40 Innovation Policy and Sustainable Development - IWT

IWT-Studies 

IWT-Observatory 

Institute for the Promotion of Innovation 

by Science and Technology in Flanders 

Innovation 

Science 

Technology 

Innovation Policy and Sustainable Development: 

Can Innovation Incentives make a Difference 

40 

Contributions to a Six Countries Programme Conference, 

February 28 - March 1, 2002, Brussels 

EDITOR: PATRIES BOEKHOLT

COLOFON 

IWT-Studies is published by IWT-Vlaanderen 

as part of the work programme of the IWT- 

Observatory. However, the authors are 

personally responsible for the standpoints 

adopted in the development of these studies. 

Editors 

Ann Van den Bremt (secretariat) 

Jan Larosse (co-ordination) 

Production 

N’lil 

Copyright 

Reproduction and use is permitted subject 

to acknowledgement of source 

IWT-Observatory 

Jan Larosse, Co-ordinator 

Donald Carchon, Information system 

Ann Van den Bremt, Secretariat 

Henri Delanghe, Policy analysis 

Bischoffsheimlaan 25 

1000 Brussels 

Phone: 02/209 09 00 

Fax: 02/223 11 81 

E-mail: iwt-observatorium@iwt.be 

Web-site: http://www.iwt.be 

Registration number: D/2002/7037/5 

Published in August 2002

IWT-STUDIES > >> 40 

CONTENTS 

ABSTRACT 4 

FOREWORD 7 

PART 1: SETTING THE SCENE 

Introduction 9 

• Position paper (Patries Boekholt and Jan Larosse) 11 

• Innovation and environmental policies for sustainable development 

(Ken Guy) 19 

• Policies for innovation and the environment: 

toward an arranged marriage (George R. Heaton, Jr.) 29 

• An integrated policy for innovation for the environment (René Kemp) 37 

PART 2: DOES INNOVATION POLICY MAKE A DIFFERENCE 


• Impact and additionality of innovation policy (Luke Georghiou) 57 

• The Norwegian systemic approach to impact estimation of R&D subsidies: 

focus on additionality and the contra-factual problem 

(Lasse Bræin, Arild Hervik, Erik Nesset, Mette Rye) 67 

• Measuring ‘relative effectiveness’ 

can we compare innovation policy instruments 

(Erik Arnold and Patries Boekholt) 87 

PART 3: COUNTRY CASES 


SUMMARY 

• Green innovation (Jesper Holm, Ole Erik Hansen, Bent Søndergård) 103 

• Sustainability research in Austria (Hans-Günther Schwarz) 117 

• Stimulation of sustainable technology development in the Netherlands: 

the E.E.T. programme (Corine van As, René Wismeijer) 125 

• Flanders’ new scheme of innovation subsidies 

for sustainable development (Paul Zeeuwts) 135 

• Towards policy-integration (Patries Boekholt) 141 

3

ABSTRACT 

This volume of IWT-Studies brings together 

most of the contributions to the Conference 

on ‘Innovation Policy and Sustainable 

Development. Can public incentives make a 

difference’ held in Brussels, in February 2002. 

The Flanders 2002 Conference of the Six 

Countries Programme (SCP), organised by 

IWT with the assistance of Technopolis 

Group, can be regarded as a follow-up to 

the earlier SCP-Conference organised in 

1996 by IWT in Ghent under the title ‘R&D 

subsidies at stake In search of a rationale 

for public funding of R&D’. The issue of the 

‘additionality’ of innovation policy now has 

been enlarged to the contribution to innovation 

performance for wider, societal goals 

as sustainable development. 

The papers are organized in three blocks. 

The first, introductory one is on the interplay 

between innovation and environmental 

policy. The second one deals with the question 

of evaluating the effectiveness of innovation 

policy as such: does it ‘make a difference’ 

(the additionality issue). The third 

block presents country case studies of environmentally 

sustainable innovation that all 

entail a particular kind of integration of 

innovation and environmental policy. A 

summary presents conclusions from this conference 

and implications for further work. 

1. SETTING THE SCENE 

The Position Paper by Patries Boekholt and 

Jan Larosse starts from the observation that 

the rationale for innovation policy has been 

strengthened from a system perspective. It 

can meet the agenda of sustainable development 

in terms of innovation for societal 

objectives. From an economic perspective 

the missions of innovation policy -reducing 

market failures due to positive externalities 

of knowledge production and usage- and of 

environmental policy -reducing the negative 

externalities of non-sustainable technologies- 

can be combined. On the agenda is 

how to operate an horizontal policy that 

effectively realizes this goal: is innovation 

policy effective as a policy instrument, and 

can it be part of a policy mix that levies 

progress to sustainable development 

A first series of conference papers was delivered 

by the keynote speakers, invited to ‘set 

the scene’, and introduce the two policy traditions 

that are challenged to contribute 

together to environmentally sustainable 

growth. 

Ken Guy starts with drawing some lessons 

on the effectiveness of present day innovation 

policy from a system perspective. 

Innovation policy is now expected to serve 

several masters. Exploring the possibilities of 

interaction between innovation and environmental 

policies, he suggests that a beneficial 

combination has to answer two different 

questions: what can innovation policy 

do for sustainable development, and what 

can environmental policy do for innovation. 

Applying a successful broad mix requires 

high levels of ‘strategic intelligence’. 

The second paper was presented by George 

Heaton, who introduced the metaphor of 

‘an arranged marriage’ to describe the difficulties 

and opportunities involved in matching 

two parties that have a lot in common, 

both are not able to get acquainted without 

special commitment and special changeagents. 

There is a clear under-investment in 

next generation technologies by the environmental 

industry, but there is a need to 

frame these technology developments in 

new, transformative technological pathways. 

The paper of René Kemp is included in this 

introductory series because it outlines the 

need for an integrated policy approach 

towards innovation for the environment. It 

outlines the model of transition management 

as a policy approach that can add topdown 

elements of system innovation (longterm 

ambition) to bottom-up initiatives 

(short-term concerns). 

2. DOES INNOVATION POLICY MAKE 

A DIFFERENCE 

The three following papers deal with the 

question of additionality - what would have 

4


happened if no intervention had taken 

place - as a way to establish whether policy 

instruments have any effect. Identifying and 

measuring the socio-economic effects and 

impact of innovation policies is high on the 

agenda of many OECD countries, as a consequence 

of an increased requirement for 

accountability in public sector management. 

In order to establish the positive externalities 

of knowledge production and diffusion, 

and not only the benefits for those directly 

involved in a policy programme, we need to 

find ways to measure these wider effects. 

Policy makers are increasingly required to 

demonstrate effects of policy instruments in 

quantitative terms. 

The first paper by Luke Georghiou takes stock 

of the progressively broadening approach 

towards the key issue of additionality of innovation 

policy: from quantitative approaches 

and effects on R&D investment (flows), 

towards behavioural approaches and qualitative 

effects on capacities in the innovation system 

(stocks). The paper explores what the evidence 

for the impact of innovation policy has 

been so far. Governments can make a difference 

in the stimulation of innovation in public 

goods. But measures to promote R&D in 

general are less a support mechanism than 

establishing a public-private partnership to 

produce innovations in an area that might not 

have benefited. 

There is a school of scholars aiming to measure 

the longer-term impacts of innovation 

policy with the help of econometrics. 

Norway has a long tradition of evaluating a 

number of innovation policy instruments 

using quantitative analytical methods. The 

second paper by Lasse Bræin, Arild Hervik, 

Erik Nesset and Mette Rye discusses these 

methods and their problems and provides 

empirical evidence from a number of 

Norwegian cases. 

The third paper by Erik Arnold and Patries 

Boekholt discusses how evaluations can help 

us deciding whether innovation policy 

works. It also discusses what we do not 

know because of methodological difficulties. 

The paper is based on a large number 

of evaluation studies in various countries. 

3. COUNTRY CASES 

The following country case studies demonstrate 

the wide variety of initiatives that are 

developed to stimulate innovation for environmental 

sustainability, depending on different 

development paths and different 

institutional settings. They illustrate the 

importance of a broad range of (complementary) 

options in the achievement of the 

overall goal. 

The Danish study by Jesper Holm et al refers 

to evidence from sector specific case studies 

on development of new institutional practices 

and capacity building and the development 

of green products on the basis of the 

win-win philosophy of the predominant 

model of ‘ecological modernisation’. It was 

possible to start to integrate environmental 

and business development policies and to 

create an interactive ‘green’ responsiveness 

from business to new regulations and programmes 

starting from a company focussed 

policy view. The success of these reflective 

learning processes is based on the interfacing 

of different actors and the presence of a 

social component that enhances the change 

in business behaviour. 

The Dutch EET programme, presented by 

Corine van As and René Wismeyer, is an 

example of a mission-oriented approach, 

coordinating the efforts of three Ministries. 

It aims to stimulate breakthroughs in 

themes put forward by government to 

achieve sustainable development. The latest 

version stresses the transition to sustainability 

by systemic innovation as the distinctive 

objective of funding. 

In contrast, the new programme of stimulation 

of sustainable environmental technology 

development in Flanders, presented by 

Paul Zeeuwts, is an example of a bottom-up 

approach, linked to generic technology and 

innovation funding programmes. The aim is 

to stimulate a shift in innovation behaviour 

towards more environmental benefits in all 

projects, by selectivity and financial incentives. 

The paper by Hans-Günther Schwarz gives 

5

an account of the Austrian programme on 

Technologies for Sustainable development, 

in particular the two thematic sub-programmes 

on the House and the Factory of 

Tomorrow. They are not limited to technology 

development but include the necessary 

complements of demonstration projects and 

knowledge dissemination. The integration 

of multiple goals and multiple types of innovation 

is a characteristic of the programme 

design. As in the other cases evaluation 

therefore is a rather complex issue. 

SUMMARY: 

TOWARDS POLICY-INTEGRATION 

The Summary of Patries Boekholt attempts 

to draw some conclusions referring back to 

the objectives of the conference, namely to 

see whether a closer integration of innovation 

policy and environmental policy on 

the overlap of innovation for sustainable 

development, is feasible. If the effectiveness 

of general innovation policy is not clearly 

established by direct measurement we do 

know that it is the policy mix that works 

on the level of innovation systems. 

Environmental policies have demonstrated 

some effectiveness in changing behaviour in 

addressing focussed technological challenges. 

The challenge now is to extend these more 

targeted approaches towards a systemic 

change process affecting the broader business 

community. We will need further policy 

innovations in the field of policy integration. 

6


FOREWORD 

What can innovation policy do for Sustainable 

Development In particular, can innovation 

policy provide incentives to promote 

technological solutions for environmental 

sustainability 

This theme was proposed by IWT for the 

Spring 2002 Conference of the Six Countries 

Programme. The SCP, in which IWT has been 

participating for several years, is an international 

forum for discussions on new policy 

developments and advanced policy research 

in innovation. The Conference in Brussels 

brought together representatives of the two 

policy communities to investigate the possibilities 

of beneficial combination. A lot of 

experience has already been accumulated at 

the level of environmental technological 

policies but the issue of incentives for 

Sustainable Development from an innovation 

policy perspective is still emergent. 

For IWT this is not an academic discussion 

because it was given the task to implement 

the policy objectives formulated by the 

Flemish Government to promote sustainable 

technological development. After thorough 

preparations – in which this conference also 

played a role – IWT proposed a new scheme 

for ‘Sustainable Technological Development’ 

that was accepted by the Flemish Government 

in May 2002. A specific characteristic 

of the new IWT scheme is its integration into 

the general subsidy scheme. All projects in 

the ecological existing programmes that 

meet specified criteria are given priority and 

can also benefit from a subsidy bonus of 

10%. This generic approach is rather unique. 

But it is directly linked to the main 

challenges of innovation policy. 

The first one is the additionality of innovation 

policy. If innovation policy wants 

to make a difference it is acknowledged 

that this should be at the level of behavioural 

changes in innovation management. 

Incentives should stimulate more and 

better innovation projects. The modalities 

of general support thus have been altered 

to discriminate in favour of STD-projects. 

In addition support is given to introduce 

new tools as Life Cycle Analysis in the 

project planning. 

The second is the complementary interaction 

of policy instruments. The IWT scheme 

is but one of the instruments of the Flemish 

government to advance the agenda of 

Sustainable Development. It is the policy mix 

of regulatory change, indirect tax incentives 

and direct subventions that determines its 

success. But IWT in its capacity of the technology 

agency of the Flemish Government 

didn’t stop with the new scheme. Within the 

framework of its mission to provide co-ordination 

between innovation intermediaries 

IWT recently inaugurated an ‘Environmental 

Technology Platform’ that should improve 

relations between ‘supply’ and ‘demand’ for 

sustainable technology. With these two initiatives 

IWT contributes to a wider horizontal 

policy approach. 

Sustainable Technological Development is 

an important theme for years to come. The 

Conference constituted one of the first 

attempts to combine innovation and environmental 

policies. In the future these discussions 

will be continued on the basis of 

evaluation and monitoring of the experiences 

of the new programmes. 

Paul Zeeuwts 

President IWT 

7

Part 1 

SETTING THE SCENE 

> 

INTRODUCTION 

The Position Paper by Patries Boekholt and 

Jan Larosse starts from the observation that 

the rationale for innovation policy has been 

strengthened from a system perspective. It 

can meet the agenda of sustainable development 

in terms of innovation for societal 

objectives. From an economic perspective 

the missions of innovation policy – reducing 

market failures due to positive externalities 

of knowledge production and usage – and 

of environmental policy – reducing the negative 

externalities of non-sustainable technologies 

– can be combined. On the 

agenda is how to operate an horizontal 

policy that effectively realizes this goal: is 

innovation policy effective as a policy 

instrument, and can it be part of a policy 

mix that levies progress to sustainable 

development 

A first series of conference papers was delivered 

by the keynote speakers, invited to ‘set 

the scene’, and introduce the two policy traditions 

that are challenged to contribute 

together to environmentally sustainable 

growth. 

Ken Guy starts with drawing some lessons 

on the effectiveness of present day innovation 

policy from a system perspective. 

Innovation policy is now expected to serve 

several masters. Exploring the possibilities of 

interaction between innovation and environmental 

policies, he suggests that a beneficial 

combination has to answer two different 

questions: what can innovation policy 

do for sustainable development, and what 

can environmental policy do for innovation. 

Applying a successful broad mix requires 

high levels of ‘strategic intelligence’. 

The second paper was presented by George 

Heaton, who introduced the metaphor of 

'an arranged marriage' to describe the difficulties 

and opportunities involved in matching 

two parties that have a lot in common, 

both are not able to get acquainted without 

special commitment and special changeagents. 

There is a clear under-investment in 

next generation technologies by the environmental 

industry, but there is a need to 

frame these technology developments in 

new, transformative technological pathways. 

The paper of the third keynote speaker Luke 

Georghiou features in Part 2, introducing the 

concept of additionality to assess whether 

innovation policy makes a difference. 

The paper of René Kemp is included in this 

introductory series because it stresses the 

need for an integrated policy approach 

towards innovation for the environment. It 

outlines the model of transition management 

to work towards sustainability in various 

domains. Transition management consists 

of a deliberate attempt to bring forth 

long-term change in a step-wise manner, 

utilising dynamics. 

1 

9


POSITION PAPER 

INNOVATION POLICY AND SUSTAINABLE DEVELOPMENT 

CAN PUBLIC INNOVATION INCENTIVES MAKE A DIFFERENCE 

PATRIES BOEKHOLT 

JAN LAROSSE 

Technopolis 

IWT 

Patries Boekholt is founder and director of 

Technopolis BV in Amsterdam. Before joining 

Technopolis she worked for the TNO 

Centre for Technology Policy Studies since 

1989. She has over ten years of experience 

working on Research, Technology and 

Innovation policies in various countries and 

for international organisations such the 

European Commission and the OECD. This 

includes evaluation studies, strategic advice, 

international comparative, benchmark studies 

and improving policy implementation. 

Jan Larosse obtained a MA in Economics and 

a Baccalaureat in Philosophy at the KU of 

Leuven. Since 1992 he is Scientific Advisor for 

IWT, the technology agency of the Flemish 

Government. In 1996 the Observatory unit 

was started, of which he is the co-ordinator. 

His work is mainly directed to innovation 

monitoring on the level of indicators and 

databases, and to innovation studies covering 

themes as the knowledge economy, cluster 

analysis and innovation systems, innovation 

policy and additionality. 

11

Position paper 

> 

2 

See for instance 

the OECD: 

A new Economy 

The changing role 

of innovation 

and information 

technology in growth, 

2000, Paris. 

1. INNOVATION POLICY IN THE 

KNOWLEDGE ECONOMY 

In the Knowledge Economy permanent 

renewal has become the driving engine of 

the economic fabric. This pivotal role of 

innovation and the central position of 

knowledge production and distribution in 

this process are reshaping the institutional 

settings of economic development. In the 

1990s the interaction between private and 

public actors, networking between firms, 

and the flow of knowledge between organisations 

and people, both nationally and 

internationally, have become an important 

prerequisite for good performance. 

The focus of most recent innovation studies 

and policy actions has mainly been on the 

impact on productivity and economic 

growth 2 . Public intervention in innovation is 

justified if it helps in achieving these goals, 

and in such a way that would not have happened 

if left to market forces. Policymakers 

have designed and implemented a wide 

range of incentives and schemes to make 

this happen, either by improving general 

framework conditions for markets to work 

or by remediating market failures in particular 

domains with important ‘externalities’. 

Tax incentives, R&D subsidies, collaborative 

R&D schemes, technology transfer agencies 

and many more type of measures have been 

put in place to spur innovation in the private 

sector. 

To accommodate the ever more wide-ranging 

and interactive character of innovation 

processes the traditional science and technology 

policy agenda has been enlarged to an 

innovation policy agenda that links economic, 

industrial and research policies. Innovation 

policy has become a horizontal policy that is 

playing a vital role in strengthening economic 

performance. Its impact is conditioned by the 

ability to align a broad spectrum of policy 

measures, ranging from a conducive regulatory 

and macro-economic framework, over a 

high quality education, developed risk capital 

markets that support start-ups, up to facilities 

for R&D partnerships. 

Today improving competitiveness and 

growth performance is not the only challenge 

for innovation policy. Government 

institutions and policies are also urged to 

make ‘policy innovations’ in order to contribute 

to an environmentally sustainable 

growth. This challenges innovation policy to 

widen its scope even more. 

A new kind of ‘innovation paradox’ seems 

to arise. At the very moment that public 

intervention to stimulate innovation is 

challenged to widen its scope, the rationale 

of public intervention is under scrutiny. 

Success models of technological change like 

Silicon Valley have confirmed the central 

role of enterprises and market forces in 

innovation. Does public intervention really 

make a difference in stimulating innovation 

And how should it operate The correct 

answers to these questions will be very 

much determining the effectiveness of the 

operation of the innovation system as a 

whole. 

In this context the 2002 6CP will address the 

following key questions: 

• Should innovation policy, which is primarily 

focused on improving competitiveness 

and growth performance in the private 

sector incorporate ‘public goals’ in its 

policy instruments, particularly those 

goals related to minimising the negative 

externalities that accompany economic 

growth 

• What difference has innovation policy 

really made in the last decades Has it 

been effective Are we able to measure 

this 

• In this context which innovation policy 

incentives will work to also have a major 

contribution to sustainability 

The remainder of this position paper will 

discuss these questions in more detail. In 

section 5 we will discuss what the Brussels 

6CP aims to achieve. We have translated this 

into a programme structure (section 6) 

which addresses these questions in plenary 

and parallel sessions. 

12



> 

2. DEFINING A BROADER RATIONALE 

FOR INNOVATION POLICY 

Governments are under constant pressure to 

justify public interventions in private sector 

innovation. Policy makers are not only held 

accountable for the effects of their instruments, 

the rationale for intervention is 

under permanent scrutiny and needs to be 

reassessed with the advancement of innovation 

theory as well as the implementation of 

good public management practices. 

Main stream economics accepts policy 

intervention on the level of improving 

framework conditions (regulation) and 

factor conditions (e.g. education) stimulating 

markets to work and to maximise 

social welfare. Welfare economics provides 

a specific theoretical framework for 

policy intervention in the neo-classical 

tradition, referring to market failures. 

Maximising ‘social returns’ is then the 

policy objective that incorporates the 

(positive and negative) external effects 

beyond the scope of private cost-benefit 

calculation. An external effect is any value 

creation or destruction that is not mediated 

through the price system. A narrow 

definition will limit these external effects 

to ‘non-excludabilities’ and ‘non-appropriabilities’ 

in consumption or production of 

private persons, but it can be extended to 

any social (non-price) valuation, e.g. 

involving strategic objectives of governments 

in innovation. 

The traditional rationale for public intervention 

in knowledge production and distribution 

is based on the characteristics of 

knowledge as a quasi-public good, due to 

its spillover effects. They are at the origin 

of under-investment in private R&D-activities 

compared to the socially desirable 

level. Government can compensate for this 

market failure by public R&D or financial 

incentives to private R&D. The objectives 

for policy are mainly focused on increasing 

the competitiveness of firms and the sectors 

they operate in. The market failure 

approach is particularly weak in helping to 

make policy choices where intervention is 

needed and at what level. 

The evolutionary economists have moved 

away from this neo-classical ‘market failure’ 

approach and widened the set of rationales 

for innovation policy. The national systems 

approach identifies four types of failures 

that provide a justification for public 

intervention (see Smith 1996): 

1. Failures in infrastructural provision and 

investment; e.g. universities, publicly supported 

technical institutes, regulatory 

agencies, libraries and databanks, or even 

government ministries. 

2. ‘Transition failures’, e.g. serious problems 

for firms and sectors in adapting to transitions 

such as (radical) technological 

change. As Smith states many public 

policies are in fact aimed at these issues, 

frequently without any explicit rationale. 

3. Lock-in failures; Just as firms are not able 

to switch away from their existing technologies, 

so industries and indeed the 

whole socio-economic system can be 

“locked-in” to a particular technological 

paradigm. External agencies, with powers 

to generate incentives, to develop technological 

alternatives, and to nurture emerging 

technological systems are required. 

4. Institutional failures; an integrated set of 

public and private institutions, regulatory 

systems and the policy system makes up 

an overall context of economic and technical 

behaviour which shapes the technological 

opportunities and capabilities of 

firms. Failures in this system can form 

bottlenecks for innovation act as rationale 

for policy action such as a change in the IPR 

regulations. 

Thus the systems approach has widened the 

rationale for public intervention in innovation. 

All those different types of ‘failures’ can 

be summed-up as ‘system failures’, inhibiting 

the interactive process of innovation. 

The question is whether these types of 

system failures also apply for sustainable 

development, given that negative environmental 

externalities are already a ground 

for policies that encourage internalisation 

of these external costs. It is easy to argue in 

favour of that: environmentally friendly 

products and services are difficult to fund 

since mass markets need to be stimulated. 

13


It requires a major transition particularly in 

manufacturing practices and distribution of 

goods. The system is locked in the resource 

intensive (mass) production modes that are 

low cost but also environmentally damaging, 

and the institutional system needs to be 

made more aware of advantages and good 

practices of sustainable growth paths. 

Thus integrating technology innovation 

with environmental sustainability, using the 

innovation systems approach will strengthen 

the rationale for public policy. Given the 

pressure on accountability of public expenditures, 

a proper integration of the two policy 

goals could not only provide a better justification 

for action but also extend the 

social impacts of government action. 

The basic programme of sustainable development 

– in environmental protection - is 

the decoupling of economic growth from 

the increased usage of materials and 

energy, and from pollution. The intervention 

of government is based on the public 

good nature of the environment, as a common 

heritage of mankind, spanning the 

generations. The negative externalities of 

private economic activity on the stock of 

material resources, the climate and biodiversity, 

have to be corrected and prevented 

by specific incentives and regulations. 

Because of the increasing innovation 

intensity of economic growth the orientation 

of technological innovation is of strategic 

importance for the success of a sustainable 

development strategy. Two missions 

for public intervention are than to be combined: 

the promotion of innovation in general 

(because of market failures due to the 

positive externalities of knowledge production) 

and the promotion of ecological innovation 

in specific (because of market failures 

due to the negative externalities of 

using non-sustainable technologies). A winwin 

situation for innovation and environmental 

policy might be obtained by substituting 

technologies of the ‘old’ industrial 

economy, characterised by decreasing 

returns in the exploitation of its material 

resources, by far less material and energy 

dependent technologies of the ‘new’ economy, 

that is characterised by increasing 

> 

returns in the production and use of knowledge 

as prime resource. In general innovation 

policies and other public policies that 

are linked to other social missions of government 

impact each other. To achieve 

‘innovation performance’, they need to be 

co-ordinated in a horizontal manner. But 

the globalisation of the economy has also 

spurred the globalisation of the social 

agenda. The new mission of sustainable 

development has therefore to be translated 

in new institutional settings and usage of 

public incentives that combine innovation 

policy with goals of sustainable development, 

on a national and international scale. 

3. INNOVATION POLICY FOR 

SOCIETAL OBJECTIVES: 

THE CASE OF SUSTAINABLE GROWTH 

A broader rationale for public innovation 

incentives is a consequence of the pervasiveness 

of innovation in public affairs. 

Innovation has become a prime mover in the 

achievement of a wide range of societal 

goals, in particular summed up by the 

agenda of ‘sustainable development’. 

Whereas innovation policy is still very much 

focused on improving competitiveness, it is 

also used as a means to solve a wider set of 

issues such as traffic congestion, social 

exclusion, security and pollution. Very often 

the implementation of innovation policies 

for these goals has occurred in separate 

realms. Whereas the innovation policy community 

is mainly dealing with encouraging 

applied research, dissemination of technologies, 

innovations management for the 

private sector, these other policy domains 

are developing policy instruments in whole 

different policy communities, dealing with 

different sometimes conflicting policy 

objectives. In the 2000 OECD report 

Innovation and the Environment, Heaton 

makes the observation that “[t]he complementarity 

between technological change 

and environmental quality is not newly 

recognised; nor is the realisation that environmental 

and innovation policies should 

work in tandem. But these desiderata are 

still far from achieved: environmental and 

14



technology policies are routinely pursued in 

separate sphere, and environmental technology 

is too often lodged outside the 

mainstream of industrial innovation.” 

The 6CP conference wishes to discuss how 

the goals of these policy communities can be 

reconciled in order to make more comprehensive 

and effective policy choices. It does 

this mainly from the perspective of innovation 

policy: what can innovation policy 

instruments contribute to sustainability 

A question to be addressed is also whether 

‘traditional’ innovation policy can learn from 

the experiences with public interventions for 

environmental protection. In discussing this 

we need also to learn from the experience of 

those involved in environmental (technology) 

policies. Yukiko Fukasaku argues that 

innovation can play crucial role in contributing 

to the environmental sustainability of 

growth since the latter requires radical 

changes in the nature of goods and services 

that are produced, as well as the way they 

are produced, distributed and used. He sees 

as the main obstacle the diffuse nature of 

environmental innovation and the difficulties 

in defining the boundaries of both environmental 

R&D and the environmental 

goods and services industry. 

Studies on the innovation effects of environmental 

policies mainly discuss regulation as 

the key form of public intervention. The most 

common response to regulation according to 

Kemp (2000) is incremental innovation in 

processes and products and diffusion of existing 

knowledge (in the form of end-of-pipe 

solutions and non-innovative substitutions of 

existing substances). Other instruments have 

been developed world-wide, including subsidies, 

convenants, R&D programmes and so 

on. The effects of these instruments have also 

been modest. A problem that scholars raise is 

that these are often end-of-pipe solutions, 

repairing damaging effects that occur in the 

manufacturing process. The 6CP want to discuss 

the possible early integration of policy 

objectives aimed at the positive externalities 

as well as the negative externalities of technology 

development. Will this have an effect 

on both competitiveness and sustainability 

> 

Or does the experience with environmental 

policies show that to have a real effect, environmental 

issues should not be tackled 

through innovation projects but using other 

policy mechanisms 

4. HAS INNOVATION POLICY REALLY 

MADE A DIFFERENCE SO FAR 

If innovation policy claims to contribute to 

sustainability we first need to establish 

whether the traditional incentives have 

made a real difference in its core domain: 

competitiveness. The justification for innovation 

policy is under scrutiny in many countries, 

because despite several decades of policy 

instruments, the impact of these policies 

on the key policy goals (increasing economic 

growth by improving the innovation 

performance) has not been established 

unambiguously. The systems approach has 

widened the rationale for public intervention 

in innovation. Most important is the 

stimulation of technology diffusion. The 

knowledge distribution power of the 

national innovation system – basic to the 

productivity of the innovation process - has 

to be structurally improved through the 

enhancement of connectivity or inter-linkages. 

The most obvious place for public 

intervention is therefore on the interfaces 

between different institutions and actors. 

What does this mean for the portfolio of 

policy incentives that have been used over 

the past decades Are the traditional R&D 

subsidies less effective than for instance 

mechanisms to create university-industry 

links What empirical evidence do we have 

for demonstrating the state-of-the-art in 

innovation policy incentives Has the impact 

only affected the private level or can we 

show the wider social impacts that different 

incentives have 

The overall assessment of public intervention 

in the framework of social returns is 

mirrored in the assessment of effectiveness 

and ‘additionality’ of piecemeal policy 

incentives. Do they make a difference compared 

to the results of market forces left on 

their own Additionality in R&D performance 

is defined as a measure of the extent to 

15


which public support stimulates new R&D 

activity as opposed to subsidising what 

would have taken place anyway (Cameron, 

et al. 1996). It consists of three categories of 

effects: input, behavioural and output. 

Cameron et al. conclude that in considering 

the impact of government for R&D on firms’ 

strategies, additionality is most relevant at 

portfolio level. However most evaluations 

are performed at individual scheme level. 

Additionality can be restricted to a narrow, 

quantitative understanding of the economic 

activities to be stimulated, i.e. R&D expenditures, 

but can also be broadened to more 

qualitative effects, i.e. on the project management 

or the propensity to collaborate. 

Additionality in the welfare economic sense 

is achieved when governments can stimulate 

private investors to provide public goods as 

new generic knowledge or a healthy environment 

(maximising positive externalities 

of knowledge production or minimising 

negative externalities of pollution and 

depletion of resources). These quasi-public 

goods in the realm of innovation policy can 

be the consequence of knowledge spillovers 

(knowledge products itself). But additionality 

in innovation stimulation policies can 

also be assessed according to additional 

social objectives for innovation, in which 

governments have a complementary or 

correcting role with respect to market 

forces, as the case of sustainable development 

demonstrates. This links up with a 

systemic approach towards additionality in 

which the structural and institutional 

features of the innovation system are object 

of ‘management’. 

One question for the debate is whether current 

analytic tools are adequate to measure 

this wider social impact and additionality 

Do policy makers have the appropriate 

information to decide on effectiveness 

Establishing the socio-economic effects and 

impact of innovation policies is high on the 

agenda of many OECD countries, as a consequence 

of an increased requirement for 

accountability in public sector management. In 

order to establish the positive externalities of 

knowledge production and diffusion, and not 

only the benefits for those directly involved in 

a policy programme, but also the social benefits 

accruing from their activities, we need to 

find ways to measure these first and second 

order effects. Policy makers are more and 

more required to demonstrate effects of policy 

instruments in quantified terms. 

Can innovation policy be used to achieve other 

societal goals or should it be restricted to 

innovation as such And if social goals can be 

integrated with the traditional innovation policy 

incentives, how does this work in practice 

Impact analyses suffer from some key 

methodological problems, which the evaluation 

expert community summarises as 

• The attribution problem: how can you isolate 

the effect of a policy instrument on 

the performance of one firm or a group of 

firms, given the many additional factors 

that influence that performance The mirror 

problem: how to measure and evaluate 

the impact of the ‘spillovers’ of innovation 

activities 

• The time lag between research, innovation 

and economic effects for those directly 

involved in the programme and even more 

for those not participating in the programme. 

Years can go by before a commercial 

return of investment can be 

achieved at the individual firm level. The 

effects of technology support programmes 

should be measured short, medium and 

long term. The data on the medium and 

long-term effects are usually lacking. 

There are few programmes that have a 

sufficient long history to be able to analyse 

effects in the longer term, which many 

experts and practitioners estimate at 20 

years. (Norway’s user-directed R&D programmes 

are unusual in that their effects 

on participating firms have been investigated 

at periods up to ten years after project 

completion.) 

• The quantification of the many qualitative 

effects that are included in the objectives 

of the programme such as networking, 

improving the absorptive capacity and 

competences of firms 

16



Exercises to develop these methodologies 

are currently being undertaken in several 

countries and international platforms. Has 

the innovation policy community achieved 

any progress on these matters What is the 

state-of-the-art at this moment Can innovation 

policy continue to acquire political 

support if it does not progress on these 

matters Have technology and innovation 

policies in other domains, such as environmental 

policies, achieved better successes in 

producing and demonstrating effectiveness 

Policy choices are not only dependent on 

the effectiveness and additionality of individual 

policy measures, they should be 

made assessing a whole portfolio of policy 

instruments in a systemic way. Only few 

countries have developed exercises to evaluate 

their policies in a systemic manner. In 

those cases multiple methods were used to 

come to a conclusion on what worked well 

and which instruments did not make a difference. 

We propose to re-examine the empirical evidence 

on the real effects of major policy 

choices in perspective of their required (systemic) 

‘additionality’. Enhancing the positive 

externalities of knowledge production 

(and also minimising the negative externalities 

of environmental damage) and enhancing 

the provision of public goods of different 

nature is thereby a guideline. 

The empirical problem is twofold: how can 

impact be measured, and what is the real 

impact of different incentive schemes Is the 

impact in line with the policy goals of those 

schemes What type of scheme is really 

effective for innovation stimulation and 

what doesn’t make a difference How to 

measure impact on the level of the innovation 

actors (e.g. additional growth of private 

R&D) How to measure impact on the level 

of the innovation system (e.g. the growth of 

total factor productivity) How to measure 

the impact on ‘societal returns’ (e.g. knowledge 

spillovers, environmental benefits, 

employment effects) 

On the level of policy co-ordination new 

institutional mechanisms need to be created 

to cope with the tasks of public authorities 

to manage better the public good aspects of 

the innovation system. Different sectoral 

ministries and agencies have to mutually 

adjust their policies. 

Two preconditions are therefore important 

to progress on the road to policy co-ordination: 

the establishment of organisational 

communication between departments that 

are involved in the stimulation of innovation 

about their objectives, and the availability 

of specific indicators that measure the 

(interaction) effects of their policy actions to 

achieve these objectives. At European level 

the need for convergence and subsidiarity of 

national and European level is already well 

perceived. The recent advances in scoreboard 

and benchmarking exercises are producing 

instruments for better co-ordination. 

In some countries ‘Innovation Summits’ (e.g. 

Australia) bringing together all actors to formulate 

the policy needs spur the co-ordination 

efforts of government. Impact on 

behaviour of innovation actors very much 

depends on anticipations of strategic goals 

becoming self-fulfilling. 

The cycle of making policy goals explicit and 

assessment of its effect can generate convergence 

to effective policy mixes stimulating 

innovation on the level of the individual 

firm, of clusters or of the innovation system 

as a whole. This (self-organising) interaction 

of policy objectives, given the ever more 

prominent role of innovation policy, has to 

be embodied in new governance structures 

and methods. Impact assessment of the 

additionality of different policies should 

play a steering role. ‘Additionality’ as an 

analytical concept has its roots in the context 

of measuring ‘substitution effects’ for 

R&D subsidies. Welfare economics indicates 

the complementary roles of market and 

government to achieve social goals. The 

recent systemic approaches stress the interaction 

effects of different institutional levels 

and the challenge of re-adjusting the 

‘linear’ policy instruments in this context. 

The policy area, which can provide an excellent 

test case for this question, is that of sustainable 

development. 

17


> 

> 

5. WHAT WE WOULD LIKE TO ACHIEVE 

IN THE CONFERENCE 

The Brussels 6CP aims to further the discussion 

on the rationale and effectiveness of 

innovation policy in the context of: 

• an increased pressure for accountability 

and bench-marking of innovation policy 

instruments 

• the wish to contribute to environmental 


• the achievements and limitations of present 

evaluation methodologies to measure 

the social impact of policy incentives 

• the co-existence of a range of, sometimes 

conflicting, policy goals in which innovation 

can play a role, which also include 

societal goals 

• the growing importance of more integrated 

policy approaches versus the still 

fragmented implementation of policy 

instruments leading to sub-optimal results 

• the need to bring the existing innovation 

policy instruments further 

Not only do we want to launch policy 

debate, we also hope to discuss these 

themes on the basis of cases that provide 

empirical evidence and practical operational 

experiences. For instance examples of policy 

measures that have combined the ‘traditional’ 

innovation policy objectives with sustainability. 

Cases of innovation policy programmes 

with empirical evidence of their 

success in terms of a wider social impact. 

SOME REFERENCES 

Arthur B., Increasing returns and Path 

Dependence in the Economy, 1994 

Amable B. et al, Les Systèmes d’Innovation à 

l’ère de la globalisation, Paris, 1997 

Cameron, H., Luke Georghiou, Tim Buisseret, 

“What difference does it make” Additionality 

in the public support of R&D in 

large firms., PREST, 1996. 

David, P. & Forray D., Accessing and Expanding 

the Science and Technology Knowledge 

Base, in STI Review no 16, 1995. 

C. DeBresson, An Entrepreneur Cannot Innovate 

Alone; Networks of Enterprises Are 

Required, DRUID Conference, Aalborg, 1999. 

MERIT, Innovation Policy in a Knowledge- 

Based Economy, Innovation Directorate EC, 

June 2000 

OECD, Boosting Innovation, the Cluster 

Approach, 1999 

OECD-TIP, New Policies for a New Economy: 

Emerging Issues in Public Funding for R&D, 

2000a 

OECD, Innovation and the Environment, 

2000b 

OECD-TIP, Technological Innovation and 

Sustainable Development, 2000c 

Smith K., 1996, Systems Approaches to 

Innovation: Some Policy Issues, STEP-Group, 

Oslo, Norway. 

TAFTIE, Additionality, efficiency and quality 

of public funding, in TAFTIE Guidelines on 

performance indicators for evaluation and 

monitoring, www.taftie.org. 

18


INNOVATION AND 

ENVIRONMENTAL POLICIES 

FOR SUSTAINABLE DEVELOPMENT 

KEN GUY 

Wise Guys Ltd. 

Ken Guy is currently the Director of Wise 

Guys Ltd., a flag of convenience which 

allows him to continue working with 

friends on topics which interest him. After 

taking Natural Sciences at Cambridge in the 

late 1960’s, Ken Guy’s interest in innovation 

policy was stimulated during his postgraduate 

days at Manchester University in 

the Department which later became PREST. 

His interest in environmental policy began 

shortly after his first academic appointment 

at Clark University, Worcester, Massachusetts, 

where he was a Research Fellow for the 

Scientific Committee on Problems of the 

Environment (SCOPE). 

Over the years he has maintained his interest 

in both areas, first at the SCOPE/UNEP 

Monitoring and Assessment Research 

Centre (MARC) at the University of London 

and subsequently at SPRU, University of 

Sussex, during the 1980s, where he found-ed 

the group concerned with the Evaluation 

of Government and Industry Strategies for 

Technology, and at Technopolis, an innovation 

policy consultancy which he founded 

in 1989. 

Recent tasks relevant to his talk today 

include evaluations of environment-related 

R&D programmes in a number of countries, 

jointly authoring an EU report on climate 

change and the challenge for R&D policy, 

and involvement in a number of Committees 

and Panels concerned with the 

development of EU innovation policies. 

19

1. INTRODUCTION 

When I was invited to give the introductory 

talk at this conference, I was asked to set the 

scene for the next two days by running 

through some of the key issues preoccupying 

all those concerned with innovation policy, 

environmental policy and sustainable 

development, concentrating in particular on 

how innovation and environmental policy 

might be brought together to promote sustainable 

development. Underpinning this 

was a desire to reorient innovation policy 

such that it could become an effective tool 

in the process of attaining sustainable development. 

In effect, I was told that the main 

aim of the conference was to ask the question: 

“What can innovation policy and innovation 

policymakers do for sustainable 

development”. 

Any attempt to cover all this ground in 

detail would be doomed to failure in the 

time allocated for my talk. I shall only 

attempt, therefore, to scratch the surface – 

leaving it to those who follow to dig deeper 

and unearth the real treasures. In so doing I 

shall touch fleetingly upon the nature of 

innovation and innovation policy and what 

we know about the effectiveness of policy in 

this sphere. I shall also do the same for environmental 

policy and environmental technology 

policy. I shall then consider their 

interaction and some of the ways in which 

they can be beneficially combined, following 

the plot through to ask what innovation 

policy can do for sustainable development. 

As a twist to the plot, however, I shall 

also ask: “What can environmental policy 

and sustainable development do for innovation” 

20

Innovation and Environmental Policies for Sustainable Development 


> 

2. POLICY SPHERES 

To be sure that we’re all talking and thinking 

about the same things, I’m going to start 

with some simple clarifications and definitions. 

For example, as the first Exhibit shows, 

I shall take innovation policy as a sub-set of 

industrial policy; the overlap between industrial 

policy and environmental policy as the 

domain of sustainable development policy; 

and the overlap with innovation policy as 

the area we are most concerned with in this 

conference. 

Policy Spheres 

Industrial Policy 

Innovation Policy 

Environmental Policy 

Innovation Policy for 

Sustainable Development 

Sustainable Development 

Policy 

It follows immediately from this diagram 

that some of the issues we should cover concern 

the relative size and position of these 

spheres, e.g. should environmental policy be 

a small adjunct to industrial policy Should 

innovation policy dominate industrial policy 

Can or should all of innovation policy be 

located within the sustainable development 

policy domain Should innovation policy 

dominate the sustainable development 

overlap etc. 

> 

Hopefully we will return to some of these 

issues during the course of the conference, 

but let’s have the simple definitions first. In 

terms of aims and objectives, we can take 

the aim of industrial policy to be the promotion 

of economic ‘well-being’, typically 

defined in terms of economic growth and 

the robustness of the economy as a whole. 

Similarly, we can take environmental ‘wellbeing’ 

as the aim of environmental policy, 

with actions often resembling remedial 

rather than preventative medicine. At the 

juncture of these two spheres, we can consider 

the aim of sustainable development to 

be the maximisation of environmental productivity 

and efficiency within economic 

production processes, i.e. producing more 

output from less material resources with the 

minimum amount of waste and pollution. 

For innovation policy, however, a simple definition 

is less forthcoming, for the aim of 

innovation policy is not simply to promote 

more innovation, nor is it simply to promote 

economic growth. Although innovation is 

one of the key drivers of modern economies, 

innovation policy has historically been 

applied only when the aim has been to 

achieve socially desirable externalities which 

would not, or could not, be achieved by the 

private sector alone in the absence of public 

policy interventions. Frequently these externalities 

manifest themselves in terms of economic 

benefits, and innovation policy does 

then become an instrument which enhances 

economic growth. But increasingly these 

externalities, or expected externalities, have 

come to have a broader socio-economic 

complexion, which has meant that innovation 

policy is now expected to serve other 

masters, including the goal of environmental 

well-being. This in turn means that innovation 

policy is concerned not only with 

increasing the absolute amount or rate of 

innovation, but also with shaping the course 

of scientific and technological developments 

in ways deemed socially desirable. 

3. INNOVATION POLICY 

Before examining the consequences of this 

broader remit for the attainment of sustainable 

development, however, we need to ask 

whether innovation policy has fulfilled 

expectations apropos of its narrower, historical 

economic remit. In other words, has 

innovation policy worked and, if it has, 

what’s the magic formula for success 

Broadly speaking, the answer to the first 

part of the question is both ‘yes’ and ‘don’t 

know’. Efforts to evaluate individual innovation 

policy mechanisms, ranging from R&D 

programmes to tax incentives and diffusion 

initiatives, have become commonplace over 

21


the last twenty years, and my own general 

impression – which I’m sure will be shared by 

many of the other professional innovation 

policy evaluators in the audience – is that 

most single initiatives actually achieve their 

immediate, pragmatic goals and contribute 

to the higher level goals which litter the 

rhetoric of their rationales – though to what 

extent remains an open question. Also 

unknown is the effectiveness of different 

combinations of policy instruments – i.e. of 

the overall policy mix – vis-à-vis the attainment 

of these high-level goals. 

All this should come as no surprise, however, 

for we are dealing here with complex social 

systems, not conducting simple laboratory 

experiments or controlled trials. We have 

developed fairly crude social science methods 

to assess the attainment of the more 

directly realisable goals of individual policy 

initiatives (e.g. the knowledge, networking 

and even commercial exploitation goals of 

R&D programmes), and we have also developed 

sets of macro-economic indicators 

which allow us to track many aspects of 

innovative and economic behaviour at the 

level of whole economies, but we have not 

cracked the problems of causation and attribution 

which bedevil attempts to link microand 

macro- phenomena, nor have we managed 

to build adequate models of the ways 

in which innovation policy initiatives interact 

with and influence overall economic 

behaviour. And neither are we likely to in 

the foreseeable future. 

Our inability to measure the overall impact of 

innovation policies on economic performance, 

however, is no argument for their 

removal from the armoury of public policymakers. 

We cannot measure the effect of 

many other individual initiatives on the performance 

of complex social systems, but we 

persist in their use if they satisfy lesser goals 

and, more importantly, if theory suggests 

that there is a role for them to play even if 

the tools to measure overall effectiveness are 

inadequate. And this is where developments 

over the past twenty years in our understanding 

of the ways in which ‘innovation systems’ 

operate come to the rescue, for even a 

cursory inspection of current theory provides 

a few simple clues concerning the desirability 

and applicability of innovation policies. 

In its simplest form, innovation systems theory 

draws upon general systems theory by 

defining ‘systems’ in terms of a number of 

‘actors’ (often represented diagrammatically 

by ‘boxes’) and the relationships between 

them (which are usually depicted by ‘arrows’ 

suggesting flows of information, money, 

influence etc.). Individual ‘reductionist’ studies 

then focus in on the evolving relationships 

between specific ‘boxes’ and ‘arrows’, while 

more ‘holistic’ studies attempt to understand 

the functioning of the system as a whole, 

albeit via an informed appreciation of the 

functioning of individual elements and often 

cavalier theorising about how they combine 

and evolve. There are many lessons to be 

learnt for the formulation of individual policy 

instruments from the plethora of reductionist 

studies conducted over the years, but there 

are also many lessons which stem from more 

holistic appraisals. 

The first of these is based on our current 

understanding of the complexity of modern 

innovation systems, each of which is composed 

of many different types of actor (multiple 

boxes) interacting in multifarious ways 

(multiple arrows). In such systems, system 

performance is often determined or regulated 

by the weakest node (i.e. the weakest 

link in the chain). The implication for policy 

formulation, therefore, is that policy interventions 

should target the weakest links. 

Similarly, attempts to benchmark innovation 

systems and the impact of policies on system 

performance should also concentrate on the 

identification and characterisation of weakest 

links. 

A second lesson which stems from the complexity 

of innovation systems is that individual 

policy instruments applied in isolation 

are unlikely to have a dramatic impact on 

overall system performance. In theory this is 

exactly what could happen if policies are targeted 

accurately at extremely critical weak 

links, but in practice the ‘strategic intelligence’ 

required to identify critical nodes is 

woefully inadequate. In complex systems, 

too, there are likely to be many weak nodes, 

22



> 

and even accurate targeting of an individual 

weak link is only likely to produce incremental 

improvements if other weak links are 

neglected by policymakers. 

A corollary of all of this is that successful 

attempts by public policymakers to improve 

the performance of complex innovation systems 

are more likely to consist of the application 

of a broad portfolio of policy instruments 

than the application of any one 

instrument in isolation. In this context it is 

salutary and dispiriting to note the upsurge 

in popularity in countries such as Sweden 

(and, possibly, the UK) for the unitary application 

of policies designed solely to support 

the science base rather than the application 

of a broad mix of policies across the whole 

innovation spectrum. In the context of complex 

innovation systems, policies such as these 

are very unlikely to bear the expected fruit. 

Applying a successful broad mix, however, 

again requires high levels of ‘strategic intelligence’ 

about the existence of weak links 

and the efficacy and appropriateness of individual 

policy instruments in particular contexts. 

In turn, this emphasises the need for 

constant experimentation and evaluation in 

the use of different instruments and combinations 

of instruments, with the results of 

these assessments continually feeding back 

into policy formulation discussions and 

benchmarking exercises. 

4. MORE LESSONS 

The process of segmenting whole innovation 

systems into constituent parts which 

interact with each other provides many 

more lessons for policymaking and benchmarking, 

even when the segmentation is 

simplistic in the extreme. Exhibit 2 segments 

a national innovation system into four constituent, 

interacting groups of actors, 

defined in terms of their membership of the 

public and private sectors and their roles as 

either ‘knowledge creators’ or ‘knowledge 

users’. Typical activities conducted by these 

different groups are also included in the figure. 

In reality the situation is obviously much 

more complicated than this, but even this 

gross simplification is sufficient to illustrate 

some lessons for innovation policy. 

One of the first things which current innovation 

theory tells us about a system such 

as this is that all parts need to interact well 

if the system is going to function smoothly. 

There is no point in knowledge creation, 

for example, if the routes to knowledge use 

are blocked (which is precisely the reason 

why the funding of basic science will not 

lead to improvements in innovation systems 

if the weak nodes in the system correspond 

to the barriers between knowledge 

creation and use). 

Our knowledge of innovation systems also 

tells us that the amounts of money spent by 

the private sector on applied R&D and product/process 

development dwarf the sums 

which the public sector has to support civil 

innovation generally. Using any of these 

funds to subsidise work in the private sector 

similar to that already being undertaken by 

firms would thus lead only to minor changes 

at the margin. In practice, policymakers have 

moved away from straightforward subsidy 

mechanisms in this area and will now only 

contemplate direct funding of private sector 

R&D if additionality, leverage or catalytic consequences 

can be demonstrated. 

A Simple Inovation System 

Public Sector 

Private Sector 

Knowledge Universities Intermediate and End Consumers 

Users S&T Training and Education Market for Goods and Services 

Knowledge 

Creators 

Universities 

Basic Scientific Research 

Firms 

Applied R&D and 

Product/Process Development 

23


It is still arguable, however, that innovation 

policies aimed to date at stimulating the 

scale of private sector investment in R&D 

activity have had limited success – via both 

direct funding and indirect schemes such as 

tax incentives. Given the relative size of 

these R&D budgets, however, and their critical 

role as motors of innovation systems, this 

is one of the most important areas for creative 

thinking in the whole innovation 

policymaking domain. 

The importance of S&T education and training 

within the Knowledge Use/Public Sector 

quadrant of the simple innovation system 

depicted in Exhibit 2 is emphasised by the 

current stress within the innovation theory 

literature on concepts such as social capital 

and absorptive capacity. Without adequate 

policies supporting strong scientific and 

technological cadres within the population 

and, perhaps even more importantly, a population 

at large equipped to be ‘knowledge 

consumers’ within ‘knowledge societies’, the 

prospects for healthy and dynamic innovation 

systems look bleak. 

The importance of adequate links between 

all the constituent parts of innovation 

systems is probably most easily demonstrated 

when considering private sector 

links between Knowledge Creators and 

Knowledge Users, for lack of an effective 

linkage here would simply mean that firms 

were unable to sell their goods and services. 

It is not surprising, therefore, that many 

innovation policies in recent years have 

taken the form of ‘bridging’ actions 

designed to ensure links are strengthened. 

Exhibit 3 uses the simple model of an innovation 

system outlined earlier to distinguish 

between different types of ‘bridging’ policies 

and different types of ‘reinforcement’ 

policies. The latter are aimed at strengthening 

or supporting mainstream activities 

within each quadrant. In contrast, the former 

are specifically designed to link quadrants 

together via actions which encourage 

or enable the actors in these quadrants to 

benefit from increased exposure to each 

other. Although the list of mechanisms in 

Exhibit 3 is by no means exhaustive, the preponderance 

of measures which can be 

described as ‘bridging’ measures - especially 

between public and private sector actors of 

all kinds and private sector knowledge creators 

and users - exemplifies their current 

importance in the arsenal of modern-day 

innovation policymakers. 

So is there a magic formula for successful 

innovation policy Can consideration of 

these simple models point the way forward 

All the discussion so far suggests that the 

following virtuous steps should be taken 

during the policy formulation process: 

• Base policy prescriptions on an analysis of 

as much ‘strategic intelligence’ as possible, 

making every effort to identify weak 

nodes within the innovation systems 

addressed and targeting these first; 

• Attempt to construct a policy portfolio 

which addresses as many of these weaknesses 

as possible, and don’t rely on a 

single policy instrument; 

• Include measures aimed at human 

resource development which strengthen 

competence and lead to increases in social 

capital and absorptive capacity; 

• Include ‘bridging’ measures aimed at 

improving knowledge and information 

flows; 

• Experiment, evaluate and feedback this 

‘strategic intelligence’ into policy formulation. 

24



A Simple Taxonomy of Innovation Policies 

PUBLIC SECTOR 

PRIVATE SECTOR 

KNOWLEDGE 

USERS 

Reinforcement Policies for 

Public Sector 

Knowledge Users 

• Support for science/technology 

educational institutions 

• Support for government 

‘knowledge workers’ 

Bridging Initiatives between 

Public and Private Sector 


• Public sector bodies 

providing technical information 

services 

• Public sector bodies 

providing metrology services 

• Public sector bodies providing 

patent and licence information 

• Public sector bodies providing 

consultancy advice to firms 

• Enhanced public private/ 

public sector interaction via 

proximity on science parks 




• Improved access to venture 

capital for ‘knowledge 

intensive’ sectors 

• Preferential tax regimes for 

‘knowledge intensive’ sectors 


Public Sector 

Knowledge Users and 

Knowledge Creators 

• ICT network 

infrastructure support 




Creators 

• Government procurement 

mechanisms 




Creators 

• Subsidised technology transfer 

and adoption schemes 

• Technology brokerage 

schemes 

• Support for the growth 

of private sector ‘intermediaries’ 

• Awareness campaigns 

• Technology demonstration 

initiatives 

KNOWLEDGE 

CREATORS 


Public Sector 


• Support for research equipment 

infrastructure in universities 

• Support for government 

labs 

• Support for basic science 

• Support for generic collaborative 

research between 

universities 

• Support for networks of 

research excellence 




• Support for targeted 

collaborative research 

between universities and 

industry 

• R&D services provided by 

Government labs to industry 

• R&D services provided by 

universities to industry 

• R&D partner search 

initiatives 

• University/industry 

personnel exchange 

schemes for R&D staff 




• Subsidised market-oriented 

R&D 

• Product development 

assistance 

• R&D tax incentives 

• Government support for 

inter-firm R&D collaboration 

(e.g. reciprocity agreements) 

• Innovation credits 

• Favourable IPR regimes 

• Inward investment schemes 

for ‘R&D intensive’ industries 

25


> 

1 

D. Anderson et al 

(2001), Innovation and 

the Environment: 

Challenges and Policy 

Options for the UK, 

London: Imperial 

College Centre for 

Energy Policy and 

Technology and the 

Fabian Society 

2 

R. Kemp (2000), 

‘Technology and Environmental 

Policy: Innovation 

Effects of Past 

Policies and Suggestions 

for Improvement’, in 

OECD (2000), Innovation 

and the Environment: 

Sustainable Development, 

Paris: OECD 

5. ENVIRONMENTAL POLICY 

In due course we will return to some of the 

generic lessons which innovation systems theory 

has for policy formulation and implementation. 

In the meantime, though, we should turn 

our attention to environmental policy, asking 

again what works and what doesn’t work. 

We can begin immediately by narrowing the 

focus, for the sphere of environmental policy is 

vast and the part we are interested in – the interaction 

with technological development, innovation 

and economic well-being – is a relatively 

small, if not unimportant, component of the 

whole. I shall argue, however, that this domain 

could and should grow in relative size and 

importance in the future, for technological innovation 

is a crucial route to environmental wellbeing, 

not only via ‘end-of-pipe’ modifications 

to existing technologies, but also via the wholesale 

shifts in production paradigms implied via 

the concept of sustainable development. 

A recent UK report 1 stated that the importance 

of technological and managerial improvement 

in enhancing environmental efficiency is 

becoming increasingly accepted. In support of 

this claim, it cites two examples of policies 

which successfully stimulated technological 

developments and led to environmental 

improvements. Both involved regulatory 

changes and the subsequent introduction of 

best practice technologies to reduce air and 

water pollution, the most striking of which was 

the fall in lead tailpipe emissions following the 

introduction in 1986 of lead-free petrol in the 

UK. Regulatory change helps define markets 

and can thus be an important stimulus to innovation. 

The historical record suggests that 

much of this innovation is modest and incremental 

in nature but, as René Kemp has noted 2 , 

stringent regulatory changes, e.g. product 

bans, can induce more radical innovation. 

We know too that some traditional innovation 

policy instruments also ‘work’ for environmental 

technology. Evaluations of environmental 

technology R&D programmes in which I have 

been personally involved (in Finland, Sweden 

and the EU generally) all suggest that they are 

no more and no less successful than R&D programmes 

in other technology areas. 

> 

Unfortunately, we also know -– or at least we 

strongly suspect – that efforts to stimulate R&D 

in these areas are sub-critical given the scale of 

the environmental challenges which confront 

us. Moreover, there is the additional suspicion 

that many of these initiatives need to shift 

away from a short-term focus on incremental 

technology improvem ents and refocus on 

more radical solutions to the longer-term 

problems posed by phenomena such as climate 

change, loss of biodiversity and the depletion 

of freshwater resources. 

We also know that the barriers to successful 

innovation are particularly severe in many 

environment-related areas. Although there is 

scope in theory for innovations which bring 

about improvements in environmental productivity 

and efficiency to be accompanied by cost 

reductions, in reality many in industry regard 

environmental innovation as an unwanted cost 

rather than as an opportunity to reduce costs 

in the longer term. Many of the industry sectors 

in which environment-friendly technologies 

are most needed also have ponderous 

technology infrastructures and very low rates 

of capital stock turnover, making innovation a 

slow and laborious process. 

6. WHAT CAN INNOVATION POLICY DO 


Many of the observations made about innovation 

systems and the lessons for innovation policy 

can be applied to the special case of innovation 

policy for sustainable development. 

Probably the most important of these is the 

proposition that an appropriate policy mix is 

likely to be much more effective than the use of 

a single, isolated instrument. Innovation is a 

complex process , different parts of which are 

best tackled via different types of instrument – 

some of which deal with ‘supply-side’ issues (e.g. 

funding for R&D and the training of scientific 

personnel), while others focus on demand-side 

stimulation (e.g. technology adoption incentives) 

and a whole host of other activities in 

between -– including all the ‘bridging’ initiatives 

depicted in the earlier Exhibit 3. Combinations 

of these instruments, customised to the needs of 

different industry and technology sectors, are 

needed to improve overall system performance. 

26



3 

D. Anderson et al 

(2001), op. cit. 

> 

This fact was recognised in the UK study mentioned 

earlier 3 . The authors recommended combinations 

of three types of instrument: target 

setting; financial support, including mechanisms 

for raising finance; and political and corporate 

leadership. The setting of outcome-based targets 

– based on assessments of what innovations can 

realistically deliver and backed by the promise (or 

threat) of legislative or economic enforcement – 

constituted the first part of the tripartite scheme. 

Next came a series of measures designed to provide 

financial support for innovation. These 

included public support for environmental R&D 

and the training of researchers; ‘back-loading 

support for innovation, i.e. prizes for technologies 

which secured particular environmental 

objectives; tax incentives and credits for innovation 

with beneficial implications for the environment; 

and the setting up of a National 

Environment facility to manage and channel 

environmental innovation. The third leg of the 

scheme envisaged much more efforts at a political 

level to provide leadership and deliver clear 

statements of long-term intent which industry 

could use to shape its investment plans. 

The second important lesson from our examination 

of innovation policy, however, is that 

we still don’t know which policy mixes are 

appropriate in different settings and which are 

likely to deliver the most desirable outcomes. 

But this is not a recipe for inaction. Rather, it is 

a clarion call for more determined efforts to 

evaluate and assess the performance of individual 

instruments, experiment with different 

combinations of policies, benchmark the performance 

of overall systems and slowly but 

surely increase our collective understanding of 

how these incredibly complex social and physical 

systems interact and function. The need 

here is for the type of ‘strategic intelligence’ 

which, in the long term, will allow us to 

identify and rectify weaknesses in particular 

innovation systems and make more enlightened 

and informed policy choices. 

7. WHAT CAN ENVIRONMENTAL 

POLICY DO FOR INNOVATION 

So far I have kept to the suggested story-line 

but, as promised, here’s a twist to the plot. 

Rather than asking what innovation policy can 

do for sustainable development, let’s turn the 

spotlight on what environmental policy can do 

for innovation. 

The answer is certainly not trivial. Innovation 

policymakers around the world are still making 

strenuous efforts to overcome systemic 

and context-dependent hurdles to the 

improvement of innovation system performance. 

In the EU, for example, the informal 

seminar of Industry and Research Ministers 

which took place in S’Agaró-Girona at the 

beginning of this month concluded that the 

EU still has many barriers to overcome if it is 

to achieve the so-called ‘Lisbon objective’ of 

becoming the most competitive and dynamic 

knowledge-based society in the world. 

Primary amongst these is the need to close 

the existing gaps with its main competitors in 

terms of the relative amounts of R&D conducted 

in the different settings and the efficiency 

of its translation into innovation and, 

ultimately, enhanced economic performance. 

Stimulating private sector investment in R&D 

is critical to the success of this venture and 

constitutes a very real challenge for innovation 

policy. Specific policy measures include 

direct measures (e.g. R&D grants and subsidies), 

indirect measures (e.g. tax incentives), 

loan guarantee schemes and schemes to make 

venture capital available for R&D, but finding 

the right policy mix remains an elusive goal. 

One suggestion, however, would be to take a 

leaf from the environmental policymakers’ 

handbook. Any comparison of the types of 

measures used in the two separate domains 

of innovation and environment policy soon 

reveals a glaring disparity. Whereas innovation 

policy has historically been dominated by 

efforts to improve inputs to the innovation 

process, with other actions gaining popularity 

only in the recent past, environmental policy 

has, in contrast, been dominated by regulatory 

measures which shape and influence 

markets and the demand for innovations. 

Would it be possible, therefore, to make better 

use of regulatory instruments to stimulate 

private sector spend on R&D 

In many technology spheres this will be neither 

possible nor desirable, particularly where 

the end goal of innovation policy is to 

27


enhance competitiveness irrespective of the 

nature or type of technology supported. 

It does start to make sense, however, when 

innovation policy is expected to serve other 

socio-economic goals – those related to 

health and the environment, for instance. 

The scale of the problems society will face in 

these two areas in the future also suggests 

that the markets for innovative products and 

processes in these domains will soar once 

successfully stimulated. The trick, therefore, 

is to find the correct policy mix, and it is 

here that combinations of traditional innovation 

policy instruments and environmental 

regulatory policies might prove their worth. 

Undoubtedly there will be some in the audience 

– and amongst these I expect to include 

my co-speakers this morning, George Heaton 

and Luke Georghiou – who are not convinced 

of the efficacy of regulatory mechanisms, for 

such instruments have often lacked teeth in 

the past. To my mind, however, the solution is 

to complement them with other innovation 

policy mechanisms and to lobby for their 

improvement rather than to dismiss them 

wholesale. But enough of my opinions. It’s 

time now for others to have their say. 

28


POLICIES FOR INNOVATION 

AND THE ENVIRONMENT: 

TOWARD AN ARRANGED MARRIAGE 

GEORGE R. HEATON, JR. 

Worcester Polytechnic Institute 

Massachusetts (USA) 

George R. Heaton is on the faculty at 

the Worcester Polytechnic Institute, an 

engineering university in Worcester, 

Massachusetts (USA), and is managing partner 

of Technology Policy International, 

an international consulting group in the 

US and Japan. Trained as a lawyer, 

George Heaton began his career at the 

Massachusetts Institute of Technology 

(MIT), where he taught in the engineering 

and management schools and undertook 

some of the first research focusing on the 

relationship between environmental policy 

and technological change in industry. The 

connection between public policy and technology 

– whether environmental or economic 

– is the central focus of his professional 

interest. In addition to academic pursuits, 

Mr. Heaton has consulted widely to 

national and international organizations, 

including long-term relationships with the 

World Resources Institute, the World Bank, 

the OECD, USAID, and several agencies of 

the Japanese government. In the international 

context, he has been a visiting professor 

at the Wissenschaftzentrum-Berlin 

and Saitama University in Japan, and has 

worked extensively on environmental and 

technology policy issues in France, China, 

Chile, Germany and elsewhere. George 

Heaton was rapporteur of the OECD workshop 

Innovation and the Environment. 

29

Policies for Innovation and the Environment: Toward an Arranged Marriage 

> 

1 

A standard cartoon in 

the U.S. shows a man 

shaking hands with his 

girlfriend’s father, 

as she introduces him 

with: “Dad, this is 

Mike; we’re getting 

married.” 

2 

In Japan, aunts of 

a certain age are 

stock-character 

go-betweens. 

So are superiors 

at the office. 

> 

1. A MARRIAGE ANALOGY 

Different cultures approach marriage differently, 

with widely differing results. In my 

own country, the United States, arranged 

marriages are universally scorned – although 

it should be noted that we have the world’s 

highest rate of internet-formed couples! In 

Japan, the process known as “omiyaikekkon” 

(literally, marriage by taking a 

look) is a respected institution, employed in 

about half of all marriages. The paradigm 

assumptions of these two approaches – 

“love match” and “arranged marriage” – 

diverge significantly: in the former, the natural 

affinity of the couple is valued above all 

else; in the latter, practical considerations 

affecting long-term compatibility tend to 

take greater weight. Their institutional 

mechanisms differ as well. In one case, the 

couple finds each other, falls in love and 

announces marriage – independently 1 . In 

the other case, intermediaries choose candidates, 

offer venues for getting acquainted 

and smooth the union of two families 2 . 

Obviously, each of these approaches has its 

virtues and drawbacks. But one fact is clear: 

the divorce rate in Japan is dramatically 

lower than in the U.S. 

The argument implicit this analogy is that 

environmental policy and innovation policy 

could make a good arranged marriage. That 

they are not an obvious love-match should 

be clear after 30 years of parallel co-existence 

accompanied by little interaction. But 

their compatibility is also certain: the fundamental 

goal of each is to put today’s technology 

on new trajectories. As with any 

arranged marriage, skillful intermediaries 

will be needed – to make the case, to make 

the contacts and to craft the mechanisms. 

The attendees at this meeting could well be 

in the vanguard of this effort. 

2. THE SITUATION OF THE PARTIES 

“Innovation policy” is a relatively new term 

for an older concept. Starting life in the 

post-World War II period as “science policy,” 

this set of initiatives has focused above all 

on increasing the quantity of inputs to the 

process of technological change – R&D 

funds, skilled personnel and technical information. 

The goal has been to spur economic 

growth through the development of new 

technologies that increase productivity and 

offer new functionality. Today’s focus on the 

“innovation system” as the locus of policy 

concern does not bespeak a change in 

purpose, but rather, a new understanding of 

the process of technological change. 

Perhaps the central realization has been 

captured by this meeting’s Position Paper, in 

its assertion that “permanent renewal” is 

the driving force in the knowledge economy. 

What this means is that all technologies 

are constantly challenged and 

improving – except for those that perish. 

Innovation policy has also developed an 

impressive infrastructure of expertise – people 

and institutions abreast of the latest 

developments in science and technology, 

adept at funding good research and alert to 

possibilities for its commercialization. What 

this infrastructure has generally lacked is 

much commitment to other societal goals, 

of which sustainable development offers but 

one example. To a large extent, this “neutrality” 

has been conscious and deliberate, 

often rationalized by the need to keep 

science and technology free from politicization. 

As a practical matter, what this has 

meant is that the technology policy community 

has largely ignored the environment. 

Modern environmental policy is now about 

30 years old. In its early years, the primary 

energy was placed on curbing the “adverse 

effects” of industrial technologies – especially 

pollution – via mandatory regulatory 

standards. More recently, “sustainability” 

has become the watchword, and economic 

incentives and other institutional mechanisms 

have come more into play as policy 

instruments. Nevertheless, the analytical 

focus of environmental policy remains overwhelmingly 

geared to the identification of 

particular environmental problems, with 

programmatic actions closely targeted to 

ameliorate them. 

Environmental policy is rooted in a profound 

political consensus. It draws on a 

30



3 

This issue is discussed 

at length in Heaton, 

Repetto and Sobin, 

Transforming 

Technology: An Agenda 

for Environmentally 

Sustainable Growth in 

the 21st Century, World 

Resources Institute, 

Washington, D.C., 

April 1991. 

> 

4 

The 1998-99 edition of 

World Resources: 

A Guide to the Global 

Environment (Oxford 

Press, 1998) gives a 

good overview of 

these issues. 

5 

Asia’s Clean 

Revolution, D. Angel 

and M. Rock, eds. 

Greenleaf Publishing, 

Sheffield, UK, 2000.See 

articles by editors and 

Heaton and 

Resosudarmo. 

large administrative apparatus and a high 

degree of professionalism, in environmental 

science, economics and law – all vibrant 

fields. This is not to say that environmental 

technology is ignored; quite the contrary. 

However, the approach to technology has 

tended to be reductionist, concentrating on 

the application and improvement of today’s 

“best available technology.” In short, diffusion 

and incrementalism – rather than radical 

innovation or systemic change – occupy 

the field. Rare indeed is the environmental 

policy-maker with license to focus on the 

transformative possibilities 3 of emerging 

technologies, much less one equipped with 

the tools to promote their realization. 

3. WHY A UNION MAKES SENSE 

Although environmental and innovation 

policy both stand to benefit from a marriage 

of their purposes and programs, environmental 

policy probably has the more to 

gain. This is so for two main reasons. First, 

the world’s environmental problems are far 

from solved – indeed they may in a number 

of respects be getting worse. And second, 

the potential of new technology to contribute 

to environmental improvement has 

been grossly underexploited. 

Looking at environmental quality from the 

vantage point of a highly developed country 

in northern Europe, it is easy enough to feel 

momentarily sanguine. The visible battles of 

the past – auto emissions, PCBs, and CFCS to 

name only a few – have been won. Local 

environmental indicators – air and water 

quality, toxic releases, waste streams – have 

indeed improved. But there are two less-perceived 

trends which cause this optimism to 

fall short. One is the shift from the shortterm, 

local, toxic-incident paradigm of environmental 

threat – which most policies were 

designed to combat – to a pattern of longterm, 

chronic, global and potentially irreversible 

environmental damage. It is not to 

minimize the gravity of global climate 

change and loss of biodiversity to say that 

other dangers may well loom just as large – 

alarming increases in male sterility, immune 

system deficiencies and continued largescale 

destruction of critical ecosystems, 

which may all presage global catastrophes 

of unprecedented scope 4 . 

The second cause for alarm is the ascendancy 

of developing countries as the principal 

source of environmental damage. It is 

not an overstatement to say that the world’s 

environmental future lies with the developing 

countries, particularly those in Asia – 

China, India and Indonesia above all – where 

rates of rates of increase in the pollutionintensity 

of production even outstrip population 

growth 5 . This pattern is intimately 

related to technology, as it is by and large 

the consumption patterns of Europe and 

North America that these societies are mimicking, 

and the accompanying resourceintensive, 

dirty technologies that they have 

adopted. 

Both of these trends suggest that the technology 

base on which the world has been 

relying is badly flawed. They suggest as well 

that the “environmental” technologies of 

today – generally “end-of-pipe” – will not 

be sufficient to balance the magnitude of 

these new threats. This is precisely why environmental 

policy needs innovation policy so 

much: to move it toward a new generation 

of environmentally oriented technologies, 

as well as to generate production technologies 

that are as environmentally benign as 

they are economically superior. 

The theoretical case for the expansion of 

innovation policy to include an environmental 

mission has been well made in the 

Position Paper. It positions an environmental 

innovation policy securely within the realm 

of market failures that justify government 

intervention, arguing that both the negative 

externalities of environmental degradation 

and the positive externalities in knowledge 

production are persuasive. Clearly this is true. 

There is another, partly theoretical, partly 

empirical argument to make. It rests on the 

perception that there is something very 

“wrong” in the innovation process for new 

environmental technologies, both in the private 

sector and in the allocation of public 

R&D funds. 

31


6 

See Heaton and Banks, 

“Toward a New 

Generation of 

Environmental 

Technology: The Need 

for Legislative 

Reform,”Journal 

of Industrial Ecology, 

Vol 1, No. 2, 1997. 

7 

Rene Kemp’s paper, 

presented at the OECD 

conference on the 

Environment and 

Innovation in Paris in 

June 200, makes this 

point for Europe as well: 

“Technology and 

Environmental Policy 

Innovation Effects of 

Past Policies and 

Suggestions for 

Improvement.” 

8 

U.S. Department 

of Commerce, Office 

of Technology Policy, 

The U.S. Environmental 

Industry, 

Washington D.C., 1998. 

9 

National Science and 

Technology Council, 

Technology for 

A Sustainable Future, 

Washington, D.C., 

July 1994. 

10 

See Heaton and 

Banks, “Toward a New 

Generation of Environmental 

Technology,” in 

Investing in Innovation 

(Branscomb and 

Keller, eds), MIT Press, 

Cambridge, MA 1998. 

11 

This is a figure put 

forward by the OECD 

in the mid 1990s. 

The overwhelming majority of environmental 

technologies – like all new technologies - 

- will be developed and deployed by private 

firms. What distinguishes environmental 

technology from other sectors, however, is 

the degree to which public policy “makes 

the market,” through regulatory standards 

and other means. If the signals from public 

policy are certain and strong, then one can 

expect innovation to follow. Indeed, private 

sector innovation may well lead public 

policy if the administrative apparatus is 

flexible enough to incorporate these innovations 

into its standards. Unfortunately, 

what we have seen in the U.S. suggests that 

the reverse is often the case 6, 7 . 

The chilling effect of technologically conservative 

environmental regulation on innovation 

in the private sector is demonstrated by 

data on the structure of the industry and its 

R&D portfolio. For the U.S., at least, we 

know that R&D in the environmental industry 

is significantly lower than in other 

technology-oriented sectors 8 . We have also 

come to refer to a “valley of death” for 

environmental technologies, in which many 

perish for lack of development funds while 

waiting to be sure that the market (i.e. regulators) 

will accept them 9 . To the extent 

that the structure of today’s environmental 

markets – throughout the OECD – is a predictor 

of tomorrow’s global technology, the 

outlook is also bleak. The table appended 

afterwards shows environmental industry 

revenues in the U.S., Europe and Japan. 

Most tellingly, one sees that only a trivial 

amount – no more than 0.5% – is prevention-oriented. 

In all three venues, water 

treatment works and utilities absorb about 

30% of the total. Beyond this, the pattern of 

expenditure derives from the regulatory 

regimes that are ubiquitously based on pollution 

control standards for individual 

media: air, water, waste. 

If the environmental industry is “underinvesting” 

in R&D for the next generation of 

technologies, one would hope that public 

investment would fill the void, given that 

such a situation offers the classic rationale 

for public R&D funding. This does not seem 

to be the case. Certainly in the U.S., the 

> 

prospect for any large-scale funding of 

environmental technology evaporated with 

the failure of the Clinton Administration’s 

Environmental Technology Initiative in the 

mid-1990s 10 . Environmental science – which 

is an entirely different matter – has continued 

to do well in terms of funding. What 

the situation is globally is difficult to say, as 

meaningful funding figures for environmental 

technology are hardly to come by – and 

valid international comparisons even more 

difficult. The table appended below on patterns 

of international R&D expenditures 

hardly begins to answer the question – 

clearly there is a need for better data on 

environmental technology. But the sense 

one gets is that its prominence within innovation 

policy universally falls woefully short 

of the need. 

What the above suggests is that there is a 

clear “market opportunity” for innovation 

policy to expand its focus in the direction of 

environmental technology. Beyond the theoretical 

rationale, the move would probably 

also be good politics, both in the industry 

and among the general public. There is also 

the argument to be made that support for 

environmental technology R&D is securely in 

line with innovation policy’s consistent focus 

on economic competitiveness. The global 

environmental industry already garners 

more than $400 billion per year 11 . And as 

environmental consciousness shows no signs 

of abating, it is a good bet that the positive 

environmental features of new products, 

processes and systems will increasingly be a 

competitive selling point 12 . 

4. AN IDEAL ENVIRONMENTAL 

INNOVATION POLICY 

Defining an ideal policy to promote environmental 

innovation is not terribly difficult. 

Almost two years ago, an OECD workshop 

put forward the following sensible threepart 

framework 13 : 

• Combine a flexible, incentive-based environmental 

policy – which focuses on the 

direction of technological change – with a 

technology policy designed to increase the 

rate of innovation 

32



12 

Stephan 

Schmidheiny’s 

Changing Course 

(MIT Press 1992) 

remains as eloquent 

a statement of this 

point as any. 

13 

Rapporteur’s Report, 

Workshop on the 

Environment and 

Innovation, 

OECD Directorate for 

Science, Technology 

and Industry, 

Paris, June 2000. 

14 

This is not to be 

taken as an 

environmental impact 

statement requirement 

for R&D, which could 

be disastrous 

and unworkable. 

15 

Of course, this is already 

occurring in some 

countries. But it is far 

from ubiquitous, or 

enough. 

16 

The ETI was actually 

developed by Vice 

President Gore, and 

was tied closely with 

his overall technology 

policy and the 

“Reinventing 

Government “plan.The 

history of the ETI – as 

well as other technology 

programs during 

the Clinton-Gore years 

are reviewed in 

Branscomb and Keller, 

Investing in 

Innovation, op. cit. 

The Heaton/Banks 

article covers ETI. 

• Embed innovation policy in environmental 

policy, by targeting policy instruments 

toward technologies that will offer new 

means of potential environmental prevention 

and remediation. 

• Embed environmental policy in innovation 

policy, by increasing the saliency of 

environmental criteria in policies and 

programs that support technology development. 

These generalities can also be put into more 

concrete terms. The basic idea from the first 

bullet is that environmental and innovation 

policies as they need to adopt a common 

complementary mission and mindset. But 

their separation of function must continue. 

For its part, environmental policy needs to 

adopt mechanisms that are much more flexible 

– economic instruments, incentives to 

go beyond compliance, special exemptions 

for new technologies and the like. But it 

needs to do so with a clear view of new trajectories 

for technological change in industry, 

agriculture, and infrastructure that can 

move in the direction of sustainability. 

Innovation policy, for its part, can continue 

with its traditional mechanisms – R&D funding, 

training, technology transfer – that 

increase flows through the innovation 

pipeline. But it should target a much greater 

proportion of these resources explicitly 

toward environmental sustainability. 

The second and third bullets are essentially 

expansions of the first. In the case of environmental 

policy, we should not expect 

its administrative authorities to engage 

directly in the development of new environmentally 

improved technologies. (I.e., 

the Ministry for the Environment does not 

become an R&D agency.) But what can be 

demanded is for environmental policy to 

move away from its all-too-frequent preference 

for the technological status quo. This 

implies developing a capability to envision 

possible technological futures and work 

toward their achievement. It means discarding 

standards that do no more than enfranchise 

today’s average end-of-pipe pollution 

control technology. And it requires taking 

on the underlying consumption patterns 

that are at the root of the problem. 

> 

In the case of innovation policy, it does not 

seem unreasonable to require R&D programs 

and projects to at least consciously 

consider the environmental implications of 

their research 14 . Thus, sustainability could 

well become an important funding criterion. 

It also seems essential to set aside a portion 

of the overall public R&D portfolio for the 

development of commercially viable environmental 

technology 15 . Two caveats need 

to be inserted with respect to this: that 

generic technology, not science is the goal; 

and that the development needs to take 

place in firms, not government labs. 

5. LESSONS FROM REALITY 

The U.S. and Japan have both made 

attempts to integrate some of the traditional 

mechanisms for the promotion of 

innovation with environmental policy. Their 

efforts, which have yielded radically different 

results, offer lessons from actual experience 

about the design of an environmental 

innovation policy. 

In the U.S., certain elements in the policy 

community – on both the left and right -- 

routinely oppose any government program 

to aid industry. Other elements – typically 

environmentalists – have tended to oppose 

non-regulatory approaches to environmental 

policy. These forces have long made it 

difficult to enact an environmental innovation 

policy; indeed, it was not until the early 

years of the Clinton-Gore Administration 

that one was proposed 16 . This Environmental 

Technology Initiative (ETI) had three main 

features: 1) a new program of R&D grants to 

companies for the development of innovative 

environmental technologies; 2) a broad 

regulatory reform movement, with the purpose 

of encouraging innovation; and 3) various 

commercial assistance measures to firms 

in the environmental industry. After a few 

years’ effort, the ETI was abandoned. The 

Congress had been unwilling to create 

authority or commit funds for the R&D program; 

regulatory reform, though continually 

debated was far from consensus; and the 

promotion of the environmental industry 

suffered from lack of a programmatic base. 

33


17 

The ATP gives R&D 

grants to industry for 

leading edge technologies.It 

came to light 

during the ETI that a 

surprising number of 

ATP’s projects had an 

important environmental 

dimension.But ATP 

was never enlisted in 

the initiative. 

18 

In former days 

known as MITI, 

the Ministry of 

International Trade 

and Industry, METI 

now stands for the 

Ministry of Economics, 

Technology and 

Industry. 

Many of the reasons for the ETI’s lack of success 

are peculiar to the politics and style of 

the Clinton era. But others may be more 

generalizable. Certainly ETI shows that it is 

dangerous to attempt a great deal. The 

combination of such large and diverse elements 

– an R&D program, regulatory reform 

and export promotion – makes the prospects 

for implementation difficult, especially in 

times of fiscal stringency. The patterns of 

responsibility for various parts of the ETI 

puzzle also show that responsibility and 

expertise must be assigned appropriately. 

Giving the proposed R&D program to regulators 

in the Environmental Protection 

Agency gave it neither a naturally receptive 

home nor expertise. Establishing an environmental 

export promotion program in the 

Department of Commerce thrust it into 

unfamiliar ground. Oddly enough, however, 

it was the Commerce Department’s 

Advanced Technology Program that was 

best able to show how environmental R&D 

could be funded in the general context of 

innovation policy 17 . 

In Japan, the willingness to create and 

endow new public programs with significant 

resources is high in comparison to the U.S. 

Often, public spending – in science and technology 

as well as elsewhere – is employed for 

its stimulus effect as well as to respond to 

urgent national needs. This may be said to be 

the case for NEDO, the New Energy and 

Industrial Development Organization, whose 

budget has moved up to over $2 billion during 

the last decade. NEDO’s initial mission 

was almost entirely energy, but its R&D support 

has broadened to include a wide range 

of critical industrial technologies such as, 

most recently, nanotechnology. 

Japan’s most focused effort to mount an 

environmental innovation policy occurred 

during the mid 1990s, with the establishment 

of RITE, the Research Institute for 

Innovative Technology for the Earth. With 

funding on the order of $1 billion per year, 

RITE’s projects tend to be basic in nature, 

often carried out by university personnel 

and very long-range. Perhaps the most 

remarkable feature of Japan’s posture in 

both of these cases is their placement within 

> 

the framework of the industrial development 

ministry, METI 18 . 

6. BECOMING ENGAGED 

At this juncture, environmental policy and 

innovation policy look much like two attractive, 

unattached people with a lot in common 

– except an acquaintanceship. As they 

have different constituencies and very different 

styles of operating, they are not likely 

to combine forces without assistance. 

Someone needs to convince them that their 

essential missions are basically the same: to 

change the course of today’s technology. 

The process of arranging this marriage presupposes 

the acceptance of a new concept 

or mindset: an environmental innovation 

policy. In meetings like this one, this concept 

may mean something, but to most of the 

world it does not. Papers, speeches, symposia 

and conversation in the relevant policy 

communities are needed to disseminate it. 

As to the operational side of the equation, in 

my judgment, the two parties to the engagement 

will continue to need separate domiciles 

for the foreseeable future – cohabitation 

is not likely to be harmonious. In the case of 

environmental policy, I have long been of the 

view that it needs a new household organization. 

Its administrative apparatus is typically 

structured according to the media in 

which environmental problems occur – air, 

water, waste, and so on. This problem-centric 

approach may inhibit thought about solutions 

– long-term changes to technology and 

the economy. If the agency subunits were 

focused on economic functions or sectors – 

housing, transport, energy, etc. – a radical 

reenvisioning of their technological trajectories 

might better be put forward. 

Whatever the organization of environmental 

policy, envisioning the technological 

future is where it could enlist the capabilities 

of innovation policy immediately, to 

best effect. What might work is a standing 

taskforce of individuals from both environmental 

and technology agencies, combined 

with industrial and academic representa- 

34



19 

Some experiments 

like this have been 

tried. “Backcasting” 

is a common rubric. 

These often focus on 

environmentally desirable 

futures.What is 

suggested here is to 

focus on the technological 

possibilities of 

the future, which is 

different. 

tives. They would be charged, first, with 

charting the technological possibilities for 

different sectors, and second, with proposing 

the policy mechanisms to move down 

these trajectories 19 . 

It almost goes without saying that new, 

flexible policy instruments are needed in environmental 

regulation to pave the way for 

technological innovation. I am certainly not 

against this. But I am certainly not convinced 

that a larger, faster flow of the type of environmental 

technology that is being produced 

today will be enough to wrench the world out 

of its current destructive pattern. Again, new 

technological pathways hold the key. 

For innovation policy, a marriage to environmental 

policy will require a new commitment, 

new criteria, new or redirected money 

– and a guest house. By this I mean that 

some sort of discrete operation should be 

mounted to provide financial support for 

the development of environmental technology 

in industry. Members of the environmental 

policy community should be an integral 

part of its decision-making apparatus. 

An alternative possibility – the application 

of environmental factors as one criterion in 

funding decisions throughout the innovation 

policy arena – would also make sense. 

But it has the drawback of not bringing the 

two policy communities together. 

One hesitates to recommend study. But in 

crafting an environmental innovation policy, 

at least one study is desperately 

needed: a thorough-going empirical analysis 

of environmental technology and its 

modes of innovation. Some study of the 

environmental industry began in the U.S. 

Office of Technology Assessment and in the 

OECD a half dozen years ago; it is my 

understanding that both are defunct. At 

the moment, very little is understood about 

patterns of environmental R&D – and 

hence we have hardly any idea of what the 

next generation of technology will look 

like. Flying blind into this future is not the 

way to create a fruitful relationship 

between these two policy areas. 

The last need I would cite is for gobetweens. 

The marriage between environmental 

and innovation policy will not spring 

up naturally. A cadre of change-agents 

needs to push it forward. Conferences like 

this one are the beginning. 

Revenues of the Environmental Industry:U.S., Europe, Japan 1996 

US Eur. Japan 

$bill. (%) $bill. (%) (%) $bill. 

Equipment 

Water and Chemicals 16 9.3 2.7 7.9 9.3 2.7 

Air Pollution Control 14 9.0 7.3 5.5 3.3 3.8 

Instruments, Info. 1.8 1.0 1.6 1.2 1.0 1.1 

Waste Management 10.7 6.2 9.1 6.8 8.6 9.9 

Process and Prevention 0.9 0.5 0.5 0.4 0.5 0.6 

Services 

Solid Waste Mgmt. 32.7 19.0 29.5 22.1 29.6 34.0 

Hazardous Waste Mgmt. 5.9 3.4 5.2 3.9 3.8 4.4 

Consulting, Engineering 14.2 8.3 8.4 6.3 1.1 1.3 

Remediation 8.3 4.8 3.7 2.8 1.1 1.3 

Analysis 1.2 0.7 1.0 0.7 0.5 0.6 

Water Treatment Works 24.6 14.3 21.8 16.3 9.6 11.0 

Resources 

Water Utilities 27.0 15.7 19.7 14.8 12.2 14.0 

Resource Recovery 11.6 6.8 13.6 10.2 9.2 10.6 

Environmental Energy 1.4 0.8 1.5 1.1 1.0 1.1 

Total 171.7 100% 133.4 100% 87.1 100% 

Source:US Department of Commerce, 1998, with data from Environmental Business International 

35


Appendix > 

Distribution of government R&D budget appropriations, by socloeconomic objective : 1997 or 1998 (percentages) 

Country (year of coverage) 

Objective US Japan a Germany France UK Italy Canada 

(1998) (1997) (1997) (1997) (1997) (1997) (1998) 

Total (millions of U.S. dollars b ) 73 569 18 309 15 619 13 178 8 887 6 211 3 395 

Agriculture, forestry 

and fishing 2.1 3.4 2.6 3.6 4.4 2.3 11.7 

Industrial developement 0.5 6.6 12.8 5.2 1.8 9.1 13.3 

Energy 1,3 20,2 3,5 4,8 0,7 4,0 5,7 

Infrastructure 2,5 2,7 1,6 0,6 1,7 0,4 4,2 

Transport 

and telecommunications 2,5 1,4 0,8 NA 0,3 NA 4,2 

Urban and rural planning 0,1 1,3 0,8 NA 1,4 NA 0,0 

Environmental protection 0,8 0,6 3,7 2,0 2,2 2,5 3,3 

Health 19,3 4,0 3,4 5,3 14,5 8,5 9,5 

Social development 

and services 1,0 0,9 2,4 0,9 2,0 4,5 3,6 

Earth and atmosphere 1,3 1,3 2,0 0,7 1,7 1,4 4,9 

Advancement of knowledge 5,9 48,2 53,6 35,7 30,3 59,6 27,1 

Advancement of research 5,9 10,8 15,6 19,2 11,8 12,1 8,4 

General university funds – 37,4 38,1 16,5 18,5 47,4 18,7 

Civil Space 11,1 6,3 4,8 11,0 2,7 4,0 9,2 

Defense 54,1 5,8 9,6 27,7 37,7 3,5 5,0 

Not elsewhere classified 0,0 0,0 0,0 2,4 0,4 0,0 2,6 

NA = not separately available but included in subtotal; 

– = the United States does not have an equivalent to general university funds 

NOTES: Percentage may not add to 100 because of rounding. U.S. data are based on budget authority. For all countries, 

because of the inclusion of general university funds and slight differences in accounting practices, the distribution of government 

among socioeconomic objectives may not completely reflect the actual distribution of government - funded research in 

particular fiels. 

a 

Japanese data are based on science and technology budget data, which include items other than R&D. Such items are a small 

proportion of the budget; therefore, the data may still be used as an approximate indicator of relative government emphasis 

on R&D by objective. 

b 

Conversions of foreign currencies to U.S. dollars are calculated with OECD purchasing power parity exchange rates. (See 

appendix table 2-2) 

SOURCES: National Science Foundation, Division of Science Ressources Studies (NSF/SRS), Federal R&D Funding by budget 

Function: Fiscal years 1998-2000, NSF 00-303 (Arlington, VA: 2000); Oraganisation for Economic Co-operation and 

Development, Basic Science and Technology Statistics (unpublished tabulations). 

Science & Engineering Indicators - 2000 

36


AN INTEGRATED POLICY 

FOR INNOVATION 

FOR THE ENVIRONMENT 

RENÉ KEMP 

Merit 

René Kemp (PhD in economics) is senior 

research fellow at Maastricht Economic 

Research Institute on Innovation and 

Technology (MERIT) from Maastricht 

University in the Netherlands and former 

research director of STEP in Oslo. He has 

worked on clean technology adoption, 

green innovation policy and led several 

research projects about environmental policy 

and innovation. His current work is on 

transitions to sustainability, focussing on 

how these transitions may be managed (see 

www.meritbbs.unimaas.nl/rkemp). He can 

be contacted at r.kemp@merit.unimaas.nl 

37

An Integrated Policy for Innovation for the Environment 


> 

1 

It would be interesting 

to investigate the relative 

contribution of 

environmental technologies 

(technologies 

that are environmentally 

motivated) and 

normal technologies to 

environmental improvement 

for specific 

types of pollution. 

Energy saving in manufacturing 

has almost 

certainly more to do 

with energy prices than 

with energy covenants. 

Energy savings in 

consumer devices and 

housing owes probably 

more to energy efficiency 

regulations and 

subsidies coupled to 

energy labels. Greene 

(1990) tested the relative 

influence of 

energy prices and CAFE 

on automobile fuel efficiency 

and found that 

the regulations were 

twice as important 

than energy prices. 

1. INNOVATION EFFECTS 

OF INNOVATION POLICIES 

All OECD countries have an innovation policy, 

superseding the science and research 

policies of the past. Such policies are oriented 

for the most part to private companies 

and to economic goals, not environmental 

sustainability goals. The goal of the 

policies is to help companies innovate. New 

technology is also important for environmental 

gains. It may alleviate tradeoffs 

between economic wellbeing and environmental 

quality (Jaffe and Stavins, 1990, p.1) 

or produce environmental benefits as a 

gratis effect. Advanced technologies often 

are more resource efficient and generic 

technologies such as information and communication 

technologies may be applied to 

environmental purposes. Sensors are a good 

example of this. They aid waste management 

and help to limit pollution by offering 

information for process operators. The environmental 

benefits from innovations may 

thus be accidental, in the sense that the 

original research leading to these innovations 

was not environmentally motivated or 

because the innovation are inherently more 

environmentally benign 1 . 

This shows that environmental technologies 

should not be privileged as the source for 

achieving environmental gains (Williams 

and Markusson, 2002) There are other 

sources for obtaining environmental improvement 

as well, such as new processes 

that are more resource efficient, a shift 

towards cleaner industries and behavioural 

change. This suggests that it is better to talk 

about “innovation for the environment” 

which includes innovations that are not 

environmentally motivated (Berkhout, 

2002). There is also second reason for not 

privileging environmental technologies 

which is that they may lead to a transfer of 

environmental problems or add to the costs 

of the company. This holds true for end-ofpipe 

technologies that cause a waste problem 

unless the captured pollution and 

treated emissions are re-used. Whether the 

captured or treated pollution is utilized 

depends on the existence of markets for 

waste and the economics of re-use vis-à-vis 

the costs of waste disposal. Fly ash is used in 

cement production and sewage sludge from 

communal wastewater treatment plants is 

burned in cement kilns. 

Environmental technology often is supported 

through special innovation programmes. 

An example of such a programme 

in the Netherlands is the subsidy programme 

Milieu en Technologie. In this programme 

technical research is subsidised on a project 

basis but there are also subsidies for making 

a business plan for innovations for the environment. 

The programme is a successor of 

the “Stimuleringsregeling Milieu-technologie” 

(STIR-MT). An important question is: 

Did R&D subsidies stimulate firms to undertake 

research in environ-mental technology 

they would not have done otherwise The 

evidence that is available suggests that R&D 

subsidies in the Netherlands for environmental 

technology have been of limited 

effectiveness. According to Olsthoorn, 

Oosterhuis and Verbruggen (1992, p.18) the 

STIR-MT did not elicit new research projects. 

This is in line with the observation by de 

Jong and van der Ven (1985, pp.78-79) that 

innovator firms develop environmentally 

beneficial technologies not because a subsidy 

is available but because they believe a 

market exists for the new technology. The 

conclusion is at odds with two other evaluation 

studies, quoted in Cramer et al. (1990), 

that find that of the 10 projects that 

received financial support under the Clean 

Technology programme in the Netherlands, 

five would never have been initiated. 

However, it turned out that many of the 

projects funded under the programme were 

second-rate projects: of the ten projects only 

seven were technically successful and only 

four of them were applied in practice. 

The outcomes of the German research programme 

of the BMBF for environmental 

research and environmental technology are 

more favourable. A total of 1402 projects 

were supported, receiving 646 million 

which amounted to an average subsidy 

quota of 51 %. 

The funds were distributed among various 

environmental fields as shown in Figure 1. 

Most of the projects consisted of treatment 

39


2 

The weak incentive 

for developing technologies 

with greater 

environmental efficiencies 

has to do with the 

fact that most regulations 

are based on 

existing technologies. 

Such regulations provide 

no incentive to go 

beyond existing control 

efficiencies, they only 

provide an incentive 

to develop solutions 

that are less expensive 

than those in use. 

technology, or cleaning technology, keeping 

in with the past orientation of environmental 

policy towards end-of-pipe solutions. 

40% of the projects were oriented to water 

pollution, 33% to waste management, 6% 

to air pollution and 15% to clean technologies 

and environmentally sound products. 

The share of 15% for clean technologies and 

environmentally sound products, in which 

pollution is prevented rather than dealt 

with, is very low. 

To what extent did the programme encourage 

companies to do research they 

would not have done otherwise, or 

encourage them to do projects earlier or 

in a more elaborated way The findings 

from an evaluation study by ISI (Angerer 

et al., 1997) are quite positive. 36% of the 

projects would not have done without the 

support and 38% of the projects were 

done in a more elaborated way, with 

greater scope and budget. Only 5% of the 

projects would have been done in the 

same way (free rider effect). Of course 

there is a free rider effect in all projects 

that did not depend on the subsidy for 

their undertaken.The evaluators noted 

that the free rider effect was low compared 

to other programmes. The high 

additionality of the German programme is 

probably caused by the fact that the subsidy 

was quite high (51% on average).A 

second explanation might be that many of 

the projects are end-of-pipe solutions for 

which the incentives are less favourable 

than for normal innovations 2 . 

Figure 1 > Distribution of funds of the German Environmental Technology programme between 1980 and 1992 (source: Angerer, 2002) 

Water pollution prevention 

and waste water 

treatment 


Waste management and 

hazardous waste site 

remediation 

33 

Clean technologies and 

environmetally sound 

products 

15 

Share of total funds in % 

Air pollution control 

6 

0 5 10 15 20 25 30 35 40 45 

Figure 2 > The additionality of the German Environmental Technology programme between 1980 and 1992 (source: Angerer, 2002) 

Project would not be 

done without support 

36 

Project would be done 

but with smaler scope 

38 

Project would be done 

over longer period 

11 

Project would be done at 

later date 

13 

No additionallity at all 

5 

0 5 10 15 20 25 30 35 40 

40



3 

The project was more 

than a subsidy 

programme. The 

programme brought 

together firms with an 

environmental problem 

and firms and research 

institutes that could 

provides solutions to 

these problems. 

4 

EET stands for 

Economy, Environment 

and Technology. 

Additionality is just one element to evaluate 

programmes. Another criterion is whether 

the innovations are used and the environmental 

benefit connected with their use. 

There is no information on the environmental 

impact and actual use of the innovations 

supported by the German programme but 

the evaluation study by ISI found that 16% 

of the projects had a decisive influence on 

the design or implementation of environmental 

regulation. A further 20% were supportive 

but not decisive to policy changes. 

This compares favourably to the experiences 

with the Sustainable Technology Programme 

called DTO in the Netherlands. The additionality 

of the programme was very high but 

the results were not utilized for policy nor 

utilized by industry. DTO is an interdepartmental 

research programme for sustainable 

technologies, which ran from 1993-1997. The 

goal of the programme was to identify and 

work towards technology options offering a 

factor 20 improvement in environmental 

efficiency while satisfying human needs in 

terms of nutrition, transport, housing, and 

water supply and protection. Industry was an 

important actor in the programme. Industrial 

opinion leaders were asked to think about 

long-term technological solutions offering 

magnitude environmental benefits. They 

were selected for their imagination and their 

position within industry, because the programme 

wanted to influence the industrial 

research agenda. Many of the industry people 

were research directors. In total 25 million 

guilders (€11.3 mln) was spent under 

the programme by the Dutch government. 

The financial contribution from industry was 

low, about 10% of the costs of the illustration 

projects, in the form of money and time. 

The DTO programme led to the development 

and articulation of 14 illustration processes 

for sustainability. 

The project was successful in tapping people’s 

mind and imagination and led to ideas 

for system innovation and networks of collaboration 

but failed to influence industries’ 

research agenda in an important way for the 

simple reason that the technologies were 

not economical. Their use would require a 

change in the frame conditions, giving the 

sustainable technologies a competitive edge. 

A 5 million guilder (€2.3 mln) programme of 

knowledge transfer called DTO-KOV followed 

the programme but like the first programme 

this programme did not address the 

root problem of unfavourable frame conditions. 

The absence of a pull mechanism frustrated 

the further development of these 

technologies and the occurrence of processes 

of co-evolution resulting in transformations 

and the creation of new systems. 

A programme that was very successful in 

terms of the immediate utilisation of the 

results is the Danish Clean Technology 

Development Programme, described in 

Georg et al. (1992). Under the programme, 

industries, private and semi-governmental 

research institutions could apply for financial 

aid for developing and implementing clean 

technology. The programme was oriented at 

stimulating preventive process solutions and 

cooperation among technology suppliers, 

research institutes, consultancy firms and 

users. The Danish Environmental Protection 

Agency played an active role in selecting 

environmentally beneficial projects and in 

finding the right partner with whom to cooperate. 

That is, the agency acted as a match 

maker to elicit environmentally innovative 

solutions, something that previous subsidy 

programmes had failed to do 3 . According to 

the authors, the Danish pro-gramme was a 

success. In almost all cases, appropriate technical 

solutions were found for the environment 

problems at hand. In more than half of 

the projects, substantial environmental 

improvements were achieved at low costs. 

Some projects led to net economic gains for 

the polluting firms. 

A quite novel programme is the EET programme 

in the Netherlands. EET is a 

research programme for breakthrough innovations 

offering economic and environmental 

benefits in a time space of 5-20 years 4 . 

So far (at the beginning of 2001) 70 projects 

have been funded (plus 38 KIEM projects, 

technical feasibility studies). An average EET 

project has a size of 8 million guilders 

(€3.6 million) of which half is funded by 

the government (Willems & van den 

Wildenberg, 2000). The minimal size is 1 mln 

41


5 

A programme for 

sustainable technology 

that is less technology 

oriented is NIDO 

(Nationaal Initiatief 

Duurzame 

Ontwikkeling). NIDO is 

a programme which 

supports ‘jump projects’, 

initiatives which 

offer sustainability 

benefits. It is less technology 

focussed than 

DTO and EET and more 

oriented towards practical 

implementation. 

The NIDO budget for 

2001 is 8.5 million guilders 

(3.9 million €). 

The private contribution 

to these projects is 

3.5 million guilders 

(1.6 million €). Apart 

from supporting the 

programmes financially 

NIDO helps participating 

parties with 

obtaining additional 

funds and the dissemination 

of knowledge. 

The small size of the 

projects and short 

period of support (2 

years) means that for 

some type of changes 

(such as the shift to an 

emission-low energy 

system or a different 

type of transport systems) 

the support from 

NIDO will be too little 

to have much of an 

impact. 

6 

Transition agendas 

are an element of 

transition management 

for sustainability, 

a concept 

explained later on. 

guilders (€0.45 mln). The total size of the 

EET projects funded between 1995 and 2001 

is 529 million guilders (€240 mln) of which 

the government paid 280 million guilders. 

It is a very large and perhaps unique subsidy 

programme through its focus on both economic 

and environmental benefits. 

EET complements environmental technology 

programmes that have a more narrow focus 

on environmental benefits and that offer little 

opportunities for system innovation. The 

focus on radical innovation is good, given 

the long development times for such innovation 

and positive spillover effects. It is a 

programme for high-risk, high reward projects. 

About 10% of the projects are believed 

to be economical, but profits from those 

projects are estimated at 1 billion € per 

annum, giving extremely good value for taxpayers 

money. The selection of the projects 

is done on the basis of the project’s market 

potential and environmental benefits, but 

the programme is not linked to transition 

agendas or to road maps made by industrial 

actors. As it stands, the EET programme is 

not well-aligned with environmental policy 

and oriented towards one aspect of system 

innovation: technology. 

Suggestions for innovation policy for the 

environment 

The above shows that the experiences of 

innovation policy in promoting innovation 

for the environment are mixed. Two key 

considerations for innovation policy are 

additionality of support and utilization of 

results. One way of making sure that the 

results will be utilized is by funding projects 

that are close to the market or by reducing 

the government share of support. Many 

countries have followed this road. But this 

disadvantages projects for which no market 

yet exists, as is often case for environmental 

innovations. The additionality of support for 

projects that bound to be economical may 

also be low. There is a conflict between 

achieving greater additionality of support 

and commerciability of the results. The first 

suggests the support of high-risk, highreward 

projects that are unlikely to produce 

immediate results, the second just the opposite. 

This really suggests the use of multiple 

programmes. Especially in the area of environmental 

innovation there is a need for a 

portfolio of programmes. Most countries 

have programmes for environmental technologies 

and sometimes clean technology. 

Few countries have programmes for “factor 

10 innovations” (or even factor 20 solutions). 

The Netherlands is an exception, 

although the programmes are very technology-oriented 

5 . Support for radical innovations, 

clean technology and environmental 

technologies is probably best organized 

through separate programmes. It is hard to 

factor in environmental criteria in normal 

research programmes because it is hard for 

evaluators to assess the environmental merits 

(noted by Ken Guy at the 6CP conference 

Innovation Policy and sustainable development, 

Brussels, 28 febr-1 march). 

There is also a need for a more integrated 

innovation policy for the environment, in 

order to increase the usefulness of innovation 

research and support policies. There are 

several ways to do this. 

First, innovation policy for the environment 

could be targeted to areas in which innovation 

is needed. Innovation support could 

also be informed by sustainability agendas 

and by transition agendas in which the sustainability 

goals are translated into specific 

policy goals 6 . Countries without transition 

agendas could set up task forces to this end. 

Second, the experiences with the Dutch 

programme for sustainable technologies 

(DTO) suggest that there is a need for support 

programmes that go beyond the support 

of research. There is a need for programmes 

for system innovation such as 

integrated mobility or industrial ecology, to 

explore visions of sustainability through 

research and the real use of new technologies 

in society. Such support programmes 

should be time-limited and flexible to prevent 

the creation of “white elephants”. 

System innovation in the sociotechnical 

realm involves changes in sociotechnical 

systems beyond a change in (technical) 

components. It is associated with new linkages, 

new knowledge, different rules and 

42



7 

A related distinction is 

that between sustaining 

innovations and 

disrupting innovations 

(Christensen, 2000). 

8 

AThree other 

examples, described in 

Ashford et al. (2001) 

are: biomass-based 

chemistry, multiple 

sustainable land-use 

(the integration of the 

agricultural function 

with other functions in 

rural areas) and flexible, 

modular manufactured 

construction. 

9 

The factors are 

hypothetical and will 

differ per case. Some 

estimates are given in 

Weaver et al. (2000). 

10 

This suggestion is 

made by Angerer in 

his paper for the 

BLUEPRINT workshop 

on environmental 

innovation systems. 

The paper can be 

downloaded from 

www. 

blueprint-network.net 

roles, a new ‘logic of appropriateness’, and 

sometimes new organisations 7 . System innovation 

usually consists of a combination of 

new and old components and may even consist 

of a novel combination of old components, 

as in the case of industrial ecology – 

the closing of material streams through the 

use of waste output from one company by 

another 8 . 

It is important to have programmes for system 

innovation because system innovation 

may provide factor 10 improvements in environmental 

impact compared to the factor 2 

improvements associated with incremental 

changes or factor 5 improvements connected 

with partial system design. The 

hypothesised time path of environmental 

benefit for various types of innovation is 

shown in figure 3 9 . 

System innovation is not about environmental 

innovation but about system changes 

offering environmental benefits alongside 

other types of benefits – economic ones and 

social ones although there may be tradeoffs 

especially in the early phase. 

Third, more attention should be paid to the 

hazards, negative side-effects of new technologies 

for the environment and society at 

large. Part of the budget for innovation support 

could be earmarked for environmental 

assessment or more generally technology 

assessment. The share does not have to be 

high, 1% could already suffice 10 . 

Fourth, apart from the support for demonstration 

projects there should be programmes 

for technology experimentation. 

Sustainability requires experimentation. The 

primary goal of experiments should be to 

learn: about the product but also about the 

sustainability effects. Technology experimentation 

is an important element of programmes 

for system innovation. They should 

be set up towards the goal of learning and 

not as demonstration projects. In the book 

Experimenting for sustainable transport. 

The approach of strategic niche management, 

analysing 8 experiments with sustainable 

transport, it was found that more could 

have been learned if they were set up 

differently. Little was learned about different 

configurations as the design was fixed. 

Most of the learning consisted of technical 

learning. The projects were largely selfcontained; 

they hardly stimulated 2nd order 

learning and wider learning processes and 

as a result of this contributed little to 

processes of co-evolution involving social 

change (Hoogma et al., 2002). With one 

exception, they were not undertaken from a 

long-term sustainability vision, to be 

explored through a series of experiments. 

Fifth, apart from embedding environmental 

policy in innovation policy through the 

above policies, there is also a need for 

embedding innovation policy in environmental 

policy. Innovation research findings 

about enhanced possibilities to deal with 

Figure 3 > 

The time path of environmental benefit for various types of innovation (based on Weterings) 

Improvement in environmental efficiency 

Factor 10 

System innovation 

= new system 

Factor 5 

Partial system redesign 

Factor 2 

System optimisation 

5 10 20 

Time horizon (years) 

43


environmental problems technologically 

could be used for setting long-term emission 

limits – to create a market for new environmental 

solutions. Ashford (1994) has argued 

for a technology options analysis for environmental 

policy, which appears a very good 

idea. Of course a challenging environmental 

policy will also promote innovation, as 

stated by Ashford in various publications, 

but it is hard to pull appropriate innovation 

through environmental policy. Such policies 

can be disruptive in the sense of compromising 

user services (such as detergents with 

less washing performance) or creating a 

large economic cost. The use of catalytic 

converters in response to regulation led to a 

fuel penalty (an increase in fuel consumption) 

of as much as 7 miles per gallon in 1981 

(White, 1982), whereas the converters themselves 

were ineffective at low temperature 

(during the first kilometres), which meant 

that they were ineffective at short trips. The 

superiority of market-based instruments in 

pulling innovation still needs to be demonstrated—which 

is one reason why they 

should be used, apart from the efficiency 

gains associated with their use. 

These are some suggestions as to how innovation 

policy may be oriented towards environmental 

goals and sustainability. Let me 

recapitulate them. 

The first suggestion is to target innovation 

policy to areas in which innovation for the 

environment is needed. 

Suggestion 2 is to have programmes for 

system innovation besides programmes for 

clean technology and environmental technology. 

Suggestion 3 is to do environmental assessment 

and broader system assessment. 

Suggestion 4 is to support sociotechnical 

experiments for sustainability which are 

more than market learning experiments. 

Suggestion 5 is to disseminate the findings 

from innovation research towards regulators 

who may use such findings for setting 

future standards and goals. 

> 

The next section will deal with the innovation 

effects of environmental policy and 

how these may be enhanced. 

2. THE INNOVATION EFFECTS OF 

ENVIRONMENTAL POLICY 

There is much talk about environmental 

policies being faulty. Past policies are being 

criticised for failing to achieve environmental 

goals (the environmentalist complaint), 

for being overly expensive (the industrialist 

complaint) and for failing to encourage 

innovation and dynamic efficiency (the complaint 

of innovation students). In my book 

Environmental Policy and Technical Change I 

give an overview of the effects of past environmental 

policies on technical change. 

I found several examples of environmental 

policies that stimulated innovation but the 

common technology response is the use of 

end-of-pipe solutions and incremental 

process changes offering limited environmental 

gains. Several harmful substances 

have been substituted through product bans 

but there were few examples of radical 

change in response to environmental policy. 

This begs the question: why did the policies 

fail to promote more radical innovation and 

dynamic efficiency One explanation – wellrecognised 

in the economic literature – is 

the capture of government policies by special 

interests.This indeed is the case. 

Companies have an influence on the regulations 

to which they will be held and the way 

in which they are implemented. A second, 

complementary, explanation is that regulations 

are often based on available solutions. 

Sometimes particular solutions are prescribed. 

This stifles innovation. There have 

been successful attempts though to stimulate 

innovation. PCBs have been phased out, 

lead has been removed from gasoline, and 

CFCs have been successfully substituted by 

HCFCs and HFCs. There was innovation in 

end-of-pipe technology and innovation in 

products such as detergents, paints and cars 

as a direct response to environmental policy 

but on the whole environmental policy has 

been too unchallenging to provoke innovative 

responses. 

44



Also new policy instruments such as innovation 

waivers and environmental covenants 

failed to bring forth innovative solutions 

(Kemp, 2000). Probably more could have 

been achieved through better designed 

instruments. Perhaps economic instruments 

will encourage innovation more by providing 

an incentive to go beyond what is required 

by law in terms of emission reductions. I am 

a bit doubtful whether they will. In order to 

stimulate technological innovation, probably 

a more focused approach may be needed. 

One way of doing this is through R&D programmes 

for environ-mental technologies or 

more environmentally benign energy technologies. 

But as noted with R&D support 

there is always the danger that the programmes 

promote second-rate technologies 

and provide windfall gains to the recipients. 

Another strategy is by specifying strict environmental 

stan-dards that require the development 

of new technologies. However, this 

should be done only in situations where the 

environmen-tal risks are large and acute and 

when there is consen-sus about the most 

viable technological solution or trajectory. If 

there is no such consen-sus there is a danger 

that technology-forcing standards lock 

industry into overly expen-sive and suboptimal 

tech-nical solutions. In such circumstances 

there is a need for further research 

and experiments to learn more about the 

technological possibilities, about the disadvantages 

of particular solutions (and how 

they may be overcome), the economic costs 

and environmental gains of the technologies, 

and their acceptability to society. When 

using direct regulation, policy makers should 

give careful attention to the actual design 

of standards: their strictness, differentiation, 

timing, ad-ministration, flexibility and 

enforcement. The experiences in the US with 

innovation waivers and tradeable permits 

(described in Hahn, 1989) illustrate that the 

ways in which the instruments are designed 

and implemented are important determinants 

of the technological responses of 

industry. It is important to find a balance 

between certainty, stringency and flexibility. 

Technology compacts, described in Banks 

and Heaton (1995), appear useful way to 

promote technological innovation by setting 

an agenda of phased increments of technological 

change. But there is the danger of 

strategic behaviour on the part of industry 

that may claim that it is impossible to 

develop technology that is both environmentally 

superior and economically feasible. 

Generally, policy instruments should be 

combined with one another to benefit from 

synergistic effects (Kemp, 1997; Blazejczak 

et al. 1999; Norberg-Böhm, 1999; OECD, 

2000). A combination of standards with 

economic instru-ments is particularly useful 

by combining effectiveness with efficiency. 

A good example of an effective and economically 

efficient environmental policy are 

the US automobile fuel economy standards 

which set progressive fuel economy targets 

for automobile manufacturers in the 1979- 

85 period under penalty of a fine of $50 per 

car sold for each mile per gallon of shortfall. 

This worked very well as shown by Greene 

(1990). Tradeable pollution permits also 

deserve to be used more as they too combine 

effectiveness with efficiency. At this moment 

a nation-wide market exists for SO2 in the 

US where utilities can trade SO2 rights at the 

Chicago Board of Trade. According to a 

recent evaluation the tradeable permits for 

sulphur dioxide emissions reduce the costs 

of the 1990 acid rain pro-gramme by 55 

percent (Ellerman et al, 2000), quoted in 

Cramton, 2000) but this system has failed to 

promote innovation (Burtraw, 2000). 

In my view there is a need for government 

authorities to be explicitly concerned with 

technical change (rather than implicitly 

through a change in the economic frame 

conditions) and to be concerned with institutional 

arrangements beyond the choice of 

policy instruments. Government has to act as 

a change agent, making sure that societies’ 

problem solving capabilities are utilized. 

This requires different roles for policy 

makers: that of a sponsor, planner, regulator, 

matchmaker, alignment actor and a ‘creative 

game regulator’ who intervenes cleverly in 

regulatory games. 

I also feel there is a need for policy to be 

oriented towards system innovation involving 

structural change rather than the develop- 

45


11 

The project team 

consisted of Jan 

Rotmans, Marjolein 

van Asselt and Kirsten 

Molendijk from ICIS, 

René Kemp from 

MERIT (in Maastricht), 

Frank Geels from the 

University of Twente 

and Geert Verbong 

from TUE. 

12 

The idea of development 

rounds comes 

from Teisman (2000). 

> 

ment and diffusion of environmental technologies. 

This should be done in a forwardlooking, 

reflexive manner. A planning and 

implementation approach won’t work 

because apart from practical problems of 

implementation it may be insufficiently 

responsive to peoples needs and wants. In 

general one should try to utilize dynamics in 

technology and markets and bottom-up initiatives 

with new technology and systems 

for achieving long-term change that goes 

beyond a pale greening of existing systems. 

Policies should be used to stimulate both 

incremental change and step changes. Ideas 

about how to do this are worked in a policy 

model called transition management. 

3. TRANSITION MANAGEMENT 

The experiences with the above Dutch programmes 

(especially DTO) and the continuing 

non-sustainability of fossil-based energy 

and transport systems and risk-prone chemistry 

and food production led Dutch policy 

makers to look for a more integrated and 

comprehensive approach to work towards 

transitions. Together with others, I was 

involved in a project for the 4th National 

Environmental Policy plan (NMP-4) analysing 

the concept of transition and the possibilities 

for transition management 11 . 

Transition management can be defined as 

a deliberate attempt to bring about structural 

change in a stepwise manner. 

Transition management for sustainability is 

oriented towards long-term transition 

goals but does not attempt to achieve a 

particular transition goal at all cost. It is 

based on a philosophy of modulation: it 

tries to utilise existing dynamics and orient 

these dynamics to transition goals that are 

chosen by society. The policies to further 

the goals are periodically adjusted in development 

rounds 12 . The goals are not set into 

stone but may also change in the course of 

time. Existing and possible policy actions 

are evaluated against two criteria: first, the 

immediate contribution to policy goals (for 

example in terms of kilotons of CO2 reduction 

and reduced vulnerability through 

climate change adaptation measures), and 

second, the contribution of the policies to 

the overall transition process. Policies thus 

have a content goal and a process goal. 

Learning, maintaining variety and institutional 

change are important policy aims 

and policy goals are used as means. The use 

of development rounds brings flexibility to 

the process, without losing a long-term 

focus. 

A schematic view of transition management 

is given in figure 4. 

Figure 4 > 

Short-term versus long-term policy 

Current policy: only short term goals 

Political 

margins for 

change 

Transition 

goals 

State of 

development 

of solutions 

reassessment reassessment reassessment 

Transition management: 

short term and long-term goals and re-evaluation 

Sustainability 

visions 

46



Figure 5 > Role of the government in various phases of a transition process 13 

13 

Strategic niche 

management is the 

exploiting windows of opportunity 

creation and management 

of a niche for an 

innovation with the 

aim of promoting processes 

of co-evolution. 

The innovation is 

used by real users. 

This helps to promote 

interactive learning 

(between suppliers 

and users) and helps 

to build product constituencies 

(which 

Pre-development phase: 

Keeping a wide playing field 

Promoting participative discussions 

Strategic niche management 

Acceleration phase: 

Selection of options 

Policies for structural change 

Monitoring of outcomes 

Adjustment 

Take off phase: 

Mobilizing actors programmes for system innovation quality images 

Internalisation of external costs 

Stabilisation phase: 

Stimulating a new regime (consolidation) 

include policy actors). 

The approach of SNM 

is described in Kemp 

et al. (1998a), Kemp et 

al. (1998b), Kemp et 

al. (2001), Elzen et al. 

(2001) and Hoogma et 

al. (2001). Loiter and 

Norberg-Bohm (1999) 

also argue for ‘niche 

policies’. 

Transition management is based on a twopronged 

strategy. It is oriented towards 

both system improvement (improvement of 

an existing trajectory) and system innovation 

(representing a new trajectory of development 

or transformation). The role of government 

differs per transition phase. For 

example, in the predevelopment stages 

there is a need for social experimentation 

and creating support for a transition programme. 

In the acceleration phase there is a 

special need for controlling the side effects 

of large-scale application of new technologies. 

Throughout the entire transition the 

external costs of technologies should be 

reflected in prices and windows of opportunity 

should be utilized. The changing nature 

of policy is shown in Figure 5. 

Transition management breaks with the 

planning and implementation model and 

policies aimed at achieving particular outcomes. 

It is based on a process-oriented philosophy 

of modulation and long-term goals. 

Both elements help to deal with complexity 

and uncertainty in a constructive way. 

Transition management is a form of process 

management against a set of goals set by 

society whose problem solving capabilities 

are mobilised and translated into a transition 

programme, which is legitimised 

through the political process. 

Key elements of transition management are: 

• Long-term thinking (at least 25 years) as a 

framework for shaping short-term policy 

• Thinking in terms of more than one 

domain (multi-domain) and different 

actors (multi-actor) at different scale levels 

(multi-level); how developments at one 

level with one type of actors gel with 

developments in other domains 

• A focus on learning and a special learning 

philosophy (learning-by-doing and doingby-learning) 

• An orientation towards system innovation 

• Learning about a variety of options (which 

requires a wide playing field). 

Transition management does not aim to 

realize a particular path. It may be enough 

to improve existing systems;problems may 

also turn out to be less severe than at first 

thought. 

Transition management is not an instrumental 

activity. The actual policies are the outcome 

of political negotiations and processes 

of co-evolution which inform further steps, 

but the basis steps are: 

The transition goal 

This consists of a basket of goals. The transition 

goal is multi-dimensional and should 

not be defined in a narrowly technological 

sense. The goals should be democratically 

chosen and based on integrated risk analysis. 

This will constitute a radical break with current 

practice in environmental policy where 

quantitative standards are set on the basis of 

studies of social risk, and adjusted for political 

expediency. Risk-based target setting is 

47


Figure 6 > 

14 

The idea of a policy 

corridor is described 

and applied in 

Rotmans and den 

Elzen (1993). 

Optimalisation and innovation 

Optimalisation 

and innovation 

doomed to fail when many issues are at 

stake and when the associated risks cannot 

easily be expressed in fixed, purely quantitative 

objectives. This holds true for climate 

change but also for sustainable transport. 

Transition management relies on integrated 

risk analysis and the setting of minimum levels 

for certain stocks (e.g. health, ecosystem 

diversity and capital) and aspiration levels. 

The estimates of various types of risk are subjective, 

since the risks are surrounded by structural 

uncertainties, legitimating the incorporation 

of various perspectives (van Asselt, 

2000). The net result is a policy corridor for 

key variables, indicating the margins within 

which the risks are considered acceptable 14 . 

The use of transition visions 

Transition management is based on longterm 

visions that function as a framework 

and a frame for formulating short-term and 

long-term objectives and evaluating existing 

policy. To adumbrate transitional pathways, 

these visions must be appealing and imaginative 

and be supported by a broad range of 

actors. Inspiring final visions are useful for 

mobilizing social actors (such as ‘underground 

transport’ and ‘multifunctional land 

use’), although they should also be realistic 

about innovation levels within the social 

subsystem in question. 

The ‘basket’ of visions can be adjusted as a 

result of what has been learned by the players 

in the various transition experiments. 

The participatory transition process is thus a 

goal-seeking process, where both the transition 

goals and visions change over time. This 

differs from so-called ‘blueprint’ thinking, 

which operates from a fixed notion of final 

goals and corresponding visions. 

Interim objectives 

Figure 6 shows the similarities and differences 

between current policy-making and 

transition management. In each case, 

interim objectives are used. However, in 

transition management these are derived 

from the long-term objectives (through socalled 

‘backcasting’), and contain qualitative 

as well as semi-quantitative measures. In 

other words, the interim transition objectives 

contain content objectives (which at 

the start can look like the current policy 

objectives, but later will increasing appear 

to be different), process objectives (quality 

of the transition process, perspectives and 

behaviour of the actors concerned, unexpected 

developments) and learning objectives 

(what has been learned from the 

experiments carried out, have more options 

been kept open, re-adjusting options and 

learning objectives). 

Evaluating and learning 

Transition management involves the use of socalled 

‘development rounds’, where what has 

been achieved in terms of content, process 

dynamics and knowledge is evaluated. The 

actors who take part in the transition process 

evaluate in each interim round the set interim 

transition objectives, the transition process 

itself and the transition experiments. 

The set interim objectives are evaluated to 

see whether they have been achieved; if this 

is not the case, they are analysed to see why 

not. Have there been any unexpected social 

developments or external factors that were 

not taken into account Have the actors 

involved not complied with the agreements 

that were made 

The second aspect of the evaluation concerns 

the transition process itself. The set-up 

and implementation of the transition 

process is put under the microscope. How do 

the actors concerned experience the partici- 

48



> 

pation process Is it dominated by certain 

parties (vested interests) Is it too consensual 

(too cosy), or is there too little commitment 

Are there other actors who should be 

involved in the transition process Are there 

other forms of participation that must be 

tried out 

The final issue for evaluation is the amount 

of learning or ‘enrichment’ that has taken 

place in the previous period. A special point 

of attention is what has been learned from 

the experiments carried out to stimulate the 

transition. What have been the most important 

learning moments and experiences 

Have these led to new knowledge and new 

circumstances And what does this means 

for future policies 

Creating public support 

A continuing concern is the creation and 

maintenance of public support. This is 

important for the process to keep going and 

preventing a backlash, which may occur 

when quick results do not materialize and 

setbacks are encountered. One way to 

achieve this is through participatory decision-making 

and the societal choice of goals. 

But societal support can also be created in a 

bottom-up manner, by engaging in experiences 

with technologies in areas in which 

there is local support for their use. The experience 

may take away fears elsewhere and 

give proponents a weapon. With time solutions 

may be found for the problems that 

limit wider application. Education too can 

allay fears but real experience is probably a 

more effective strategy. Through the prudent 

use of new technologies in niches, societal 

opposition may be circumvented. 

4. TRANSITION MANAGEMENT IN 

RELATION TO CURRENT POLICY 

Transition management should be seen as 

an integrative framework for government 

policy but also for NGOs and business. The 

concept of transition places short-term policy 

within a time frame of one, two or three 

generations (50-100 years) rather than the 

maximum of 5-10 years, which is typical of 

current policy. It is also oriented towards system 

innovation. Perhaps unfortunately, the 

fruits of technical fixes will contribute more 

quickly to policy objectives in the short term. 

An example of this is CO 2 collection and 

storage. Another example is the catalytic 

converter which helped to achieve reductions 

in automobile NOx emissions but 

increased energy use and that did not deal 

with the many social and economic problems 

related to car use. Technical fixes are no 

solution for complex social problems. 

This does not mean that transition management 

rejects the improvement of existing systems 

as a route towards sustainability. It says 

that you must aim for both system optimisation 

and system innovation instead of one of 

the two. It also should be noted that the two 

strategies are not necessary mutually exclusive: 

cleaner cars can go hand-in-hand with innovative 

public transport systems. System improvements 

may thus act as a stepping-stone for system 

innovation. An example is organized car 

sharing, which facilitates intermodal travel. 

A characteristic of transition management 

when successful is that structural change is 

achieved gradually, without too much 

destructive friction in the form of social 

resistance or high costs. This is done through 

the use of ‘two-world’ technologies and 

exploitation of niches, attractive domains of 

application. You do not need centralised 

planning for the creation of a new system. It 

can also be achieved through in a gradual, 

unplanned way, through hybrid technologies 

and by adding new elements to an 

existing system that facilitate further 

change (Geels and Kemp, 2000). 

Transition management tries to utilize the 

opportunities for transformation that are 

present in an existing system. It joins in 

with ongoing dynamics instead of forcing 

changes. Transition management also implies 

refraining from large-scale investment in 

improvement options that only fit into the 

existing system and that, as a result, create a 

‘lock-in’ situation. 

The role of government in transition management 

is a pluralistic one: facilitator-medi- 

49


ator-controller-director, depending on the 

stage of the transition. The most effective 

(but least visible) is the guidance in the predevelopment 

phase, and to a lesser extent, in 

the take-off phase. Much more difficult is the 

guidance in the acceleration phase, because 

the direction of development in this phase is 

mainly determined by reactions that reinforce 

(or weaken) each other and cause 

autonomous dynamics. It is still possible at 

this stage to adjust the direction of development, 

but it is almost impossible to reverse it. 

This paper has described a method for managing 

the change process, called transition 

management. It is called transition management 

because the challenge of sustainability 

involves the management of transition problems: 

the costs of adaptation, resistance of 

vested interests, and uncertainty about the 

best option. Through transitions environmental 

benefits may be achieved, through 

systems that are inherently more environmental 

benign, but that also may produce 

wider sustainability benefits in the form of 

health benefits and social well-being. 

> 5. CONCLUSIONS 

Transition management 

This paper looked at the experiences with 

innovation policy and environmental policy 

in bringing forth innovations for the environment—innovations 

that have environmental 

benefit compared to existing products 

and processes. 

The paper offers practical suggestions both 

for innovation policy and for environmental 

policy to stimulate innovations for the environment, 

something that is needed because 

the current incentives for environmental 

innovations are weak. I have made a plea for 

programmes for system innovation offering 

sustainability benefits, to complement existing 

programmes for environmental technology 

15 . Programmes for system innovation are 

programmes for system changes offering 

environmental benefits alongside other 

types of benefits. The programmes should 

be time-limited and flexible. I am thinking 

here about programmes for novel protein 

foods in which proteins are produced in factories 

using biotechnology (to substitute for 

meat), and programmes for integrated 

mobility and industrial ecology. 

Four other suggestions are: 

1. to target innovation policy to areas in 

which innovation for the environment is 

needed. 

2. to do environmental assessment and 

broader system assessment 

3. to support technology experiments for 


4. to embed innovation policy in environmental 

policy. 

… is a collective, cooperative effort to work 

towards a transition in a flexible, stepwise 

manner, utilising dynamics and visions 

… involves a wide range of policies with 

their choice and timing gauged to the 

particular circumstances of a transition 

… involves system innovation 

and system improvement 

Although strictly speaking transitions cannot 

be managed, one can work towards 

them. This is what transition management 

attempts to do. Transition management consists 

of a deliberate attempt to bring about 

structural change in a stepwise manner. It 

tries to utilise existing dynamics and orient 

these dynamics to transition goals that are 

chosen by society. The goals and policies to 

further the goals are constantly assessed and 

periodically adjusted in development round. 

Through its focus on long term ambition 

and its attention to dynamics it aims to overcome 

the conflict between long-term ambition 

and short-term concerns. 

Transition management is based on a twopronged 

strategy. It is oriented towards 

stimulating system improvement and system 

innovation to meet the transition goals. The 

role of government in transition management 

is a plural one: facilitator-mediatorcontroller-director, 

depending on the stage 

of the transition. 

The value of transitions management is that 

it orients myopic actors to the future and to 

societal goals, that it helps to create societal 

50



> 

support for a transition (resulting in a transition 

programme which is politically legitimised) 

and commits societal actors to 

change. Transition management provides a 

basis for coordination of public and private 

action. It does not fix a path but explores 

various options. It adds a top-down element 

to bottom-up initiatives. 

In my view, transition management offers a 

promising alternative for a planning and 

control approach and the use of economic 

incentives that both suffer from serious 

problems: economic incentives are likely to 

be too weak and probably too general to 

promote system innovation whereas a planning 

and implementation approach is likely 

to be disruptive, by failing to include the 

multitude of microconcerns at the decentralized 

level. 

Transition management involves a change in 

policy making, which will be more oriented 

toward long-term goals of sustainability 

(instead of short-term goals), to system innovation 

and to new actors. Transition management 

is not something consensual. 

Transition management does not exclude the 

use of control policies, such as the use of 

standards and emission trading. We need 

corrective policies besides push policies. The 

policies can be chosen and legitimised as part 

of the transition endeavour or independently 

from it. For example the use of CO2 

taxes and other types of economic incentives 

can be legitimised by the economic principle 

that says that one should internalise external 

costs. Perhaps the commitment to a transition 

facilitates their introduction. We’ll see. 

Transition management is not a magical tool 

but does appear to be a promising perspective. 

Dutch authorities think so. It may be 

used to achieve a greater coherence in policy 

and in societal actions for sustainability. It is 

a different type of governance model, ained 

at modulating dynamics, not an instrument. 

REFERENCES 

Aichholzer, Georg, and Gerd Schienstock 

(eds.), (1994), Technology Policy: Towards an 

Integration of Social and Ecological 

Concerns, Berlin: Walter de Gruyter. 

Angerer, Hipp, Holland, Kuntze (1997) 

Umwelttechnologie am Standort Deutschland. 

Der Ökologische un ökonomische 

Nutzen der Projektforderung des BMBF, 

Evaluation of Fraunhofer ISI. Physica-Verlag, 

Heidelberg. 

Angerer, Gerhard, (2002) Orienting research 

policy to sustainability goals, paper for BLUE- 

PRINT workshop “Environmental Innovation 

Systems” on Jan 23-24, 2002, in Brussels, available 

from www.blueprint-network.net. 

Ashford, Nicholas A., Christine Ayers, and 

Robert F. Stone (1985), `Using Regulation to 

Change the Market for Innovation', Harvard 

Environmental Law Review, 9, 419-466. 

Ashford, Nicholas A.(1994) ‘An Innovation- 

Based Strategy for the Environment’, in Adam 

M. Finkel and Dominic Golding (eds.) Worst 

Things First: The Debate over Risk-Based 

National Environmental Policies, Resources 

for the Future, Washington D.C., 275-314. 

Ashford, Nicholas A., (2001), Government and 

Environmental Innovation in Europe and North 

American, paper presented at Conference 

“Towards Environmental Innovation Systems”, 

sept27-29 2001, Garmisch Partenkirchen, (forthcoming 

in Environmental Transformation, 

Governance and Globalization, ed.s. David A. 

Sonnenfeld and Arthur P.J. Mol, American 

Behavioral Scientist, 45.) 

Ashford, Nicholas, Wim Hafkamp, Frits 

Prakke and Philip Vergragt (2001) Pathways 

to Sustainable Industrial Transformation: 

Cooptimising Competitiveness, Employment 

and Environment, Ashford Associates, 

Cambridge, MA. 

Ayres, R.U., and U.E. Simonis (eds.) (1994), 

Industrial Metabolism, United Nations 

University Press, Tokyo. 

Banks, D.R. and G.R. Heaton, Jr. (1995), “An 

Innovation-Driven Environmental Policy”, 

Issues in Science and Technology, Fall. 

Berkhout, Frans (2002) Response to William’s 

paper, paper for BLUEPRINT workshop 

“Environmental Innovation Systems” on Jan 

23-24, 2002, in Brussels 

51


Blazejczak, J., D. Edler, J. Hemmelskamp and 

M. Jänicke (1999), “Environmental Policy 

and Innovation: An International Comparison 

of Policy Frameworks and Innovation 

Effects”, in P. Klemmer (ed.), Innovation 

Effects of Environmental Policy Instruments, 

Analytica, Berlin. 

Burtraw, Dallas (2000) Innovation under the 

Tradeable Sulphur Dioxide Emission Permits 

Programme in the US Electricity Sector, in 

OECD proceedings Innovation and the 

Environment, OECD, Paris, 63-84. 

Butter, Maurits (2002) A Three Layer Policy 

Approach for System Innovations, paper for 

BLUEPRINT workshop “Environmental Innovation 

Systems” on Jan 23-24, 2002, in Brussels. 

Christensen, Clayton (2000), The Innovator’s 

Dilemma. When New Technologies Cause 

Great Firms to Fail, Harvard, Harvard 

Business School Press. 

Cramer, J., J. Schot, F. van den Akker, and G. 

Maas Geesteranus (1990), “Stimulating 

Cleaner Technologies through Economic 

Instruments; Possibilities and Constraints”, 

UNEP Industry and Environment Review, 

13(2), 46-53. 

Clayton, Anthony, Graham Spinardi and 

Robin Williams, (1999), Policies for Cleaner 

Technology. A New Agenda for Government 

and Industry, Earthscan, London. 

Cramton, Peter (2000) ‘A Review of Markets 

for Clean Air: The US Acid Rain Programme 

by A.D. Ellerman et al.’, Journal of Economic 

Literature, 38, 627-633. 

De Jong, M. and K. van de Ven (1985), 

“Milieu-innovaties in kleine ondernemingen” 

(Environmental Innovations in Small 

Enterprises), publicatiereeks Milieubeheer 

21, VROM.` 

Downing, Paul B., and Lawrence J. White, 

(1986), ‘Innovation in Pollution Control’, 

Journal of Environmental Economics and 

Management 13: 18-29. 

Elzen, Boelie, Remco Hoogma and René 

Kemp (2001) Managing the Transition to 

Sustainable Transport through Strategic 

Niche Management, paper for Conference 

“Managing Transitions in the Transportation 

Sector: How Fast and How Far”, Sept 11-14, 

2001, Asilomar, California. 

Geels, F. (1999), “Sociotechnical Scenarios as 

a Tool for Reflexive Technology Policies”, 

paper presented at the 4S Conference, 28-31 

October, San Diego. 

Geels, F. and Kemp, R. (2000), “Transities 

vanuit socio-technisch perspectief” (Transitions 

from a socio-technical perspective), 

background document for chapter 1 of the 

report “Transities en transitiemanagement” 

(Transitions and transition management) 

(Rotmans et al, 2000), Maastricht, The 

Netherlands. 

Georg, S., I. Røpke and U. Jørgensen (1992), 

“Clean Technology-Innovation and Environmental 

Regulation”, Environmental and 

Resource Economics, 2(6), pp. 533-550. 

Green, Kenneth, Andrew McMeekin (1995), 

“Excellent at What Environmental Business 

and Technology Strategies”, Working paper 

9505, CROMTECH, Manchester. 

Greene, David L. (1990), “CAFE OR PRICE. An 

Analysis of Effects of Federal Fuel Economy 

Regulations and Gasoline Price on New Car 

MPG, 1978-89”, The Energy Journal, 11(3), 

37-57. 

Groenewegen, Peter, and Philip Vergragt, 

(1991), ‘Environmental Issues as Threats and 

Opportunities for Technological Innovation’, 

Technology Analysis and Strategic 

Management 3(1): 43-55. 

Hahn, R.W. (1989), “Economic Prescriptions 

for Environmental Problems: How the Patient 

Followed the Doctor”s Orders”, Journal of 

Economic Perspectives, 3(2), pp. 95-114. 

Hansen, O.E., J. Holm, and B. Søndergård 

(2000), Construction of a selective milieu of 

environmental developmnent in the Danish 

Textile Industry, paper presented at the 3rd 

international conference of the European 

52



Society for Ecological Economics, Vienna, 3-6 

May 2000. 

Hartje, V.J. (1985), “Environmental Product 

Regulation and Innovation: Limiting Phosphates 

in Detergents in Germany”, International 

Institute for Environment and Society (IIUG), 

IIUG dp 85-5, Wissenschaftszentrum Berlin. 

Heaton, G.R., Jr. (1990), “Regulation and 

Technological Change”, paper presented at 

the WRI/OECD Symposium Toward 2000: 

Environment, Technology and the New 

Century, Annapolis, Maryland, 13-15 June. 

Hemmelskamp, J. (1997), “Environmental 

Policy Instruments and their Effects on 

Innovation”, European Planning Studies, 

5(2), pp. 177-193 

Hoogma, Remco, René Kemp, Johan Schot 

and Bernhard Truffer (2002), Experimenting 

for Sustainable Transport Futures. The 

Approach of Strategic Niche Management, 

EF&N Spon, London. 

Irwin, A. and P.D. Hooper (1992), “Clean 

Technology, Successful Innovation and the 

‘Greening of Industy’”, Business Strategy 

and the Environment, 1(2), pp. 1-12. 

Jaffe, Adam B., and Karen Palmer (1996), 

Environmental Regulation and Innovation: 

A Panel Data Study, Washington DC: RFF. 

Kemp, R., Schot, J. and Hoogma, R. (1998), 

‘Regime Shifts to Sustainability through 

Processes of Niche Formation. The Approach 

of Strategic Niche Management’, Technology 

Analysis and Strategic Management, 

10(2): 175-195. 

Kemp, R., Truffer, B. and Harms, S. (2000), 

‘Strategic Niche Management for Sustainable 

Mobility’, in K. Rennings, O. Hohmeier, 

and R.L. Ottinger (eds.) Social Costs and 

Sustainable Mobility - Strategies and Experiences 

in Europe and the United States, 

Physica Verlag (Springer), Heidelberg, New 

York, 167-187. 

Kemp, René (2000), Technology and Environmental 

Policy – Innovation effects of past policies 

and suggestions for improvement, OECD 

proceedings Innovation and the Environment, 

OECD, Paris, 35-61. 

Kemp, R., Rip, A. and Schot, J. (2001), 

Constructing Transition Paths through the 

Management of Niches, in Raghu Garud and 

Peter Karnoe (eds.), Path Creation and 

Dependence, Lawrence Erlbaum: Mahwah 

NJ and London, 269-299. 

Kemp, René, and Ellen Moors (2001), 

Modulating Dynamics in Transport for 

Climate Protection, paper for conference 

“Institutions and instruments to control 

global environmental change” in Maastricht, 

21-22 June, 2001, organised by METRO. 

Kemp, René, and Jan Rotmans (2001), The 

Management of the Co-Evolution of 

Technical, Environmental and Social Systems, 

paper for international conference Towards 

Environmental Innovation Systems, 27-29 

Sept, 2001, Garmisch Partenkirchen, Germany. 

Kemp, René (2002) Environmental Protection 

through Technological Regime 

Shifts, forthcoming in Andy Jamison & 

Harald Rohracher (eds.), Technology Studies 

and Sustainable Development. Profil Verlag, 

Munich, forthcoming. 

Kemp, René (2002) Synthesis report of 1st 

BLUEPRINT workshop “Environmental Innovation 

Systems” on Jan 23-24, 2002, in 

Brussels, available from www.blueprint-network.net. 

Loiter, J.M., and V. Norberg-Bohm (1999) 

‘Technology policy and renewable energy: 

public roles in the development of new energy 

technologies’, Energy Policy, 27(2) pp. 85-97. 

Meyer-Krahmer, Frieder (1998) Innovation 

and Sustainable Development. Lessons for 

Innovation Policies, Physica Verlag, 

Heidelberg, New York. 

Milliman, Scott R., and Raymond Prince, 

(1989), ‘Firm Incentives to Promote 

Technological Change in Pollution Control’, 

Journal of Environ-mental Economics and 

Management 17: 247-265. 

53


Nationaal Milieubeleidsplan 4 (2001) Een 

wereld en een wil, werken aan duurzaamheid 

(A world and a willing),VROM, Den 

Haag, Juni 2001. 

Norberg-Bohm, Vicki (1999) ‘Stimulating 

'Green' Technological Innovation: An 

Analysis of Alternative Policy Mechanisms’, 

Policy Sciences 32: 13-38. 

OECD, (2000) Innovation and the Environment, 

Paris. 

Olsthoorn, X., F. Oosterhuis and H. 

Verbruggen (1992), “Inhoudelijke evaluatie 

van de stimuleringsregeling milieutechnologie” 

(Evaluation of the Subsidy Programme 

on Environmental Technology), IvM report 

R-92/01, IVM, Amsterdam. 

Palmer, K., W.E. Oates and P.R. Portney (1995), 

“Tightening Environmental Standards: The 

Benefit-Cost or the No-cost Paradigm”, 

Journal of Economic Perspectives 9 (4), pp. 

119-132. 

Peacock, Alan (ed.) (1984), The Regulation 

Game. How British and West German 

Companies Bargain with Government, 

Oxford: Basil Blackwell. 

Rennings, Klaus (1998) “Towards a Theory 

and Policy of Eco-Innovation—Neoclassical 

and (Co-) Evolutionary Perspectives”, ZEW 

Discussion Paper 98-24. 

Rip, A. and Kemp, R. (1998), ‘Technological 

Change’, in S. Rayner and E.L. Malone 

(1998), Human Choice and Climate Change. 

An International Assessment, Vol. 2, Batelle 

Press, Washington D.C., 327-399. 

Rotmans, J. and den Elzen, M. (1993), 

Halting Global Warming: Should Fossil Fuels 

by phased out In M. Lal (editor) (1993) 

Global Warming. Concern for tomorrow, 

Tata McGraw-Hill Publishing Company 

Limited, New Delhi. 

Rotmans, J, Kemp, R., van Asselt, M.B.A., 

Geels, F., Verbong, G. and Molendijk, K. 

(2000), ‘Transitions and Transition Management: 

the case of a low-emission energy supply’, 

ICIS-Report, Maastricht, October 2000. 

Rotmans, Jan, René Kemp, and Marjolein 

van Asselt (2001) ‘More Evolution than 

Revolution. Transition Management in 

Public Policy’, Foresight 3(1): 15-31. 

Sahal, D., 1985, ‘Technological guideposts 

and innovation avenues’, Research Policy. 

14: 61-82. 

Smith, Keith (2002) Environmental Innovation 

in a Systems Framework, paper for BLUE- 

PRINT workshop “Environmental Innovation 

Systems” on Jan 23-24, 2002, in Brussels. 

Stewart, R. (1981), “Regulation, Innovation 

and Administrative Law: A Conceptual 

Framework”, California Law Review, 69(5), 

pp. 1256-1377. 

Teisman, G. (2000), Sturen als ontwikkelingsopdracht 

(Guidance as a development 

assignment), in: Report of discussions with 

scientists during the NMP4 process, RMNO, 

The Hague, The Netherlands, pp. 64-67. 

VROM (2000), ‘Transities naar Duurzaamheid: 

Milieubeleid als Transitiemanagement’ 

(Transitions to sustainability: Environmental 

policy as transition management), Vision 

document for Knowledge and Technology 

workgroup, draft 10 April 2000. 

Weaver, Paul, Leo Jansen, Geert Grootveld 

and Philip Vergragt (2000). Sustainable 

Technology Development. Sheffield: Greenleaf 

Publishing. 

Williams, Robin and Nils Markusson (2002) 

Knowledge and Environmental Innovation, 

paper for BLUEPRINT workshop “Environmental 

Innovation Systems” on Jan 23-24, 

2002, in Brussels, available from www.blueprint-network.net. 

54

Part 2 

DOES INNOVATION POLICY 

MAKE A DIFFERENCE 

The three following papers deal with the 

question of additionality – what would have 

happened if no intervention had taken place 

– as a way to establish whether policy instruments 

have any effect. Identifying and measuring 

the socio-economic effects and impact 

of innovation policies is high on the agenda of 

many OECD countries, as a consequence of an 

increased requirement for accountability in 

public sector management. In order to establish 

the positive externalities of knowledge 

production and diffusion, and not only the 

benefits for those directly involved in a policy 

programme, we need to find ways to measure 

these wider effects. Policy makers are increasingly 

required to demonstrate effects of policy 

instruments in quantitative terms. 

The first paper by Luke Georghiou takes 

stock of the progressively broadening 

approach towards the key issue of additionality 

of innovation policy: from quantitative 

approaches and effects on R&D investment 

(flows), towards behavioural approaches 

and qualitative effects on capacities in the 

innovation system (stocks). The paper 

explores what the evidence of the impact 

of innovation policy has been so far. 

Governments can make a difference in the 

stimulation of innovation in public goods. 

Georghiou argues that measures to stimulate 

R&D in a systems context, need to be 

designed as public-private partnerships, 

rather than as support mechanisms. 

There is a school of scholars aiming to measure 

the longer-term impacts of innovation 

policy with the help of econometrics. 

Norway has a long tradition of evaluating a 

number of innovation policy instruments 

using quantitative analytical methods. The 

second paper by Lasse Bræin, Arild Hervik, 

Erik Nesset and Mette Rye discusses these 

methods and their problems and provides 

empirical evidence from a number of 

Norwegian cases. 

The third paper by Erik Arnold and Patries 

Boekholt discusses how evaluations can help 

us in deciding whether innovation policy 

works. It also discusses what we do not 

know due to methodological difficulties. 

The paper is based on a large number of 

evaluation studies in various countries. 

2 

55


IMPACT AND ADDITIONALITY 

OF INNOVATION POLICY 

LUKE GEORGHIOU 

Policy Research in Engineering, Science and Technology (PREST), 

University of Manchester, United Kingdom 

Luke Georghiou (1955) is Professor of 

Science and Technology Policy and Management 

and Director of PREST (Programme 

of Policy Research in Engineering, Science 

and Technology) University of Manchester. 

He is responsible for management of 

research and postgraduate institute of 25 

staff and 28 doctoral researchers. Luke 

Georghiou completed his PhD from the 

University of Manchester in 1982. The thesis 

is called “Technical Characteristics and Inter- 

Fuel Substitution” and was supervised by 

of Professor Michael Gibbons. He has published 

extensively in the area of technology 

policy and evaluation. Publications include 

numerous invited conference presentations 

and keynote addresses, academic journal 

articles, book chapters and reports. Other 

activities in the past include Member 

Quinquennial Review of the Research 

Councils (2001), Chair Advisory Committee 

on Evaluation of Science Foundation Ireland 

(2000), Chairman Strategic Review Panel 

of EUREKA Initiative and Chairman 5 Year 

Assessment Panel of EU Framework 

Biotechnology Programmes (1996). 

57

Impact and Additionality of Innovation Policy 

> 

> 


The design of innovation policy is both a 

practical art and one which draws upon current 

theories to provide a rationale, a justification 

and some specific principles of 

design. Just as success in innovation itself 

comes from the ability to sustain a series of 

improvements over time rather than from 

an achievement at a fixed point in time, so 

the design of innovation policies needs to be 

adaptive to circumstances and to build on 

lessons learned in practical situations. 

Informal feedback is always present but 

policy evaluation provides the most systematic 

approach to learning. In this paper, a key 

dimension of evaluation, the additionality 

of the intervention, is reviewed and linked 

to different perspectives on and rationales 

for innovation policy. This is set in the context 

of the evolving behaviour of innovative 

firms. Finally some brief conjectures are 

drawn about the implications choice of 

policy rationale would have for addressing 

the issue of sustainability in the context of 

innovation policy. 

2. ADDITIONALITY 

It is tempting for the policymaker to compile 

dossiers replete with “success stories” which 

may be used to justify continuation or 

expansion of the policy measure in question. 

Even better if some proxy for rate of return 

or economic activity generated can be integrated. 

However, it has long been realized 

that the critical question that an evaluation 

needs to ask must go beyond the level of 

effects achieved by the beneficiaries of a 

policy and pursue the issue of the contribution 

to those effects made by the existence 

of the public intervention. In an earlier 

paper, the author and colleagues encapsulated 

this question in terms of what difference 

is made by the policy, or to use the terminology 

of evaluation, the additionality of 

the support measure (Buisseret et al, 1995). 

Conceptually, additionality appears relatively 

simple on superficial examination. It 

involves comparison with the null hypothesis 

or counterfactual – what would have 

happened if no intervention had taken 

place. The framework we developed treated 

additionality in three manifestations: 

a) Input additionality: a concern with 

whether resources provided to a firm are 

additional, that is to say whether for every 

Euro provided in subsidy or other assistance, 

the firm spends at least an additional 

Euro on the target activity. Two 

other scenarios are worthy of note, the 

first being that the firm does not increase 

its expenditure on the activity in question 

by an amount equal to or greater than the 

subsidy, thus diverting the public funds to 

enhance its profits. This is an outcome easily 

possible when a fiscal incentive for R&D 

is introduced which allows the whole 

spend rather than incremental spend to be 

offset against the allowance. The second 

scenario is that the firm accepts the subsidy 

for an activity that it would have carried 

out anyway and spends the resources on 

another project, thus rendering the subsidy 

as “deadweight”. This is a typical scenario 

in grant-funding for R&D where it is 

very difficult for the policymaker to judge 

the firm’s initial intentions. The incremental 

spend in R&D may be achieved but the 

policymaker is funding a different project 

(normally the marginal one which the firm 

would have undertaken if additional 

resources were available). The advantage 

to the firm in a situation of competition for 

grants is that it can put forward a more 

attractive project which (obviously) promises 

higher returns than the marginal 

project. 

b) Output additionality: At its simplest this 

measure concerns the proportion of outputs 

which would not have been achieved 

without public support. It has appeal for 

evaluators as they are more interested in 

assessing the achievement of programme 

goals than in auditing the expenditure of 

participants. However, the evaluator is 

faced with two serious impediments in 

operationalising this concept. The first is 

that the outputs of an innovative project 

in receipt of support (say reports, patents, 

prototypes, business plans, new partnerships) 

are essentially intermediate 

58



achievements. They do not in themselves 

constitute innovation. However, if the 

concept is shifted in the direction of outcome 

additionality (improved business 

performance as a result of new or 

improved products, processes or services) 

it becomes increasingly difficult to attribute 

the effect to the intervention. 

Multiple competences need to be applied 

to realize an innovation and separating 

them is non-trivial. Furthermore, if outcome 

additionality is the issue in question 

then outcomes beyond the economic performance 

of the beneficiary become relevant. 

These include unintended effects 

(including negative ones) and all types of 

spillovers, Jaffe (1996). The timing of the 

assessment is also relevant – there are 

many different time-profiles over which 

outcomes are manifested. These include 

short-term effects which terminate 

abruptly in the face of changed market 

conditions or corporate strategy. At the 

other end of the scale is the “sleeper technology” 

which finds its applications some 

years later, perhaps because complementary 

technologies have been developed or 

simply because a new application has 

been perceived. 

c) Behavioural additionality: this is a category 

we introduced following the observation 

that a common effect of innovation 

policy was not to alter a stop-go decision 

by the firm in respect of the project but 

rather to modify in some way the way in 

which the project was carried out. The UK 

Department of Trade and Industry has 

articulated these changes in three sub-divisions 

– scale additionality when the activity 

is larger than it would otherwise have 

been as a result of government support 

(perhaps creating economies of scale); 

scope additionality, where the coverage of 

an activity is expanded to a wider range of 

applications or markets than would have 

been possible without government assistance 

(including the case of creating a collaboration 

in place of a single-company 

effort); and acceleration additionality 

when the activity is significantly brought 

forward in time, perhaps to meet a market 

window. 

As Bach and Matt (2002) have correctly 

observed, behavioural additionality has a 

further dimension in capturing permanent 

or persistent changes in firm behaviour as a 

result of the policy intervention. This can be 

at the strategic level (incentivising the firm 

to move into a new area of activity, or to 

alter its business processes) or at the level of 

acquired competences. These changes are 

potentially more significant in the long run 

than the short-term boost to resources 

afforded by a subsidy. The two types of 

intervention are not wholly separable – support 

for collaborative R&D often has a 

rationale of overcoming firms’ natural resistance 

to collaboration through information 

failures or lack of the necessary competences 

to manage a partnership. 

Bach and Matt (2002) have sought to 

extend this dimension of additionality by 

introducing a further category, that of cognitive 

capacity additionality. With this perspective 

they see the key question as being 

whether the policy action changes the cognitive 

capacity of the agent. They argue 

that if output additionality reflects a neoclassical 

perspective, then cognitive capacity 

additionality occupies the equivalent position 

in an evolutionary-structuralist perspective. 

Since the introduction of the behavioural 

concept some further evidence has accumulated 

on this topic. Davenport and Grimes in 

assessing the effects of company support in 

New Zealand found that the behavioural 

additionality concept provided the best 

explanation for their findings. They discuss 

how managers and policy administrators can 

exploit the occurrence of behavioural additionality 

to maximize the impact of a 

research policy, on the basis that modified 

behaviour is likely to strengthen a policy's 

latent ability to influence the creation of 

output additionality. In such circumstances, 

the study suggests that managers and policy-makers 

should be identifying those interventions 

that lead to sustained improvements 

in managerial practice, and in 

competitiveness, and should be managing 

their diffusion within firms and throughout 

industries. 

59


On the other hand Luukkonen has criticised 

the additionality concept on the grounds 

that it is insufficient to reveal the usefulness 

of public support. She cites empirical evidence 

to show that projects deemed as trivial 

by firms at the time of support may in the 

long run turn out to have been highly significant 

in their impacts, for example 

because they may build capacity in areas 

where firms have suffered from what 

Salmenkaita and Salo (2001) have subsequently 

labeled “anticipatory myopia”. In 

response to this criticism it may be argued 

that whether a project is additional or not is 

a separate question from that of the success 

of a project. Indeed high additionality may 

easily be associated with an increased risk of 

failure because the intervention has 

tempted a firm to move beyond its competences 

or to undertake a project which was 

more risky than usual. Both of these may be 

positive effects overall. There is also the case 

of high additionality where the policymaker 

incentivised the firm to move in the wrong 

direction because the policymaker misjudged 

the direction of technology or the 

market. Empirical evidence from Hervik 

(1997) in a study of successive policies in 

Norway found a clear trade-off between 

additionality and economic impact probably 

for the first reason given above. 

How can the different types or manifestations 

of additionality be reconciled with current 

thinking on rationales for innovation 

policy The market failure rationale needs 

little explanation here. Following Arrow 

(1962) the argument follows the general 

line of positive spillovers, non-appropriability 

and uncertainty creating a situation in 

which there is under-investment in research 

(and by implication in other knowledgebased 

innovative activities) in comparison 

with the socially desirable level. As argued 

previously (Metcalfe and Georghiou, 1998) 

the market failure perspective has been 

highly successful in providing a general 

rationale for policy intervention but it is 

inherently unable to provide specific guidance 

on policy prescriptions. 

Lipsey and Carlaw (1998) in a study aiming 

to show that neo-classical and structuralist 

evolutionary policies lead one to different 

conclusions in a technology policy evaluation, 

engage in a discussion of how additionality 

(or in Canadian terminology, incrementality) 

would be assessed under each 

perspective. They argue that a neo-classical 

approach would insist at least on what they 

term a “narrow test of incrementality” 

being that “some technology is developed 

or installed that would not have been produced 

in the absence of the policy or programme 

under consideration”. This corresponds 

to output additionality as discussed 

above. They argue that the neo-classical 

approach could also go further to demand a 

test of “ideal incrementality” in which the 

policy is demonstrated to be an optimal use 

of government expenditure. This invokes a 

series of tests attributed to the Canadian 

economist Dan Usher: 

• The project must be the least costly way to 

undertake the desired level of R&D investment; 

• Social benefits must exceed the subsidy 

(including transaction costs, deadweight 

and other leakages); and 

• Discounted benefits must exceed discounted 

costs of intervention. 

It is clear that the information requirements 

of these test far exceed what is likely to be 

available in any practical situation and may in 

themselves place undue transaction costs 

upon the subsidy. The crucial criticism which 

Lipsey and Carlaw make is that the structuralist/evolutionary 

perspective would apply only 

a “weak test of incrementality”, defined as 

“something the policy makers are trying to do 

has happened as a result of their expenditure 

of funds”. The difference from the neo-classical 

perspective is that, with no attempt at 

optimality, the desired effects are less clearly 

specified (to allow for inherent variability 

between firms) and the effects looked for 

include structural changes and enhancements 

of firms’ capabilities. For innovation policies 

such as R&D subsidies where the main aim is 

to provide resources to the firm it seems reasonable 

to expect both kinds of effect to be 

evident (the targeted product and the longer 

term enhancements). However, when we 

come to consider innovation policies which do 

not involve the provision of finance this distinction 

becomes crucial. 

60



It is noticeable that the operational level for 

additionality tests is the firm. Indeed the 

whole structure is founded upon conjectures 

about firms’ decision-making and 

intentions. This sits comfortably with policies 

anchored in market failure but if we 

shift to the systems failure rationale for 

innovation policy, the level of analysis and 

the range of actors in the frame both 

broaden. Consider for example Smith’s 

articulation of systems failure (Smith, 2000) 

with the four types of manifestation, failure 

in infrastructure provision, failure to 

achieve transitions to new technological 

regimes, failure from lock-in to existing 

technological paradigms, and institutional 

failure (regulation, standards and policy culture), 

it can be seen the first and fourth 

exist only at the system level while transition 

and lock-in failures are manifested at 

both firm and system levels. 

As we have seen, innovation policy also 

works both at the systemic level and at the 

level of the firm. The preceding discussion 

of additionality is clearly cast at the level of 

the firm. It can be seen that discussions of 

input additionality and of output additionality 

are dealing with the questions raised 

in a market failure context. Hence, the policy 

measure presumes to correct a failure of 

the firm to allocate sufficient resources and 

the additionality question is whether it now 

allocates those resources. For output or outcome 

additionality the question is whether 

the socially desirable innovation has been 

achieved. 

For behavioural and cognitive capacity 

evaluation, the issues at the level of the 

firm are closer to the systems failure perspective. 

A typical goal for a technology 

programme is to direct firms towards a 

transition (see for example the strenuous 

efforts most governments are now pursuing 

to stimulate an interest in nanotechnology). 

Larger public initiatives, involving 

coordination in the establishment of new 

standards (for example in mobile communications) 

can be seen as directed towards 

lock-in failures and again are aiming to 

modify the firms capabilities and behaviour 

in a particular direction. 

> 

3. BEHAVIOUR OF INNOVATIVE FIRMS 

3.1 The Project Fallacy 

The situation for R&D support is further 

complicated by what we term the “project 

fallacy” whereby the policymaker concludes 

a contract with a firm to perform a set of 

work packages which it considers to be the 

innovation project. The fallacy lies in the 

assumption that the contract and package 

of associated deliverables against which the 

firm will be monitored is the real innovation 

project. Empirical evidence has established 

that for firms, the real project often starts 

before the contracted work, continues after 

it, and integrates the contract work with a 

suite of other innovative activities which are 

privately funded (or even draw upon other 

sources of public support). In effect the firm 

reduces its overall innovation costs by scanning 

public programmes for funding opportunities 

and matching relevant parts of its 

activities to cause the release of funds. This 

situation becomes relevant when evaluations 

seek evidence on socio-economic 

effects. The project outputs which constitute 

the deliverables were never intended to 

lead to such effects in isolation. The real 

evaluation question is what did the publicly 

supported contract contribute to the wider 

effort The answer requires a much deeper 

understanding of the strategic positioning 

of the project within the firm than would 

normally be sought by an evaluation. 

While this concept was developed to 

describe a situation in large firms, the project 

fallacy concept can also be applied to 

small firms and academic research. A determined 

innovator in a small firm can often go 

from one source of funding to another to 

keep a development programme going, 

with successive supporters each believing 

that they are supporting the phase before 

commercialization. The academic analogue 

is the (normal) situation where a researcher 

or research team is pursuing a long term 

agenda and writes proposals to a variety of 

sources over time to maintain funding for 

the programme through a succession of 

shorter term grants. Key findings may only 

emerge after several funding rounds. 

61


3.2 Implications of Trends in Industrial R&D 

We have earlier stressed the close link 

between policy design, additionality and 

conceptions of how firms make decisions. 

Following this argument it is worth asking 

what have been the major trends in industrial 

R&D and what implications have these 

for innovation policy and its evaluation. 

Quantitatively, there has been a forward 

surge (35% growth in real terms in OECD 

countries between 1993 and 1998) and companies 

are doing their best to maintain R&D 

levels during the present economic downturn 

(Butler and Giles, 2001). The growth was 

in part fuelled by a large increase in venture 

capital funding in the USA (Sheehan, 2001) 

though this has declined dramatically from 

its peak in 2000. This increase in R&D could 

be ascribed to a range of explanations: 

• A wider range of technological opportunities; 

• Increases in R&D productivity through 

availability of new tools and methods; 

• Increased competitive pressure to innovate; 

greater ability to afford R&D in 

favourable economic conditions; and 

• Increased returns from stronger and 

broader intellectual property rights. 

All have their advocates but, in a discussion 

of R&D policy, few would ascribe a large role 

to direct government support which was 

generally declining as a proportion of BERD 

(from 12.6% in 1993 to 9.9% in 1998 for 

OECD countries and from 11.3% to 9% of 

BERD in the EU). However, qualitative 

changes in industrial R&D provide a closer 

link to policy. A similar list (Coombs and 

Georghiou, 2002) includes: 

• Increased acquisition of technology rich 

smaller companies emerging from the venture 

capital sector; 

• Growth in outsourcing of R&D to specialist 

firms and universities; 

• Globalisation making R&D facilities more 

“footloose”; and 

• Continuing high significance of technological 

alliances. 

All of these trends constitute a changed 

industrial ecology. The common factor is 

that all stress relationships between companies 

or between companies and other innovation 

actors; put more broadly a distributed 

innovation system (Coombs et al, 

2001). Even globalisation can be seen as 

partly driven by the desire to find more 

effective linkages. 

Strengthening such relationships and providing 

the framework conditions in which 

they can thrive is nowadays seen as a central 

thrust of innovation policy. A list of such 

policies includes fostering industry-academic 

links, commercialization of public 

sector research and promotion of venture 

capital, particularly in circumstances where 

it is perceived to be deficient, notably at the 

seed funding stage. 

What does the concept of additionality 

mean in such circumstances Policies are less 

concerned with market failures than with 

creating markets for knowledge that did 

not previously exist and with introducing 

innovation actors to new relationships. The 

aim of most such policies is not to redress a 

temporary problem, the failure to invest in 

a given project, but rather to introduce new 

capabilities which will, if successful lead to a 

self-sustaining cycle of investment in innovation. 

Par excellence this is the territory of 

behavioural additionality. 

3.3 Policies for SMEs and Traditional 

Industries and the Measurement 

of Additionality 

A similar line of argument can be applied to 

innovation policies directed at SMEs and at 

traditional industries with low levels of innovative 

activity. In addition to innovation 

financing and support for networking, the 

policies of the last decade have focused on 

the provision of advice, information and 

infrastructure. Government and its agents 

shift from the role of provider to that of broker 

(Metcalfe and Georghiou, 1998) with the 

aim of guiding firms to suitable partners or 

financiers, technology acquisitions or management 

capabilities. Broader infrastructural 

initiatives address educational deficiencies, 

barriers to mobility and improvement of reg- 

62



ulatory frameworks governing innovation. 

Such policies certainly consume public financial 

resources and hence can be and are evaluated. 

But on which criteria 

The desire for evidence-based policy and the 

current vogue for performance indicators 

makes it likely that at least the advisory and 

information schemes will have objectives. 

Many of these will be essentially activity 

related (eg the number of firms assisted). 

Policymakers still wish to know what effects 

their schemes are having upon the innovative 

performance of firms and ideally the 

relative effectiveness of different measures. 

This latter goal is complex to pursue for two 

reasons: 

• Comparison of widely varying schemes is 

highly dependent upon the choice and 

weighting of criteria; and 

• Comparison often presumes independence 

when in fact the range of schemes 

should be considered as a portfolio, covering 

a range of deficiencies in the system 

and often interacting with each other. 

Taking the simpler task of effects at the level 

of the firm, the additionality question 

becomes hard to pursue by direct investigation. 

Reconstruction of the consequences of 

an advisory or training input is much harder 

than the same process for a large grant. 

While the grant will usually involve dedicated 

staff and a detailed accountability 

trail, this is unlikely for the “softer” measures. 

Evaluations in this domain tend to rely 

on econometric comparisons of matching 

samples of companies (see for example 

Shapira et al, 1996 for a review of evaluations 

of manufacturing extension programmes). 

Such approaches have an implicit 

model of additionality built into them which 

substitutes the counterfactual by presuming 

that the matched firm will behave in the 

way that the original firm would have without 

the intervention. The model is that of 

outcome additionality. Leaving aside the 

problem of how satisfactory the matching 

process is, the concerns which arise are 

clearly those of attribution. Findings of this 

kind should always be combined with realtime 

case study evidence to allow some 

insight into the causality mechanisms. 

Taking the issue of additionality measurement 

more generally there are three main 

approaches, all of which have their problems: 

• The matched sample approach, as just discussed 

always raises questions of just how 

similar the firms are. From the early days 

of innovation studies Project Sappho illustrated 

the difficulties of achieving true 

matching. 

• The second approach is a variant on the 

matched sample, involving the use of 

failed applicants for assistance. This is 

generally criticized on the grounds that 

“failures” are by definition different. This 

problem is reduced if there is significant 

over-subscription and the failures can be 

perceived as close equivalents in terms of 

the selection criteria. Only random selection 

above a quality threshold would 

guarantee this approach. However, this 

approach raises another difficulty. Even 

Problems in calculating returns to R&D 

Timing 

Attribution 

RESEARCH 

EFFECTS 

INNOVATION 

63


> 

the process of application can be seen as a 

policy intervention and it is often the case 

that raising the issue with the firm, or 

causing it to expend thought and 

resources on an application, will lead it to 

act in the direction of the policy. The comparison 

is then not with an unaffected 

state of events. 

• The third approach comes closest to the 

core of additionality, by asking the firm 

directly about the counterfactual – would 

it have done the work without assistance 

and if so how would it have been different. 

This presumes that the respondent is 

honest (usually they are) but more importantly 

that they are capable of performing 

the counterfactual analysis themselves. 

Ideally the intentions of the firm should 

be explored before the decision to apply. 

This begins to approach the situation of 

the Usher tests where the transaction 

costs of evaluative information exceed the 

benefit. 

CONCLUSIONS 

Faced with a barrage of methodological 

and practical difficulties the policymaker 

and evaluator are faced with a choice. 

Pursuit of input and output/outcome additionality 

can be treated ex ante as a design 

criterion and ex post as an area where some 

evidence can be collected but where full 

measurement may be impossible and in any 

case is not justified in resource terms (or in 

terms of the disruptive effect on intended 

beneficiaries).On the other hand, if the 

behavioural perspective is adopted then a 

more pragmatic strategy is possible. Again, 

other forms of additionality can be measured 

within the bounds of practicality but 

the primary interest will be in the qualitative 

changes which have been stimulated in 

the innovation system. Empirical work then 

focuses less upon the events during and 

immediately following a project (a measurement 

of flow) and instead switches to a 

desire to understand the present capacities 

of the innovation system and the players in 

it (a measurement of stock). 

This perspective also has implications for a 

conference which is interested in exploring 

how innovation could be influenced in the 

direction of sustainability. A project-based 

approach would suggest that efforts should 

be targeted on clean technologies. This 

would provide incentives to specialists in 

the field and aim to draw in others to produce 

cleaner technologies than they would 

otherwise contemplate. A regulatory 

approach would set standards with which 

innovations would have to comply. Practical 

experience has shown that there are significant 

failings possible with this approach. 

Evaluation findings have often indicated 

that firms (especially small ones) fail at the 

commercial stage because they have 

wrongly anticipated the level of regulation 

(in either direction) (see for example 

Ormala et al, 1993). 

A systems-perspective would start from a 

different position by questioning what 

capabilities and incentives a firm would 

need to engage in sustainable innovation. 

This would include better information and 

interaction with regulation, better understanding 

of the social context in which the 

innovation would perform and a command 

of the necessary technological skills, which 

could include a Smith-style transformation 

to a new technological regime based on different 

materials or processes. A shift of this 

kind requires coordination. Measures to 

stimulate R&D and innovation in this context 

need to be designed as public-private 

partnerships rather than as support mechanisms. 

The partnership element is one in 

which both sides contribute and both benefit. 

Government provides resources and 

information and industry produces innovations 

in an area which it might not otherwise 

have done and feeds back into the 

information and regulatory process. Most 

developed societies face a much greater 

challenge in stimulating innovation in public 

goods and services than they have in the 

present wave of consumer and business 

innovations. To address these it is necessary 

to employ flexible and pragmatic rationales 

with the underpinning aim of achieving a 

persistent shift in capabilities and hence in 

behaviour. 

64



> 

REFERENCES 

Arrow K (1962) Economic welfare and the 

allocation of resources for invention in 

Nelson R (ed), The Rate and Direction of 

Inventive Activity Princeton: Princeton 

University Press 

Bach L and Matt M (2002) “Rationale for 

Science and Technology Policy” in 

Georghiou L, Rigby J and Cameron H (eds) 

Assessing the Socio-Economic Impacts of the 

Framework Programme, Report to DG 

Research, http://les.man.ac.uk/PREST/Download/ASIF_report.pdf 

Buisseret TJ, Cameron H and Georghiou L 

(1995) “What difference does it make 

Additionality in the public support of R&D in 

large firms”, International Journal of 

Technology Management, Vol.10, Nos 4/5/6 

pp587-600 

Butler D and Giles J, (2001) Nature 413, 448 

Coombs R and Georghiou L (2002), A New 

“Industrial Ecology”, Science Vol 296, p471, 

19th April 2002 

Coombs R, Harvey M and Tether B (2001) 

Analysing Distributed Innovation Processes, 

CRIC Discussion Paper No 43 

http://les.man.ac.uk/CRIC 

Davenport S, Grimes C, Davies J (1998) 

Research collaboration and behavioural 

additionality: A New Zealand case study 

Technol Anal Strateg 10 (1): 55-67 Mar 1998 

Hervik A (1997) Evaluation of user-oriented 

research in Norway: the estimation of longrun 

economic impacts in Papaconstantinou 

G and Polt W (eds) Policy Evaluation in 

Innovation and Technology- Towards Best 

Practices, OECD 

Jaffe AB (1996) Economic Analysis of 

Research Spillovers: Implications for the 

Advanced Technology Program, NIST, US 

Dept. of Commerce Technology Administration 

NIST GCR 97-708 

Lipsey RG and Carlaw K, (1998) Technology 

Policies in Neo-Classical and Structuralist- 

Evolutionary Models, STI Review No. 22 

Special Issue on “New Rationale and 

Approaches in Technology and Innovation 

Policy”, OECD Paris 

Luukkonen T (2000) Additionality of EU 

Framework Programmes Research Policy 29 

711-724 

Metcalfe JS and Georghiou L (1998) 

Equilibrium and Evolutionary Foundations 

of Technology Policy, STI Review No. 22 

Special Issue on “New Rationale and 

Approaches in Technology and Innovation 

Policy”, OECD Paris 

Ormala E et al (1993) Evaluation of EUREKA 

Industrial and Economic Effects, Brussels: 

EUREKA Secretariat 

Salmenkaita J-P and Salo A (2001) Rationales 

for Government Intervention in the 

Commercialisation of new Technologies, 

Helsinki University of Technology, Systems 

Analysis Laboratory Research Reports, 

September 2001 

Shapira, P., J. Youtie, and J.D. Roessner, (1996) 

“Current Practices in the Evaluation of US 

Industrial Modernization Programs.” Research 

Policy, vol. 25, no. 2 (March 1996): 185-214. 

Sheehan J (2001) Changing Business 

Strategies for R&D and their Implications for 

Science and Technology Policy: OECD 

Background and Issues paper, OECD document 

DTSP/STP(2001)29, OECD Paris 

Smith K (2000) Innovation as a Systemic 

Phenomenon: Rethinking the Role of Policy 

Enterprise and Innovation Management 

Studies, Vol.1, No.1, 73-102 

65


THE NORWEGIAN 

SYSTEMIC APPROACH 

TO IMPACT ESTIMATION 

OF R&D SUBSIDIES: 

FOCUS ON ADDITIONALITY 

AND THE CONTRA-FACTUAL PROBLEM 

LASSE BRÆIN 

ARILD HERVIK 

ERIK NESSET 

METTE RYE 

Møre Research Molde 

Mette Rye worked as a consultant at the 

international department of the Central 

Bank of Norway (1995-1997). Research 

Fellow at Møre Research since 1997 working 

mainly with innovation and evaluation 

research. 

Erik Nesset worked as a researcher at the 

department of Social Services in 1987. 

From 1987-92 he was consultant at the 

research department of the Central Bank of 

Norway. Research Fellow at Møre Research 

since 1992 working mainly with transport 

economics, business development, innovation 

and evaluation research. He is also working 

at the Aalesund College since 1992. 

Arild Hervik, is professor at Molde College 

and Research Leader of Møre Research 

Molde. 

Lasse Ove Bræin worked as a researcher at 

Møre Research since 1991, working mainly 

with innovation and evaluation research. 

Managing director of Møre Reserach Molde 

since 1997. 

67

The Norwegian Systemic Approach to Impact Estimation 

of R&D subsidies focus on additionality and the contra-factual problem 


> 1. INTRODUCTION 

During the last two decades Møre 

Research has been involved in the evaluation 

of a number of policy instruments 

held by what today are the main 

institutions responsible for carrying out 

Norwegian R&D policy: the Research 

Council of Norway (RCN) and the State 

Industrial and Regional Fund (SND). 

This paper will focus on our work 

related to the evaluation system of the 

Research Council of Norway. Møre 

Research has been involved in developing 

the evaluation system which can be 

described in four steps: 

1. Selection procedures: PROVIS is a system 

of indicators for selecting research projects 

for funding where additionality and 

programme relevance are among the 

indicators that are rated. This rating is 

performed by the RCN agents. This step 

was established in 1999. 

2. Ex Ante evaluation of projects. Projects 

are interviewed upon 1 year after they 

have received the funding to register 

expectations and additionality. This step 

was carried out first time in 1995. 

3. Ex Post evaluation: Projects are interviewed 

upon 3 years after they have 

received the funding. This evaluation 

focus on reporting preliminary economic 

results and further expectations on economic 

impacts. 

4. Long run economic impacts: Projects are 

interviewed upon three years after leaving 

the RCN system and a stronger focus is 

put on evaluating along indicators of economic 

impacts, spin-offs and competence 

change. This step is being carried out for 

the first time in 2002. 

Since R&D projects usually carry a long term 

perspective, the system try to evaluate the 

effects as they evolve, responding with a 

long term system for evaluation. 

This paper is based on our data collected on 

commission from RCN involving the main 

instruments of RCN’s department of Industry 

and Energy. One of the main instruments is 

the User Directed Research Scheme (UDR1). 

This instrument was launched in 1990 as a 

strategic policy tool for increased innovation. 

RCN had two main goals; increase the 

share of applied, marked oriented research, 

and strengthen the competitiveness in the 

Norwegian trade and industry by developing 

the network for more efficient R&D 

services (Hervik, 1997). Companies may 

apply funding on projects either alone or in 

co-operation with other firms or a research 

institute. To create good incentives the support 

is granted directly to the company 

who pays the R&D-institute. The funding is 

sorted into sectors based on industrial 

branches which each are holding different 

programmes. The government has a strategic 

say in the matter on what industry 

branch the research are focusing on 

through the funding available for each sector. 

UDR can be characterised as a selective 

public measure, as opposed to a general 

measure working through the tax system. 

The R&D activity in Norway, and in 

Norwegian industry in particular, is low 

compared to the average of the OECD countries. 

The Norwegian innovation system is 

characterised by a relatively large R&D institute 

sector compared to other countries, 

accounting for more than 25% of the total 

R&D expenditure of the country (RCN, 

2001). The Research Council of Norway 

(RCN) has been given the strategic responsibility 

for the R&D institutes towards the 

Ministry. UDR implied a change in policy 

that reduced the basic funding to the institutes, 

and increased the dependence on 

customer based funding in the Norwegian 

R&D institute sector. 

The present study is also based on customer 

inquiries of 19 technical institutes that were 

carried out during the period of 1996-2000 

on commission from the Research Council of 

Norway (RCN) as a part of RCN’s overall evaluations 

of the institutes. We will also use the 

additionality reports of RCN projects 

observed through the yearly study performed 

by Møre Research on commission 

from RCN. 

69



> 

2. ADDITIONALITY OF PUBLIC FUNDING 

The most obvious guideline for governmental 

R&D Policy is that one should not subsidise 

activities the firms would undertake 

themselves without the support. Additionality 

is a conceptual framework meant to 

capture the intended “catalyst” effects of 

public support. It is a critical concept to public 

bodies that support/carry out R&D policy 

when rationalising the support, since any 

benefit arising from a R&D project can only 

be credited the public policy if the activity 

would not have taken place otherwise. 

“More generally the problem of additionality 

remains a core issue for evaluation and 

one which reaches the most sensitive areas 

of policy, going to the core of whether there 

is a rationale for intervention.” 

(Georghiou, 1998) 

Mistakenly thought of as the only additionality 

criteria, the question whether the public 

funding has increased the total amount of 

R&D investment is just one aspect of the concept. 

In addition, the public support has a 

wider set of intentions, for instance, to contribute 

to the building of R&D networks, 

improving the R&D competence and managerial 

skills in industry, building the national 

knowledge base to improve the innovation 

capability, etc. The additionality of these elements 

is more challenging to measure since 

they are more intangible and unquantifiable. 

To make an operational concept sorting the 

different effects, Buisseret et al. (1995) operates 

with three different forms of additionality; 

input, output and behavioural additionality. 

The catalyst effect on the 

investment decision and R&D efforts in terms 

of man-hours and capital investments represent 

the input additionality. This is what traditionally 

has been thought of as additionality, 

and the measure of success has been 

whether one unit of currency in R&D support 

has lead to at least one additional currency 

unit of total company R&D investment. If 

not, the public support is crowding out or 

replacing private funding. 

R&D theory has been focusing on market 

failures related to the incentives to invest in 

R&D. Another reason for the traditional 

focus on input additionality is theory 

related to efficiency in public policy means. 

One of the main reasons for choosing selective 

support of projects as opposed to a 

general support system, is that it is less 

costly and allows for a larger individual support. 

Providing funding to projects that 

would have been undertaken anyway 

makes this policy tool less efficient and targeted. 

In this perspective input additionality, 

and especially project input additionality 

(whether the project would have been 

realised without the public support) are 

important indicators of policy efficiency and 

success. However, demanding full project 

input additionality may leave out significant 

other additionality aspects, especially 

related to behavioural additionality . 

Behavioural additionality is a more complex 

concept describing the “...change in a company’s 

way of understanding R&D which can 

be attributed to policy action” (Buisseret 

et.al.1995). This includes the impact on the 

strategic behaviour of the firm in terms of for 

instance engaging in collaborative research, 

or the improvement of the R&D management 

skills or R&D absorptive capacity of the firm. 

The change of company behaviour is a difficult 

task to achieve and to measure. As with 

output additionality it usually takes time to 

develop and observe any effects from policy 

measures. The time lag involved further complicates 

the measuring of impacts. 

Often the most interesting aspect from a 

policy view, but also quite difficult to measure, 

is the output additionality, describing 

the additional commercial effects rising 

from new products, processes or services due 

to the public support. As pointed out by 

Georghiou (1998) the return on public 

investment approach which presumes a linear/sequential 

model of innovation, do not 

reflect the realities in how R&D projects 

evolve and leads to measuring problems. 

There is a growing understanding that innovations 

rarely can be subscribed to single 

projects within the company, let alone the 

selective public support of single projects. 

How the public support affects the R&D 

70




strategy and links with other firms is therefore 

interesting and incorporated in the 

behavioural additionality. 

The focus has traditionally been on the input 

and output additionality, which coincides 

with the development in innovation theory 

from a“technology push” model, where it 

was believed that developments in science 

was driving innovation, to a “demand pull” 

model where more focus was put on the 

demand in the market for new innovations. 

The more complex and interactive view of the 

innovation process and the firm as a learning 

organisation have given increased attention 

to the behavioural additionality. To get a 

deeper understanding of behavioural additionality, 

we need to further develop the theoretical 

basis for public intervention, or 

sources of market failure, related to firms 

behaviour in the R&D market. 

To obtain the most reliable data of change in 

R&D strategy or orientation a comparative 

study through interview studies ex ante and 

ex post is preferable. However, these data are 

rarely readily available. In comparative analysis 

the self-selection problem should be 

adjusted for. For instance, the difference in 

behavioural additionality between groups of 

firms may be due to inherent differences in 

for instance pre-programme attitute towards 

R&D co-operation and the importance of 

R&D for long term survival. We end up measuring 

the inherit differences in stead of measuring 

the change in behaviour. This is illustrated 

in Nesset (2001) where questions 

regarding pre-programme R&D attitude are 

used as control variables for self-selection. 

Questions regarding how the project has 

changed the company focus from a short 

term to a long term R&D orientation, enabled 

larger and more resilient R&D projects and 

whether the project has led to higher internal 

R&D spending are indicators of behavioural 

additionality. He find that increased R&D orientation 

has a significant positive effect on 

the triggering of new projects in terms of 

increased internal R&D spending. However, 

he cannot find that changes in R&D support 

level has a positive impact on either R&D orientation 

or internal R&D spending. He find 

that changes in R&D orientation depend 

strongly on pre-program R&D attitudes, indicating 

that the change in R&D orientation is 

due to inherit differences. These questions 

are further discussed in chapter 4. 

To increase the efficiency of policy instruments 

there has been an increasing demand 

for continous feed-back and reports of 

ongoing programmes. Both behavioural and 

output additionality is difficult, if possible, 

to observe in an ongoing programme 

because of the timelags involved in developing 

these impacts. The reports of behavioural 

and output additionality at this stage 

may easily be over-optimistic, and should be 

followed up by ex post studies. 

However, evaluations of on-going R&D programmes 

also limit the data and methods 

available in analysing the input additionality. 

Therefore, for mid-way evaluation purposes, 

the most common method to measure 

input additionality is through verbal 

reports/questionnaires. The question regarding 

project input additionality is typically 

posed like this; 

Q1: What would have come of the project 

without the public support 

• Project realised without modifications 

• Project realised, but over a longer period 

of time. 

• Project downsized 

• Project Canceled 

This way of measuring additionality is quite 

common, but criticized for the possibility of 

strategic answering and that hypothetical 

questions are difficult to answer. The hypothetical 

question reflects the nature of the 

problem as counter factual. Looking at verbal 

reports of additionality over the last two 

decades in Norway, Rye (2002) could not 

find that strategic answering is significantly 

reducing the validity of the data.In verbal 

reports of additionality timing of the interview 

may affect the data. Asking Q1 5 or 10 

years after receiving the support, reduce the 

reliability of the answers simply because 

more time has passed since the project decision 

was taken makes it more difficult to 

recall the situation and answer the hypothetical 

about what would have come of the 

71



Figure 2.1 > 

Additionality RCN supported projects interviewed upon 1 year after receiving support (ex ante interviews). 

100 

90 

80 

70 

66 

Low/no additionality (Done without subsidy) 

Medium additionality (Delayed or diminished) 

High additionality 

Don’t know 

66 

60 

55 

55 

56 

58 

50 


41 

39 

42 

39 

30 

31 

28 

48 49 0 

20 

10 

0 

0 

3 4 

2 

3 4 

1 

2 

2 

3 

2 

0 

1 

1995 1996 1997 1998 1999 2000 2001 

Figure 2.2 > 

Additionality of RCN supported projects interviewed upon 1 year after receiving support (ex ante interviews) 

sorted by RCN programmes. 

100 

90 

80 

Low/no additionality (Done without subsidy) 

Medium additionality (Delayed or diminished) 

High additionality 

Don’t know 

70 

60 

50 


30 

63 

33 

48 

41 

50 

46 

52 

48 

54 

44 

68 

27 

39 

61 

62 

34 

20 

10 

0 

3 

0 

Building and 

Construction 

(51 pro.) 

7 

Maritime 

(29 pro.) 

4 4 

New tech 

(28 pro.) 

0 0 0 

Offshore 

(21 pro.) 

1 

1 

Progit 

(75 pro.) 

4 

1 

prosmat 

(85 pro.) 

0 0 

Service-info 

(23 pro.) 

2 2 

Varp 

(44 pro.) 

project without public funding. The project 

leaders may also have left the company, being 

replace by others that don’t know the history. 

Figure 2.1 present the verbal reports of projects 

interviewed upon 1 year after receiving 

supports, i.e. the 1995-additionality reports 

are of projects that received funding in 1994, 

while the 1996- additionality reports are of 

projects that received funding in 1995, etc. 

The figure show that the reports of high additionality 

tended to increase in the period of 

1995-2000, while the report of 2001 show a 

“high additionality” share which is in line 

with the reports of 1998. This may illustrate 

the focus on additionality in the selection procedure 

or it may reflect the true variation of 

additionality in the different portfolios. As 

will be discussed in chapter 4, there is a need 

for implementing the effect of RCN selection 

procedures/ criteria to increase our understanding 

of additionality. Figure 4.2 show the 

additionality reports of RCN supported projects 

interviewed upon 1. year after receiving 

support (ex ante interviews) sorted by RCN 

programmes. The figure illustrate the difference 

between different programmes. 

72




> 

3.USE OF CONTROL GROUPS 

IN COMPARATIVE ANALYSIS 

WHAT TO COMPARE AGAINST 

When there are no absolute yardstick, performance 

can be observed though a relative 

measure, or a comparative analysis. This is 

the main method either the objective is to 

measure the performance of the public support 

or the performance of for instance an 

individual research institute. As will be discussed, 

finding a suitable reference group or 

counter factual situation to compare against 

is difficult when trying to measure the 

impact of public support. As pointed out by 

Nesset (2001) there are three possible 

counter factual cases; 

1) alternative R&D support, 

2) alternative use of public funds 

3) status quo. 

The first is creating a comparative analysis 

evaluating the effectiveness of the policy tool 

used, compared to other policy tools with the 

same set of receivers. This is clearly relevant, 

but not the kind of experiments that are 

granted social scientist. The second is an 

analysis of the alternative cost of the support, 

in terms of alternative use of the public 

money. An approach to this analysis is the use 

of an estimated alternative cost rate of public 

funds, that in Norway is estimated to 7%, 

however this approach easily leaves out the 

more intangible benefits of R&D like the 

building of a knowledge base and R&D network. 

In practise, the third option is used, 

constructing a comparison with status quo, 

i.e. the situation without the support. The main 

methodological approaches used to estimate 

the counter factual of the status quo is: 

1) User surveys, obtaining experimental data 

of the receivers perceived benefits of the 

supported R&D. 

2) Econometric analysis based on non-experimental 

data. 

3) Combinations of the two. 

The usefulness of the methods depends on 

the purpose of the study. However, neither 

method is unproblematic when it comes to 

measuring, and bias problems. 

In a comparative analysis of differences 

between receivers and non-receivers of 

public support it is preferable that the reference 

group has as similar characteristics 

as the “treated” group as possible to keep 

internal validity experiencing similar extraneous 

effects, since it is a proxy for what 

would happen without support (i.e. the 

counter factual situation). However, when 

the technology and market characteristics 

between the two groups are very similar, 

the benefits from the public supported 

R&D investments might spill over to the 

non-receivers, making it difficult to 

observe the difference between the two 

groups. This is creating a measuring problem, 

underestimating the effect of the 

public funding. Klette, Møen and Griliches 

(2000) call this a “catch 22” situation. The 

more successful the innovation, the more 

likely the spillover to competitors and 

other firms that can use the new technology 

and the more difficult it is to observe 

additional benefits compared to the 

“counter factual situation”. 

There is also a problem of using nonreceivers 

as a reference group, due to selfselection. 

Usually, the firms or projects did 

not end up in the two groups by chance. 

The public funded projects were selected 

for support, and those who did not receive 

support, might not even have applied. 

Therefore the difference we observe in the 

two groups might result from inherit differences, 

which does not represent the difference 

from a situation without the support. 

This might create a selection bias problem. 

Experimental data obtained through survey 

based studies, may supplement non-experimental 

data with a richer data set that 

allows for alternative way of constructing 

reference groups and analysing additionality. 

To be able to obtain the richness in the 

data needed to describe complex concepts, 

like for instance behavioural additionality, 

non-experimental data has to be supplemented 

with experimental data through 

user surveys. Ordinal data allows the use of 

quantitative methods, like for instance LIS- 

REL, which are designed to handle and 

describe latent variables. An example of this 

method is presented in chapter 4. 

73



2 

Since this is ordinal 

data we used a nonparametric 

test 

comparing the average 

score of research in 

the two groups 

Table A > 

Project content 

Average scores 

Contents of project - on a scale from 1-7 

Simple testing/measurements 2,8 

More advanced testing/measurements 3,5 

Other simple technical consulting 2,4 

Other 

more advanced technical consulting 3,4 

Studies/advising/problem-solving 4 

More advanced advising/analysis 3,7 

Product development 2,9 

Applied research 3,8 

Elements of basic research 2,1 

Survey data usually includes qualitative data 

that can give interesting observations about 

the effect of public funding on the project 

content and performance, observations that 

are not affected with the “spillover” problem 

since we are capturing a wider set of 

indicators than the pure economic effects, 

and are observing on a project level. 

However, usually survey data include only 

“treated” respondent, i.e. projects that have 

received support, which does not allow for a 

comparative analysis on a project level with 

similar projects that have not received support. 

Our data from customer inquiry of the 

Norwegian technological R&D institutes 

include both supported and non-supported 

projects, since the projects interviews upon 

was selected from the R&D institutes total 

project lists. In fact, the majority (75%) are 

non-supported. The total database which 

includes responses of both private firms and 

public institutions and are representative for 

the customers of R&D institutes in Norway. 

Since the R&D institutes are quite specialised, 

most of the firms operate within 

the same industrial branch. They represent 

the buyers of R&D services, and fulfils the 

criteria of a similar reference group which 

gives us a unique opportunity to make a 

comparative analysis between supported 

and non-supported R&D projects. This can 

be illustrated by one of the most interesting 

indicators reported in the customer inquiries 

which is the content of research reported in 

the projects by the customers and the project 

leaders. This study is based on customer 

inquiries of 19 different Norwegian technological 

research institutes involving a total 

of 818 telephone interviews carried out by 

Møre Research. For the last 10 institutes, we 

extended the study with a smaller questionnaire 

handed out to the project leader or 

researcher at the institute that was working 

on the projects selected for customer interview. 

They were posed some parallel questions 

that gave us a second opinion on the 

work done on the project in addition to an 

indicator on the researchers’ awareness of 

the customer’s views. We received a total of 

391 answers to this postal inquiry. 

To improve the customer’s understanding of 

the research content question, and help 

them to recall the project we started out 

asking the customer to describe the content 

of the project indicated by different elements 

as listed in table A. The table shows 

the average customer score for all the institutes. 

It can be seen that Studies/ advising/ 

problem-solving, More advanced advising/ 

analysis and Applied research are the elements 

given the highest average score, and 

Elements of basic research the lowest. 

In the next question, the customer was 

asked to rate the content of research in the 

project on a scale from 1 to 7, where 1 is 

consulting service, 4 is applied research and 

7 is leading edge research.What we found 

most interesting is the fact that projects 

receiving RCN support through UDR show a 

significant higher content of research, than 

the projects not receiving this public support 

2 (Figure 3.1). This is supported by the 

reports of the project leaders regarding the 

same projects (Figure 3.2) 

This confirms that projects with a higher content 

of research are selected for public funding 

through the UDR programme. However, 

this may also indicate that the marginal benefit 

of public support of projects within the 

R&D institutes is a heightening of the level of 

research within the research institutes. Without 

the UDR support, the overall research content 

in the institute portfolios of projects will 

decrease. We found that the willingness to 

pay for long term projects with a high content 

of research was low among the institute customers. 

Without the public support there is 

74




Figure 3.1 > 

Customers report: content of research in public supported vs. privately financed projects. 

45 


Public supported through RCN (Average score: 4,5) 

Fully funded by customers (Average score: 3,3) 

35 

30 

25 

20 

15 

10 

5 

0 

1 

Consulting 

service 

2 3 4 

Applied 

Research 

5 6 7 

Leading-edge 

Research 

Figure 3.2 > 

Customers report: content of research in public supported vs. privately financed projects. 

45 


Employees answers - public supported projects 

Employees answers - Non supported projects 

35 

30 

25 

20 

15 

10 

5 

0 

1 

Consulting 

service 

2 3 4 

Applied 

Research 

5 6 7 

Leading-edge 

Research 

reason to believe that share of consulting 

service in the project portfolio will increase. 

The observed additionality within the public 

supported projects confirms this hypothesis. 

38% of the projects were fully additional 

(would not have realized the project without 

support). In 48% of the projects, the support 

affected the size or progress of the project. 

Only 12% reported that they would have carried 

out the project without changes if they 

did not receive support. This indicate that the 

public support through UDR is giving an 

incentive to project realisation and to increase 

the R&D content of the projects. As illustrated 

in figure 3.3 the marginal effect on the institute’s 

R&D profile may be an important input 

to the R&D infrastructure over time. 

If firms are constrained by their R&D capability 

through lack of R&D absorptive capacity, 

R&D networks may be a cost-effective way of 

importing knowledge and increasing the 

skills of the staff. This is one of the intentions 

behind the UDR programme. Table B illustrate 

indicators of behavioural additionality 

regarding R&D absorptive capacity based on 

75



Table B > 

Behavioural additionality; indicator of change in R&D absorptive capacity 

Average scores: 

What is your firm’s overall assessment of the consequences Customer: Customer: 

of the co-operation with the institute regarding: Public supported Non-supported 

projects 

projects 

New knowledge, technology development 5,18 4,65 * 

Developing the skills of /motivating the staff 5,06 4,56 * 

The * indicate significant difference in average scores between supported and non-supported projects. 

Figure 3.3 > 

the R&D institute data. The public supported 

firms report a significantly higher average 

rating of knowledge import through the cooperation 

with the institute using a nonparametric 

test. However, this can also 

reflect the self-selection problem. The companies 

selected for support may carry a 

higher R&D orientation and absorptive 

capacity which enables them to draw larger 

benefits from the co-operation with the R&D 

institute supporting the hypothesis that to 

be able fully to make use of the research, you 

have to be engaged in research yourself 

possible scenarios 

R&D content 

(Cohen and Levinthal, 1989). This illustrates 

the measuring problems related to behavioural 

additionality, and furter work regarding 

the identification of indicators and 

how they are interlinked needs to be done. 

To identify the most important indicators, 

further work related to market failure 

regarding firm behaviour in R&D would be 

useful. 

As can be seen from figure 3.4 when evaluating 

the quality of the institute’s work, value 

for money and whether the project has had 

an effect on competitive position are among 

the lowest rated on a scale from 1 (not at 

all/poor) to 7 (substantial/outstanding). 

Leading edge 

research 

Applied 

Consulting 

service 

With 

public support 

Without 

public support 

Time 

Also, when looking at the customers’ overall 

evaluation of the institute in figure 3.5 

and 3.6 the productive value of the co-operation 

and the economic motives for using 

the institutes are not ranked high compared 

to indicators of knowledge transfer. The 

fact that we have not interviewed upon 

projects elder than three years, may be part 

Figure 3.4 > 

Average score project related questions, 19 institutes 

7 

6 

5 

4 

3 

2 

1 

0 

Ability to 

communicate / 

co-operate 

Understanding 

the customers 

requirements 

Knowledge, 

results-tranfer 

Quality of the 

scientific work 

Speed of work / 

efficiency 

Value for money 

/ profitability 

On the whole 

Has the project contributed 

to the improvement 

of competitive position 

76




Figure 3.5 > 

Average score institute related questions, 19 institutes 

7 

6 

5 

4 

3 

2 

1 

0 

Contribution to new 

knowledge, 

technology development 

Contribution to product 

development 

Developing the skills 

of the staff 

Contribution to the productive 

value of the firm 

(competitiveness) 

Total judgement 

Figure 3.6 > 

Average score institute related questions, 19 institutes 

7 

6 

5 

4 

3 

2 

1 

0 

Well equipped More cost-effective 

laboratory/facilities than doing it 

in-house 

The institute itself 

takes initiatives 

that are useful for 

the customer 

Strategically 

important R&Dcompetence 

for 

the customer 

Prefer to keep a 

small in-house 

department 

(Outsourcing) 

Joint industry 

projects (low cost 

way of looking at 

long term issues) 

Offers services 

relevant to 

customers need 

of the reason why economic results are 

given a low rating as the outcome of cooperation 

with the institute.However, the 

overall rating of the institute is based on 

the customer’s total experience with the 

institute, and may therefore include their 

experience of elder projects. Also, the fact 

that 73% are returning customers, support 

our believes that the customers benefit considerably 

from their co-operation with the 

R&D institutes, and that these benefits 

exceeds the short term economic benefits. 

However, that 70% did not consider other 

offers can both indicate a monopolistic situation, 

or that the customers are satisfied 

with earlier experiences. 

> 

4. AN ECONOMETRIC APPROACH FOR 

ESTIMATING A CONDITIONAL R&D 

SUPPORT MODEL 

4.1 Introduction 

The problems - and thus challenges - connected 

to the evaluation of public R&D policy 

are many and large. First, the main concepts 

(additionality and external effects) are 

difficult to grasp because they are not 

directly observable – i.e., they are latent variables. 

Both concepts are multidimensional, 

and each dimension should probably be 

measured by more than one observable indicator. 

Second, the effects are based on a 

77



counter-factual situation, where we try to 

compare the outcome of public R&D support 

with a hypothetical situation with no support. 

We therefore have to “construct” the 

counter-factual by using a quasi-experimental 

design. 

Both survey analyses based on ”subjective” 

data and microeconometric analyses based 

on ”objective” registered data have been 

employed in order to measure additionality 

effects of R&D support. The main objections 

raised against analyses within the former 

approach are problems connected to strategic 

answers, lack of a structural approach, 

and lack of control groups. The microeconometric 

approach, on the other hand, has 

mainly been criticised for using inadequate 

matching control groups in the case of selection 

biases. By combining the strengths of 

both approaches we may, however, overcome 

some of these objections. To do so, we 

outline a structural evaluation model based 

on a quasi-experimental design, where the 

problems connected to matching control 

groups are focused. The data to be fitted by 

such a model is constructed by linking R&D 

survey panel data at the project level with 

official registered R&D and industry panel 

data at the firm level. The survey data typically 

include indicators of R&D-attitudes and 

R&D cooperation both between firms and 

R&D institutions and among different firms, 

as well as various expectational (forwardlooking) 

indicators. These indicators will 

mainly serve the purpose of being control 

variables in the econometrical panel data 

model to be estimated. 

4.2 The fundamental causality problem 

A fundamental problem with any evaluation 

of public policy is to determine the causal 

ordering. Causality isconnected to explanations 

which necessitates an understanding 

of the underlying economic structure. Three 

criteria must be met in order to give such a 

causal explanation: 

1) The correlation and control criterion: 

there must be significant correlation 

between the cause and the effect variables 

after controlling for other possible 

covariates. 

2) The time criterion: the cause variable 

must precede the effect variable. 

3) The invariance criterion: the effect-model 

must have a structure which is reasonable 

invariant with respect to changes over 

time as well as structural changes in the 

marginal cause-model. 

The last criterion is the most problematic 

one, and often overlooked in the traditional 

evaluation literature. In order to make conditional 

inference (efficient estimation) for a 

given set of parameters of interest without 

loss of information, a cause variable must be 

weakly exogenous- i.e., the parameters of 

interest are variation free with respect to 

changes in the exogenous variables. If the 

parameters of interest in addition are invariant 

for changes in policy ( i.e., structural 

changes in the marginal cause-model), the 

cause variable is also “super exogenous”. 

Super exogeneity may be tested according 

to a procedure proposed by Engle and 

Hendry (1993): A simple marginal model for 

the supposed exogenous variable is estimated. 

If we need variables (e.g., dummies 

for important changes in policy) to stabilize 

this marginal model, we can test for invariance 

of the parameters of the conditional 

model by including these variables and test 

for their significance. If these variables have 

no impact in the conditional model and the 

cause variable is also weakly exogenous, the 

super exogeneity condition is fulfilled. In 

order to conduct such a test within our 

approach we thus need a marginal model of 

the public R&D support determination. 

4.3 Construction of 

the counter-factual 

The effect of R&D support is in the mainstream 

econometrical analysis represented 

by the difference in performance between 

supported and non-supported firms. The 

group of non-supported firms thus acts as 

a control group. Both traditional ”nonstructural” 

estimation models and structural 

estimation models, e.g., Euler equations, 

are employed.3 The effects of public 

R&D support are analysed with respect to 

both privately financed R&D (input) and 

pay-off (output). Klette et. al. (2000) 

78




review five different studies within this 

approach. They conclude that all the studies 

except for the Norwegian ones show 

positive and statistical significant effects 

on output. According to the Norwegian 

analyses of high-tech firms (reported in 

Klette and Møen (1998a and 1998b)), the 

effect of support on total factor productivity 

is significantly negative. But when looking 

at a broader set of measures, they are 

not able to find any systematic difference 

in performance between supported and 

non-supported firms. All the analyses show 

positive effects of R&D support on private 

R&D. But in a more comprehensive review 

of the empirical evidence accumulated 

over the past 35 years, David et. al. (2000) 

conclude that ”..[The] findings overall are 

ambivalent and the existing literature as a 

whole is subject to the criticism that the 

nature of the ”experiment(s)” that the 

investigators envisage is not adequately 

specified.” Typically,”objective” cross-section 

or panel data in such analyses will 

include both support-receiving firms and 

firms not even applying for support, neither 

of them constituting random samples. 

This non-random nature of the data may 

be due to both administrative and selfselection 

biases, the latter being the far 

most difficult case to cope with. 

4.4 A tentative causal model of long term 

effects from R&D support 

To be able to estimate effects of public R&D 

support we must distinguish different additionality 

effects as well as external effects. 

At the same time we must take account of 

possible feedback effects from output to 

both private R&D investment and the firm’s 

decision of applying for R&D support. This 

demands a simultaneous, structural evaluation 

model. Such a model is presented in 

figure 4.1. In this figure the circles represents 

unobservable latent variables while 

the arrows indicate the causal effects 

between them. 

R&D support will in this model be influenced 

by the support receiver’s ex ante R&D 

attitude. This variable serves the purpose of 

being a control variable to reduce the selectivity 

bias problem. R&D support is dependent 

on how the Research council selects 

projects that are support-worthy. In our 

model this is simply represented by a multidimensional 

variable called “criteria and 

other conditions”. The criteria for support 

are described in the council’s guidance documents 

for evaluation and ranging of projects, 

and will necessarily partly be based on 

subjective evaluations, e.g., expectations 

with respect to risk and additionality. Other 

conditions may be more objective aspects 

like age and size of the firm. However, R&D 

support may also be influenced by past 

experiences and results (output) of earlier 

supported projects – i.e., there is an arrow 

from output to R&D support. R&D support 

has a direct effect on private R&D investment 

as well as an indirect effect via 

changes in R&D orientation. The direct 

effect may be classified as the degree of 

“project input additionality” while the indirect 

effect will express the degree of 

“behavioural additionality”. Private R&D 

investment will also be influenced by “other 

conditions” like cash flow, sales, and so on. 

Included in “other conditions” are variables 

affecting marginal return (X) and marginal 

cost of capital (Y) of the R&D investment. 

Both private R&D investment and the firm’s 

R&D-orientation (e.g., R&D cooperations) 

can influence the “common R&D capital”, 

which may have an effect on output via 

internalized knowledge spillovers. Output is 

of course also influenced by “other conditions”, 

and will probably feed back to private 

R&D investments via effects on the 

marginal return (X). 

The model in figure 4.1 is complicated. To 

achieve a better grips with the model we 

therefore breaks it down into a partial additionality 

model which may be analysed separately. 

In this partial model we exclude the 

common R&D capital and the output variables 

to avoid the long term feedback effects. 

We can thus test this model on a cross-section 

of firm/project data. The model must accordingly 

be regarded as a model estimating only 

short term effects of public R&D-support. 

In figure 4.2 a combined structural and measurement 

model for input- and behavioural 

79



Figure 4.1 > 

A tentative causal model of effects from R&D support 

OTHER COND. 

OTHER COND. 

SELECT 

CRITERIA AND 

OTHER COND. 

R&D SUPPORT 

PRIVATE R&D 

INVESTMENTS 

OUTPUT 

(PRODUCT- 

EX ANTE 

R&D-ATTITUDE 

R&D 

ORIENTATION 

“COMMON” 

R&D 

Self-selection 

Project-input additionality 

Multidimensional latent variables 

(vector of explanatory variables) 

Behavioural additionnality 

Knowledge spilovers 

Unidimensional latent variables 

X = vector of variables influencing marginal return (MR) of the R&D investment 

Y = vector of variables influencing marginal cost of capital (MCC) of the R&D investment 

Figure 4.2 > 

A structural and measurement model of input and behavioural additionality 

I 1 

I 7 

I 5 

SELECT 

I I 

I 8 

9 

R&D SUPPORT 

PER EMPLOYEE 

EX ANTE 

CRITERIA AND 

R&D-ATTITUDE 

OTHER COND. 

I 2 

3 

I 10 

I 11 

R&D 

ORIENTATION 

FIRM SIZE 

I 4 

PRIVATE R&D 

INVESTMENT 

I 6 

80




Table 3.1 > 

Estimated coefficients and fit statistics 

Cause – effect standardised t-value* 

coefficient 

Criteria and other cond. 

• Expected input-additionality R&D support per employee 0.21 2.79 

• (Age of the firm) -1 R&D support per employee 0.37 2.70 

• Expected risk R&D support per employee - - 

Firm size R&D support per employee -0.09 -2.89 

Ex ante R&D attitude R&D support per employee - - 

R&D support per employee R&D orientation 0.11 1.64 

Ex ante R&D attitude R&D orientation 0.43 4.39 

Firm size private R&D investments 0.04 0.54 

R&D support per employee private R&D investments -0.22 -3.76 

R&D orientation private R&D investments 0.73 7.48 

χ 2 (26) = 31.80* RMSEA = 0.032* 

Marks: RMSEA - Root Mean Square Error of Approximation – is a test for good fit adjusted for the degrees of freedom. 

RMSEA < 0.05 indicates good fit. See e.g., Browne & Cudeck (1993). 

* based on robust estimation (corrected for non-normality) 

• Indicates that the coefficient is set to zero (is not significantly different from zero when freely estimated, 

and improves the fit by zero restriction) 

additionality is illustrated. The squares represent 

observable indicators (I 1 -I 11 ) which fill 

the latent variables (circles) with “empirical 

content”. The model is formulated stochastically, 

i.e., residuals are also estimated. The 

residuals are indicated by open arrows 

directed at the squares and the circles representing 

endogenous latent variables. 

In Nesset (2001) such a model is estimated 

on data from a survey among Norwegian 

firms in 1996. The sample consists of 181 

firms receiving R&D support in the User 

Oriented Research Schemes in 1993-95. 

Most of the questions in this survey (indicators) 

are formulated as 7-point Likert-scale 

indicators, while some as e.g., R&D support 

and firm size are measured on a continuous 

scale.The latent variable Ex ante R&D attitude 

is measured by three indicators (attitude 

towards R&D-collaboration with other 

firms; attitude towards R&D-collaboration 

with R&D institutes; attitude towards R&D 

in general as a strategy for long term survival), 

while R&D orientation is measured 

by two indicators (to what extent the firm 

has changed its behaviour from short term 

to long term R&D orientation; to what 

extent the firm is willing to engage in 

larger and more resilient R&D projects). The 

> 

parameters of the model are estimated by 

LISREL 8.30 (Sörbom, DuToit and DuToit 

(2000)). The fit of the estimated model is 

good according to a large set of tests. Table 

3.1 shows the main results. 

A main result from this analysis is that R&D 

support per employee (support intensity) 

has a significant negative effect on private 

R&D efforts. An explanation for this result 

might be that higher support intensity to a 

certain extent crowds out internal R&D 

efforts due to lack of R&D absorptive capacity, 

while less support intensity is more complementary 

to internal R&D efforts. 

5. EMPIRICAL FINDINGS 

OF RCN FUNDING SYSTEM 

Regretfully a unique identification number 

was not registered among the receivers of 

RCN within their system until 1999. However, 

through matching on names and addresses 

with the firm register at the Norwegian register 

authority and manual quality check on 

each firm performed by Møre Research we 

have managed to obtain a match with 853 

unique firm identification codes for 50% of 

the projects (1536). We will continue this 

81



Figure 5.2 > 

Risk classification of firms receiving RCN support compared to average risk classification of Norwegian firms (Central Bank of 

Norway). Projects ended 1993. 

100 

90 

80 

70 

Project ended 1993 - Low risk 

Project ended 1993 - High risk 

Norway - Low risk projects 

Norway - High risk projects 

60 

50 


30 

20 

10 

0 

1993 

1994 1995 1996 1997 1998 1999 2000 

Figure 5.3 > 

Risk classification of firms receiving RCN support compared to average risk classification of Norwegian firms (Central Bank of 

Norway). Projects ended 1994. 

100 

90 

80 

70 

Project ended 1994 - Low risk 

Project ended 1994 - High risk 

Norway - Low risk projects 

Norway - High risk projects 

60 

50 


30 

20 

10 

0 

1994 1995 1996 1997 1998 1999 2000 

work to hopefully increase the match. 

Figure 5.1 shows how the funding is distributed 

among 853 identified firms receiving 

support from RCN. One interesting aspect 

about this figure is the fact that when 

adding up the funding of these 853 firms 

and looking at the distribution, we find that 

10% of the firms have received 58% of the 

total RCN funding. We also find that this 

distribution is consistent over time. The 10 

firms with the largest share of funding are 

large well known Norwegian firms. 

Figure 5.1 and 5.2 illustrate the development 

in risk classification of firms receiving 

RCN support compared to average risk classification 

of Norwegian firms. The two figures 

illustrate this difference related to 

82




Figure 5.1 > 

Share of RCN support 

6% 

35% 

58% 

The 10% firms with the largest share of total RCN support during the period 

The 40% firms with the medium share of total RCN support during the period 

The 50% firms with the smallest share of total RCN support during the period 

firms with projects that ended in 1993 and 

1994 respectively. The risk classification system 

is based on accounting statistics with 

focus on developments inearnings and the 

share of long term debt based on the risk 

classification system of the Central Bank of 

Norway. The figures indicate that among 

the firms that received support to projects 

ended in 1993 and 1994, the share of low 

risk is increasing while the share of high 

risk is decreasing compared to the overall 

development within Norwegian firms. In 

other word, there is a tendency that the 

firms that have received support show a 

more positive development in risk classification 

than what is the average of 

Norwegian firms. 

83



> 

REFERENCES 

Bagozzi, R.P. and Y.Yi (1988): “On the 

Evaluation of Structural Equation Models”, 

Journal of the Academy of Marketing 

Science, Spring,Vol. 16, No. 1, 74-94. 

Bollen, K.A. (1989): Structural Equations 

with Latent Variables, John Wiley & Sons, 

Inc.New York. 

Browne,M.W. and R.Cudeck (1993): ”Alternative 

ways of assessing model fit”, in Bollen, 

K.A. and J.S.Long (eds): Testing Structural 

Equation Models, Sage Newbury Park, CA. 

Brown, Marylin A., Curlee Randall T.,Elliott, 

Steven R. (1995), “Evaluating technology 

innovation programs; the use of comparison 

groups to identify impacts” Research Policy 

24,1995 pp.669-684 

Buisseret, T.J., Cameron,H.M; Georghiou, L., 

(1995) “What difference does it make 

Additionality in the public support of R&D in 

large firms”, IJTM, Vol. 10, Nos 4/5/6, 

Evaluation of Research and Innovation 

Cohen,W.M., Levinthal, D.A, (1989), “Innovation 

and learning:The two faces of R&D”, 

The Economic Journal, no. 99, pp.569-596 

David, P.A., B.H. Hall, and A.A. Toole (2000): 

”Is public R&D a complement or substitute 

for private R&D A review of the econometric 

evidence”.Research Policy 29, 497-529. 

Fazio,R.H. 1989, On the Power and 

Functionality of Attitutes, Attitute, Structure 

and Function: A.R. Pratkanis, S. J. Breckler, 

A.G. Greenwald. Hillsdale, New Jersey: 

Lawrence Erlbaum Associates, Publishers 

Georghiou, L. (1998) “Issues in the 

Evaluation of Innovation and Technology 

Policy” Evaluation Vol 4(1), 1998 

Griliches Z. (1979), ”Issues in assessing the 

contributions of Research and development 

to productivity growth”, Bell Journal of 

Economics 10 (1), pp. 92-116 

Nesset, Erik, (2001): “Does the level of commercial 

R&D support make any difference” 

Mimeo, Aalesund University College, 

Norway. 

Hervik, Groth, (1996); “Customer Inquiry - 

Norwegian Building and Construction 

Institutes”, Report - Research Council of 

Norway 

Hervik, A. Rye, M (1997-2000), “Customer 

Inquiry”, Møre Research, Working paper no. 

M9904,M9905,M9906,M0010,M0011,M0012, 

and M0013 

Hervik, A., (1997); “Evaluation of User 

Oriented Research in Norway- The 

Estimation of long run economic impacts”, 

OECD Proceedings: Policy Evaluation in 

Innovation and Technology – Towards best 

practices. OECD, 1997 

Heckman, J. (1979): ”Sample selection bias 

as a specification error”, Econometrica 46, 

931-961. 

Heckman, J.J., H. Ichimura, J. Smith and P.E. 

Todd (1998): ”Characterizing selection bias 

using experimental data”. Econometrica 66, 

1017-1098. 

Hervik, A. og L. Bræin (2001): ”Mål og resultatstyring 

i NFR/IE. Analyser av PROVIS. Før 

og etterundersøkelser.” Arbeidsrapport 

M0108, Møreforsking Molde. 

Jöreskog, K.G., S. Du Toit, and M. Du Toit 

(2000): LISREL8: New Statistical Features. 

Chicago: Scientific Software International. 

Kauko, K. (1996): ”Effectiveness of R&D subsidies 

– a sceptical note on the empirical 

literature”. Research Policy 25, 321-323. 

Klette, T.J. and J. Møen (1998a): ”From 

growth theory to technology policy – coordination 

problems in theory and practice”, 

Nordic Journal of Political Economy (forthcoming). 

Klette, T.J. and J. Møen (1998b): ”R&D 

investment responses to R&D subsidies: 

A theoretical analysis and a microeconometric 

study”, Mimeo, presented at NBER 

Summer Institute 1998. 

84




Klette, T.J., J. Møen, and Z. Griliches (2000): 

”Do subsidies to commersial R&D reduce 

market failures Microeconometric evaluation 

studies”. The economics of technology 

policy, a special issue of Research Policy. 

Research Policy Vol. 29 (4-5) pp. 471-495 

Olsson, U.H., S.V. Troye and R.D. Howell 

(1999): ”Theoretic Fit and Empirical Fit: The 

Performance of Maximum Likelihood versus 

Generalized Leased Squares Estimation in 

Structural Equation Models”, Multivariate 

Behavioural Research 34, no 1, 31-59. 

Lynch, Chakravarti, Mitra, 1991: Contrast 

effect in consumer judgements: changes in 

mental Representations or in the anchoring 

of Rating scales Journal of consumer 

research 18, p. 248-98) 

Rye, Mette(2002); Evaluating impacts of 

public support to commercial R&D projects - 

can we trust verbal reports of additionality; 

Forthcoming in: Evaluation. 

RCN (2001): “Det norske forsknings- og innovasjonssystemet 

– statistikk og indikatorer 

2001” The research council of Norway, 2001. 

85


MEASURING 

‘RELATIVE EFFECTIVENESS’ – 

CAN WE COMPARE INNOVATION 

POLICY INSTRUMENTS 

ERIK ARNOLD 

AND 


Technopolis Group 

Erik Arnold is the co-founder and managing 

director of Technopolis, and has been active 

in R&D evaluation for almost 20 years. He 

has worked on over 50 R&D/innovation programme 

evaluations internationally, including 

national IT programmes such as the 

Alvey programme in the UK, the national 

Industrial IT programme in Sweden and the 

Irish national telecommunications programme. 

Trained at SPRU, he was for 5 years 

a university researcher in science policy and 

for 6 years a management consultant and 

project leader at Booz.Allen & Hamilton, 

where he focused on the management of 

technology-based companies. 

Patries Boekholt is founder and director of 

Technopolis BV in Amsterdam. Before joining 

Technopolis she worked for the TNO 

Centre for Technology Policy Studies since 

1989. She has over ten years of experience 

working on Research, Technology and 

Innovation policies in various countries and 

for international organisations such the 

European Commission and the OECD. 

This includes evaluation studies, strategic 

advice, international comparative, benchmark 

studies and improving policy implementation. 

87

Measuring ‘relative effectiveness’ – Can we compare innovation policy instruments 


> 


Lots of people ‘know’ that comparing the 

relative effectiveness of innovation policy 

instruments is easy to do, and that this can 

most usefully be done using cost-benefit 

analysis. One thing these people tend to 

have in common is that they have never 

actually tried to do the measurement 

involved in those comparisons. Their belief 

that such comparisons are feasible and 

meaningful is therefore based not in practice 

but in theory: specifically, in the assumptions 

of mainstream economics. Therefore, if 

we want to move innovation policy analysis 

and evaluation away from concern with the 

‘mission impossible’ of comparative costbenefit 

analysis and on to a more tractable 

and useful mix of approaches, we have to 

tackle theory as well as practice. 

In this paper, we show that different 

theoretical perspectives imply different things 

about the feasibility and usefulness of comparing 

innovation policy instruments. The lack 

of major practical achievements internationally 

supports the idea that monetary comparisons 

of effectiveness are problematic. 

Our shifting understanding of innovation 

processes and their links to socio-economic 

well being suggests the need to shift some 

of our attention in evaluation and policy 

development towards the systems level. 

We conclude that, in the real world, it is 

more realistic to think of evaluation and 

policy analysis as helping us continually to 

improve the performance of innovation 

systems in a somewhat ad hoc and pragmatic 

way, rather than imagining we can 

optimise at the overall level. 

89


> 

1 

See Christopher 

Freeman, Technology 

Policy and Economic 

Performance: Lessons 

from Japan, London: 

Frances Pinter, 1987; 

Bengt-Åke Lundvall, 

National Systems of 

Innovation: Towards a 

Theory of Innovation 

and Interactive 

Learning, London: 

Pinter, 1992; RR 

Nelson, National 

Innovation Systems, 

New York: Oxford 

University Press, 1993 

2 

This is generally 

called a National Innovation 

System in the 

literature – essentially 

because the key authors 

are researchers on 

questions to do with 

economic innovation. 

This is a very unhelpful 

name, since all three 

words are misleading. 

In fact, such systems 

can be regional and 

international as well 

as national. They incorporate 

the whole of 

knowledge production 

and consumption, 

not just economic 

innovation. 

2. WHAT KIND OF EFFECTIVENESS 

What we mean by ‘effectiveness’ in the context 

of innovation policy – and what kind of 

effectiveness it is important to understand 

and measure – depends strongly on which set 

of theoretical spectacles we choose to wear. 

Over the past ten years or so, there has been 

a revolution – a ‘paradigm shift’ – in the way 

we understand the relationship between 

research, innovation and socio-economic 

development. Conventional, neo-classical 

economics viewed firms, in effect, as 

autonomous and rational robots using perfect 

information. In the context of technological 

change, much of the traditional, neoclassical 

framework has been superseded 

during the 1990s, through a convergence of 

evolutionary economics (which stresses firms 

as ‘learning organisations’) and research on 

the innovation process. 

This new National Research and Innovation 

Systems (NRIS 1 ) 2 approach stresses the idea 

that firms and other economic actors have 

‘bounded rationality.’ This makes knowledge, 

learning and institutions key to overall 

economic performance. In the new view, 

economic actors are no longer autonomous 

robots, but are deeply interwoven into the 

economic fabric. The unit of analysis is no 

longer only the individual firm but also the 

‘system’ of networks within which firms operate. 

National economic performance is explained 

as the performance of this total system. 

A second key idea, which stems from the 

central role of learning, is that of historical 

path dependence. What a company or institution 

can do today depends upon what it 

could do yesterday 3 and what it has learnt in 

the meantime 4 . Another implication of path 

dependence is co-evolution among institutions 

such as funding agencies, which strive 

to improve their performance within the 

existing institutional division of labour. As a 

result, they adapt to each other’s presence 

and different National Research and Innovation 

Systems develop different institutional 

ways to achieve similar ends. 

Exhibit 1 indicates some of the differences 

among the core assumptions of the neoclassical 

and NRIS approaches. Most neoclassical 

economists would object to this characterisation. 

They would rightly point out that 

much of twentieth century microeconomics 

devoted itself to refining these core assumptions 

(which derive, essentially, from a simplified 

view of a special case: trading corn in 

agricultural markets) and taking account of 

imperfect information, imperfect competition 

and so on. But this is the central point. 

Just as the Copernican hypothesis that the 

sun and all the heavenly bodies go round the 

earth can be protected by adding epicycles to 

their orbits, so the assumptions of neoclassical 

economics can successively be refined to 

improve the fit with reality. However, the 

description that results is complex and messy. 

In contrast, the NRIS approach offers the 

simplicity and elegance that attract us to 

theories. If the sun goes round the earth, we 

can lose all those inconvenient epicycles and 

some of the rather odd consequences that 

we get if we combine Copernicus with more 

recent astronomical observations! 

Exhibit 1 > 

Some Core Assumptions of Neoclassical and NRIS Approaches 

And they are only 

‘systems’ in the sense 

of many things being 

System 

behaviour 

Neoclassical 

Generally predictable 

Can be derived from micro level 

NRIS 

Not fully deducible from micro level 

Time and place dependent 

connected together. 

They are not systems 

in the sense that their 

behaviour can readily 

be predicted and 

Interactions 

Rational, with perfect capabilities 


Factors and actors are substitutable 

Rational with imperfect capabilities 


Learning, path dependency, co-evolution, 

hence actors are not substitutable 

controlled, or planned 

in the old soviet 

sense. 

Actors 

Rational robots 

No history 

Boundedly rational, imperfectly capable 

Have history 

90



Exhibit 2 > 

Implications of Theoretical Perspectives for Comparisons 

Across different 

categories of 

instruments 


time horizons 


NRIS 

Neoclassical core assumptions 

Unproblematic, since instruments 

are substitutable and both costs 

and benefits can be monetarised 

Can be done to some extent. 

Economic actors act in response to 

short run economic opportunities, 

but there can also be long run 

effects. The length of the ‘long 

run’ is not given, and may vary 

Unproblematic, as all NRIS follow 

the same set of general economic 

laws 

NRIS 

Instruments have different purposes. 

Many can therefore not be substituted 

for each other. 

Actors act both for reasons of short 

run economic advantage and in order 

to secure ‘softer’, long run benefits 

such as learning. Comparability of 

benefits can therefore vary over time 

Achievable benefits may vary among 

NRIS, depending on initial conditions 

at the start of the intervention 

3 

Nathan Rosenberg, 

Perspectives on 

Technology, Cambridge 

University Press, 1976 

4 

Bengt Åke Lundvall 

(ed), National Systems 

of Innovation: 

Towards a Theory of 

Innovation and 

Interactive Learning, 

London: Pinter, 1992 

5 

Charles Edqvist (ed), 

Systems of 

Innovation, London: 

Frances Pinter, 1997 

6 

Rodrigo Arocena and 

Judith Stutz, ‘Looking 

at national systems of 

innovation from the 

South,’ Industry and 

Innovation, Vol 7, No 1, 

June 2000, pp 55-75 

For many purposes we do well to cling to 

the achievements of mainstream economics. 

It is far better articulated than the NRIS 

approach. Like Newtonian physics, it lets us 

do comparatively simple sums, and it works 

as long as we understand and steer clear of 

its limitations. Unlike in Newtonian physics, 

however, we are rather more likely to trip 

over these limitations in our everyday lives – 

not least in innovation policy making. 

Our choice of theory matters, not only 

because some theories fit the facts better 

than others, but also because – in a policy 

making context – theories are normative. 

Neoclassical economics pushes us towards 

policies and regulations that make the 

world behave more like the economic textbook 

description, making money the measure 

of all things and treating policies and 

policy instruments as subsitutable. Neoclassical 

systems are capable of being optimised, 

using a common set of instruments. 

As Edqvist argues 5 “the notion of optimality 

is absent from the systems of innovation 

approaches. Hence, comparisons between 

an existing system and an ideal system 

are not possible.” The NRIS approach is 

nonetheless normative, in the sense that it 

claims that certain system characteristics – 

such as strong network links between 

actors – are likely to improve performance. 

6 An NRIS world requires a differentiated 

set of policy instruments to improve 

performance. 

> 

What do these two theoretical perspectives 

then imply about whether comparisons of 

innovation policy instruments are feasible or 

meaningful As Exhibit 2 indicates, different 

theoretical perspectives lead to different 

answers. 

3. WHAT DOES EVALUATION TELL US 

Finance ministries tend to be run by mainstream 

economists, who wear neoclassical 

spectacles. They want to know about the 

cost-effectiveness of policy instruments, 

which is a question we all (as tax payers) 

care about. Often, however, their neoclassical 

spectacles can blind them to important 

characteristics of the NRIS world. They ask 

questions which make neoclassical sense but 

are naïve in NRIS terms. 

The body of evaluation literature dealing 

with innovation policy is immense. There is 

neither time nor space to do it justice here. 

Some ‘stylised facts’ which emerge from that 

literature include 

• A large variety of intervention types exists 

• There is policy learning: different NRIS imitate 

each others’ interventions, so there 

are trends – and also fashions 

• Well-managed interventions tend to 

achieve their medium-term aims – even if 

we rarely know much about their sideeffects 

(displacement, for example, or 

dead weight) 

91


7 

Patries Boekholt, 

Maureen Lankhuizen, 

Erik Arnold, John 

Clarke, Jari Kuusisto, 

Bas de Laat, Paul 

Simmonds, Susan 

Cozzens, George 

Kingsley and Ron 

Johnston, An 

International Review 

of Methods to 

Measure Relative 

Effectiveness of 

Technology Policy 

Instruments, 

Amsterdam: 

Technopolis, 2001 

• Many of these aims and achievements 

relate to learning at the level of actors and 

networks 

• Short-term economic benefits can be difficult 

to identify, especially where research 

is involved 

• More generally, benefits of different interventions 

can appear with different time 

lags following the intervention 

Policy makers generally hope to establish a 

positive correlation between the use of government 

programmes and the effect on for 

instance innovation behaviour, R&D expenditure, 

productivity, competitiveness, export 

position, and employment. This effect 

should be established at the private level 

(those directly benefiting from government 

support) and at the public level (the effect 

beyond those directly involved, whether 

economic or social). 

We can see that usually impacts are defined 

in terms of direct outputs of a government 

instrument such as the change in the level of 

R&D expenditure, or the number of patents 

registered, commercial exploitations of project 

results. Second order effects have been 

measured at the level of the direct participants 

of technology programmes and 

described in terms of commercial returns of 

products which had input from the support 

programme, new business opportunities, 

etc. Few evaluation studies have measured 

third order effects on the wider economy 

and society. 

To establish the cost effectiveness of a policy 

programme requires aggregating the 

impact on single actors (micro-data), to the 

sector (meso-data) or even economy wide 

level (macro-data). Techniques used to 

aggregate this information have their own 

methodological challenges and drawbacks. 

The international evaluation community 

acknowledges that methods and tools 

developed to date are far from perfect. 

The most commonly used effectiveness and 

cost-benefit analysis methods used are 

• Micro-level case studies and in depth interviews 

which results are aggregated for the 

wider population of programme users 

• Large scale telephone and postal surveys 

which ask participants to make self-assessments 

of achieved results and effects on 

criteria such as turnover, export and 

employment 

Cost-benefit studies suffer from some key 

methodological problems, which the evaluation 

expert community summarises as 

• The attribution problem: how can you isolate 

the effect of a policy instrument on 

the performance of one firm or a group of 

firms, given the many additional factors 

that influence that performance 

• The time lag between research, innovation 

and economic effects for those directly 

involved in the programme and even more 

for those not participating in the programme. 

Years can go by before a commercial 

return of investment can be achieved 

at the individual firm level. The effects of 

technology support programmes should be 

measured short, medium and long term. 

The data on the medium and long-term 

effects are usually lacking. There are few 

programmes that have a sufficient long history 

to be able to analyse effects in the 

longer term, which many experts and practitioners 

estimate at 20 years. (Norway’s 

user-directed R&D programmes are unusual 

in that their effects on participating 

firms have been investigated at periods up 

to ten years after project completion.) 

• The quantification of the many qualitative 

effects that are included in the objectives 

of the programme such as networking, 

improving the absorptive capacity and 

competences of firms 

In a recent study 7 for the Dutch Ministry of 

Economic Affairs, we found that in most of 

the benchmark countries considered, policy 

makers are not convinced that the empirical 

basis for the current cost-benefit analysis 

methods is solid enough to use this as the 

sole approach to decide upon the effectiveness 

and usefulness of certain individual 

programmes. The uncertainty of the direction 

of the many causal linkages in the innovation 

process, prohibit solid conclusions on 

the exact impact of the public funded 

contribution. 

92



Impact assessments are regularly undertaken 

in the USA, Australia, and Norway. 

They are used incidentally in Canada, 

Finland, and by the European Commission. 

Germany, France, New Zealand and the UK 

hardly ever use these more quantitative 

evaluation approaches to assess the effectiveness 

of technology and innovation policy 

instruments. 

While present quantitative methods – especially 

cost-benefit analysis – have severe limitations 

at the level of explaining the effects 

of single programmes or interventions, they 

do have a useful contribution to make 

in combination with more qualitative 

methods. The reality is that, while we have 

many tools available to us in conducting 

R&D evaluations, none of them works all 

that well on its own. It is important to use a 

combination of methods in any evaluation 

and to test for consistency between the findings 

emerging from each. Only then can we 

reasonably be confident that our overall 

evaluation findings are trustworthy. 

Second, we often have a prejudice in favour 

of findings and conclusions that can be 

expressed in numbers. Responsible use of 

quantitative effectiveness measurement 

includes analysis of the sources and sizes of 

potential errors and a discussion of the 

methodological difficulties involved, in 

order to help the reader interpret the numbers. 

If this is not done, many readers are 

inclined to treat as ‘hard’ numbers that are 

very ‘soft’ indeed. 

Third, quantification and this systematic and 

honest analysis of potential sources of error 

involve tight discipline in experimental 

design and documentation. In our experience, 

qualitative research designs can often 

benefit from such increased discipline, so that 

incorporating a quantitative aspect can help 

raise the quality of the overall evaluation. 

Last, but not least, some quantitative effectiveness 

evaluation is now being conducted 

within the context of tighter programme 

planning and monitoring processes, providing 

a feedback loop in a larger system of 

policy development and implementation. 

The case of the Research Council of Norway 

provides a useful example. Other systematic 

approaches include the ROAME (Rationale, 

Objectives, Assessment, Monitoring, Evaluation) 

scheme used by the UK civil service and 

the Logical Framework approach now 

adopted by SENTER in the Netherlands and 

parts of the European Commission. This 

systematisation of planning and evaluation 

is likely to improve our collective ability 

to reach goals, not least by improving the 

quality of goal definition. 

Our comparative study found very few 

examples of relative effectiveness measurement 

in the countries under review (Exhibit 

3). We need to distinguish here between 

those attempts that have compare the 

degree to which different instruments have 

contributed to government objectives or 

targets and attempts where a comparison is 

made on cost effectiveness in terms of public 

rates of return on investment. The few 

attempts we have found concentrated on 

the former rather than the latter type of 

comparison. 

Where comparisons have been made they were 

• Part of a systems analysis, where the comparative 

effectiveness measure was done to 

help decide the optimal policy mix in relation 

to the key objectives of ministries. 

Individual instruments were compared on 

the basis of their contributions to these 

general objectives. The systems of instruments 

were analysed: whether they 

overlapped, whether gaps in the policy 

portfolio existed, and what the relative 

appreciation was of these programmes by 

their users. The relative effectiveness measurement 

was not made on the basis on best 

performance in terms of rate of return on 

public investment, although ex-post evaluations 

provided strategic information 

• More an element of an ex-ante strategic 

policy formulation process than an ex-post 

evaluation exercise based solely on cost 

benefit analyses 

• Based on a mix of evaluation and assessment 

tools 

• Focused on a limited set of programmes 

with more or less similar objectives 

93


Exhibit 3 bis > 

Effectiveness assessment in the benchmark countries and the EU 

Qualitative 

assessment of 

effectiveness 

Output 

measurement 

quantitative 

analysis 

Outcome 

measurement 

quantitative 

analysis 

Quantitative 

assessment 

wider economic 

impact 

Cost-benefit 

analysis 

(C-BA) 

Comparative 

analysis on basis 

of performance 

indicators 

Australia 

Canada 

Finland 

France 

Germany 

New Zealand 

Netherlands 

Norway 

UK 

USA 

EU 

Note: Full (black) squares indicate that this kind of assessment is undertaken in this country; Half (grey) squares 

indicate that this kind of assessment is undertaken in this country, but does not have a major emphasis; Empty 

(white) squares indicate that this kind of activity is not undertaken in this country 

Our study identified a number of objections 

and drawbacks from the perspectives of policy 

practitioners and evaluators 

• The time lag on effects differs from programme 

to programme. There is not a single 

moment in time when the ultimate 

effects can be compared between more 

than one instrument 

• Given the attribution problem, there is a 

risk of double counting the financial effect 

of various policy measures used simultaneously 

by companies. The more instruments 

are included in the comparison the more 

difficult it will be to define control groups 

• The variation in risk levels of different 

policy instruments where some low risk 

programmes have many incremental and 

short term effects whereas high risk programmes 

have fewer but potentially more 

radical effects in the longer term. 

Comparing the two in the medium term 

would always favour the low risk programme 

and therefore lead to a certain 

risk averseness of public action, whereas 

the ‘market failure’ justification assumes 

that government acts when risks are too 

high for the private sector 

• The failure to quantify the ‘softer’ effects 

that governments want to achieve in 

changing the behaviour of the target 

groups in the current cost benefit analyses 

• Possible changes in the context of the 

‘problem’ that a government action wants 

to address. Even though an instrument 

performed perfectly well in terms of costbenefit 

analysis, it could be that the context 

of the firms and the innovation system 

has changed drastically in the meantime, 

making the same instrument ineffective in 

the future. This would argue against using 

ex-post cost benefit analysis as the sole 

input for policy decisions 

• Effectiveness measurements typically 

measure only some of the effects of programmes. 

In particular, cost-benefit analyses 

tend to focus on the private returns to 

intervention rather than the social returns 

or ‘externalities,’ which normally justified 

the intervention in the first place, but 

which are harder to measure. Because the 

ratio between the measured and unmeasured 

effects of individual programmes is 

not known, different cost-benefit measures 

are typically incommensurable and 

cannot validly be compared 

94



8 

Ken Arrow , 

‘Economic Welfare 

and the Allocation of 

Resources for 

Invention,’ in Richard 

Nelson (Ed.) The Rate 

and Direction of 

Inventive Activity, 

Princeton University 

Press, 1962; see also 

Richard Nelson, 

‘The simple economics 

of basic scientific 

research,’ Journal 

of Political Economy, 

1959, vol 67, 

pp 297-306 

9 

His argument is, 

however, conceptually 

flawed.It simply 

assumes that there is 

under-investment in 

basic research compared 

to an imagined 

welfare-economic 

optimum. It makes this 

assumption because it 

implicitly accepts the 

‘linear model’ account 

of the role of science 

> 

as causing economic 

innovation and 

development. In fact, 

no one has observed 

or calculated what 

such an optimum 

would look like 

10 

Erik Arnold and 

Ken Guy, ‘Diffusion 

policies for IT: the way 

forward,’ OECD/ICCP 

Expert group on the 

economic implications 

of Information 

Technologies, Paris: 

OECD, 1991 

• Policy instruments should be seen in the 

context of their role in the innovation system 

and the specific objectives and target 

groups they address. You can not compare 

on one parameter, i.e. effectiveness alone. 

As long as the basic methodological problems 

surrounding effectiveness measurement of 

single programmes are not solved, the same 

problems will affect any relative effectiveness 

measurement comparing more programmes. 

Given these methodological difficulties and 

policy objections, we have found only three 

instances where some form of relative effectiveness 

measurement has been conducted. 

The use of relative effectiveness measurement, 

tried in only a small number of countries 

suggests that 

• This should always be done in the context 

of how the policy portfolio addresses the 

different issues in the innovation system 

• It should not be based on merely ex-post 

evaluations using cost-benefit techniques. 

The methodological problems of using 

these techniques on single programmes prevent 

a robust outcome of any comparative 

approach. Those that have engaged in these 

activities have used a mix of methods including 

expert views and foresight activities 

4. INNOVATION POLICY IN A SYSTEMS 

PERSPECTIVE 

The idea of ‘market failure’ leading to 

under-investment in research has been the 

principal rationale for state funding of R&D 8 

in the post-War period. Of course, governments 

had been funding research long 

before the economics profession produced a 

reason. Arrow is generally credited with 

describing the three major sources of market 

failure which – from a neo-classical perspective 

– make it useful for government to 

fund research 

• Indivisibility, because of the existence of 

minimum efficient scale 

• Inappropriability of the profit stream from 

research, leading to a divergence between 

public and private returns on investment. 

This results from two essential (and economically 

efficient) freedoms that scientific 

researchers have: namely to publish 

and to change jobs 

• Uncertainty, namely divergences in the 

riskiness of research respectively for private 

and public actors 

Arrow’s argument was particularly relevant 

to more ‘basic’ (and, by implication, generally 

applicable) forms of knowledge because 

capitalists’ inability to monopolise the 

results of such research meant they would 

be least likely to invest in it. 9 

More recent developments in theory do not 

invalidate but extend the neo-classical idea 

of market failure. The market failure 

approach assumes away key deficiencies of 

real companies, not least what we many 

years ago called ‘capability failures’ 10 as well 

as failures in systems. We can think of these 

failures as belonging to four types 

• Capability failures 11 . These amount to 

inadequacies in companies’ ability to act in 

their own best interests, for example 

through managerial deficits, lack of technological 

understanding, learning ability 

or ‘absorptive capacity’ 12 to make use of 

externally generated technology 

• Failures in institutions 13 . Not only companies 

but also other social institutions such 

as universities and research institutes, 

patent offices and so on need to work well 

if the NIS is to facilitate innovation and 

growth. Rigid disciplinary orientation in 

universities and consequent inability to 

change with changes in knowledge would 

be an example of such an institutional failure. 

Failure to provide adequate investment 

in knowledge institutions would be 

another 

• Network failures. These relate to problems 

in the interaction among actors in the 

innovation system, and can themselves be 

of several types 

- Inadequate amounts and quality of interlinkages, 

as where there is low trust 

among companies or where universities 

are isolated from their social context 

- ‘Transition failures’ and ‘lock-in’ failures, 

where clusters or innovation systems fail or 

take on board new technological opportunities 

or remain locked into old ones 14 

95


11 

For analysis, see 

Erik Arnold and Ben 

Thuriaux, Developing 

Firms’ Technological 

Capabilities, report to 

OECD, Brighton: 

Technopolis, 1997, 

which may be downloaded 

from www. 

technopolisgroup.com 

12 

W Cohen and D 

Levinthal, ‘Absorptive 

capacity: a new perspective 

on learning 

and innovation,’ 

Administrative Science 

Quarterly, No 35, 

1990, pp 128 - 152 

13 

This is similar to the 

idea of ‘hard institutional 

failure’ discussed 

by Bo Barlsson 

and Staffan Jacobsson, 

‘In search of useful 

public policies: Key 

lessons and issues for 

policy makers,’ in Bo 

Carlsson (ed) 

> 

Technological Systems 

and Industrial 

Dynamics, New York, 

Kluwer Academic 

Publishers, 1997. 

Their ‘soft institutional 

failures’ correspond to 

part of our ‘framework 

failures’ 

14 

Keith Smith, ‘Economic 

infrastructures and 

innovation systems,’ 

in Charles Edqvist 

(ed), Systems of 

Innovation: Technologies, 

Institutions and 

Organisations, 

London: Cassel, 1997 

- Various problems in industry structure, 

such as too intense competition or 

monopoly, which stifle innovation, or the 

absence of key complementarities (such 

as when a cluster’s development is stifled 

by the lack of a crucial type of producer) 15 

• Framework failures. Effective innovation 

depends partly upon regulatory frameworks, 

health and safety rules etc as well as 

other background conditions, such as the 

sophistication of consumer demand, culture 

and social values 16 . Deficiencies in these 

frameworks can have a negative effect on 

innovation and economic performance 

These failures justify state intervention not 

only through the funding of basic science, 

but more widely in ensuring that the 

Innovation System performs as a whole – 

always provided that the state is actually 

capable of reducing failure. Because systems 

failures and performance are highly 

dependent upon the interplay of characteristics 

in individual systems, there can be no 

simple rule-based policy as is possible in relation 

to the static idea of market failure 17 . 

Rather, a key role for state policy making is 

‘bottleneck analysis’ – continuously identifying 

and rectifying structural imperfections. 

5. EVALUATION AT A SYSTEMS LEVEL 

The R&D evaluation community does not have 

an opportunity to dodge the issue of systems 

evaluation. Without evidence, it is not possible 

to operationalise the systemic understandings 

emerging from newer theory. Policy makers 

have read the new books, and are increasingly 

asking about the systems level. (In Sweden, the 

prioritisation of the systems view has gone so 

far that the new agency responsible for funding 

technological development and absorptive 

capacity in called the Swedish Agency for 

Innovation Systems – VINNOVA.) Increasingly, 

R&D funders are moving from single-point 

interventions relating to individual research 

projects or companies to network and cluster 

interventions. There is an increasingly explicit 

desire to evaluate policies and portfolios as well 

as individual programmes – partly in order to 

understand systems effects but also in part to 

reduce the ‘evaluation tax’ on funding activity. 

Exhibit 4 sketches what we mean by a 

‘National Research and Innovation System’: 

namely, all the actors and activities in the 

economy which are necessary for industrial 

and commercial innovation to take place 

and to lead to economic development. 

The current orthodoxy is that economic 

well-being is founded on a well-functioning 

NRIS, in which not only the actors shown in 

Exhibit 4, but also the links between them, 

perform well. In contrast to earlier views, 

which focused on entrepreneurs as individual 

heroes, innovation and learning are 

now seen more as network or collective 

activities. 

We can understand systems of research and 

innovation in much the same way as we 

understand sub-atomic physics: not by 

observing the whole system at once, but by 

looking at individual pieces and – through a 

mixture of experiment and conjecture – figuring 

out how they relate to each other. 

Evaluation does not get easier as we move 

from the project and programme levels 

towards considering sub-systems and systems. 

The scale and complexity of the phenomena 

mean that we often cannot treat 

them in as much detail as is possible when 

we operate at a smaller scale. Tactics for 

dealing with this include using methods to 

reach judgements at rather aggregated 

levels (such as modified peer review of programme 

portfolios) and increased, explicit 

use of theory and best practice as benchmarks 

against which to judge performance. 

We can use theory and experience-based 

ceteris paribus assumptions to hold constant 

certain system characteristics while successively 

investigating sub-systems. Evaluation, 

like the policy making process, becomes 

increasingly evolutionary, no longer seeking 

an overall optimum. (To think in optimisation 

terms remains more useful at the project/programme 

levels.) 

Evaluation therefore becomes in a certain 

sense less rigorous (because it is less complete) 

as we move to higher levels in the 

innovation system. This means we must build 

evaluation systems and frameworks that can 

96



Exhibit 4 > 

Major Components of a National Research and Innovation System 

Demand 

Consumers (final demand) 

Producers (intermediate demand) 

➐ 

➐ 

Framework Conditions 

Financial environment; taxation and incentives; 

propensity to innovation and entrepreneurship; mobility 

➐ 

❷ 

Industrial System 

❷ 

Education 

and Research System 

Political System 

Large companies 

➐ 

Mature SMEs 

➐ 

❷ 

New, 

technology-based firms 

➛ 

➛❿ 

➛ 

Intermediaries 

Research institutes; 

Brokers 

❿ 

➛❿ 

❿ 

Professional education, 

training 

Higher education 

and research 

Public sector research 

➛ 

➛ 

Government 

Governance 

RTD policies 

➐ 

➐ 

❷ 

Banking, 

venture capital 

IPR 


Infrastructure 

Information and 

business report 

❷ 

Standards 

and norms 

Source: Erik Arnold and Stefan Kuhlman 

15 

Franco Malerba 

‘Public policy and 

industrial dynamics: 

An evolutionary perspective,’ 

in Charles 

Edqvist (ed), Systems 

of Innovation: 

Technologies, 

Institutions and 

Organisations, 

London: Cassel, 1997 

16 

Smith, Ibid 

17 

Johan Hauknes and 

Lennart Norgren, 

Economic Rationales 

of Government 

Intervention in 

Innovation and the 

Supply of Innovation- 

Related services, 

STEP Report 08 1999, 

Oslo: STEP Group, 

downloadable from 

www.step.no 

operate with an imperfect evidence base and 

which therefore involve a mixture of systemic 

and micro evaluation. Micro work is done 

sometimes to follow up issues identified at 

the systems level but often still to provide the 

traditional benefits of such evaluation: learning; 

accountability; quality control. 

As Exhibit 5 illustrates, changing the focus 

from lower to higher levels of aggregation in 

the innovation system implies a shift in the 

types of methods that are appropriate. While 

traditional, quality-oriented peer review evaluation 

remains extremely important at the 

level of projects, this methods needs increasing 

modification for use at higher levels. As 

higher levels of aggregation, the notion of 

‘peer review’ in a strict sense breaks down. 

‘Peers’ are peers in the sense that they are the 

scientific equals of those whose work they 

judge. At the programme level and above, it 

would probably be more correct to speak of 

‘expert review’ by people who qualify as 

‘peers’ on the project level. Equally, at high 

system levels, less and less of the evaluation 

problem is amenable to expert or peer judgement. 

There is a corresponding increase in 

the relative importance of socio-economic 

methods at higher system levels. 

At all levels, the job of evaluation is to ask 

1. Are we doing the right thing 

(appropriateness) 

2. What are the results of our action 

(impacts) 

3. Could we do it better (effectiveness) 

As we move up the systems hierarchy, too, 

so the balance among the traditional 

evaluation questions shifts. To some 

degree, good evaluations tend always to 

question assumptions made at higher 

policy levels. That is, they raise issues about 

appropriateness. However, at lower system 

levels, evaluations tend to deal with 

actions which implement policies, and 

therefore to place more effort on impacts 

and effectiveness than on the more fundamental 

question of appropriateness. For 

example, an evaluation of a mechanical 

engineering R&D programme is more likely 

to find that there are gaps in the population 

of users served than to challenge the 

principle that such a programme is needed. 

Evaluation at higher system levels is more 

likely to produce a challenge to policy, 

because the evaluators have to analyse 

some of the same systemic problems as 

policy makers. For example, an evaluation 

of a research council can rapidly lead to 

questioning of national research policy. 

What, then, are the roles of evaluation in a 

complex systems policy world We see 

broadly three: support to the policy making 

97


Exhibit 5 > 

Appropriate Evaluation Methods at Different System Levels 

System 

Portfolio 

Programme 

Project 

Peer review 

and modified 

peer review 

Socio-economic methods 

* Scientometrics 

* Surveys 

* Case studies 

* Micro- and macro-economics 

* Comparative study, best practice analysis 

* Growing use of non-R&D specific methods, eg cluster analysis 

Balance of Appropriate Methods 

process; quality assurance; and providing 

accountability. 

Support to the policy making process 

involves different types of bottleneck 

analysis 

• One relates to new and existing NIS 

performance indicators, which provide 

policy puzzles. For example, ‘If we are 

spending more on R&D than any other 

country in the world, why is our GDP per 

person declining’ 

• A second involves responding to more ad 

hoc policy problems. ‘The research community 

seems unhappy with the way we 

organise research funding. Should we 

reform the funding institutions’ Other 

examples would be include evaluations of 

- Networks: industrial clusters or districts, 

university-industry relations, the role of 

research institutes in the NIS 

- Framework conditions: the effects of tax 

law on research funding, intellectual 

property rights and innovation 

• The third is routine review of sub-systems 

of the NIS, such as evaluations of 

- Institutions: research councils, technology 

development agencies, universities, 

research institutes, technology-transfer 

organisations, change agencies 

- Areas of knowledge: national positions 

in various basic sciences and developing 

technologies 

It is clear from these examples that, at the 

systems level, the act of evaluation converges 

> 

with the provision of the evidence base 

needed for policy making. (Evaluative activities 

at this level deal with specific institutions, 

fields, etc in a specific time and place. 

In parallel, research goes on to provide general 

evidence upon which to base policymaking. 

Inevitably, the categories overlap.) 

Quality assurance and accountability are 

partly supported by these systemic types of 

evaluation. But the fact that these higherlevel 

questions are tackled does not abolish 

the need for more ‘routine’ forms of evaluation 

at portfolio, programme and projects 

levels. Regular evaluation (and meta-evaluation) 

at these levels not only supports their 

operation but also provides a stream of inputs 

into higher-level policy making and the generation 

of systems-level evaluation questions. 

CONCLUSIONS 

A number of important conclusions follow 

from the theoretical shift and the practice 

described in this paper. 

• Even though they may be attainable in 

theory, accurate, scientifically defensible 

impact assessments of innovation policy 

instruments seem to be beyond our reach 

• It follows from this that comparisons 

among attempted assessments are likely to 

be misleading 

• If we accept the NRIS view of non-substitutability 

of instruments, then comparisons 

will not in any case help much. The 

98



natural policy implication of a comparison 

of rates of return to different instruments 

is that the state should invest where the 

return is highest. In the NRIS view, this is 

about as sensible as trying to drive faster 

by throwing away your car and buying a 

bigger engine 

• If we introduce sustainability as an additional 

impact criterion for innovation 

policy, we need to widen our current concept 

of the innovation system. This would 

mean for instance adding other framework 

conditions (e.g. environmental regulation) 

and organisations in the pictuce. 

Understanding this wider complexity is 

a joint challenge for innovation and 

environemental policy analysts. 

We conclude that the economics-inspired 

desire to optimise based on rates of return 

has to be given up. Rather, in the NRIS view, 

we can take our inspiration from manufacturing 

engineering. We are all familiar with 

the idea of kan-ban, or ‘just in time’ manufacturing. 

Generally, we think of this as a 

way to substitute information and logistics 

for stock. Reducing stocks reduces work-inprogress 

capital requirements and thereby 

saves production cost. But the most 

interesting aspect of kan-ban is not stock 

reduction. It is the fact that reducing stock 

stresses the production system, exposing 

weaknesses and bottlenecks. This tells you 

where to direct engineering effort. Having 

cleared one bottleneck, you take away more 

stock and look for the next. In this way, the 

productive system itself guides the engineering 

interventions. Combined with creative 

use of the intelligence of those who 

work in the production system, we attain a 

process of continuous improvement, or kaizen. 

None of this removes the opportunity 

or need for occasional radical innovations 

and restructuring, though these are typically 

followed by more normal periods of 

renewed incremental improvement. What 

the manufacturing engineering analogy 

offers us, however, is a vision of the way we 

can work in piecemeal ways to improve the 

NRIS, without needing complete knowledge 

or falling into the trap of imagining that we 

can plan everything. 

99

100

Part 3 

COUNTRY CASES 

The following country case studies demonstrate 

the wide variety of initiatives that are 

developed to stimulate innovation for environmental 

sustainability, depending on different 

development paths and different 

institutional settings. They illustrate the 

importance of a broad range of (complementary) 

options in the achievement of the 

overall goal. 

The Danish study by Jesper Holm et al refers 

to evidence from sector specific case studies 

on development of new institutional practices 

and capacity building and the development 

of green products on the basis of the 

win-win philosophy of the predominant 

model of ‘ecological modernisation’. It was 

possible to start to integrate environmental 

and business development policies and to 

create an interactive ‘green’ responsiveness 

from business to new regulations and programmes 

starting from a company focussed 

policy view. The success of these reflective 

learning processes is based on the interfacing 

of different actors and the presence of a 

social component that enhances the change 

in business behaviour. 

The Dutch EET programme, presented by 

Corine van As and René Wismeyer, is an 

example of a mission-oriented approach, 

coordinating the efforts of three Ministries. 

It aims to stimulate breakthroughs in 

themes put forward by government to 

achieve sustainable development. The latest 

version stresses the transition to sustainability 

by systemic innovation as the distinctive 

objective of funding. 

In contrast, the new programme of stimulation 

of sustainable environmental technology 

development in Flanders, presented by Paul 

Zeeuwts, is an example of a bottom-up 

approach, linked to generic technology and 

innovation funding programmes. The aim is 

to stimulate a shift in innovation behaviour 

towards more environmental benefits in all 

projects, by selectivity and financial incentives. 

The paper by Hans-Guenther Schwarz gives 

an account of the Austrian programme on 

Technologies for Sustainable development, 

in particular the two thematic sub-programmes 

on the House and the Factory of 

Tomorrow. They are not limited to technology 

development but include the necessary 

complements of demonstration projects and 

knowledge dissemination. The integration 

of multiple goals and multiple types of innovation 

is a characteristic of the programme 

design. As in the other cases evaluation 

therefore is a rather complex issue. 

101 

3

102


GREEN INNOVATION 

BETWEEN CORPORATE REFLEXIVITY AND SECTOR INTEGRATED 

ENVIRONMENT & TECHNOLOGY POLICIES 

JESPER HOLM, 

OLE ERIK HANSEN 

AND 

BENT SØNDERGÅRD. 

Department of Technology, Environment and Social Studies, 

www.teksam.ruc.dk, Roskilde University 

Jesper Holm is an Associate Professor in 

Environmental regulation at the dep. of 

Environment, Technology and Social Studies, 

Roskilde University Denmark. He has done 

international research in the impacts on 

markets and business from environmental 

policy and political strategies. Secondly he 

has given focus to issues of democracy and 

the environment. 

103

Between corporate reflexivity and sector integrated 

environment & technology policies 

> 

1 

The dogma or story 

line of ecological 

modernisation is, in 

short, that current 

industrial societies 

may be guided, 

through the prevailing 

institutions of markets, 

politics and cultures, 

towards reconciliation 

with nature in 

a sum plus game 

(Hajer 1995, Gouldson 

& Murphy 1997). 

2 

The paper summarises 

the Danish results 

from an international 

two-year study of 

policy stimuli and 

environmental innovation 

The study 

Towards an 

Integration of 

Environmental and 

Ecology-oriented 

Technology Policy. 

Stimulus and 

Response in 

Environment Related 

Innovation Networks 

(ENV1NNO) was funded 

by the Target 

Socio-Economic 

Research Pro-gramme, 

EU. The project includes 

partners from 

Austria, Spain, 

Germany, the 

Netherlands and UK 


An ecological modernisation effort has characterised 

environmental policy in many northern 

European states since beginning of the 90’ties. 

Ecological modernisation is a current belief 

system among institutions in the industrialliberal 

world, holding that economic development 

and environmental protection are 

compatible, and focuses on the positive role 

of green corporate managers and consumers. 

Environmental corporate management 

regimes, cleaner technology programmes, 

and institutional greening may be 

seen as some of the core elements in this 

dominant discourse. 1 

The targeting of corporate behaviour in ecomodernistic 

environmental policy schemes 

has often stimulated and rested upon environmental 

policy co-ordination with business- 

and technology policy. Subsequent 

institutional transitions following these attempts 

have sought to promote eco-modernisation 

by enhancing and stabilising environmental 

communication on the market, in 

business chains etc. 

It is obvious that the environmental-technological 

outcomes from the corporate 

focus in eco-modernistic efforts differs 

from the perhaps more promising outcomes 

from the radical belief systems such 

as structural change, industrial transition or 

strategies of dematerialization like Factor 

10. But eco-modernistic market, institutional 

and management focus is still predominant, 

and the general political focal 

point of the sovereignty of the single firm 

is ever growing. Thus, it is important to 

recognise the actual outcomes and restrictions 

from these company-focused efforts, 

and to address what may be learned 

regarding policy options. 

> 

The paper takes up the issue by looking at 

failures and successes in a rather advanced 

eco-modernistic policy milieu - Danish experiences 

at policy (macro), sector-institutional 

(meso)and corporate (micro) levels. Thus it 

might serve as an extreme case exploring 

options and thresholds for green innovations 

by eco-modernistic approaches, in 

countries with a liberal democracy, polycentric 

network culture, a fair amount of 

SME´s and a mixed economy. At the policy 

level we have analysed how the macro-policy 

frameworks conditions- co-ordination 

and discursive integration between environmental 

and business/technology policy in 

general and related to selected manufacturing 

sectors. At a sector level we have 

analysed the institutional milieu for industries, 

environmental competencies, perceptions 

and communication within R&D units, 

trade organisations, municipal administrations 

etc. Lastly, at the corporate level we 

have studied the interactive responses to the 

policy and institutional outputs among 

selected industries embedded in various networks. 

Attention has been paid to identification 

of the institutional development, 

business responses and shifts in environmental 

perceptions and performances. 2. 

The paper examines how the general sector-integrated 

environmental policy and 

the targeted policy programmes have 

installed specific institutional and discursive 

formations within four industries: textile, 

electronics, food and construction. In a 

dynamic perspective we look at how systemic 

competencies and environmental 

perceptions are reproduced and produced 

within the industries, resulting in a selective 

milieu that favours specific paths of 

technology development. These processes 

and environmental performance are studied 

using a bottom-up approach exploring the 

interpretative responses to environmental 

and technological policy programmes 

among firms and network stakeholders. 

2. THE GENERAL POLICY FRAMEWORK 

FOR CORPORATE ENVIRONMENTAL 

INNOVATION – THE INTERLINKAGE 

BETWEEN ENVIRONMENTAL POLICY 

(EP) AND TECHNOLOGY POLICY (TP) 

If we take a rough look at how the interlinkage 

or crossover between Environmental 

Policy (EP) and Technology Policy (TP) has 

developed in Denmark we may conclude the 

following stages: 

104




1972-84:Consensus oriented building of 

guiding environmental standards in close 

consultation with Danish industrial trades, 

balancing environmental requirements to 

available purification and dilution techniques. 

Subvention schemes were launched 

for the industries having difficulties in complying 

with purification demands. The period 

was characterised by no specific technological 

innovation focus and no specific EP focus 

in the emerging TP. An exemption though 

was a subsidy scheme from 1981 for investments 

in renewable energy technologies. 

1985-91:Technology pushes by strategic coorientation 

of TP and EP. Improving the technological 

capacity options among Danish subcontractors 

to public and private purification 

infrastructure, balanced by increasing 

demands to environmental protection. 

Strategic environmental modernisation by 

stressing the competitive advantages that 

might emerge from a Danish environmental 

front running. Research and Development 

(R&D) in TP programs in environmental oriented 

topics. R&D EP programs in cleaner 

technology and recycling related to processes 

and waste. A new risk- and environmental 

oriented concern in TP was launched with the 

1986 act on Environment and Genetic 

Engineering, which banded all experiments 

with genetically modified organisms (GMO) 

unless they were given a dispensation. 

1992-2000: An institutionalisation of 

technology orientation of EP in acts and 

orders occurred. Strategic systems export 

became a new orientation for the support of 

developing Danish environmental infrastructure, 

institutionalised in various aid 

programs under both ministries. New R&D 

programs were started in cleaner technology 

changes towards environmental management 

(EMAS), technology diffusion and 

design support for cleaner products. 

Enhanced green public procurement and 

market pull policies were given priority. 

Finally new TP and EP schemes for creating 

new green jobs have been initiated at the 

end of the period. There have been developed 

a general discursive environmental coorientation 

of TP, but only very few specific 

strategic initiatives have been launched. 

Enhanced administrative co-ordination since 

1998. Currently, June 2002 a new liberalconservative 

government has made a radical 

restructuring effort on environmental policy, 

including cuts in cleaner product and 

process subsidies, deregulation and 

enhanced voluntary policies. The results in 

the public sphere and on the market are not 

yet foreseeable but even more stress are put 

to the eco-modernistic belief in corporate 

self-determination and green markets. 

The integration between EP and TP has 

developed in an institutional framework 

with the following characteristics: 

• Denmark has up until end of 2001 a relatively 

strong Ministry of the Environment and 

Energy (MEE). The ministry has managed to 

establish an eco modernistic discourse as a 

part of the general policy of the government. 

Strategies and support schemes for 

cleaner technologies have become a part of 

the Ministry’s competence. The Ministry has 

established and has been involved in policy 

networks with branch organisations, private 

firms and other actors in order to influence 

the technology and business policies. 

Furthermore, the Ministry has had many programmes 

focusing on the development and 

diffusion of technology. Thereby the Ministry 

has been an important actor in the development 

of the technological infrastructure. 

• The Ministry of Business and Industry (MBI) 

has also in recent years become a strong 

Ministry and business related research and 

technology policy is a part of the Ministry’s 

competence area. Business policy and technology 

policies have been consensus oriented 

and branch organisations and private 

firms have played an important role 

in the formulation of new policies. 

• The technological infrastructure is an important 

result of the consensus-oriented business 

and technology policy. This infrastructure combines 

consultative work and business related 

technology research financed by the 

Government. A big part of the money used for 

integrated EP and TP efforts is canalised 

through the infrastructure or is used to develop 

the basic competencies in the local R&D units. 

105



• There is a vertical division of labour 

between the state level, the county level 

and the municipals. Counties and municipalities 

have a big part of the responsibility for 

the implementation of as well EP and TP. 

Therefore, there can be big regional variations 

according to the degree to which local 

authorities trie to make for example cleaner 

technologies an important part of the regulation. 

Likewise, there have been established 

regional Technological Information 

Centres in order to support the development, 

adaptation and diffusion of new 

technologies especially in SME’s. Also 

business policy institutions are established 

on the local or regional level in order 

to attract enterprises or in order to 

strengthen networking in the local business 

community. 

These institutional characteristics mean 

that there are different arenas for the 

interaction between TP and EP. At the 

governmental level both the MEE and the 

MBI have tried to integrate environment 

and technology from different perspectives. 

Both ministries have emphasised that it is 

not a hostile relation. At some occasions 

from the end-eighties the ministries have 

formulated a common strategy and they 

have administered programmes for the 

development of environmentally friendly 

technologies in collaboration. But most of 

the technology oriented EP activities have 

been administered by the MEE, whereas the 

environment oriented TP elements has been 

administered by the MBI. 

It is important to notice that EP is a 

rather strong policy area while TP must 

be seen as a weak area. In recent years, 

the development of new smart technologies 

and new materials has been seen 

as the cornerstone in the development 

of a more environmentally friendly and 

sustainable production. On the other, technology 

policy is traditionally a weak area 

and environmental concern is normally not 

understood, as an important element of TP. 

The technological capability to enhance 

competitiveness has been the important 

part of TP. 

> 

3. CASE STUDY APPROACH 

The object of our case studies has been to 

examine how the transition of the environmental 

regulation have induced corporate 

reflection in the Danish bakery, textile finishing, 

electro plating and insulation industry. 

The aim has been to analyse the content of 

this transition, focusing on how new institutional 

structures and environmental perceptions 

have developed within the industry and 

how they have been interpreted and influenced 

the environmental activities at the 

manufacturing level. The analytic endeavour 

does have a normative motive - the process is 

observed and analysed with the aim to assess 

and develop a better understanding of which 

policy options we can have in relation to the 

development of environmentally more 

friendly production schemes 

The interface between policy programs and 

the enterprises in the industry is at the centre 

of the analysis. On the one hand the 

cases raises the question, how policy programs 

are communicated to the enterprises 

and how they may succeed to install ecological 

modernisation and innovative 

responses in the industry. On the other 

hand the cases raise the question, how 

these stimuli (changed institutional framework 

and communicated perceptions of 

environment) have been reflected and 

interpreted by the enterprises within their 

strategic scheme The basic tenet of the 

case studies is that the study of the 

processes of interaction, studied as institutional 

and enterprise practices, are vital to 

develop the means of profound environmental 

transition of the industrial production. 

To obtain this, we have integrated two 

lines of analysis; an analysis of the institutional 

development concerning the environmental 

transition of the industry sectors, 

and an analysis of the processes at the 

enterprises reflecting the changes in their 

milieu. The case studies have focused on a 

meso level within the institutional regulatory 

complex of Danish textile, insulation, 

bakery and electro plating industries: how 

the policy programs have supported specific 

institutional and discursive formations, and (in 

a dynamic perspective) how systemic compe- 

106




> 

tencies and environmental perception have 

been produced. Here a profound number of 

varying constellations of R&D institutions, 

trade associations, labour unions, green 

NGO´s etc. has been examined in order to 

identify main drivers and their success in institutionalising 

their preferences in a certain eco 

modernistic development.The aim has been 

to specify the resulting selective milieu that 

favours specific paths of technology development 

and environmental actions. At a micro 

level the impacts of this conditioning milieu is 

studied from a bottom-up perspective: 

Exploring how firms in networks reflect upon, 

interpret and perform technology in response 

to the environmental policy programmes. 

4. CASE SELECTIONS 

The point of departure for the selection of 

case studies was: 

1. There had to be inputs from the policy side, 

and there had to be some form of crossover 

between Environmental Policy (EP) and 

Environmental Technology Policy (ETP). 

2. In four cases an environmentally innovative 

output (technological, network) had to be at 

hand, whereas one case had to be a failure. 

3. The focus was on the end users, not on the 

inventors of new process-technology for 

purchasing. 

Although it is not possible to make statistical 

generalisations from five case studies, we 

decided that in order to cover different 

aspects of EP/ETP crossover it was important 

to have a large degree of variation in our 

cases on the input side, the output side and 

the mediation. Furthermore, it was important 

to select cases from branches of industry 

that are important part of the production 

and consumption structure or have a 

considerable amount of environmental 

impact. The criteria resulted in five case 

studies in four branches of industry (construction, 

food, textile and electronics). We 

chose to make some kind of direct comparison 

of a failure innovation and a success 

innovation in the textile industry. 

The textile industry is relatively important in 

Denmark, there has been several integrated 

EP/ETP programmes addressing the industry, 

and the industry’s environmental impact is 

well known, and there is a national user-producer 

network linked to international production 

facilities. Our aim was to examine 

Figure 1 > 

Case Study Characteristics 

COMPANY 

RETAIL CHAIN, BAKERY 

Kvickly/Superbrugsen 

All products in bakeries turned 

into organic 

Eco-labelling, R&D subvention 

TEXTILE: 

Soedahl Design 

Heavy metal- and toxic substance 

free colouring 

R&D programs in EP, rules, 

TP competence schemes 

Trade discussions, local 

enforcement of EP, NGO 

DTI, textile R&D units, 

regulatory 

TEXTILE: 

Danish Colour Design 

New no-low VOC containing 

paints 

R&D subvention programs, 

working health regulation 

Pressure from unions, local 

enforcement of rules 

DTI, business, trade organisation, 

regulatory 

Change 

EP/ETP policies 

Mediation 

Public debates, GMO discourses 

Networks 

Supplier networks, green NGO´s 

Impacts 

COMPANY 

Changed not only products of 

72 shops but the whole market 

ELECTROPLATING 

Printline 

Lowered emittants, 

new market profiles 

New network constructions 

CONSTRUCTION: 

Dansk Naturisolering 

Lower emission of VOC´s, 

improved working conditions, 

new technology path, cost gains, 

new standards 

Change 

EP/ETP policies 

Mediation 

Networks 

Impacts 

Finishing: Closed loop, ion-exchange 

R&D programs in EP, rules, regional ambient 

water quality plans, voluntary regulation 

Local enforcement, trade consultancy 

R&D from TP policies, trade, consultancies 

Reduced emissions of heavy metals from waste 

water, reduced water consumption 

Environmental monitoring (product development) 

Innovator TP programs, subvention of R&D 

Network (Center contracts) 

Co-operation, marked driven. Pressure from unions 

Research institutions, unions 

Development of environmental technologies 

107



how the same regulatory and institutional 

frame produced different innovative outputs. 

Likewise, the construction industry has 

a national network, very important environmental 

impacts and there have been different 

kinds of environmental oriented technology 

programmes addressing the industry. 

Furthermore, the industry is characterised 

with an institutional framework making 

shifts in technological paths very difficult. An 

important aspect of the environmentally oriented 

development of the industry is the 

social movement for ecological housing. The 

food industry is also characterised by a 

national network and the importance of a 

social movement for organic food. In this 

case, however, the eco-labelling scheme has 

framed the industry and supported the production 

of organic food. The fourth industry, 

electronics, is in many aspects different from 

the others. There have not been specific programmes 

targeting the industry, and the 

Danish production is linked to an international 

production chain setting the standards 

for as well product and production technologies. 

Therefore, this case is very important 

because it demonstrate some of the linkages 

between national policies focusing on the 

implementation of new technology paths in 

industries fully dependent of their ability to 

comply with international standards. 

The selection of companies, the focal innovations, 

the relevant ETP/EP policies, the mediators, 

the network and the impact analysed 

is summarised in figure 1. We presumed that 

the way policies may cause impacts on corporate 

behaviour are varying a lot, depending 

on both the input side, the side of the 

respondents and their form of mediation. 

Whereas others may focus on the type of 

firms or polices as dependent variables we 

tended to focus on the institutional, discursive 

and regulatory dynamics in the relationships 

between market and state. Therefore, 

we focused on the following points in the 

five case studies: 

1. How was the crossover between EP/ETP 

communicated Institutionally (service, 

R&D), discursively, rules, rules enforcement, 

funds and aid schemes. 

> 

2. Within the case: which were the main 

institutions, organisations and networks 

that used and interpreted the EP/ETP communication 

3. How were the rules and the EP/ETP communication 

interpreted by case-companies 

and their regulatory, R&D, and business 

networks 

4. How were these inputs influencing the 

technological innovations that were at 

stake within the companies 

5. Did the EP/ETP inputs have any innovative 

impacts on the constellation of actors 

within the firm’s networks 

6. What kind of innovations in relation to 

the ambitions of the EP/ETP policies, mediation 

forms and corporate conditions 

were the result of the innovation process. 

In our opinion this focusing of the analyses 

has made it possible to come up with some 

results of general importance according the 

communication and the institutionalisation 

of the EP and the ETP. Additionally the consequences 

for the specific configuration of 

networks surrounding the innovations, the 

radicality of the innovations from a technological 

and environmental point of view, 

and the grade of success and failure. 

5. EP-TP POLICY AND REGULATION AND 

CORPORATE RESPONSES: FINDINGS 

In our perspective the institutional learning 

process within the Danish industries have 

shown a green responsiveness. This can be 

seen as an interplay between a national ecological 

modernisation strategy, a construction 

of policy programmes and institutions 

directed towards the industry and enterprises 

responding to the programmes by their own 

strategic interpretation of the programmes. 

This interaction has established a dynamic 

learning process with the construction of new 

actors, the establishing of a new technological 

selection environment, and the building of 

new competencies within the enterprises and 

their network. We have identified two types 

of industrial responsiveness: the environmen- 

108




tal learning of integrating environmental concern 

into management, and secondly the construction 

of new ecological or green products 

as a respond to general eco-discourses and 

ecological movements 

The incremental learners 

Within the environmental regulation of the 

textile industry a wastewater track of development 

has been dominating, giving fewer 

impulses to other paths of development such 

as more radical substitution schemes and optimising 

material flow (reducing the materials 

suspended in waste water). Thus it was here 

we found our failure case, were the technological 

trajectory of a certain membrane technology 

filter was pushed as the solution. 

The targeted cleaner technology program 

within the textile industry, alternative insulation 

and Electro plating has produced 

knowledge resources concerning cleaner 

technologies. Technological options, that 

offer substantial reduction in the consumption 

of water, chemicals, heavy metals and 

energy and reduces the discharge of suspended 

material in the waste water, is documented 

and is accessible in the environmental 

knowledge network. 

The implementation of the cleaner technology 

program within the industries has been 

based on a strategy program brought forward 

by the stakeholders in the industry and 

the related network of knowledge institutions. 

This has focussed and strengthened 

the program, but it has also produced a closure 

in terms of actors involved and in terms 

of the environmental perception in the program 

and industry. The program has been 

technology oriented and focussed on wastewater, 

spills and single substances. However, 

the program has also conditioned a learning 

Figure 2 > 

EP stage characteristics and various impacts 

EP stages 

Permits 

1986- 

Cleaner Technology 

1991- 

Environmental 

Management 

systems 1995- 

Product oriented 

1997- 

Construction 

of actors 

Command and 

control oriented 

regulators, 

reactive firms 

Closed corporate network 

of actors 

(Knowledge institutions, 

proactive branch 

organisation, proactive 

firms focusing on winwin 

technologies) 

Proactive firms with 

enhanced capacity to 

control the process 

operations 

Proactive firms actively 

trying to construct new 

markets for environmentally 

friendly products. 

Corporate institution 

focusing on eco-labelling, 

new stakeholders from for 

example retail industries 

Technology 

paths 

Reduction of over 

consumption, 

reuse of water, 

cleaning 

technologies 

Technology push – 

development of - 

winwin technologies 

combining cost reduction 

and environmental 

improvements 

Environmental management 

systems, 

chain management 

and documentation 

of environmental 

performance 

Product and life cycle 

orientation 

Competencies 

in knowledge 

network 

Knowledge of 

wastewater and 

hazardous chemicals 

in environmental 

agencies 

Construction of new 

technologies in knowledge 

institutions, 

capacity building in 

business network 

Supply chain management 

in business 

network 

Chain and market 

oriented, development 

of instruments for environmental 

documentation 

Competencies 

in firm 

Water reduction 

and reuse of 

water 

Lack of ownership to 

cleaner technologies, 

no focus on competence 

building 

Supply management 

system 

eco-labelling, ability to 

communicate and document 

environmental performance 

Strategic 

interpretation 

in firms 

Environmental 

action as cost 

Win-win perspective 

focusing on cost reduction, 

reactive response 

to public demands 

Competitive 

advantages related 

to environmental 

documentation and 

‘green cotton’ 

concept 

Win-win perspective 

focusing on new business 

opportunities environmentally 

friendly products 

109



process. New schemes on eco-management, 

eco-design and product development are 

currently being taken up within the industries. 

The environmental regulation and the 

resources produced in the cleaner technology 

programs highly influenced the way the 

enterprises have dealt with their environmental 

problems. It has influenced how they 

defined their problem, and it has influenced 

the solutions that were considered, it has 

influenced their decision to implement or 

not and it has greatly influenced their final 

choice of technologies. 

The textile and electronics industries have 

been object to a number of targeted programs 

and regulatory initiatives aimed at 

supporting an environmental upgrading of 

the industry. Parts of these programs have 

directly addressed the task to spur and 

define an environmental innovative behaviour 

in the industry. Taken all together they 

have defined a changed playing field, which 

has influenced the pace and the direction of 

environmental innovations in the industry. 

The construction sector has kicked of later 

but resembles the development path in 

regulation and industry response. 

Consensus and close co-operation among 

the main actors in the textile, electronics 

and partly construction industry (the trade 

organisations, the knowledge institutions, 

DEPA, the municipalities and counties, and 

national R&D institutions) have characterised 

the programs and regulatory initiatives 

within the industries. The fact, that the 

trade organisations have adopted a strategy 

of restructuring of the branch, which has 

included an environmental modernisation as 

a strategic option, has paved the way to 

environmental initiatives in the industry. In 

general, the programs and initiatives have 

benefited of this consensus and co-operation. 

It has contributed to the consistency in 

the programs and initiatives and has 

enabled concerted action on the framework 

program of cleaner technology, introduction 

of environmental management, the elaboration 

criteria to eco-labelling, and the 

product-oriented program. 

The outcome of the programs and initiatives 

has been a high environmental awareness 

in the industry. Environmental considerations 

are now integrated in the search 

and decision processes of the enterprises. 

Throughout the process a group of frontrunners 

has participated actively – and 

have in this process developed their environmental 

perception, i.e. the problems 

addressed and the horizon of solution 

strategies. Adoption of technological solutions 

may have been limited, but they have 

participated in a collective learning 

process, in which they have moved from 

good housekeeping and substitution, to 

environmental management systems, to 

product oriented and to LCA and ecolabelling 

schemes. A change of the way 

enterprises reflects environment may show 

to be the main achievement of the program. 

Looking at the Danish textile finishing 

industry, the construction sector industry, 

and the electronics industry the environmental 

and technological policy programmes 

has led to a wide incremental 

introduction of cleaner technology. In this 

process a selective milieu of the industry has 

been produced: 

a) Within the Danish technology programme 

a stable constellation of consultants 

and research institutions have 

formed a interwoven net of competencies 

on textile dyeing processes, 

b) The industry and (local) environmental 

authorities have developed a high degree 

of consensus on environmental perception 

(e.g. regulatory score-system of the 

environmental load of dyes based on a 

common prioritisation). The industry and 

the produced constellation of actors have 

been bound together in a specific environmental 

perception. 

c) A progression in the systemic competencies 

were build up within this network. In 

a dynamic learning process the sector as a 

whole has gone through the stages of dissemination 

of information, development 

and diffusion of technology, introduction 

of environmental management and product 

oriented strategies. 

110




The construction of new ecological or 

green products 

Concerning the second type of industrial 

responsiveness, the construction of new ecological 

or green products, we have found a 

radical approach inventing or co-shaping 

new green markets. Within the construction 

and in the bakery sector we have traced a 

number of bottom-up initiatives among 

green lifestyle promoters, green consumers 

and political movements that have formed 

the discursive and pragmatically background 

for new green entrepreneurs. Thus it were 

these groups that formed the front running 

organic movement in the 1970´s that pawed 

the way for an incremental institutionalisation 

during the 1980´s – the organic 

labelling scheme that confirmed and normalised 

organic food. The same type of 

movement occurred in the 80´s and 90´s 

within the housing sector, and a number of 

ecological housings and constructions 

pupped up during 1982-1997, before an 

institutional response took place. 

The Ministry of Food has been innovative in 

taking over the various private labels on the 

marginal green markets in order to establish 

a state guaranteed green market. By time 

the health, drinking water, and environmental 

regulation from the Ministry of Food and 

the Ministry of Environment and Energy 

have benefited from this market based 

recognition of organic food. They have been 

able to use the precautionary principle by 

ruling and giving subventions for the 

enlargement of organic manufactured fields 

in Denmark. The normalisation and stabilisation 

policies have together with solid supports 

for organic agriculture formed the 

business model opportunity for the retail 

chain Kvickly to establish their 100% organic 

bakeries. Thus, the transition could take 

place as organic flour and sugar was on the 

market, whereas Kvickly could pass impetuses 

for subcontractors on margarine, 

marzipan, chocolate etc. to make an organic 

switch. Secondly the knowledge networks 

and R&D institutions within the food sector 

provided the necessary primary insights to 

make the organic transition. The market stabilisation 

of organic foods have discoursively 

institutionalised an alternative to all negative 

risk connotations in the food sector: 

GMO´s, straw shortens, penicillin, pollution 

of ground water etc. 

The construction sector has gone through a 

parallel development, but the Ministry of 

City & Housing has been very late responsive 

to these development paths, and has been 

hesitant to leave a co-dependency upon the 

major construction, housing and construction 

material industries. Regulatory it has 

been the Ministry of Environment and the 

Ministry of Food that has been issuing subvention 

schemes for radical innovations in 

construction technologies and materials. In 

fact these extra-construction/housing policy 

sectors have been able to avoid the path 

dependency and technological trajectory 

that the Ministry of City and Housing have 

been locked in. Of special interest is that 

labour unions and occupational health 

interests have been major forces in pushing 

forward a need for finding technological 

alternatives with both an occupational 

health and an environmental benefit. By the 

gradual articulation of a joint and coherent 

eco-health discourse within housing and 

construction a stabile front has been met by 

a responsive policy that has issued R&D, subvention 

schemes and eco-standards for ecological 

housing and construction materials. 

New green entrepreneurs have on this background 

of state supported R&D been able to 

penetrate the regulative, economic and 

technological barriers to alternative, ecological 

housing technology. The result has been 

new materials and knowledge that have 

kicked off a number of local ecological housing 

projects. 

The latter type of policy responsiveness to 

green social movements and discourses has 

been important for the second wave of entrepreneurs 

who have dared to form new 

technological innovations and profiles. 

The most obvious strength of Danish EP has 

been that it has developed in quite a responsive 

way, enabling the MEE to take up new 

environmental problems and initiatives from 

research results, public debates and business 

technology options at very short warning. 

111



The consensus policy style is, according to 

general evaluation studies of Danish EP 

(Andersen 1995, 1997, Christiansen 1996, 

Holm 1999, Mortensen 1999, Nielsen 1997, 

Wallace 1995), both strength and a weakness. 

A strength, because industrial trades 

are open to communicating on technological 

options and to participating in finding 

environmental protection measures they 

may benefit from. There is a practical capacity 

for co-ordinating consulting and environmental 

monitoring with R&D initiatives in 

cleaner substitutes or technologies, which 

have established effective feedback mechanisms 

between the DEPA, the political system, 

research and networking. The weakness 

is that the minister and DEPA very often 

have hesitated to enforce legal and binding 

rules that industry strongly opposes. A 

weakness is also the risks of falling into technological 

trajectories, which fit to what 

industrial suppliers of environmental equipment 

offer. 

General findings 

> 

The first finding of general importance 

relates to the often experienced ambiguity 

of EP and TP communication and institutionalisation. 

The industry cases in mind here are 

from textile industry where the local authorities 

have enforced demands upon the firms 

that have forwarded traditional end-of-pipe 

technologies, on the expense of cleaner 

technology options for reducing hazardous 

chemicals. The other case is the construction 

sector, where mineral wool insulation has 

been favoured by the public rules on construction, 

by the R&D institutions and by 

occupational health regulation. But at the 

same time new programs for alternative 

insulation from other ministries have made 

a brake trough. One may consider it as an 

advantage that several technology paths are 

being stimulated, but on the other hand it 

may leave firms confused when they try to 

be innovative adjusting to public demands. 

The best case in mind here is our electronics 

case where a perceived market advantage in 

converting to a non-led silver technology 

has withered away from lack of public 

enforcement of a phasing out heavy metal 

strategy. The same picture, though somewhat 

successful, may be the case in our 

Danish Colour Design case, where the weak 

market pull in non-toxic dying partly is due 

to hesitating public regulation on ecofriendly 

clothing. 

The second general conclusion is that communication 

and institutionalisation vary 

a lot. Among successful environmental cases 

there is a constructive co-appearance 

between communicated environmental perceptions, 

institutional configuration, prohibitions, 

subventions etc. But this does not mean 

that there has to be a pre-planned co-ordination 

of policy inputs, a genuine EP and ETP 

fusion. What is of importance is that there is 

some kind of shared discursive horizon which 

the diverse inputs relate to – e.g. an eco-modernistic 

community perception. Plus the fact 

that the co-existence of the various incitements 

and rules actually do stimulate the 

entrepreneur into innovative processes. 

The third general conclusion is that the different 

policy-inputs in all cases have stimulated 

an open interpretation of scale and 

scope of what is to be innovated. From here 

it is the institutional embeddedness of the 

focal firm in value chains, user-producer networks 

and relations to R&D institutions that 

determines the way inputs are transformed 

into innovative processes. Thus, it is vital 

what the network of the focal firm may 

offer in innovative solutions. 

Fourthly, the state driven environmental 

innovation by R&D schemes, purchasing 

policies, labelling etc. is of outmost importance 

for the establishment of a new technological 

path, when a deeply institutionalised 

technological path controlled by few 

“dinosaurs” is the case. The enhancement of 

new insulation materials and methodologies 

depended upon basic research stimuli and 

policy stimuli for new entrepreneurs. 

6. CONCLUDING ON ECOLOGICAL 

MODERNISATION POLICES 

AND STUDIES 

The aim of the paper was to analyse the 

dynamic interplay between: 

112




1. The establishment of a societal discourse 

on ecological modernisation 

2. The construction of policies and policy 

integration between the Ministry of 

Business and Industry and the MEE in order 

to facilitate ecological modernisation 

3. The set up of new institutions in some 

important branches of industry 

4. The practices of firms in order to carry out 

environmentally sound innovations through 

networking, interpreting policis and using 

institutional features. 

In relation to the discourse of ecological 

modernisation we note that the perspective 

of win-win has been established as the way 

to understand the interplay of environment 

and economy. Furthermore we note that this 

discourse has led to change of policies and 

to the establishment of new policies. As well 

in the MBI as in the MEE it was reflected 

how to set the agenda in relation to ecological 

modernisation. From the perspective of 

the MBI ‘Environment’ has been seen as an 

element in a strategy to secure international 

competitiveness for Danish firms and the 

national economy, thus the market is seen as 

the prime ‘driver’. On the other hand, from 

the perspective of the MEE there has been a 

focus on environmental capacity building in 

order to support a transition of the environmental 

behaviour of the companies. These 

two different focuses in relation to ecological 

modernisation result in significant differences 

in relation to the institution building 

and the constellations of actors. 

On the other hand, a very important conclusion 

of the studies is that institution building 

in order to support ecological modernisation 

has to be understood as a reflective 

learning process. In a top-town perspective 

we note that policies have resulted in a long 

range of institutional transitions in the studied 

sectors. The programmes have resulted 

in capacity building and new ways to support 

innovative conduct. In relation to this 

process there have been established learning 

processes based on as well internal as 

external interpretations of strengths and 

weaknesses on the existing configurations. 

This conclusion has to be seen in relation to 

the marked differences concerning as well 

the form of institutional development as the 

characteristics of the involved actors. 

A common horizon for the four studied sectors 

is that there have been developed a 

dominant technology trajectory in connection 

with the sector specific capacity building, 

establishing of standards and so on. This 

trajectory has cognitively framed interpretations 

among the stakeholders in the sector 

whereby a certain choice of technology and 

environmental performance of the companies 

have been fostered. Another common 

aspect is that the development of new technologies 

is seen as the way to reduce environmental 

consequences of the production 

or related to the product. Then there are distinct 

differences between the sectors that 

result in substantially different dynamics 

and thereby give some important insights 

concerning how the discourse of environmental 

modernisation’s results in the 

change of industrial practices. 

An important starting point for policy considerations 

is the characterisation of the technological 

change, i.e. whether it represents a 

radical break with the technology path in the 

sector. This concerns as well the degree of 

change in relation to the substituted technology 

as it concerns how the innovation fits into 

the existing technological system. The not to 

surprising conclusion is that changes not representing 

a radical break with existing trajectories 

are easier to realise because of a strategic 

selectivity in the technological system. A 

somewhat less trivial conclusion from the 

study of the most radical innovation, the transition 

to organic bakery, is that the radical 

innovation is preconditioned by an interaction 

of a range of simultaneous processes. A 

dominant company with the capacity to construct 

its own markets and a central role in the 

value chain is staging the transition process in 

combination with a sector specific capacity 

building and institutionalisation process in 

relation to organic food. Furthermore, there is 

a support from other social actors especially 

the consumers. On the other hand, companies 

that lacks these preconditions have severe difficulties 

to stage radical transitions in spite of 

environmentally related capacity building and 

demands from regulation. 

113



This could be read as if market is the most 

important driver, but the case studies show 

that the existence of a market is a necessary 

but not sufficient condition. Furthermore, it 

would be wrong to consider the market an 

external variable. The cases show that the 

markets are constructed in a social process. 

For example in electronics it was difficult to 

transform an incremental innovation to a 

commercial success because regulation could 

not deliver the promised market for lead 

free circuits.In the textile industry the innovation 

was realised in an interplay with a 

niche market for ‘green cotton’ established 

as an element of a business strategy, while 

the failed innovation did not succeed 

because it was in no way related to the business 

strategy of the firm. Organic bread was 

an extensive construction of a market based 

on an established market for organic food. 

And lastly, the case in the construction 

industry is an example of an innovation that 

is integrated in a niche market for ecological 

building supported by a social movement. 

The market as a necessary but not sufficient 

condition is also illustrated by the building 

and textile cases. In both cases the innovation 

was established in an interaction with 

the customer or the consumer. In the first 

example it is a business to business relation 

to a company with a green market niche. In 

the second case it is a social movement. The 

first innovation is from a technological point 

of view not especially radical. Nevertheless, 

there is a very limited diffusion of the innovation. 

Although the second innovation 

from a technical point of view very easily can 

substitute an existing product, it is slightly 

more radical because it confronts established 

standards in the industry. Therefore, 

it is situated as a niche product. Thus it can 

be concluded that the diffusion of radical 

innovations and more radical breaks with 

existing trajectories, presupposes: an institutional 

learning process, support of green 

markets, capacity building among a web of 

R&D, political and entrepreneurial networks 

for enacting new technology paths. 

114




> 

REFERENCES 

Beck, U, Giddens, A. & Lash, S. (1994), 

Reflexive Modernisation – Politics, Tradition 

and Aesthetics in the Modern Social Order, 

Polity Press, Cambridge. 

Beck, U. (1994), The reinvention of politics: 

Towards a theory of reflexive modernisation, 

(in Beck, Giddens, Lash 1994) 

Berger, P.L. and Luckmann, T (1967), The 

Social Construction of Reality: A Treatise in 

the Sociology of Knowledge, Doubleday, 

London. 

Buttel, F. H. (2000), Ecological modernisation 

as social theory, Geoforum, 31 (1), 57 – 66. 

v. Dijken, K et.al. (1999), Adoption of environmental 

innovations – the dynamics of 

innovation as interplay between business 

competence, environmental orientation and 

network involvement. Kluwer, Dordrecht 

Giddens, A. (1994), Risk, trust, reflexivity, (in 

Beck, Giddens, Lash 1994). 

Gouldson, A. & J. Murphy 1997: Ecological 

Modernisation: Restructuring Industrial 

Economies. In M. Jacobs (ed.) 1997: 

Greening the Millennium. The new politics 

of the Environment. Blackwell Publishers. 

pp. 74-86 

Hajer, M. A. (1995), The Politics of 

Environmental Discourse: Ecological Modernisation 

and the Policy Process, Clarendon 

Press, Oxford. 

Hansen, O.H., Kerndrup, S and Søndergård B. 

(1998), Cleaner technology in an innovative 

perspective (in Danish), in Holm, J et al. 

Miljøregulering (environmental regulation), 

Tek-Sam forlaget, Roskilde University Centre. 

Hansen, O. E., Holm, J. and Søndergård, B. 

(1999) “Technological Innovation and Environmental 

Policy in Denmark - On technology oriented 

environmental policy and environmental 

oriented technology policy”, October 1999. 

Working paper in The ENVINNO- project. 

(Forthcoming on Kluwer, 2000.) 

Hansen, O. E., Søndergård, B. and Meredith, 

S., (1998), Environmental innovations in 

Small and Medium Enterprises, Paper to the 

European environmental conference: 

Advances in European Environmental Policy, 

14-15 Sept. 1998, London School of 

Economics. 


(2000a), Targeted regulation and program: 

environmental orientated innovations in the 

Danish Textile Industry. Case: Membrane filter 

technology at Södahl Design (36 p), 

Working paper in The ENVINNO- project, 

February 2000 (unpublished). 


(2000b), Targeted regulation and program: 

environmental orientated innovations in the 

Danish Textile Industry. Case: Colour savings, 

re-use and heavy metal elimination in textile 

printing: Adoption of cleaner technology at 

Danish Colour Design (27 p), Working paper 

in The ENVINNO- project, April 2000, 

(unpublished). 

Hansen, O. E. Jesper Holm, Bent Søndergård: 

Network Driven Environmental Innovation 

within the Electronic Industry. Case: Leadfree 

Surface Finish on Printed Circuit Boards. 

Adoption of Silver Plating Technology at 

Printline A/S Department of technology, 

Environment and Social Studies, Roskilde 

University, Working paper in The ENVINNOproject, 


Hansen, O. E.Jesper Holm, Bent Søndergård 

ETP Support of Entrepreneurs and Labor 

Unions networking for an eco-work friendly 

alternative to a monopolized trajectory. -the 

innovative construction of crops grown insulation 

slabs in Denmark, Working paper in 

The ENVINNO- project, February 2000 

(unpublished). 

Hansen, O. E.Jesper Holm, Bent Søndergård 

Technological R&D in Organic Agriculture, 

Subvention for Markets and Normalisation 

through Environmental Policy and Eco-labeling 

The Retail Chain Kvickly - first mover on 

full organic transformation of bakeries, 

Working paper in The ENVINNO- project, 


115



Hukkinen, J. (1999), Institutions of Environmental 

Management. Constructing Mental 

Models and Sustainability, Routledge, London 

Johnson, B. (1992), Institutional Learning 

(p23-44) in Lundall, B-Å, National Systems of 

Innovation - Towards a Theory of Innovation 

and Interactive Learning, Pinter, London. 

Lash, S. (1994), Expertsystems or situated 

interpretation Culture and institutions in 

disorganised capitalism, (in Beck, Giddens, 

Lash 1994) 

Miljøstyrelsen (1996), En styrket produktorienteret 

miljøindsats, en debatoplæg 

(Strenthenedproduct-oriented environmetally 

effort – discussion paper). 

Miljøstyrelsen (1999a), Program for Renere 

produkter m.v. 1999 – 2002 (Program on 

cleaner products), (February 1999). 

Miljøstyrelsen (2000), Prioriteringsplan for 

program for renere produkter m.v. 2000 

(Priority plan for the program of cleaner 

products), (January 2000). 

Mol, A. (1995), The Refinement of Production- 

Ecological Modernisation Theory and 

the Chemical Industry, International Books, 

Utrecht. 

Mol, A. (2000) The environmental movement 

in an era of ecological modernisation, 

Geoforum, 31 (1), 45-56. 

Murphy, J. (2000), Editorial – Ecological modernisation, 

Geoforum, 31 (1), 1-8. 

116


SUSTAINABILITY RESEARCH 

IN AUSTRIA 

THE AUSTRIAN PROGRAM 

ON TECHNOLOGIES FOR SUSTAINABLE DEVELOPMENT 

HANS-GÜNTHER SCHWARZ 

Department of Technology, Environment and Social Studies, 

www.teksam.ruc.dk, Roskilde University 

Hans-Günther Schwarz is responsible for 

research on technologies for sustainable 

development & environmental technologies 

in Austria; Energy and Environmental 

Technologies, Austrian Federal Ministry of 

Transport, Innovation and Technology. 

He is currently building up and managing the 

sub programme “Factory of Tomorrow” in the 

framework of the Austrian Programme on 

Technologies for Sustainable Development. 

117

Sustainability research in Austria 

1. OVERVIEW 

The Austrian Program on Technologies for 

Sustainable Development is a five-year 

research and technology program. It initiates 

and supports trend setting research and 

development projects and the implementation 

of exemplary pilot projects. 

The program pursues clearly defined 

emphases, selects projects by means of tendering 

procedures and is characterized by 

networking between individual research 

projects and by accompanying project management. 

Currently two subprograms are in 

operation: 

• Subprogram “Building of Tomorrow” 

• Subprogram “Factory of Tomorrow” 

A third subprogram entitled “Energy Systems 

of Tomorrow” will start in late 2002. 

Project proposals are currently accepted in 

German language only. 

“Building of Tomorrow” refers to residential 

and office buildings that feature the following 

improvements as compared to the 

present practice in Austria: 

• Improved energy efficiency over the whole 

life cycle 

• Pronounced use of renewable sources of 

energy, in particular solar energy 

• Increased use of renewable raw materials 

and efficient use thereof 

• Increased consideration of service and use 

aspects for the benefit of users of residential 

and office buildings 

• Costs comparable to those of conventional 

building designs 

The subprogram “Factory of Tomorrow” 

addresses the trade and industry as well as 

service enterprises that produce and provide 

products of tomorrow using materials of 

tomorrow to meet tomorrow’s needs. The 

following aspects have to be taken into consideration: 

• Aiming at zero-waste and zero-emission 

technologies and methods of production 

• Increased use of renewable raw materials 

for materials and products 

• Increased use of renewable sources of 

energy in the production process and in 

the enterprise as a whole 

• Development of new partnerships and cooperations 

as well as in-house models 

for further training and participation of 

employees in order to achieve these objectives. 

Other subprograms are in preparation. 

The Austrian Program on Technologies for 

Sustainable Development has been developed 

at the Federal Ministry of Transport, 

Innovation and Technology (BMVIT) in cooperation 

with a network of experts and is 

being implemented with the assistance of 

the “Austrian Industrial Research Promotion 

Fund”. Individual subprograms are being 

attended by “umbrella-managements”. 


Only an economy based on the principles of 

sustainability will be able to secure our prosperity 

and quality of life in the long run. 

This, however, requires a radical reduction in 

the consumption of resources, which, in 

turn, can be achieved only by a fundamental 

change in our way of life and our economy. 

In addition to an appropriate political 

framework and increased awareness in consumer 

behaviour the economy itself will be 

an important factor in sustainable development. 

The Austrian Program on Technologies 

for Sustainable Development aims 

at supporting the economy with future-oriented 

innovations and developments. 

It is a well known fact that an orientation 

towards sustainability not only contributes 

to a relief of the environment, it also opens 

up completely new opportunities for the 

economy. For, saving on energy and raw 

materials also means developing considerably 

smarter and more efficient solutions 

and products to meet our needs. 

3. OBJECTIVES OF THE PROGRAM 

• New opportunities for the economy 

• Economical use of natural resources 

• Consolidation of Austria’s position in the 

field of technology 

• Positive effects on the economy and on 

118



employment achieved through 

- Strengthening of R&D competence 

- Interdisciplinarity and networking 

- Diffusion and application of R&D Results 

4. WHY “SUSTAINABLE DEVELOPMENT” 

A linear continuation of the present development 

of industrialized nations would 

cause growing global inequality at the 

expense of future generations (e.g. greenhouse 

effect). Therefore, fundamental 

changes in our way of life and economic system 

are imperative. On an international 

scale, there is general agreement that the 

concept of sustainability is a constituent element 

of future-oriented research and development 

and that research will play a key 

role in this area. In addition to innovation in 

the field of technology, structural and social 

innovation will be of great importance. 

5. GUIDING PRINCIPLES 

OF SUSTAINABILITY 

In a preparatory phase, a set of guiding principles 

of Sustainability was selected by an 

expert group. The principles also serve as criteria 

for the international jury when evaluating 

project proposals for the different calls 

for tenders: 

• Orientation towards benefit and need 

• Efficient use of resources 

• Use of renewable resources 

• Multiple use and recycling 

• Flexibility and adaptability 

• Fault tolerance and risk precaution 

• Securing employment, income and quality 

of life 

6. WHAT ACTIVITIES WILL BE SUPPORTED 

BY THE PROGRAM 

• Generation of innovative approaches and 

project definitions 

• Activities focusing on fundamental research 

• Applied research and development 

• Networking and cooperation between 

individual projects 

• Support for implementation (promotion, 

trainings, etc.) 

• Pilot and demonstration projects 

7. SUBPROGRAM 

BUILDING OF TOMORROW 

What is the Building of Tomorrow 

The “Building of Tomorrow” makes use of 

the two most important developments in 

solar and energy efficient building: the passive 

house and the low energy solar building 

method. For the purposes of the “Building 

of Tomorrow” subprogram, these energy 

centred innovations are expanded to take in 

ecological, economical and social concerns 

(see graphic). 

The “Buildings of Tomorrow” are residential 

and office buildings, and differ from current 

building practice in Austria by fulfilling the 

following criteria: 

• higher energy efficiency throughout the 

whole life-cycle of the building 

• greater use of renewable energy sources, 

especially solar energy 

• greater use of sustainable raw materials, 

and efficient use of materials 

• increased consideration of user needs and 

services. 

However, the costs are comparable with 

conventional building methods 

Goal of the subprogram 

The subprogram’s goal is the development 

and market diffusion of components, prefabricated 

building parts and building methods 

which correspond to the above criteria 

and to the main principles of sustainable 

development. 

Combining all of these demands is very challenging. 

Conflicts of aims can arise which 

need somehow to be reconciled. On the 

other hand, when social, economic and ecological 

aims can be integrated, the chances 

of success for the concept are vast. The key 

to realising this goal lies in innovation - not 

only technological but social, technological 

and institutional innovation. 

119


User and service aspects 

Low energy solar house 

Renewable energy sources 

Passive house 

Energy efficiency 

Building 

of Tomorrow 

Ecological building 

materials and systems 

Renewable raw materials, 

building ecology 

Comparable costs 

It is precisely in the combination of all these 

criteria that the chance arises to make technological 

leaps which actually have a high 

market potential. 

After careful consideration of over-development 

of the countryside, land use and mobility 

demands, priority has been given to multiple 

dwellings, rather than single family homes. 

8. STRUCTURE OF THE SUBPROGRAM 

The “Building of Tomorrow” subprogram 

has a planned duration of five years. It comprises 

the following elements, which build 

on each other logically. 

• Generation, preparation and dissemination 

of know-how in order to support the 

technology development process in a way 

that focuses on the project’s aims 

• Concept-led technology and component 

development 

• Development of innovative building concepts 

for residential and office buildings 

• Setting up and evaluating demonstration 

projects 

• Market diffusion of the “Buildings of 

tomorrow” 

The current call for tenders concentrates on 

the first three elements of the subprogram, 

and considers especially the projects that 

have already been financed or supported by 

the programme. 

In subsequent calls for tenders, demonstration 

buildings will be set up and evaluated, 

although concept-led technology and component 

development will continue. In order 

Collection and dissemination 

of relevant knowledge 

and experiences 

Concept-led technology 

and component 

development 

Development 

of innovative 

building concepts 

Setting up and 

evaluating demonstration 

projects 

Disseminating the results, 

market diffusion of the 

“Building of Tomorrow” 

120



to focus future calls effectively, the 

“Building of Tomorrow” subprogram will 

draw strongly on the results of the call topic 

“innovative building concepts”. 

The close of the programme will involve 

activities to disseminate the results and to 

support the market diffusion of the 

“Building of Tomorrow”. 

9. SUBPROGRAM FACTORY OF TOMORROW 

What is the “Factory of Tomorrow” 

The “Factory of Tomorrow” aims at zerowaste, 

zero-emission production, and 

produces and provides products and services 

of tomorrow using materials of tomorrow to 

meet tomorrow’s needs. Thus, the “Factory 

of Tomorrow” will aspire to generate maximum 

benefit with minimum consumption of 

resources and minimum environmental 

impact. The subprogram “Factory of 

Tomorrow” focuses on innovative development 

in the following fields: 

• Sustainable technologies and innovations 

in production processes 

• Use of renewable raw materials 

• Products and services 

• In addition to technological topics, projects 

dealing with ecological, economic, 

social, and structural problems are invited 

within the scope of the subprogram. 

The program addresses 

• Companies 

• Researchers and research institutions 

• Consulting and service companies 

How are the calls for tenders organized 

Each year there is an invitation of tenders on 

a specific topic. The projects submitted will 

be evaluated by an international jury. More 

detailed information about the tenders can 

be obtained at www.fabrikderzukunft.at. 

“New partnerships”, cooperations and 

networks should be developed within the 

different fields of activity. Improved staff 

qualification and better quality of work 

through participative processes and the use 

of information and communication technologies 

are pivotal concerns in this context. 

In addition to technological aspects the 

objectives of sustainability also comprise 

economic and social issues. 

10. THE FIRST CALL FOR TENDERS 2000 

The subprogram ”Factory of Tomorrow“ was 

initiated in the year 2000. The first call for 

tenders was based on a rather broad definition 

of the range of topics covered by the 

subprogram. This stage resulted in 20 projects 

from various thematic fields, which received 

funds amounting to a total of some € 2.54 

million. More than 100 institutions participated 

as partners in these 20 projects. 

Information on the contents and objectives 

of the research projects currently under way 

as well as contact persons can be found at 

www.fabrikderzukunft.at. 

11. THE SECOND CALL FOR TENDERS 2002 

The second call for tenders in 2002 concentrates 

on topics in the fields “Material Use of 

Renewable Raw Materials” and “Technologies 

and Innovative Developments in Production 

Processes”. 

Targeted projects are the results of which 

can be further developed into concrete 

demonstration and pilot projects within the 

duration of the subprogram, and that may 

serve as model projects demonstrating the 

feasibility of a sustainable economy in 

practice. These so-called “lighthouses” 

should also serve as role models on an 

international scale. 

The third call for tenders aims to further pursue 

topics of the second call and advance them 

towards technology development projects. 

A special accompanying measure of the second 

call for tenders consists in a contest 

focusing on “Intelligent Applications for 

Renewable Raw Materials”; distinctions will 

be awarded to already realized, trend-setting 

“Factory of Tomorrow” projects in the field 

121


of “Renewable Raw Materials”, and the 

projects will also be presented to the public. 

12. EVALUATION CRITERIA OF THE 

INTERNATIONAL JURY 

In addition to the “Guiding Principles of 

Sustainability” the following criteria will be 

taken into account 

• Contribution to the overall goal 

• Scientific excellence 

• Technical novelty and level of innovation 

• Appropriateness, clarity and consistency of 

the method 

• Professional aptitude and managerial skill 

• Quality of the presentation of possible 

barriers to implementation 

• Integration of several thematic aspects of 

the subprogram 

For “General and Economy-Oriented Basic 

Research” 

• Quality of the exploitation strategy 

• Potential benefit 

For “Development of Technologies 

and Components” 

• Market potential 

• Cofinancing quota 

• Balanced and complementary consortium 

• Potential for implementation of planned 

cooperation 

• Overall contribution to Austria’s national 

economy 

Activities supported by the subprogram 

“Factory of Tomorrow” 

• Generation of innovative approaches and 

project definitions 

• Activities focusing on fundamental research 

• In-process research and actor-oriented issues 

• Feasibility concepts and comprehensive 

strategies 

• Applied research and development 

• Networking and cooperation between 

individual projects 

• Support for the implementation, coordination 

of projects, networking and dissemination 

activities 

• Pilot and demonstration projects 

Program management and sponsor 

Austrian Federal Ministry of Transport, 

Innovation, and Technology 

Division of Energy and Environmental 

Technologies 

Contact: Michael Paula (head of unit) 

Hans-Günther Schwarz 

(email: hans-günther.schwarz@bmvit.gv.at) 

13. SUPPORT AND INFORMATION 

Umbrella - Management Building of 

Tomorrow: 

Austrian Society for Environment and 

Technology (ÖGUT) 

Mrs. Manuela Schein 

Hollandstrafle 10/46 

A-1020 Vienna, Austria 

phone: +43 (0) 1 315 63 93 

E-mail: office@hausderzukunft.at 

Umbrella - Management Factory of Tomorrow: 

TRUST CONSULT 

Management Consultancy Ltd. 

Mrs. Birgit Reiss 

Lothringerstrasse 16 

A-1030 Vienna, Austria 

phone: +43 (0)1 712 17 07-313 

E-mail: office@fabrikderzukunft.at 

122


14. EXAMPLES 

Building of Tomorrow results from projects 

funded through the sub-program and from 

the two project competitions new building 

of tomorrow and old building / retrofit of 

tomorrow 

Apartment building Oelzbuendt (Dornbirn, 

Vorarlberg) 

Three-storied residential building constructed 

in wood designed as passive house 

Row houses in Batschuns (Vorarlberg) 

The first row houses in passive house standard 

in Austria 

Multi-family-villa constructed in wood 

(Feldkirch/Tosters, Vorarlberg) 

South-orientation, excellent insulation and 

glazing and a ventilation system with heat 

recovery result in the building’s excellent 

energy performance. 

123


15. “FACTORY OF TOMORROW” 

results from projects funded through the sub-program “green bio-refinery” 

fossil raw materials 

* crude oil 

* natural gas 

oil 

refinery 

products / commodities 

* fuels / combustibles 

* chemical base materials 

* basic materials (plastics) 

renewable raw materials 

* field crops 

* forest plants 

* biomass from waste 

* green biomass 

(grass, trefoil, lucerne,...) 

green 

bio-refinery 

products / commodities 

* fuels / combustibles (ethyl alcohol) 

* chemical base materials (organic acids) 

* basic materials (bioplast materials) 

* food-products (oils, starch products,...) 

ref.: Wachter, Mandl, Kromus, Narodoslawsky, Interrim Report Green Bio-Refinery, Vienna, 2002 

Introduction Hans-Günther Schwarz 

grass, trefoillucerne,... 

proteins, 

animalfeed 

dilactide, 

ethyllactate, 

amine lactate,... 

silo 

proteinseparation 

separation of 

lactic acid 

juice 

press 

cake 

fibre 

pulping 

heat, powder 

biogasplant 

insulation, gardening, construction materials (additive),... 

separation of 

special materials 

value added products 

(chlorophyl, flavouring substances) 

ref.: Wachter, Mandl, Kromus, Narodoslawsky, Interrim Report Green Bio-Refinery, Vienna, 2002 

124


STIMULATION OF 

SUSTAINABLE TECHNOLOGY 

DEVELOPMENT 

IN THE NETHERLANDS: 

THE EET PROGRAMME 

CORINE VAN AS AND RENÉ WISMEIJER 

Project officers at the E.E.T. programme office 

Corine van As (1971) holds a degree in 

Environmental Sciences of Wageningen 

Universiteit (Netherlands). For several years 

she worked in the field of metal industry 

and the environment. 

Since 1999 she has worked as a programme 

officer for the E.E.T. programme office in the 

field of renewable energy, and with special 

focus on improving the management of 

projects. 

René Wismeijer (1961) holds a degree in 

Environmental Chemistry from the Free 

University of Amsterdam. He worked for 

several years for the Dutch Ministry of the 

Environment in the field of soil protection. 

Since 1997 he works as a programme officer 

for the E.E.T. Programme Office in the field 

of renewable energy and programme monitoring 

and evaluation 

125

Stimulation of Sustainable Technology Development 

in the Netherlands: the E.E.T. Programme 


> 

> 


The Dutch EET programme (Economy, 

Ecology and Technology) supports strategic 

middle to long-term research and development. 

One of the main goals is a sustainable 

economic development. The market launch 

of the project results should be between 5 or 

20 years after the project-start. 

Three Ministries finance the EET programme: 

the Ministry of Economic Affairs, 

the Ministry of Spatial planning, Housing 

and the Environment and the Ministry of 

Education, Culture and Science. Every year 

approximately EUR 40 million is granted. 

This paper will give an introduction to the 

EET programme. It will explain the background 

and the outlines of the programme, 

and gives some project examples and project 

results. Furthermore, developments on policy 

level are described. 

2. BACKGROUND 

In 1995 the EET programme was announced 

in a policy document named “Knowledge on 

the move” by the Ministry of Education, 

Culture and Science and the Ministry of 

Economic Affairs. 

This policy document states that technological 

innovations can build a bridge between economy 

and the environment. To achieve both 

substantial economic and ecological benefits, 

technological breakthroughs are necessary. 

The speed of regular, autonomic technology 

development (incremental) is not fast enough. 

Favourable conditions have to be created in 

order to achieve these breakthroughs. 

The EET programme should initiate strategic 

long-term research and should mobilise 

knowledge of industry, universities and 

technological institutes, by subsidising R&D 

projects. 

The EET programme started off in 1996. The 

programme is executed by the EET 

Programme Office, which is a partnership 

between Novem and Senter. 

In the policy document “Environment and 

Economy” (1997) the principles of the EET 

programme are further specified. The complete 

integration of environmental aspects 

into the development of new processes, 

products or services should lead to the disconnection 

of economic growth and pressure 

on the environment. 

As a result, the EET budget was raised and 

the programme was expanded with socalled 

‘embryonic projects’. These are 

smaller feasibility projects that should lower 

the threshold for multiyear EET projects. 

In this period of time, the Ministry of 

Housing, Spatial Planning and the Environment 

entered the EET programme by contributing 

to it. 

A striking feature of the scheme is the 

emphasis laid on co-operation. Practice 

shows that companies or researchers on 

their own do not create breakthroughs. Cooperation 

is an absolute precondition for 

the development and nurturing of complex 

technologies. 

In the Netherlands there’s a wide range of 

schemes to stimulate technological innovations. 

It varies from tax incentives until 

schemes for specific sectors e.g. ICT. The figure, 

which is a simplification of reality, illustrates 

the position of the EET programme. 

The EET programme focuses on system 

innovations and process innovations. 

Process innovations are defined as a radically 

re-engineering of production processes. 

Process innovations go beyond reengineering. 

System innovations are 

innovations that can radically change the 

social infrastructure, production chains and 

consumer behaviour (examples are the use 

of renewable energy instead of energy from 

fossil sources and the use of renewable 

natural resources). System innovations 

require an integral approach in which various 

parties from a wide vision on sustainable 

development may find each other. The combination 

of experience and insights of 

people in business, the government, social 

127



Position EET programme 

Transitions 

Systeminnovations 

Changes in social 

infrastructure 

Technological or 

processinnovation 

Processoptimization 

5 20 

Time (y) 

Shemes 

EIA 

MIA 

M&T 

TS... 

EET... 

> 

institutions and science. The EET programme 

only stimulates the technological component 

of system innovations. 

3. OUTLINES OF THE PROGRAMME 

Goal 

Goal of the EET programme is to stimulate 

large long-term projects, aiming at technological 

breakthroughs. These breakthroughs 

are intended to lead to substantial economic 

and ecological benefits. 

This should reinforce both the competitiveness 

of the Dutch industry and the Dutch 

knowledge infrastructure. 

Themes 

The EET programme focuses on five environmental 

themes: 

1. Renewable raw materials 

2. Sustainable energy sources 

3. Sustainable industrial production 

processes 

4. Traffic and transport 

5. Eco design. 

These themes are selected because they 

offer solutions for the main environmental 

problems as signalled in the Dutch National 

Environmental Policy Plans. 

Target groups 

EET is especially aimed at 

- (large) companies 

- technological research institutes 

- universities 

- SME (small to medium-sized enterprises) 

- intermediates 

Execution 

EET subsidies are granted on the basis of a 

competitive tender procedure. The budget 

put out to tender every eight months is EUR 

23 million of which EUR 1.1 million is reserved 

for ‘embryonic projects’ (one year feasibility 

projects). Accidental budget increases occur. 

On average EUR 40 million a year is granted. 

In each such eight-month period proposals 

for multiyear projects and for one-year 

‘embryonic projects’ can be submitted. The 

latter are pre studies in the case that the 

uncertainty about a new technology is so 

great that a full project proposal for a multiyear 

project is not feasible yet. 

In the case of multiyear project proposals 

the EET scheme provides for both an advisory 

round and a qualification round. The 

advisory round is intended to make a first 

selection of projects and to improve the proposals. 

In the case of one-year embryonic 

projects there is no official advisory round, 

only the qualification round. 

128




An external EET Advisory Committee 

appointed by the three Ministries assesses 

the proposals and makes recommendations 

regarding the form and content of the 

project. On the basis of their advice the coordinator 

of the project makes a final proposal 

for the qualification round. In this 

round, the EET Advisory Committee ranks 

all the proposals in order of quality: the 

extent to which they contribute to the 

objectives of EET 

Subsidies are then granted in ranked order 

until the funds are exhausted. 

The EET Programme Office plays an important 

role in advising organisations regarding 

new proposals. 

The subsidy percentage is highest for fundamental 

research remote from the market: 

62,5%. This reduces to 40% for industrial 

research directed at the development of 

new processes, products and services and 

25% for pre-competitive development. 

Assessment criteria 

The assessment criteria for new projects 

focus at economy, ecology, technology and 

collaboration. 

Each assessment criterion comprises approximately 

twenty subcriteria, of which some 

examples are given below: 

Economy 

On what scale will the project lead to economic 

effects, such as profits, turnover, scale 

of the market, cost savings, etc. This has to 

be expressed in millions of Euro. 

Ecology 

On what scale will the project lead to ecological 

effects This has to be quantified, 

e.g. in Peta Joules, kilograms etc. Will the 

project contribute substantially to solving 

major environmental problems 

Technology 

Will the project lead to a technological 

breakthrough Is the proposal a new development 

Collaboration 

Is there collaboration between the knowledge 

sector and the industry Will the collaboration 

lead to a successful project and a 

successful market launch Is there enough 

commitment from the industrial partners 

For economic and ecological effects, risk 

(chance of success) and reward (potential 

impact) are distinguished. There is a tendency 

towards rewarding high risk/high 

reward projects better than low risk projects, 

since these projects are best in line 

with the overall EET goal. 

At the start of the project a collaboration 

agreement is required to ensure collaboration 

between industry and knowledge institutes. 

The agreement describes the terms 

on which co-operation take place and 

arranges the distribution of knowledge and 

exploitation rights. 

Monitoring and evaluation 

For monitoring and evaluation in order to 

measure the outcome of the projects and to 

improve the EET programme, ‘input and 

outcome performance indicators’ are used. 

Input indicators describe the input into the 

R&D activity in terms of various resources. 

Typical parameters are R&D costs, public 

funding, type of technology, type of the 

companies participating, etc. 

The outcome indicators describe the outcome 

that is realised or expected from a 

project. These indicators include data of 

new products, processes, services or methods, 

new companies, applied new technologies, 

increased turnover, export, employment, 

ecological effects (e.g. the decrease 

in emissions of pollutants), etc. 

The input performance indicators are 

mostly collected from the project proposal. 

A standard form is used for the collection of 

the outcome performance indicators. By 

means of this form the project co-ordinator 

must report the current insights on the 

expected outcome of the project every six 

months and after completion of the project. 

The standard form is an annex along with 

129



the half-yearly progress reports and the final 

report. 

Though relatively simple, the outcome indicators 

currently used give reasonable indications 

for the expected outcome. Yet, at this 

moment the EET Programme Office is investigating 

if indicators can be developed that 

give more reliable results on the expected 

outcome (see Chapter six). An important 

condition is that these indicators must be 

relatively easy to determine, since the workload 

for the project co-ordinator should be 

kept at a minimum. 

Eco-design 

19% 

Renewable 

resources 

21% 

Themes 

Transport 

9% Sustainable energy 

16% 

Participants 

Sustainable 

industrial processes 

42% 

> 

4. PROJECT CHARACTERISTICS AND 

EXAMPLES 

Universities 20% 

Other 

4% 

SME 27% 

Characteristics 

There is no such thing as an average EET 

project. Nevertheless, based on statistics, an 

average (multiyear) EET project can be 

described by the following characteristics: 

number of project partners: 5 

Mostly a consortium comprises the whole 

chain of knowledge providers, and suppliers 

and customers (see figure ‘participants’ 

below, for participation in EET per target 

group). 

duration of the project: 5 years 

costs and subsidy: The total mean costs are 

EUR 4 million per project of which half is 

subsidised (EUR 2 million). 

time-to-market: The time-to-market for 

most projects is between 7 and 10 years. 

type of project: The major part of EET projects 

fits into the scope of the‘sustainable industrial 

production processes’ theme (about 40%). 

The ‘traffic and transportation’ theme is least 

represented (see figure ‘themes’). 

Examples 

Some examples of EET projects: 

Mega Wind Turbines 

The development of 5 to 6 Mega Watt offshore 

wind turbines. 

Research centres 

21% 

Companies 

21% 

All-PP 

The development of polypropylene, reinforced 

with polypropylene, instead of reinforced 

with other fibres. 

Eco drive 

The development of an optimised power 

train comprising a combustion engine, continuous 

variable transmission and flywheel 

for passenger vehicles. 

Food preservation 

The development of a competitive process 

to ensure longer shelf life using ultra high 

pressure. 

Signature analysis 

Prognosis of the lifetime of xerox machines 

on the base of changes in behaviour of 

electromechanical modules. 

Micro-algae 

Combined production of fine chemicals and 

energy from micro-algae. 

130




> 

5. RESULTS 

The programme office evaluates the expected 

results of the EET programme on a 

regular basis, every two years. The most 

recent evaluation was in 2000, based on the 

first six tenders (53 projects). 

As the EET programme supports projects of 

which the expected market-introduction lies 

at least five years ahead, hardly any results 

(products, processes, services etc) have 

entered the market yet. That is why we are 

discussing expected results here. 

Currently, nearly one hundred multiyear 

projects received EET subsidy and about 70 

‘embryonic projects’ started off. Only a few 

EET projects have been completed. 

With each tender the EET Programme Office 

receives a sufficient number of project proposals. 

The quality of the proposals shows a 

gradual increase. The scores given by the 

Advisory Committee measure this. 

About 60% of the embryonic projects result 

in an EET project. This is according to the 

original intention of these projects. These 

one-year projects especially meet with the 

needs of research institutes and SME in the 

research-field. 

If all EET projects succeed, this would lead to 

an expected increase in turnover and cost 

reductions of EUR 10 billion. Experts of the 

53 evaluated projects estimated that about 

10% of the projects will succeed, which 

would lead to an economic effect of approximately 

EUR 1 billion, compared to subsidies 

amounting to EUR 0.13 billion. 

A substantial contribution to ecological 

aspects is estimated, especially in the fields 

of climate (CO 2 emissions), acidification, soil 

degradation caused by decrease of thegroundwatertabel, 

and NOx. In case of 

100% success of the projects, this could lead 

to an abatement of the above mentioned 

environmental problems with resp. approximately 

10%, 25%, 45% and 5%. 

EET projects are highly innovative regarding 

> 

technology. Most projects are aimed at 

process innovation and/or the technology 

component of system innovations. Until the 

year 2000 67 patens were submitted. 

It should be stated again that the results 

mentioned above were based on a 2000 

survey – since then three more tenders have 

been put out. 

6. RECENT AND FUTURE 

DEVELOPMENTS 

Horizon explorations 

In the so-called Horizon Explorations it is 

investigated every two years if the ecological 

ambitions (themes) of the EET programme 

are still in accordance with the 

major environmental problems, and if 

adjustments are necessary in the programme 

execution to increase the impact of 

EET subsidies. 

The last horizon exploration was carried out 

in 2001 by a consultancy firm. One of its conclusions 

was that the themes were still in 

accordance with the topical major environmental 

problems. Recommendations have 

been made to make minor changes in the 

evaluation criteria, so that there will be 

more focus on: 

- system innovations for sustainability rather 

than eco efficiency, 

- high risk/high reward, and 

- non-technological activities, such as business 

planning, Lifetime Cycle Analysis and 

competitive technology analysis. 

In addition it was recommended to increase 

the efforts on monitoring by developing a 

new monitoring system, named “long-term 

impact monitoring”. The objective is to 

have a more reliable view on the contribution 

of both individual projects and the 

overall programme to sustainability, at the 

start, during en after completion. This monitoring 

system should have more advanced 

performance indicators compared to the 

ones currently used. The indicators must be 

applicable on ecology, economy, technology, 

the transition to sustainability, community 

added value, technology and collabora- 

131



tion between industries, research institutes 

and universities. 

The assessment criteria have been changed 

according to the recommendations. No decision 

has been made yet on developing the 

“long-term impact monitoring system”. The 

EET Programme Office has been asked to 

investigate the options and the feasibility. 

This investigation will be performed in the 

second half of 2002. 

Evaluation of the programme 

A consultancy firm currently evaluates the 

effectiveness of the EET programme. Almost 

thirty subjects of investigation are formulated. 

Two of the main subjects are: 

- Does the EET programme contribute to a 

change of company investments and a 

direction of R&D towards innovation 

towards sustainability (e.g. influence on 

investment decisions, influence on the 

organisation structure of R&D and influence 

on technology strategy), 

- Does the integration of fundamental 

research, industrial research and pre-competitive 

development and the collaboration 

between companies, research institutes 

and universities lead to a better 

chance of a successful market introduction 

or to an improvement in speed or quality 

of R&D. 

Preliminary results show that the scheme is 

effective in the way that it stimulates the 

type of projects that was aimed at at the 

start. Since none of the projects has reached 

the market yet it is hard to determine the 

effects on economy and ecology. The EET 

programme has made a clear contribution 

to the formation of new collaborations and 

networks of excellence in the field of sustainability 

and has led to in increase in 

awareness that collaboration benefits both 

the knowledge institutes and the companies. 

Participants are positive about the 

objectives and outlines. 

The technological institutes and to a lesser 

extent the universities play a large role in 

the projects. These groups make over fifty 

percent of the total project costs and in 

quite a lot of the cases they are the initiators 

of the projects. This “dominance” by 

the technological institutes is seen as undesirable 

by the consultancy firm that evaluates 

the EET programme. They interpret it 

in the way that the majority of the projects 

are so called “technology push”. For these 

project there is some concern about the 

chances of a successful market introduction. 

It is recommended to take more effort 

into increasing the role of the industrial 

parties and to extend the share of business 

development (including a technology 

implementation plan) within the projects. 

The dominance of the knowledge sector 

doesn’t come as a surprise. On average the 

time to market for an EET project is 8 years 

or more. Furthermore it is possible to divide 

EET projects in two or more project phases 

of at least 2 years each. For each phase a 

separate apply for subsidy must be submitted. 

Most of the EET projects are still in the 

first phase, this means that the emphasis is 

on fundamental research. In this phase of 

fundamental research companies are reluctant 

to take a leading role because of the 

high risk. In the second phase, when there is 

more confidence in the technology and the 

business opportunities, we see an increase in 

the share of the industries and often they 

take the lead. 

Within companies you see a change from a 

central research organisation to research 

steered by the business units. For their innovation 

process companies are more dependent 

on third parties like the technological 

institutes and universities. At that point 

the EET programme fits in perfectly. 

Companies participating within the EET 

projects have recognised that sustainable 

enterprise also is a precondition for the 

continuity of a business. 

The evaluation mainly focuses on the 

programme results so far. But also recommendations 

will be made on the necessity to 

make adjustments within the scope of the 

latest developments in Dutch policy on 

sustainability (e.g. the 4th National Environmental 

Policy Plan). 

132




The results of the evaluation will be 

reported in July 2002. 

Policy 

The 4th National Environmental Policy Plan 

was presented in 2001 (NEPP4). The plan outlines 

the Dutch government's environmental 

strategy up to the year 2030 both in a 

national and an international context. Seven 

important environmental problems have 

been identified. To solve these problems 

‘transition to sustainability’ was introduced to 

the government’s strategy. Solving the major 

environmental problems requires system 

innovations: in many cases this can take the 

shape of a long drawn-out transformation 

process comprising technological, economic, 

socio-cultural and institutional changes. 

The concept of transition is definitely not new. 

What is new, however, is the use of the concept 

of transitions to describe broad social changes 

and to explain their mutual connection. 

At this moment 4 action plans for transitions 

are being prepared: 

• sustainable energy supply 

• sustainable mobility 

• sustainable agriculture 

• sustainable use of biodiversity and natural 

resources 

The striving towards sustainable development 

raises the question whether transitions 

can be managed in such a way that economic, 

ecological and social goals are met in one 

concerted process. In order to explore the 

possibilities of transition management the 

Ministry of Environment committed a 

research project on this subject in the course 

of the NEPP4 process. Thus it was found that 

managing transitions requires in the first 

place a thorough analysis of the actors 

involved. This analysis should explicit the perspectives 

of the various actors, on which one 

transition objective is based. The transition 

objective should be described in such a manner 

that it is considered as a shared perspective. 

Moreover the transition objective should 

be reachable via various transition paths, 

which makes it attractive for various actors. 

Transition management describes the concerted 

actions of all relevant parties sharing 

a common perspective. 

In transition management, the following 

principles are important: 

• long term goals are leading for short term 

policies 

• thinking in terms of multi-domain, multiactor 

and multi-level is required 

• learning processes should be organised 

• improvement and system change should 

be combined 

• a large number of options should be kept 

open 

Transition management is not a blueprint, 

but a model for development focused on 

learning, experimenting and innovation. 

Transition management tries to combine 

public and private interests by creating conditions 

(e.g. prices that tell the truth) that 

direct innovation to sustainability 

Recent developments in policy and society 

have led to several initiatives within the 

Ministry of Economic Affairs to give sustainability 

a more prominent place in their policy. 

One of the examples is the institution of a 

project team that will develop new visions and 

strategies on technological innovations for 

sustainable development. Yet, it is too early to 

say what consequences the outcome of these 

initiatives may have on the incentive schemes 

for ‘innovations towards sustainability’. 

The view towards innovations in the 

Netherlands is based on a cyclic innovation 

model. Characteristics of this model are: 

- innovation on boundaries between disciplines 

- thinking multi- & interdisciplinary 

- innovation is result of influencing developments 

in technology, science, society and 

market 

What should and can the government do to 

promote innovation towards sustainability 

In the first place provide a clear and consistent 

vision on the direction of sustainable 

development. This vision should give the 

boundary conditions and long term perspective, 

and must give companies and knowl- 

133



edge infrastructure the opportunity to find 

the most effective and efficient way to reach 

these goals. Providing the right conditions 

for sustainable development includes a 

good legislation and transparent competitive 

markets. This also includes that environmental 

costs are shown in the prices of products 

so these become visible for customers. 

The keys to successful innovations are cooperation 

and networks. Often new partnerships 

must be started in order to react in 

a flexible manner to the changing demand 

of society. The government needs to stimulate 

R&D and co-operation between the 

different partners in the innovation 

process. The Dutch government now has 

the task to melt together the insights of 

NEPP4 on transitions and system innovations 

with visions concerning incentives 

towards sustainability. 

National Initiative for Sustainable 

Development 

Another initiative from the Dutch government 

to be mentioned is NIDO, the Dutch 

National Initiative for Sustainable Development 

established in 1999. 

NIDO’s aim is to make a quantum leap 

forward in sustainable development. 

By acting as a spur to new development, 

NIDO can help organisations to fulfil their 

ambitions with regard to sustainable development. 

The Dutch government believes 

that major advances can be made if firms, 

government bodies, scientists and civil-society 

organisations are prepared to pool 

their resources and expertise. The job of 

NIDO, as an independent trailblazer, is to 

bring the various parties together and to 

enable each of them to find its own particular 

route towards sustainable development 

(www.nido.nu). 

> 

Epilogue 

As described above the EET programme is 

reviewed on a regular base and in various 

ways in relation to its effectiveness and its 

accordance with the topical developments 

in policy, market and society. 

Until now, six years from the start, the 

formula of the EET programme still is effective 

and only minor adjustments have been 

made in the way of execution. Examples are 

the introduction of the embryonic projects 

and the small changes made in the assessment 

criteria. 

These regular reviews by the Programme 

Office, the Ministries and third parties, will 

be continued to make sure the connection 

with topical developments in policies, market 

and society will remain to guarantee 

that the main objective of the EET 

programme, ‘breaking through to sustainability’, 

will be realised. 

MORE INFORMATION 

If you wish more information, feel free to 

contact EET Programme Office through our 

web site www.eet.nl or through our secretariat, 

phone number +31 30 2393 436. 

Publications available in English: 

- Breaking through to sustainability – innovation, 

co-operation, knowledge-building 

in joint service of economy and environment; 

published by EET; December 2000. 

- Innovation for sustainability – technology 

meets the market; published by EET on the 

occasion of the EET conference on March 

13th and 14th, 2002. 

134


FLANDERS’ NEW SCHEME 

OF INNOVATION SUBSIDIES 


PAUL ZEEUWTS 

President IWT-Vlaanderen 

Paul Zeeuwts is Civil Engineer since 1971, 

Master in Economic Science since 1975 and 

holds a Bachelor Degree in Philosophy 

(1972), all from the Catholic University of 

Leuven. 

He started his professional career as a consultant 

at the Office for industrial promotion 

(Dienst voor Nijverheidsbevordering). 

He became project leader in very different 

subjects in the energy field, industrial sector 

studies (glass, chemical, electricity, cokes, 

refineries, etc.), industrial policy, etc. 

In 1985 he was transferred to the Research 

Service of the Flemish administration (Dienst 

Onderzoek). 

For two years he became manager in the private 

sector, dealing with 'Third party financing' 

in energy conservation. 

From '88 till '91 he was head of the Cabinet 

of the Secretary of State on Science Policy. 

On 24 July 1991 he was nominated President 

of IWT. 

135

Flanders’ new scheme of innovation subsidies for Sustainable Development 

> 

1. CONTEXT1. 

Context 

At the start of the new Flemish Government 

in 1999 the general option was taken to 

organize subsidies schemes for R&D and 

Innovation on a horizontal basis, not specific 

for technology domains or sectorial issues. In 

the Belgian context innovation policy is a 

regional matter. As for Flanders the following 

schemes are in application: 

- R&D subsidies for company projects in general, 

eventually in collaboration with universities 

or research centres. 

- A subsidy scheme for SME’s (Innovation 

Studies and Innovation projects). 

scheme or regulation, which could be directly 

of use for the scheme that was ambitioned. 

Three main issues were: 

- The identification of technology projects 

that could be considered as having an 

innovation goal focused on sustainable 

development. 

- The determination of an ambition level 

towards sustainable development, which 

would be sufficient to justify extra incentives. 

- The kind of extra incentives that would be 

feasible to be worthwhile within the general 

objectives of the scheme. 

- Projects of long-term strategic basic research 

at universities and research centres. 

- Technology diffusion projects at technical 

schools for higher education. 

- A scheme for clusters of companies with 

projects in the fields of innovation stimulation 

and synergies, technology assistance 

and collective research. 

- PhD and post-doctoral grants in applied 

research. 

The new Flemish Government as a result was 

not in favour to launch specific programs in 

the field of Sustainable Development, 

Environmental technology or Energy savings 

and renewables. Instead it was in favour of 

integrating the goals of Sustainable 

Development in general subsidy schemes. 

Sustainable Development goals were however 

restricted to their ecological dimensions 

(environment and energy). 

In 2001, IWT was given the task to elaborate a 

regulation to tackle these general objectives 

in practical terms. The IWT-proposal was accepted 

by the Flemish Government on May 3rd of 

2002 and has been introduced since then. 

The general outlines of this regulation are 

further developed. 

International bench marking didn’t procure a 

> 

2. THE DEFINITION OF TECHNOLOGY 

INNOVATION PROJECTS 

CONTRIBUTING TO A SUSTAINABLE 

DEVELOPMENT SUSTAINABLE 

DEVELOPMENT IS A NOTION WITH 

VERY DIFFERENT DIMENSIONS 

(AGENDA 21 E.A.). 

The option was taken to start with a scope 

limited to the ecological dimension of ‘sustainable 

development’. A broader ambition 

will be analysed within the next two years. 

Anyhow, the objective of the Flemish 

Government was from the beginning in 

1999 limited to the ecological dimension. 

Following 7 innovation objectives were 

selected to characterize projects as “potentially 

contributing to sustainable development”: 

1. Savings of raw materials 

2. Energy savings 

3. Reduction of emissions 

4. Reduction of waste and environmental 

nuisance (ex. noise, smell, light, 

electro-magnetic waves, etc …) 

5. Use of renewable energy and raw 

materials 

6. Recycling and re-use 

7. Enhancing life cycle of products and 

processes 

Having at least one of these innovation 

136



> 

objectives is a necessary condition to be considered 

as projects entering the scheme. 

3. THE AMBITION LEVELS IN THE 

POTENTIAL RESULTS OF INNOVATION 

3.1 In general terms, the importance of a 

technological innovation for its environmental 

impact has to be linked to its 

exploitation potential. The technological 

performance on itself is not really meaningful 

if the related new technology is 

not sufficiently diffused through commercial 

exploitation. It is preferable to 

favour technology with rather limited 

environmental improvements per unit of 

product or service but with broad commercial 

potential (e.g. large series of 

products, wide application) than technology 

with huge environmental performances 

but with a very limited 

exploitation potential. Anyhow, all technology 

projects supported by IWT are 

assessed not only on the criterion of 

their scientific or technological quality 

but also on the criterion of their 

exploitation potential 

3.2 The ambition level of a project is considered 

contributing to the SD-goals in its 

technological environmental performance 

as “sufficiently high” with an innovation 

objective, which is the primary 

goal of the project. 

What is “sufficiently high” 

This has to be situated in the kind of 

project. For projects of strategic basic 

research with exploitation potential in 

the middle or long term, eco-efficiency 

improvements have to be in order of at 

least 75% (rather “break-through” technologies; 

factor 4 or 5). For projects with 

a rather short-term exploitation goal, 

typically company projects, the eco-efficiency 

improvement should reach at 

least 50% (factor 1,5; reduction of environmental 

impact at least 30%). Both 

targets are set in comparison with the 

actual performance of technology today. 

And in any case, the eco-efficiency levels 

of the actual Best Available Technology 

(BAT) should be improved. 

If these objectives are assessed to be 

realistic and feasible, the project is considered 

as “potentially contributing to a 

sustainable development”. 

3.3 Different cases could be less clear in the 

sufficient ambition level in relation to 

the SD-goals: 

- Projects can have different innovation 

objectives with an improvement of 

environmental impacts as one of them. 

The environmental goal could be 

important or marginal in comparison 

with the overall innovation objectives. 

- Projects can have a more limited ambition 

level in eco-efficiency improvement 

but with an important commercial 

potential and as a result an important 

improvement on the environment in 

comparison with the actual (industrial) 

situation they could substitute. 

In these cases a more in-depth analysis is 

requested from applicants in order to be 

assessed as “projects potentially contributing 

to SD”. 

The methodology in this approach has 

been elaborated in close collaboration 

with VITO (Vlaamse Instelling voor 

Technologisch Onderzoek, a research 

centre in Mol (Flanders) specialised in 

energy and environmental technologies). 

The methodology proposed is the ECOpoints 

approach (ECO-indicator 1999) 

closely linked to the LCA-approach(*). The 

environmental improvements (as the negative 

effects) of the new technology have 

to be compared and measured in terms of 

ECO-points in comparison to the (industrial) 

situation that would be substituted 

(“reference situation”). Not only per 

“unit” but also in relation to the scenarios 

of commercial diffusion over a period of 

10 years after commercial introduction. 

The average expected ECO-points 

improvement has than to be “monetarised”, 

assuming that on average 1 

137


> 

ECO-point = 3 . This gives us a (rough) 

magnitude of the ecological externalities 

ambitioned with a technological and 

commercial successful innovation. 

If the ratio of these calculated externalities 

related to the total subsidies applied 

for, exceed a factor 4, than the project 

also can be assessed as sufficiently “contributing 

potentially to SD”. 

3.4 Overview of the assessment procedures 

The procedures described before result 

in the following procedure scheme in 

order to assess if a project is considered 

as contributing to SD or not. 

4. EXTRA INCENTIVES FOR 

SUSTAINABLE DEVELOPMENT 

Two incentive mechanisms are introduced 

for projects considered as being potential 

contributors to SD. A pre-condition is in any 

case that the projects can be considered as 

sufficiently good in relation to the other 

assessment dimensions (essentially the scientific/ 

technological quality of the project 

and the exploitation potential of its results). 

Indeed, it would not be acceptable to subsidize 

qualitatively poor projects even if they 

aim to contribute to SD. 

4.1 The mechanism of “priority projects”. 

In subsidy schemes where a high level of 

subsidies (80% - 100%) is foreseen a subsidy 

bonus is not realistic (strategic basic 

research at universities; technology diffusion 

projects at technical schools for 

higher education, etc…). Instead, a 

mechanism of “positive discrimination” 

in the selection is put forward. This 

implies that given the budgetary constraints 

for the program a project earmarked 

as “contributing to SD” can 

obtain a subsidy even if otherwise it 

would have been ranked as “good” but 

not supported given the overall budget. 

This treatment of “positive discrimination” 

is however limited to a quota that 

should be reached for SD-projects (as a 

% of the overall budget). These quotas 

are fixed per program and increases 

gradually over the years. 

This mechanism of “priority projects” is also 

applicable in programmes with a lower level 

of subsidies, in particular to company projects 

of R&D. Details on this mechanism would 

bring us too far within this contribution. 

4.2 A subsidy bonus 

Essentially for company projects of R&D 

a subsidy bonus of 10% is given if the 

project is assessed as “potentially contributing 

to SD”. This is compatible with 

the R&D-subsidy regulation based upon 

the EC-guidelines. 

This bonus could also be applied to a 

part of the project if this has specific SDgoals 

on itself. 

A feasibility study linked at an R&D-project, 

aiming at SD-goals can also be supported 

(50% subsidy; max. 10% of the overall project-budget). 

Innovation studies in the SMEscheme 

can include LCA and Ecodesign 

studies up to 50% of the overall budget. 

Project proposal with potential contribution 

to SD applying for SD incentive 

Y 

Condition 1: Innovation goal 

primarily focused on SD-objectives, cfr.2 

N 

Condition 2: 

Sufficient eco-efficiency 

ambition level; cfr. 3.2 

N 

Evt. Condition 3: 

Ecopoints improvements 

in / subsidies > 4; cfr.3.3 

Y 

Y 

Project considered as having a sufficient potential contribution to SD 

Project not considered as SD-contributor 

SD-incentive (cfr.4) 

No SD-incentive 

138



> 

5. EVALUATION OF THE IWT-PROJECT 

PORTFOLIO IN THE PAST 

5.1 Projects with SD-goals as considered in 

this scheme are not new, of course. The 

project portfolios of recent years have 

been screened within this respect. 

In the period October 1997 - September 

2000 subsidized company projects 

account for about 16% projects that 

could (roughly) be considered as SD-projects 

within the new scheme. The following 

characteristics for these projects can 

be analysed: 

SD-projects in relation to their ecological 

goals 

Preventive technology 74% 

Reduction of resource use 70% 

Environmental 

or process monitoring 7% 

SD-projects in relation to their 

technological ambition 

Incremental optimalisation 

of products or processes 19% 

Product innovation 48% 

Process innovation 52% 

System innovation 11% 

SD-projects in relation to the ecological 

impact and goals 

Ecological impact High (1) Limited (2) 

1. Savings in raw materials 19% 26% 

2. Energy savings 4% 41% 

3. Reduction of emissions 37% 33% 

4. Reduction of waste a.o. 0% 30% 

5. Renewable energy 4% 4% 

6. Recycling and re-use 11% 0% 

7. Enhancing Life cycle 7% 4% 

(1) 

More than 30% improvement 

(2) 

10% to 30% improvement 

5.2 In other programmes and schemes the 

following proportions of SD-earmarked 

projects in the portfolios could be 

noted, although the SD-definitions were 

taken into account in a more qualitative 

manner: 

- Strategic Basic Research at universities: 

20% (2000 - 2001) 

- Technology diffusion from Technical 

Higher education: 22% (2000 - 2001) 

> 

6. SOME GENERAL COMMENTS 

6.1 An “exclusion criterion” has also been 

defined: if a project could lead in commercial 

exploitation (or project execution) 

to (major) problems with actual or 

future environmental requirements every 

subsidy will be denied. 

A more restrictive approach, excluding 

for instance subsidies for potential commercialisation 

leading to more energy or 

raw material consumption, was considered 

problematic in application. 

6.2 No distinction has been made for ‘endof-the-pipe’ 

technologies or for projects 

aiming at meeting future environmental 

requirements. If the technological innovation 

is leading to a sufficient high positive 

impact it is considered as having a 

potential contribution to SD. 

6.3 “Window dressing” in project description 

could be the case on the SD issue. 

However, this is not essentially different 

for the issues of scientific/ technological 

or exploitation potential in the project 

description. 

A qualitative assessment and a good control 

on execution of R&D-activities is the 

core responsibility of the subsidy agency 

in all aspects of the supported projects. 

6.4 It must be clear that a shift of technologies 

towards a more sustainable development 

cannot be the result of a subsidy 

policy for R&D and technological innovation 

on itself. 

A broader set of policy instruments has 

to be developed, essentially in the fields 

of environmental legislations, economical 

and fiscal instruments, etc… Innovation 

policy could only be part of an 

overall policymaking towards SD. But it 

can be one of the signals that the (only) 

future for companies and for society is in 

a Sustainable Development. 

139


SUMMARY 

TOWARDS POLICY-INTEGRATION 


Technopolis 

The summary by Patries Boekholt attempts 

to draw some conclusions referring back to 

the objectives of the conference, namely to 

see whether a closer integration of innovation 

policy and environmental policy on 

the overlap of innovation for sustainable 

development, is feasible. The effectiveness 

of innovation policies appears to be difficult 

to measure. We do know that individual 

policy instruments can not change the 

whole innovation system, but the right policy 

portfolio possibly could. Environmental 

policies have demonstrated some effectiveness 

in changing behaviour in addressing 

focussed technological challenges. The challenge 

now is to extend these more targeted 

approaches towards a systemic change 

process affecting the broader business community. 

We will need further policy innovations 

in the field of policy integration. 

141

Summary 

> 

> 

1. THE AMBITIONS OF 

THE CONFERENCE 

Government institutions and policies are 

urged to make ‘policy innovations’ in order 

to contribute to environmentally sustainable 

growth. Yet do we know whether today’s 

innovation policies really make a difference 

Has the innovation policy community been 

able to demonstrate the ‘additionality’ of 

innovation policies of the last decades The 

Flanders 2002 Conference of the Six 

Countries Programme (6CP) debated these 

two interdependent questions: what has 

innovation policy really achieved so far and 

how could it possibly contribute to stimulating 

environmental sustainability And what 

can the innovation policy community learn 

from the achievements of environmental 

technology policies, in order to give its mainstream 

policy instruments a greener focus 

All these were very ambitious questions that 

the conference aimed to address in two days. 

The conference was one of the first attempts 

to see whether a closer integration between 

the two worlds is feasible and as such 

offered the opportunity for ‘a first flirtation’. 

The audience of the conference came 

from both communities: practitioners and 

policy makers dealing with innovation policy 

and those dealing with environmental policies. 

Over a hundred people attended the 

conference. The following paper gives an 

overall summary of the presentations and 

discussions during the conference. Without 

aiming to be comprehensive, the paper 

highlights some of the key issues that were 

debated. For more detailed accounts of the 

contributions we refer to the papers in these 

proceedings. 

2. SETTING THE SCENE: IS INNOVATION 

POLICY READY FOR A CLOSER 

INTEGRATION WITH 

ENVIRONMENTAL SUSTAINABILITY 

The opening session of the conference set 

the scene of the discussion for the following 

two days. Whereas innovation policy focuses 

on economic well being, environmental policy 

focuses on environmental well being. 

How can these two seemingly conflicting 

objectives meet in order to achieve both A 

mix of speakers was chosen with the first 

speaker (Ken Guy) spanning the bridge 

between innovation and environmental policy 

issues, the next speaker (George Heaton) 

looking at the issues from an environmental 

perspective and the third keynote (Luke 

Georghiou) adding to the debate from the 

innovation policy perspective. 

In the words of the first key note speaker, 

we are addressing the area where innovation 

and environmental policy overlap, 

which can be called Innovation Policy for 

Sustainable Development. One even smaller 

component of this overlap area is 

Environmental Technology Policy: policies to 

stimulate the development of new technologies 

and processes that help reduce 

emissions and pollution or save energy. How 

this overlapping area is positioned and 

whether it needs widening up is one issue of 

debate. The merger of the two domains was 

compared with an arranged marriage. “That 

they are not an obvious love-match should 

be clear after 30 years of parallel co-existence 

accompanied by little interaction” 

(George Heaton). 

The lessons from the environmental policy 

perspective is that the end-of-pipe solutions, 

which are sought for in current policies, are 

not sufficient to tackle the serious problems 

that the world is faced with today. In addition, 

focusing environmental technology 

policy only on those companies in the environmental 

industry has a limited impact on 

the business sector as a whole. The polluting 

industries must be addressed as well. The 

public sector could have used the opportunity 

of regulatory standards in a more visionary 

way than has been the case so far. Both 

environmental and innovation policy are in 

need for more strategic knowledge about 

the development of technologies in the 

future. 

Remains the question whether innovation 

policy has really made a difference 

In terms of evaluators and policymakers this 

is translated in whether policies have addi- 

142

Summary 


tionality (what would have happened without 

the public support or what has happened 

thanks to this support) and impact. 

Current thinking on additionality identifies 

three manifestations: input, output and 

behavioural additionality. All three are difficult 

to measure due to problems how to 

attribute an effect to a policy intervention, 

time lags and so on. Particularly the behavioural 

additionality is meaningful since policies 

aim to have a lasting effect on firm 

behaviour and competence levels. The 

importance of bringing about behavioural 

change, in particular improvement of capabilities 

and skills, is one of the key lessons 

from a vast amount of studies on public 

incentives. But showing that this has indeed 

occurred is not an easy task. 

The three keynote speakers confirmed that 

we are dealing with two policy worlds, that 

are starting to acknowledge they need each 

other, in order to have a greater impact on 

sustainable economic growth. We have seen 

from the contributions of the country cases 

that attempts are being made to combine 

the two policy domains in integrated instruments. 

What came back in all presentations 

and the discussions was the certitude that 

we need to address the policy challenges in 

relation to the innovation system: not one 

policy instrument can have an impact on this 

system, but the right portfolio of policies 

could have. If the innovation system consists 

of various boxes with for instance the business 

sector in one and the research organisations 

in another, the policy challenges are 

to form linkages between the different components 

that constitute the innovation systems. 

Addressing the innovation and sustainability 

issues for each of the boxes 

separately is not sufficient. From a system 

perspective one could begin by questioning 

what capabilities and incentives are needed 

for engagement in sustainable innovation. 

This asks for strategic intelligence to analyse 

which weak nodes in the system could be 

‘repaired’ and if so whether the state has a 

role to play. 

An additional barrier in bridging the innovation 

and environmental policy domains 

are the institutional barriers that we can 

> 

find in most countries, where the walls 

between the responsible departments are 

still high, and much is to be done to develop 

integrated policy approaches. 

3. DO INNOVATION POLICIES MAKE 

A DIFFERENCE 

We can roughly say that Europe has witnessed 

one to two decades of innovation 

policy with a wide variation of instruments 

aiming to achieve, among others, higher levels 

of private R&D investment, more collaboration 

between industry and public 

research organisations, a greater absorptive 

capacity of small firms and better access to 

advice on innovation. The second parallel 

sessions in the conference addressed the 

question: after years of innovation policy 

instruments: do we know what difference 

they have made Have they been able to 

achieve their objectives 

The contributions in this conference came 

up with a mixed picture: 

1 Yes innovation policy has made a difference, 

as is shown by many evaluation studies. 

Several presentations showed examples 

of evidence from evaluations that 

public R&D instruments have made a 

difference. Jon Hekland from Norway 

explained how the Norwegian evaluation 

system is gradually improving the understanding 

of the additionality of Norwegian 

instruments. There seems to be a positive 

effect on the R&D activity and business 

performance. What is more difficult to 

capture is the effect on behaviour and the 

soft factors involved in innovation. Mette 

Rye’s presentation showed that public 

incentives have a positive effect on the 

type of activities business perform: evaluations 

show that publicly funded R&D projects 

have a higher research component 

than privately sponsored R&D projects. 

Tom Poot’s contribution was based on a 

study of R&D collaborations between firms 

and between firms and research organisations. 

It showed that there is a positive correlation 

between firms entering in S&T 

collaborations, often stimulated by public 

incentives, and their economic perform- 

143

Summary 

> 

ance. It also showed that the public funding 

of these collaborations has added 

value. 

2 We can also argue that the impact of public 

incentives has been minimal if we look 

at a longer time span: in the years 1993- 

1998 the proportion of government 

support for the Business Expenditure in 

R&D (BERD) in OECD countries has gone 

down, whereas private R&D investment 

went up in this period with 35%, as Luke 

Georghiou pointed out 

3 The main line of discussion was however 

that there is a lot that we do not know 

about the effect of innovation policies on 

innovation and competitiveness. There are 

still fundamental methodological problems 

in measuring socio-economic impacts. 

So we can not say with great certainty 

which policy instruments work best. Erik 

Arnold’s presentation highlighted that: 

- The ‘soft’ factors of innovation are crucial 

for economic success 

- When assessing the success of policy 

instruments, the effect on behavioural 

additionality needs to be taken on 

board, yet we are not capable of measuring 

these 

- Thus, if we take input and output additionality 

concepts as the main test 

whether innovation policy works it will 

lead to a very narrow definition of what 

works 

Again it was stressed that it is not sufficient 

to look at individual instruments, but how 

the policy mix works in relation to the innovation 

system. Members of the audience 

also stressed that this should not be limited 

to innovation policy instruments alone, but 

also include general framework conditions 

necessary for a knowledge based economy. 

4. WHAT CAN INNOVATION POLICY 

LEARN FROM ENVIRONMENTAL 

TECHNOLOGY POLICIES 

Environmental technology policy for years 

has had an approach to technology which 

aimed at incremental improvements, 

addressing specific environmental bottlenecks, 

with today’s best available technology. 

Some speakers in the conference, particularly 

René Kemp and Jesper Holm’s contributions, 

pleaded for a more systemic 

approach to address urgent environmental 

problems such as global climate change. This 

requires better integration of both policy 

realms. 

It appears that innovation and environmental 

technology policies are not that different 

in their effect. Evaluations show similar patterns 

of additionality, when measured by surveys 

of programme participants. One of the 

key challenges in environmental technology 

policy is finding the right balance between 

programmes that support incremental innovations 

(the safe route) and more integrated 

approaches that push radical paradigm shifts 

(the risky route). But extremely so than with 

‘traditional’ innovations, addressing the 

behavioural factor is even more important. 

In the debate with the audience a point that 

came up was how to define who are the 

drivers for sustainable solutions Where is 

the market that will act as a catalyst for these 

changes. Has enough been done to involve 

consumers in acting as drivers It has often 

been public demand or public regulations 

that served as the driving force. But it was 

questioned whether this is sufficient to support 

the more radical changes that are necessary 

to address long term environmental 

problems. Kemp suggested transition management 

as an effort to work step by step 

towards more radical systems innovations, 

based on visions of goals that society wants 

to achieve. In the audience debate such a 

planned approached was questioned in 

terms of finding the right stakeholders to 

bring such a process further. 

The conference gave the floor to a number 

of practical examples of policy measures 

which are already in place, and which combine 

both innovation and sustainability as a 

goal. These were: 

• The German Green Technology Programmes 

of the Ministry of Education and 

Science (BMBF) presented by Karl-Ulrich 

Voss. These are programmes that aim to 

reach mostly traditional industries to 

improve competitiveness by introducing 

environmental processes and products 

144

Summary 


• The Austrian example of Technologies for 

sustainable development presented by 

Hans-Günther Schwarz of the Austrian 

Ministry for Industry. This programme aims 

to create a number of green projects with 

a high visibility and demonstration effect. 

Examples are the Green Factory 

• The Dutch Economy and Ecology and 

Technology programme that supports R&D 

projects to develop innovative environmentally 

friendly products and processes 

• The Flemish initiative by IWT, presented by 

Paul Zeeuwts, to include a green label as 

an additional bonus in their mainstream 

R&D support instruments. This initiative 

will be launched shortly. 

The policy practitioners stressed that a key 

challenge is to convince the mainstream 

companies of the potential benefits (cost 

reduction, customer satisfaction, …) of 

greener production processes. Another feature 

was that most of these instruments, 

with the exception of the IWT initiative, are 

not well integrated in mainstream innovation 

policy development. They usually function 

as a policy niche with little interaction 

> 

with other R&D and innovation policy 

instruments. There is a definite lacking sense 

of urgency with the ‘mainstream’ innovation 

policy makers, of the pressing environmental 

problems that we face. So from their 

side there are few incentives to also take on 

board the effects of their policies on the 

environment. 

5. OPEN QUESTIONS FOR THE FUTURE 

The conference debates revealed that the 

two policy spheres have different foci in relation 

to the business sector. If the two are to 

enter a marriage they would have to find 

ways in addressing more similar target 

groups. Innovation policy usually aims at the 

business sector as a whole, in particular those 

firms that have the willingness and capability 

to conduct some form of innovation. 

Although innovation policy instruments can 

be sectorally or technologically focused, dissemination 

to a wider business (and research) 

community is usually the objective. Another 

way of influencing innovation behaviour and 

efforts is setting technical standards. 

Different target groups and mechanisms of the two policy spheres 

145

Summary 

Environmental technology policy on the 

other hand, mostly focuses on a sub-set of the 

business community, e.g. that part of the private 

sector that develops and sells products 

or processes that help sustain the environment. 

These instruments have more problems 

reaching the wider business communities, 

since it requires a much greater effort to 

change mindsets of the private sector, if the 

economic gains are not immediately evident. 

Environmental technology policies have 

relied strongly on regulation to change 

behaviour of the larger group of firms. This 

difference in focus of the two domains is illustrated 

in Exhibit 1 below. If innovation policy 

and environmental policy are really to integrate, 

and if the two want to address the 

more systemic changes, more should be done 

to reach the business sectors not directly 

operating on the environmental technology 

markets. 

Up to now little has been done in both 

spheres to change behaviour by influencing 

(consumer) demand. Environmental technology 

policy can show some examples where 

public demand has stimulated sustainable 

solutions (transport, construction, ekolabelling). 

In the conference debate it was 

put forward that more could be done here. 

The challenge of merging innovation and 

environmental technology policies is to 

widen the scope of public supported innovations 

by setting additional incentives. The 

limited number of practical examples does 

not yet give us a wide pool of experience, to 

show how this can be done best. 

And as was stated in one of the debates: we 

have to keep in mind that a green development 

is only one societal objective, alongside 

creating of jobs and economic growth. 

This tension between objectives will not simply 

vanish by breaking down the walls of 

bureaucracy. 

> 

6. CONCLUSIONS: 

ACHIEVEMENTS OF THE CONFERENCE 

The conference is one of the first attempts 

to bring together the ‘traditional’ innovation 

policy and the environmental technology 

policy communities. 

We can not expect that it has already contributed 

to the ‘marriage plans’, but at the 

least it has facilitated a first flirtation. There 

are still many questions raised during the 

conference, that remain to be dealt with: 

• How do we create a ‘sense of urgency’ 

with innovation policy makers regarding 

sustainability problems Sustainability is 

not high on the policy agenda and still 

asks for more ‘mainstreaming’. 

• The institutional barriers in the public sector 

are a major impediment. There is a 

need for innovation of policy making 

• The debates pointed out that we need 

strategic intelligence to understand how 

to change mindsets in businesses. But what 

type of intelligence is this exactly 

• Some have argued that environmental policies 

do not take sufficient account of the 

market forces and the competitive environment 

of the business sector. Will the marriage 

between the two enhance the market 

orientation of sustainability policies 

• The incremental approach of most environmental 

technology programmes is not 

sufficient to address the pressing issues. 

Environmental policy experts argue that 

we need paradigm shifts. But how to bring 

about such a radical shift in mind sets 

How does this relate to the innovation system 

approach which innovation experts 

tend to apply. And what does it mean for 

policy implementation 

The Panel debate at the end of the conference 

took up most of these challenging 

questions, for which an hours discussion is 

barely sufficient. The Six Countries programme 

organisers have taken it upon them 

to continue the discussion on the theme of 

integrated policy approaches, reflected in 

the programme for the autumn conference 

in Germany (see.6cp.org). 

146

REEDS VERSCHENEN BIJ HET IWT-OBSERVATORIUM 

VTO-STUDIES: 

1/ Het Vlaams Innovatiesysteem: een nieuw statistisch beleidskader 

1annex/ Theoretische en empirische bouwstenen van het ‘Vlaams Innovatie Systeem’ 

2/ Innovatiestrategieën bij Vlaamse industriële ondernemingen 

3/ Octrooien in Vlaanderen: technologie bekeken vanuit een strategisch perspectief 

Deel 1: Octrooien als indicator van het technologiesysteem 

4/ De impact van technologische innovaties op jobcreatie en jobdestructie in Vlaanderen 

5/ Strategische verschillen tussen innovatieve KMO’s : Een kijkje in de zwarte doos 


Deel 2: Analyse van het technologielandschap in Vlaanderen 

7/ Diffusie van belichaamde technologie in Vlaanderen: een empirisch onderzoek op basis 

van input/outputgegevens 

7 annex/ Methodologische achtergronden bij het empirisch onderzoek naar de Vlaamse 

technologiediffusie 

8/ Schept het innovatiebeleid werkgelegenheid 

9/ Samenwerking in O&O tussen actoren van het “VINS” 


Deel 3: De internationale technologiepositie van Vlaanderen aan de hand van octrooi 

posities 

Deel 4: Sporadische en frequent octrooierende ondernemingen : profielen 

11/ Technologiediffusie in Vlaanderen. Enquêteresultaten - Product- en diensteninnovatie: 

evolutie 1992-1994-1997 

12/ Technologiediffusie in Vlaanderen. Enquêteresultaten - Hoogtechnologische producten: 

evolutie 1992-1994-1997 

13/ Technologiediffusie in Vlaanderen. Enquêteresultaten - Procesautomatisering: 

evolutie 1992-1994-1997 

14/ Technologiediffusie in Vlaanderen. Methodologie en vragenlijst 

15/ Financiering van innovatie in Vlaanderen. Het aanbod van risicokapitaal. 

16/ Product- en diensteninnovativiteit van Vlaamse ondernemingen. Enquêteresultaten 1997 

17/ Adoptie van procesautomatisering en informatie- en communicatietechnologie in 

Vlaanderen. Enquêteresultaten 1997 

18/ Performantieprofiel en typologie van innoverende bedrijven in Vlaanderen. Waarin verschillen 

innoverende bedrijven van niet-innoverende bedrijven. Enquêteresultaten 1997 

19/ De werkgelegenheidsimpact van innovatie: is de aard van de innovatie-strategie belangrijk 

20/ Samenwerking in O&O tussen actoren van het “VINS” 

Deel 2: Samenwerking in een aantal specifieke technologische disciplines

IWT-STUDIES: 

21/ Clusterbeleid: Een innovatie instrument voor Vlaanderen 

Reflecties op basis van een analyse van de automobielsector 

22/ Benchmarken en meten van innovatie in KMO’s 

23/ Samenwerkingsverbanden in O&O en kennisdiffusie 

24/ Financiering van innovatie in Vlaanderen. De venture capital sector in internationaal 

perspectief 

25/ De O&O-inspanningen van de bedrijven in Vlaanderen - De regionale uitsplitsing van 

de O&O-uitgaven en O&O-tewerkstelling in België 1971-1989 

26/ De O&O-inspanningen van de bedrijven in Vlaanderen - Een perspectief vanuit de 

enquête voor 1996-1997 

27/ Identificatie van techno-economische clusters in Vlaanderen op basis van input-outputgegevens 

voor 1995 

28/ The Flemish innovation system: an external viewpoint 

29/ Geïntegreerd innovatiebeleid naar KMO’s toe. Casestudie: Nederland 

30/ Clusterbeleid als hefboom tot innovatie 

31/ Resultaten van de O&O-enquête bij de Vlaamse bedrijven 

32/ ‘Match-mismatch’ in de O&O-bestedingen van Vlaamse en Belgische bedrijven in termen 

van de evolutie van sectoriële aandelen 

33/ ‘Additionaliteit’- versus ‘substitutie’-effecten van overheidssteun aan O&O in bedrijven in 

Vlaanderen: een econometrische analyse aangevuld met de resultaten van een kwalitatieve 

bevraging 

34/ Het innovatiebeleid in Ierland als geïntegreerd element van het ontwikkelingsbeleid: van 

buitenlandse investeringen naar ‘home spun growth’ 

35/ ICT Clusters in Flanders: Co-operation in Innovation in the New Network Economy 

36/ Het fenomeen spin-off in België 

37/ KMO-innovatiebeleid levert toegevoegde waarde aan Vlaamse bedrijven 

38/ Technology watch in Europa: een vergelijkende analyse 

39/ ICT-Monitor Vlaanderen: Eindrapport van een haalbaarheidsstudie

WHAT IS THE IWT 

The Institute for the promotion of innovation through science and technology 

in Flanders (IWT-Vlaanderen) is an autonomous public body, established by the 

Flemish government in 1991 to support industrial R&D in Flanders. For this IWT 

has various sources of finance through which financial assistance totalling euro 

200 millions is provided annually. 

It also provides services to Flemish companies in the area of technology transfer, 

partner search, preparation of projects under European programmes, etc. 

Through these activities and others IWT is developing into a knowledge centre 

for R&D and innovation in Flanders. 

WHAT IS THE IWT-OBSERVATORY 

The IWT-Observatory (Innovation - Science - Technology) is a division of IWT-Vlaanderen which 

focuses on policy support through policy indicators and policy studies. The IWT-Observatory organises 

technology surveys and collects indicators on the R&D and innovation activities of companies in 

Flanders. 

The most important task of the IWT-Observatory, however, is the organisation of innovation studies, 

with support from external research groups, for the purpose of deepening knowledge of the Flemish 

innovation system, bench-marking against foreign (policy) experience, introduction of new insights 

from innovation theory, and providing access to data from specialised surveys and databases. Until 

the end of 1998 the IWT-Observatory was known as the Vlaams Technologie Observatorium (VTO).

40 Innovation Policy and Sustainable Development - IWT

Create successful ePaper yourself

Delete template?

Save as template?