Stretching—Blending and Hubbing
The cluster has already expanded from its original concept and, in consequence, a stretching of the definition has occurred, resulting in growing detachment from Porter’s concept (Desrochers & Sautet, 2004; Floysand, Jakobsen & Bjamar, 2012; Malmberg & Power, 2006; Martin & Sunley, 2003; Todtling & Trippl, 2005). Stretching can materialise via hubbing, i.e. expansion in geographical scale, and blending, i.e. expansion in industrial scope (Njos et al„ 2017a).
It is likely that stretching processes might be further compounded by the implications of the fourth industrial revolution (Gotz & Jankowska, 2017).
The twin strategies of hubbing and blending (i.e. expanding geographically and sectorally) depict the presumed modification of the cluster concept we might face in the light of 14.0. It should be noted that ‘stretching’ bears resemblance to the term of ‘global cluster network’, coined by Batlielt and Li (2014), which covers the system of ties and relations among firms located in clusters. It refers to both trans-local linkages between industrial clusters, but also local linkages within the same cluster. As argued by Turkina, Van Assche and Kali (2016), the recognition in the literature that the network of both local and trans-local (external) linkages are essential for a cluster firm’s access to knowledge, encouraged scholars to go beyond the simple local-global dichotomy and adopt a network view of industrial clusters.
Internationalisation can be a chaimel facilitating more cluster heterogeneity, allowing the incorporation of the unrelated knowledge, preventing a lock-in situation and helping to sustain innovativeness and growth (Dohse, Fomahl & Vehrke, 2018). The need of heterogenous external knowledge can be derived from the cluster-life-cycle (CLC) model (Brenner & Schlump, 2011; For- nahl, Hassink & Menzel, 2015; Menzel & Fomahl, 2010). It assumes that the cluster goes through different developmental stages (emergence, growth, sustainment and decline). As argued by Buciuni and Pisano (2015), clusters can decline for two primary reasons. First, because other competing clusters can gain an advantage (out-perform these clusters) or second, because agglomeration economies would dissipate, due to changes in technology, competitive dynamics or some other reason. Similarly, as shown by Turkina and Van Assche (2018), the cluster’s technological base is far more than previously subject to the resident firms’ capabilities of sourcing the distant knowledge. Mien talking about stretching, it should be mentioned that Chapman, MacKinnon and Cumbers (2004) distinguish between geographical diversification and sectoral diversification as forms of cluster renewal (adaptability leading to change and continuity, simultaneously). It thus can be considered as stimulated by the influx of novelty from other sectors within a region or through extra-regional linkages (Njos & Jakobsen, 2016).
The processes of hubbing and blending could also be seen hi the context of inter-clustering—still a nascent research field (Lorenzen & Mudambi, 2013; Goerzen. 2018). Inter-clustering might be regarded as a specific form of increasing the geographical scope and/or sectoral scope. Franco and Esteves (2020) see inter-clustering centred around knowledge sharing and learning as specific inter-organisational relations, which contributes to regional competitiveness (Dohse, 2007; SchiiBler, Decker & Lercli, 2013). Inter-clustering is also perceived as a synonym for co-operation (Cusin & Loubaresse, 2018), providing benefits for participating clusters (SchiiBler, Decker & Lercli, 2013).
Stretching processes might be also seen as the accessing and sourcing of different knowledge bases (synthetic, analytical or symbolic) from different geographical scales (from regional to global). It can happen thanks to different mechanisms—market forces, networks (e.g. alliances), spill-overs (e.g. mobility) and hierarchies (e.g. FDI) (Bellandi, Chaminade & Plechero, 2018).
The processes of blending and hubbing described below also mirror some calls (Park, 2018) that innovative cluster policies must respond to new industrial challenges through facilitating cross-sectoral value chains and strengthening internationalisation. Stretching (hubbing, but also to some extent blending, providing it involves an external dimension) might be seen as a strategic coupling of regional and extra-regional assets, which enables the path creation and emergence of new economic activities (MacKinnon, Dawley, Pike & Cumbers, 2019; Chandrashekar & Subralmianya, 2019).
Blending strategy aims at the upgrading of the cluster and buttressing the innovation capabilities of incumbent firms through facilitating the mixing of different but somehow related competencies (Gancarczyk & Bohatkiewicz,
2018). This strategy implies expanding the industrial scope of the cluster by stimulating the collaboration between firms in related branches and those with different but similar knowledge. It stresses the regional dimension and can be associated with RV (Boschma & Frenken, 2011; Cooke, 1992; Cooke, Uranga & Etxebania, 1997; Uyarra, 2010).
Blending means the co-operation between related entities within a region (Njos & Jakobsen, 2016; Njos et al„ 2017a). It is not only linked to the concept of RV (Frenken, Van Oort & Verburg, 2007), but also regional branching (Boschma & Frenken, 2011; Boschma & Giannelle, 2014), and stresses different dimensions of proximity, such as cognitive and organisational, rather than purely industrial specialisation and (only) geographical vicinity. The region is in the centre of this strategy. Blending strategy stimulates knowledge spill-overs between related sectors and then players, i.e. it encourages the cross-industry innovation (Enkel & Gassmann, 2010) and facilitates ‘mixing’ of different but associated competences. Blending should lead to the expansion of the cluster’s industrial scope by promoting co-operation, knowledge exchange and learning between companies representing related branches. This strategy, however, runs the risk of stimulating ‘unproductive’ networking (too many differences). Blending is meant as a counter-balance to traditional sector specialisation (Cooke, 2012) as it backs a more diverse system. In consequence, it modifies the definition of cluster, stipulating that this is an agglomeration of units belonging to related industries.
Hubbing strategy means that a cluster creates new connections or assemblage points outside the original cluster core area, and as such, is linked to the exploitation of geographic scale. This process enables the development of specialised clusters by building extra-regional pipelines and establishing relations with specialised external actors. Hubbing can be associated with the concept of global pipelines, regarded commonly as drivers of innovation (Bathelt,
48 Knowledge, Business, Policy in Cluster
Malmberg & Maskell, 2004) and learning, thanks to their role of connecting the highly competent actors of innovation systems (Malerba, 2002).
Hubbing denotes the geographical expansion of cluster linkages, i.e. the increase of the geographic areas of impact (Njos & Jakobsen, 2016). Hubbing strategy underscores the building of external linkages, based on a cluster’s sector-specific expertise and utilisation of scale economies. Hubbing brings the risk that fostering extra-regional specialised ties happens at the expense of building linkages among local firms in related branches. Hubbing., nevertheless, allows the cluster to establish ‘satellites’ or ‘nodes’ in external, highly relevant national or even international milieux. These extra- regional networks may stimulate the innovativeness, learning and development, but may pose a challenge to encourage and maintain a local buzz.
In this place, the relations between internationalisation of the cluster and the stretching processes, in particular, the hubbing should be accurately defined (Gotz, 2020). Internationalisation, as presented commonly in existing literature, encompasses mostly, the processes of foreign expansion by local constituting firms, either in the form of the export or a more advanced mode of an international venture, including FDI, or by attracting new investors from abroad. CO may obviously, and usually does, play an essential role in facilitating these relations. Hubbing is understood as a geographical expansion of the cluster, as gaining new members and new territory. Here, the ‘reaching out’ is more the search for sustained/increased competitiveness, rather than the result of revealed competitiveness. This activity, though, might be considered as a particular type/mode of cluster internationalisation along export or FDI lines. It is clear that the relationship between pure internationalisation and hubbing is complicated and requires some de-limitation (Figure 4.6).
Figure 4.6 Internationalisation vs. hubbing Source. Author’s own proposal
In many studies so far, the external cluster linkages seem to have been understood in terms of internationalisation (Njos et al., 2017a, b). The importance of extra-regional ties for cluster evolution requires taking into account nuances, such as context, as well as the diversified role of MNEs and their relations. By investigating one of Norway’s strongest industry clusters, the sitb-sea cluster in Hordaland county, Njos et al. (2017a) demonstrate that MNEs-out (cluster inhabiting firms which venture abroad) bring about more specialisation to cluster renewal, by contrast, MNEs-in (incoming foreign enterprises) contribute to diversification. In light of the above, it is justified to argue that ‘MNEs-in’ and ‘MNEs-out’ satisfy the basic definition of cluster internationalisation (Jankowska & Gotz, 2018). Yet, it contributes to the cluster adaptability and renewal, i.e. simultaneous processes of continuation (extension of practices) and change (novelty), by ensuring some balance between specialisation and diversification (Njos et al., 2017a) (Figure 4.7). It can, however, also be provided, thanks to the stretching—/nibbing, and most probably, blending (both related and unrelated variety).
Grillitsch et al. (2018) argue that positive regional/structural change, which can happen through the industrial development path, is subject to the ‘opportunity space’. This ‘opportunity space’, i.e. the room for manoeuvre for makeover of the region, depends on whether the region is characterised by the specialisation (interactions traded and non-traded in a given field, including learning and innovation), the RV (potential of diversification drawing on similar knowledge) or the unrelated variety> (combination of
Figure 4.7 Internationalisation and hubbing of the cluster Source. Author’s own proposal analytical, synthetic and symbolic knowledge), and whether the potential change is confined to the concrete geographic location, or it can happen in an abstract economic space (which allows for some external linkages). According to Grillitsch et al. (2018), it is, in fact, this under-appreciated concept of unrelated variety (typical for metropolitan areas), which seems to offer most possibilities for the industrial development path. Findings of research on the ongoing digital business transformation allow us to claim that this fourth industrial revolution might constitute a particular evolution from ‘specialisation and concrete localisation’ towards ‘unrelated variety and the growing importance of abstract space’. In other words, 14.0 could imply the transition of clusters (understood as specialised geographically and concentrated) towards diversified and more spatially abstract concepts; thus, supporting the hypothesis of more diversified and less focused clusters in the era of 14.0.
Digital technologies and infrastructure can enhance, extend and enrich interactions among economic agents (Autio & Thomas, 2016), thereby influencing the stretching processes. The digital disruption materialises via ‘digital affordances’—defined as possibilities to perform existing functions more effectively or to perform entirely new functions (Autio, Nambisan, Thomas & Wright. 2018; Autio, 2017). These affordances can drive the business model innovation (BMI) and emergence of a distinctive type of cluster—entrepreneurial ecosystem (see also Feldman, Siegel & Wright,
2019). Whereas classic clusters focus on spatial affordances, entrepreneurial ecosystems are built upon digital affordances. It means that an entrepreneurial ecosystem constitutes such a cluster type, which is not specific to a particular sector or technology (Autio, Nambisan, Thomas & Wright, 2018). Entrepreneurial ecosystems, in contrast to classic clusters, put the emphasis on the exploitation of so-called digital affordances. They are organised around the discovery and pursuit of entrepreneurial opportunity and concentrate on BMI. Entrepreneurial ecosystems foster voluntary horizontal knowledge spill-overs and searching for cluster-external entrepreneurial opportunities. Although, not all, many features of entrepreneurial ecosystems seem to resemble the combination of RV and IC explored in this volume.
The issues incorporated in this volume seem to relate to the four aspects identified by Hassink, Isaksen and Trippl (2019), as critical for regional industrial path development. ICs might be linked to the (1) ‘expectations’, and to (2) ‘how non-firm actors such as users, universities, intermediaries and policy actors shape the regional development'. The problem of (3) ‘inter-path relations and interdependencies between multiple established paths, established paths and new paths, as well as multiple emerging paths’ may all offer insight to the concept of RV and blending processes as
Developing a Conceptual Model 51
Figure 4.8 Key concepts which frame the analysis Source: Author’s own proposal
discussed in this volume. Hubbing, meanwhile, might be seen as reflecting the need for more attention to be paid to the (4) ‘multi-scalarity of sources, relations and influences’.
Summing up, the framework for discussion is built around the concepts of (Figure 4.8):
- • the RV, associated with such terms as ‘smart or diversified specialisation’ (McCaim & Ortega-Argiles, 2014; Foray, David & Hall, 2009; Foray, 2014) or ‘territorial servitisation’, which implies local execution of various GVC activities, rather than single specialised slices.
- • the ICs, which are understood as locally bound resources, agglomeration externalities, which draw on the triple helix concept (Leydesdorff. 2012; Etzkowitz, 2012).
If ICs are viewed as local externalities, scale economies or middle-gr ound activities; then RV, understood in terms of dynamic, complementary externalities, including the knowledge flows, might be seen as part of IC; or even as a specific section of IC.