Handbook of innovation policy impact

Series Editors: Susana Borrás, Department of Business and Politics, Copenhagen Business School, Denmark, Jakob Edler, Manchester Institute of Innovation Research, Manchester Business S

Series Editors: Susana Borrás, Department of Business and Politics, Copenhagen Business

School, Denmark, Jakob Edler, Manchester Institute of Innovation Research, Manchester Business

School, UK, Stefan Kuhlmann, Science, Technology and Policy Studies, University of Twente, the

Netherlands and Ismael Rafols, INGENIO (CSIC-UPV), Polytechnic University of Valencia, Spain

and SPRU, University of Sussex, UK

The aim of this series is to present some of the best and most original research emanating

from the Eu-SPRI Forum on Science, Technology and Innovation Policy The typical

questions addressed by the books in the series will include, but not be limited to:

security or the environment?

public policies?

competitiveness and economic growth?

innovation?

higher education policies for universities?

Titles in the series include:

The Governance of Socio-Technical Systems

Explaining Change

Edited by Susana Borrás and Jakob Edler

Public Procurement for Innovation

Edited by Charles Edquist, Nicholas S Vonortas, Jon Mikel Zabala-Iturriagagoitia and Jakob Edler

Handbook of Innovation Policy Impact

Edited by Jakob Edler, Paul Cunningham, Abdullah Gök and Philip Shapira

Handbook of Innovation

Policy Impact

Edited by

Jakob Edler

Professor, Manchester Institute of Innovation Research, Alliance Manchester

Business School, University of Manchester, UK

Paul Cunningham

Senior Research Fellow, Manchester Institute of Innovation Research,

Alliance Manchester Business School, University of Manchester, UK

Abdullah Gök

Lecturer, Manchester Institute of Innovation Research, Alliance Manchester

Business School, University of Manchester, UK

Philip Shapira

Professor, Manchester Institute of Innovation Research, Alliance Manchester

Business School, University of Manchester, UK, and Georgia Institute of

Technology, USA

EU-SPRI FORUM ON

SCIENCE, TECHNOLOGY AND INNOVATION POLICY

Cheltenham, UK • Northampton, MA, USA

All rights reserved No part of this publication may be reproduced, stored in a retrieval

system or transmitted in any form or by any means, electronic, mechanical or photocopying,

recording, or otherwise without the prior permission of the publisher.

Edward Elgar Publishing, Inc.

William Pratt House

9 Dewey Court

Northampton

Massachusetts 01060

USA

A catalogue record for this book

is available from the British Library

Library of Congress Control Number: 2016932480

This book is available electronically in the

Social and Political Science subject collection

Jakob Edler, Abdullah Gök, Paul Cunningham and Philip Shapira

Philippe Larédo, Christian Köhler and Christian Rammer

3 The impact of direct support to R&D and innovation in firms 54

Paul Cunningham, Abdullah Gök and Philippe Larédo

Barbara Jones and Damian Grimshaw

5 The impact and effectiveness of entrepreneurship policy 129

John Rigby and Ronnie Ramlogan

6 The impact of technology and innovation advisory services 161

Philip Shapira and Jan Youtie

Elvira Uyarra and Ronnie Ramlogan

8 The impact of innovation policy schemes for collaboration 239

Paul Cunningham and Abdullah Gök

Paul Cunningham and Ronnie Ramlogan

10 The impact of policy measures to stimulate private demand for

innovation 318

Jakob Edler

Elvira Uyarra

12 The impact of pre- commercial procurement on innovation 382

Jennifer Cassingena Harper

Paul Cunningham, Jakob Edler, Kieron Flanagan and Philippe Larédo

18 Conclusions: Evidence on the effectiveness of innovation policy

intervention 543

Jakob Edler, Philip Shapira, Paul Cunningham and Abdullah Gök

and Management at the Technische Universität Berlin, Germany He also

holds the endowed Chair of Standardisation at the Rotterdam School

of Management at Erasmus University He received a BA from Brock

University, Canada, and his Diploma in Economics and doctorate from

Freiburg University, Germany Between 1996 and 2010, he worked at the

Fraunhofer Institute for Systems and Innovation Research ISI, Karlsruhe,

Germany Since 2010, he has been associated with the Fraunhofer Institute of

Open Communication Systems FOKUS in Berlin

Institute of Innovation Research at the Alliance Manchester Business School,

University of Manchester, UK His research interests encompass a range of

related fields in science, technology and innovation (STI) policy, including:

innovation and R&D evaluation methodologies; STI policy governance;

col-laboration between higher education institutions and industry; quantitative

measures of R&D performance and STI indicators; transnational scientific

collaboration; and international STI policy He has undertaken numerous

evaluations, reviews and studies for a wide range of bodies, and his work has

been influential in the development and formulation of STI policy at a variety

of levels, within and outside of the UK

Director of the Manchester Institute of Innovation Research at the Alliance

Manchester Business School, University of Manchester, UK His fields of

expertise comprise the governance of science, technology and innovation

systems, responsible research and innovation, and the analysis and

con-ceptual development of research, technology, development and innovation

policies and instruments, including demand- based innovation policy, the

internationalisation of science, technology and innovation policy and

cor-porate innovation strategies He has advised the European Union, OECD

and a range of governments and has led projects for numerous international

funding bodies He was elected into the German National Academy of

Science and Engineering in 2013

Manchester Institute of Innovation Research, University of Manchester, UK

Over more than 15 years he has taught and researched a wide range of science

policy issues He has special interests in the roles science and technology

play in local and regional economic development and in the implications of

an increasingly globalised scientific enterprise for national science policies

He has also written on policy dynamics, including work on rationales for

science and technology policies and on the implications for innovation policy

analysis of taking the ‘policy mix’ seriously He is an active commentator on

science policy issues in the specialist press, on social media and as a

contribu-tor to the Guardian newspaper’s science policy blog.

at the Manchester Institute of Innovation Research at the Alliance Manchester

Business School, University of Manchester, UK His research focuses on

the formulation, evaluation and impact of science and innovation policy as

well as the management, economics and governance of emerging

technolo-gies He was a member of the leadership team of the project underpinning

this Handbook He also teaches at undergraduate, post- graduate and

execu-tive levels, including directing the Execuexecu-tive Short Course on Evaluation

of Science and Technology Policies He holds a PhD in Innovation Studies

(Manchester), an MSc in Science and Technology Policy Studies and a BSc in

Economics Prior to joining the Manchester Institute of Innovation Research,

he worked at the Scientific and Technological Research Council of Turkey

(TUBITAK)

European Work and Employment Research Centre (EWERC) at the Alliance

Manchester Business School, University of Manchester, UK His research

interests include non- standard employment, minimum wages, low- wage

service work and procurement practices Sponsors of his research projects

include the European Commission (DG Employment), the International

Labour Organization, the Equality and Human Rights Commission, and

EuroFound

Science and Technology (MCST) since 1989 in various capacities Until 2011,

she was Director of Policy, Strategy, FP7 and International with core

responsi-bility for National Research and Innovation Strategy and Foresight She

coor-dinated and participated in European Union Framework Programme projects

on foresight and research and innovation policy She retains a part- time

con-sultancy role with MCST on strategy, policy and foresight and an expert role

on foresight with European Training Foundation She is active at European

and international levels as adviser, reviewer and expert group member

Research and the European Work and Employment Research Centre at the

University of Manchester, UK, and is Visiting Research Fellow, Facultad de

Educación, the University of Salamanca, Spain An economics graduate of

the London School of Economics, she completed her master’s and doctoral

studies at the University of Manchester Her research interests include the

political economy of technological change and innovation and the

relation-ship between work, learning, skills and training in new and emerging

tech-nology areas She is the author (with Bob Miller) of Innovation Diffusion in the

New Economy: The Tacit Component (Routledge, 2008)

Christian Köhler is a researcher at the Centre for European Economic

Research (ZEW), Mannheim, Germany He works in the Department of

Industrial Economics and International Management His research focuses

on microeconometric analyses of innovation behaviour at the firm level,

including determinants of R&D investment such as public funding,

competi-tion and vertical relacompeti-tionships He has contributed to numerous studies in the

area of innovation policy for the European Union and national governments

des Ponts, IFRIS), France, and a Professor with the Manchester Institute of

Innovation Research at the Alliance Manchester Business School, University

of Manchester, UK His research interests are in emerging sciences and

break-through innovation and in research and innovation policies Recent work on

the former focuses on market construction, while policy- oriented work deals

with new evaluation approaches for assessing the societal impacts of public

research, and the development of ‘positioning indicators’ in a distributed

European research infrastructure

School, and a member of the Manchester Institute of Innovation Research,

University of Manchester, UK He has broad research interests in the area

of innovation studies, focusing on such issues as the economics of

innova-tion, innovation management, growth of knowledge, health innovation and

university–industry dynamics He has published numerous articles and book

chapters and co- edited three books on issues related to innovation

Research (ZEW), Mannheim, Germany, and Deputy Head of ZEW’s

Department of Industrial Economics and International Management His

main research interest is innovation economics, with a focus on innovation

activities of firms and technology transfer between industry and academia

He directs the German innovation survey as part of the European Union

Community Innovation Survey Programme He has been involved in a large

number of innovation policy studies at national and European levels,

includ-ing evaluations of R&D support programmes

Innovation Research at the Alliance Manchester Business School, University

of Manchester, UK He read history at Cambridge and then completed a PhD

at the University of Manchester on public policy design and evaluation His

research interests extend around the whole policy cycle from design and

development through implementation to evaluation and impact assessment

His work is concerned mainly with innovation policy programme design and

with science policy, where he uses bibliometric methods to examine the

inter-action between research performers and funding bodies He has led a number

of high- profile studies, including the DG Enterprise study on the feasibility of

European Union support to the procurement of innovation in 2010 and, more

recently, the evaluation of the UK SBRI for Innovate UK in 2015

Philip Shapira is Professor of Innovation, Management and Policy at the

Manchester Institute of Innovation Research at the Alliance Manchester

Business School, University of Manchester, UK, and Professor of Public

Policy at Georgia Institute of Technology, Atlanta, USA His interests include

science, technology and innovation management and policy, the analysis

and governance of emerging technologies, regional innovation, and policy

evaluation He co- edited The Theory and Practice of Innovation Policy: An

International Research Handbook (Edward Elgar, 2010) and chaired the US

National Academies Panel on 21st Century Manufacturing: The Role of the

Manufacturing Extension Partnership Program of the National Institute of

Standards and Technology (2013) He is a Fellow of the Royal Society of Arts

a member of the Manchester Institute of Innovation Research, University of

Manchester, UK Her research activities focus on: regional science and

inno-vation policy; spatial dimensions of knowledge and innoinno-vation; evolutionary

approaches to public policy, universities and regional development; and the

innovation impact of public procurement She has published in leading

jour-nals in economic geography, innovation studies and management, including

Research Policy , Technovation and Regional Studies.

Associate at Georgia Tech’s Enterprise Innovation Institute and an adjunct

with the School of Public Policy at Georgia Institute of Technology, USA Her

research focuses on technology- based economic development, emerging-

technology assessment, manufacturing competitiveness, regional

innova-tion clusters, and innovainnova-tion and knowledge measurement and evaluainnova-tion

She is a founder of the Georgia Tech Program in Science, Technology, and

Innovation Policy and serves as Co- Principal Investigator of the Center for

Nanotechnology in Society at Arizona State University, USA

The Rt Hon Lord Willetts

I welcome this review of innovation policies, especially as its origins lie with

a challenge I put as minister responsible for technology and innovation in

the UK: ‘Given that there are now so many initiatives across the world to

promote innovation, what is the evidence about the ones that work and the

ones that don’t?’

This Handbook is an important attempt at answering that question

Perhaps inevitably the conclusions so far are limited Investing in relevant

skills, both formal and informal, looks helpful So does advice and

network-ing support for firms Technology foresight exercises can be useful, especially

when guiding the use of specific instruments in specific priority areas Tax

reliefs may promote incremental improvements but are not so good at

stimu-lating radical innovation

There are however major uncertainties Short- run gains do not

neces-sarily point to more transformational long- term effects And, vice versa,

long- term benefits may not show up in the short run Most profoundly there

are uncertainties about the effects of combining different policies – does that

yield a kind of multiplier effect or do they just get in each other’s way and

overlap with little extra impact? This is an area where we know little as yet

Often innovation policy is linked to universities, which are at the heart

of all the great innovation clusters But despite their commitment to research

it does not look as if innovation policy itself is always studied by

universi-ties with the rigour and sophistication one would expect It is thus pleasing

to see the Manchester Institute of Innovation Research at the University of

Manchester take up this challenge

One of the most valuable features of this Handbook is the typology of

dif-ferent kinds of innovation policy This on its own is a contribution to good

policy making by helping us to think through what we are trying to do and

why And when policy makers gather at discussions promoted by the OECD

and others this Handbook means there will be a framework within which

ini-tiatives can be compared and contrasted

Above all, policy makers themselves should keep trying As a minimum

their experiences provide evidence from which others can learn And

some-times things really do work My experience has been that we exaggerate what

governments can do in the short run but underestimate what they can do in

the long run I believe this may be true of innovation policy

The Rt Hon Lord Willetts was Minister for Universities and Science in the

UK 2010–14

Writing and publishing this Handbook of Innovation Policy Impact has been

an extensive and collective exercise In addition to the authors of each of

the chapters, we need to profoundly thank many others for their valuable

contributions

Our acknowledgements start with Nesta, the innovation charity that was our initial sponsor Stian Westlake and Kirsten Bound at Nesta raised

the need for a broad, systematic account of innovation policy impact which

could be used by academics and policy makers alike Together with Albert

Bravo- Biosca and Jen Rae, our Nesta colleagues were encouraging yet

criti-cal friends throughout the entire study, reflecting and commenting on our

findings Nesta also helped to mobilise the wider stakeholder community,

hosting three stakeholder workshops in London We are deeply grateful for

the encouragement and support of Stian and his team

Throughout the writing of this Handbook, we received exceptional

support from an advisory committee The advisory committee comprised

Luke Georghiou (University of Manchester), Mark Glover (formerly of

the Technology Strategy Board), Fergus Harradance (formerly with the

Department for Business, Innovation and Skills and now at Her Majesty’s

Treasury), Mark Franks (Department for Business, Innovation and Skills),

Michael Keenan (OECD) and Stefan Kuhlmann (University of Twente) These

advisers commented on our approach, methodology and selection of policy

instruments from both academic and policy- making perspectives Most

notably, they provided feedback on initial drafts of the policy instrument

chapters and discussed them in workshops Advisory committee comments

proved to be indispensable, and we are deeply grateful

We greatly appreciate the numerous academic colleagues and policy makers who participated in four workshops at which draft versions of the

Handbook were discussed We also thank academic colleagues and other

stakeholders who commented on various presentations about the Handbook,

including at sessions of the 2014 Conference of the European Forum for

Studies of Policies for Research and Innovation (EU- SPRI), the 2015 Globelics

International Conference, and the 2015 Atlanta Conference on Science and

Innovation Policy These discussions improved the chapters and also

con-firmed great interest in the innovation and policy communities for the

com-prehensive treatment of evidence on impact offered in the Handbook.

We further thank Evgeny Klochikhin, Carlos Ramos Perez, Jaime Humberto Sierra Gonzalez and Omid Omidvar These were all doctoral stu-

dents at the Manchester Institute of Innovation Research during the

prepa-ration of this Handbook We much appreciate their support in our search for

existing evidence and in addressing the technicalities involved

It must be stressed the book would not have seen the light of day

without the tremendous efforts and diligence of Kalle Stahl- Nielsen and

Kathryn Morrison, who shouldered the burden of editing and formatting the

Handbook These tasks were performed with great patience and

understand-ing for their academic colleagues

We extend our appreciation to David Willetts, UK Minister of State for

Universities and Science 2010–14, for contributing the Foreword and for his

continued interest in our activities

Finally, as the book editors we sincerely thank all the chapter authors for

their creativity, endurance, resilience and critical discussions throughout the

entire process The authors retain responsibility for each of their chapters,

while any further errors and shortcomings in the Handbook are attributable to

the editing authors and not to any of the colleagues named above

Jakob Edler, Paul Cunningham, Abdullah Gök and Philip Shapira

Manchester, December 2015

Abbreviations and acronyms

€ euro

ANVAR French innovation agency (now incorporated into OSEO)

BERD business expenditure on research and development

BERR Department for Business, Enterprise and Regulatory Reform

(UK) (now replaced by BIS)BIS Department for Business, Innovation and Skills (UK)

CIR French research tax credit

EUREKA intergovernmental organisation for pan- European R&D funding

GERD gross expenditure on research and development

HRST human resources in science and technology

IFRS International Financial Reporting Standards

ISO International Organization for Standardization

KIBS knowledge- intensive business services

NASA National Aeronautics and Space Administration (US)

NTBF new technology- based firms

OECD Organisation for Economic Co- operation and Development

OSEO French organisation for growth and innovation in SMEs

R&D research and development

RDA Regional Development Agency (UK) (abolished 2012)

RDI research, development and innovation

SBDC Small Business Development Center (US)

SBIR Small Business Innovation Research (US)

SME small and medium- sized enterprise

STI science, technology and innovation

Tekes Finnish Funding Agency for Innovation

UI university–industry

1 Introduction: Making sense of innovation policy

Jakob Edler, Abdullah Gök, Paul Cunningham and

Philip Shapira

1.1 AIM OF THIS HANDBOOK

‘Wouldn’t it be good if we knew what works in innovation policy, to inform

future policy making and be more efficient and effective in designing future

innovation policy instruments as a result?’ This earnest request was made

recently by a senior UK politician Yet, in the last 20 years, hundreds of

evalu-ations and academic studies have been conducted on a wide variety of

inter-ventions that, by various means, have an impact on innovation input, output,

processes, practices and capabilities So why is there still uncertainty about

this area and what does this new volume add to our understanding of the

impacts of innovation policy?

Previous studies have mainly focused on specific policies, programmes

and projects – to assess their past performance and, in some cases, to

improve their future design and implementation The editors and authors

of this book have themselves performed numerous evaluations, and have

also tried to learn from existing evaluation evidence Over the years,

however, we have realised that, in the policy making community, learning

from existing evidence has its limits It can often be introspective, drawing

lessons from one’s own activities and evaluations or from a limited number

of narratives labelled as ‘best cases’ Less common in the academic and

policy making communities are systematic attempts to take advantage

of the numerous existing evaluations of innovation policy instruments

Furthermore, academic studies tend to highlight the specific contribution

that their method or data makes, rather than producing systematic

compari-sons or syntheses of the effects of policy instruments For these reacompari-sons the

idea of a structured effort to learn from the extensive array of evidence on

innovation policy impact both fills a gap and offers promising opportunities

for new insights

Supported by Nesta,1 an international team of innovation policy experts,

led by the editors of this book, turned our politician’s request into something

tangible Between 2011 and 2013, we conducted a study to gather and

syn-thesise the most relevant and recent evidence on the impact of innovation

policy measures The study was titled the Compendium of Evidence on the

Effectiveness of Innovation Policy.2 This Handbook presents the result of that

study In line with practice in the international evaluation and academic

lit-erature, in this Handbook we use the terms policy ‘instrument’, ‘intervention’

and ‘measure’ interchangeably

The contributions to the Handbook present a unique and systematic

analy-sis of secondary evidence on the impact of interventions in innovation policy

Such analyses are not new – they are found in the areas of education, health

and international development amongst others Recently these have been

further reinforced with systematic attempts such as the UK government’s

‘What Works?’ centres, which aim at analysing and sharing existing evidence

(Halpern et al., 2014) While the practice of systematically collecting and

ana-lysing existing evidence is well established in other policy areas, it has been

underdeveloped in innovation policy

The idea of comprehensive secondary analyses in innovation policy was first proposed in the late 1990s (Georghiou, 1999) It took a further

decade to elaborate an operational framework (Edler et al., 2008) This

framework distinguishes two types of secondary analyses The first is

meta- analysis, whereby primary data from different studies is pooled and

analysed, improving robustness and validity Given the idiosyncrasies

of policy interventions, their contexts and uneven data availabilities, it

remains difficult to undertake meta- analysis in innovation policy The

second approach is evaluation synthesis, which systematically compiles,

qualitatively analyses and interprets the findings of existing studies, taking

into account differences in contexts and methods, thus allowing in- depth,

yet contextualised, learning

To provide contextualised learning from existing evidence, this Handbook

pursues an evaluation synthesis approach3 which allows the reader to

obtain a thorough overview of a broad range of policy instruments and their

effects The main aim of this book is to provide the opportunity for critical

reflection and – we hope – enlightened policy learning for policy makers,

academics and all innovation policy stakeholders It offers an entry point for

those who seek specific support in designing and implementing innovation

policy instruments and aims to foster academic debate about policy

ration-ales, intervention logics and the opportunities and limits of analysing and

understanding impact

In order to introduce the reader to the wealth of evidence in this

Handbook, and to provide some guidance for the interpretation of its

find-ings, this introduction presents our understanding of innovation policy

(section 1.2) and rationales (section 1.3) This does not, however, include

the specific intervention rationale for each instrument or the mechanisms by

which they exert impact – this is done in the individual chapters Since most

of the existing evidence and analysis is based on individual instruments,

section 1.4 reflects on the nature of policy instruments and their impact We

present a typology of innovation policy instruments to systematise the

evi-dence and allow distinct entry points for readers interested in different kinds

of instruments A short explanation of our methodology is then provided

(section 1.5) This is followed by an explanation of the structure of the book

(section 1.6) The introduction closes with reflections on what we believe

are important conditions for the interpretation of the findings of this book

(section 1.7)

1.2 INNOVATION POLICY – DEFINITION AND DELINEATION

For the purpose of this Handbook, we define innovation policy as public

inter-vention to support the generation and diffusion of innovation, whereby

an innovation is a new product, service, process or business model that is

to be put to use, commercially or non- commercially Innovation policy, as

we delineate it, is intervention that is designed and administered by

gov-ernment, including multiple agencies at various spatial levels We do not

include  private, corporate policies or strategies for innovation within this

definition, although organisations that originate these are often the targets of

innovation policies

Our definition includes innovation generation, market introduction and

dif-fusion The generation of innovation involves the production of underlying

knowledge, artefacts and practices that are needed to produce something

novel Thus innovation policy overlaps with and is linked to science, research

and technology policy Often the distinction is not straightforward, and some

of the instruments discussed in this book would also qualify as technology or

research policy However, measures are included under our umbrella

defini-tion if they are designed to develop artefacts and models for the marketplace,

rather than being restricted to the production of underlying knowledge or

technology We also include the introduction and diffusion of innovation in

our definition, since the bottleneck for innovations is often not their design

and development, but their absorption by users Therefore, what qualifies

a ‘policy intervention’ as an ‘innovation policy intervention’ is its purpose

to provide support to the process of the generation, introduction, diffusion,

adoption and use of novelties

In our understanding, the target groups of innovation policies are in

prin-ciple all those actors who generate innovations from the supply side and also

those who ask for, absorb and use innovations from the demand side As

discussed further below, we realise that this distinction is somewhat artificial;

while the target groups of innovation policy instruments on the supply side

will mainly be companies as the prime generators of innovations, supply- side

policy instruments often also incentivise companies to link with public sector

organisations or other users of innovations Equally, demand- side measures

often support the linkage between supply and demand and have systemic

effects on markets more broadly

The locus of the design and implementation of innovation policy is

varied Innovation policy is frequently designed and implemented through

ministries or agencies explicitly responsible for the ‘economy’ or for

‘inno-vation’, whereby the division of labour between such bodies differs

accord-ing to national and regional contexts However, two additional important

points must be highlighted First, as already mentioned, the delineation

between innovation policy and science, research and technology policy

is blurred Second, many of the measures that support innovation are

designed and implemented by functional ministries or agencies (such as

energy, health, or transport) and as such are not labelled innovation policy

measures; rather they serve the purpose of supporting innovation as a

means to achieve an ultimate policy goal This we label functional innovation

policy An example would be a subsidy for photovoltaic installations which

catalyses the absorption and diffusion of this technology to contribute to

a reduced national carbon footprint Most of these functional innovation

policies target the diffusion of innovative solutions and thus address the

demand side As this Handbook considers the demand side as an important

dimension of innovation policy, we have included numerous examples of

such functional innovation policies Similarly, industrial policies that aim

to support selected sectors of the economy also use a raft of sectoral

inno-vation policy measures Consequently, this book does not apply a strictly

institutional approach to innovation policy; that is, while it focuses on

measures that are labelled ‘innovation policy measures’ and which are

designed and implemented in dedicated innovation ministries or agencies,

it is not limited to these We pursue a functional understanding and thus

include policy instruments that fall within our definition of innovation

policy

1.3 INNOVATION POLICY RATIONALES AND THEIR LIMITS

Why should public policy intervene in the process of producing and

dif-fusing innovation? Academics and policy makers have developed and

adopted various ways to conceptualise and justify intervention in the

innovation process.4 Each chapter of this Handbook outlines the

interven-tion rainterven-tionale for the specific group of instruments it covers However,

among  the plethora of instrument- specific intervention rationales, we

can  identify  three  broad clusters that set the theoretical scene for this

Handbook

The first rationale is based on market failure This assumes the existence

of a market equilibrium and optimal level of inputs, outputs and activities,

with technology being an exogenous factor (Laranja et al., 2008) Policy, in

this thinking, has to intervene if market failures occur that would lead to sub-

optimal levels of knowledge and innovation generation to achieve that market

equilibrium (see for example Metcalfe, 1995; Metcalfe and Georghiou, 1998)

This is the classical justification theoretically underpinned by Nelson (1971)

and Arrow (1962) The main argument rests on appropriation asymmetries,

that is, that the benefits of scientific knowledge, as a major input for

innova-tion and as a public good, can and will be used, not only by the knowledge

generator, but by other actors (externalities) The creator of knowledge cannot

appropriate all its benefits alone, which leads to a disincentive to optimal

knowledge production, as private returns are lower than public returns The

need for public policy, therefore, is to provide for knowledge production in

public organisations, to financially support knowledge production and

inno-vation activities in firms and start- up activities and to help protect intellectual

property to incentivise private knowledge production and exploitation (the

temporary monopoly function) Further, market failure occurs through

infor-mation asymmetries, while innovation follows from – and leads naturally

to – information asymmetries, which can also hamper investment in

innova-tion generainnova-tion and absorpinnova-tion (Metcalfe, 1995) Market failures can occur

on the supply side (generation of basic knowledge) and on the demand side

(e.g learning externalities in early adoption) and at the interface of the two

(information asymmetries)

The second dominant school of thought follows an innovation systems

approach that is rooted in evolutionary economics This is not reliant on the

existence of an equilibrium in the market, but rather conceptualises

inno-vation as an interplay of system components within specific framework

conditions, whereby the generation of knowledge and innovation is

char-acterised by constant interaction and learning (Lundvall, 1988, 1992; Smits

et al., 2010) This has been more recently linked to the idea of ‘functions’ of

innovation policy (Bergek et al., 2008; Hekkert and Negro, 2009) and ‘policy

problems’ in innovation systems (Edquist, 2011; Borrás and Edquist, 2013)

A key assumption of these approaches is that policy must intervene in order

to support those system functions that do not perform to a level regarded

as sufficient To that end, policy needs to be supported through an analysis

of ‘problems’ (Edquist, 2011) in the system, and policy intervention follows

an assessment of whether a certain problem can actually be tackled through

policy intervention Therefore, in this volume we speak of system failures as

a policy intervention rationale if the functional performance of the system

to create and use innovation at a rate that is deemed socially desirable is

limited through:

● existing legal, regulatory and financial conditions for generating and

diffusing innovation;

● inadequate capabilities in a system;

● insufficient exchange, interaction and cooperation (Klein Woolthuis

et al., 2005)

Under these conditions, the system and its actors need to be provided

with appropriate legal and financial framework conditions and with support

to overcome the capability and cooperation failures The main idea of

inno-vation policy is thus to support broad capabilities, exchange, cooperation

and interaction so that complementarities and specialisation can be brought

together, for the production of knowledge and innovation as well as for their

uptake by producers and users It also needs supportive and stable

frame-work conditions Again, system failures occur not only in the production of

innovation (on the supply side), but on the demand side as well (e.g ability

and willingness to adopt an innovation, to cooperate with producers or to

signal a need to the market)

A last rationale is based on the idea that science and innovation

can contribute to addressing societal missions and challenges In some

countries, this has conditioned the organisation of innovation policy

A major example is the USA, where innovation policy is often linked

to specific policy objectives and designed and implemented by

depart-ments responsible for those specific missions Thirty years back already,

Ergas (1987) has labelled such systems as ‘mission- oriented’ in contrast

to ‘diffusion- oriented’ systems in which innovation policy is organised to

upgrade the innovation capabilities and system conditions for innovation

horizontally, across the system.5 In the last ten years or so, the strategic

discourse and orientation in innovation policy across the OECD countries

have somewhat shifted towards more mission or challenge orientations

(Gassler et  al., 2008; Mahroum, 2012; Weber and Rohracher, 2012) The

underlying argument is that it is a primary duty of the state to provide

direction for technological development and innovation in order to satisfy

state needs (e.g defence, security) and citizen needs (health, education,

etc.), take risks and help to create the kinds of markets that are societally

preferable (Mazzucato,  2011) Thus,  policy support incentivises actors to

invest in knowledge and  innovation production in targeted areas with a

specific need in mind

While we have outlined three general rationales for innovation policy, reality is typically more multifaceted Clearly, the linkage of rationales to

policy intervention is complex and policy interventions will often draw on

a mix of these rationales For example, the decision on which instruments

should be employed to steer innovation (mission orientation) will be based

on considerations about the existence and nature of underlying system or

market failures In addition, the concept of intervention rationales suggests

a mechanistic understanding of diagnosis and therapy based on a theory

that applies only to a limited degree in a complex and dynamic innovation

policy context (see also Laranja et al., 2008) Rationales are based on models

of innovation systems and innovation processes that are necessarily

sim-plifications and standardisations of complex, idiosyncratic processes, and

thus can never entirely fit a given situation Furthermore, policy makers

have bounded rationalities; they are often unable to obtain the information

and knowledge needed to comprehensively contextualise the instrument

and define the failure or problem they seek to address (Linder and Peters,

1989,  p 41) We also cannot be sure that policy makers who take certain

rationales for granted and use them to justify their intervention fully grasp

their meaning or have the necessary data and strategic intelligence to be

able to assess whether certain rationales are justified for their situation

Innovation policy rationales, therefore, are often ex- post rationalisations of

interventions or simplified ex- ante justifications in the face of complexity

This is exacerbated by the fact that policy design and implementation is a

collective process, and the understanding of appropriate rationales, even

the understanding of the theoretical nature of a rationale, tends to differ

between actors (political decision makers, designers, implementers) and

may change over time As Laranja et al (2008) convincingly argue, policy

design and implementation are driven by a number of factors beyond a

theoretically derived intervention rationale

1.4 THE NATURE OF POLICY INSTRUMENTS AND THEIR

IMPACT

1.4.1 Different Understandings of the Nature of Policy Instruments and

Biases in Existing Evidence

As this Handbook is organised around sets of actual policy instruments, we

need to critically discuss the concept of a policy instrument The existing

evidence underpinning this Handbook is itself based on particular

construc-tions of what a policy instrument is, how it is designed and implemented and

what effects it has To assist readers in interpreting the synthesis and lessons

we provide, we need to briefly step back and reflect on the nature of policy

instruments and their role

The political science literature has defined – broadly speaking – three

dominant understandings of instruments over time (van Nispen, 2011)

In the 1970s, the ‘classical’ (van Nispen, 2011, p 1) approach conceived of

instruments as policy mechanisms for goal- attainment This included

‘tra-ditional functionalist’ perspectives which defined instruments as ‘the set of

techniques by which government authorities wield their power in

attempt-ing to ensure support and effect or prevent social change’ (Vedung, 1998,

p 21) In this top- down view, governmental actors have the information and

absorptive capacity to identify the gap, understand the cause, design the

appropriate instrument and implement it without distortion Governmental

actors are in charge at every step of the process, instruments are selected on

the basis of their specific characteristics to tackle the gap and their

imple-mentation is largely a technical and mechanistic matter and largely context

insensitive once the specific gap is identified Finally, if designed and

imple-mented appropriately, the instrument will be effective; that is, it contributes

to achieving its aim, which is to close the identified gap

This school of thought was subsequently modified, based on the

realisa-tion that it is not the characteristics of the instrument alone that determine

the effectiveness of the instrument, but the context and process of its

imple-mentation (‘the instrument- context approach’; van Nispen, 2011, p 2) In

other words, while an instrument is still understood as a technical device

or a tool that can be applied to tackle problems, its performance will differ

strongly from context to context Therefore, one instrument with a similar

delivery structure, tackling a problem of the same nature, might still perform

very differently in different contexts, given the variability of actor landscapes

and capabilities and the interplay with other existing instruments, broader

framework conditions or broader socio- political dynamics

A more recent, sociological school of thought takes it one step further

(Lascoumes and Le Gales, 2007) According to this approach, an instrument

is, by definition, only one of a whole range of variables that intervene in the

system and affect target groups Here, the implementation structure and

process of the instrument are seen as major factors, with implementation

agents creating their own specific understanding of the problem, the target

group and the mechanisms of the instruments In this sociological view,

instruments are not technical, neutral devices that can be selected and

imple-mented to solve a ‘given’ problem Rather, they are to be conceived as

‘institu-tions’ (Lascoumes and Le Gales, 2007) that structure collective action and that

must be actively constructed or adapted In this understanding, instruments

represent certain normative and causal ideas, as part of a broader policy

(Borrás and Edquist, 2013), that are changeable in the process of adoption and

implementation (Linder and Peters, 1989; Lascoumes and Le Gales, 2007) In

this social and interactive process, problems may be redefined and

instru-ments adjust their nature and create their own dynamics that necessarily are

different with each application, between contexts and over time Moreover, in

this perspective, instruments and the way they are implemented are a result

of political processes, with the instrument being ultimately a manifestation of

a dominant worldview and influence These political processes are thus not

independent of power and interest They involve the principal (the owner of

the instrument, with ultimate responsibility), the agent (responsible for

imple-menting), individuals and specific groups representing principal and agent,

and interested stakeholders more generally, all with their own initial views

and interests In sum, instruments are not understood simply as neutral tools

to be used for specific, clearly defined problems, but as social constructs that

are often contested and highly context dependent, that are linked to

the dom-inant problem definition and which develop their own dynamics over time

This more complex and less technical or functional understanding of instruments as institutions has become more relevant, it appears, with shifts

from hierarchical, top- down governance to interactive, collective governance

(Borrás and Edler, 2014) and the enormous growth in the number of

instru-ments In what Salamon (2000, p 1612) calls a ‘revolution’, the number and

variety of policy instruments in all fields have greatly increased and, more

importantly, their associated delivery structures and logics have changed

Governmental action to design and implement instruments has been

com-plemented by governance processes whereby the design and implementation

not only are often the collective actions of a principal (e.g a ministry) and an

agent (e.g a funding agency), but also involve other stakeholders, such as

interest organisations, private consultancies, network managers, firms and

universities These stakeholders take greater responsibility for instrument

design and implementation and increasingly, in the case of public–private

instruments, for the co- funding of collective action While many instruments

covered by this Handbook are still dominated by governmental actors, an

increasing number are implemented in concert with stakeholders (e.g

clus-ters, networks, pre- commercial procurement, and foresight exercises) This,

in turn, makes a critical appraisal of the mechanisms by which those

instru-ments are designed and implemented more compelling, and the increase in

complexity and idiosyncrasy at the same time confounds the transferability

of findings regarding their impact

While many analysts, evaluators and policy makers would agree that this

is an overly simplified model of how things actually work (e.g Borrás and

Edquist, 2013), the majority of evaluations and impact analyses examined in

this Handbook follow a conventional, somewhat functional approach, looking

at the effects of instruments as technical, neutral devices for intervention by

state actors Often, this is done in a sensible way involving a contextualisation

of the analysis and a commentary on the importance of context However,

there are multiple cases in which the contextualisation is implicit, and seldom

do we see studies that elaborate on the actor constellation and the power

games that influence instrument design and implementation, on the changes

in its meaning during implementation or on the role of non- state actors in

design and implementation

1.4.2 Limitations of Impact Analyses

We also have to stress limitations as regards the notion of impact Modifying

the concept of Reale et al (2014, p 37), we can define the impact of an

instrument as the change that can be wholly or partially attributed to it

Conceptually, impact can be intended or unintended and it can be expected

or unexpected (Reale et al., 2014) Much of the existing evidence on impact,

especially from the evaluation studies examined in our analyses, focuses

largely on intended and expected impacts, that is, on assessing performance

against the initial goals of an instrument In other words, while this Handbook

strives for a broad understanding of innovation policy impact, existing

impact evaluations (and many academic studies) are often reduced to the

examination of goal attainment, albeit differentiated for different types and

levels of goals While the chapters in this book report on and analyse impact

as broadly as the underlying data allows, there frequently remains a certain

restriction to goal attainment This is a conclusion we discuss in the final

chapter but which needs to be stressed already here in the introduction

1.5 INSTRUMENT CLASSIFICATION, DATA AND METHODS

1.5.1 Our Classification of Innovation Policy Instruments

One of the first challenges in compiling a synthesis of evidence on

innova-tion policy is to organise the evidence base systematically into manageable

and logical subsets through a classification process There are numerous

logics for classifications of policy instruments (Salamon, 2000; Hood, 2007;

Lascoumes and Le Gales, 2007), and each classification must be appropriate

to the analytical purpose it is intended to serve (Salamon, 2000; Hood, 2007)

While a number of typologies exist for innovation policy based on political

priorities or other ad hoc considerations, they are not suited to the purpose

of this Handbook, which is to give academics and policy makers an overview

of existing evidence to enable them to learn about the impact of instruments

according to their own specific policy objective Existing typologies of

inno-vation policy instruments did not fit our objective- oriented purpose and

tended to be overly complicated, mainly because they reflect the complex

political priorities at the time they were devised For instance, the European

Commission’s (EC) ERAWATCH and Trend Chart policy framework presents

37 types of innovation policy instruments at three levels.6 Not all categories

were relevant for our purpose, and some appear very specific and represent a

small niche of instruments, typically reflecting prior political priorities Other

typologies are not organised according to intervention goals or target groups,

but according to the different modes by which they influence their target

group For example, the three- fold typology of Borrás and Edquist follows

a well- established tradition in political science that distinguishes between

regulatory instruments, economic and financial incentive instruments, and

‘soft’ instruments Since this does not allow the classification of instruments

according to policy goals, the authors create an elaborated matrix of ‘policy

problems’ and instruments in their three- dimensional classification (Borrás

and Edquist, 2013)

For the classification7 used in the Handbook, we start from the standing that many readers will enter from the perspective of a policy problem

under-or a policy goal; that is, policy makers and other stakeholders would wish to

look at prior experience and evidence to better understand how they could

achieve a specific policy goal Thus, we identified seven major innovation

policy goals within the realm of our innovation policy definition.8 On the basis

of these, we identified and allocated instruments that are perceived, in the

existing literature and evidence, as being the most typical and critical for

achieving these goals

We then distinguished between those instruments that target

prelimi-narily producers of innovation (i.e intervene at the supply side) and those

that target (potential) users of innovation (i.e intervene at the demand side).9

Innovation policy is geared towards generating and diffusing innovations,

but, as also discussed earlier, this can be done by supporting or influencing

the supply side, the firms themselves, or the demand side and the context in

which firms operate (Edler and Georghiou, 2007) In the former, those

gen-erating innovation are supported to do things differently, that is, to innovate

more quickly, be more interactive, or do so with different kinds of partners

In the latter, public and private actors are supported in order to formulate

demand for innovations and be in a better position to apply them We realise

that the distinction between supply- and demand- side policies can be a crude

one, as many interventions are based on policy failures (see below) that arise

out of a lack of interaction between demand and supply and stem from

coor-dination problems in the interplay of demand and supply, across markets

(Bleda and Del Rio, 2013) However, we have chosen to categorise policy

interventions along demand and supply side because most interventions

initially target actors in their primary function to either generate or demand

and use innovation Therefore, this Handbook includes both instruments to

support the supply side and those supporting (potential) demanders for

innovation and – as far as the available evidence permits – the deliberate

combination of the two Table 1.1 depicts our taxonomy

Table 1.1

Chapter number and instr

This goal- driven approach also allows the inclusion of a range of

instru-ments that are not primarily geared towards improving innovation

capabili-ties and activicapabili-ties, but – as introduced above – are designed for other policy

goals and can also affect innovation behaviour and performance at the same

time This is especially true for instruments that are geared towards

improv-ing the skills base and demand conditions for innovation Although such

pol-icies were rarely designed for or evaluated against their impact on innovation

activities per se, it is important to discuss innovation policy in such a broader,

inclusive approach, to overcome any narrow compartmentalisation In

addi-tion several instruments may serve multiple goals Thus, while the individual

chapters deal with broad groups of instruments and goals, the allocation of

instruments is generally based on their primary goal and their discussion is

limited mainly to a single chapter Since this taxonomy cannot take account of

the potentially important role of the interplay of instruments, the book

con-tains a specific chapter on policy interrelationships and mixes of instruments

1.5.2 Approach and Underlying Data

This Handbook consists of 18 chapters, including this introduction Fifteen

chapters are devoted to syntheses of the evidence on categories of innovation

policy instruments, one chapter considers evidence on policy mixes and

inter-actions, and a concluding chapter synthesises the findings and reflects on the

quality, availability and appropriate use of evidence The chapters that

con-sider specific innovation policy instruments all follow the same basic

struc-ture: they start with a broad discussion of scope of the instruments, continue

with an overview of their specific rationales, comment on the underlying data

sources, provide a detailed synthesis and analysis of the existing evidence and

finally reflect on the specific lessons that can be learned about the evaluation

of the instruments and on the performance of the instruments and – as far as

evidence was available – the context- specific requirement for performance

The chapters in the Handbook draw on existing available evidence After a

broad scanning exercise, sources were selected based on expert judgement of

both relevance and quality We reviewed formal evaluation reports, academic

analyses of innovation policy impact10 (primarily from peer- reviewed journals),

and other relevant documents Each chapter presents details on the search

strat-egy used In total, more than 1200 items were reviewed, 725 of which provided

evidence (216 evaluation reports and 509 academic publications providing

evidence), while 600 provided other conceptual and empirical background to

better understand the nature and setting of the specific instruments (Table 1.2

summarises the evidence used by core chapters in the Handbook.)

1.6 THE STRUCTURE OF THE HANDBOOK

The book is structured around the innovation policy goals outlined in

Table 1.1 Chapters 2 and 3 cover evidence on the effectiveness of indirect and

direct support instruments which primarily aim at increasing R&D

spend-ing Evidence on the instruments for increasing non- financial capabilities

are analysed in Chapter 4 (policies for training and skills), Chapter 5

(entre-preneurship policy) and Chapter 6 (technical services and advice) The next

set of chapters, Chapter 7 (cluster policy), Chapter 8 (policies to support

col-laboration) and Chapter 9 (innovation network policies), cover instruments

that target systemic capabilities and complementarities Instruments that

enhance the demand for innovation are discussed in three consecutive

chap-ters (Chapchap-ters 10–12) analysing instruments to stimulate private demand for

innovation, public procurement policies and pre- commercial procurement,

respectively Instruments covered in Chapter 13 (innovation inducement

prizes) encompass a number of different policy goals ranging from increasing

R&D spending to enhanced demand for innovation Evidence on the

instru-ments related to framework conditions shaping both supply and demand

are addressed in Chapter 14 (standardisation and standards) and Chapter

15 (regulation) Chapter 16 (technology foresight) discusses the evidence on

instruments that design and apply discourse approaches to define

innova-tion policies and support the communicainnova-tion between supply, demand and

policy The penultimate chapter (Chapter 17) concerns the relatively scarce

explicit evidence on policy mixes Despite the scarcity of evidence, we regard

this as a crucial chapter because it highlights the importance of further

devel-oping an understanding of the interplay of instruments, both for attempts to

deliberately design and coordinate a mix of instruments and to understand

Table 1.2 Handbook evidence sources

background and context analyses)

Academic articles (evidence- giving)

Evaluation reports

3 Direct support – firm R&D and

the inevitable influences of an instrument on others in the system Finally,

Chapter 18 synthesises the evidence and critically discusses the implications

of our findings

1.7 INTERPRETING RESULTS AND DRAWING LESSONS

Systematically collating and analysing the available evidence on the

perfor-mance of policy instruments is a useful basis for policy learning However,

it will be of limited use or even counter- productive if its limitations are not

understood Three caveats should be kept in mind when interpreting the

results discussed in the Handbook.

First, following from the above discussion on the nature of policy instruments, one has to be very careful to understand the sensitivity of

context in its broadest sense Meaningful lessons from particular policy

and programme assessments can be drawn and transferred only if specific

contextual circumstances are considered Socio- economic and ‘ instrumental’

contexts differ, as do political and administrative dynamics As far as

pos-sible, this Handbook attempts to reflect on this context sensitivity, but the

underlying evidence often lacks a thorough analysis of context and of

implementation structures and processes The evidence and insights about

innovation policy impacts identified are most useful if used to prompt and

inform reflection and to spark debate While analysts and decision makers

may draw many lessons from the significant body of work amassed in this

Handbook, lessons on ‘what works’ will only be appropriate at the level of

basic mechanisms and against an understanding of contextual differences

When interpreting insights gained from the Handbook, policy makers need to

reflect upon their own context, the objectives they seek to achieve, and the

capabilities available

A second factor influencing the generalisation of lessons relates to methodology and conceptualisation The use and interpretation of existing

evidence are highly dependent on the methods used and the underlying

theoretical assumptions of the studies synthesised The chapters in this book

illustrate that the same programme can be assessed in different ways

depend-ing on the kind of method used and the theoretical lens applied This is a

source of richness and reflection, but needs to be kept in mind when

inter-preting individual evidence on the effectiveness of policy interventions and

in drawing more general lessons

Third, the standpoint and absorptive capacity of the analyst and the reader may condition what is gleaned from existing evidence Evidence can

be interpreted differently as actors, often implicitly, apply varied theories

and assumptions about policy intervention rationales and the mechanisms

of interventions, and have different understanding of methods applied

Moreover, over time, new evidence and methods may be developed, and

circumstances evolve, turning what once appeared to be a general lesson into

a specific case of limited general value

These three caveats are, of course, common problems in policy analysis

and deliberation However, we need to re- emphasise these cautions in the

context of this volume precisely because we are convinced that it is a

valu-able reference for academics and policy makers We position this Handbook

as a source to prompt users to reflect on their own situation and problems,

to foster appreciation about the relationships and varieties of contexts and

instruments, and to consider the opportunities and limits of evaluation and

learning from evaluation in innovation policy We anticipate that, if readers

approach this book with the necessary critical mind towards transferring

lessons, this Handbook can make a major difference in their understanding of

innovation policy

NOTES

1 Nesta is a non- profit British innovation charity with ‘a mission to help people and organisations bring

great ideas to life’ See http://www.nesta.org.uk/.

2 More information about the project can be found at the project website (http://www.innovation-

policy.org.uk) Preliminary versions of the documents along with annotated references can be found

at the website, but the material produced in this Handbook is extended, updated and refined.

3 A prior effort to systematically collect and analyse evidence in innovation policy was the INNO-

Appraisal project, which collected 171 evaluations in Europe in order to understand methodological

features of innovation policy evaluations (Edler et al., 2012) This study focused exclusively on

meth-odological issues and thus fell short of a full- fledged evaluation synthesis, as it did not interpret the

actual findings However, the study provided a useful dataset, which subsequently contributed to the

evidence base included in this book.

4 See Laranja et al (2008) and Mytelka and Smith (2002) for a broader discussion of the evolution of

policy rationales and the relation between policy models and rationales and policy making.

5 Mission orientation has often been linked to very specific large- scale challenges supported by

‘mission programmes’ (Manhattan Project, Man on the Moon) (Foray et al., 2012).

6 See http://erawatch.jrc.ec.europa.eu/erawatch/opencms/research_and_innovation/.

7 We use the term ‘classification’ here because typologies should have mutually exclusive categories,

while our goal- driven approach produces overlaps between categories.

8 The advisory board of the underlying study was helpful in commenting on initial choices and

definitions made.

9 For a similar suggestion in the political science literature see Salamon (2000).

10 In the remainder of this Handbook we use the term ‘evaluation’ in a broad sense, encompassing

com-missioned studies to ascertain the effects and efficiency of policy interventions as well as broader,

academic analyses which often focus on specific aspects of an intervention Equally, we use the term

‘evidence’ when referring to the results of both commissioned evaluation and broader academic

studies.

REFERENCES

Arrow, K (1962), ‘Economic welfare and the allocation of resources for invention’, in Universities–National

Bureau Committee for Economic Research, Committee on Economic Growth of the Social Science

Research Council, The Rate and Direction of Inventive Activity: Economic and Social Factors, Princeton, NJ:

Princeton University Press, pp 609–929.

Bergek, A., M Hekkert and S Jacobsson (2008), ‘Functions in innovation systems: A framework for

ana-lysing energy system dynamics and identifying goals for system- building activities by entrepreneurs

and policy makers’, in T.J Foxon, J Köhler and C Oughton (eds), Innovation for a Low Carbon Economy:

Economic, Institutional and Management Approaches, Cheltenham, UK and Northampton, MA, USA:

Edward Elgar, pp 79–111.

Bleda, M and P Del Rio (2013), ‘The market failure and the systemic failure rationales in technological

innovation systems’, Research Policy, 42, 1039–1052.

Borrás, S and J Edler (2014), ‘Introduction: On governance, systems and change’, in S Borrás and J Edler

(eds), The Governance of Socio- technical Systems: Explaining Change, Cheltenham, UK and Northampton,

MA, USA: Edward Elgar, pp 1–22.

Borrás, S and C Edquist (2013), ‘The choice of innovation policy instruments’, Technological Forecasting and

Social Change, 80, 1513–1522.

Edler, J and L Georghiou (2007), ‘Public procurement and innovation – Resurrecting the demand side’,

Research Policy, 36, 949–963.

Edler, J., B Ebersberger and V Lo (2008), ‘Improving policy understanding by means of secondary

analy-ses of policy evaluation’, Research Evaluation, 17, 175–186.

Edler, J., M Berger, M Dinges and A Gök (2012), ‘The practice of evaluation in innovation policy in

Europe’, Research Evaluation, 21, 167–182.

Edquist, C (2011), ‘Design of innovation policy through diagnostic analysis: Identification of systemic

problems (or failures)’, Industrial and Corporate Change, 20 (6), 1725–1753.

Ergas, H (1987), ‘Does technology policy matter?’, in B.R Guile and H Brooks (eds), Technology and

Global Industry: Companies and Nations in the World Economy, Washington, DC: National Academy Press,

pp 191–245.

Foray, D., D.C Mowery and R.R Nelson (2012), ‘Public R&D and social challenges: What lessons from

mission R&D programs?’, Research Policy, 41, 1697–1702.

Gassler, H., W Polt and C Rammer (2008), ‘Priority setting in technology policy: Historical development

and recent trends’, in C Nauwelaers and R Wintjes (eds), Innovation Policy in Europe: Measurement and

Strategy, Cheltenham, UK and Northampton, MA, USA: Edward Elgar, pp 203–224.

Georghiou, L (1999), ‘Meta- evaluation: Evaluation of evaluations’, Scientometrics, 45, 523–530.

Halpern, D., D Mason, L Baynton and L Moore (2014), What Works? Evidence for Decision Makers, London:

What Works Network.

Hekkert, M.P and S.O Negro (2009), ‘Functions of innovation systems as a framework to understand

sus-tainable technological change: Empirical evidence for earlier claims’, Technological Forecasting and Social

Change, 76, 584–594.

Hood, C (2007), ‘Intellectual obsolescence and intellectual makeovers: Reflections on the tools of

govern-ment after two decades’, Governance, 20, 127–144.

Klein Woolthuis, R., M Lankhuizen and V Gilsing (2005), ‘A system failure framework for innovation

policy design’, Technovation, 25, 609–619.

Laranja, M., E Uyarra and K Flanagan (2008), ‘Policies for science, technology and innovation: Translating

rationales into regional policies in a multi- level setting’, Research Policy, 37, 823–835.

Lascoumes, P and P Le Gales (2007), ‘Introduction: Understanding public policy through its

instruments  –  From the nature of instruments to the sociology of public policy instrumentation’,

Governance, 20, 1–21.

Linder, S.H and B.G Peters (1989), ‘Instruments of government: Perceptions and contexts’, Journal of Public

Policy, 9, 35–58.

Lundvall, B.- Å (1988), ‘Innovation as an interactive process: From user–producer interaction to the

national system of innovation’, in G Dosi, C Freeman, R Nelson, G Silverberg and L Soete (eds),

Technical Change and Economic Theory, London: Pinter, pp 349–369.

Lundvall, B.- Å (1992), National Systems of Innovation: Towards a Theory of Innovation and Interactive Learning,

London: Pinter.

Mahroum, S (2012), Innovation Policies and Socio- economic Goals: An Analytic- diagnostic Framework, Working

Paper, Abu Dhabi: INSEAD.

Mazzucato, M (2011), The Entrepreneurial State, London: Demos.

Metcalfe, J.S (1995), ‘Technology systems and technology policy in an evolutionary framework’, Cambridge

Journal of Economics, 19, 25–46.

Metcalfe, J and L Georghiou (1998), ‘Equilibrium and evolutionary foundations of technology policy’, STI

Review, 22, 75–100.

Mytelka, L.K and K Smith (2002), ‘Policy learning and innovation theory: An interactive and co- evolving

process’, Research Policy, 31, 1467–1479.

Nelson, R.R (1971), ‘The simple economics of basic scientific research’, Journal of Reprints for Antitrust Law

and Economics, 3, 725.

Nispen tot Pannerden, F.K.M van (2011), Policy Instruments, http://hdl.handle.net/1765/33101.

Reale, E., M Nedeva, D Thomas and E Primeri (2014), ‘Evaluation through impact: A different viewpoint’,

fteval: Journal for Research and Technology Policy Evaluation, 39, 36–41.

Salamon, L.M (2000), ‘The new governance and the tools of public action: An introduction’, New Fordham

Urban Law Journal, 28, 1611–1674.

Smits, R., S Kuhlmann and P Shapira (eds) (2010), The Theory and Practice of Innovation Policy: An

International Research Handbook, Cheltenham, UK and Northampton, MA, USA: Edward Elgar.

Vedung, E (1998), ‘Policy instruments: Typologies and theories’, in M.- L Bemelmans- Videc, R.C Rist and

E Vedung (eds), Carrots, Sticks, and Sermons: Policy Instruments and Their Evaluation, New Brunswick, NJ:

Transaction Publishers, pp 21–58.

Weber, K.M and H Rohracher (2012), ‘Legitimizing research, technology and innovation policies for

transformative change: Combining insights from innovation systems and multi- level perspective in a

comprehensive “failures” framework’, Research Policy, 41, 1037–1047.

2 The impact of fiscal incentives for R&D

Philippe Larédo, Christian Köhler and Christian Rammer

2.1 INTRODUCTION

Offering fiscal incentives to stimulate business research and

develop-ment (R&D) has emerged as an increasingly popular policy tool over

the past decade In 2013, 27 countries belonging to the Organisation for

Economic Co- operation and Development (OECD) provided tax

incen-tives to support business R&D, up from 18 in 2004 and 12 in 1995 (OECD,

2013, p 106) Non- OECD countries, such as Brazil, China, India, Russia,

Singapore and South Africa, have also developed business R&D tax

incentives

R&D tax incentives allow a firm to reduce its tax burden (or other types

of mandatory contributions imposed by law such as social security

contribu-tions) depending on the size of – or increase in – eligible R&D activities Tax

incentives lower the costs of private R&D, though they are delivered only

after the R&D activity has been performed Tax incentives are an indirect

means of supporting R&D, in contrast to the direct government funding

of business R&D through grants or contracts The volume of government

funding through R&D tax incentives is significant and can reach a similar

magnitude to direct R&D funding (see Figure 2.1) In several countries, such

as Australia, Austria, Belgium, Canada, Denmark, France, Ireland, Japan,

Korea and the Netherlands, indirect support through tax incentives exceeds

direct funding

This chapter reviews the experience of business R&D tax incentives in OECD countries and presents findings on using this instrument for achieving

certain R&D policy objectives We begin by discussing the design features for

R&D tax incentives and their relationships with the political rationales The

flexibility of this instrument and its ability to address varied policy objectives

are highlighted (section 2.2)

Evaluations of R&D fiscal incentives have been undertaken in tiple countries We summarise the findings of these evaluation studies

mul-(section  2.3) Most of these studies focus on input additionality – the

con-tribution of the tax incentive to increased business R&D expenditure There

is less evidence on output additionality – the effects of R&D tax incentives

on innovation and economic impact No study has yet tackled behavioural

additionality – whether there are lasting structural changes in enterprise

innovation practices and societal impacts mostly linked to jobs We link the

evaluation results to features of the evaluated instruments to derive

conclu-sions about how R&D tax incentives can be designed for meeting particular

policy goals (section 2.4)

2.2 DESIGN FEATURES

The key economic rationale for public intervention is the presence of

knowl-edge spillovers from R&D activities: tax credits are one way to compensate

for this, and, by reducing the unit cost of R&D, this should promote ‘input

additionality’: increasing firm R&D efforts For OECD (2011), this dimension

is all the more important as uncertainty of results and asymmetries of

infor-mation tend to drive financial institutions away from supporting firm R&D

efforts There is an implicit rationale behind fostering ‘input additionality’:

that more R&D will drive more innovations; that more innovations will drive

better competitiveness of firms, both in their home market and even more in

exports; and finally that better competitiveness will drive more jobs, which

is the final declared aim of nearly all R&D tax credit policies Tax credits

have also been mobilised to address a growing policy rationale termed as

system failures: actors in the system do not cooperate enough, losing the

effects of such synergies The objective is to change the behaviour of actors –

encouraging more relations between university and industry, or between

large and small firms

How tax credits actually foster these objectives depends on their selected

design features This section reviews different design dimensions We address

seven key dimensions: the type of incentive selected, the approach (volume

versus increment based), the definition of eligible operations, the generosity

Indirect government support through R&D tax incentives Direct government funding of BERD

Note: * Data on the amount of tax incentive support not available.

Source: Own presentation by the authors based on data from OECD (2013, p 106).

Figure 2.1 Volume of tax incentives for R&D and direct government funding for business

R&D, 2013, as a percentage of GDP

of the tax credit, the beneficiaries, the rules of credit consumption and the

duration This offers governments endless possibilities of combinations and,

de facto, a large variety of implemented designs

2.2.1 The Type of Incentive

The first choice relates to the type of R&D tax incentive Currently, four types

of R&D tax incentives are applied:1

● Accelerated depreciation schemes for investments (machinery,

equip-ment, buildings, intangibles) used for R&D activities This has been the case in Italy, for example, which was one of the first countries to start such a scheme

● Special R&D allowances that enable firms to deduct more than 100

per cent of their current eligible R&D expenditures from their taxable income This is the case for the UK, where two levels of deduction are offered: 130 per cent for firms in general, and 175 per cent for small and medium- sized enterprises (SMEs) (2009 figures)

● Special exemptions on wage and/or social taxes for employees in R&D

activities The Dutch scheme WBSO (Wet bevordering speur- en kelingswer) allows the deduction of R&D labour costs only (for a more detailed explanation, see van Pottelsberghe et al., 2003)

ontwik-● Tax credits, which allow firms to directly deduct a specific share of their

R&D expenses from their corporate tax liabilities This type of R&D tax incentive is currently the most widespread

A further type of fiscal support to R&D that is closely related to R&D tax incentives is the so- called Patent Box A Patent Box grants a lower corporate

tax rate on profits generated from patents that are held in a certain country

Since patents are typically the result of R&D activities, the lower tax rates

represent a preferential treatment of R&D investment over other investments

The governments of the Netherlands and Belgium first introduced the Patent

Box in 2007, followed by Spain and Luxembourg in 2008 It has become a

major tool for tax optimisation by large firms (see debates about the Irish

situation)

Governments may combine different types of fiscal incentives Austria, for example, offered both an R&D allowance and an R&D tax credit, but

repealed the allowance in 2011

2.2.2 Volume versus Incremental Basis

The basis of calculation, either volume- based or incremental, is the second

major dimension defining an R&D tax incentive A volume- based scheme

allows the deduction of all eligible R&D expenditure in a given year In

contrast, an incremental scheme allows the deduction only of the increase in

R&D expenditure during the fiscal year

The latter was the initial choice made by numerous countries The central

argument was that public support is an incentive for applying more effort,

rather than a recurrent support for doing R&D, whatever the amount Such

a choice had one further critical fiscal advantage: it was easier over time to

identify fraud One should not underestimate the importance of ease of

veri-fication by the tax authorities in designing incentive schemes In the debates

of the early 1980s, this issue was central For instance, France and Germany

shared preparatory studies before the introduction of their schemes It was

anticipated that 30 per cent of firms would initially overvalue their R&D

efforts The German minister judged that this was an unacceptably high

percentage in the short term The French minister focused on the 70 per

cent of truthful beneficiaries, and the fact was, with a system based upon

increase and not volume, it was not sustainable over the long term to over- or

underestimate R&D expenditures

However, this approach was considered too complex for SMEs, and, in a

period of uncertainty, it gave rise to strong yearly variations that did not help

firms to plan This is why most systems have progressively, over a period of

time, moved toward volume- based solutions

2.2.3 The Definition of Eligible Operations for Tax Deductions

The definition of R&D differs among countries (see OECD, 2010) A

rela-tively narrow definition is to qualify all expenditures on wages related to

R&D as eligible R&D expenses, and thus the tax credit becomes an

incen-tive for investment in human capital (e.g the Netherlands) More generous

approaches add other current costs to the eligible R&D expenditure (e.g the

UK) and depreciation on capital R&D expenditure (including an option for

accelerated depreciation, e.g Australia)

The debate on the definition of R&D has developed along two

dimen-sions The first dimension relates to the harmonisation of definitions in

order to minimise ‘fiscal uncertainty’ (i.e the interpretation of R&D by

fiscal authorities) The current trend is to move towards an internationally

harmonised definition, relying on concepts used in collecting data on the

R&D expenditure of firms The main reference is the OECD’s Frascati Manual

(OECD, 2002) However, some countries have chosen wider definitions in

order to support specific sectors or types of research (e.g Belgium for green

technology or China for high- tech industries) Others extended the R&D

definition towards innovation A few countries have for instance included

the acquisition of intangibles (patents, licences, designs, etc.) in their

defini-tion (e.g Spain) More recently, some firms have argued that the Oslo Manual

would be a better reference Fiscal specialists have tended to oppose this

posi-tion because of the loose definiposi-tion provided by the Oslo Manual (OECD and

Eurostat, 2005) and the difficulty in identifying and measuring

correspond-ing expenses One direction that has been put forward by a number of firms

is to make use of the classifications used by a number of agencies (the DoD

and NASA in the US, ESA in Europe) of nine ‘technology readiness levels’

or TRLs The firms promoting this view (mostly in the aeronautics sector)

suggest enlarging the Frascati definition to include tax incentives for TRL 6

and 7, which deal with technology demonstrations up to an operational level

The second central, though often low- key, issue lies in the calculation of overheads A number of systems have chosen the simple solution of a given

percentage of all the direct costs accepted The generosity of the tax incentive

scheme largely depends on this For instance it has been calculated that, in

France, moving the percentage from its present level of 75 per cent to a level

of 45 per cent would represent a reduction of the tax credit by over 10 per cent

2.2.4 The Generosity of the Tax Credit

The generosity of the tax credit is a design element that largely determines the

cost of the measure for a country Two elements determine the generosity: the

percentage of R&D expenditure that can be deducted and the maximum

amount of tax reduction that can be claimed In addition, a tax

incen-tive system may differentiate the level of generosity by type of firm, R&D

activities, technologies, regions or sectors

For R&D tax credits, the first component refers to the percentage of R&D expenditure that can be deduced from the tax burden or contribution

This percentage differs widely between countries, from 10 per cent in Italy,

18  per  cent in the Netherlands, 20 per cent in Canada and Korea, up to

30 per cent in Spain and France For R&D tax allowances, governments have

to determine the multiplier for R&D expenditures that can be deducted from

the taxable income (e.g 130 per cent for firms in the UK)

The second component deals with the maximum amount of tax reduction that can be claimed within one year The level of the ‘cap’ selected is a central

issue when evaluating tax credits (e.g Norway; see below)

Both dimensions can be linked through thresholds While in most cases only the amount below the threshold can be taken into account by one firm, in

some countries the threshold means a change in the percentage considered A

typical example of such a combination is France: the percentage is 30 per cent

below €100 million and 5 per cent above A simulation showed that moving

the cap can have important consequences: in the above- mentioned case,

moving it down to €30 million in 2009 would have impacted upon only

40 large companies but would have represented a 16 per cent reduction in the

total cost for public authorities.2

In order to compare the generosity of R&D tax incentives, the OECD has established the B- index (see Warda, 2001; Figure 2.2) The index shows

the share of R&D tax incentives that can be deducted through an R&D tax

incentive.3 There is a wide variation in the generosity of R&D tax

incen-tives within the European Union While Germany, Finland, Switzerland and

Sweden do not offer R&D tax incentives, Portugal, France and Spain do run

quite generous schemes Some countries offer significantly higher incentives

for SMEs (particularly Canada, the Netherlands, the UK, France, Korea and

Australia)

2.2.5 The Beneficiaries

The definition of the subjects that are entitled to claim R&D tax incentives

builds the fifth major design element While one approach focuses on legal

entities, other approaches apply the concept of ‘enterprise groups’, based on

majority ownership or on ‘fiscal integration’ (a feature which enables groups

to balance the different results of their subsidiaries) In this case, considering

only legal entities might entail a vast increase of the group amount below the

cap (to take again the French situation, the parliamentary report estimated

that levels of deduction can vary for one group up to 300 per cent taking into

account different definitions)

Beyond the general lines that apply to all, many schemes tend to

differen-tiate beneficiaries They may be more generous for SMEs (which requires the

country to define what is meant by an SME), as in Canada, Japan, Norway,

the Netherlands and the UK They may be more generous for recently created

firms (as in the French case, where the percentage deducted is higher during

the first two years)

The tax credit can also privilege certain aspects of R&D activities It

can boost high- level employment by giving a higher reduction for the

salary of recently recruited doctoral holders (e.g in France) However, the

most common feature is to give a higher reduction to all expenses paid to

public  research and in particular universities Such a device that supports

industry–university collaboration then participates in fostering the linkages

Note: Tax subsidy rates for profitable firms.

Source: Own presentation by the authors based on data from OECD (2013, p 107).

Figure 2.2 Tax subsidy rates on R&D expenditures (‘B- index’), 2013

within the national innovation system One could imagine other uses, such

as fostering collaborations by firms with SMEs (though no such case has been

yet identified)

Addressing specific sectors (e.g sectors that are considered of strategic importance by the government or that face economic challenges) through

R&D tax incentives is difficult, since most fiscal laws require very clear- cut

and transversal discriminations (like age or size), while a sector is less easy

to delineate EU competition law also restricts the use of R&D tax incentives

for supporting specific sectors One solution is to identify a type of R&D

activity that is specific to a set of firms that mostly belong to the sector

tar-geted Some countries have chosen to target specific fields of R&D (such as

biotechnology or nanotechnology) or types of technologies (such as green

technologies; cf. Belgium) Under such a design, an R&D tax incentive may

become complex and give room for interpretation both by firms and by fiscal

authorities, and will impose significant compliance costs for both parties

A final differentiation is geographical: China for instance targets specific regions or development zones The Italian government targets some southern

regions In federal countries where regional authorities have fiscal power,

they can establish state or provincial R&D tax credits, as is the case in about

40 US states (the tax credit is based on state taxes otherwise payable, rather

than federal taxes)

2.2.6 Rules for Tax Credit Consumption by Firms

Policy makers can choose whether tax credits apply only for firms that make

a profit in the same fiscal year as the R&D expense took place, or whether

claims can be carried backward or forward, or whether claims can be

dis-bursed in the case of a firm recording a loss Governments can design R&D

tax incentives in such a way as particularly to address recession periods,

when cash flows of firms tend to fall sharply The French government, for

example, responded to the 2008 crisis by making all tax credits accumulated

by firms (which were until then paid over a period of four years) available

within 2009, providing a significant boost to the cash situation of many firms

2.2.7 Permanent versus Temporary Measures

The last dimension deals with the duration of the measure itself In most

countries these fiscal decisions are taken for a limited duration (often four to

five years) and thus need to be renewed periodically It is striking to note that

very few countries abandoned the principle once they adopted it However,

the trend has been, while keeping the principle, periodically to change the

conditions of operation (often for reasons other than R&D issues) This has

then led to industry asking for longer time frames for such measures to be

efficient An article by Arque- Castells and Mohnen (2011) opens another

perspective: looking at behavioural change in Spain, they argue that a

perma-nent system that would support ‘entry into R&D’ by firms might be a very

effective option, but this would require a high though transitory incentive

Ways of implementing such an approach still remain to be invented

This tour has enabled the reader to see the variety of concrete decisions

required to implement an R&D tax incentive It shows how flexible the

instru-ment is and the multiplicity of additional objectives it can include, from

supporting human capital to targeting specific firms, technologies, activities

or regions We shall see that evaluations have focused mostly on one central

issue, whether or not it increases business R&D, for how much and how long

(sections 2.3.1 and 2.3.2) Section 2.3.3 will show that we can say little from

the available evidence about the specific effects of different design features

2.3 EFFECTIVENESS OF R&D TAX INCENTIVES: FINDINGS

FROM EVALUATION STUDIES

This section summarises the empirical findings on the effectiveness of R&D

tax incentives.4 The section starts with some notes on the scope of R&D tax

incentive evaluations and their methodological challenges and limitations

We proceed with a brief discussion of the results of about 20 evaluation

studies conducted between the early 1990s and 2013 in different countries

Tax incentives are typically implemented at the national level as part of

national taxation laws;5 consequently evaluation results refer to the effects

of tax incentives under the specific legislative situation in that country In

addition, we relate the evaluation findings to the design features of the R&D

tax incentives under consideration in order to allow conclusions on the

effectiveness of different policy designs

2.3.1 Scope of the Literature Review

2.3.1.1 Studies on input additionality

The literature review focuses on studies that econometrically analyse the

impact of R&D tax incentives on key policy goals of the instrument Since a

primary goal of R&D tax incentives is to raise the R&D spending of

enter-prises, most studies look at input additionality, that is, the change in private

R&D expenditure that can be attributed to the tax incentive Table 2.1 lists the

most important studies consulted for this review Some of these studies were

official evaluations commissioned by governments and conducted as part

of policy implementation and monitoring activities, while others originated

from academic work based on publicly available or dedicated survey data

The studies are typically based on firm- level panel data and either cover

periods before and after the introduction of a tax incentive or analyse the

effects of changes in the generosity of R&D tax incentives Methodologically,

they estimate R&D demand equations using a dummy variable for the tax

credit or R&D price elasticity (see Hall and Van Reenen, 2000) In recent years

control group approaches have been used too (see Corchuelo and Martínez-

Ros, 2009; Duguet, 2010; Czarnitzki et al., 2011) that compare firms that use an

Manufacturing (publicly listed enterprises only)

R&D demand estimation with tax credit shift parameter (pooled OLS with fixed ef

Manufacturing (country level)

Estimation of R&D price elasticities using dynamic panel models (OLS, instr

Manufacturing and services (publicly listed enterprises only)

Estimation of R&D price elasticities (generalised T

Manufacturing and services

R&D demand estimation with tax cr

Manufacturing (publicly listed enterprises only)

Estimation of R&D price elasticities (pooled OLS, instr

US$1.30 to 2.00