kiese: national styles of cluster promotion

37

Structural Change in Europe – Innovative City and Business Regions

burg, London, Paris and Munich. But the actual structur-ing of the globally valid flexibilisation requirements hap-pens in Berlin – place matters! – in a specifically Berlin style. Their habitual practice is constituted in the City of Berlin in accordance with its own rules, processes and narratives that correspond with the contradiction, hetero-geneity and improvisation that are the bedrock of this city’s identity. (English editing: Blossom Stefani) ::

ReferencesLange, B. 2007 Die Räume der Kreativszenen. Culturepreneurs und ihre Orte in

Berlin (Bielefeld: Transcript Verlag).Senate Administration for the Economy 2008 Kulturwirtschaft in Berlin – Entwick-

lungen und Potenziale (2. Kulturwirtschaftsbericht). Senate Administration for the Economy 2008 (Berlin).

Wellmann, I. 2009 Schnittstellenkulturen – Hybride Akteure, Patchworkökonomien, intermediäre Institutionen in Lange, B./Kalandides, A./Stöber, B./Wellmann, I. Governance der Kreativwirtschaft. Diagnosen und Handlungsoptionen (Bielefeld: Transcript Verlag).

see following page

38


Kie

se

Now that almost two decades have passed since the publication of Porter’s (1990) enquiry into ‘The Competi-tive Advantage of Nations’, the cluster concept has appar-ently not lost its appeal to academics, policymakers and practitioners thinking about the promotion of innovative capabilities and economic growth.

Defined by Porter himself (1998: 197 f.) as “geographic concentrations of interconnected companies, specialised suppliers, service providers, firms in related industries, and associated institutions (for example, universities, standards agencies, and trade associations) in particular fields that compete but also cooperate,” clusters are wide-ly regarded as a panacea for national, regional and local competitiveness.

Business-led versus government-led initiativesWith the ongoing popularity of cluster promotion, it is

often overlooked that the most shining cases of successful spatial concentrations of industry emerged and grew with-out explicit government intention. Of over 800 clusters identified world-wide in the Harvard Business School’s Cluster Meta Study, only one emerged through a public initiative, while the emergence of a further 40 was influ-enced by public action such as wars and expulsions (van der Linde 2005: 28). Therefore, it is now widely accepted that governments can only create favourable conditions for the emergence of clusters and facilitate their growth and restructuring once they have emerged. Cluster poli-cies can be defined as “efforts of government to develop and support clusters in a particular region” (Hospers & Beugelsdijk 2002: 382). Their degree of public agency sets them apart from business-led cluster initiatives in which cluster firms assume centre stage, while govern-ment and/or the research community only play a support-ive role (cf. Sölvell et al. 2003: 31).

Knowledge economy – from technology policy to triple helix relationsThe global spread of the cluster concept in policy and

practice can be explained by a set of common challenges faced by countries and regions at all stages of develop-ment. First and foremost, globalisation exposes nations and regions to intensified locational competition, which in turn accelerates structural change and promotes the competitiveness of such territories to a new paradigm for economic development (cf. Siebert 2005, Martin et al. 2006). Further challenges include the transition of ad-vanced industrial countries into knowledge economies

(OECD 1996) and the dearth of budgets making clusters an attractive means of focussing public funds. A growing concern with technological capabilities was joined by a mounting dissatisfaction with traditional technology poli-cies based on linear models of innovation. As a result, the interactive and systemic character of technological change is now commonly acknowledged, as is the assumption that despite globalisation, placebound ‘sticky’ assets such as the exchange of implicit knowledge assume even more importance for the competitiveness of firms. In policy and practice, the cluster approach is often employed to develop so-called triple helix relations between universities, indus-try and government (Etzkowitz & Leydesdorff 2000).

Diffusion channels for cluster know-howFurther to these common challenges, research high-

lights a variety of channels available for the diffusion of policies across space and time. Knowledge of and about cluster promotion may diffuse through literature, either academic, best practice case studies, or the growing sup-ply of manuals for cluster development distilled from the former. Such know-how is also disseminated by special-ised consultants acting as transfer agents (Stone 2004), and by the mobility of practitioners between regions and countries. Practitioners may also use cluster conferences like those held annually by The Competitiveness Institute to link into international communities of practice that might lead to knowledge flows in more informal networks later on (Brown & Duguid 1991). Finally, politicians and practitioners may travel abroad to study cases of cluster promotion perceived as best-practice, or invite their peers to give presentations. Within the German-speaking area, for instance, politicians and practitioners tend to refer to Upper Austria as a source of inspiration, and as an occa-sional destination for policy tourism.

In the European context, the Commission plays a cen-tral role in disseminating best practice and promoting policy-learning across borders. This aim is generally pur-sued through the open method of co-ordination and the principle of yardstick competition. As for cluster policy, the Commission focuses on mapping, the identification and dissemination of best practice, the provision of plat-forms for know-how exchange between policy-makers and practitioners, as well as on cross-border networking of clusters. Under the Commission’s Europe INNOVA Initiative, the installation of a High Level Advisory Group on Clusters led to the European Cluster Memoran-dum as a common agenda for cluster policy action signed in January 2008.

National Styles of Cluster Promotion Cluster policies between variety and convergence

Dr Matthias Kiese, Research Associate, Institute of Economic and Cultural Geography, Leibniz University of Hannover, Germany

39


Central or federal – path-dependent learningDespite these significant forces of convergence, struc-

tural and institutional differences between nations and re-gions mean that there is no such thing as a one-size-fits-all cluster policy. Furthermore, cumulative and path-depend-ent learning by doing in policy and practice contribute to persisting variety not only in the interpretation and appli-cation of the cluster concept, but also in the rationales and targets of cluster polices. Cluster policies emerge at the interfaces of hitherto isolated policies, especially science and technology policy, industrial policy, and regional policy which converge into regionalised innovation poli-cy. A first source of international variety is thus the policy area which embraces the cluster notion first and most forcefully, and how well these formerly separate policies are integrated. Further institutional variety may be expected from the role of the state, and a country’s centralised vs. federal set-up.

The role attributed to the public vis-à-vis the private sector is prominently captured by the concept of varieties of capitalism that places market economies on a spectrum reaching from liberal on the one hand to co-ordinated on the other (Hall & Soskice 2001). While the United States are commonly regarded as the archetypal liberal market economy, this model is best represented by the United Kingdom (UK) within the EU. Co-ordinated market econ-omies are epitomised by continental European countries like France or Germany. The degree of centralisation vs. federalism is the third major source of variety since cluster promotion is a typical case of multi-level governance, as it is pursued at all administrative levels from the suprana-tional via nations and regions all the way down to the eco-nomic development efforts of cities and municipalities (cf. Bovaird et al. 2008, Burfitt & MacNeill 2008, Conzel-mann & Smith 2008). Varieties of capitalism and federal-ism are thus expected to directly translate into distinct styles of cluster promotion that would ultimately determine the scope for cross-country policy learning.

To illustrate this variety, three countries have been se-lected as case-studies for cluster promotion at the national level. From a varieties of capitalism perspective, the UK is an obvious choice as it comes closest to the liberal ver-sion of a market economy, while Germany and France are the two largest economies of the ‘old’ continental Europe. On the centralism-federalism axis, Germany was purpose-fully founded as a federal republic, while the UK and France have all embarked on varying decentralisation efforts, but their centralist past is still far from a distant memory.

–––––––– Country Case Studies: Germany ––––––––

To prevent any return to centralism after World War II, Germany was founded as a federal republic in 1949. As a consequence, all 16 federal states pursue autonomous cluster policies today, but for brevity’s sake this paper can only focus on the federal level. Associated with the co-or-dinated nature of its market economy, Germany’s system of innovation is focused on incremental innovation and diffusion, but has comparative weaknesses in radical and breakthrough innovations, such as biotechnology (cf. Casper 2007). German policymakers praise the country’s research excellence, but lament that German inventions such as the MP3 standard are often commercialised abroad. Clusters are hence seen as vehicles to bridge the perceived gap between science and industry to accelerate innovation. However, a too consequent spatial concentra-tion of public resources is at odds with Germany’s tradi-tionally redistributive regional policy, given that spatial equity is a constitutional goal. Unification in 1990 sud-denly increased spatial disparities in productivity and in-novative capabilities. Technological and socio-economic convergence of the new Länder towards the West German level is a special priority of federal government since, and a regionalised innovation policy including the promotion of cluster structures in the new Länder is one way of pursuing this aim.

Federal government started to embrace the cluster concept with the BioRegio contest in 1995, an attempt to jump-start Germany’s embryonic biotech industry. Policy lessons learnt from BioRegio were adapted to promote innovative capabilities in the new Länder with the Inno- Regio initiative in 1999 and the subsequent Entrepreneurial Regions family of programmes. The federal government’s cluster policy received its most recent impetus with the leading-edge cluster competition in 2007, which is part of a broader High-Tech Strategy for Germany.

The BioRegio contest Germany’s federal government got hooked on the

cluster notion in the mid-1990s when trying to promote its fledgling biotechnology industry which was estimated to lag twenty years behind the U.S. and ten years behind the UK at that time (Cooke 2001: 267). The experience of those countries suggested that clusters like San Diego, Boston or the English Cambridge were important sources of national competitiveness in biotech. To close this gap, federal government decided to leverage on the competi-tive potential of federalism. In 1995, the BioRegio contest

40


was launched to identify and promote Germany’s most promising potential biotech clusters (cf. Dohse 2007). 17 regions entered the contest, and in November 1996, Munich, the Rhineland and the Rhine-Neckar area emerged as winners, with a special vote awarded to Jena in the new federal state of Thuringia. The three winners received around 25 million € each over five years, plus preferential access to R&D funding from the federal Bio-technology 2000 programme.

However, it is worth pointing out that not only these winners benefitted from the competition. The contest mo-bilised actors in most other regions, too, with the effect that by the mid-2000s, there were 25 regional networks and cluster initiatives and five state-level associations in charge of regional biotech promotion (BMBF 2005: 5). Some of them received support from subsequent pro-grammes like BioFuture, BioChance and BioProfile (Dohse 2007: 77 f.), but it remains questionable if the almost ubiquitous promotion of biotech networks is the most efficient way of growing internationally competitive clusters. However, the BioRegio contest is now regarded as an important vehicle to jumpstart the biotech industry in Germany which scored spectacular growth in the sec-ond half of the 1990s, although this was helped by legis-lative changes, a favourable business cycle and ample supply of venture capital. Following the burst of the New Economy bubble in 2000/2001, the industry consolidated at around 400 companies with about 10,000 employees (Ernst & Young 2008).

The InnoRegio contestIn the mid-1990s, the initial convergence of the new

Länder vis-à-vis West Germany had come to a halt, and significant disparities in innovative capabilities and eco-nomic wealth threatened to become very persistent. The federal Ministry of Education and Research thus adapted its acclaimed BioRegio model to the specific needs of the new Länder: While BioRegio strove for the mobilisation of regional assets for the benefit of national competitive-ness, the InnoRegio contest was designed to narrow the gap between the eastern and the western states. In contrast to BioRegio, the new contest was not only confined to the new Länder, but also open to all industries and technolo-gies. In 1999, the initial call triggered 444 applications from diverse consortia of individuals and organisations such as businesses, research, education, politics, public administration and associations at the sub-state level (Dohse 2007: 75 f.). Out of these applications, 25 projects

were selected by an independent jury, and 23 ultimately qualified for funding. Convinced by the success of Inno-Regio, the federal ministry differentiated the initial con-cept into a whole new family of programmes called Entre-preneurial Regions (Unternehmen Region) to support innovative networks in the new Länder.

The Spitzencluster competitionIn September 2006, Germany’s federal government for

the first time announced an inter-ministerial high-tech-strategy (BMBF 2006). Of 14.6 billion € earmarked for 2006–2009, 11.94 billion € are designated for a set of 17 industries and technologies, while the remaining 2.66 billion € are reserved for generic measures of innovation policy. Of the latter, 600 million € are earmarked for measures to join the forces of science and industry, of which the leading-edge cluster competition (Spitzencluster-wettbewerb) is the key thrust. The aim is to promote up to 15 already well-developed clusters irrespective of technol-ogy or industry in three rounds over a period of five years. Essentially, this means an extension of the BioRegio con-cept beyond the narrow confines of the industry. Conse-quently, the aims are the same: to identify and strategically promote clusters to achieve leading positions in interna-tional competition, to accelerate the commercialisation of new knowledge, to stabilise and create growth and employ-ment, and to make Germany a more attractive business location. Following the first call for applications in August 2007, 38 regional projects applied by the closing date in December. A dozen of those projects qualified for the final, before the winners of the first round were disclosed in September 2008:

• Cool Silicon – Energy Efficiency Innovations from Silicon Saxony

• Solarvalley Mitteldeutschland• Aviation Cluster in the Metropolitan Region of Hamburg• Forum Organic Electronics in the Rhine-Neckar

Metropolitan Region• Biotech-cluster “cell-based and molecular medicine”

in the Rhine-Neckar Metropolitan Region.

These five winners qualify for a total funding of up to 200 million € over a five-year period. While the first two winners are from the new Länder, the bottom project was filed by one of the BioRegio winners, regional network BioRN. Soon after the first round of selection was com-pleted, the call for the second round was issued in January 2009, with the application period closing in April.

41


Metropolregion Rhein-Neckar - Forum Organic Electronics - Bio RN Cell-based & Molecular Medicine

Metropolregion Hamburg Aviation

Dresden Cool Silicon Energy Efficiency Innovations

Solar Valley Mitteldeutschland

Stuttgart / Neckar-Alb Bioregion STERN Regenerations- biologie

Braunschweig / Göttingen / Hannover Funktionelle Genomanalyse

Potsdam / Berlin Genomforschung & Pflanzenbiologie

BioRiver

BioInstrumente Jena

BioRegion Rhein-Neckar-Dreieck

BioRegio München

MannheimHeidelberg

Ludwigshafen

Braun-schweig

Göttingen

RegionNeckar- Alb

EsslingenReutlingenTübingen

7

13, 14

18

Gardelegen 9

Ohrdruf 16

Barleben 1

Parchim 17

Gatersleben 10

Wismar 22

Rostock 18

Wildau 21

Buch 2Hennigsdorf 12

23

ZwickauFreiberg 8

Markneukirchen 15

3, 4Chemnitz

5 6

Greifswald 11

Schmalkalden 20Bonn

Köln

AachenJena

Schleswig-Holstein

North Rhine-Westphalia

Hesse

Rhineland-Palatinate

Baden-Württemberg

Bavaria

Saarland

Brandenburg

Mecklenburg-Western Pomerania

Saxony

Saxony-Anhalt

Thuringia

Lower Saxony

Stuttgart

Berlin

Bremen

Dresden

Düsseldorf

Erfurt

Hamburg

Hannover

Kiel

Magdeburg

Mainz

München

Potsdam

Saarbrücken

Wiesbaden

Schwerin

BioRegio (1996)InnoRegio (1999)BioProfile (2001)Spitzencluster (2008)

Contest 0 100 km

Cartography: Stephan Pohl

1 MAHREG Innovationsnetzwerk AUTOMOTIVE

2 Gesundheitsregion Berlin-Buch3 Innosachs Innovationsregion

Mittelsachsen4 INNTex Textilregion

Mittelsachsen 5 BioMeT - Innovationsnetzwerk

Dresden6 KONUS - Kooperative Nutzung

von Datennetzen7 Bautronic8 RIST: Mit neuen Netzen zur

Wissensregion 9 NinA Naturstoff-

Innovationsnetzwerk Altmark10 InnoPlanta

Pflanzenbiotechnologie Nordharz/Börde

11 Diabetes Informations- und Service-Center DISCO

12 RIO - Regionales Innovationsbündnis Oberhavel

13 Innovationspotential Börde (REPHYNA)

14 INNOMED Innovative Medizintechnologien

15 Musicon-Valley16 Barrierefreie Modellregion für

Integrativen Tourismus17 NUKLEUS - Netzwerk

Präzisionsmaschinenbau18 BioHyTec19 Maritime Allianz20 inprosys21 FIRM22 Kunststoffzentrum

Westmecklenburg23 IAW 2010 - Industrie- und

Automobilregion Westsachsen

The winners of Federal Government cluster contests in Germany, 1996–2008

42


France’s Pôles de Compétitivité, 2008

0 100 200 km

0 0 0200 km 50 km 50 km

St-Denis 52

French Guiana Guadeloupe

Reunion

Source: DGE/DIACT, quoted in Longhi 2008, p. 33 Cartography: Stephan Pohl

Montpellier69

Epinal 45

Chartres 8

Metz 50

Mulhouse 56

Besançon 49Dijon 58, 67

Saint-Étienne 57, 71

Pont-Saint-Esprit 53

Grasse 64

Oyonnax 28

Annecy 40

Marseille 25, 26, 27

Avignon 51

Bordeaux 32, 47, 68

Limoges 11, 29

Clermont-Ferrand 60

Chappes 44

Poitiers 63

Orléans 10

Cholet 65

Caen 37Mondeville 13

Rennes 38, 42

Arques 62Boulogne-sur-Mer 66

Le Havre 61

Tours 70

Perpignan 9

Aix-en-Provence 46

Strasbourg 3

Lyon 5, 21 / 16, 20, 35

Grenoble 22, 36

Sophia Antipolis-Rousset 33

Toulon 31

Toulouse2, 43 / 1

Brest 30

Lannion 15

Angers 55

Rouen 23

Paris

Laon 18

Lille 19 / 17, 24, 54

Nantes 12, 41, 59

6, 14, 34, 48 / 4, 39

St-Paul-lez-Durance 7

global clusters

national clusters

1 Aerospace Valley2 Agrimip Innovation3 Alsace Biovalley4 Astech Paris Région5 Axelera6 Cap Digital Paris Région7 Capénergies8 Cosmetic Valley9 Derbi10 Elastopôle11 Elopsys12 EMC213 Filière équine14 Finance innovation15 Images et réseaux16 Imaginove17 Industries du commerce18 Industries et agro-ressources19 i-Trans Vocation20 Lyon Urban Truck&Bus21 Lyonbiopôle22 Minalogic23 MOV‘EO24 Nutrition Santé Longévité25 Optitec26 Orpheme

27 Pégase28 Plastipolis29 Pôle européen de la céramique30 Pôle Mer Bretagne31 Pôle Mer PACA32 Route des lasers33 Solutions communicantes sécurisées (SCS)34 System@tic Paris Région35 Techtera36 Tenerrdis37 Transactions électroniques sécurisées (TES)38 Valorial39 Ville et mobilité durables40 Arve Industries41 Atlantic Biotherapies

42 Automobile haut de gamme43 Cancer-Bio-Santé44 Céréales Vallée45 Fibres Grand‘Est46 Gestion des risques47 Industries et pin maritime du futur48 Medicen Paris Région49 Microtechniques50 Matériaux innovants produits intelligents (MIPI)51 Pôle européen d‘innovation fruits et légumes52 Qualitropic53 Trimatec54 Up-Tex55 Végépolys56 Véhicule du futur

57 Viaméca58 Vitagora59 Génie civil Ouest 60 InnoViandes61 Logistique Seine Normandie (Nov@log)62 Matériaux à usage domestique (MAUD)63 Mobilité et transports avancés64 Parfums, arômes, senteurs, saveurs (PASS)65 Pôle Enfant66 Pôle filière produits aquatiques67 Pôle Nucléaire Bourgogne68 Prod‘Innov69 Q@LI-MEDéditerranée70 Sciences et systèmes de l’énergie électrique (S2E2)71 Sporaltec

43


Policy learning – yet conflicting aimsFederal innovation policy in Germany has firmly em-

braced the notion that national competitiveness depends on localised assets. To unleash the hitherto underutilised potential of competitive federalism, the federal govern-ment employs contests as a device for discovery and mo-bilisation, and shows signs of cumulative policy learning when differentiating initial concepts like BioRegio and InnoRegio into entire programme families. Owing to the power of the Länder in Germany’s federal set-up, federal government acts as a facilitator by organising competi-tions and selecting regions, but does not intervene in state policies, nor is it actively involved in programme man-agement which is left to the federal states or to independ-ent agencies. However, it may be criticised that the pre-vailing approach promotes intraregional networks at the expense of interregional and international networking, and that the InnoRegio family to promote innovation net-works in the new Länder is intrinsically trapped between the conflicting aims of growth and competitiveness on the one hand, and spatial equity on the other.

––––––––– Country Case Studies: France –––––––––

In sharp contrast to Germany’s federal system and poly-centric spatial structure, France has a long history of cen-tralism and dominance of its capital region Île de France. This is still evident today despite the decentralisation thrust initiated in 1982. While industrial policy tradition-ally promoted large firms as national champions, technol-ogy policy used to follow the linear model of innovation through Grandes Programmes or the open science model (Brette & Chappoz 2007). In regional policy, France pio-neered the promotion of growth poles in a tradition linked to the works of Francois Perroux in the 1950s, but that approach fell into disgrace with the demise of industrial policy in the 1980s (Longhi & Rainelli 2007: 2 f.). More recently, the government responds to perceived weakness-es in high-tech industries as highlighted by Beffa (2005), a lack of collective processes involving firms, universities and research, and to the Lisbon agenda by e.g. adding a competitiveness dimension to its regional policy from 1999 and 2002 onwards. Central government in France first embraced the cluster approach with a programme to support local production systems in 1998 and responded to the German BioRegio example with a Genopole initia-tive to spur the commercialisation of life sciences in 1999, before embarking on its massive Pôles de Compétivité programme in the mid-2000s.

The Systèmes Productifs LocauxThe Systèmes Productifs Locaux (SPL) programme

was launched as part of French regional policy in 1998 to promote clusters of small and medium-sized enterprises (SMEs) in low-technology sectors and in peripheral areas (OECD 2007). Such industrial districts were required to exhibit a concentration of activities in a specific industry, a high level of inter-firm linkages, supporting infrastruc-ture and operators qualified to stimulate interaction. Out of 180 business plans submitted by co-ordination bodies, 96 were initially selected, with two more specific calls to follow in 2001 and 2003, respectively. A typical SPL con-sists of around 100 SMEs, of which between 30 and 40 actually participate in collaborative projects. Co-operation with a university or research institute is encouraged but no funding requirement. Central government funds up to 3.6 million € per SPL for collective management expenses to engage local actors, but SPL receive most of their fund-ing from other sources. Many of them are networked in the Club des Districts Industriels Français (CDIF), a vol-untary association of French industrial districts.

The Genopole programmeAkin to Germany’s BioRegio initiative, the French

government embarked on its Genopole programme in 1999 to spur the commercialisation of life sciences. The support of innovation networks was expected to boost France’s competitive position in life sciences through scien-tific advances and improved science-industry relationships (Quéré 2008). Following a call for applications in 1999, eight Genopoles were selected in 2000, and a further two in 2001. The government spent around 75 million € on the establishment of technological platforms for co-operation over the first three years and promoted a nationwide divi-sion of labour between the ten specialised Genopoles. Three main assessments carried out between 2001 and 2003 found that the initiative did have some impact on the science infrastructure and improved the networking of scientists. However, the programme’s economic impact appeared unsatisfactory since networks were dominated by academic actors and raised output mainly in the form of academic publications. Most Genopoles outside the Île de France region still lack a critical mass of firms and hence failed to mobilise public funding.

The Pôles de Compétitivité First outlined by the French regional planning Author-

ity DATAR (2004, now DIACT = Interministerial Delega-tion for Territorial Competitiveness and Attractiveness),

44


the Pôles de Compétitivité programme was originally designed to improve the international competitiveness of 10–15 clusters. However, a call for proposals in September 2004 stimulated 105 project applications, of which 67 clusters were ultimately selected. This figure was further increased to 71 in a second round in 2007. The expansion of the original programme design meant that the orienta-tion towards international competitiveness had to be com-promised. As a result, only six clusters were classified as international, with ‘international orientation’ attributed to a further nine, adding up to the number originally targeted. The remainder consists of 15 inter-regional and 37 regional clusters. Spatially, the programme was expanded from the most competitive clusters nation-wide to include applicants from almost all regions across the country, which means a shift from national to regional objectives. Sectorally, this inflation led to the inclusion of industries like meat processing and construction (OECD 2007: 191). It is no surprise that the most advanced “global competitiveness poles” can be found in those regions commonly known for their outstanding technological capabilities, i.e. Île de France (Medicin Paris Region, System@tic, Finance), Rhône-Alpes (Lyonbiopole, Minalogic), Midi-Pyrénées (Aerospace Valley) and the Provence-Alpes-Côte d’Azur (Secured Communication Solutions).

A Pôle typically includes up to a handful of very large firms, up to hundreds of SMEs, and three to four institu-tions of higher education and research institutions, respec-tively. Responding to the expansion of the programme, central government doubled its funding from 750 million € originally earmarked for 10–15 international clusters to 1.5 billion € for 67 Pôles over three years, which may be topped up by EU, regional and local government funds. 80% of this funding is reserved for the 15 clusters with at least international orientation, while only 11 million € go into the cluster management of the Pôles, averaging around 55,000 € per cluster and year. The bulk of the money is used as competitive funding for projects accessible for consortia of at least two firms and one research institute from a territory designated as Pôle, the demarcation of which was part of the Pôle’s initial application (Longhi & Rainelli 2007: 10).

Evaluation of the Pôles programmeBetween November 2007 and June 2008, the Pôles

programme has been subject to an official evaluation (BCG & CM International 2008). It was found that 54.9% of the Pôles achieved their policy targets fully, while a further 26.8% achieved them partially at least. The re-

maining 18.3% of Pôles were found not to have met their targets, they were thus recommended for a reconfigura-tion. While Pôle status will expire for the latter group af-ter one grace year at the end of 2009, the first two groups will remain eligible for funding under the Pôles 2.0 pro-gramme that is endowed with another 1.5 billion € from 2009 to 2011.

Academics are more critical about the Pôles pro-gramme than the official evaluation commissioned by the French government. Longhi & Rainelli (2007) acknowl-edge a positive mobilisation effect since most Pôles did not have formal co-operation structures prior to their ap-plication. Furthermore, the programme has improved poli-cy co-ordination within and across the various level of government, including a diffusion of knowledge on clus-ter promotion. Their main critique, however, surrounds the inflation of clusters supported by the programme. A committee staffed mainly by civil servants rather than business representatives selected many former low-tech SPL which have thus been relabelled Pôles, while other Pôles would rather suit the definition of SPLs rather than the original understanding of Pôles. As a result of their heterogeneous economic and innovative capabilities, the 71 clusters scored very differently on the generation of R&D projects. Further critique include the tight time-frame of the application process, the dominance of large firms in many clusters, different expectations especially between large firms and SMEs, governance structures complicated by a multiplicity of oversight bodies, as well as a lack of inter-cluster networking. To improve their chance for selection, applicants moved towards the forma-tion of larger areas, leading to a gradual shift from region-al towards national clusters. In sum, the Pôles de Com-pétitivité programme integrates top-down elements (initiation, selection of Pôles) with a bottom-up approach, but the decision on R&D funding remains a centralised and very demanding multi-step process: “The label pole is not a guarantee of financing, it is only a prerequisite to compete for subsidies” (Longhi/Rainelli 2007, S. 11). Despite a quarter-century of decentralisation effort, the French government’s recent cluster policy is still marked by a considerable degree of centralism.

––––– Country Case Studies: United Kingdom –––––

Within the EU, the United Kingdom is the most promi-nent case of a liberal market economy. While centralism reached its latest climax under the Thatcher administra-tion, New Labour engaged in a major devolution effort

45


from 1997 by decentralising powers to the Devolved Administrations (DAs) in Scotland, Wales and Northern Ireland, as well as to the Regional Development Agencies (RDAs) in England. In general, UK policy has a strong focus on improving labour productivity by removing ob-stacles to innovation, skills and infrastructure. As part of this overarching thrust, regional policy focuses on improve-ments in lagging regions which are seen as a drag on national productivity and income levels in the light of an ever-deepening North-South divide.

DTI – top-down mapping and best practice manualWhile Scottish Enterprise has been a European pioneer

of cluster policies in the early 1990s (cf. Brown 2000), Whitehall did not open its doors for clusters until the De-partment of Trade and Industry’s 1998 Competitiveness White Paper included an enquiry into the UK’s biotech clusters. From 1999 to 2003, a High-level Cluster Policy Steering Group looked at barriers to cluster development and recommended appropriate policy measures. This was accompanied by a top-down cluster mapping exercise that identified a total of 154 ‘clusters’, amounting to between eight and 18 per UK region. These clusters accounted for 43 per cent of regional employment in London, but only 15 per cent in the North West (DTI 2001: 8 f.). The DTI’s 2001 White Paper “Opportunity for all in a World of Change” encourages the newly-formed RDAs to develop existing and embryonic clusters in their regions. The DTI stresses that the focus should be on creating favourable conditions for the formation and growth of clusters, but not on attempting to create clusters artificially. It promotes clusters to RDAs as a key instrument to support business excellence and to promote innovation. Initially, the DTI tied its funding of RDA cluster initiatives to the results of its 2001 mapping exercise. However, following two years of consultation the RDAs decides to pursue cluster poli-cies autonomously of the DTI to better suit their regional needs (Borrás & Tsagdis 2008: 90). To provide guidance, the DTI (2004) issued a best practice manual on cluster development which highlights functioning networks and partnerships, a strong innovation base with supporting R&D activities and a strong skills base as the most critical success factors.

RDAs – a range of different cluster initiativesThe flexibility of the DTI-RDA framework has encour-

aged a rather diverse set of regional initiatives ranging from single RDA projects to collective cluster initiatives involving the joint effort of several RDAs such as Motor-

sport Development UK which pulls together the four RDAs covering England’s Motorsport Valley (cf. Aston 1998). Cluster initiatives supported by RDAs include high-tech and more traditional industries alike and com-bine a range of activities from regional mapping studies via the identification and forging of links with important regional clusters to employing clusters as vehicles for wider economic development initiatives (OECD 2007: 318).

Flexibility at the expense of strategic coherenceMirroring the French situation, the UK way of cluster

promotion retains a significant degree of central control despite noticeable devolution. It is through their funding that regional institutions such as DAs and RDAs remain “closely linked to and strongly associated with central government” (ibid: 315). However, there are significant differences to France: In line with the UK’s classification as a liberal market economy, cluster policy does not in-volve top-down prescriptions but rather provides a flexible framework for cluster promotion through the regional agencies. The cluster concept is used intensively for map-ping and localisation of industrial dynamics, but less for specific policy action. Although the UK pioneered the use of the cluster idea through Scottish Enterprise in the early 1990s, cluster policy remains “a peripheral isolated area of piecemeal policy development” (Borrás & Tasgdis 2008: 89). RDAs are the lead agencies in charge of cluster promotion, but since they are fraught with a wide range of tasks, cluster policy has to compete with issues like lifelong learning or the ageing workforce for attention. Under a liberal framework, the greater flexibility in cluster promo-tion obviously comes at the expense of strategic coherence.

–––––– National Styles of Cluster Promotion ––––––

What scope for cross-border policy learning?Cluster promotion represents a textbook case of multi-

level governance involving supra-national, national, re-gional and local actors and hence the interaction of top-down and bottom-up policies. This is evidenced by all country cases studied in this paper, irrespective of their variety in terms of capitalism and federalism. Despite decentralisation efforts in the UK and France, top-down elements in the cluster policies of both countries remain stronger than in an originally federal country like Germany. Varieties of capitalism help explaining why the UK ap-proach provides a flexible framework for cluster promo-tion, but lacks strategic coherence. However, this does not necessary imply that the more co-ordinated efforts of

46


Germany and France are more effective or even more effi-cient. More research and independent evaluation is clearly needed to investigate this connection. Porter-style cluster policies with a lot of mapping and little intervention pre-vails in the UK to raise productivity, while national cluster policies in Germany and France follow the triple-helix model of university- industry-government relations, using clusters as a vehicle to accelerate the commercialisation of new technological knowledge.

Institutional context and policy traditions matterEspecially the comparison of Germany and France

shows that learning through mutual observation has already taken place. For instance, French Genopoles are a reflec-tion of the older German BioRegio initiative but failed to boost the commercialisation of research. Conversely, Germany’s leading-edge cluster competition may well be inspired by the French Pôles de Compétitivité, but was not subject to the same pressures that led to the inflation of clusters. In Germany, lobbyism and the specific rationali-ties of politicians and bureaucrats are absorbed by Länder policies where cluster inflation is commonplace. Take tiny Brandenburg for example: With only 2.5 million inhabit-ants, the federal state surrounding Berlin embarked on an ambitious policy in 2006 targeting no less than 16 clusters.

Despite globalisation and strong forces of convergence, national varieties of cluster promotion exist and tend to prevail. Learning from another country’s experiences thus requires a careful look at that country’s institutional con-text and policy traditions. As a first approximation, institu-tional context may be read as the relative weight attached to public vs. private initiative captured by the varieties of capitalism approach, while policy traditions can be located on a highly persistent scale from centralism to federalism. Taking these differences into account may help to extract valuable lessons from seemingly incomparable settings. However, the current quest for best practice lessons tends to overlook the heuristic value of failed attempts and poor practice – which are by the way more easily transferable across borders. (English editing by the author) ::

ReferencesAston, B., 1998: Advanced Auto-Engineering: A Technology-Led UK Cluster. In: Business

Strategy Review, 9(1): 45–53.BCG; CM International, 2008: Evaluation des Pôles de Compétitivité: Synthèse du rap-

port d’évaluation. http://www.competitivite.gouv.fr/IMG/pdf/synthese_BCG-CMI_evaluation_des_poles_de_competitivite.pdf, accessed 10.02.2009.

Beffa, J.-L., 2005: Pour une Nouvelle Politique Industrielle. http://lesrapports.ladocumentationfrancaise.fr/cgi-bin/brp/telestats.cgi?brpref=

054000044 &brp_file=0000.pdf, accessed 09.02.2009.BMBF, 2005: BioRegionen in Deutschland. Starke Impulse für die nationale Technologie-

entwicklung. (3rd ed.) Berlin: Federal Ministry for Education and Research.BMBF, 2006: Die Hightech-Strategie für Deutschland. Berlin: BMBF.Borrás, S.; Tsagdis, D., 2008: Cluster Policies in Europe: Firms, Institutions, and Govern-

ance. Cheltenham: Edward Elgar.Bovaird, A. G.; Löffler, E.; Parrado-Díez, S. (eds.), 2008: Multi-level Governance:

New Approaches to Decentralising Power in European Countries. Cheltenham: Edward Elgar.

Brette, O.; Chappoz, Y., 2007: The French Competitiveness Clusters: Toward a New Public Policy for Innovation and Research? In: Journal of Economic Issues, 41(2): 391–398.

Brown, J.S.; Duguid, P., 1991: Organizational Learning and Communities of Practice: Toward a Unified View of Working, Learning, and Innovation. In: Organization Science, 2(1), S. 40–57.

Brown, R., 2000: Cluster Dynamics in Theory and Practice with Application to Scotland. (=Regional and Industrial Policy Research Paper, 38). Glasgow: EPRC.

Burfitt, A.; MacNeill, S., 2008: The Challenges of Pursuing Cluster Policy in the Congested State. In: International Journal of Urban and Regional Research, 32(2): 492–505.

Casper, S., 2007: Creating Silicon Valley in Europe: Public Policy Towards New Tech-nology Industries. Oxford: Oxford Univ. Press.

Conzelmann, T.; Smith, R. (eds.), 2008: Multi-Level Governance in the European Union: Taking Stock and Looking Ahead. Baden-Baden: Nomos.

Cooke, P.N., 2001: New Economy Innovation Systems: Biotechnology in Europe and in the USA. In: Industry and Innovation, 8(3): 267–289.

DATAR, 2004: La France, Puissance Industrielle – Une Nouvelle Politique Industrielle par les Territoires: Réseaux d‘Entreprises, Vallées Technologiques, Pôles de Compétitivité. http://lesrapports.ladocumentationfrancaise.fr/cgi-bin/brp/telestats.cgi?brp_ref= 044000090&brp_file=0000.pdf, accessed 09.02.2009.

Dohse, D., 2007: Cluster-based Technology Policy: The German Experience. In: Industry and Innovation, 14(1): 69–94.

DTI, 2001: Business Clusters in the UK: A First Assessment. London: DTI.DTI, 2004: A Practical Guide to Cluster Development: A Report to the Department of Trade

and Industry and the English RDAs by Ecotec Research & Consulting. London: DTI.Ernst & Young, 2008: Biotech-Branche mit stabilem Wachstum. Press Release, Frankfurt/

Main, 7. Mai 2008.Etzkowitz, H.; Leydesdorff, L., 2000: The Dynamics of Innovation: From National Sys-

tems and “Mode 2” to a Triple Helix of University-Industry-Government Relations. In: Research Policy, 29(2–3): 109–123.

Hall, P. A.; Soskice, D., 2001: An Introduction to Varieties of Capitalism. In: Hall, P.A.; Soskice, D. (eds.): Varieties of Capitalism: The Institutional Foundations of Compara-tive Advantage. Oxford: Oxford Univ. Press, 1–68.

Hospers, G.-J.; Beugelsdijk, S., 2002: Regional Cluster Policies: Learning by Comparing? In: Kyklos, 55(3): 381–402.

Longhi, C., 2008: The French Regional Policy in the 21st Century: Old Wine in New Bottles? Unpublished paper prepared for the Korea Development Institute International Conference on “International Experiences of Regional Policy and Policy Implications for Korea,” Seoul, July 15–16, 2008.

Longhi, C.; Rainelli, M., 2007: Poles of Competitiveness, a French Dangerous Obsession? mimeo, Sophia Antipolis.

Martin, R.A.; Kitson, M.; Tyler, P. (eds.), 2006: Regional Competitiveness. London: Routledge.

OECD, 1996: The Knowledge-based Economy. Paris: OECD.OECD, 2007: Competitive Regional Clusters: National Policy Approaches. (=OECD

Reviews of Regional Innovation). Paris: OECD.Porter, M.E., 1990: The Competitive Advantage of Nations. New York: Free Press.Porter, M.E., 1998: Clusters and Competition: New Agendas for Companies, Governments

and Institutions. In: Porter, M.E.: On Competition. (=The Harvard Business Review Book Series). Boston: The Harvard Business School Publishing, 197–287.

Quéré, M., 2008: Innovation Networks in the Life Sciences Industry: A Discussion of the French Genopoles Policy. In: European Planning Studies, 16(3): 411–427.

Siebert, H., 2005: Locational Competition: A Neglected Paradigm in the International Division of Labour. (=Kiel Working Paper, 1258). Kiel: Institute of World Economics.

Sölvell, Ö.; Lindqvist, G.; Ketels, C.H.M., 2003: The Cluster Initiative Greenbook. Gothenburg: Ivory Tower AB.

Stone, D., 2004: Transfer Agents and Global Networks in the “Transnationalization” of Policy. In: Journal of European Economic Policy, 11(3): 545–566.

van der Linde, C.M., 2005: Cluster und regionale Wettbewerbsfähigkeit: Wie Cluster ent-stehen, wirken und aufgewertet werden. In: Cernavin, O.; Führ, M.; Kaltenbach, M. (eds.): Cluster und Wettbewerbsfähigkeit von Regionen: Erfolgsfaktoren regionaler Wirtschaftsentwicklung. (=Volkswirtschaftliche Schriften, 543). Berlin: Duncker & Humblot, 15–34.