence to the EU aerage perormanceleel could be epected in about 10 years time. These countries are Estonia, Czech

11 The Technical Anne (section 7.4) proides more details.

Colour coding is conorm the groups

o countries as identied in Section

2.1: bright green is Sweden, green are

the innoation leaders, yellow are the

innoation ollowers, orange are the

moderate innoators, blue are the

catching-up countries, dark blue is

Turkey. The ordering o the countries

ollows the rankings o their SII score

or that year (see Anne D).

These country groups were determined

using hierarchical clustering techniques

(with between-groups linkage using

squared Euclidean distances as the

clustering method) and SII scores or

each year between 2003 and 2007.

Cluster results or 2007 as shown in

other sections o the EIS 2007 report

were determined using SII scores or 5

years between 2003 and 2007 and

thus dier rom those shown in igure

3 where the cluster results are or SII

scores or 2007 only. Hence LU, LT andMT are in dierent groupings based

on their 2007 SII than or the 5 year

period shown in igure 1.

igure 4: Cluster membership oer time

SE

2003

FI

DK

CH

IL

US

JP

DE

UK

BE

IE

LU

NL

IS

FR

CA

AT

NO

AU

EEIT

CZ

SI

ES

CY

MT

EL

HU

HR

LT

SK

PL

PT

BG

LV

RO

TR

SE

2003

CH

FI

DK

IL

JP

US

DE

UK

LU

IS

BE

NL

IE

FR

CA

AT

NO

AU

SI

EE

CZ

IT

ES

CY

MT

EL

HU

PT

LT

HR

SK

PL

BG

LV

RO

TR

SE

2003

CH

FI

DK

IL

JPDE

US

UK

LU

IE

IS

BE

NL

FR

AT

CA

NO

EE

AU

SICZ

IT

ES

CY

MT

EL

HU

LT

PT

HR

SK

PL

BG

LV

RO

TR

SE

2003

CH

FI

DK

IL

JPDE

LU

US

UK

IE

IS

BE

FR

NL

AT

CA

NO

EE

SI

AUCZ

IT

ES

CY

MT

LT

PT

HU

EL

SK

PL

HR

BG

LV

RO

TR

SE

2003

CH

FI

IL

DK

JPDE

UK

US

LU

IS

IE

AT

NL

FR

BE

CA

EE

AU

NO

CZSI

IT

CY

ES

MT

LT

HU

EL

PT

SK

PL

HR

BG

LV

RO

TR


15/60

013

3.CONVERGENCEI

N

INNOVATION

PERFORMANCEBETwEENEU

MEMBERSTATES

Republic, Lithuania and Cyprus. or Sloenia short-term conergence could be

epected in about 15 years time, or Poland, Portugal, Latia, Bulgaria, Sloakia,

Malta and Romania conergence would take at least 20 years. or Hungary and

Italy the catching up process would take more than 30 years. On the other hand,

countries like Belgium, rance, the Netherlands and Denmark: these countries still

show an aerage alue o the SII aboe the EU aerage, but might regress to the

EU aerage, possibly within the net 5 to 10 years, as the aerage EU perormance

increases aster than their indiidual innoation perormance. inally, based on this

analysis, some countries seem to stay outside the conergence process (and are not

thereore represented in the chart) as they are either moing away rom the

aerage in a negatie direction (Spain, Greece, Croatia, Norway and Turkey) or in

a positie direction (UK, Iceland, Austria and Luembourg).

Howeer, linear etrapolations o trends are likely to become less reliable oer

longer time periods, as maintaining the aboe EU growth rates will become more

and more dicult when countries start to approach the EU aerage leel o

perormance. A non-linear catching-up process was thereore modelled by

assuming that the growth rate o each country would diminish oer time12. The

catching-up process now looks dierent, with only Estonia and the Czech Republic

as likely candidates to complete their catching-up process in the short-run.

Belgium, rance and the Netherlands are still in danger o alling back to the

aerage EU leel o perormance within a relatiely short time period. While

Sweden was predicted to all back to the EU leel in 17 years time using the linear

approach, in the non-linear approach it would take more than 100 years.

Understanding how countries innoation perormance can change oer time is

one o the key uture challenges identied in Section 6. The analysis conducted in

this section shows that oer a e year time period there has been a relatiely stable

grouping o countries, with each group at a dierent leel o innoation

perormance. This nding points to the diculty o bringing about major changes

in oerall innoation perormance. This may be because innoation has many

dimensions along which countries need to improe in order to increase their oerall

12 The Technical Anne proides more details.

Aerage or moderate innoators does

not include Australia, aerage or

innoation ollowers does not include

Canada and aerage or innoation

leaders does not include Israel, Japan

and the US.

igure 5: Conergence between groups o countries

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

CATCHING-UP VS. MODERATE MODERATE VS. FOLLOWERS FOLLOWERS VS. LEADERS

MODERATECATCHING-UP FOLLOWERS LEADERS

2003 2004 2005 2006 2007

0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0.16

0.18

2003 2004 2005 2006 2007


16/60

014

EUROPEAN

INNOvATION

SCOREBOARD2007


PERORMANCE

perormance; but also because changing innoation perormance simply takes

time.

Howeer, oer a longer time period we do obsere a more dynamic situation. irst,

there are some countries that appear to hae made a transition between dierent

leels o innoation perormance and it would appear that some other countries are

on track to making such a transition in uture. Second, there appears to be a long-term trend towards conergence between the dierent groupings. I this continues,

it may mean that the dierent groupings merge oer time or alternatiely it may

lead to new patterns and trends emerging.

or countries haing either both aboe

aerage SII and growth rates or both

below aerage SII and growth rates,

years to catch up could not be

calculated as these countries are either

epected to increase their lead,

respectiely gap, towards the EU (AT,

EL, ES, HR, IS, LU, NO, TR AND UK).

Time to catch up eceeding 100 years

is not shown (linear: DE; non-linear:

BG, CH, DE, I, IE, HU, IT, Lv, PL, PT,

RO, SE, SK).

igure 6: Time to catch up or all back to EU aerage perormance

53

34

22 22 21 20 20 18 18

13

10 10 10 9

1

5

89

17

2226

49

76

1916

22

13 12

2

6

9

17

31

1

10

100

IT HU RO MT SK BG LV PT SI CY LT CZ EE BE FR NL DK SE FI IE CH DEPL

Years to catch up linear Years to fall down linear Years to catch up non-linear Years to fall down non-linear


17/60

015

4.ThEEUINNOVATION

GAPwIThThEUSAND

JAPAN4. Te EU Innovation Gap

it te US and Japan

The data used or the 2007 EIS (igure 7) shows that the US and Japan are stillahead o the EU, but the innoation gaps hae been declining13. The EU-US gap

has dropped signicantly between 2003 and 2006 and shows a urther but ery

modest reduction in 2007. The EU-Japan gap rst increased in 2004 and then

dropped more signicantly between 2004 and 2006 and ery modestly in 2007.

There are 15 indicators with ull data or the US and EU, and o these the US

perorms better than the EU in 11 indicators (Table 1), while the EU scores aboe

the US in 4 indicators (S&E graduates, employment in medium-high and high-tech

manuacturing, community trademarks and community designs). Although the US

is leading in 11 indicators, on 9 o these indicators the US is outperormed by at

least one European country. Only in tertiary education and USPTO patents the US

is perorming better than any European country.

Japan perorms better than the EU in 12 indicators, while the EU only scores aboeJapan in 2 indicators (community trademarks and community designs). Although

Japan is leading in 12 indicators, on 9 o these indicators Japan is outperormed by

at least one European country. Only in tertiary education, USPTO patents and triad

patenting Japan is perorming better than any European country.

or the EU, EU innoation leaders, US and Japan the latest aailable data are

shown (c. Annees A and B). or indicator 3.4 or the EU and the EU innoation

leaders data or 2005 are used instead o the 2006 data as shown in Anne A.

European early-enture capital data fuctuate on aerage by 150% between 2005

and 2006 turning a long-lasting EU-US gap suddenly in an EU-US lead assuming an

the same US perormance in 2006 as in 2005. Pending the release o 2006 US data

showing the true nature o this possible lag reersal, we hae adopted to compare

perormance leels in 2005.igure 8 shows those areas where there is an increasing or stable EU lead oer the

US, where there is a decreasing gap and where there is an increasing gap. The EU

is eperiencing a stable lead with the US in Community designs where it would be

epected to hae a home adantage oer the US. The EU is increasing its lead in

S&E graduates, medium-high and high-tech manuacturing employment and

13 A direct comparison o the 2003-2006 gaps shown in igure 7 with those reported in the EIS 2006 report is not possibleor seeral reasons. irst, not or all indicators data has been updated with one year, or some indicators data eithercould not be updated or the update is or more than one year, so the gap shown or 2006 in igure 2007 will be

dierent rom the gap shown in the EIS 2006 report. Second, last year the gap was calculated as the dierence betweenthe SII using all indicators, thus by comparing the SII or the EU with the estimated SII scores or the US and Japan. Thisyear, in order to improe the comparability, the gap is calculated as the dierence between the SII scores only usingthose indicators or which data are aailable or the US respectiely Japan.

The ertical ais represents the dier-

ence between SII scores o EU and US

and Japan respectiely. SII scores are

calculated using the re-scaled alues

or those indicators only or which

data or the US respectiely Japan are

aailable. or the EU-US comparison

these are the ollowing indicators: S&E

graduates, population with tertiary

education, broadband penetration,

public R&D, business R&D, share o

medium/high-tech R&D, early-stage

enture capital, ICT ependitures,

high-tech eports, medium/high-tech

manuacturing employment, EPO pat-

ents, USPTO patents, triad patent,

trademarks and designs. or the

EU-Japan comparison the same indica-tors are used ecept early-stage en-

ture capital.

igure 7: EU Innoation Gap towards US and Japan

-0.164

-0.146

-0.116-0.100 -0.098

-0.20

-0.18

-0.16

-0.14

-0.12

-0.10

-0.08

-0.06

-0.04-0.02

0.00

2003 2004 2005 2006 2007

EU-US

-0.162 -0.169 -0.162-0.151 -0.150

-0.20

-0.18

-0.16

-0.14

-0.12

-0.10

-0.08

-0.06

-0.04-0.02

0.00

2003 2004 2005 2006 2007

EU-Japan


18/60

016

EUROPEAN

INNOvATION

SCOREBOARD2007


PERORMANCE

Community trademarks. or community trademarks a similar home adantage

applies or Community designs, but here the EU is steadily increasing its lead rom

haing about twice as many new Community trademarks in 2002 to more than 3

times as many new Community trademarks in 2006. The increase in the lead in

S&E graduates and medium-high and high-tech manuacturing employment is

more moderate.

The EU is eperiencing a gap in all other indicators, but this gap is decreasing or

the broadband penetration rate, early-stage enture capital14, ICT ependitures and

triad patents. The gap or the broadband penetration rate has almost disappeared

in 2006, with the US haing only about 10% more broadband lines per 100

population as compared to almost 100% in 2002-2003. The gap or ICT

ependitures has also almost disappeared with the US GDP spending share on ICT

only about 5% higher than that o the EU. or early-stage enture capital we rst

see an oerall decline, but with some periods o increase which may refect the

cyclical nature o enture capital markets. Neertheless the gap remains large, with

the GDP share o early-stage enture capital still being more than 50% higher in

the US. The gap or triad patents has been steadily decreasing since 2000, when

the US had more than twice the amount o triad patents per million population. In

2006 the US still had 60% more triad patents per million population, the gap thus

remains large.

14 US data are aailable up until 2004, EU data up until 2005. Until 2004 the EU has been eperiencing a lag which, asshown in igure 8, has been decreasing. The early-stage enture capital perormance o the EU improed with 150% in2005, thus reersing this gap in a hypothetical lead as shown in Table 1 assuming that the US perormance leel in 2005would remain unchanged.

Table 1: Dierences in EU-US and EU-Japan Perormance by Indicator

EU US JP European Innovation leaders

INNOVATION DRIVERS

1.1 S&E graduates 12.9 10.6 13.7 IE (24.5) R (22.5) LT (18.9)

1.2 Tertiary education 23.0 39.0 40.0 I (35.1) DK (34.7) NO (33.6)

1.3 Broadband penetration rate 14.8 18.0 18.9 DK (29.6) NL (29.0) IS (28.1)

KNOwLEDGE CREATION

2.1 Public R&D ependitures 0.65 0.69 0.74 IS (1.17) I (0.99) SE (0.92)

2.2 Business R&D ependitures 1.17 1.87 2.40 SE (2.92) I (2.46) CH (2.16)

2.3 Share o medium-high/high-tech R&D 85.2 89.9 86.7 SE (92.7) DE (92.3) CH (92.0)

INNOVATION & ENTREPRENEURShIP

3.4 Early-stage enture capital 0.022 0.035 DK (0.051) UK (0.047) I (0.044)

3.5 ICT ependitures 6.4 6.7 7.6 BG (9.9) EE (9.8) Lv (9.6)

APPLICATIONS

4.2 High-tech eports 16.7 26.1 20.0 MT (54.6) LU (40.6) IE (28.9)

4.5 Employment in medium-high/

high-tech manuacturing 6.63 3.84 7.30 DE (10.75) CZ (10.33) SK (9.72)

INTELLECTUAL PROPERTY

5.1 EPO patents 128.0 167.6 219.1 CH (425.6) DE (311.7) I (305.6)

5.2 USPTO patents 49.2 273.7 274.4 CH (167.5) I (133.2) DE (129.8)

5.3 Triad patents 19.6 33.9 87.0 CH (81.3) DE (53.8) NL (47.4)

5.4 Community trademarks 108.2 33.6 12.9 LU (902.0) CH (308.3) AT (221.5)

5.5 Community designs 109.4 17.5 15.2 DK (240.5) CH (235.7) AT (208.8)


19/60

017

4.ThEEUINNOVATION

GAPwIThThEUSAND

JAPAN

igure 8: Conergence and Diergence in EU-US Innoation Gap

Increasing or stable EU-US lead

0

50

100

150

200

250

300

350

1999 200 0 2001 2002 2003 2004 200 5 2006

0

200

400

600

800

S&E graduates

Med-hi/high-tech manufacturingemployment

Community trademarks

Community designs (2nd axis)

Decreasing EU-US gap

0

20

40

60

80

100

120

1999 2000 2001 2002 2003 2004 2005 2006

Broadband penetration rate

Early-stage venture capital

ICT expenditures

Triad patents

Stable EU-US gap

0

20

40

60

80

100

120

1999 2000 2001 2002 2003 2004 2005 2006

Population with tertiary education

Business R&D expenditures

Share of med-high/high-tech R&D

EPO patents

USPTO patents

Increasing EU-US gap

0

20

40

60

80

100

120

1999 2000 2001 2002 2003 2004 2005 2006

Public R&D expenditures

Exports of high technology products

All alues are relatie to the US with

the US = 100.

The EU-US gap is stable or population with tertiary education, business R&D,

medium-high and high-tech manuacturing R&D, EPO patents and USPTO patents.

The gap is smallest or the share o medium-high and high-tech manuacturing

R&D, but gien the act that most R&D ependitures in the manuacturing sector

come rom so-called high-tech and medium-high-tech manuacturing industries, it

should not come as a surprise that these shares are almost equal in the US and the

EU as both hae similar R&D specialisation patterns. The EU is eperiencing a gap

in EPO patents despite its home adantage, and a large gap in USPTO patents where

the US has a home adantage. The decreasing gap in Triad patents may thereore

be a more important indicator. There is a large gap in business R&D ependitures,

1.17% o EU GDP as compared to 1.87% in the US which is not becoming smaller.

The EU-US gap in the share o population with tertiary education is also large with

almost 40% o US adults in 2005 haing completed a tertiary education as compared

to 23% in the EU in 2006. This gap might be an indicator o a relatie shortage o

the supply o adanced skills in Europe, but dierences in US and European

educational systems might lead to relatiely oerrated US scores on this indicator.

The EU-US gap is increasing in public R&D ependitures and eports o high-tech

products. Up until 2001 the EU was leading in public R&D ependitures, but in


20/60

018

EUROPEAN

INNOvATION

SCOREBOARD2007


PERORMANCE

2002 this lead turned into a small but increasing gap. This switch in leadership was

both caused by a decline in the public R&D intensity in the EU and an increase in

public R&D intensity in the US, in particular by decreasing EU R&D ependitures

and increasing US R&D ependitures in the goernment sector (GOvERD). The US

is also increasing its lead in high-tech eports, in particular rom 2005 to 2006.

The trends in the EU-Japan innoation gap show greater stability with no signicantchanges in the indicators or population with tertiary education, public R&D

ependitures, medium/high-tech manuacturing R&D, ICT ependitures, eports

o high-tech products, employment in medium/high-tech manuacturing, USPTO

patents and triad patents. As is the case with the US, the EU is eperiencing an

increasing lead oer Japan in Community trademarks and Community designs

(igure 9). The EU-Japan gap is decreasing in S&E graduates and broadband

penetration. The share o S&E graduates is almost equal in the EU and Japan in

2006. Japan is still enjoying a lead in broadband penetration but this lead

disappearing ast. The EU-Japan gap is increasing or business R&D ependitures

and EPO patents.

igure 9: Conergence and Diergence in EU-Japan Innoation Gap

All alues are relatie to Japan with

Japan = 100.Increasing EU-Japan lead

0

200

400

600

800

1000

1999 2000 2001 2002 2003 2004 2005 2006

Com munity trademarks

Com munity designs

Decreasing EU-Japan gap

0

20

40

60

80

100

1999 2000 2001 2002 2003 2004 2005 2006

S&E graduates

Broadband penetration rate

Stable EU-Japan gap

0

20

40

60

80

100

120

1999 2000 2001 2002 2003 2004 2005 2006

Po pulation with tertiary education

Public R&D expenditures

Share of m ed-high/high-tech R&D

ICT expenditures

Exports of high technology products

Med-hi/high-tech manufacturing empl.

USPTO patents

Triad patents

Increasing EU-Japan gap

0

20

40

60

80

1999 2000 2001 2002 2003 2004 2005 2006

Business R &D expenditures

EPO patents


21/60

019

5.ThEMATICS5. Tematics

5.1. Innovation in services

This section proides a summary o the thematic paper on serices innoation 15.

The serices sector16 is becoming more and more important in deeloped countries,

both in terms o its share in total alue-added or GDP and employment. On

aerage, the serices sector contributed to 40% o total EU25 employment in 2004

and to 46% o EU25 alue-added. This contribution is oer twice as large as the

contribution o the manuacturing sector. Within the serices sector, Knowledge

Intensie Business Serices (KIBS)17 hae attracted policy interest because o rapid

rates o growth in some countries and because they are considered to be highly

innoatie. The relatie economic contribution o KIBS has been increasing oer

time. The share o manuacturing alue-added in real prices declined by 2.5%

between 1999 and 2004 while the share o serices sector alue-added decreased

by 0.3% and KIBS increased by 6.8%. Based on these trends and the larger

contribution o serices to the economy, KIBS are likely to be one o the main actors or uture growth within the EU. The economic importance o serices

suggests that improements in European liing standards are likely to depend more

on productiity improements in the serices sector than in manuacturing. This

has been demonstrated or the US, where serices contributed three-quarters o the

increase in productiity ater 199518. Much o the productiity increase is due to

dierent types o innoation, deeloped both in-house by serice rms and rom

serice rms adopting productiity enhancing innoations such as ICT.

Although both the economic weight o the serices sector and the importance o

serice sector innoation to economic prosperity hae been recognized or well

oer a decade, there has been a lag in the collection o European innoation

statistics or serices and in the deelopment o innoation policies o releance to

serice sector rms. There are partly good reasons or this. or instance, themanuacturing sector is the source o many o the technical product and process

innoations that are adopted by serices sector rms. Howeer, a growing

awareness o the role o non-technological innoation, sotware, and logistics in

innoation has meant that the serice sector is no longer (i it eer was) a passie

adopter o manuacturing innoations. This is also leading to a rethink o European

innoation policy and an ealuation o the steps that might be needed to remoe

or reduce the policy bias towards manuacturing19.

A common concern is that innoation policy is not adequately sering the needs o

serice sector rms. By comparing innoation indicators or rms in the serice and

manuacturing sectors one can eamine whether rms responses to the CIS

support this concern or not. This comparison indicates two areas where serice

rms responses dier markedly rom those o manuacturing rms: publicprocurement and support rom innoation programmes. or three policy areas,

support could be required under specic conditions: use o intellectual property,

use o and access to public science, and aailability o nancing. or three areas

there is no eidence to suggest that policy is biased against serice rms: supply o

qualied personnel, support or start-ups, and regulatory burdens. Howeer, in

15 http://www.proinno-europe.eu/inde.cm?useaction=page.display&topicID=282&parentID=5116 The Serices Sector is dened as NACE classes G (Wholesale and retail trade; repair o motor ehicles, motorcycles and

personal and household goods), H (Hotels and restaurants), I (Transport, storage and communication), J (inancialintermediation), and K (Real estate, renting and business actiities). Not included are the serices included in NACEclasses L (Public administration and deence; compulsory social security), M (Education), N (Health and social work) andO (Other community, social and personal serice actiities) as these sectors are not coered by the CommunityInnoation Surey (CIS).

17 KIBS includes Computer and related actiities (NACE K72), Research and deelopment (NACE K73), Architectural andengineering actiities and consultancy (NACE K74.2) and Technical testing and analysis (NACE K74.3).

18 Bosworth BP, Triplett, J. The early 21st Century US productiity epansion is still in serices. International ProductivityMonitor, No. 14, pp 3-19, Spring 2007.

19 Eamples include the report by the European Commission, Staff working document on innovation in Services, 2007andthe report by the Epert Group on Innoation in Serices, Fostering Innovation in Services - Final Report, 2007.


22/60

020

EUROPEAN

INNOvATION

SCOREBOARD2007


PERORMANCE

these areas the particular needs o serices rms may dier rom manuacturing

rms een though the oerall leels o concern are similar.

Another important concern or policy is whether innoation perormance diers

signicantly between manuacturing and serices sectors. Analyzing composite

innoation indicators using CIS-4 data shows that seeral o the new Member

States perorm better on serice sector innoation than on general innoation asmeasured by the Summary Innoation Inde. The results suggest that innoatie

serice sector rms in the new Member States could benet as much rom

innoation as rms in more innoatie countries, een though the nature o the

innoation could be ery dierent. The results o an analysis o Knowledge

Intensie Business Serices (KIBS) proide no eidence that KIBS drie oerall

innoatie perormance, as measured by a change in the Summary Innoation

Inde. Howeer, the KIBS share o total employment and alue-added in 2004 is

positiely correlated with innoatie perormance on the 2006 Summary Innoation

Inde. This is probably because o the high leel o innoatie actiity within KIBS

itsel, such as in sotware deelopment. The lack o eidence or a driing role or

KIBS could be due to a lack o data or many countries or NACE 73, which is a key

KIBS sector that includes R&D serices and high technology start-up rms.

A nal important concern is whether current indicators properly capture serices

innoation. The Community Innoation Surey (CIS) is the main source o

innoation indicators and was at rst designed to measure technological innoation

in the manuacturing sector. Oer time improements hae been made to coer a

large share, but not all, o the business serices sector and improe questions

dealing with both technological and non-technological innoation. But urther

improements are needed to measure serices innoation in the uture, either

through modications to the CIS or through other sureys:

1. Research on serice sector innoation (and on innoation in the manuacturing

sector) would be considerably improed i disaggregated results were aailable

or the CIS questions on rms introducing new or signicantly improed goods

and/ or serices. Results or these two options could be used to obtain a better

measure o the types o new products introduced both by manuacturing andserice rms. Similarly, disaggregated results are needed on rms introducing

new or improed methods o manuacturing or producing goods or serices,

new or signicantly improed logistics, deliery or distribution methods, and

new or improed supporting actiities such as maintenance systems or

purchasing operations.

2. CIS data are missing or ar too many countries. Eery eort should be made to

ensure ull coerage or all CIS questions.

3. All countries should be encouraged to surey NACE sector 73 to improe the

measurement o innoation in KIBS.

Many other new indicators could be constructed using CIS data, such as a measure

o new to market innoations that controls or large dierences in what constitutes

a market20.

5.2. Socio-economic and regulatory environment

This section proides a summary o the thematic paper on socio-economic and

regulatory enironment21. Economic growth is at the heart o increases in peoples

well-being. Innoation including technological progress is one o the main sources

o economic growth. variations in economic growth and well-being can be partially

eplained by ariations in innoation perormance. This section seeks to identiy

actors that can help eplain why countries innoation perormance aries so

markedly.

20 See Arundel, A., Innoation Surey Indicators: What Impact on Innoation Policy?, in: Science, Technology and InnovationIndicators in a Changing World: Responding to Policy Needs, OECD, September 2007.

21 http://www.proinno-europe.eu/inde.cm?useaction=page.display&topicID=282&parentID=51


23/60

021

5.ThEMATICS

Preious EIS Thematic Papers the NIS 2003 and ExIS 2004 report hae

identied innoation categories and indicators which eplained ariations in

innoation perormance as measured by the Summary Innoation Inde (SII). This

section builds upon the ndings o the NIS 2003 and ExIS 2004 report and etends

the analysis to the 5 innoation dimensions as identied in the EIS: Innoation driers,

Knowledge creation, Innoation & entrepreneurship, Applications and Intellectual

property. Based on the ndings o the NIS 2003 report, the ExIS 2004 report, the

World Economic orums Global Competitieness Report 2006-2007 and the World

Banks Worldwide Goernance Indicators project 7 categories o policy indicators

Table 3 Relatie importance o socio-economic and regulatory enironment or eplaining dierences in innoation perormance

SIIInnovation

drivers

Knoledge

creation

Innovation

& entrepre-

neursip

Appli-

cations

Intellectual

property

DEMAND CONDITIONS

Youth share

Buyer sophistication

Goernment procurement

Demanding regulatory standards

SOCIAL CAPITAL

Trust

Perception o corruption

INSTITUTIONAL FRAMEwORK

Burden o administration

Quality o educational system

Intellectual property protection

Price stability

MARKET EFFICIENCY

Intensity o local competition

oreign ownership restrictions

leibility o wage determination

inancial market sophistication

TEChNOLOGY FLOwS

Brain drain

irm-leel technology absorption

Uniersity-industry research

collaboration

SOCIAL EQUITY

Social protection ependiture

Income equality

Employment rate

(INNOVATION) GOVERNANCE

voice and accountability

Political stability

Goernment eectieness

Regulatory quality

Rule o law

Control o corruption

: Strong correlation between ariation in indicator and innoation perormance; : Moderate correlation; : Weak correlation.


24/60

022

EUROPEAN

INNOvATION

SCOREBOARD2007


PERORMANCE

hae been identied coering 26 indicators. The eplanatory power o each o these

on the e dierent innoation dimensions was eplored using linear regressions

controlling or dierences in per capita GDP22. Table 3 summarises or each o the

innoation dimensions the eplanatory power o the indicators.

The main conclusions o the analysis are as ollows. The two categories that seem to

correlate best with dierences in oerall innoation perormance are social capitaland technology fows. These categories are also highly signicant or the Innoation

& entrepreneurship aspect o innoation perormance. This is important because

this aspect is not highly correlated with GDP, meaning that actors other than oerall

income leel are important in determining country perormance. This nding

suggests that policies that build trust and collaboration such as promoting

innoation networks and collaborations should be releant or countries at

arious income leels that under perorm on innoation and entrepreneurship.

Social capital and technology fows are also highly correlated with innoation

driers, but in this case the causality may be in the other direction. or eample

inestments in innoation driers (education, public research, broadband access)

may help build social capital which in turn improes technology fows and

innoatie perormance.The other e categories inestigated also appear to hae some infuence on oerall

innoation perormance, but here the linkages are less clear. Within the demand

category, the indicators or goernment procurement and demanding regulatory

standards appear to be most important, suggesting an important role or

goernment in raising innoation perormance through these mechanisms. These

indicators are not strongly correlated with any o the innoation dimensions,

suggesting that their impact is diused oer dierent parts o the innoation

process.

Most indicators o market eciency and the institutional ramework hae some

correlation with dierences in innoation perormance, o which price stability,

intensity o local competition and feibility o wage determination appear to be

the most important. This result might be related to the importance o macroeconomicstability and strong competition or raising innoation perormance. The indicator

or burden o administration is particularly correlated with the innoation driers

and innoation & entrepreneurship dimensions, suggesting the need or

goernments to reduce administratie burdens in order to oster innoation and

entrepreneurship.

The result or feibility o wage bargaining is more curious, particular as it is most

strongly correlated with the innoation driers dimension o innoation perormance.

Linked to this, the indicators o social equity also correlate relatiely strongly with

some dimensions o innoation perormance, with the notable eception o social

protection ependiture. There are no clear cut causal eplanations or this, but it is

consistent with earlier work (e.g. NIS paper) and could warrant urther

eamination.

There are some correlations between indicators o goernance and oerall

innoation perormance. This is particularly the case between goernment

eectieness and innoation driers, and to some etent or eplaining dierences

in innoation and entrepreneurship23.

It is noticeable that relatiely ew o the indicators correlate with the applications

dimension o innoation perormance (which includes employment in high tech

serices, eports o high tech products, sales o new to rm and o new to market

22 Correlation analyses show that innoation perormance measured by the SII and innoation perormance in each o theinnoation dimensions correlates moderately to highly with the leel o per capita GDP. By controlling or ariations inper capita GDP, we minimize the risk o so-called spurious correlations where two unrelated occurrences would show asignicant correlation coecient due to the a third, unseen actor, i.e. per capita GDP, which is correlated with each o

the two occurrences.23 See Celikel Esser, . 2007, The Link between Innoation Perormance and Goernance, JRC Scientic and TechnicalReports (JRC42104), or an analysis between innoation and goernance or a sample including many more non-EUcountries.


25/60

023

5.ThEMATICS

products, and employment in medium high and high tech manuacturing),

particularly as this is the dimension which is least correlated with GDP. The most

highly correlated indicator with applications is that or income equality. One

possible eplanation might be that more equal societies hae a higher demand or

innoatie products and serices, i.e. that income equality creates innoation

riendly demand conditions. Another eplanation is that this dimension o

innoation perormance is the most dicult to measure, and hence improements

in the indicators are needed beore causal actors can be properly identied.

5.3. Innovation efciency:linking inputs to outputs

This section proides a summary o the thematic paper on innoation eciency24.

ollowing the Lisbon strategy and the Barcelona target o an R&D intensity o 3%

in 2010, many countries hae taken steps to increase their innoation eorts.

Innoation eciency is related to the concept o productiity. Higher productiity

is achieed when more outputs are produced with the same amount o inputs or

when the same output is produced with less input. Innoation eciency will here

be dened similarly: innoation eciency is improed when with the same amounto innoation inputs more innoation outputs are generated or when less innoation

inputs are needed or the same amount o innoation outputs. Although innoation

is not a linear process where inputs automatically transer into outputs, it is

worthwhile to eamine dierences in eciency by assuming that eciency can be

dened as the ratio o outputs oer inputs. In the EIS the indicators are diided into

3 innoation input dimensions coering 15 indicators and 2 innoation output

dimensions coering 10 indicators25. Innoation eciency will be measured by

comparing the ratio between the composite indicator scores or one or more input

dimensions and one or more output dimensions. Inputs and outputs can be plotted

in a multidimensional space where the most ecient perormers will be on or close

to the eciency rontier. The larger the distance to this rontier, the smaller will

be the leel o innoation eciency26

. In a two-dimensional graph with inputs onone ais and outputs on the other ais, the rontier can be isualised as the

enelope cure connecting those dots with the most ecient output/input ratios.

In our analysis we hae employed a constant-returns-to-scale output-oriented DEA

(Data Enelopment Analysis27) on all combinations o the 3 input and 2 output

dimensions. Missing alues hae been imputed using the techniques used in the

2005 EIS Methodology Report. The analyses were done separately or the most

innoatie countries (Sweden, the innoation leaders and innoation ollowers) and

or the least innoatie countries (moderate innoators and catching-up countries).

Aerage eciency scores or both output dimensions are shown in igure 10.

All innovation leadersecept Sweden hae aboe aerage eciency in transorming

inputs into Applications. Despite its oerall leadership in innoation perormance,

Sweden has the lowest eciency in Applications o these countries indicating thatdespite its ery good oerall perormance it has room to make improements here.

Germany and Switzerland show high eciency in generating Intellectual property.

Some o the innoation leaders, in particular the UK, hae relatiely low eciency

in transorming inputs into Intellectual property outputs. This may because the

type o their innoation actiities does not lead to ormal IPRs but it could also

indicate that these countries could be creating more IPRs or their leel o inputs.

24 http://www.proinno-europe.eu/inde.cm?useaction=page.display&topicID=282&parentID=5125 Intellectual property, one o the output dimensions, can also be seen as an intermediate dimension with the reenues

earned rom the use o patents, trademarks and designs in the production process or the licensing o these representingthe nal output.

26 or an introduction into and more detailed discussions o eciency measures see Coelli, Timothy J., D.S. Prasada Rao,

Christopher J. ODonnell and George E. Battese, An Introduction into Eciency and Productiity Analysis, Springer, 2deedition, 2005.27 DEA inoles the use o linear programming methods to construct a non-parametric piece-wise surace (or rontier) oer

the data. Eciency measures are then calculated relatie to this surace. (Coelli et al., 2005, p.162).


26/60

024

EUROPEAN

INNOvATION

SCOREBOARD2007


PERORMANCE

The innovation followershae aboe aerage eciency in transorming inputs into

Applications, with Luembourg and Belgium showing highest eciency rates. Only

Austria, the Netherlands and Luembourg show aboe aerage eciency in

Intellectual property, and hence Belgium, rance and Iceland could seek to improe

their eciency rates by generating more IPRs rom their innoation inputs.

The moderate innovators show a range o dierent eciencies: we nd thesecountries in all our quadrants in igure 10 combining aboe or below aerage

eciency perormance. Italy combines aboe aerage eciency scores in both

output dimensions. This result suggests that it may be dicult or Italy to improe

its innoation perormance without increasing innoation inputs. Australia, Cyprus,

Norway and Spain show aboe aerage eciency in Intellectual property28 and the

Czech Republic shows aboe aerage eciency in Applications. Estonia and

Sloenia combine below aerage eciency in both Applications and Intellectual

property.

The catching-up countries also show a ariety o eciencies in transorming

innoation inputs into Applications. On Intellectual property eciency all countries

are signicantly below aerage with the eception o Portugal. This may be because

IPR is o less releance or the innoatie actiities o these countries or that thereis the potential to generate higher leels o IPR rom the eisting inputs. Some o

these countries are also still in a process o replacing national patent applications

by EPO patent applications. or Sloakia and Romania the eciencies or

Applications are relatiely high, suggesting that these countries need to increase

inputs to increase perormance in generating more Applications. The majority o

catching up countries hae below aerage eciencies and this suggests that or

these countries an important ocus should be improing innoation eciencies.

Based on their relatie position in igure 10, peer countries in eciency terms can

be identied as those countries with higher eciency scores in either Applications

or Intellectual property. or eample, Austrias possible peer countries include

Germany, Luembourg, the Netherlands and Switzerland, which combine higher

eciency scores in both Applications and Intellectual property. The innoation

28 We also hae to keep in mind that the eciency scores or the moderate innoators were calculated within the group oleast innoatie countries thus not including the innoation leaders and innoation ollowers.

Colour coding is conorm the groups

o countries as identied in the EIS

2007: bright green is Sweden, green

are the innoation leaders, yellow are

the innoation ollowers, orange are

the moderate innoators, blue are the

catching-up countries. The size o the

bubble gies the alue o the 2007Summary Innoation Inde (SII). The

dotted lines gie the unweighted

aerage o the eciency scores or the

EU27 Member States.

igure 10: Eciencies between innoation inputs and application and intellectualproperty outputs

FI

NL

FR

IS

US CASE

DK

JP

UK

ILIE

BESK

HU

LT

BGHR

SI

EL

PLLV

LUDE

CH

AT

IT

NO

AUCY

PTES

CZ

RO

EE

0

10

20

30

40

50

60

70

80

90

100

0 10 20 30 40 50 60 70 80 90 100

Efficiency in Applications

Efficiency

in

Intellectualpro

perty


27/60

025

5.ThEMATICS

policies implemented in these countries could be compared with those in Austria

to identiy options or policy improements to increase the eciency o transorming

innoation inputs into outputs29.

5.4. Non-R&D innovators

This section proides a preliminary summary o a orthcoming thematic paper on

non-R&D innoators30. Until recently R&D has been synonymous with technology

and innoation in many discussions on science, technology and innoation. Most

support measures or innoation on the national and the EU leel are or R&D

actiities. The Lisbon strategy, which aims to build Europe by 2010 the most

competitie and dynamic knowledge-based economy in the world, incorporates a

policy goal that the R&D ependiture in the European economies should reach 3

percent o GDP by 2010. As emphasized in the Lisbon strategy, R&D intensity is

etensiely used by scholars and policy makers as a benchmark or measuring the

innoatieness o a rm, an industry, a region and a country.

There is no doubt about the importance o R&D: it is the source o many

productiity enhancing innoations; it is essential to competitieness in ast-

growing high technology industries such as pharmaceuticals, it is critical to theabsorptie capacity o a rm or an industry and is associated with terms o trade

adantages o a country; and R&D actiities create demand and supply or high

skilled people which gie impetus to the deelopment o the education system in

a country.

Howeer, although R&D is ital or many innoation actiities o rms and the

competitieness o an industry and a country, the Community Innoation Surey

shows that almost hal o the European innoators do not conduct intramural or

in-house R&D (igure 11). Such non-R&D innoation includes the purchase o

adanced machinery and computer hardware specically purchased to implement

new or signicantly improed products or processes, the purchase o rights to use

29

The INNO-Policy Trendchart proides a database o innoation policies, see http://www.proinno-europe.eu/inde.cm?useaction=page.display&topicID=52&parentID=5230 http://www.proinno-europe.eu/inde.cm?useaction=page.display&topicID=282&parentID=51 (orthcoming January

2008)

igure 11: Share o innoators not perorming R&D

Share of non-R&D innovators

8 17 22 23 30 38 40 40 41 45 46 47 49 49 52 56 57 57 61 68 70 76 82 91

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

DK NL SE NO FR BE DE SK IT CZ E U MT LU EL UK PT H U E E ES RO LT CY PL BG

NON-R&D INNOVATORS R&D INNOVATORS

Results based on CIS-4 data. R&D

innoators are dened as all innoa-

tors perorming in house or intramural

R&D. Non-R&D innoators innoate

by acquiring or by buying etramural

R&D (i.e. R&D perormed by other

companies or research organisations),

by buying adanced machinery,

equipment and computer hardware

or sotware, by buying or licensing

patents and non-patented inentions,

by training their personnel, or by

spending resources on the design and

market introduction o new goods or

serices.


28/60

026

EUROPEAN

INNOvATION

SCOREBOARD2007


PERORMANCE

patents and non-patented inentions, licenses, know-how, trademarks and

sotware, internal or eternal training actiities or rms personnel aimed at the

deelopment or introduction o innoations, and internal and eternal marketing

innoations aimed at the market introduction o new or signicantly improed

products.31 The shares o non-R&D innoators tend to be higher in the new

Member States. Breaking down the data o non-R&D innoators by sector, we nd

that non-R&D innoators are concentrated in low technology manuacturing and

serice sectors. The distribution o these non-R&D innoators is also skewed

towards small and medium sized rms (or SMEs).

Non-R&D and R&D innoators are similar and dissimilar. The eect on innoation

actiities on the perormance o the enterprise is not that much dierent (Table 4),

but non-R&D innoators do consider uniersities and goernment research

institutes as less important sources o inormation or their innoation actiities.

Non-R&D innoators also introduce less products which are also new to their

market and the share o non-R&D innoators receiing public support rom their

central goernment or the EU is less than hal that o the R&D innoators. Both

non-R&D and R&D innoators ace almost the same barriers to innoation and

share similar objecties o innoation. The act whether or not a rm engages in

R&D is still an etremely important rm characteristic rom a policy perspectie as

R&D perormers are the target o most policy actions. A ailure to dierentiate

between non-R&D and R&D innoators reduces the eectieness o both

(academic) analyses o innoatie rms and the eectieness o public policies to

stimulate innoation.

Gien that a signicant number o rms innoate without any R&D, non-R&D

innoation actiities should hae drawn considerable attention rom academics and

policy makers. In act, the Oslo Manual proides a broad denition o innoation in

31 Non-R&D innoation is not the same as non-technological innoation. The latter includes organisational and marketing

innoations, where an organisational innoation is the implementation o new or signicant changes in rm structure ormanagement methods intended to improe a rms knowledge, quality o goods and serices or the eciency o workfows and a marketing innoation is the implementation o new or signicantly improed designs or sales methodsintended to increase the appeal o goods or serices or to enter new markets.

Table 4 Dierences between Non-R&D and R&D innoators

Non-R&Dinnovators

R&Dinnovators

Ratio

Percentage o frms:

Receiing unding rom local goernments 10 13 0.77

Receiing unding rom central goernment 5 16 0.33

Receiing unding rom the EU 3 8 0.44

Reported tat inormation source as used or innovation:

Internal sources within the enterprise 75 92 0.82

Internal sources other enterprises within the same group 16 28 0.59

Market sources suppliers 70 77 0.90

Market sources clients or customers 67 83 0.81

Market sources competitors 61 72 0.85

Institutional sources uniersities 21 45 0.46

Institutional sources research institutes 15 31 0.48

Other sources conerences, meetings 58 76 0.76

Other sources airs, ehibitions 68 81 0.85

Sales sare due to:

New to rm products 25 29 0.86

New to market products 5 10 0.54

Results based on CIS-3 data.


29/60

027

5.ThEMATICS

recognition o the acts that diusion is crucial to realizing the economic benets

o innoation and that R&D only coers a part o all o the dierent methods that

rms use to innoate. Howeer, there is lack o systematic studies on other means

that rms use to innoate and through research that links dierent types o

innoation to perormances o rms.

The Community Innoation Surey (CIS) collects only a limited amount oinormation on precisely how non-R&D innoators innoate. In order to proide

more statistical inormation on how these rms innoators, the Innobarometer (IB)

2007 surey was perormed to dele urther into the methods used by non-R&D

perorming rms to innoate and to see i one o the methods is based on user

drien innoation. The orthcoming EIS thematic paper on non-R&D based

innoation proides results based on an econometric analysis o the IB data.


30/60

028

EUROPEAN

INNOvATION

SCOREBOARD2007


PERORMANCE

Table5:ChangesintheEuropeanInnoationScoreboard

EIS20

00(Pilot)

EIS2001

EIS2002

EIS2003

EIS2004

EIS2005

EIS2006

EIS2007

Numberoindicators

16

18

18

22

22

26

25

25

Dissim

ilaritywithpreiousEIS

(28%)

3%

34%

14%

35%

4%

0%

Numberogroups/dimensions

4

4

4

4

4

5

5

5

Indica

torsbasedonCIS

4

4

4

5

6

7

7

7

Summ

aryInnoationInde

No

Yes

No

Yes

Yes

Yes

Yes

Yes

Countries

17:EU

15,US,JP

17:EU15,US,JP

33:EU25,US,JP,

IS,NO,CH

,BG,

RO,TR

33:EU25,US,JP,

IS,NO,CH,BG,

RO,TR

33:EU25,US,JP,

IS,NO,CH,BG

,

RO,TR

33:EU25,US,JP,

IS,NO,CH,BG,

RO,TR

34:EU25,US,JP,

IS,NO,CH,BG,

RO,HR,TR

37:EU27,US,JP,

IS,NO,CH,HR,

TR,AU,CA,IL

Input

Innoationdriers(EIS2005)

S&E(

ScienceandEngineering)graduates

Share

opost-

secondarygraduates

Shareopopulation

aged20-29

Share

oworking-agepopulationwith

tertiaryeducation

Broad

bandpenetrationrate

Participationinlie-longlearning

Youth

educationattainmentleel

Input

Knowledgecreation(EIS2005)

Public

R&Dependitures(%oGDP)

GOvERDonly

GOvERD+HERD

GERD

BERD

GOvERD+HERD

BusinessR&Dependitures(%oGDP)

Share

omedium-high/high-techR&Din

manu

acturing

Share

oenterprisesthatreceiepublic

undingorinnoation(CIS)

Share

ouniersityR&Dundedbypriate

sector

Input

Innoation&entrepreneurship

(EIS2

005)

Share

oSMEsinnoatingin-house(CIS)

Manuacturingsector

+

SericessectorTotalbusinesssector

Share

oSMEsco-operatingininnoation

(CIS)

Manuacturingsector

+


Inno

ationependitures(%oturnoer)

(CIS)

Manuacturingsector

+


ventu

recapital(%oGDP)

Earlystageand

epansionstage

Earlystageonly


31/60


32/60

030

EUROPEAN

INNOvATION

SCOREBOARD2007


PERORMANCE 6. Future Callenges

Since the 2000 pilot report, seen ull ersions o the European Innoation Scoreboard

hae been published. The list and number o indicators has undergone major

changes oer time as highlighted in Table 5. The number o indicators has increasedrom 18 to 25 and those deried rom the Community Innoation Surey rom 4 to

732. With major reisions in 2003 and 2005 (the dissimilarity percentages eceed 30

in both years), only 13 indicators eature in all Scoreboards. The number o countries

has increased to 37, although actual data aailability aries rom ery good (90% or

more) or most EU27 countries, Norway and Switzerland, to good or Bulgaria,

Cyprus, Latia, Sloenia, UK and Iceland (between 75% and 90%), to moderate or

US, Israel and Australia (between 60% and 70%) and to poor or Croatia, Turkey,

Japan and Canada (less than 60%). The EIS indicators are grouped in dierent

categories to capture key dimensions o the innoation process. In 2005 the current

e dimensions were introduced. Oerall innoation perormance is captured by a

composite inde, the Summary Innoation Inde, which has also been reised seeral

times, most recently in 2005 ollowing the EIS 2005 Methodology Report.

Current and past ersions o the EIS and accompanying thematic papers hae

continuously tried to improe measurement o innoation perormance by

countries, sectors and regions. uture editions o the EIS will hae to deal with a

number o eisting and new challenges under the ollowing our headings:

Measuringnewformsofinnovation

Assessingoverallinnovationperformance

Improvingcomparabilityatnational,internationalandregionallevels

Measuringprogressandchangesovertime

Across these areas, there is a need to maimise the releance and utility o the EIS

or policy makers, programme managers, and the wider innoation community.

Measuring ne orms o innovationThe changes in indicators and denitions o indicators used in the dierent EIS

reports all refect changes in our perception and understanding o the innoation

process33. Innoation is a comple phenomenon where rms can use dierent

models o innoation. Science-based innoation has been used by certain industries

and large rms or a long time. Innoation and technological progress is here

drien by rms by their new scientic discoeries. Innoation sureys were at rst

designed to measure science-based or R&D-based innoation. But new concepts o

the innoation process hae emerged. The model o user innoation, which was

introduced in the 1980s, states that consumers and end users deelop innoations.

More recently the model o open innoation has emerged: companies can no

longer rely on their own research but must instead combine own ideas and

research with eternal research e.g. by buying licenses and other eternalknowledge. Many o the current EIS indicators are better suited to capture science-

based innoation. Thereore, new indicators are increasingly required to better

capture new trends in innoation as portrayed in the models o user and in

particular open innoation, or eample on measuring knowledge fows.

Serices innoation is becoming more and more important as the relatie size o

the serices sector in the economy is continuously increasing. Innoation in serices

may dier rom that in manuacturing e.g. by greater use o marketing and

32 Also see Arundel, A. and H. Hollanders, Innoation Scoreboards: Indicators and Policy Use, in C. Nauwelaers and R.Wintjes (eds.), Innoation Policy in Europe, Edward Elgar: Cheltenham, 2008 or a history o the EIS and a comparisonwith other (innoation) scoreboards.

33 Alternatie indicators and approaches to measure innoation were eplored in two thematic papers in 2003 and 2004.

The 2003 NIS thematic report inestigated arious structural and socio-cultural indicators and their impact on acountrys innoation perormance. The 2004 ExIS 2004 thematic report deeloped an alternatie scoreboard with aocus on innoation at the rm-leel including a more dierse range o non-technological innoatie actiities (e.g.market and organisational innoation). This approach is ollowed up in the 2007 thematic report on innoation andsocio-economic and regulatory enironment.


33/60

031

6.FUT

UREChALLENGES

organisational innoation. Also serice innoations may be increasingly prealent

within manuacturing sectors. Current statistics and innoation policies are biased

towards measuring technological innoation and thereore new deelopments in

both statistics and policies may be needed or better understanding and stimulating

non-technological innoation.

To improe the measurement o new orms o innoation in uture editions o theEIS we need to deelop and implement new indicators measuring e.g. open

innoation, user innoation and non-R&D innoation. New indicators can draw on

new data, in particular the improed measurement on marketing and organisational

innoation and serices innoation in the latest editions o the Community

Innoation Surey, but more improements are needed to ully capture all

innoation process in the European economies.

Assessing overall innovation perormanceThe EIS proides a composite inde, the Summary Innoation Inde, which

summarises innoation perormance by aggregating the arious indicators or each

country in one single number. The 2005 Methodology Report studied in detail

alternatie computation schemes or the SII, but recent deelopments in compositeindicator theory may call or changes in the scheme. The SII transorms each

indicator on a relatie basis, i.e. each indicator is measured relatie to the best and

worst perorming country. Some o the indicators are highly skewed, e.g. patent

applications. The question emerges whether or not to transorm the indicators as

or many o the indicators the distribution o the data dier rom the normal

distribution on which composite indicator theory is based.

In addition, the EIS proides innoation perormance by 5 groups o indicators, the

innoation dimensions. This h

Documents

European Innovation Scoreboard 2007