Innovation modes and design as asource of innovation: a firm-level

analysisAndrea Filippetti

Italian National Research Center – CNR-IRPPS, Rome, Italy andSchool of Business, Economics and Informatics, Birkbeck College,

University of London, London, UK

Abstract

Purpose – The purpose of this article is to contribute to the empirical literature, which investigatesinnovation modes, by exploring the role of design as a source of innovation.

Design-methodology/approach – The empirical analysis is carried out at the firm-level, on theground of a recent survey covering more than 5,000 European firms. A factor analysis is carried outfirst, followed by a cluster analysis based on identified factors in order to ensure a significant numberof homogeneous groups of firms.

Findings – The paper finds that: design and R&D are complementary sources of innovation; designis predominant in firms characterized by a complex innovation strategy and intense interactions withthe external environment; and these types of firms also show better economic performance.

Social implications – Policies should recognize the importance of design-based competences, asthey differ from those related to R&D activities.

Originality/value – To date, in this empirical research, R&D activity is regarded as the majorinternal source of knowledge, as well as a fundamental driver of firms’ competitiveness. This paper’sresults show how design enters in this framework and suggests future research directions.

Keywords Innovation, Design, Companies

Paper type Research paper

1. Introduction: design as a (neglected) source of innovationOver the last decade the literature investigating patterns of innovation has devotedincreasing attention to empirical analysis at the micro level, as the relevance of theheterogeneity across firms with regard to innovation behavior has been increasinglyrecognized (OECD, 2009). A stream of empirical studies (reviewed in the secondsection) has tackled the issue of the heterogeneity across firms by putting forward theconcept of innovation modes. This is grounded on the awareness that firm-specificfactors play a fundamental part in shaping a complex phenomenon such as innovation.Accordingly, the presence and the source of regularities – patterns – in the innovativebehavior should also be investigated at the micro level. The innovation modes research

The current issue and full text archive of this journal is available at

www.emeraldinsight.com/1460-1060.htm

This research has been carried out during the author’s visiting period at the Department ofManagement, School of Business, Economics and Informatics at Birkbeck College supported bythe DIME Mobility Fellowship. The author is very grateful to Daniele Archibugi, Marion Frenz,Rinaldo Evangelista and Sergio Bruno for reading and commenting on a previous version of thearticle. He would also like to thank two anonymous reviewers for useful comments andsuggestions. Thanks also go to Keith Sequeira and to the DG Enterprises and Industry of theEuropean Commission for allowing the use of the data of the Innobarometer Survey.

Innovationmodes and

design

5

European Journal of InnovationManagement

Vol. 14 No. 1, 2011pp. 5-26

q Emerald Group Publishing Limited1460-1060

DOI 10.1108/14601061111104670

addressed here, empirical in nature, aims at grouping firms depending on a number ofcharacteristics of innovation. They include, together with structural characteristics ofthe firm, several innovative dimensions such as the formal research and development(R&D) activities, collaboration activities, knowledge management and organizationalinnovation, as well as marketing and service innovation.

Among the characteristics of the firm that define an innovation mode, the sources ofinnovation are of fundamental importance. In this research, R&D activity is regardedas the major internal source of knowledge generation and learning, as well as afundamental driver of firms’ competitiveness. In more recent years, innovation andmanagement scholars have devoted growing attention to the role played bynon-technological innovation such as innovation in services (Evangelista, 2000; Galloujand Djellal, 2010; Kandampully, 2002; Mothe and Thi, 2010), organizational innovation(Chanal, 2004; Camarero and Garrido Marıa, 2008; Jimenez-Jimenez et al., 2008; Tetherand Tajar, 2008). However, the role of design has thus far received little attention in theempirical research on innovation modes.

A central issue when dealing with design is that design is a multifaceted and broadconcept with no commonly agreed upon definition. As Walsh claims “the term ‘design’covers a wide range of activities: architecture, fashion design, interior design, graphicdesign, industrial design, engineering design” (Walsh, 1996, p. 512). In what follows,she also points out that engineering design has evolved into a separate discipline fromindustrial design. Interestingly enough, while in English the term design refers both toindustrial and engineering design, in other languages, such as French and Italian theword design refers only to industrial design. In this article we are bound to refer to thedefinition of design provided in the Survey we use in the empirical analysis. TheInnobarometer Survey definition of design includes “graphic, packaging, process,product, service or industrial design”. Clearly, this is a wide definition of design, and itis not limited to manufacturing but also to the service sector. Importantly, engineeringdesign is not included, and R&D activity is separated.

Previous studies show that design activity represents an important internal sourceof knowledge generation and learning, with its own organizational structure andprocesses, deep interactions with other functions of the firms, as well as externallinkages (Verona and Ravasi, 2003; von Stamm, 2003a, b). In addition, there is agrowing body of evidence showing the positive impact of design on differentdimensions of company performance (see Lorenz, 1986, Hertenstein et al., 2005; Marsiliand Salter, 2006 among the others), as well as the central contribution design activitymakes to innovation (Walsh, 1996; von Stamm, 2004; Verganti, 2008).

Thus, there is a considerable gap between innovation studies addressing patterns ofinnovation at the micro level, on the one hand, and a growing evidence supporting theprominence of design activity in enhancing innovation capabilities and thecompetitiveness of the firm, on the other hand. This paper seeks to fill this gap inthe literature, trough a micro-level analysis of more than 5,000 European firms. Its aimis to show how design enters the innovation modes, by identifying those modes offirms that evolve around design. Which are the characteristics of firms that rely ondesign as a source of innovation? Is design activity an alternative or complementarysource to R&D activity? These are the research question this study seeks to answer.This is followed by a final empirical section that compares the relative importance ofthe innovation modes on firm performance.

EJIM14,1

6

A fundamental tool to tackle the heterogeneity of innovative activities has been theincreasing availability of data at the firm level. The major role has been played by theCommunity Innovation Survey (CIS) (Smith, 2005). Over the last few years a secondsurvey managed by the European Commission, namely the Innobarometer Survey, hasbeen developed and carried out across Europe. In this paper the last Innobarometercarried out on April 2009 covering 5,234 firms is used (European Commission, 2009).

The paper is organized as follows. The next section reviews the empirical researchon innovation modes of firms. Section three presents the survey and the data. Sectionfour discusses the innovation modes through an explorative analysis. Section fiveinvestigates the dynamic of economic performance across the identified innovationmodes. Finally, section six discusses the main findings, some limitations andimplications for policy.

2. Patterns of innovation activities across heterogeneity of firms: a reviewof the “innovation modes” literatureA great deal of literature has focused on the importance of industry-specific factors toexplain patterns of innovation of firms and the dynamic of industrial structure (Pavitt,1984; Archibugi et al., 1991; Breschi et al., 2000; Malerba, 2004). The main assumptionlying behind these studies is that patterns of innovation of firms are sector-specific,depending on the very nature of the technological domain. Even though this body ofliterature has provided important insights about the way firms innovate, furtheranalysis shows that sectors matter to a certain extent, but heterogeneity among firmsplays a crucial role within both sectors and countries (Srholec and Verspagen, 2008).

Within this perspective, over the last decade, a growing empirical literature hasfocused on the sources of heterogeneity at the firm level. Different dimensions ofheterogeneity have been investigated, including the typology of innovation (e.g.product, process, service), the sources of innovation and the related strategies (i.e.in-house vis-a-vis outsource R&D), and the growing importance of non-technologicalinnovation. This has led to the concept of innovation modes, aiming at grouping firmsdepending on a number of characteristics of the innovation activities, behaviors andstrategies. In what follows, the main studies and empirical evidence of this literatureare briefly reviewed.

Laursen and Foss (2003) explore the complementarity between new human resourcemanagement practices and their impact on innovation performances using data from aDanish survey of 1,900 business firms. They identify two main practices across firmsand they find that the adoption of a package of these practices significantly affectsinnovation performance of firms. On a similar note, Arundel et al. (2007) compare thework organization environment and innovation patterns across fourteen Europeancountries. They show how differences in the organizational forms – discretionarylearning, lean production, Taylorism, traditional organization – lead to differentinnovation behavior as identified by three main innovation modes – Leaders,Modifiers, and Adopters. Building on firm level data – the Third European Survey ofWorking Conditions and the third CIS (CIS-3) – they put forward a taxonomy of fourdifferent modes of organization of work and three innovation modes, i.e. Leadinnovators, Technology modifiers, and Technology adopters. They find significantdifferences across countries (even after controlling for industrial structure) in the waywork is organized and how firms innovate.

Innovationmodes and

design

7

The identification of innovation modes in the service sector is also the aim of anempirical paper by Hollenstein (2003) based on a survey including 2,731 Swiss firms.Based on different groups of variables – innovation typologies, sources of innovationand investments – the author identifies five different innovation modes: science-basedhigh-tech firms with full network integration, IT-oriented network-integrateddevelopers, market-oriented incremental innovators with weak external links,cost-oriented process innovators with strong external links along the value chain,low-profile innovators with hardly any external links. He concludes that innovation inservices differs from that in manufacturing in terms of lower levels of R&D and themajor role played by non-technological factors in some segments of the service sector.

Jensen et al. (2007) emphasize how different strategies of knowledge creation andlearning processes are relevant sources of firms’ heterogeneity which are related totheir innovation behaviors. They propose two different innovation modes, one basedon the production and use of codified scientific and technical knowledge, and the otherrelying on informal processes of learning and experience-based know-how. Using aDanish dataset including 700 firms they show that those which adopt strong versionsof both practices tend to improve their innovative performance. Additionally, they findevidence that mixed strategies tend to perform significantly better than those relyingsolely on one or the other of the two modes.

Tether and Tajar (2008) use the Innobarometer survey carried out in 2002 toinvestigate the importance of organizational innovation. Their major finding is that,together with well-established patterns of innovation based on product innovation andprocess innovation, a third relevant “organizational-cooperation mode of innovation”emerges. This mode of innovation results particularly relevant amongst service sectorfirms. Evangelista and Vezzani (2008) explore the relationships between technologicaland non-technological innovation modes (and a combination of the two) and the effectson economic performance and employment using Italian CIS-4 data.

Srholec and Verspagen (2008) assess the heterogeneity of the innovation processesusing exploratory factor analysis on micro data from the CIS-3 across 13 countries.Their analysis shows that, while sectors and countries matter to a certain extent, thelarger share of variance in terms of innovation strategies has to be related to theheterogeneity within both sectors and countries. Finally, in a recent study based on theUK data, Frenz and Lambert (2009a) using micro-level innovation survey dataconcerning innovation find that productivity measured by value added per capita isinfluenced more strongly by non-technological innovation modes, whereas output percapita is relatively more strongly and significantly related to product, process andtechnology based innovation.

All the studies discussed above share the same theoretical underpinning and asimilar methodological approach. The former is the need to take into account theheterogeneity that characterizes firms’ behaviors in relation to innovation activities,dimensions and strategies. From a methodological standpoint, they all addressheterogeneity in firms’ innovation activities relying on explorative multivariateanalysis such as factor analysis and cluster analysis.

Overall, this empirical research that looks at innovation activities at the firm level hasconsiderably enriched our understanding along several directions. First, it has shed somelight on the weight and the sources of heterogeneity across firms in terms innovationcharacteristics, strategies and behavior, as well as the consequences upon economic

EJIM14,1

8

performance. Second, it has highlighted systematic differences in innovation across themanufacturing and service sector. Third, it has called attention to the importance ofnon-technological forms of innovation, such as organizational innovation, humanresources management practices, innovation in services. Also, the fact that differentforms of innovation have been addressed together has shown the presence of patterns ofcombined innovation strategies. Finally, different data sets have been used showing agood deal of consistency among the results. However, three major weakness characterizethis series of studies. It is mainly static and therefore fails in capturing underlyingdynamics of innovative activities. Next, the relationships with the market structure arerarely addressed. And finally, so far design has been barely addressed.

This article seeks to address the fact that the role played by design innovation is stillunder-explored in this kind of literature[1]. When included in the analysis it is associatedwith marketing innovation and is not specifically addressed (Srholec and Verspagen,2008). This can be explained by two main reasons. First, although some scholars haverecognized the important role of design concerning innovative activities (Walsh, 1996) ithas not been taken into full account in the economics of innovation literature yet. In thisresearch R&D is regarded as the major source of knowledge generation. Accordingly, alot of attention is devoted to R&D activity, in terms of internal processes, externalsources and complementarities (see for example, Cassiman and Veugelers, 2006). This isalso well reflected in innovation policy which is still strongly R&D-centric.

The second (and related) reason is the lack of data. In the harmonized CIS-4questionnaire the role of design is explicitly considered as a marketing innovation: “amarketing innovation is the implementation of new and significantly improved designand sales methods to increase the appeal of your goods and services or to enter to newmarkets”. This derives from the last edition of the Oslo Manual in which one can findthat “marketing innovations involve the implementation of new marketing methods.These can include changes in product design and packaging, in product promotion andplacement, and in methods for pricing goods and services” (OECD, 2005, p. 17). Thenext section presents the Innobarometer Survey and discusses its advantages anddrawback compared to the CIS.

3. The data and the variables: the Innobarometer Survey3.1 The Innobarometer Survey 2009The data derives from the Innobarometer Survey 2009 designed and collected by theEuropean Commission (2009). Innobarometer has been conducted on a yearly basissince 2001. The current Innobarometer was conducted in April 2009 in the 27 MemberStates of the EU, Norway and Switzerland. The current wave of (2009) placed the focuson innovation, and the reference period of the various activities surveyed was thatbetween 2006 and 2008. The targeted number of main interviews varied somewhat inaccordance to the size of the respective country; however, the default sample size was200 in most EU Member States. In total 5,238 enterprises from the 27 EU MemberStates, plus Norway and Switzerland responded to the questionnaire. The sample is arandom one stratified by country, enterprise size, and industry (2-digit) (Table AI in theAppendix). The Innobarometer reports further details on the survey procedure(European Commission, 2009).

Most of the research addressing innovation modes relies on the CIS. The latterincludes a large number of firms and also allows for dynamic investigation. The

Innovationmodes and

design

9

Innobarometer Survey used here has a fundamental advantage when compared withthe CIS in relation to this paper’s topic: it includes a specific question about design as asource of innovation. Additionally, the question about design is included in the samequestion related to the other innovation sources, such as R&D (both in-house andexternal) and acquisition of external know-how and machineries (see Table I). This isnot negligible when we compare it with the CIS, where design is explicitly consideredas a marketing innovation, and therefore an output. This article’s findings about thecharacteristics of the innovation modes are in tune with prior empirical evidence. Thissupports the case for the use of this dataset, as allows to better address the role ofdesign.

A peculiar characteristic of the “design” of this survey is the use of an “openapproach” in terms of definition, in opposition for example to the CIS, where verydetailed instructions are attached to explain the terminology at great length. There arebenefit and costs of using this kind of approach when dealing with innovation. Themain benefit is that it does not impose a specific view of innovation upon therespondents, whilst the main cost is a lack of preciseness of the answers.

3.2 The variablesIn Table I the variables feeding into the explorative analysis are presented. All thevariables are categorical dichotomies (yes or no type of answers) and they are dividedinto the following five groups:

Sources of innovation. Among the different sources of innovation it is possible toidentify R&D performed in-house and design activity. It is also possible to distinguishamong three external sources of innovation, that is external R&D, the acquisition ofknowledge (i.e. patents, inventions, know-how) and the acquisition of capital (i.e.machinery, equipment and software). It is worth observing that the number of firmswhich rely on design as an internal source of knowledge in relation to innovationactivities is relevant when compared to other sources (Table II).

Collaboration activities. Among the different sources of external collaboration it ispossible to distinguish between collaboration with customers, suppliers and otherfirms, in contrast to collaboration with universities and research centres.

Non-technological innovation. Two different forms of non-technological innovation areincluded, namely, organizational innovation and marketing innovation. At this stage,product, process and service innovation are not included for two main reasons. First,their importance has been already addressed in section three. Second, there is a greatoverlap between the three innovation outputs as most innovative firms reported product,service and process as innovation outputs. However, their relevance as well as the maindifferences among the manufacturing and service sectors are addressed below.

Activities and methods in support of innovative activities. Methods in support ofinnovative activities are also included. Specifically, the use of patent and designregistration in order to capture whether or not firms follow an appropriability strategylinked to the intellectual property rights (IPRs). The adoption of knowledgemanagement practices, as discussed in the literature ( Jensen et al., 2007) is alsoaddressed. Finally, a set of questions related to the so-called open innovation paradigm(Chesbrough, 2003) is included. Open innovation practices have hardly ever beenaddressed in this series of literature. In this way, this analysis is able to contribute tothis emerging issue.

EJIM14,1

10

Typologies of variables Variable Survey question

Sources of innovation Has your company had expenditureson any of the following activities tosupport innovation since 2006?

R&D performed in house Research and development within yourcompany

R&D acquired outside Research and development performedfor your company by other enterprisesor by research organisations

Acquisition of know-how Purchase or licensing of patents,inventions, know-how, and other typesof knowledge

Acquisition of machinery Acquisition of new or significantlyimproved machinery, equipment andsoftware

Design Design (graphic, packaging, process,product, service or industrial design)

Collaboration activities Has your company developed anystrategic relationships in support ofyour innovation activities with:

Customers Some specific customers or clientsSuppliers SuppliersOther companies in the same field Other companies active in your fieldUniversities and research centres Research institutes and educational

institutions

Non-technologicalinnovation

Has your company introduced any ofthe following innovations since 2006

Marketing innovation New or significantly improvedmarketing strategies

Organizational innovation New or significantly improvedorganisational structures (e.g.knowledge management, workplaceorganisation or external relations)

Methods in support ofinnovative activities

Patents and design registration Has your company had expenditureson application for a patent orregistration of a design

Knowledge management practices Has your company started or increasedany of the following initiatives tointegrate different company activities(R&D, design, marketing/sales,production etc.) in support ofinnovation?Knowledge management systemsInternal mechanisms for employees tosubmit innovative ideasStaff rotations or secondments betweendifferent functionsCreation of cross-functional or cross-departmental teams on innovationprojects

(continued )

Table I.Variables and surveyquestions used in the

factor analysis

Innovationmodes and

design

11

Innovation drivers. Finally, some factors which can drive the innovation strategies offirms are also included. On the one hand, firms’ innovative efforts can be driven bycost-reducing reasons and are usually linked to process innovations (Hollenstein, 2003).While on the other hand, firms can be more attracted to exploiting new technologicalopportunities (Klevorick et al., 1995) and new markets through their innovativeactivities.

4. Design innovation within the innovation modes of the firms4.1 Explorative analysisIn this section the variables discussed above are used to carry out a factor analysis. Afactor analysis, which is commonly used for exploratory, inductive research is carriedout first. Second, a cluster analysis based on identified factors is performed in order toensure a significant number of homogeneous groups of firms. In line with the previousempirical studies using categorical variables (see for example Frenz and Lambert,

Typologies of variables Variable Survey question

Open innovation practices Has your company used any of thefollowing methods to support itsinnovative activities?Create or participate in internet-baseddiscussion forumsGive away or allow free access to testproducts or services to potential usersInvolve potential users in your in-houseinnovation activitiesShare or exchange your intellectualproperty

Innovation drivers Cost reducing Reduce costs of existing products andservices

Technological opportunities Emergence of new technologies to beexploited

Market opportunities New opportunities to enter newmarkets or expand sales in existingmarkets

Source: Innobarometer Survey 2009 (European Commission, 2009b)Table I.

No. firms %

Design 1,846 43R&D performed in house 2,277 54R&D acquired outside 1,498 35Acquisition of machineries 3,747 83Acquisition of external knowhow 2,525 59

Notes: The sample is the total of 4,664 innovative firms. Responses are not mutually exclusiveSource: Elaboration on Innobarometer Survey 2009

Table II.Sources of innovation

EJIM14,1

12

2009b), a principal component analysis is carried out based on the tetrachoriccorrelation matrix between the variables. Table III displays the results of the factorsanalysis of the innovation variables (as in Table I) in terms of factor loadings. The fourfactors extracted account for 55 percent of the total variance, with the first two factorsaccounting for approximately the same amount of variance, around 18 percent (seeAppendix Table AII and Figure AI).

The factors seem to reflect well some of the most relevant dimensions of theinnovation activities of the firms. In particular they can be summarized as follows:

. Factor No. 1 – Technological creation and adoption. it reflects the importanceattached to knowledge, both developed inside the firms and absorbed outside ofthe firms’ boundaries. These activities include design as an input and a strongappropriation strategy through the use of patents and design registrations;

. Factor No. 2 – Interacting and searching outside of the firm’s boundaries. thisfactor reflects the attitude of firms to interact and explore new opportunitiesoutside their boundaries. Specifically, by collaborating with suppliers, customersand universities, by implementing open innovation practices, and exploring newtechnological and market opportunities;

. Factor No. 3 – Non-technological innovation. this factor accounts for theimportance attached to non-technological forms of innovation and specifically toorganizational innovation and marketing innovation, as well as knowledgemanagement practices;

. Factor No. 4 – Cost saving strategy. finally, the last factor accounts for therelevance of cost-saving strategies of innovation activities.

Technologycreation and

adoption

Interacting andsearching out of the

boundaries

Non-technological

form ofinnovation

Costsaving

strategyVariable (Factor 1) (Factor 2) (Factor 3) (Factor 4)

Marketing innovation 0.12 0.08 0.80 20.06Organizational innovation 0.05 0.16 0.77 20.03Knowledge management 0.24 0.37 0.52 20.10Design 0.60 0.12 0.30 0.20R&D performed in house 0.75 0.25 0.20 20.16R&D acquired outside 0.66 0.20 0.33 20.30Acquisition of know-how 0.70 0.12 0.17 0.20Acquisition of machinery 0.33 0.22 0.18 0.67Customers 0.15 0.78 0.10 0.03Suppliers 0.07 0.72 0.16 20.03Universities and res. centres 0.43 0.54 0.09 20.32Other firms 0.06 0.72 0.08 20.04Patents and design reg. 0.79 0.10 0.12 0.03Open innovation practices 0.30 0.51 0.23 20.32Cost reducing 20.16 20.13 20.22 0.49Technological opportunities 0.24 0.57 0.15 0.12Market opportunities 0.24 0.40 0.38 0.18

Table III.Factor analysis of theinnovation variables:

factors loading andunique variances

Innovationmodes and

design

13

4.2 Identified innovation modesThis section identifies the innovation modes on the basis of the results of anon-hierarchical cluster analysis. Different solutions from different clustermethodologies have been compared, also considering different numbers of clusters.The results are quite robust across the different methodologies. The five clusters andthe methodology have been chosen trying to balance three different criteria:

(1) statistical significance;

(2) economic interpretation; and

(3) a sufficient number of observations within each cluster.

Following these criteria five groups have been chosen together with the k-meansmethodology. The number of observations ranges from 488 of the third cluster to 732of the first cluster. In Table IV the distribution of the variables within the fiveidentified clusters is reported. On the basis of these results the following fiveinnovation modes have been identified:

Mode 1: “outward-oriented non-technological innovation”. This group of firmsconsists of 732 firms (24 percent of the sample) and it is characterized by an intensenon-technological innovation activity, namely organizational innovation, marketinginnovation and knowledge management practices. The other distinctive feature of thisgroup is the importance attached to collaborative activities mainly with customers,suppliers and other firms in the same field. This group of firms also adopts openinnovation practices extensively and is oriented towards the exploration of newtechnological and market opportunities. Regarding typologies of innovation, processinnovation and service innovation are more relevant (Table V). Concerning size

Cluster1 2 3 4 5

Marketing innovation 0.76 0.1 0.26 0.81 0.8Organizational innovation 0.86 0.19 0.36 0.82 0.83Knowledge management 0.76 0.24 0.5 0.54 0.8Design 0.4 0.32 0.24 0.41 0.86R&D performed in-house 0.43 0.42 0.7 0.35 0.97R&D acquired outside 0.21 0.16 0.59 0.17 0.78Acquisition of know-how 0.12 0.17 0.08 0.12 0.34Acquisition of machinery 0.91 0.94 0.49 0.79 0.97Customers 0.86 0.4 0.59 0.14 0.71Suppliers 0.86 0.38 0.67 0.12 0.78Universities and res. centres 0.49 0.18 0.66 0.15 0.78Other firms 0.66 0.25 0.38 0.07 0.44Patents and design registration 0.04 0.11 0.14 0.08 0.62Open innovation practices 0.77 0.29 0.79 0.42 0.85Cost reducing 0.28 0.62 0.15 0.34 0.2Technological opportunities 0.65 0.36 0.38 0.22 0.64Market opportunities 0.83 0.52 0.46 0.57 0.86

Notes: Cluster no. 1: outward-oriented non-technological innovation; cluster no. 2: cost-savinginnovation; cluster no. 3: R&D-focus with strong basic collaboration; cluster no. 4: inner-oriented non-technological innovation; cluster no. 5: outward-oriented multifaceted innovation

Table IV.The characteristics of thefive clusters based on theindicator used for thefactor analysis

EJIM14,1

14

distribution, small firms are more important than large ones (Table V), while in terms ofsectoral distribution the less knowledge-intensive sector is over-represented (Table VI).

Mode 2: “cost-saving innovation”. Similarly to previous empirical findings(Hollenstein, 2003) one cluster is mainly characterized by the relevance ofcost-competiveness innovation together with the importance of the acquisition ofexternal machinery[2]. It accounts for 21 percent of the sample equal to 655 firms. Inthis cluster, large firms are under-represented while small firms are above the average(Table V). By looking at Table VI it clearly emerges that cost-saving strategies aremore likely to be pursued in the manufacturing sector with respect to services, andspecifically in the medium-low tech sector.

Mode 3: “R&D-focus with strong basic collaboration”. This cluster of firms is thesmallest, accounting for 488 firms equal to 16 percent of the sample. Firms here presenta strong technological focus of their innovative activities based on R&D, both in-houseand external. They present a very low propensity to acquire know-how and machineryfrom outside. They also show strong linkages with universities and research centres,as well as a propensity to undertake open innovation practices. The size distributiondoes not significantly differ from that of the sample, while the knowledge-intensivesector plays an important role here with respect to the others (Tables V and VI).

Mode 4: “inner-oriented non-technological innovation”. This group of firms resemblesthe first one in terms of the importance attached to non-technological forms of

Cluster1 2 3 4 5 Total

Typologies of innovationProduct

% 67 59 63 54 88 66No. 464 370 292 346 461 1,933

Process% 74 53 52 61 85 65No. 529 340 252 400 453 1,974

Service% 68 47 55 55 65 58No. 488 302 266 357 344 1,757

Firm sizeSmall

% 40 44 36 43 25 38No. 294 289 175 288 131 1,177

Medium% 35 33 35 33 32 34No. 256 214 171 218 172 1,031

Large% 25 23 29 24 43 28No. 182 152 142 158 231 865

Total % 100 100 100 100 100

Notes: Cluster 1: outward-oriented non-technological innovation; cluster 2: cost-saving innovation;cluster 3: R&D-focus with strong basic collaboration; cluster 4: inner-oriented non-technologicalinnovation; cluster 5: outward-oriented multifaceted innovation

Table V.Typologies of innovation

and firm size by cluster

Innovationmodes and

design

15

innovation, specifically organizational innovation and marketing innovation. The majordifference lies in the very low propensity to undertake collaboration activities outsidetheir boundaries. In addition, these firms are also less likely to explore technologicalopportunities and new market opportunities. Similar to mode no. 1, in this cluster processinnovation and service innovation are more important than product innovation. In termsof size distribution, small firms are clearly over-represented with respect to the large ones(Table V). In relation to the sectoral distribution, the relevance of the service firms and inparticular of the “Wholesale, retail and trade” industry emerges.

Mode 5: “outward-oriented multifaceted innovation”. This group accounts for 17 percent of the sample equal to 534 firms. These firms are primarily characterized bywell-rounded innovation activity. Non-technological innovation is coupled with R&Dactivity as well as design activity. They are very active in absorbing knowledge andtechnology from outside in terms of acquisition of know-how and machinery. They arevery inclined to establish external collaboration, especially collaboration withsuppliers as well as with universities and research centres. They are also likely to beinvolved in open innovation practices and exploring new technological opportunitiesand new markets. This is also the only cluster in which appropriability strategy playsan important role. In this mode, the prominence of product and process innovation withrespect to service innovation emerges (Table V). In terms of firm size, in this clusterlarge firms play a big role: 43 per cent are large firms with respect to a 28 per cent of the

ClusterSector and industry 1 2 3 4 5 Total

Medium-high tech manufacturing 10 15 13 7 19 13Medium-low tech manufacturing 29 40 28 29 35 32knowledge intensive service 18 13 24 17 21 18Less-knowledge intensive service 43 32 35 47 25 37Coke, petroleum 1 0 2 2 2 1Chemicals 16 21 15 17 22 18Rubber, plastic products 3 6 8 4 4 5Metals 7 7 4 5 8 6Machinery and equipment 4 5 7 4 10 6Radio, television, medical instr, Motor vehicles 2 4 3 1 4 3Furniture, manufacturing nec. 2 3 2 2 3 2Electricity, gas water supply 2 2 3 1 1 2Construction 9 11 6 6 3 7Wholesale, retail and trade 25 19 20 31 18 23Hotels restaurants 5 2 4 7 4 4Transport, pipelines 7 7 5 5 4 6Post and telecoms, fin. int., insurance 4 3 5 5 5 4Real estate, renting of mach. 2 1 1 1 1 1Computer and related activities, R&D 2 1 4 2 4 2Other business activities 6 5 6 6 7 6Public administration, educ., health 1 1 4 1 1 1Total 100 100 100 100 100 100

Notes: Cluster 1: outward-oriented non-technological innovation; cluster 2: cost-saving innovation;cluster 3: R&D-focus with strong basic collaboration; cluster 4: inner-oriented non-technologicalinnovation; cluster 5: outward-oriented multifaceted innovation

Table VI.Industry and sectordistribution by cluster

EJIM14,1

16

sample, while only 25 per cent of firms are small firms with respect to 38 per cent of thesample (Table V). Also, the over-representation of the high-tech manufacturing sector,in particular the “chemicals” sector, “Machinery and equipment” is evident. Also theknowledge-intensive service innovation is over-represented with respect to the sample,namely the “Post and telecoms, fin. int., insurance” sector, the “Computer and relatedactivities, R&D” sector and the “Other business activities” sector (Table VI).

To sum up, thanks to the large number of firms it has been possible to delve intosome aspects of the heterogeneity of the innovation modes already highlighted in theliterature. Two different innovation modes based on non-technological innovation havebeen identified (mode no. 1 and mode no. 4). In a recent paper already discussed Tetherand Tajar (2008) stress the importance of the organizational-cooperation mode ofinnovation. In line with their results, here both non-technological modes are morerelevant in relation to the service sector (see also Hollenstein, 2003). However, the twomodes identified here differ in terms of their attitude towards their externalenvironment. That is, cluster no. 1 firms attach a large importance to externalcollaboration when compared to firms belonging to cluster no. 4. This difference intheir relationship with their environment is also reflected in the different propensity offirms in adopting open innovation practices, and about the importance of exploringnew technological opportunities and new market opportunities.

A “pure” R&D-based mode of innovation also emerges as relevant. These firms arenot very likely to couple R&D activity with non-technological innovation, and also donot tend to acquire knowledge from outside. Consistent with their focus on technology,they are more likely to establish science-based external collaborations. It is worthnoting that the knowledge-intensive sector plays an important role.

Finally, a group of firms extremely active in every typology of innovation andinvolved in every form of collaboration also emerged (cluster no. 5). These firms alsoshow a high propensity to both rely on IPRs as a means of capturing returns from theoutcomes of their innovative activities and to adopt open innovation practices. The factthat these firms pursue open innovation practices which include the sharing ofknowledge and intellectual property rights is not at odds with the importance they attachto the appropriability strategy. On the contrary, this suggests that an important elementof the open innovation strategy is about balancing the pursuit of proprietary knowledgewith some form of open access and sharing of knowledge external to the firms.

As far as the importance of design among the identified clusters is concerned, itsstrong association with mode no. 5 is clear. As for the sources of innovation, it is worthstressing that within this mode of innovation 97 per cent of firms reported carrying outR&D activities (Table IV). Design as a source of innovation accounts for 86 per cent.This suggests an overlap of these two sources of innovation and therefore lendssupport to the complementarity hypothesis between design and R&D. In terms offirm’s structural characteristics, design innovation is predominant in a clustercharacterized by the presence of large firms along with an over-representation ofhigh-tech manufacturing and knowledge-intensive services. Finally, innovationstrategies are characterized by a complex approach to innovation activities coupledwith an explorative and interactive attitude beyond their boundaries.

However, design is also partially present in both the innovation modescharacterized mainly by the role of non-technological innovation, anover-representation of the service sector as well as a predominance of process

Innovationmodes and

design

17

innovation (mode no. 1 and mode no. 4), while it is less important in the pure R&Dinnovation mode. In the next section we investigate whether there are significantdifferences across the five innovation modes in terms of economic performance.

5. Innovation modes and economic performance of firmsThe relationship between patterns of innovation and economic performance of firmshas been one of the lines of research addressed by this literature (see for exampleCainelli et al., 2006; Jensen et al., 2007; Frenz and Lambert, 2009a). This section exploresthe relationship between the identified innovation modes and the economicperformance of firms.

The results of a “robust” ordered logistic estimate are shown in Table VII. Thedependent variable is the dynamic of the turnover over the period 2006-2008. It canassume four values (¼ 1 if the turnover decreased; ¼ 2 if the turnover increased by lessthan 10 per cent; ¼ 3 if the turnover increased by 10 to 50 per cent; ¼ 4 if turnoverincreased more then 50 per cent). The independent variables of Model 1 are binaryvariables indicating whether or not the firm belongs to a particular cluster. In Model 2,three sets of control variables are included in order to control for the size of the firms,the intensity of innovation expenditures, and the industry effect.

Model 1 shows that the only two innovation modes which are significant andpositive predictors for turnover’s growth are the “outward non-tech” mode and the“multifaceted-innovation” mode[3]. Specifically, firms belonging to the “multi-facetedinnovation” mode are more likely to show faster rates of turnover’s growth withrespect to those feeding into the “outward non-tech” mode and the others.

When three sets of control variables are added in Model 2, namely firm size,intensity of innovation expenditure and macro-sectoral dummies, the odds ratiorelative to the two innovation modes is still positive and significant. Additionally, the

Model 1 Model 2Variables Odds ratio Coefficient Odds ratio Coefficient

Innovation modesOutward non-technological 1.28 * * 0.24 * * 1.24 * 0.22 *

Inner non-technological 1.11 0.1 1.17 0.16Technology-focus 0.93 20.06 0.93 20.077Multifaceted innovation 1.71 * * * 0.54 * * * 1.72 * * * 0.55 * * *

Control variablesMedium size 1.3 * * * 0.26 * * *

Large size 1.23 * * 0.21 * *

Innovation intensity 1 1.17 0.15Innovation intensity 2 1.78 0.16Innovation intensity 3 2.27 * * 0.82 * * *

Medium-high tech manufacturing 1.29 * 0.25 *

Medium-low tech manufacturing 0.94 20.06Knowledge intensive service 1.53 * * * 0.43 * * *

Observations 2,381 2,152

Notes: * p , 0.1; * * p , 0.05; * * * p , 0.01; reference categories: innovation modes: cost-savinginnovation; size: 20-49 employees; innovation intensity: low ,5% of the turnover; macro-sector: less-knowledge intensive service

Table VII.Economic performanceand innovation modes,ordered logit model (oddsratio and coefficient)

EJIM14,1

18

difference between the odds ratio relative to the two innovation modes moderatelyincreases. The fact that, in Model 2, the odds ratio of the two modes does not changesignificantly is worth discussing. In fact, firm size, innovation intensity and sectoraldummies are significant and positive. That is, medium and larger firms are more likelyto be associated with faster growth rates of turnover with respect to small firms. Firmswhich invest a large share of their turnover (mode than 50 per cent) are also more likelyto experience faster turnover growth. Finally, firms belonging to the high andmedium-tech sector and knowledge-intensive sector are also associated with fasterrates of turnover. So far this is hardly surprising. However, what is meaningful is thatthe estimated odds ratio is robust to the introduction of these control variables. Andthis is particularly remarkable in the case of the multi-faceted innovation mode, whichis characterized by both an over-representation of large firms as well as of high-techand knowledge-intensive sectors. This suggests that with respect to these twoinnovation modes, both size and industry are not a necessary condition to be associatedwith faster growth of turnover. On the contrary, idiosyncratic characteristics, such asinnovation strategies, play a role.

6. Discussion of the main findings: limitations, future research andimplications for policyThe aim of this paper was to bring design as a source of innovation into the realm ofthe empirical research that investigates the patterns of innovation at the firm level.This stream of research has so far underestimated the importance of design activity asa source of innovation and competitiveness of the firm. This is at odds with a wealth ofstudies showing the relevant contribution made by design activity in enhancing firm’sinnovation capabilities and competitiveness. This paper has tried to fill this gap bycarrying out an empirical analysis across more than 5,000 European firms by using arecent Innobarometer Survey carried out in 2009. Some interesting policy implicationarise and they are discussed below, but we first turn to discuss some limitations.

Since the data do not allow us to point out the amount of expenditure of designactivities and R&D activities, it was not possible to determine their relativeimportance. In addition, cross-section analysis does not allow to investigate thephenomenon from a dynamic perspective. It can be that in some industry design playsthe major role with respect to R&D activities (and vice versa), in terms of amount ofresources, allocation of human resources and impact on firm’s competitiveness. Also,we have pointed out that the Innobarometer Survey provides a wide definition ofdesign. It has been argued that that one of the main advantages of using a generaldefinition is that it does not impose a specific view on the respondent. When dealingwith design this can be a preferable feature since among both firms and designers (aswell as scholars) a precise definition of design is still lacking. Too much restrictiveex-ante definitions of design could risk to exclude some forms of design. On the otherhand, the fact that we are using a broad definition of design needs to be taken intoaccount when interpreting the results.

Three main findings are worth stressing in relation to our research questions putforward in the Introduction. First, design and R&D seem to be complementary sourcesof innovation rather than alternative. This raises some important questions to beaddressed in future studies about the mechanisms linking the two functions within thefirm in terms of relationships, management practices, and sources of competitive

Innovationmodes and

design

19

advantage. Further, more research needs to be done to further investigate therelationship between design and R&D taking a cross-industry perspective, along thefollowing two directions:

(1) the relative importance of R&D, and design and its evolution over time withinthe firm; and

(2) the presence of systematic differences in terms of R&D and designexpenditures.

Second, we find that design is prominent in the more dynamic type of firm. Specifically,design activity is predominant in relation to the innovation mode characterized by:

. a complex strategy in terms of innovation activities, i.e. technological innovationis coupled with service and organizational innovation; and

. an explorative and interactive attitude in relation to the external environment interms of establishing collaborations, exploiting technological opportunities andnew markets, as well as “open innovation practices”.

This raises another interesting question to be further addressed. That is, what is therelationship between design and these specific attitudes of the firm, such ascollaboration and interaction with the external environment. As we know, for example,that designer consultants often make a substantial part in developing new designsolutions within the firm (Walsh, 1996; Filippetti, 2010).

Thirdly, the multifaceted innovation mode is also associated with a relatively bettereconomic performance when compared to the other modes. As for the role played bydesign in fostering firm’s competitiveness, the results cannot be said to be conclusive,as it was not possible to single out the contribution of design. Our evidence shows thatcomplex strategies of innovation are associated with better economic performance andthat design is part of it.

It is difficult to draw bold implications for business practice related to design basedon our evidence. Sure, our results show that strategies characterized by a multifacetedapproach to innovation activity coupled with a strong interaction outside the firm’sboundaries have a positive impact on performance. This is in tune with the mainmessage of the “open innovation” paradigm, as well as some research showing howdesign considerably benefits from collaboration and an interactive attitude with thesocial environment (Walsh, 1996; Verganti, 2003; Filippetti, 2010).

The results also suggest some implications in terms of measurement methodology.The important steps forward of the third edition of the Oslo Manual in terms ofincorporating non-technological innovation as a fundamental form of innovativeactivities (OECD, 2005) have been already recognized (Smith, 2005). However, design isconsidered a sub-product of marketing innovation[4]. This is also reflected in the CISharmonized questionnaire, which includes a question about design only as part ofmarketing innovation, and therefore an output. In light of our results, it is reasonable tosuggest including a specific question regarding design separate from marketinginnovation and emphasizing its specific nature[5]. However, this should be done once amore accurate definition of design is provided. A suggestion would be to amend thedefinition of design in the next revision of the Oslo Manual, emphasizing its nature ofprocess and source of innovation.

EJIM14,1

20

From a normative perspective, this paper’s findings are in tune with those scholarsclaiming the need to broaden the definition of innovation activities beyond the focus ontechnology and R&D activities. Within this standpoint, this paper’s contribution is toadd some good reason and sound evidence to seriously take into account the roleplayed by design. This has also policy implications. To begin with, a generalacknowledgement of the role that design can play in fostering innovation and firms’competitiveness would be a fundamental first achievement (European Commission,2009). A central point that should be taken into deep account by policy makers is theimportance of qualified human resources in this field. Design activity requires a highlyskilled base of human resources ranging from designers to engineers, scientists tocraftsmen. The set of competences designers need to have is quite broad, ranging frommaterials to production techniques, technology and marketing. Universities and theeducation system as a whole, together with the vocational training system, should bewell aware of the increasing importance design is going to play in the coming years asa source of competitiveness for the firms. More generally, we show that many firmstake a multifaceted approach to innovation activities coupling formal R&D activity,design, organizational practices and so on. It has to be borne in mind that thecompetences and skills which lie behind these rising forms of innovation differ fromthose related to R&D activities. Policies should therefore recognize the importance ofdesign-based competences, along with R&D-based ones, as long as we believe theseforms of innovation are going to play a major role in fostering competitiveness andgrowth, as well as generating new job opportunities in advanced countries in thecoming years.

Notes

1. For some first attempts to investigate design at the micro-level see Cereda et al., 2005; Marsiliand Salter, 2006.

2. It should be noticed that the acquisition of external machineries is an important featureacross all clusters, with the only exception of cluster(3.

3. In this section the usual caveat regarding causality when dealing with cross-sectionanalysis applies. As already pointed out in the discussion in the literature review, therelationship between economic performance and innovation activity is very likely toresemble a self-reinforcing mechanism taking place over time rather then showing aone-way direction of causality (see for example Cainelli et al., 2006). However, the mainpoint here is to underline the presence of a significant association between the twophenomena and the differences across the independent variables (innovation inputs andinnovation modes).

4. More precisely: “marketing innovations include significant changes in product design thatare part of a new marketing concept. Product design changes here refer to changes inproduct form and appearance that do not alter the product’s functional or usercharacteristics” (OECD, 2005, p.49).

5. A good example is the UK version of the CIS, where a new item, named “all forms of design”,has been added in the “Innovation activities and expenditures” section. It has been defined asfollows: “expenditure on design functions for the development or implementation of new orimproved goods, services and processes. Expenditure on design in the R&D phase of productdevelopment should be excluded”.

Innovationmodes and

design

21

References

Archibugi, D., Cesaratto, S. and Sirilli, G. (1991), “Sources of innovative activities and industrialorganization in Italy”, Research Policy, Vol. 20 No. 4, pp. 299-313.

Arundel, A., Lorenz, E., Lundvall, B.A. and Valeyre, A. (2007), “How Europe’s economies learn:a comparison of work organization and innovation mode for the EU-15”, Industrial andCorporate Change, Vol. 27 No. 2, pp. 243-63.

Breschi, S., Malerba, F. and Orsenigo, L. (2000), “Technological regimes and Schumpeterianpatterns of innovation”, The Economic Journal, Vol. 110 No. 463, pp. 388-410.

Cainelli, G., Evangelista, R. and Savona, M. (2006), “Innovation and economic performance inservices: a firm-level analysis”, Cambridge Journal of Economics, Vol. 30 No. 3, pp. 435-58.

Camarero, C. and Garrido Marıa, J. (2008), “The role of technological and organizationalinnovation in the relation between market orientation and performance in culturalorganizations”, European Journal of Innovation Management, Vol. 11 No. 3, pp. 413-34.

Cassiman, B. and Veugelers, R. (2006), “In search of complementarity in innovation strategy:internal R&D and external knowledge acquisition”, Management Science, Vol. 52 No. 1,pp. 68-82.

Cereda, M., Crespi, G., Criscuolo, C. and Haskel, J. (2005), Design and Company Performance:Evidence from the Community Innovation Survey, DTI Report, DTI, London.

Chanal, V. (2004), “Innovation management and organizational learning: a discursive approach”,European Journal of Innovation Management, Vol. 7 No. 1, pp. 56-64.

Chesbrough, H.W. (2003), Open Innovation: the New Imperative for Creating and Profiting fromTechnology, Harvard Business School Press, Boston, MA.

European Commission (2009), Innobarometer 2009, DG Enterprise and Industry, Brussels.

Evangelista, R. (2000), “Sectoral patterns of technological change in services”, Economics ofInnovation and New Technology, Vol. 9 No. 3, pp. 183-221.

Evangelista, R. and Vezzani, A. (2008), “The employment impact of technological andorganizational innovations, firm level evidence”, paper presented at the Non-technicalInnovation Definition, Measurement & Policy Implications Workshop, Karlsruhe,October 16-17.

Filippetti, A. (2010), “Harnessing the ‘essential tension’ of industrial design: the complexrelationship between the firm and designer consultants”, available at: http://ssrn.com/abstract¼1586958

Frenz, M. and Lambert, R. (2009a), “Mixed mode and organizational innovations in theperformance of firms: an analysis of Innovation Survey and Annual Business Inquirydata”, working paper.

Frenz, M. and Lambert, R. (2009b), Innovation in Firms: A Microeconomic Perspective, OECD,Paris.

Gallouj, F. and Djellal, F. (2010), The Handbook of Innovation and Services, Edward Elgar,Cheltenham.

Hertenstein, J.H., Platt, M.B. and Veryzer, R.W. (2005), “The impact of industrial designeffectiveness on corporate financial performance”, Journal of Product InnovationManagement, Vol. 22 No. 1, pp. 3-21.

Hollenstein, H. (2003), “Innovation modes in the Swiss service sector: a cluster analysis based onfirm-level data”, Research Policy, Vol. 32 No. 5, pp. 845-63.

Jensen, M.B., Johnson, B., Lorenz, E. and Lundvall, B.A. (2007), “Forms of knowledge and modesof innovation”, Research Policy, Vol. 36 No. 5, pp. 680-93.

EJIM14,1

22

Jimenez-Jimenez, D., Valle, R.S. and Hernandez-Espallardo, M. (2008), “Fostering innovation:the role of market orientation and organizational learning”, European Journal ofInnovation Management, Vol. 11 No. 3, pp. 389-412.

Kandampully, J. (2002), “Innovation as the core competency of a service organisation: the role oftechnology, knowledge and networks”, European Journal of Innovation Management,Vol. 5 No. 1, pp. 18-26.

Klevorick, A.K., Levin, R.C., Nelson, R.R. and Winter, S.G. (1995), “On the sources andsignificance of interindustry differences in technological opportunities”, Research Policy,Vol. 24 No. 2, pp. 185-205.

Laursen, K. and Foss, N.J. (2003), “New human resource management practices,complementarities and the impact on innovation performance”, Cambridge Journal ofEconomics, Vol. 27 No. 2, pp. 243-63.

Lorenz, C. (1986), The Design Dimension: The New Competitive Weapon for Business, BasilBlackwell, New York, NY.

Malerba, F. (2004), “Sectoral systems of innovation: basic concepts”, in Malerba, F. (Ed.), SectoralSystems of Innovation, Cambridge University Press, Cambridge.

Marsili, O. and Salter, A. (2006), “The dark matter of innovation: design and innovativeperformance in Dutch manufacturing”, Technology Analysis & Strategic Management,Vol. 18 No. 5, pp. 515-34.

Mothe, C. and Thi, T.U.N. (2010), “The link between non-technological innovations andtechnological innovation”, European Journal of Innovation Management, Vol. 13 No. 3,pp. 313-32.

OECD (2005), Oslo Manual: Guidelines For Collecting And Interpreting Innovation Data, OECD,Paris.

OECD (2009), Innovation in Firms: A Microeconomic Perspective, OECD, Paris.

Pavitt, K. (1984), “Sectoral patterns of technological change: towards a taxonomy and a theory”,Research Policy, Vol. 13 No. 6, pp. 343-73.

Smith, K. (2005), “Measuring innovation”, in Fagerberg, I., Mowery, D. and Nelson, R. (Eds),The Oxford Handbook of Innovation, Oxford University Press, Oxford.

Srholec, M. and Verspagen, B. (2008), “The voyage of the beagle in innovation system land:exploration on sectors, innovation, heterogeneity and selection”, UNU-Merit workingpaper, Maastricht.

Tether, B. and Tajar, A. (2008), “The organizational-cooperation mode of innovation and itsprominence amongst European service firms”, Research Policy, Vol. 37 No. 4, pp. 720-39.

Verganti, R. (2003), “Design as brokering of languages: the role of designers in the innovationstrategy of Italian firms”, Design Management Journal, Vol. 14 No. 3, pp. 34-42.

Verganti, R. (2008), “Design, meanings, and radical innovation: a metamodel and a researchagenda”, The Journal of Product Innovation Management, Vol. 25 No. 5, pp. 436-56.

Verona, G. and Ravasi, D. (2003), “Unbundling dynamic capabilities: an exploratory study ofcontinuous product innovation”, Industrial and Corporate Change, Vol. 12 No. 3,pp. 577-606.

von Stamm, B. (2003a), Managing Innovation, Design and Creativity, John Wiley & Sons,Chichester.

von Stamm, B. (2003b), “Whose is design it? The use of external designers”, The Design Journal,Vol. 1 No. 1, pp. 41-53.

Innovationmodes and

design

23

von Stamm, B. (2004), “Innovation - what’s design got to do with it?”, Design ManagementJournal, Winter, p. 2004.

Walsh, V. (1996), “Design, innovation and the boundaries of firms”, Research Policy, Vol. 25 No. 4,pp. 509-29.

Further reading

Czarnitzki, D. and Thorwarth, S. (2009), “The design paradox: the contribution of in-house andexternal design activities on product market performance”, ZEW Discussion PaperNo. 09-068, ZEW, Mannheim.

Laursen, K. and Salter, A.J. (2004), “Searching high and low: what types of firms use universitiesas a source of innovation”, Research Policy, Vol. 33 No. 8, pp. 1201-15.

Appendix

Figure AI.Eigenvalues after thefactor analysis

EJIM14,1

24

No. firms

Size of the firmSmall 1,755Medium 1,524Large 1,340Industries (NACE 2-digit)23 7124 78225 18927 27229 24430 832 6034 6636 11240 8845 33250 1,09355 24460 27865 23770 5572 11574 28575 88Total 4,619CountriesBelgium 182Czech Rep. 180Denmark 177Germany 183Estonia 172Greece 184Spain 177France 173Ireland 182Italy 174Cyprus 63Latvia 164Lithuania 180Luxemburg 56Hungary 140Malta 61Netherland 182Austria 185Poland 187Portugal 171Slovenia 190Slovak Rep. 185Finland 186Sweden 187

(continued )

Table AI.Total sample of

innovative firms by size,industry and country

Innovationmodes and

design

25

About the authorAndrea Filippetti is Research Fellow at the Italian Research Council – CNR, and VisitingResearcher at the School of Business, Economics and Informatics, Birkbeck College, Universityof London. His reserch interests are industrial economics, innovation theory, intellectual propertyrights, technological change and economic growth. Andrea Filippetti can be contacted at:a.filippetti@irpps.cnr.it

No. firms

United Kingdom 168Bulgaria 158Romania 184Norway 92Switzerland 96Total 4,619

Note: “Non innovative firms” have been excluded from the sample, that is, those firms which have notreported any innovative activities over the considered period, equal to 615 firmsTable AI.

Factor Variance Proportion Cumulative

Factor 1 3.11 0.18 0.18Factor 2 3.05 0.18 0.36Factor 3 1.99 0.12 0.48Factor 4 1.16 0.07 0.55

Note: There are several methodologies to rotate a factor loading matrix in order to make results easierto interpret. In this case the author used the orthogonal varimax methods, in which the rotated factorsare still orthogonal

Table AII.Factor analysis of theinnovation variables:explained variance

EJIM14,1

26

To purchase reprints of this article please e-mail: reprints@emeraldinsight.comOr visit our web site for further details: www.emeraldinsight.com/reprints