The ambitious goals of the European “Digital Agenda” need active involvement by regional innovation systems. Effective regional “digital strategies” should be both consistent with the European framework and based on available evidence on the needs and opportunities of local contexts. Such evidence should be used to balance the different components of the Information Society development (e.g. eServices vs. infrastructures; ICT supply and demand), so as to ensure that they can all unleash their full potential. Therefore, EU regions should spend more money to overcome regional weaknesses than to improve existing assets. In this paper we explore the different strategies of the EU’s lagging regions through the analysis of the allocation of 2007-13 Structural Funds. Then, we verify whether such strategies respond to territorial conditions by comparing strategic choices made with the actual characteristics of local contexts. Results show that EU regions tend to invest more resources in those aspects in which they already demonstrate good relative performances. Possible causes of this unbalanced strategic approach are discussed, including the lack of sound analysis of the regional context and the path dependence of policy choices.

1.
Are
EU
regional
digital
strategies
evidence-­‐based?
An
analysis
of
the
alloca8on
of
2007-­‐13
Structural
Funds
Luigi
Reggi
and
Sergio
Scicchitano
Ministry
of
Economic
Development,
Department
for
Development
and
Economic
Cohesion,
Italy
*
EIBURS-­‐TAIPS
team,
University
of
Urbino
luigi.reggi@tesoro.it
sergio.scicchitano@tesoro.it
Regional
Innova4on
and
Compe44veness
Policy
Workshop
15
November
2012
UK-­‐Innova4on
Research
Centre
-­‐
University
of
Cambridge
*
The
views
expressed
here
are
those
of
the
authors
and,
in
parFcular,
do
not
necessarily
reflect
those
of
the
Ministry
of
Economic
Development

2. Outline
• IntroducFon
• Research
objecFves
• Regional
digital
strategies
&
EU
Cohesion
policy
• Data
and
methodology
• Results
• Discussion
and
Conclusions

3. Introduc4on
(1/2)
• ICT
diffusion
foster
producFvity
and
growth
•
“Digital
Agenda
for
Europe”
>
a
strategic
framework
for
Informa(on
Society
(IS)
development
(2010)
• A
“sustained
level
of
commitment”
at
the
regional
level
is
required
– increased
powers
in
the
field
of
InnovaFon
policy
(OECD,
2011)
– Regional
InnovaFon
Systems
(Cooke
and
Morgan,
1998;
EC,
1998)
– Within
IS
>
regions
play
an
intermediaFng
role
between
EU
frameworks
and
local
intervenFons
(Ca^aneo,
2004)

4. Introduc4on
(2/2)
• Regional
planning
for
IS
must
be
guided
by
effecFve
Regional
“digital
strategies”,
that
should
be
both:
1. consistent
with
the
EU
policy
framework
2. place-­‐based,
i.e.
based
on
available
evidence
on
the
characterisFcs
of
local
contexts
in
terms
of
IS
development
(Barca,
2009;
Tsipouri,
2002)

5. Research
objec4ves
1. InvesFgate
the
existence
of
different
regional
digital
strategies
2. Verify
whether
such
strategies
respond
to
local
territorial
condiFons

6. Regional
digital
strategies
&
the
EU
Cohesion
policy
• European
Cohesion
(or
Regional)
Policy
offers
an
ideal
opportunity
to
explore
the
key
elements
of
regional
digital
strategies
with
a
quanFtaFve
approach
shared
rules
and
regula(ons
=>
funding
is
allocated
and
classified
through
common
categories
and
definiFons,
and
that
data
on
the
financial
distribuFon
of
resources
is
fully
comparable.

7. Regional
digital
strategies
&
the
EU
Cohesion
policy
The
EC
has
put
a
great
emphasis
on
the
need
to
adopt
comprehensive
and
balanced
regional
strategies
for
the
InformaFon
Society
development
• 1994-­‐1999
>
Regional
InformaFon
Society
IniFaFve
(RISI)
• 2000-­‐2006
>
5.5
billion
€
(Vincente
and
Lopez,
2011).
• 2007-­‐2013
>
15.3
billion
€
(Reggi
&
Scicchitano,
2012)

8. "
Fig. 5. Structural funds dedicated to Information Society development in 2007-13 period.
Source: Own elaboration from European Commission - DG for Regional Policy data
Source: Own elaboration from European Commission - DG for Regional Policy data
27"
27"
"

9. Regional
digital
strategies
&
the
EU
Cohesion
policy
POLICY
RECOMMENDATIONS
FROM
THE
LITERATURE
1. Regional
digital
strategies
should
adopt
a
holis4c
and
integrated
perspecFve
(e.g.
public
&
private
actors)
(Tsipouri,
2002)
2. Regional
digital
strategies
should
be
based
on
a
sound
analysis
of
the
implementa4on
context
and
the
territory-­‐specific
condiFons,
needs
and
opportuniFes
(Tsipouri,
2002,
Technopolis,
2006)

10. Regional
digital
strategies
&
the
EU
Cohesion
policy
POLICY
RECOMMENDATIONS
FROM
THE
LITERATURE
3. A
well-­‐balanced
approach
is
needed,
since
“IS
development
involves
parallel,
mutually
reinforcing
developments
in
a
range
of
fields”
(Taylor
and
Downes,
2001).
Examples
>
Telecommunica(on
infrastructures
vs.
eServices
development
Supply
vs.
Demand
of
ICT

11. Data
&
methodology
(1/3)
• The
data
on
the
allocaFon
of
EU
Structural
Funds
are
based
on
financial
resources
programmed
by
each
2007-­‐13
Opera4onal
Programme
(OP)
of
the
EU
Cohesion
Policy.
• The
dataset
is
provided
by
the
European
Commission
–
22
DG
Regional
Policy
and
includes
data
on
the
amount
of
L. Reggi and S. Scicchitano
programmed
financial
resources
by
“category
of
Table 1 Categories of expenditure dedicated to IS and public e-Services and financial resources
expenditure”:
allocated in both CONV and COMP objectives
N. Name A.V. %
10 Broadband networks 2,257,722,464 15%
11 þ 12 Information and communication technologies 4,121,115,554 27%
(interoperability, security, etc.)
13 Services and applications for citizens 5,225,072,351 34%
14 Services and applications for SMEs 2,144,358,160 14%
15 Other measures for improving use of ICT by SMEs 1,537,162,147 10%
Total 15,285,430,676 100%
11

12. Data
&
methodology
(2/3)
A
two
step
procedure
is
adopted.
1. EU
regions
classified
into
homogeneous
groups
through
a
Principal
Component
Analysis
(PCA)
followed
by
a
hierarchical
Cluster
Analysis
(CA),
on
the
basis
of
the
allocaFon
of
Structural
Funds
(in
%)
to
strategic
IS
objecFves
in
2007-­‐13
period.
2. logit
model
to
examine
the
link
between
the
strategic
choices
idenFfied
in
the
first
step
and
the
characterisFcs
of
regional
contexts
in
terms
of
socio-­‐economic
development
and
IS
diffusion.

13. Data
&
methodology
(3/3)
Our
analysis
is
limited
to
the
lagging
regions
belonging
to
the
CONV
objec4ve
where
Structural
Funds
can
be
considered
as
a
proxy
of
the
total
amount
of
financial
resources
that
a
region
can
invest
in
IS
development.
– CONV
regions
benefit
from
a
unprecedented
amount
of
2007-­‐13
EU
Cohesion
resources
dedicated
to
innovaFon
(Bonaccorsi,
2010;
Reggi
&
Scicchitano,
2012)
– CONV
regions
tend
to
spend
the
few
locally
available
resources
to
maintain
the
current
levels
of
basic
public
services
13

14. """""""""""""""""""""""
Results:
1st
step
–
PCA+CA
Fig. 4. Identifying three strategies in allocating financial resources for IS
evelopment in CONV Regions
Dimension 2 - 27.21 %
3.0
e-Services
1.5 CLUSTER 2
29%
0
SME2
29%
CLUSTER 1 broadband
ICT
CLUSTER 3 41%
SME1
-1.5
-3.0 -1.5 0 1.5
Dimension 1 - 36.90 %

15. Results:
2nd
step
-­‐
logit
model
• We
esFmate
three
logit
models
separately
for
each
cluster.
• The
observed
dependent
variable
is
binary,
taking
the
value
of
one
if
a
region
belongs
to
one
of
the
three
clusters
and
zero
otherwise.
• Most
of
the
variables
(with
2
excepFons)
are
at
the
NUTS2
(regional)
level
• Depending
on
data
availability,
most
of
regressors
are
from
2006,
the
year
when
regional
strategies
were
approved;
from
2007
otherwise.

16. Results:
2nd
step
-­‐
logit
model
MODEL
REGRESSORS:
Eurostat
variables
on
IS
local
development
i)
Households
with
broadband
access
ii)
Households
with
access
to
the
Internet
iii)
Individuals
who
ordered
goods
or
services
over
the
Internet
for
private
use
iv)
Individuals
using
the
Internet
for
interacFon
with
public
authoriFes
v)
Enterprises
who
have
ERP
sooware
package
to
share
informaFon
on
sales/
purchases
with
other
internal
funcFonal
areas
CONTROL
VARIABLES:
vi)
Gross
DomesFc
Product
(GDP)
per
capita
vii)
Number
of
local
units
viii)
Number
of
employees
in
local
units
ix)
Total
intramural
R&D
expenditure
in
the
higher
educaFon
sector
x)
Rate
of
unemployment

17. Results:
2nd
step
-­‐
logit
model
Table 5
Determinants of regional strategies on IS and e-services: Logit results
Var Cluster 1 Cluster 2 Cluster 3
Coef. Std. Err. z Coef. Std. Err. z Coef. Std. Err. z
Broad_house -1.143005 0.4492695 -2.54** 0.0564842 0.1604883 0.35 0.4275693 0.1313482 3.26***
Internet_house 1.142004 0.448562 2.55** -0.1895048 0.1902209 -1 -0.3502617 0.1241232 -2.82***
Order_indiv -0.3475656 0.124287 -2.8*** 0.1648801 0.107555 1.53 0.1935861 0.0728091 2.66***
Interact_indiv 0.5869741 0.2845014 2.06** 0.3405171 0.2055405 1.66* -0.4194555 0.1579136 -2.66***
Integr_process -1.371032 0.5737002 -2.39** 0.406413 0.1739903 2.34** 0.2302201 0.098529 2.34**
GDP_percapita 0.0000899 0.0002822 0.32 -0.000795 0.0003868 -2.06** 0.000012 0.0001531 0.08
Local_units 0.0030452 0.0013319 2.29** 0.0029206 0.0023873 1.22 -0.0035533 0.0011161 -3.18***
Employees -0.0008222 0.0003594 -2.29** -0.0008053 0.0007944 -1.01 0.0009812 0.0003352 2.93***
R&D 0.0448095 0.0290908 1.54 0.0209459 0.026945 0.78 -0.0387537 0.0197503 -1.96**
Unemp 0.4254679 0.2084043 2.04** -0.2785523 0.1644312 -1.69** 0.0813006 0.0633516 1.28
Cons -6.153135 4.415426 -1.39 -1.577808 3.686809 -0.43 2.023108 2.100155 0.96
Number of obs. 70 70 70
LR chi2(10) 57.68 30.16 37.17
Log likelihood -14.734982 -11.777133 -29.475342
Prob> chi2 0.0000 0.0008 0.0001
Pseudo R-sq 0.6618 0.5615 0.3867
Note: Significance levels are as follows: * p<0.10, ** p<0.05, *** p<0.01.

18. Discussion
(1/3)
• Lagging
Regions
seek
to
further
improve
their
strengths
rather
than
focus
on
the
weaknesses
that
emerge
from
the
regional
IS
context.
• This
seems
to
explicitly
contradict
policy
recommenda4ons
in
the
literature
-­‐
WHY?
-­‐

19. Discussion
(2/3)
1. Inadequate
analysis
of
local
context
• qualita4ve
tools
prevailing
in
policy
evaluaFon,
no
quanFtaFve
impact
analysis
• no
real
benchmarking
• few
available
relevant
indicators
on
IS
at
regional
level

20. Discussion
(3/3)
2. Path
dependence,
the
idea
that
insFtuFonal
life
is
ooen
characterized
by
posiFve
feedback
processes
that
make
change
costly.
Example:
mulF-­‐annual
investment
decisions
to
implement
large
telecommunicaFon
infrastructures
3. Rules
of
Structural
Funds
(“N+2”)
Concentrate
resources
on
on-­‐going
projects
that
ensure
immediate
spending

21. In
conclusion...
EU
Regions
need
to
rebalance
most
of
their
current
“digital
strategies”
This
is
parFcularly
important
considering
that,
according
to
the
current
proposal
of
the
European
Commission,
a
balanced
strategy
for
the
IS
is
now
an
“ex-­‐ante
condiFonality”
for
accessing
2014-­‐2020
Structural
Funds

22. Future
research
– Compare
the
allocaFon
of
programmed
resources
with
actual
expenditure
– Extend
the
analysis
by
including
Structural
Funds
allocaFon
to
Research
&
InnovaFon,
Human
Capital
and
CompeFFveness

23.
..Thank
you
for
your
aNen(on!
Luigi
Reggi
luigi.reggi@tesoro.it
luigi.reggi@gmail.com
www.luigireggi.eu