PRELIMINARY ROADMAP
IT-enabled Growth Nodes in Europe:
Concepts, Issues and Research Agenda
(draft version)
January 2003
1. Introduction
This paper develops a research agenda for the investigation of knowledge management processes and the use of related information and communication technologies (ICTs) to foster “growth nodes” and emergent strategic growth opportunities within European regions. It is based on-going research in G-NIKE (Growth Nodes in a Knowledge Based Europe), a project sponsored by the European Commission, under the IST Programme.
G-NIKE focuses on the nature and dynamics of ICT-enabled ‘growth-nodes’ in Europe. The working hypothesis is that the development of future competitiveness in the European Union will happen through the emergence of inter-connected clusters (growth nodes) with higher than average economic growth rates, including success in equity issues and social welfare. G-NIKE studies regional and interregional growth-nodes in order to understand their role in regional development. The term ‘growth-node’ was chosen instead of ‘cluster’ or ‘growth pole’ because the focus of the research is not only be on the interrelatedness within different clusters but also on the interrelatedness between them.
The structure of the document is as follows. Firstly, we review clusters as precursors to growth nodes. We position the growth node concept as an augmented cluster. Then, we address a number of fundamental issues associated with growth nodes: Do growth-nodes exist? If so, why do they matter? How do they contribute to regional development? How can we identify them? What are the conditions to develop them? What are the desired properties of the enabling ICT infrastructure? What policies are most likely to foster them? How can they become a policy instrument?
We review what is known about clusters that may be potentially relevant, as well as what is not known but would be necessary to examine to further our understanding and implications of growth-nodes. We introduce the I-Space framework as a tool for analyzing knowledge management and learning in growth nodes. Our analysis allows us to identify key issues that are worth researching, resulting in a series of research proposals. These are summarized at the end as the G-NIKE research agenda.
From Clusters to Growth nodes
Most experts define a cluster as a geographically bounded concentration of similar, related or complementary businesses and other related organizations (or institutions), with active channels for business transactions, communication and dialogue, that share a specialized infrastructure, labor markets and services, and that are faced with common opportunities and threats (Porter 1990, 1998, 2001).
The distinction between an agglomeration and a cluster is not always clear-cut. Cities can be said to be agglomerations of economic activities, but these activities do not necessarily contain any clusters. The cluster concept entails an industrial dimension. A cluster can be viewed as a specialized agglomeration of firms with mutually supporting interactions which derive from (and also reinforce) the particular specialisation. Such interactions depend traditionally in large part on spatial proximity – i.e. agglomeration. Thus, every cluster is in a sense an agglomeration, but not every agglomeration is a cluster.
Clusters are based on systemic relationships among firms and related organizations. The relationships can be built on common or complementary products, production processes, core technologies, natural resource requirements, skill requirements, and/or distribution channels (Rosenfeld 2002). Clusters are geographically bounded, defined largely by distances and times that people are willing to travel for employment and by the observation that employees and owners of companies consider reasonable for meeting and networking. Range is influenced by transportation systems and traffic but also by cultural identity, personal preferences, and family and social demands and ties.
Porter (1998) posits that in a globalizing world the forces leading to cross-industry clustering and involving the knowledge base and social aspects have intensified. Against this background, we position the growth-node as: an evolution of the cluster concept that emphasises the external networking dimensions, in addition to the cross-industry, knowledge transfer, and social learning conventionally associated with clusters.
Thus, we define growth node[1] as:
- A high-performing geographical cluster of organizations and institutions
- networked to other clusters, i.e. other nodes, and
- potentially supported by ICT.
We chose the term growth-node because the focus of our research is not only on the interrelatedness within different clusters but also on the interrelatedness between them. Thus, the G-NIKE paradigm considers clusters as nodes in a wider network. These nodes exhibit a high degree of connectivity internally (organizations within the node) and externally (to other nodes and/or to organizations in other regions). We realise, however, that, in practice, connectedness may be uneven.
Growth-nodes are seen as an evolution of the regional cluster concept with a special emphasis on internal and external networking. “Nodality” becomes an important attribute of any cluster. The growth of a node arises from the interplay of internal and external interactions. In effect, a growth-node can be thought of as an augmented cluster.
A growth node is as an aggregation of connected organizations concentrated within a particular region but competing or collaborating on world markets (or at least in markets outside the region). Their coherence is based on knowledge sharing through the network and spillovers leading to a high rate of firm start-ups. The idea is that intra- and inter-node interrelatedness and competition as well as collaboration, taken together, will foster economic development.
In addition, we assume that the use of diverse combinations of ICTs within and between growth-nodes will have implications for the meaning of proximity. In traditional clusters, the need for physical proximity has led to regional agglomerations. The question is how will new ICTs affect traditionally perceived needs for physical proximity and introduce “virtual” proximity as a complement to physical proximity? Can growth-nodes be expected to emerge and/or develop, in part, as a result of the widespread application of ICTs? What combinations of physically proximate and ‘virtual’ arrangements best augment the social and economic performance of growth-nodes?
The next sections develop these points and the associated research agenda.
2. Do growth nodes exist?
To apply the concept of growth-nodes to policy, one must believe not just that growth nodes exist, but also that they can be brought into existence. Thus, the issue is how to identify existing growth-nodes or clusters that have nodal potential.
Underlying the assessment of growth-nodes, however, is a definitional question. The growth-node concept currently is elastic. Like clusters, growth-nodes may be hard to identify. Their multidimensional character poses problems of empirical definition, as well as methodological approach. Thus, a necessary first step for the focal research is to develop and operationalize constructs for identifying, characterizing and measuring growth-nodes, their incidence and their effects.
A preliminary list of cluster attributes that may exhibit nodality and growth potential would comprise at least the following:
- Externality: density of interactions with partners outside the growth node (GN).
- Reach: geographic scope of the GN – regional, national or international.
- Knowledge intensity: extent to which interactions are knowledge-based
- Employment structure: knowledge workers as % total employment
- ICT infrastructure: extent to which the GN is supported by a networking infrastructure for linking players internal and external to the node
The analysis of these attributes will indicate the extent to which a potential candidate cluster exhibits nodality and, hence, can be considered as a candidate for growth-node status. A key research agenda item would thus be the analysis of existing clusters in several member states or geographical areas in Europe and an assessment of their nodality, i.e. their ‘growth node’ status. By highlighting those attributes that are most relevant, such a project would also help to refine the definition of growth nodes.
If growth nodes exist, however, why do they matter? The next section examines the potential role of growth nodes in regional development.
3. Growth-nodes and regional development
The importance of clusters is well established. Porter’s identification of contemporary local agglomerations, based on a large-scale empirical analysis of the internationally competitive industries for several countries, has been especially influential, and his term ‘industrial cluster’ has become the standard concept in this field (Porter 1990, 1998, 2001). The work of Krugman has been more concerned with the economic theory of the spatial localisation of industry. Both authors have argued that the economic geography of a nation is key to understanding its growth and international competitiveness.
Clusters lead to higher growth in several ways. Concentration, or clustering, gives businesses an advantage over more isolated competitors. It provides access to more suppliers and customised support services, to experienced and skilled labor pools. Clustering enables companies to focus on what they know and do best. In addition, clusters stimulate higher rates of new business formation, as employees become entrepreneurs in spin-off ventures, since barriers to entry are lower than elsewhere.
Among all of the advantages of clustering, however, none is as important as access to innovation, knowledge, and know-how. In the ‘Knowledge-based Economy’, companies are expected to look for their main competitive advantages in terms of access to ideas and talent. This generally is assumed to require geographic proximity to professional colleagues, leading-edge suppliers, discriminating customers, highly skilled labor pools, research and development facilities, and industry leaders.
Next, we briefly discuss three elements –social capital, social learning, and networking– that are associated with clusters and are also very relevant to growth-node performance.
The Social Factor
An important influence on cluster strategies has been the accumulation of social capital and its emergence as a factor in economic growth and social development. Associative behavior rarely has been considered as an explicit factor in economic development advantage.
The contribution of social capital to economic development has roots in Europe in Northern Italy. An analytical framework for the social foundation of clusters was provided by Robert Putnam’s research, which compared Northern and Southern Italy’s economies in 1993, and by Annalee Saxenian’s research, which compared Silicon Valley and Route 128 high tech economies in 1994. Their widely cited analyses confirmed the importance of social infrastructure for competitiveness. An implication of their research is that regions should pay more attention to the roles of intermediaries and gatekeepers such as business associations, chambers of commerce, and community based organizations.
Learning Systems
In the ‘Knowledge-based Economy’, growth depends on technology diffusion and knowledge spillover. Research shows that clusters can facilitate the transmission of knowledge—particularly tacit knowledge, which is embedded in the minds of individuals and the routines of organizations and which therefore cannot move as freely or easily from place to place as codified knowledge (Cortright, 2000).
Ideas about the importance of creating structures that support and accelerate learning have been translated in the context of the ‘new economy’ in the form of strategies to create “learning cities” and “learning regions” (OECD, 2001). Within clustered economies, there invariably is more inter-firm mobility and thus more active transfer of information and knowledge among firms and workers.
Business Networks
Northern Italy is generally accepted as the proto-typical economy of clusters. The region of Emilia-Romagna was first noticed because of its unusually small, flexible, and specialized firm structure. This was described by Piore and Sabel in the Second Industrial Divide (1984). But the success of northern Italy was first attributed, not to the clustering of companies, but to the intensity of inter-firm collaboration and to the specialized services created by the government and trade associations that gave the small companies access to external economies of scale. The Italian “network” of small firms became the practice that initially piqued much interest and that many countries have molded into public policies. The “network” promised to allow small firms anywhere to survive and prosper in increasingly competitive global economies. Government agencies in other countries realized that networks were also a cost effective means for aggregating demand and delivering services to small firms.
It should be noted that whereas networking has been implicitly associated with the clustering concept it can just as easily accommodate the concept of nodality and, hence, the relationship between clusters. What, then, does the growth-node perspective add to the benefits of clusters?
Of particular importance for the growth-node concept and for the associated research is the question of “experienced proximity” and “presence” in ICT mediated environments. This topic is discussed later in more detail. Here we limit ourselves to observing that, by capitalising on the potential offered by ICTs to achieve experienced proximity between points that are distant from each other in space, a growth-node has the effect of making outside resources – cognitive, institutional, cultural and material – available inside the growth node, thus stimulating its growth beyond that which could be achieved by employing local resources alone.
Concerns about Equity
Social capital, a core asset of many clusters, has both its pluses and its minuses. Social networks expose members to new processes and markets, non-public bid requests, and innovations. But, those companies that are outside the networks miss out on many economic opportunities. Clusters create a capacity to network and learn, but the more they are defined—and correspondingly limited—by formal membership, and the more business activity depends on personal networking, the higher the hurdles for outsiders to obtain the benefits of that knowledge.
The equity question deserves particular attention. As discussed above, players outside the network miss out on many opportunities. In conventional clusters, access to the learning network may be controlled by the interests of some large companies. This has traditionally been the problem for SMEs which, as a result, have been slow to learn about and adopt new technologies, or enter new markets. The G-NIKE hypothesis is that growth nodes do lead to greater social inclusion. The rationale for such proposition is based on the premise that knowledge in growth nodes is more freely available and not limited to the local resources. There seems to be some empirical evidence supporting it.
A number of regions in the European Union classified as “less favored” have sectors specialized in traditional industries with little innovation and predominance of small family firms with weak links to external markets (Landabaso et. al., 1999). The most successful clusters in the US, on the other hand, include lead firms that are part of global networks and are exposed to global market opportunities, and that employ people active in international professional associations and networks (Rosenfeld 2002). These firms regularly benchmark themselves against the best practices anywhere.
Poorer regions and smaller companies have limited access to the benchmark practices, innovations, and markets. Without wider access, companies are limited to learning only within their regional borders and have a difficult time achieving any sort of competitive position. The question for G-NIKE research is to ascertain whether, and how, growth nodes increase social inclusion. The research agenda should thus include the analysis and comparison of the structure and behaviors of a set of growth nodes against a set of traditional clusters. Section 7 will introduce a series of tools for regional ‘institutional diagnostic’ which will be particularly useful here for the intended analysis.
4. Conditions to sustain and develop growth nodes
Research on economic growth suggests the following factors which may also be relevant to growth-node development: innovation, imitation and competition, entrepreneurship, networks, social capital (“connections”), specialized workforce, industry leaders, talent, tacit knowledge (Rosabeth Moss Kanter, 1995).
Although the success of an individual firm depends on its ability to protect its own technological advances and designs; the success of a cluster or a growth-node in which it operates depends on widespread diffusion, access to innovations and information, and spinoffs of new enterprises.
Whereas innovation builds a strong company; the imitation and the competition that follow build a strong cluster. Imitation is as important to a cluster as innovation because it is what circulates new concepts and practices among companies and spurs further innovation. Many of the imitators become innovators by improving on the practices they adopt, and this cycle of innovation and imitation drives the cluster toward excellence.
The most successful clusters have built mechanisms to speed the movement of ideas, innovations, and information from firm to firm throughout the economy. The dynamics of clusters create the learning region. The mechanisms and entities for collecting and disseminating knowledge, as well as the intermediaries that facilitate all forms of associative behavior, help build ‘social capital’ that is important for cluster competitiveness.
An important operating principle of clusters has proven to be the ability to network extensively and form networks selectively. Networking is the process that moves and spreads ideas, information, and best practices throughout a cluster and imports them from other places. The limits or constraints to active participation in a successful cluster are largely a function of “connections” or deficits in social capital. A regions’s stock of social capital resides in its civic and professional associations, and its economic value is deeply embedded in the functions of groups that bring people together to share ideas and knowledge.
To test the proposition that the above factors are relevant to the development and sustainability of growth-nodes, G-NIKE proposes a longitudinal study to track the evolution of a set of growth-nodes. Appropriate constructs and measures would have to be developed for each these factors. The analysis would test whether changes in any of the factors over time translate into changes in the dependent variable (a growth measure). The project would also test additional explanatory variables which are unique to growth nodes. For example, extra-node connectivity, availability of shared ICT infrastructure, and use of some key ICT applications.
Given the importance of networking, knowledge diffusion and social learning in growth nodes, the next section introduces a knowledge-management framework, the I-space, which will later be used to analyze growth-nodes.
5. The I-Space: a conceptual framework to assess knowledge management in growth nodes
In this section, we introduce some elements of a conceptual framework to explore characteristics of information and knowledge flows relevant to growth nodes, and how these might be affected by the new information and communication technologies (ICTs) as well as their spatial implications. The framework, known as the Information Space or I-Space, is described in detail in Boisot (1995) and (1998), and also in the appendix.
The framework highlights the relation between the extent to which data can be structured and the extent to which they can be shared. The I-Space takes codification and abstraction as the two processes through which data is apprehended and structured. Well codified and abstract data economizes on both senders’ and receivers’ data processing transmission resources. Well codified and abstract data will diffuse to more agents per unit of time than uncodified and concrete data. Thus, codification, abstraction and diffusion constitute the three dimensions of the I-Space.
This three-dimensional space is used to position learning processes. The framework distinguishes six activities that contribute to learning: codification, abstraction, diffusion, absorption, impacting, and scanning all contribute to learning. Where they take place in sequence they make up the six phases of a Social Learning Cycle (SLC) as illustrated in the figure below..
The Social Learning Cycle in the I-Space
[pic]
SLCs can have many shapes in the I-Space reflecting the different blockages that can impede the learning process and not all phases of an SLC will be immediately value adding. In addition, learners will adopt different learning strategies. Learners or agents may engage in information hoarding or information sharing, or in some cases, in both strategies:
• Information hoarding: recognizing that diffused information has no economic value, agents attempt to slow down the SLC by refraining from codifying or abstracting too much and by building barriers to the diffusion of newly codified abstract information – through patents, copyright, secrecy clauses, etc. Slowing down the SLC allows them to extract value from information in a controlled way.
• Information sharing: recognizing that, through subsequent processes of absorption, impacting and scanning, diffused information prepares the ground for further learning and knowledge creation, agents willingly share their information and watch how it is used by others. They gain first-mover advantages in being the first to initiate a new SLC and extract value from the process by participating in a succession of SLCs instead of dwelling as long as possible in a single one.
Network institutional orders
The framework introduces the role of institutions in the processes of social learning and knowledge management. Different institutional mixes embedded within a network may represent a variety of ‘strong’ and ‘weak’ cultures that interact in ways that influence the sharing of local and distant information resources and the extent to which learning progresses to enable efficient and effective knowledge management. The framework distinguishes four different types of institutional order: bureaucracies, markets, clans, and fiefs.
If codification and abstraction are taken as proxy measures for information structuring, and diffusion is taken as a proxy measure of information sharing, we can see how the different cultural features of actors connected within networks might help to institutionalize the social and economic transactions involved in the SLC.
Table 1 Institutions in the I-Space
[pic]
The four different types of institutional order can be located in the I-Space as shown in the figure below.
Institutions in the I-Space
[pic]
These institutions are potential attractors that exert a gravitational pull on transactions that are located within their field. Where these attractors are not too strong, they facilitate learning by speeding up the flow of the SLC through that region of the I-Space. However, where the pull that they exert becomes so powerful that they can attract transactions from any part of the I-Space, then they bring the SLC to a halt in their region of the Space, blocking any further progress of the learning process.
This highlights the importance of fiefs, clans, and bureaucracies, as governance structures. They can be as important as efficient markets in fostering effective collaboration and social learning as well as competition.
The I-Space and the effects of ICT
The new ICTs, by increasing data processing and transmission capacities at all levels of codification and abstraction, have the effect of shifting the whole diffusion curve to the right. Note that this shift has two quite distinct effects: 1) for any level of codification and abstraction, more people can be reached per unit of time with a given message. 2) a given size of population can now receive a message at a lower level of codification and abstraction. Another way of saying this is that the new ICTs allow more of the context to be transmitted with a message.
But even with the new ICTs, the transmission of context (e.g. through video-conferencing) remains costly. ICTs are likely to facilitate the transmission of data goods more than of information goods. In turn, this will facilitate the transmission of information goods more than that of knowledge goods.
The implication is that ICT applications serve to differentiate knowledge-based services according to their flow characteristics. Data-intensive processing and transmission can be performed virtually anywhere, whereas the knowledge-intensive interactions remain rooted in time and place and continue to depend heavily on face-to-face interactions.
Competitive advantage will accrue to those who learn how to climb the ladder, firstly from data to information, and then from information to knowledge. ICTs will get them on the bottom rung faster than they might have done on their own. But from then on, the critical assets will be an ability to learn fast rather than to access technology. Cities and regions, in effect, are engaged in learning races with each other, races in which competition and collaboration both have their place.
The predominance of a given institutional order is likely to be supported by a distinct set of ICTs. For instance, inward oriented, physically proximate networks of firms and organizations may be characterized by institutions such as ‘fiefs’ that give very high priority to investment in high bandwidth Intranets and Extranets and proprietary software applications for decision support, putting less emphasis on inter-regional networks. Alternatively, externally oriented regions and firms and other organizations may develop many global network linkages to the neglect of the ICT applications that may enable and foster local learning cycles. The institution of the ‘market’ may make it difficult or more costly to foster transactions within the regions in such cases.
The concepts introduced by this framework will now be used to propose tools for identifying and analyzing growth nodes.
6. Identifying and measuring growth nodes
The best-known model for describing the various elements of a cluster is the four-point “diamond,” developed by Michael Porter (1990). The model includes (1) firm structure and rivalry, (2) local demand, (3) related and support industries, and (4) “factor conditions,” defined as skills, infrastructure, R&D, capital, etc.
We argue that Porter’s diamond of competitive advantage needs to be further unpacked and analyzed spatially. We hypothesize that when it is, it will look like agglomerations of industry clusters linked together in a dynamic network. In pre-industrial times, given the lack of transport facilities, the network element was sparse, fragile, slow moving and barely visible. All eyes were on the spatial agglomeration itself: the Venetian republic, Genoa, London, etc. (Pirenne, 1933; Favier, 1987; Le Goff, 1986). With economic and demographic growth, the networks became denser and more visible – ie, the Hanseatic League. Some of the links were bolstered by the building of infrastructure.
Recent developments in ICTs, by substituting the flow of information goods for that of physical goods, have made us realize that the expansion of industry clusters within spatial agglomerations can usefully be considered an emergent property of network connectivity. Hence, we label expanding and interconnected industry clusters growth nodes.
Measuring growth nodes
The World Economic Forum assessed the effects of European policies and reforms, by reviewing the Lisbon objectives at the level of countries[2]. Effects were evaluated on eight dimensions of competitiveness (see graph below).
[pic]
Potential growth-node candidates in European regions could also be assessed by using similar dimensions. From a growth node perspective, however, most of these dimensions are measures of outcome / performance. They do not help us understand the elements, operation or dynamics of a growth node. The questions is how to characterize the ingredients and intervening processes that lead to growth node emergence and development. Models used in cluster analysis may be relevant for growth nodes.
Most descriptions of clusters begin by measuring numbers of establishments and employees by sector based on association directories and existing databases. The most common measures are, for each combination of sectors in the cluster, (1) numbers of employees and establishments, (2) location quotients for each that compare the local concentrations of the industry sectors included within the cluster to concentrations of the same group of sectors for the entire economy (see box below), (3) input-output tables that estimate supply chain linkages, and (4) growth rates.
Still other measures focus on rates of innovation and knowledge (stocks) associated with clusters, such as comparing the proportions of workers in occupations classified as knowledge-intensive, or comparing patent rates by organizations and employees in clusters. Each measure offers a useful but limited estimating procedure.
In the case of growth nodes, however, one should measure also those attributes that are specific to the growth node concept (as opposed to a cluster), e.g. ‘externality’, ‘reach’, ‘supporting ICT infrastructure’ as discussed earlier. The idea is to highlight the cluster or growth node’s interaction with the outside world and thus reflect its ‘nodality’.
Inventorying Knowledge Assets
Because the value of clustering is linked to the firms’ access to specialized services and resources, a region should know what those assets are and where they are located. Listing the assets available to and used by a cluster is a prerequisite for completing any picture of the growth node and understanding how it functions. Those assets include the education programmes that match the workforce needs of the cluster, the consultants who are familiar with the cluster’s industries, and the R&D that relates to the cluster. They also include the freight forwarders and exporters who know the markets; the banks and accountants who have developed relationships with the cluster; and the trade, labor, and professional associations that provide the networking opportunities (see box below).
It is important to note that these are not only resources within the cluster, but also resources between the cluster and other clusters.
Mapping Relationships
Like clusters, growth nodes depend on relationships and connections. The easiest relationships to map are the sector-based supply chains. The most difficult relationships to map are the flows of tacit knowledge and innovation, which require information from individuals about forums for associative behavior and personal relationships. It requires surveying a sample of cluster members’ relationships to learn, for example: to whom companies turn for help with business problems, with whom they trust enough to collaborate, in which business or professional associations they are active.
9
The most common map is a flow diagram in which boxes symbolize key parts of the cluster—the companies, suppliers, services, supporting institutions, and trade, business, and labor associations—and shows connections by directional arrows. The thickness of the arrows can be used to denote the intensity of the linkages.
By mapping the intensity of connections, one can discover how tightly clusters or growth nodes are bound internally; the degree to which any cluster is connected and dependent on organizations outside the region or to other regional clusters; which intermediaries are most effective; and where better information channels could be useful.
Tools to assess Growth-Nodes
An important question for growth-node research is whether growth-nodes can be considered as policy instruments for regional-development (just as clusters are today), or just an object worth of study just because of their novelty and uniqueness. In the latter case, growth-nodes will be a relatively rare phenomenon and the value of the concept for regional policy will be limited, unless one believes that growth nodes can be brought into existence in a larger scale.
If we assume that growth-nodes may become a policy instrument, then the implication for G-NIKE research is the need to develop practical tools for regional policy makers. The I-space framework discussed earlier suggests three areas for measurement of growth nodes, i.e. knowledge assets, social learning cycle, and the different institutional orders in knowledge networks. The development of the necessary instruments would in itself constitute an important research project. Such tools for regional policy makers would entail, for example:
1. Regional and territorial knowledge mapping tools.
These will allow regional and territorial authorities to develop portfolios of territorially located knowledge assets based on their value-added and their competitive potential. Such tools should allow:
- The identification of local knowledge assets, in particular, with regard to ICTs
- The assessment of the competitive position and potential of these assets with respect to outsiders
- The positioning or mapping of these assets in terms furthering European competitiveness, social cohesion, and sustainability.
2. Regional social learning tools.
These will allow regions and territorial authorities to gauge how well they are equipped to make good use of their knowledge assets at the grass roots level, based on:
- Local scanning skills (from which outside regions or clusters do you scan?)
- Local problem-solving capacities
- Local diffusion abilities (to which outside regions or clusters do you diffuse?)
- Local absorptive capacity
3.Regional institution diagnostic tools.
A well-designed network infrastructure should be able to support the development of territorial institutions that facilitate the required level of connectivity. Regional institution-building tools should promote:
- Regional access to regional and global knowledge with respect to ICT development or use
- Regional social learning
- New business models, innovative ways of working, and achieving equity in the region
- Regional solidarity and social stability
7. Enabling IT infrastructure for Growth Node development and interaction
Much has been made of the potential of information and communication technologies (ICTs) to enable a de-spatialisation of economic activity, but so far systematic analysis of how policy can stimulate European clusters in the 21st century by taking advantage of the characteristics of ICTs is lacking. The G-NIKE project seeks to addresses this problem. It considers the role of ICTs by regarding growth-nodes as clusters of geographically proximate complex organizational systems of learning and economic and social activity that are globally networked with other systems and enabled by the effective use of ICTs. Our work so far has allowed us to identify three research contributions with implications for growth nodes:
The first is associated with the ‘tacit-codified’ knowledge debate in knowledge management research. The understanding of growth nodes will be enhanced by a perspective on the potential contribution of ICTs that gives priority to socially mediated tacit skill sets and learning processes as prerequisites for the effective and efficient use of these technologies within a complex adaptive system. In this context, we can consider the conditions under which ICTs may enable new opportunities for codification which, in turn, may be expected to give rise to new clusters within the European economy. Research should also be particularly concerned with changes in the way that tacit skills sets and learning are mediated by ICTs. This aspect may be examined in terms of the extent to which these processes give rise to new hierarchies and forms of economic and social dominance that can lead to inequality and exclusion even within the context of relatively ‘flat’ and decentralized network structures.
The second is related to debates about ‘mediated co-presence’. There is considerable empirical evidence which suggests that it is important to disaggregate ICTs and their applications. Different applications seem to contribute to varying degrees to the ‘social richness’ of spatially proximate and distant encounters. Some may favor and support the growth of clusters of economic activity, while others are less favorable. G-NIKE research acknowledges that different modes of ICT supported communication are likely to have different implications for the spatial distribution of economic activity and for agglomeration economies. Much work on the application of ICTs in the context of e-commerce, e-government and intra-firm knowledge management systems, neglects the social and culturally specific ways in which mediated environments give rise to new forms of communicative interaction and their consequences of networked relationships. The application of ICTs within and between organizational settings gives rise to the question of how to achieve an appropriate balance between on- and offline relationships and overlapping networked organizations. This question can be approached from the standpoint of the economics of technological change and the social phenomena that strengthen or weaken ‘experienced proximity’.
The third is related to supporting architectures and infrastructures. Growth nodes are complex systems based on the networking of organizations, the cooperation of the players and flexible access to resources. It can be seen as a community that shares business, knowledge, and infrastructure in a highly dynamic way. Indeed, some researchers posit that the enterprise and other organizations of the future will be more fluid, amorphous and, often, with transitory structures based on alliances, partnerships and collaboration within clusters (or growth nodes). To support this scenario, a next generation of ICT inter-organizational network infrastructure (i.e. Business Digital Ecosystem) is envisaged which will provide a dynamic aggregation of network and software services to facilitate the dynamic interaction in both vertical and horizontal inter-organizational relations. Of particular interest to G-NIKE is a focus on the how new ICT infrastructures may serve as a catalyst that alters or transforms existing relationships between physical place and people’s perceptions of the value of proximity.
The next sections explain in more detail each of these three themes.
Knowledge Sharing, Proximity, and ICT
The researcher community appears to have divided views with respect to knowledge diffusion. Some suggest that knowledge can be reduced (‘codified’) to messages that can then be sent and processed as information. This view suggests that the potential exists for the universal codification of knowledge. ICTs offer a means by which codified knowledge may be disseminated as information. These authors downplay the importance of geography. From this perspective, information flows are spatially unbounded in a world that is inter-linked through the implementation of ICTs.
Others argue, instead, that knowledge cannot be considered independently from the processes through which it is generated. Comprehending and utilizing information encompass tacit skills that are intrinsically bound to social processes. These skills entail the cognitive capabilities of the agents and the organizational contexts within which they are interacting. The defining feature of this tacit knowledge is that it cannot be articulated (i.e. ‘codified’) for the purpose of exchange. Tacit knowledge can refer to specific knowledge that is mainly held and shaped by individuals. It emerges from routines, conversations, memories, stories, and repeated interactions, instead of being encrypted in rules or in organizational design.
Focusing on the properties of the knowledge that is used in innovation-related activities and on the associated knowledge exchange, there are those who argue that the transmission of new knowledge occurs more efficiently among proximate actors. (within the cluster or growth nodes). This is the knowledge spillover hypothesis. Highlighting the complexity and tacit nature of the knowledge base, proponents of this view argue that proximity helps to reduce the costs of knowledge transmission by facilitating interpersonal contacts and the inter-firm mobility of labor. The degree to which geographical proximity facilitates the sharing of knowledge, in turn, overlaps and combines with institutional, organizational and technical proximity in fostering effective processes of collective learning.
The debate about how the use of ICTs is likely to influence ‘experienced proximity’ is related to whether these technologies support knowledge codification that provides new memory aids for individuals or facilitates collective recall within group exchanges. The use of ICTs may also provide a social memory device in environments where offline social processes for guiding knowledge codification are not available.
The key issue is whether a particular form of ICT use provides sufficient cognitive context for generally tacit knowledge to be transmitted explicitly when required so as to repair any problems that occur in applying that knowledge. This is a core issue in analyzing the spatial implications of ‘experienced proximity’ and the potential of ICTs when they are used to support new network architectures and infrastructures of many different kinds.
“Presence” and ICT Mediated Environments
It is important to examine the particular types of ICT applications that are in use or may be deployed in order to understand their implications for emergent growth-nodes. Different ICT supported networks are likely to contribute to varying degrees to the ‘social richness’ of spatially proximate and distant encounters. Internet and non-Internet-based ICTs will have different implications for the different phases of the Social Learning Cycle. Thus, G-NIKE research is concerned with the several ways in which new ICTs differ from traditional modes of communication and information exchange (including face-to-face) in the configuration of constraints and processes available to those seeking to communicate in a growth-node (not only in inward-focused transactions but also in outbound transactions, i.e. interactions with partners outside the growth node).
In investigating the role of ICTs in facilitating growth-nodes, it may be helpful to conceptualise presence as ‘social richness’. Thus, for both inter and intra-organizational communication, ‘presence’ is associated with whether a medium enables the reproduction of the capabilities for comprehending and utilising information. ICTs may differ in the extent to which they ‘(a) can overcome various communication constraints of time, location permanence, distribution, and distance; (b) transmit the social, symbolic, and nonverbal cues of human communication; and (c) convey useable information.
ICTs that are high in presence or social richness enable users to adjust more precisely to physical cues, i.e. facial expressions, gestures, vocal tones, etc., and to maximise the efficacy of interpersonal communication. Visual communication generally is believed to have more social presence than verbal (audio) communication, which, in turn, is expected to embody more presence than written text. These observations are important for understanding how ICTs may influence relationships between spatial proximity and ‘experience proximity’ or ‘mediated presence’.
The richness of information in face-to-face environments is important because usually this enables actors to quickly repair problems or gaps that arise in information exchanges. Depending on the social conventions operating within virtual networks, repair is also easy, or even much easier (as in open source development networks where very high levels of shared skills can be assumed). But this may not be the case, or at least not in the same way, with many other types of knowledge that might in principal be shared virtually, e.g. non-ICT product designs, marketing strategies, organizational management and negotiating skills, etc.
Research has shown that higher resolution images in a video conferencing system elicited reports of greater ‘communicative’ presence. Many presence-evoking ICTs are promoted as enabling people to more efficiently accomplish specific tasks (i.e. transmit knowledge). However, despite the fact that these technologies often enable tasks to be completed in a new way, there is relatively little research available to indicate the extent to which these new ways are more effective or efficient in transmitting knowledge than older, more traditional methods. Although one of the most important groups of tasks for which presence evoking media have been designed and used involves specialized skills training (i.e. flight simulation), more research is required to identify the characteristics of tasks for which presence actually enhances knowledge transfer and acquisition.
A key question for the G-NIKE research agenda is, therefore, for what types of knowledge are virtual networks more likely to be sustained as effective means of sharing, developing and repairing knowledge, and for which types of knowledge are virtual networks more likely to need supplementing or replacing by face-to-face forms of information exchange? Answering this question calls for comparative studies of practices within and between organizations in order to elicit principles that might influence the emergence of growth-nodes. It also requires an analysis of different types of ICT applications and their appropriation by users including firms, universities, civil society organizations, etc., within their various social networks.
Shared ICT Infrastructure
The logic for concentrating and sharing resources in a cluster can be extended to the ICT infrastructure and related ICT services in growth-nodes. In this case, the potential benefits go beyond local access to technology and know-how. Indeed, achieving some inter-organizational ICT architecture, with standardised interfaces, flexible access and shared elements can lead to significant benefits in terms of inter-operability and flexibility. This should be of special interest to SMEs as it would lower their barriers to adopt ICT and allow them to fully engage in growth-node network(s).
Our working hypothesis is that the enterprises in growth nodes will be able to form ad-hoc, temporary alliances, partnerships and collaboration with partners within and beyond clusters (or growth-nodes). Consequently, the need for ad-hoc and flexible inter-organizational information exchange is paramount. To support this scenario, a next generation ICT inter-organizational network infrastructure is likely to be required. Such a network infrastructure would provide a dynamic aggregation of network and software services to facilitate such dynamic interaction between business partners and other institutions in the growth node (and be able to support traded and as well as non-traded transactions).
The adoption of Internet-based technologies for business, where business services and software components are supported by a pervasive software environment can be designated as a: business digital ecosystem (BDE). The key elements of the BDE structure are software components and agents, which show evolutionary and self-organising behavior, i.e. they are subject to evolution and to self-selection based on their ability to self-adapt to the local business requirements (Nachira 2002).
We reconize the need for such an infrastructure and propose research into the design and development of a DBE platform as well as its subsequent testing in several regions that exhibit growth-node potential. Such project would take into consideration the three aspects: technology, e-business models and services, and inter-organizational knowledge sharing. The holistic approach requires fundamental research on self-organization in complex adaptive systems (CAS) and on network architectures by physical and engineering scientists as well as social scientists. The following topics could be considered:
• Pervasive, adaptive, self-configuring and self healing network software architectures
• Semantic discovery and registering applications
• Distributed security and federated network identity
• Dynamic component composition, software component and knowledge sharing on the network
• Interoperability
• Multi-Agents, behavior of complex systems and agent communities
• Semantic web, knowledge sharing and cooperation mechanisms, ontologies, business process modelling and integration
As the approach is evolutionary, and therefore based on continuous development and adoption, early production of “precursor” results and preliminary links, and identification of assembly rules, it will quickly lead to operational primitive digital ecosystems. The primitive ecosystem, subject to a continuous evolution process, allows early testing and adoption in test-bed growth-nodes. The testing of the DBE architecture would be articulated and designed as a field experiment. The project is likely to entail design, simulation, development and field tests. The experiment(s) would be used to test hypotheses relating to role of ICT and development of growth-nodes.
The policy perspective
Over the past few years, the cluster concept has found a ready audience among policy makers at all levels, from the World Bank, to national governments, to regional development bodies, to city authorities. All are keen to find a new form of industrial policy or activism in which the focus is firmly on the promotion of successful, competitive economies. The argument is not that governments can create clusters or growth-nodes, but that they can help to provide the business, innovative and institutional environments vital for their success.
The first step is usually to identify the clusters in the region or the country. Organizations and government agencies that view their regions as clustered production systems are predisposed to tailoring existing policies and programs to that model and in some instances creating new strategies. The most common policy levers are those that alter the way agencies organize and deliver their services, work with employers, recruit businesses, and allocate resources. But the most popular goals are to market a political region and attract businesses and highly educated and skilled people. The following are examples of cluster-based policy levers:
- Promotion. Giving “official” recognition to a cluster represents a form of collective marketing of the cluster and its products and creates avenues for more effective lobbying efforts.
- Investment. Often the public-sector is interested in identifying clusters because it represent a way to increase the odds of attracting investment.
- Workforce. Governments adapt the appropriate degree of specialisation in higher education to meet the needs of clusters and regional economies.
- Social cohesion. Clusters offer ways to restructure equity policies to more effectively serve less-advantaged regions and lower income and less-educated populations.
- Collective awareness. A common intervention to strengthen clusters is to form and empower “cluster councils.” These get companies to articulate a collective vision and to create awareness that they represent a larger regional economic entity. It also is an attempt to build associational behavior.
- Organization of services. Many SMEs are confused by the vast array of agencies that offer solutions in a cluster. A solution is to integrate the services either by creating a hub (one-stop-shop) or creating a set of intermediaries (knowledge brokers) to serve as linking agents.
Towards Growth-Node Policies
A fundamental question is whether the growth-node paradigm can provide a new lens for policy research and practice. By introducing the ‘growth node’ concept, G-NIKE redefines the notion of cluster. The issue is whether the ‘growth node’ is a restrictive concept and, more importantly, whether it can be applied as policy tool for regional development.
To apply the concept of growth-nodes to policy, one must believe, not just that growth nodes exist, but also that they can be brought into existence. The question is how to identify existing growth-nodes or clusters that have nodal potential.
As discussed earlier, G-NIKE proposes research to develop and operationalise constructs for identifying, characterising and measuring growth-nodes, their incidence and their effects. The next step will be to apply the G-NIKE instruments to identify the growth nodes in some European countries or regions, and establish whether the regional / national economies can be effectively examined through the growth-node lens; and if so, whether policy makers can more accurately identify market imperfections, find pressure points, envisage or pinpoint systemic failures, and determine what interventions can have the greatest impacts.
Assuming that growth-nodes exist in reasonable numbers (i.e. they are significant for policy analysis), the second question for G-NIKE is whether the traditional cluster-based policies (e.g. those listed mentioned above) apply to growth-nodes. Should they be modified? Are entirely new policies required when regions are examined from the perspective of growth nodes?
If public policy makers proactively integrate advanced ICTs to link local geographically clustered firms and other organizations beyond their immediate regional surroundings, there may be substantial opportunities for a departure from the conventional emergent pattern. Global, national, regional and local ICT links and information flows may, in fact, fuel an ‘innovative milieu’ and may help to provide the catalyst for the social learning cycle (SLC) that gives rise to successful and enduring growth-nodes.
Growth-nodes differ from clusters in their nodality and the enabling role of ICTs. ICTs provide a new means of linking up local places and regions within networks of organizations. Inclusion in the network requires an adequate local technological infrastructure, a system of ancillary firms and other organizations providing support services, a specialized labor market, and a system of services required by the professional labor force. Thus, another set of relevant policies refers to actions to facilitate the adoption and usage of ICTs by small and mid-sized enterprises (SMEs). As discussed above, an inter-organizational infrastructure based on the ‘business digital ecosystem’ principles (BDE) might lower the barriers to ICTs usage by SMEs and facilitate access to the wider business network in a cluster or growth-node. If this potential exists, a policy that sponsors the development, testing and deployment of a BDE-like infrastructure may prove to be a more effective way to support SMEs than the take-up actions sponsored by earlier initiatives.
One last policy topic that G-NIKE addresses is related to growth-node dynamics. Porter (1990) has argued that fast-growing, innovative, geographically clustered firms ‘hot spots’ often turn into ‘blind spots’. More recently, other searchers have shown how rapidly the fortunes of ‘hot spots’ can be reversed, leading to the deterioration of formerly vibrant and innovative regions including both urban and rural agglomerations. Firms first begin to cluster and to forge a ‘hot spot’ identity, but convergence of clustered firms ultimately leads to a ‘hot spot’ failure.
The G-NIKE conceptual framework enables researchers and stakeholders to clarify the conditions under which emergent growth-node outcomes might be expected that are departures from the ‘hot spot’/’blind spot’ cycle. The G-NIKE perspective on emergent Complex Adaptive Systems opens the possibility for the discovery of key factors and policies that encourage divergence from historical pathways that are believed to characterise regional and local clusters.
8. Conclusion
This paper started with the elaboration of the growth-node concept as an augmented cluster. The growth-node was positioned as an evolution of the cluster concept that emphasises the external networking dimensions, in addition to the cross-industry, knowledge transfer, social learning conventionally associated with clusters.
The extensive literature and consulting work on clusters was then reviewed. Building on this background as well as our work in G-NIKE, the growth-node concept has been sharpened and growth-node attributes have been identified. Then, a series of research questions was addressed: Do growth-nodes exist? How do they contribute to regional development? How do we identify and measure growth-nodes? What are the conditions that sustain and develop them? What are the desired properties of the enabling ICT infrastructure? What policies are most likely to foster emergent growth-nodes in Europe? How can growth-nodes become a policy instrument?
In addressing these questions, we reviewed what is known about clusters that might be potentially relevant, as well as what is not known but would be necessary to examine to further our understanding and implications of growth-nodes. We introduced the I-Space framework as a tool for analyzing knowledge management and learning in growth nodes. Our analysis enabled us to identify key issues that are worth researching, resulting in a series of research project proposals. These research proposals constitute, in essence, our preliminary G-NIKE research agenda. The research proposals are summarized next.
G-NIKE Research Agenda for ICT-enabled Growth Nodes in Europe
1. Analysis of existing clusters in several member states or geographical areas in Europe to assess their nodality, and, consequently, their potential for designation as ‘growth nodes’.
An initial list of attributes would be: Externality, Reach, Knowledge intensity, Employment structure, ICT infrastructure. A composite measure, based on a combination of these attributes will indicate the degree and/or type of ‘nodality’ of the candidate clusters (i.e. potential growth node candidates). By highlighting those attributes that appear to be most relevant, this project would also help refine the definition of growth nodes and develop an appropriate typology.
2. Development of practical tools to identify and measure growth nodes for a given region or national economy.
These tools could include the following:
I. Regional and territorial knowledge mapping tools. These will allow regional and territorial authorities to develop portfolios of territorially located knowledge assets based on their value-added and their competitive potential.
II. Regional social learning tools. These will allow regions and territorial authorities to gauge how well they are equipped to make good use of their knowledge assets at the grass roots level.
III. Regional institution diagnostic tools. A well-designed network infrastructure should be able to support the development of territorial institutions that facilitate the required level of connectivity.
3. Qualitative analysis of designated growth nodes to validate the proposition that outside resources can be made available inside the growth node, thereby stimulating growth.
Growth nodes capitalise on the potential offered by ICTs to achieve ‘experienced proximity’ and ‘presence’ between points that are distant from each other in space. Our hypothesis is that a growth-node has the effect of making outside resources – cognitive, institutional, cultural and material – available inside the growth-node, thus stimulating its growth beyond that which could be achieved by employing local resources alone.
4. Identification of factors contributing to the development and sustainability of growth-nodes.
A longitudinal study is proposed to track the evolution of a set of growth-nodes. A multivariate analysis could be used to determine which factors are (statistically) significant. The analysis would test whether changes in any of the factors over time translate into changes in the dependent variable (a growth measure). Potential factors are: innovation, imitation, competition, entrepreneurship, networking, networks, connections, intermediaries, specialized workforce, industry leaders, talent, tacit knowledge and collaboration, availability of shared ICT infrastructure, use of some key ICT applications, and extra-node connectivity.
5. Assessment of virtual network properties and typologies to establish what network types are more likely to be sustained as effective means of sharing, developing and repairing knowledge.
This implies an analysis of different types of existing and emerging ICT applications (including new broadband services) and their appropriation by users including firms, universities, civil society organizations, etc., within their various social networks. The goal is to establish what types of knowledge-based virtual networks are more likely to need supplementing or replacing by face-to-face forms of information exchange.
6. Design, development and test of an infrastructure providing a dynamic aggregation of network and software services (Business Digital Ecosystem) to facilitate dynamic interaction between business partners and other institutions within and beyond the growth-node.
Research on the network infrastructure is likely to cover the following IT topics: self-configuring network software architectures, semantic discovery, distributed security, federated network identity, dynamic component composition, software component sharing on the network, multi-agents, behavior of complex systems and agent communities, semantic web, knowledge sharing and cooperation mechanisms, business process modelling and integration. The implementation approach would be based on iterative process of prototyping, development and adoption. Early production of “precursor” results and identification of assembly rules will lead to an operational primitive digital ecosystem. The primitive ecosystem, subject to a continuous evolution process, will allow for early testing and adoption in test-bed growth-nodes.
7. Assessment of usability and utility of the instruments to identify growth nodes, in order to establish whether selected regional / national economies can be examined usefully through the growth node lens.
These instruments (practical tools to identify and measure growth nodes as described in agenda item no. 2) are to be tested in some European countries or regions. The idea is to test whether policy makers find it a useful tool for policy analysis and design. An important output of such research will be a refined G-NIKE policy toolbox. In this regard, the project could explore, test and propose complementary measures / instruments that might be relevant indicators from a policy perspective.
8. Establish whether traditional cluster-based policies also apply to growth-node development, and also whether new policies are required.
Growth-nodes differ from clusters in their nodality and the enabling role of ICTs. Global, national, regional and local ICT links and information flows may, in fact, fuel an ‘innovative milieu’ and help to provide the catalyst for the social learning cycle (SLC) that gives rise to successful and enduring growth-nodes. If public policy makers proactively integrate advanced ICTs to link local geographically clustered firms and other organizations beyond their immediate regional surroundings, there may be substantial opportunities for a departure from the conventional emergent pattern. The project would an entail 1) the adoption of such policies on an experimental basis, and 2) analysis of their effects over time, 3) testing the hypothesis that inter-nodal ICT links enable a virtual ‘innovative milieu’ and its necessary social learning cycle (SLC).
9. Test whether the availability and use of a BDE inter-organizational infrastructure may lower the barriers to ICT adoption by SMEs and facilitate SME access to the wider business network in a cluster or growth-node.
If this potential exists, a policy that sponsors the development, testing and deployment of a DBE-like infrastructure (based on the ‘business digital ecosystem’ principles) may prove to be a more effective way to support SMEs than some of the take-up actions sponsored by earlier initiatives.
10. Ascertain the conditions under which one might expect emergent growth-node outcomes that depart from the ‘hot spot’/’blind spot’ cycle (i.e. rapid success followed by rapid failure).
This research is concerned with growth-node dynamics, their emergence, and their potentially chaotic behavior as complex systems. Researchers have shown how rapidly ‘hot spots’ can deteriorate and turn into into ‘blind spots’. The proposed research seeks to clarify the conditions under which one might expect emergent growth-node outcomes that depart from the ‘hot spot’/’blind spot’ cycle. The G-NIKE perspective on emergent Complex Adaptive Systems (CAS) opens the possibility for the discovery of key factors and policies that encourage divergence from historical pathways that are believed to characterise regional and local clusters.
References
Asheim, B. (1997). ‘Learning regions’ in a globalized world economy: Towards a new competitive advantage of industrial districts? In M.Taylor and S. Conti (Eds.) Interdependent and Uneven Development. Aldershot: Ashgate.
Beccattini, G. (1990). The Marshallian industrial district as a socio-economic notion. In Pyke, F. Beccattini, G. and Sengenberger, W. (Eds.) Industrial Districts and Bergman, E. (1998) Industrial trade clusters in action: Seeing regional economies whole. In M. Steiner (Ed.) Clusters and Regional Specialisation. London, Pion.
Boisot, M. (1995). Information Space: A Framework for Learning in Organizations, Institutions and Cultures, London: Routledge.
Boisot, M. (1998). Knowledge Assets: Securing Competitive Advantage in the Information Economy, Oxford: Oxford University Press.
Castells, M. & Hall, P. (1994) Technopoles of the World. London: Routledge.
Coase, R. (1937). The Nature of the Firm. Economica, N.S. 4, pp 386-405.
Coe, N. and Townsend, A. (1998) Debunking the myth of localized agglomerations: The development of a regionalized service economy in South-East England. Transactions of the Institute of British Geographers, 23.
Cooke, P. and Morgan, K. (1994) The network paradigm. Environment and Planning D,
Cooke, P. and Morgan, K. (1998) The Associational Economy: Firms. Regions and Innovation, Oxford: OUP.
Cortright, J. (2000). New Growth Theory, Technology and Learning: A Practitioners Guide to Theories for the Knowledge Based Economy. Draft report to the Economic Development Administration, Washington, DC: U.S. Department of Commerce.
Douglas, M. (1973). Natural symbols: Middlesex: Penguin Books.
DRI/McGraw-Hill. (1995). America’s Clusters, Publication for Conference “Building Industry Clusters”, June 1995. DRI/McGraw-Hill, Lexington, Massachusetts.
Enright, Michael J. (2000). Survey on the Characterization of Regional Clusters. Working Paper, Institute of Economic Policy and Business Strategy, University of Hong Kong.
Favier, J. (1987). De l’Or et des Epice: Naissance de l’Homme d’Affaires au Moyen Age. Paris: Fayard.
Fujita, M, Krugman, P. and Venables, J. (1999) The Spatial Economy: Cities, Regions and International Trade. Cambridge, Mass: MIT Press
Hall, E. (1976), Beyond Culture. New York: Doubleday.
Harrison, B. (1994) Industrial districts: old wine in new bottles? Regional Studies, 26.
Jacobs, D. and de Jong, M. (1992) Industrial clusters and the competitiveness of the Netherlands: Empirical and conceptual issues. De Economist, 14.
Jacobs, D. and de Man, A. (1996) Clusters, industrial policy and firm Strategy: A menu approach. Technology Analysis and Strategic Management, 8.
Keeble, D. and Wilkinson, F. (2000) (Eds) High- Technology Clusters, Networking and Collective Learning in Europe, Aldershot: Ahsgate.
Krugman, P. (1991) Geography and Trade. Cambridge: MIT Press.
Krugman, P. (1996) The localisation of the global economy, in Krugman, P. Pop Internationalism, Cambridge, Mass: MIT Press.
Landabaso M., Oughton C., and K. Morgan (1999). Innovation Networks—Concepts and Challenges in the European Perspective. Paper presented at the Fraunhofer Institute in Karlsruhe, Germany, November 18, 1999.
Le Goff, J., (1986). La Bourse et la Vie: Economie et Religión au Moyen Age. Paris: Hachette.
Malecki, E. and Tootle, D. (1997) Networks of small manufacturers in the USA: Creating embeddedness. In M.Taylor and S. Conti (Eds) Interdependent and Uneven Development. Aldershot: Ashgate.
Markusen, A. (1996) Sticky places in slippery space: A typology of industrial districts. Economic Geography, 72, 293-313.
Marshall, A. J. (1891). Principles of Economics. London: Macmillan.
Morgan, K. (1997). The learning region: Institutions, innovation and regional renewal.
Regional Studies, 31.
Nachira, F. (2002). Towards a network of digital business ecosystems fostering local development. European Commission, IST Programme, Brussels, September. openpub.
National Governors Association, (2002). A Governor’s Guide to Cluster-Based Economic Development. NGA, Washington, D.C.
OECD, (1999) Boosting Innovation, The Cluster Approach.
OECD, (2001) Innovative Clusters: Drivers of Innovation Systems. Paris: Organization for Economic Cooperation and Development.
Piore, M. and Sabel, C. (1984). The Second Industrial Divide. New York: Basic Books.
Pirenne, H. (1937). Economic and Social History of Medieval Europe. New York: Harcourt Brace and World.
Porter, M. (1990) The Competitive Advantage of Nations. London: Macmillan.
Porter, M. (1998) Clusters and the new economics of competition, Harvard Business Review, November/December.
Porter, M. (2000). Location, competition and economic development: Local clusters in a global economy, Economic Development Quarterly, 14.
Porter, M. (2000). Regions and the new economics of competition, in Scott, A.J. (Ed) Global City-Regions, Oxford: Oxford University Press,.
Porter, M. (2001). Clusters of Innovation: Regional Foundations of U.S. Competitiveness. Washington DC: Council on Competitiveness.
Rosenfeld, S. A. (2002). Just Clusters: Economic development strategies that reach more people and places. Carrboro, NC: Regional Technology Strategies, Inc..
Rosenfeld, Stuart A. (2002). Creating Smart Systems: A guide to cluster strategies in less favored regions. Carrboro, NC: Regional Technology Strategies, Inc.
Rosenfeld, Stuart A. (1997). “Bringing business clusters into the mainstream of economic development.” European Planning Studies 5 (1): 3-23.
Sassen, S. (2002). Global Networks, Linked Cities, New York: Routledge.
Saxenian, A-L. (1994) Regional Advantage: Culture and Competition in Silicon Valley and Route 128. Cambridge, Mass. Harvard University Press.
Williamson, O. (1975). Markets and Hierarchies: Analysis and Antitrust Implications. Free Press.
APPENDIX
The I-Space: a conceptual framework
In this appendix, we present a conceptual framework, the Information-Space or I-Space, through which to explore some characteristics of information and knowledge flows relevant to growth node research, and how these might be affected by the new information and communication technologies (ICTs) as well as their spatial implications. We look at the dynamic aspects of these information flows, how they give rise to learning processes and how competitive advantage can be extracted from these. We also examine culture and institutions in the I-Space. Four different types of cultures and institutions are identified and discussed and their spatial attributes explored.
Information flows and knowledge diffusion
Before introducing the model, however, it is useful to clarify the differences between data, information and knowledge. We take data to mean discernable differences between states of the world, i.e., hot/cold, dark/light, present/absent, etc. Such differences can be represented by binary digits, or bits, some of which will carry information. Information, however, is a relation between these discernable states and an observer. A given state is informative for someone. It may carry no information for his/her neighbor. Information is what an observer will extract from data as a function of his/her expectations or prior knowledge (Boisot, 1998).
By drawing distinctions between data, information and knowledge goods, we can see that the diffusion of data goods is entirely dependent on the physical characteristics and capacities of a communication channel. The diffusion of information goods depends both on the physical characteristics and capacities of a communication channel as well as on the efficiency of the coding scheme by means of which relevant information is extracted from data. Their effective diffusion depends on senders and receivers having shared coding schemes.
Knowledge goods provide a context within which information goods can be interpreted. Similarly, information goods may influence the creation and sharing of knowledge goods. The diffusion of knowledge goods entails the sharing of context and expectations. The more extensive the sharing of context and expectations by communicating agents, the more effective and extensive can be the transmission of information goods. However, the sharing of context and expectation is a very time-consuming, interactive and high bandwidth activity that can often only be achieved through the ‘co-presence’ of the communicating parties. The foregoing suggests the following propositions:
• The more easily data can be structured, transported and converted into information, the more diffusible it becomes.
• The less data that has been so structured requires a shared context for its diffusion, the more diffusible it becomes.
These propositions establish a relation between the extent to which data can be structured and the extent to which they can be shared. Together, they underpin a simple conceptual framework, the Information Space or I-Space that will be used as a point of reference for the material that follows. The framework is described in detail in Boisot (1995) and (1998). The following sections briefly present the basic premises on which the I-Space is built (further details can be found in the appendix).
The I-Space takes codification and abstraction as the two processes through which data is apprehended and structured (figure 1). Codification facilitates the distinction between phenomena as well as between the categories to which these are assigned. Codification can be measured by the amount of data processing required to perform a distinction either between two phenomena or between two categories. Abstraction assesses the number of categories required to apprehend a given phenomenon. The fewer the categories needed, the more abstract is the classification scheme. Well codified and abstract data economizes on both senders’ and receivers’ data processing transmission resources. Well codified and abstract data will diffuse to more agents per unit of time than uncodified and concrete data.
Figure 1: The Codification-Abstraction-Diffusion Curve in the Information Space
[pic]
Knowledge management and social learning
The activities of codification and abstraction that move one up the I-Space are cognitive and any move up the I-Space contributes to learning. The activities of codification, abstraction, diffusion, absorption, impacting, and scanning all contribute to learning. Where they take place in sequence they make up the six phases of a Social Learning Cycle (SLC) as illustrated in the figure below. The phases are described in table 1.
The Social Learning Cycle in the I-Space
[pic]
Table 2 The Six Phases of the Social Learning Cycle
|Phase |
|1. Scanning |
|Identifying threats and opportunities in generally available but often fuzzy data – i.e., weak signals. Scanning patterns such data into|
|unique or idiosyncratic insights that then become the possession of individuals or small groups. Scanning may be very rapid when the |
|data is well codified and abstract and very slow and random when the data is uncodified and context-specific |
|2. Problem Solving |
|The process of giving structure and coherence to such insights – i.e., codifying them. In this phase they are given a definite shape and|
|much of the uncertainty initially associated with them is eliminated. Problem-solving initiated in the uncodified region of the I-Space |
|is often both risky and conflict-laden. |
|3. Abstraction |
|Generalizing the application of newly codified insights to a wider range of situations. This involves reducing them to their most |
|essential features – i.e., conceptualizing them. Problem-solving and abstraction often work in tandem. |
|4. Diffusion |
|Sharing the newly created insights with a target population. The diffusion of well codified and abstract data to a large population will|
|be technically less problematic than that of data which is uncodified and context-specific. Only a sharing of context by sender and |
|receiver can speed up the diffusion of uncodified data; the probability of a shared context is inversely achieving proportional to |
|population size. |
|5. Absorption |
|Applying the new codified insights to different situations in a ‘learning by doing’ or a ‘learning by using’ fashion. Over time, such |
|codified insights come to acquire a penumbra of uncodified knowledge which helps to guide their application in particular circumstances. |
|6. Impacting |
|The embedding of abstract knowledge in concrete practices. The embedding can take place in artifacts, technical or organizational rules,|
|or in behavioral practices. Absorption and impact often work in tandem. |
Competitive advantage can be expected to accrue to those who learn how to climb the ladder, first from data to information, and then from information to knowledge. Cities and regions are engaged in learning races with each other, races in which competition and collaboration both have their place.
SLCs can have many shapes in the I-Space reflecting the different blockages that can impede the learning process. In some cases codifying things amounts to calling a spade a spade and this can be a source of conflict. In other cases, diffusion is limited to those who ‘need to know’, etc. What should be noted, however, is that not all phases of an SLC will be immediately value adding and how learners respond to this fact will tell us something about their learning strategies.
The paradox of value
In a world of physical scarcity, messages that have been made more compact through codification and abstraction will be judged more useful than messages that have not. Such messages will have had the noise squeezed out of them and will be less ambiguous or uncertain than more fuzzy ones. They will take less time to encode, transmit, and decode. They will be more portable and easier to commit to memory.
The maximum value of an item of information in the I-Space therefore will be at the point where codification, abstraction reach their highest level and diffusion is at its minimum – ie, scarcity is also at a maximum. At this point, the forces of diffusion are at a maximum, that is, the region of the I-space where the codification and abstraction of information make it most prone to extensive diffusion. The point MV is not stable and highlights a fundamental difference between information goods and physical goods: whereas in the case of physical goods, utility and scarcity are independent of each other, in the case of information goods utility and scarcity are inversely related to each other. The closer you get to maximizing the utility of an information good, the more difficult it becomes to secure its scarcity – one of the reasons that intellectual property rights pose problems of a different order to property rights in physical goods. We label the paradoxical nature of information goods with respect to value, the paradox of value.
Agents adopt two quite distinct strategies for dealing with the paradox of value:
• Information hoarding: recognizing that diffused information has no economic value, agents attempt to slow down the SLC by refraining from codifying or abstracting too much and by building barriers to the diffusion of newly codified abstract information – through patents, copyright, secrecy clauses, etc. Slowing down the SLC allows them to extract value from information in a controlled way.
• Information sharing: recognizing that, through subsequent processes of absorption, impacting and scanning, diffused information prepares the ground for further learning and knowledge creation, agents willingly share their information and watch how it is used by others. They gain first-mover advantages in being the first to initiate a new SLC and extract value from the process by participating in a succession of SLCs instead of dwelling as long as possible in a single one.
Hoarding and sharing strategies can be mixed. The skill consists in understanding the dynamics of specific SLCs and picking the strategies that fit them. It also requires strong capabilities for exploiting the full potential benefits of ICTs.
Network Cultures and Institutions
The role played by institutions in the process of economic development is well-established (North, 1990) and the role of transaction costs in shaping institutional choices is also well understood (Coase, 1937). We hypothesize that the principles of ‘internalization’ (Coase 1937, Williamson 1975) apply to territorial organizations as much as to economic ones. Problems of bounded rationality and opportunism may lead to the need for face-to-face transactions and hence to the phenomenon of spatial agglomeration. In other words, agglomeration results when transactions are internalized within a given region rather than scattered about in a wider space. These transactions lead to synergies that are the result of the dynamic interplay between actors within their social networks, potentially enabled by the use of ICTs. In essence, both territorial as well as economic organizations face internalization / externalization decisions.
Cultural features may be distinguished from each other by the way that their members structure and share information in the course of transacting with each other (Douglas, 1973; Hall 1976). Different institutional mixes embedded within a network may represent a variety of ‘strong’ and ‘weak’ cultures that interact in ways that influence the sharing of local and distant information resources and the extent to which learning progresses to enable efficient and effective knowledge management. In particular, trans-organizational social networks can generate synergies through personal exchanges that may be enabled by specific combinations of social networks, the use of ICTs, and organizational openness as a result of numerous collaborations (Castells 1996)
If codification and abstraction are taken as proxy measures for information structuring, and diffusion is taken as a proxy measure of information sharing, we can see how the different cultural features of actors connected within networks might help to institutionalize the social and economic transactions involved in the SLC. This implies that actors may locate themselves in the regions the I-Space where they feel most comfortable in terms of the cultural values associated with various technological and organizational transactional possibilities. Four different types of institutional order can be located in the I-Space that reflect the transactional possibilities offered by the respective information environments of each region (see Figure 2 and table 2).
Figure 2 Institutions in the I-Space
[pic]
Table 3 Institutions in the I-Space
[pic]
The institutions shown in figure 2 are the emergent outcomes of recurrent transactions taking place in the I-Space. Various institutions are potential attractors that exert a gravitational pull on transactions that are located within their field. Where these attractors are not too strong, they facilitate learning by speeding up the flow of the SLC through that region of the I-Space. However, where the pull that they exert becomes so powerful that they can attract transactions from any part of the I-Space, then they bring the SLC to a halt in their region of the Space, blocking any further progress of the learning process. This suggests a complex set of institutional conditions that reserve important roles for fiefs, clans, and bureaucracies, as governance structures are as important as efficient markets in fostering effective collaboration as well as competition.
-----------------------
[1] The concept of a node in a regional context was initially developed by the French Geographer, Vidal de la Blanche 1910, who borrowed the concept of nodality from the British geographer Mackinder to indicate the major crossroads that generate change of all kinds and which, as a result, have the greatest power of organization.
[2] ‘The Lisbon Review 2002-2003: an assessment of policies and reforms in Europe’, August 2002, World Economic Forum Report, Geneva ()
-----------------------
Scanning: Identifying threats and opportunities in generally available but often fuzzy data – i.e., weak signals.
Codification: The process of giving structure and coherence to such insights – i.e., codifying them.
Abstraction: Generalizing the application of newly codified insights to a wider range of situations.
Diffusion: Sharing the newly created insights with a target population.
Absorption: Applying the new codified insights to different situations in a ‘learning by doing’ or a ‘learning by using’ fashion.
Impacting: The embedding of abstract knowledge in concrete practices.
................
................
In order to avoid copyright disputes, this page is only a partial summary.
To fulfill the demand for quickly locating and searching documents.
It is intelligent file search solution for home and business.
Related download
- teaching strategies for
- terry eagleton michigan technological university
- clickersolutions home
- western literature association
- select and new scholarly activities of michael nauenberg
- preliminary roadmap
- loyola university loyola university chicago
- a review article
- smith stealthskater
- agents of socialization of children
Related searches
- how to calculate preliminary net income
- army preliminary marksmanship instruction
- preliminary change of ownership statement
- california preliminary change ownership form
- los angeles county preliminary change
- preliminary change of ownership
- preliminary change of ownership pdf
- preliminary change of ownership san diego ca
- preliminary change in ownership form
- preliminary change of ownership california
- vanguard preliminary capital gains estimates
- oregon preliminary teaching license