ArticlePDF Available

Public Procurement for Innovation as mission-oriented innovation policy

Authors:

Abstract and Figures

This article focuses on Public Procurement for Innovation as a relevant demand-side instrument to be exploited in the mitigation of grand challenges. It intends to provide some clarification on what should (and what should not) be regarded as innovation procurement. It defines what is meant by Public Procurement for Innovation and categorizes it according to three dimensions: (i) the user of the purchased good; (ii) the character of the procurement process; and (iii) the cooperative or non-cooperative nature of the process. In addition, it illustrates the main stages in innovation procurement processes and exemplifies them with six cases to provide evidence that Public Procurement for Innovation can contribute to satisfying unsatisfied human needs and solving societal problems.
Content may be subject to copyright.
Research
Policy
41 (2012) 1757–
1769
Contents
lists
available
at
SciVerse
ScienceDirect
Research
Policy
jou
rn
al
h
om
epage:
www.elsevier.com/locate/respol
Public
Procurement
for
Innovation
as
mission-oriented
innovation
policy
Charles
Edquista,b,, Jon
Mikel
Zabala-Iturriagagoitiaa
aCentre
for
Innovation,
Research
and
Competence
in
the
Learning
Economy
(CIRCLE),
Lund
University,
Sweden
bInstitut
Francilien
Recherche
Innovation
Société
(IFRIS),
Université
Paris-Est,
Paris,
France
a
r
t
i
c
l
e
i
n
f
o
Article
history:
Received
16
May
2011
Received
in
revised
form
28
March
2012
Accepted
3
April
2012
Available online 16 October 2012
Keywords:
Public
procurement
for
innovation
Innovation
policy
Functional
specification
Interactive
learning
a
b
s
t
r
a
c
t
This
article
focuses
on
Public
Procurement
for
Innovation
as
a
relevant
demand-side
instrument
to
be
exploited
in
the
mitigation
of
grand
challenges.
It
intends
to
provide
some
clarification
on
what
should
(and
what
should
not)
be
regarded
as
innovation
procurement.
It
defines
what
is
meant
by
Public
Pro-
curement
for
Innovation
and
categorizes
it
according
to
three
dimensions:
(i)
the
user
of
the
purchased
good;
(ii)
the
character
of
the
procurement
process;
and
(iii)
the
cooperative
or
non-cooperative
nature
of
the
process.
In
addition,
it
illustrates
the
main
stages
in
innovation
procurement
processes
and
exem-
plifies
them
with
six
cases
to
provide
evidence
that
Public
Procurement
for
Innovation
can
contribute
to
satisfying
unsatisfied
human
needs
and
solving
societal
problems.
© 2012 Published by Elsevier B.V.
1.
Introduction
Public
organizations
may
place
an
order
for
something
(nor-
mally
a
product
or
a
system)
that
does
not
exist;
hence,
this
“something”
has
to
be
developed
by
the
supplier
before
it
can
be
delivered.
In
other
words,
innovations
are
needed
before
deliv-
ery
can
take
place.
Until
about
10
years
ago
this
phenomenon
was
called
“public
technology
procurement”
(Edquist
et
al.,
2000a).
Since
then,
this
vocabulary
of
the
1990s
and
earlier
has
changed.
The
concept
of
“technology”
has
been
replaced
by
the
concept
of
“innovation”,
reflecting
a
widening
of
the
content
of
the
notion.
It
is
a
matter
of
using
public
demand
(or
similar)
to
trigger
innovation.
We
will
use
the
term
“Public
Procurement
for
Innovation”
(PPI)
to
denote
this
phenomenon.
However,
the
(non-existing)
product
ordered
in
the
process
of
PPI
is
not
the
beginning
of
the
process
or
its
objective.
Instead,
the
rationale
for
PPI
is:
To
satisfy
human
needs,
and/or
To
solve
societal
problems.
This
is
why
PPI
is
so
relevant
in
the
context
of
grand
challenges,
the
idea
being
to
mitigate
these
challenges
through
the
kind
of
innovation
policy
instrument
that
we
call
PPI.
However,
the
nature
of
grand
challenges
such
as
global
warming,
tightening
supplies
of
Corresponding
author
at:
Centre
for
Innovation,
Research
and
Competence
in
the
Learning
Economy
(CIRCLE),
Lund
University,
Sweden.
Tel.:
+46
46
22
239
31.
E-mail
address:
Charles.Edquist@circle.lu.se
(C.
Edquist).
energy,
water
and
food,
ageing
societies,
public
health,
pandemics
or
security
(Lund
declaration,
2009)
does
not
allow
defining
poli-
cies
to
target
them
as
a
whole,
at
the
same
time
and
with
only
one
policy
instrument.
As
it
is
simply
not
possible
to
work
at
such
lev-
els
of
aggregation,
policies
need
to
address
narrower
targets
and
partial
problems
linked
to
those
grand
challenges.
This
is
in
fact
reflected
in
the
use
of
PPI
for
meeting
human
needs
and
mitigat-
ing
societal
problems,
where
more
limited
goals
are
set
for
those
programs
(e.g.
energy
saving
of
a
certain
kind,
improving
mobil-
ity
with
regard
to
passenger
transport,
increasing
security
in
a
specific
field).
We
would
argue
that
most
mission-oriented
policy
mitigation
of
grand
challenges
has
and
must
have
a
narrower
focus
as
compared
to
the
grand
challenges
as
such,
simply
because
they
are
so
“grand”.
It
should
be
added
that
PPI
certainly
includes
innovations
intended
to
meet
needs
(‘missions’)
of
public
agencies
themselves,
if
they
are
related
to
general
human
needs
or
soci-
etal
problems
(see
direct
PPI
in
Section
3
and
four
of
the
cases
in
Section
5).
Needless
to
say,
grand
challenges
can
also
be
mitigated
through
other
means
and
instruments,
for
example
R&D
funding,
tax
credits,
environmentally
motivated
regulations
and
standards
(e.g.
mileage
standards
for
automobiles),
creation
of
markets
for
innovative
ideas,
support
for
education
and
training
or
enhancing
capacities
for
knowledge
exchange
(OECD,
2011).
Nonetheless,
instruments
other
than
PPI
will
not
be
addressed
here,
except
for
brief
refer-
ences
in
cases
when
they
are
closely
combined
with
PPI
in
the
“policy-mix”
(Flanagan
et
al.,
2011).
A
new
interest,
at
the
European
level,
has
recently
emerged
with
regard
to
demand-side
approaches
to
innovation
policy
(Edquist
and
Hommen,
1999)
and,
more
specifically,
the
use
of
public
0048-7333/$
see
front
matter ©
2012 Published by Elsevier B.V.
http://dx.doi.org/10.1016/j.respol.2012.04.022
1758 C.
Edquist,
J.M.
Zabala-Iturriagagoitia
/
Research
Policy
41 (2012) 1757–
1769
demand
as
an
engine
for
the
development
and
diffusion
of
inno-
vations.
In
early
2004
three
governments
issued
a
position
paper
to
the
European
Council
calling
for
the
use
of
public
procure-
ment
across
Europe
to
spur
innovation
(Edler
and
Georghiou,
2007;
French/German/UK
Governments,
2004).
This
development
con-
tinued
and
was
manifested
in
various
reports,
including
the
Aho
Group
Report
(Aho
et
al.,
2006).
The
Aho
Group
identified
several
application
areas
or
grand
challenges
where
demand-side
poli-
cies
could
be
used
to
a
larger
extent:
e-Health,
Pharmaceuticals,
Energy,
Environment,
Transport
and
Logistics,
Security
and
Digi-
tal
Content
(Edler
and
Georghiou,
2007,
p.
951).
There
seems
to
be
much
less
talk
about
innovation
procurement
in
the
US
than
in
Europe
(Vonortas
et
al.,
2011).
We
suspect,
however,
that
other
fields
of
US
government
policy
involving
“mission
agencies”,
such
as
defense
or
energy,
incorporate
elements
of
PPI.
Discussion
of
the
differences
between
European
and
US
practices
regarding
PPI
seems
to
conflate
the
use
of
procurement
to
meet
societal
chal-
lenges
with
the
use
of
procurement
to
meet
mission-agency
needs.
PPI
in
the
US
has
often
been
used
for
the
latter
and
less
frequently
for
the
former
(Thai,
2001).
It
has
also
been
used
for
mission-agency
needs
in
Europe
and
elsewhere.
The
aim
of
this
article
is
to
contribute
to
clarifying
the
charac-
teristics
of
(different
kinds
of)
PPI,
how
PPI
has
been
used,
as
well
as,
briefly,
its
relationship
to
other
public
innovation
policy
instru-
ments.
We
have
therefore
decided
to
base
this
article
firmly
on
empirical
experiences
by
using
a
number
of
examples
of
PPI.
Sec-
tion
5
contains
six
case
descriptions
of
PPI.
Before
that,
the
context
for
PPI
is
presented
in
Section
2.
A
few
definitions
that
are
necessary
to
structure
and
characterize
the
cases
are
introduced
in
Section
3.
The
methodology
and
the
dimensions
of
the
cases
are
found
in
Sec-
tion
4,
as
is
a
detailed
summary
of
the
case
descriptions
(Table
2).
The
conclusions
and
policy
implications
are
addressed
in
Section
6.
2.
Interaction
in
innovation
systems
Innovation
processes
occur
over
time
and
are
influenced
by
many
factors.
Because
of
this
complexity,
firms
almost
never
inno-
vate
in
isolation,
but
interact
with
other
organizations
to
gain,
develop,
and
exchange
various
kinds
of
knowledge,
information
and
other
resources.
These
interactions
among
organizations
(players)
operating
in
different
institutional
contexts
are
important
for
innovation
pro-
cesses
(Edquist,
2011).
What
we
call
‘activities’
in
the
Systems
of
Innovation
(SI)
are
the
determinants
of
the
development
and
diffusion
of
innovations.
Examples
of
activities
are
Research
and
Development
(R&D),
the
financing
of
the
commercialization
of
such
knowledge,
or
demand-side
activities
such
as
the
formation
of
new
product
markets
or
the
articulation
of
new
product
quality
require-
ments.
Hence,
the
development
and
diffusion
of
innovations
are
highly
influenced
by
demand
that
may
emanate
from
either
private
or
public
organizations
(players).1This
article
will
disregard
the
influ-
ence
from
private
organizations
(e.g.
demand
from
customer
firms
or
individual
consumers)
and
only
address
the
demand
from
public
organizations.
PPI
is
an
important
demand-side
innovation
policy
instrument
(Dalpé,
1994;
Edler
and
Georghiou,
2007;
Geroski,
1990;
Rothwell
and
Zegveld,
1981),
and
from
now
on
this
article
will
concentrate
on
PPI.
Hence,
it
is
generic
with
regard
to
research
areas
or
grand
challenges;
it
deals
with
one
innovation
policy
instrument
that
can
potentially
be
used
to
mitigate
many
different
challenges.
As
we
1Public
means
that
the
activity
is
performed
by
an
agency
or
organization
(player)
that
is
a
part
of
the
local
or
municipal
authorities,
the
regional
authorities,
the
national
state,
or
supranational
bodies.
have
seen,
and
will
see,
PPI
is
one
example
of
interaction
between
organizations
(procurers
and
suppliers),
which
is
strongly
stressed
as
a
source
of
innovation
in
the
SI
approach.
3.
Defining
and
classifying
innovation
procurement
Innovations
are
new
creations
of
economic
or
societal
signifi-
cance
mainly
carried
out
by
firms
(but
not
in
isolation).
They
may
be
new
products
or
new
processes.
New
products
may
be
material
goods
or
intangible
services;
it
is
a
matter
of
what
is
produced.
New
processes
may
be
technological
or
organizational;
it
is
a
matter
of
how
the
products
are
produced.
Non-firm
public
organizations
do
not
normally
take
part
directly
in
the
innovation
processes,
although
they
certainly
are
important
organizations
participating
in
the
research
and
invention
activities
that
influence
innovation.
They
affect
(change,
reinforce,
improve)
the
context
in
which
the
innovating
firms
operate.
As
indicated
in
Section
2,
this
context
includes
all
the
determinants
of
innovation
processes.
Innovation
policy
may
thus
be
understood
as
actions
by
public
organizations
that
influence
innovation
processes,
i.e.
the
development
and
diffusion
of
innovations
(Edquist,
2011).2
Public
procurement
means
that
a
public
organization
buys
a
product
(a
good
or
a
service
or
a
combination
of
the
two,
which
might
be
called
a
system).
Public
Procurement
for
Innovation
(PPI)
occurs
when
a
public
organization
places
an
order
for
the
ful-
fillment
of
certain
functions
within
a
reasonable
period
of
time
(through
a
new
product).3Hence,
the
objective
(purpose,
ratio-
nale)
of
PPI
is
not
primarily
to
enhance
the
development
of
new
products,
but
to
target
functions
that
satisfy
human
needs
or
solve
societal
problems.
We
must
point
out
here
that
the
diffusion
of
the
product
from
the
procuring
organizations
is
not
always
among
the
major
objectives
of
this
type
of
program.
However,
there
are
cases
in
which
diffusion
of
the
new
product
is
aimed
at
from
the
very
start
of
the
procurement
process.
This
difference
reflects
the
dis-
tinction
between
PPI
carried
out
mainly
for
the
missions
or
needs
of
the
procuring
agency
and
PPI
to
support
economy-wide
innova-
tion.
Be
that
as
it
may,
innovation
is
needed
in
all
PPI
before
delivery
can
take
place.
In
contrast
to
PPI,
regular
procurement
occurs
when
public
agencies
buy
ready-made
products
such
as
pens
and
paper
“off-the-shelf”,
where
no
innovation
is
involved.
Only
the
price
and
quality
of
the
(existing)
product
are
taken
into
consideration
when
the
supplier
is
selected.
We
will
now
present
a
taxonomy
of
different
phenomena
that
are,
or
should
be,
labeled
PPI
or
innovation
procurement
(Edler,
2009;
Edquist
et
al.,
2000a;
Hommen
and
Rolfstam,
2009;
Uyarra
and
Flanagan,
2010).
The
taxonomy
will
be
used
to
classify
the
cases
presented
later.
The
first
dimension
refers
to
whom
the
user
of
the
resulting
product
(good,
service,
system,
etc.)
is,
which
we
can
then
use
to
identify
two
different
categories
of
PPI:
direct
and
catalytic.
Direct
PPI
is
when
the
procuring
organization
is
also
the
end-
user
of
the
product
resulting
from
the
procurement.
The
buying
agency
simply
uses
its
own
demand
or
need
to
influence
or
induce
2This
implies
that
innovation
policy
also
includes
actions
by
public
organizations
that
unintentionally
affect
innovation.
3The
public
organization
may
also
financially
contribute
directly
to
the
R&D
lead-
ing
to
the
development
of
the
product.
However,
such
contributions
are
not
intrinsic
parts
of
the
PPI
as
such.
Public
R&D
funding
is
a
different
– complementary
pol-
icy
instrument
in
the
policy-mix,
one
which
is
not
in
focus
here.
The
purchase
of
a
non-existing
product
is
the
central
element
of
PPI.
However,
the
development
costs
of
the
new
product
are,
of
course,
indirectly
supported
by
the
procurer
by
(initially)
paying
a
high
price
for
the
product.
This
is
part
of
the
very
idea
of
PPI,
but
since
the
procurers’
commitment
is
only
to
buy
a
number
of
units
of
the
product
at
a
certain
price,
this
support
of
the
development
cost
is
brought
about
through
the
product
price
mechanism
and
cannot
be
regarded
as
direct
public
R&D
funding.
C.
Edquist,
J.M.
Zabala-Iturriagagoitia
/
Research
Policy
41 (2012) 1757–
1769 1759
innovation;
this
type
of
PPI
includes
the
procurement
undertaken
to
meet
the
(‘mission’)
needs
of
the
public
agencies
themselves.
However
the
resulting
product
is
often
also
diffused
to
other
users.
Hence,
innovations
resulting
from
PPI
can
be
useful
for
the
performing
agencies,
as
well
as
for
society
as
a
whole.
Catalytic
PPI
is
when
the
procuring
agency
serves
as
a
catalyst,
coordinator
and
technical
resource
for
the
benefit
of
end-users.
The
needs
are
located
‘outside’
the
public
agency
acting
as
the
‘buyer’.
Hence,
the
public
agency
aims
to
procure
new
products
on
behalf
of
other
actors.
It
acts
to
catalyse
the
development
of
innovations
for
broader
public
use
and
not
for
directly
supporting
the
mission
of
the
agency.
The
second
dimension
refers
to
the
character
of
the
result
of
the
procurement
process,
i.e.
the
character
of
the
innovation
(if
any)
embedded
in
the
resulting
product,
according
to
which
three
types
may
be
distinguished:
pre-commercial,
adaptive
and
developmen-
tal
procurement.
Pre-commercial
procurement
(PCP)
refers
to
the
procurement
of
(expected)
research
results
and
is
a
matter
of
direct
public
R&D
investments,
but
no
actual
product
development.
Moreover,
it
does
not
involve
the
purchase
of
a
(non-existing)
product,
and
no
buyer
of
such
a
product
is
therefore
involved.
This
type
of
procurement
may
also
be
labeled
“contract”
research,
and
may
include
development
of
a
product
prototype.
Adaptive
PPI
is
when
the
product
or
system
procured
is
incre-
mental
and
new
only
to
the
country
(or
region)
of
procurement.
Hence,
innovation
is
required
in
order
to
adapt
the
product
to
spe-
cific
national
or
local
conditions.
It
may
also
be
labeled
‘diffusion
oriented’
or
‘absorption
oriented’
PPI.
Developmental
PPI
implies
that
completely
new-to-the-world
products
and/or
systems
are
created
as
a
result
of
the
procure-
ment
process.
It
can
be
regarded
as
‘creation
oriented’
PPI
and
involves
radical
innovation.
Let
us
initially
focus
on
the
first
of
these
types
of
procure-
ments.
A
recent
communication
from
the
European
Commission
addresses
the
concept
of
PCP,
which
concerns
the
acquisition
of
expected
research
results,
i.e.
an
R&D
phase
that
may
precede
com-
mercialization
(European
Commission,
2007,
p.
6).
However,
the
commercial
development
of
new
products
is
not
part
of
the
PCP
as
such
(see
Art
16f
of
2004/18/EC).
Accordingly,
no
actual
product
development
and
no
buyer
of
such
a
product
are
involved
in
PCP.
The
product
prototype
that
may
result,
may
be
possible
to
commer-
cialize
or
not.
It
is
a
matter
of
R&D
funding;
i.e.
it
is
a
supply-side
policy
instrument
in
relation
to
innovation.
Hence
PCP
is
some-
thing
very
different
from
PPI,
which
is
a
dedicated
demand-side
policy
instrument
in
relation
to
innovation.
Therefore,
PCP
cannot
be
considered
as
PPI
in
our
sense
of
this
term,
since
PPI
largely
includes
instruments
other
than
public
R&D
funding.
PPI
and
PCP
may
thus
supplement
each
other
as
parts
of
a
policy-mix,
but
they
should
not
be
mixed
up.
A
PCP
may
be
an
important
preparation
and
specification
phase
before
a
PPI
process
is
started.
Table
1
classifies
the
six
cases
to
be
presented
in
the
next
two
sections
by
using
the
previous
concepts.
As
the
reader
will
know,
direct
PPI
has
historically
been
a
much
more
commonly
used
instrument
than
catalytic
PPI.
We
believe
that
this
can
be
seen
as
an
indication
that
catalytic
PPI
is
an
underused
innovation
policy
instrument.
Hence,
we
would
like
to
open
up
for
a
much
wider
use
of
this
instrument
in
order
to
contribute
to
the
mitigation
of
various
grand
challenges.
In
addition,
it
is
important
to
add
a
third
dimension
to
the
clas-
sification,
i.e.
the
fact
that
PPI
can
be
performed
in
a
more
or
less
cooperative
and
interactive
way.
All
the
five
categories
discussed
above
may
be
more
or
less
cooperative.
Cooperation
implies
that
the
(public)
procurer
and
the
potential
(private)
supplier(s)
com-
municate
and/or
collaborate
for
the
purpose
of
learning
in
the
procurement
process,
sometimes
over
a
long
period
of
time.
By
def-
inition
and
by
design,
there
is
always
some
cooperation
between
procurers
and
(potential)
suppliers
in
PPI.
Cooperation
between
the
procurer
and
the
(potential)
supplier
may
concern
the
whole
pro-
cess
of
procurement,
but
can
also
apply
to
only
one
or
more
of
the
stages
it
consists
of
(see
below).
Such
cooperation
is
obviously
related
to
the
degree
of
competition
between
potential
suppliers.
Cooperation
is
a
matter
of
degrees,
not
a
dichotomous
variable.
As
we
know,
interactive
learning
is
a
key
dimension
of
innovation
(see
Section
2)
and
even
more
of
PPI,
as
the
cases
addressed
illustrate.
The
typical
PPI
process
can
be
divided
into
the
following
stages
(adapted
from
Edler
et
al.,
2005;
Expert
Group
Report,
2005):
1.
Identification
of
a
grand
challenge
(or
a
public
agency/mission
need),
and
its
formulation
in
terms
of
a
lack
of
satisfaction
of
a
human
need
or
an
unsolved
societal
problem.
2. Translation
of
the
identified
challenge
into
functional
specifica-
tions.
3. Tendering
process:
(a)
Opening
of
the
bidding
process
through
a
tender.
(b) Translation
of
the
functional
specification
into
technical
specifications
by
potential
suppliers.
(c) Submission
of
formal
bids
by
potential
suppliers.
4.
Assessment
of
tenders
and
awarding
of
contracts.
5.
Delivery
process:
(a)
Product
development.
(b)
Production
of
the
product.
(c)
Final
delivery
to
the
purchasing
agency.
This
general
structure
does
not
imply
by
any
means
that
the
PPI
process
is
of
a
linear
nature.
As
we
will
see
in
the
six
cases
below,
these
general
steps
are
very
much
interrelated
and
intertwined.
4.
Summary
of
case
descriptions
The
cases
emanate
from
three
countries,
although
Sweden
dom-
inates
with
four
well-documented
cases.
The
other
two
cases
represent
Norway
and
the
USA,
and
are
of
more
recent
origin.
The
bias
towards
Sweden
is
explained
by
its
long
history
with
regard
to
innovation
procurement
(Edquist
et
al.,
2000a).
However,
the
Swedish
extensive
tradition
does
not
imply
that
many
new
PPI
ini-
tiatives
have
been
undertaken
since
the
end
of
the
1990s.
The
lack
of
examples
thus
represents
the
momentum
that
PPI
policies
have
lost
during
the
last
couple
of
decades.
However,
this
situation
appears
to
be
changing
(see
Section
1).
The
methodological
approach
followed
is
exploratory.
Infor-
mation
about
each
case
has
been
complied
by
accessing
relevant
documents
such
as
tender
calls,
scientific
literature,
policy
docu-
ments
and
evaluations
and
other
written
materials
and
reports.
Table
2
provides
a
detailed
summary
of
the
six
cases
described
according
to
the
following
dimensions:
1.
PPI
process
a. Challenge/need:
illustrates
the
point
of
departure
of
the
pol-
icy.
What
was
the
challenge/problem/unsatisfied
need?
Was
it
carried
out
to
meet
the
(mission)
needs
of
the
public
agency?
b.
Result
of
the
procurement
process:
describes
whether
a
prod-
uct
(material
good
or
intangible
service)
or
system
was
the
intended
result
in
order
to
mitigate
the
challenge.4
4Cases
where
R&D
results
were
the
intended
output,
e.g.
cases
of
Pre-Competitive
Procurement
(PCP),
were
excluded
from
this
article.
We
will
address
PCP
in
a
later
article
(Edquist
and
Zabala-Iturriagagoitia,
forthcoming).
1760 C.
Edquist,
J.M.
Zabala-Iturriagagoitia
/
Research
Policy
41 (2012) 1757–
1769
Table
1
Overall
summary
of
cases.
User
Character
of
the
procurement
process
Adaptive
procurement
Developmental
procurement
Direct
procurement X2000
(Case
1)
AXE
Telephone
Switch
(Case
2)
NødNett
Norge
(Case
5)
ADS-B
(Case
6)
Catalytic
procurement Light
corridor
(Case
3) a
Swedish
refrigerator
(Case
4)
aThe
fact
that
we
could
not
find
any
cases
for
this
box
is
significant,
which
relates
to
the
statement
we
have
made
on
the
underutilization
to
date
and
the
future
potential
of
catalytic
procurement.
c. Degree
of
cooperation
and
type
of
call:
was
there
an
open
call
where
potential
suppliers
could
“bid”
in
competition
or
was
the
call
restricted
to
selected
suppliers?
This
dimension
also
captures
the
degree
to
which
the
procurer
cooperated
with
the
supplier(s)
during
(some
stages
of)
the
PPI
process.
d.
Type
of
PPI:
as
specified
in
Section
3
(Table
1).
e.
Other
intended
consequences
that
the
results
had
on
the
iden-
tified
challenge/need/problem.
f.
Other
unintended
consequences
of
the
policy
(e.g.
regarding
profits,
exports,
etc.).
g.
Type
of
subsidy
(incentive):
shows
how
the
PPI
process
was
funded,
promising
the
purchase
of
a
future
order,
offering
economic
rewards,
etc.
h.
Policy-mix:
illustrates
whether
other
supplementary
policy
instruments
were
also
used
as
a
complement
to
the
PPI
pro-
cess.
2.
Procurer
a.
Who
was
the
procurer:
identifies
the
organization
acting
as
a
procurer
of
the
intended
result.
b. Functional/technical
specifications:
illustrates
whether
the
procurer
had
developed
functional
or
technical
specifications,
or
both,
prior
to
launching
the
PPI
process.
How
did
the
pro-
curer
develop
the
specifications?
c.
End-user:
identifies
who
was
the
end-user
of
the
(intended)
result
of
the
PPI.
3.
Supplier
a.
Who
was
the
supplier:
identifies
the
organization/firm
acting
as
the
supplier
of
the
intended
result.
b.
Award
criteria:
defines
the
criteria
by
which
the
supplier
was
awarded
the
contract.
Following
the
previous
dimensions,
Section
5
contains
more
thorough
descriptions
for
each
of
the
six
cases.
5.
Detailed
case
study
descriptions
5.1.
X2000
high
speed
train
The
first
case
deals
with
the
procurement
of
the
X2000
high-
speed
train
in
Sweden
(Edquist
et
al.,
2000b),
one
where
the
only
existing
user,
the
Swedish
State
Railway
Company
(SJ),
represented
the
final
demand
for
the
trains.
Hence,
it
constitutes
a
case
of
direct
procurement
which
faced
human
mobility
as
a
major
social
need.
The
entire
process
of
bidding
was
open
to
foreign
firms,
through
several
rounds
occurring
between
the
first
tender
request
in
1982
and
the
final
contract
negotiation
in
1986.
One
of
the
most
notable
aspects
of
the
X2000
procurement
was
the
length
of
time
required
for
its
completion.
An
insufficient
level
of
competence
on
the
procurer’s
side
was
one
of
the
primary
rea-
sons
for
this
slow
process.
A
technical
requirement
demanded
by
SJ
at
the
time
was
that
of
a
“single-locomotive,
tiltable
train”
(Edquist
et
al.,
2000b,
p.
86).
The
problem
of
defining
(realistic)
requirements
took
some
years
and
it
was
finally
solved
in
1985
when
SJ
sent
out
a
supplementary
request
for
a
second
round
of
tenders.
ASEA
(later
AD
Tranz,
now
wholly
owned
by
Daimler)
was
then
awarded
the
contract
in
summer
1986
as
a
“development
partner”.
Develop-
ment
partner
is
here
understood
as
the
blending
that
allows
the
involved
actors
(purchaser
and
supplier)
to
develop
a
high
level
of
competence
through
interactive
learning
(Fridlund,
2000,
p.
147).
Accordingly,
we
may
also
deem
this
case
a
cooperative
procure-
ment
process.
The
requirement
that
it
had
to
be
locomotive-drawn
not
only
resulted
in
the
X2000
losing
the
international
technological
com-
petition;
it
also
made
success
in
export
markets
impossible,
and
it
did
not
become
the
dominant
design.
In
the
competing
Italian
Pendolino,5every
wagon
had
its
own
engine,
while
X2000
neces-
sarily
had
to
consist
of
one
single
locomotive
and
a
fixed
number
of
wagons.
This
made
Pendolino
much
more
flexible
in
terms
of
demand
requirements
and
maintenance,
reflecting
the
lack
of
com-
petence
of
SJ
as
a
procurer
at
the
time,
and
the
devastating
influence
that
having
too
specific
and
strong
technical
requirements
(instead
of
only
functional
ones)
can
have
on
innovation
outputs.
The
X2000
resulted
in
significant
improvements
to
the
infra-
structure
and
thereby
conditions
for
economic
growth.
These
included
increased
commercial
profitability
for
SJ,
reduced
infrastructure
costs
for
building
traffic
routes,
lower
energy
con-
sumption
costs
from
decreased
use
of
automobiles,
shorter
travel
times
and
reduction
of
accidents
and
pollution
due
to
highway
traffic.
The
procurement
of
the
X2000
can
thus
be
regarded
as
a
case
of
adaptive
public
procurement;
it
was
not
the
first
high
speed
train
to
be
developed,
nor
has
it
come
to
be
widely
used
outside
Sweden.6In
fact
it
did
not
result
in
any
radical
change
in
the
speed
or
direction
of
technical
change,
as
most
of
the
elements
in
the
train
system
had
existed
previously
and
a
high-speed
train
with
similar
functions
(the
Italian
Pendolino)
had
appeared
well
before.
Thus,
the
pro-
curement
was
not
that
successful
from
an
innovation
policy
point
of
view.
Nonetheless,
the
X2000
case
can
be
deemed
to
have
had
limited
success
in
terms
of
industrial
policy.
5.2.
AXE
switching
technology
The
second
case
concerns
the
procurement
of
the
AXE
telephone
switch
(Fridlund,
2000),
where
the
Swedish
Telecommunications
Administration
(STA)
as
a
user
also
known
as
Televerket
(nowa-
days
Telia
Inc.),
and
the
private
manufacturer
LM
Ericsson
(LME,
today
Ericsson)
as
a
producer
were
the
main
actors
involved.
5The
Pendolino
had
already
been
developed
and
came
into
full
operation
in
1976
(Giuntini,
1993),
so
Swedish
public
procurers
should
have
been
aware
of
this
tech-
nical
development
by
the
time
the
X2000
procurement
process
started
officially
in
1982,
even
if
the
whole
discussion
of
developing
a
high
speed
train
in
Sweden
started
in
1969.
6Despite
SJ’s
belief
there
could
be
an
enormous
export
opportunity
for
the
high
speed
train
(e.g.
Norway,
Finland,
Germany,
Austria,
France,
Portugal,
China,
Australia,
USA),
only
one
train
set
was
sold
to
China.
Between
1998
and
2010,
a
trainset
called
Xinshisu
was
in
traffic
between
Hong
Kong,
Shenzhen
and
Guangzhou.
C.
Edquist,
J.M.
Zabala-Iturriagagoitia
/
Research
Policy
41 (2012) 1757–
1769 1761
Table
2
Detailed
summary
of
case
descriptions.
1.
PPI
process
A.
Challenge/need
B.
Results
C.
Cooperation
and
type
of
call
consulta-
tion/dialogue/partnerships
D.
Type
of
procurement
E.
Intended
consequences
F.
Unintended
consequences
G.
Type
of
subsidy
H.
Policy-mix
1.
X2000
Human
mobility
as
a
challenge
Faster
and
reliable
passenger
transport
on
existing
tracks
High
speed
train
(rapid
passenger
traffic)
Open
call
Continuous
dialogue,
“development
partners”
Direct
Adaptive
Cooperative
Reduced
journey
times
Industrial
development
Reduced
infrastructure
costs
Delay
Lack
of
exports
Innovation
in
some
components
Knowledge
transfer
Industrial
policy
Innovation
policy
Legislation
2.AXE
Telephone
Switch
Communication
as
a
challenge
Introduce
new
services
for
subscribers
Improvements
in
efficiency,
capacity
and
maintenance
costs
Construction
and
development
of
equipment
for
electronic
switching
Closed
call
(fixed
switch)
Continuous
dialogue
of
the
3
parties
during
the
entire
process
Exchange
of
employees
Direct
Developmental
Cooperative
Computerized
switching
technology
stations
Change
from
electro-mechanic
to
electronic
technologies
Exports
(40%
of
world
market)
National
digital
coverage
Low
telephone
fees
Adaptability
to
mobile
telephony
Procurement
contract
(including
R&D
funding)
Industrial
policy
Joint
R&D
(public-private)
funding
3.Light
corridor
Energy
efficiency
as
a
challenge
Stimulate
the
development
of
energy-efficient
products
Make
the
results
marketable
Stimulate
the
market
Improve
the
efficiency
of
lighting
Restricted
call
(light
corridor)
Open
call
(HF
ballasts)
Pre-procurement
discussions
Dialogue
with
largest
Swedish
manufacturers
Agreements
with
several
organizations
Catalytic
Adaptive
Cooperative
(in
early
stages)
Introduction
of
more
efficient
technology
Development
of
requirements
for
lighting
power
density
Use
of
Life-cycle
cost
estimation
Development
of
testing
methods
Standard
setting
Subsidy
per
kWh
saved
Energy
policy
Fiscal
instruments
Regulations
Financial
instruments
4.Swedish
refrigerator
Energy
efficiency
as
a
challenge
Produce
market
transformations
towards
more
energy-efficient
technologies
Reduce
energy
consumption
and
environmental
impact
of
refrigera-
tors/freezers
Open
call
Consultation
with
experts
and
key
purchasers
Catalytic
Adaptive
Cooperative
(in
early
stages)
Efficiency
improvements
Energy-efficient
labeling
Products
sold
after
STPP
Growing
market
share
Exports
Energy
savings
Follow-up
by
other
manufacturers
Adaptation
to
other
kitchen
appliances
Subsidized
purchase
STPP
covered
a
portion
of
the
buyers’
cost
Energy
policy
Innovation
policy
Industrial
policy
5.
NødNett
Norge
Security
as
a
challenge
but
also
mission-oriented
Coordinate
independent
analogue
mobile
radio
networks
Enable
interdepartmental
communication
Development
of
a
digital
mobile
radio
system
Open
call
Continuous
dialogue
and
cooperation
during
the
pre-procurement
and
procurement
phases
Direct
Adaptive
Cooperative
Single
nationwide
digital
radio
system
Cost
savings
Technical
improvements
Efficiency
gains
Higher
security
Employment
creation
Potential
commercialization
to
other
countries
Investments
granted
in
the
national
budget
R&D
policy
Defense
6.
ADS-B
Security
as
a
major
challenge
but
also
mission-oriented
Transform
air
traffic
control
systems
from
a
radar-based
system
to
a
satellite-based
one
Constant
broadcast
of
the
precise
location
of
all
aircraft
to/from
other
aircraft
and
traffic
controllers
Open
call
Continuous
dialogue
before,
during
and
after
the
bidding
process,
with
industry
representatives,
service
providers
and
experts
Direct
Developmental
Replace
ground-based
radar
systems
for
air-traffic
control
Lower
cost
Full
airspace
coverage
Improved
safety
Improve
accuracy,
integrity
and
reliability
of
satellite
signals
Reduce
separation
between
aircraft
Cost-plus
incentive
fee
in
phase
1
Fixed-price
arrangement
in
phase
2
Defense
policy
1762 C.
Edquist,
J.M.
Zabala-Iturriagagoitia
/
Research
Policy
41 (2012) 1757–
1769
Table
2
(Continued
).
2.
Procurer
3.
Supplier
A.
Who
was
the
procurer
B.
Func-
tional/technical
specifications
C.
End-user A.
Who
was
the
supplier
B.
Award
criteria
1.
X2000
SJ,
Swedish
State
Railway
Company
Functional
and
technical
requirements
SJ
ASEA
(now
wholly
owned
by
Daimler-
Chrysler)
Trust
Long-term
costs
2.AXE
Telephone
Switch Swedish
Telecommuni-
cations
Administration
– Televerket
Functional
Televerket
in
Sweden
and
foreign
telephone
operators
Ellemtel
(jointly
owned
by
STA
and
LM
Ericsson)
Joint-venture
between
procurer
and
supplier
3.Light
corridor
NUTEK
Swedish
Council
for
Building
Research
Functional
Technical
Energy
utilities
and
real
estate
companies
(as
energy
providers)
13
large
Swedish
real
estate
management
companies
and
owners
of
public
and
commercial
buildings
Helvar
Oy
(Finland)
Ease
of
arranging
the
funding
of
their
own
share
of
costs
(light
corridor)
Technical
expertise
(HF
ballasts)
4.Swedish
refrigerator
Purchaser
group:
HBV,
NUTEK,
etc.
Functional
40–50%
more
efficient
than
existing
products
on
the
market
1.0–0.9
kWh/l
Individuals
Electrolux
AB
Price
Use
of
standard
and
established
technology
5.
NødNett
Norge
Norwegian
Ministry
of
Justice
and
the
Police
(in
cooperation
with
other
national
organizations)
Functional
Several
public
organizations
(mostly
related
to
Fire,
Police
and
Health
brigades)
Siemens
Norway
Price
Requirements
6.
ADS-B
Federal
Aviation
Administration
(FAA)
Performance-
based
FAA
UPS
(outcome)
ITT
Corp.
Best
value
and
least
risk
in
the
implementation
C.
Edquist,
J.M.
Zabala-Iturriagagoitia
/
Research
Policy
41 (2012) 1757–
1769 1763
The
goal
was
the
creation
of
a
computerized
switching
telephone
network
that
would
meet
the
social
demands
related
to
commu-
nication
needs.
AXE
was
developed
by
the
semi-public
company
Ellemtel,
which
was
jointly
owned
by
Televerket
and
LME.7Con-
sequently,
we
can
talk
about
a
procurement
process
that
was
cooperative
to
a
very
large
degree,
in
which
a
separate
joint
orga-
nization
was
created
as
a
result
of
a
public–private
partnership,
without
a
previous
bidding
process,
in
order
to
deal
with
the
chal-
lenge.
The
first
stage
of
the
PPI
process
focused
mainly
on
develop-
ing
functional
specifications,
with
Ellemtel
assessing
the
proposals
Televerket
and
LME
had
elaborated
in
terms
of
potential
computer-
ized
switching
systems.
The
main
problem
of
the
AXE
project
was
to
come
up
with
a
solution
to
meet
the
demands
of
both
Ellemtel’s
shareholders.
While
Televerket
was
interested
in
a
switch
to
ful-
fill
the
needs
of
the
Swedish
telephone
system,
LME
was
export
oriented
and
required
a
switch
fitting
as
many
systems
as
pos-
sible.
Ellemtel
finished
the
study
by
proposing
two
versions
of
a
switching
technology:
one
for
local
switches
and
up
to
20,000
subscribers
(for
the
domestic
market),
and
one
for
larger
switches
(foreign
markets).8The
functional
requirements
to
be
met
by
the
AXE
system
were
agreed
upon
by
the
three
parties
in
1972.
The
next
stage
in
the
PPI
process
implied
the
develop-
ment/production
of
the
new
switching
technology.
Coordinating
committees,
product
and
function
committees,
including
expert
groups
and
steering
committees,
had
members
from
all
levels
in
the
three
organizations.
This
involved
an
important
mobility
of
employees
among
the
three
partners,
which
facilitated
the
transfer
of
knowledge
among
them.
We
can
thus
say
that
this
procurement
case
included
all
the
three
stages
that
a
cooperative
procurement
policy
should
have:
consultation,
demonstration
and
coordination
(Gavras
et
al.,
2005).
The
first
test
of
the
AXE
technology
was
carried
out
in
1976
in
a
Televerket
switching
station
in
Södertälje,
which
had
3000
subscribers
and
was
regarded
as
a
suitable
first
trial
for
the
new
technology.
The
next
steps
were
oriented
towards
increasing
the
scale
of
operation
of
the
stations
and
selling
the
AXE
system
in
foreign
mar-
kets.
As
was
noted
above,
LME
viewed
exports
as
an
intended
and
desirable
consequence
of
this
PPI.
During
the
following
years,
Telev-
erket
received
orders
for
10
new
AXE
stations
in
Sweden
(up
to
240,000
subscriber
lines),
which
were
manufactured
by
LME
after
knowledge
had
been
transferred
from
Ellemtel
to
LME.
This
made
Sweden
the
first
country
in
the
world
to
offer
national
digital
cov-
erage
which
also
led
to
lower
telephone
fees.
At
about
the
same
time,
LME
received
requests
for
new
stations
in
Finland,
France,
Denmark,
Mexico,
Brazil,
Australia
and
Saudi-Arabia,
which
gave
LME
a
share
of
40%
of
the
world
market
in
1992.9Accordingly,
we
may
conclude
that
the
AXE
case
was
certainly
a
direct
and
devel-
opmental
PPI,
as
a
new-to-the-world
digital
switch
was
introduced
in
the
market
and
directly
purchased
by
Televerket.
7The
contract
creating
Ellemtel
was
signed
in
1970
with
the
objective
of
devel-
oping
strategic
telecommunications
technologies
in
general
and
the
AXE
electronic
switch
in
particular.
The
procurement
contract
stated
that
the
major
advantages
of
the
new
telecommunication
technology
(the
societal
needs
it
had
to
meet)
were,
in
addition
to
increasing
operational
reliability
and
reducing
maintenance
costs,
the
adaptation
of
the
new
switches
to
the
varying
conditions
and
the
provision
of
new
services
to
customers.
8For
a
complete
list
of
the
functional
requirements
agreed
on,
see
Fridlund
(2000,
p.
157),
Vedin
(1982,
p.
138)
and
Meurling
and
Jeans
(1995,
p.
38).
9Most
of
these
buyers
were
also
public
agencies,
so
they
also
constitute
new
cases
of
innovation
procurement
of
an
adaptive
kind.
5.3.
Light
corridor
In
1988
the
Swedish
Government
decided
to
establish
a
new
energy-efficiency
program
which
included
a
sub-program
for
technology
procurement,
managed
by
the
Department
of
Energy
Efficiency
at
NUTEK
(Swedish
National
Board
for
Industrial
and
Technical
Development).10 One
of
these
projects
for
increasing
energy-efficiency
was
the
light
corridor
project,
initiated
jointly
by
NUTEK
and
the
Swedish
Council
for
Building
Research
(BFR).
Since
NUTEK
initiated
this
project
on
behalf
of
private
companies
included
in
the
BFR,
this
can
be
categorized
as
a
catalytic
case.
The
objectives
of
the
program
were
to
stimulate
the
devel-
opment
of
energy-efficient
products,
systems
and
processes,
to
demonstrate
their
function,
to
stimulate
market
penetration
and
to
commercialize
the
results
in
residential
and
commercial
buildings
and
in
industry
(Stillesjö,
1993,
p.
219;
Suvilehto
and
Öfverholm,
1998).
The
project
started
with
the
appointment
of
a
reference
group
that
included
representatives
from
the
authorities,
users,
con-
sumers,
real
estate
owners
and
managers,
energy
utilities,
lighting
manufacturers,
scientists
and
lighting
consultants
to
discuss
strate-
gies
for
achieving
more
efficient
use
of
electricity
in
buildings.
As
a
result,
the
functional
specifications
for
lighting
systems
were
pub-
lished
in
order
to
guide
bids.
In
order
to
spread
the
previous
specifications
so
that
efficiency
gains
(i.e.
savings)
would
be
as
large
as
possible,
NUTEK
invited
the
largest
energy
utilities
and
real
estate
companies
to
sign
an
agree-
ment
to
participate
in
the
procurement
program.11 NUTEK
funded
the
initiative
through
entering
into
agreements
with
BFR
and
pro-
viding
it
with
financial
incentives
to
stimulate
the
purchase
of
more
efficient
equipment
(Stillesjö,
1993).
Agreements
were
signed
with
thirteen
of
Sweden’s
largest
real
estate
management
compa-
nies
and
owners
of
public
and
commercial
buildings
(representing
30%
of
the
total
floor
space
of
such
buildings).
The
participants’
involvement
was
governed,
among
others,
by
the
following
clauses
(Stillesjö,
1993,
p.
221):
(a)
participants
will
receive
an
investment
bonus
of
SEK
1.5
for
every
kWh
of
electricity
per
year
saved
compared
to
conven-
tional
design,
up
to
a
ceiling
of
SEK
2.5
million;
(b)
participants
will
try
to
meet
the
minimum
efficiency
standard
of
10
W/m2in
new
or
retrofit
lighting
design;
(c)
if
participants
are
successful
and
implement
the
new
standards
throughout
their
whole
organization,
the
ceiling