1. Definition and classification of industrial clusters
Mike E. Porter (1998) believes that industrial clusters are interconnected in a specific field and are in A geographically concentrated collection of companies and institutions. Industrial clusters include a group of interconnected industries and other entities that play an important role in competition. Industrial clusters often extend down to sales channels and customers, and laterally extend to manufacturers of ancillary products and industrial companies related to skills, technology or inputs. Industrial clusters include government and other institutions that provide specialized training, education, information research, and technical support. J A Theo, Rolelandt and Pimden Hertog (1998) define industrial clusters as: in order to acquire new complementary technologies, obtain benefits from complementary assets and knowledge alliances, accelerate the learning process, reduce transaction costs, overcome or build market barriers, and achieve collaborative economies. Benefits, diversification of innovation risks and highly interdependent enterprises (including specialized suppliers), knowledge-producing institutions (universities, research institutions and engineering design companies), intermediaries (brokers and consultants) and customers are interconnected through value-added chains The network formed is a group.
In their study of industrial clusters in developing countries, Peter Knorringa and J¨orgMeyer Stamer (1998) drew on Markusen's (1996) classification method of industrial areas and divided industrial clusters into the following three categories (for details, see Table 1).
Lynn Mytelka and Fulvia Farinelli (2000) divided industrial clusters into three categories based on their intrinsic relationships (see Table 2 for details).
Table 1 Industrial cluster classification Italian-style industrial cluster cluster Satellite-type industrial cluster cluster Hub-type industrial cluster
The main characteristics are mostly small and medium-sized enterprises;
Specialization Strong;
Local competition is fierce, cooperation network;
Relationships based on trust are mostly small and medium-sized enterprises;
Reliance on external enterprises;
Large-scale local enterprises and small and medium-sized enterprises based on low labor costs;
Obvious hierarchical system
The main advantages are flexible specialization;
High product quality;
p>
Innovation potential, large cost advantage;
Skill/tacit knowledge cost advantage;
Flexibility; large enterprises play an important role
Main weaknesses: path dependence ;
Facing economic environment and technological changes, adapting to slow sales and investment and relying on external players;
Limited know-how affects competitive advantage. The entire cluster relies on the performance of a few large enterprises
Typical development trajectory stagnation/decline;
Changes in the internal division of labor;
Outsourcing of some activities to other regions;
The emergence and upgrade of the wheel-and-axle structure ;
Integration of forward and backward processes to provide customers with a complete set of products or services stagnation/recession (if a large enterprise declines/stagnates);
Upgrade, changes in internal division of labor
Policy intervention collective action creates regional advantages;
Typical tools for public and private sector joint ventures to upgrade SMEs (training and technology diffusion) Large enterprises/associations and SME support agencies Cooperation, thereby enhancing the strength of small and medium-sized enterprises
Source: Peter Knorringa/J¨orgMeyer Stamer. New Dimensions in Enterprise Co operation and Development: From Clusters to Industrial Districts. 1998, (10)
Table 2 Types of industrial clusters and its performance types Spontaneous industrial clusters
Informal clusters Organized industrial clusters Innovative clusters
Examples SuameMagazine Auto Parts Cluster in Kumasi, Ghana Nnewi Auto Parts Manufacturing Cluster in Nigeria , Sialkot Surgical Instrument Cluster in Pakistan, Jutland Furniture Cluster in Denmark, Belluno Eyewear Cluster in Italy
Key players’ participation is low to high
Enterprises Size Individuals, small and medium-sized enterprises, small and medium-sized enterprises and large enterprises
Innovation is almost non-existent and somewhat sustainable
Trust is almost not high and high
Skills are low, medium and high
Technology is low, medium, medium
Relationships are somewhat extensive
Cooperation is almost non-existent, not sustained high
Competition is high, medium to high
Product innovation is almost not sustainable
Exports are almost medium to high
Source: Lynn Mytelka and Fulvia Farinelli (2000) adapted from UNCTAD (1998P8).
2. Research on the mechanism of industrial clusters
The research on clusters can be traced back to Marshall. Marshall (1920) explained the phenomenon of enterprises based on external economies concentrating in the same location. He discovered the close relationship between the external economy and industrial clusters. He believes that industrial clusters are caused by externalities. Marshall believes that the external economy includes three types: the scale effect of intermediate inputs brought about by the expansion of market scale; the scale effect of the labor market; information exchange and technology diffusion. The first two are called pecuniary externalities, that is, the external economy formed by scale effects.
The latter is the technical external economy. Alfred Weber (1929) first proposed the concept of agglomeration economy. He first used agglomerative factors when analyzing the location distribution of a single industry. Subsequently, August Losch (1954) and P Sargant Florence (1948) further elaborated on agglomeration economy.
2 Overview of foreign industrial cluster research
Krugman developed his agglomeration economic perspective through his new trade theory, and the theoretical basis is still increasing returns. His industrial agglomeration model assumes that a country has two locations and two production activities (agriculture and manufacturing). Under the combined effects of economies of scale, low transportation costs and high manufacturing inputs, through mathematical model analysis, it proves that industrial Agglomeration will lead to the formation of manufacturing hubs. In addition, his monopolistic competition model, based on the integration of traditional economic geography theory, comprehensively considers a variety of influencing factors: increasing returns, self-organization theory, the role of centripetal force and centrifugal force, proving low transportation costs, high manufacturing proportion and Scale is conducive to the formation of regional agglomeration. Andersen (1994) analyzed the shortcomings of traditional Schumpeterian analysis of innovation relevance, advocated the use of evolutionary economics to analyze innovation relevance, and constructed a two-industry model and three-industry model of interactive innovation within the framework of evolutionary economics. model, exploring issues of innovation linkages and international specialization.
UNCTAD (1977) divided the cooperation models between enterprises into: groups, networks and strategic partners, discussed the effects of different cooperation models on enterprise capabilities and competition, and proposed from the levels of government, enterprises and intermediaries policy recommendations. Alex Hoen (1997) classified groups from a theoretical perspective: the concept of groups is divided into micro-level (enterprise groups), meso-level and macro-level groups (industry clusters); enterprises within the group are usually connected through innovation chains and product chains. LynnMytelka and FulviaFarinelli (2000) adopted a different classification method of industrial clusters than Markusen (1996). They divided industrial clusters into: informal clusters, organized clusters and innovation clusters. Discuss how to cultivate innovation groups and establish innovation systems in traditional industries so that traditional industries can maintain sustainable competitive advantages.
Magnus Holmen and Staffan Jacobsson (1998) discussed the problem of determining industrial clusters. The traditional input-output analysis and user-supplier relationship are based on products and industries, which are not suitable for determining industrial clusters based on knowledge externalities and diffusion. Appropriately, it proposes a new patent-based method for determining industrial clusters. Gabriel Yoguel, Marta Novick and Anabel Marin (2000) studied Volkswagen's enterprises in Argentina and explored the relevance, innovation capabilities and social management skills (organization of work processes and contract formation mechanisms) within the group from the perspective of the production network (group). . J VernonHenderson, ZmarakShalizi and AnthonyJ Venables (2000) explored why industries cluster, how new clusters are formed, and the consequences of breaking away from clusters from the perspectives of economic development and geography. In order to explain the above issues, they conducted an empirical study on the geographical characteristics of international and domestic economies. SumaS Athreyr (2001) conducted an empirical study on the growth and changes of the Cambridge high-tech cluster, and explored how the Cambridge high-tech cluster grew and changed, which microeconomic factors can explain these phenomena, and why Cambridge high-tech has not reached the level of Silicon Valley, etc. question. The theoretical basis of its research is the relationship between economic organization and agglomeration.
Aldo Romano, Giuseppina Passiante and Valerio Elia (2001) analyzed 29 virtual groups and replaced the traditional concept of geographical proximity with the concept of organizational proximity. They believed that organizational proximity is a new source of motivation for the formation of virtual groups, and organizational proximity Proximity is achieved through supply chain and customer relationship management.
They break through the geographical limitations of traditional industrial clusters, use the advancement of information and communication technology to place industrial clusters in a global virtual learning environment, and expand the space for industrial cluster activities. HenryG Overman, Stephen Redding and AnthonyJ Venables (2001) discussed the mode of trade flow, factor prices and the location of production from the perspective of economic geography, analyzed the determinants of trade costs and the impact of trade costs on trade flows, and believed that geographical conditions are factor prices important determinants, and proposed the mechanism by which geographically based trade flows and factor prices influence the emergence and development of industrial clusters.
3. Research on the relationship between industrial clusters and economic growth
Philippe Martin and GianmarcoI P Ottaviano (2001) integrated Krugman’s new economic geography theory and Romer’s endogenous growth theory and established It provides a self-reinforcing model between economic growth and the spatial agglomeration of economic activities; it proves that the spatial agglomeration of regional economic activities stimulates economic growth by reducing the cost of innovation. In turn, since centripetal force makes new enterprises tend to locate in this area, economic growth further promotes spatial agglomeration, further verifying the famous Myrdal's "circulation and causal accumulation theory." In other words, companies prefer regions with larger market sizes, and market expansion is related to the number of regional companies. AnthonyJ Venables (2001) believes that new technologies have changed the impact of geography on us, but it has not eliminated our dependence on geography; geography is still an important factor in international income imbalance and an important condition for industrial agglomeration. NicholasCraft and AnthonyJ Venables (2001) used the new economic geography theory to explore the important role of geographical agglomeration on economic performance, scale and location. They reviewed the decline of Europe and the rise of the United States and the future renaissance of Asia from a geographical perspective. They believed that despite the lack of high-end Quality system is an important reason for backwardness, but the important role of geographical agglomeration in economic development cannot be ignored.
3 Overview of Foreign Industrial Cluster Research
Catherine Beaudry and Peter Swann (2001) studied the ways in which the intensity of industrial clusters affects the performance of enterprises within industrial clusters. They used the number of employees as an indicator to measure the intensity of industrial clusters and conducted an empirical analysis on dozens of industries in the UK: there are positive and negative effects of industrial clusters in different industries, and there are very different effects in the computer, automobile, aviation and communication equipment manufacturing industries. Strong cluster positive effect. D Norman and J Venables (2001) discussed the scale and number of industrial clusters within the world economy based on increasing returns to scale, studied the relationship between national industrial cluster policies and the balanced development of the world economy, and the relationship between industrial clusters and the maximization of world economic welfare. It is believed that under the conditions of balanced development, the number of industrial clusters is too large and their scale is too small.
Lura Paija (2001), through an empirical analysis of the Finnish ICT industry cluster, believed that the ICT industry cluster is the engine of Finland’s knowledge-based economic growth, optimizes Finland’s industrial structure, and builds Finland’s national competitive advantage; and The development of ICT industry clusters in Finland is reviewed from the perspective of industrial policy.
4. Research on the relationship between industrial clusters and technological innovation, organizational innovation, and social capital
Dalum Holmen and Jacobsson (1999) discussed North Jutland from the perspective of the national innovation system The formation of knowledge-based clusters in the peninsula and western Sweden. Machiel van Dijk and Onder Nomaler (2000) differ from the traditional thinking that focuses on knowledge spillover and accumulation, explaining industrial dynamics from the supply side. They explained industrial dynamics from the perspective of demand. Under the premise of diversified consumer preferences and different compatibility of related technologies, they explored how the time and frequency of new technology application affect industrial dynamics, and verified the new technology application model. and the number of firms in the industry.
Lucia Cusmano (2000) discussed the role of enterprise-related research capabilities (that is, the ability of enterprises to evaluate, integrate and utilize knowledge flows generated in interactions) in technology policy and cooperative R&D, and its theoretical basis is evolutionary economics. Taking technology as knowledge and interaction as the unit of analysis, it is assumed that cooperative enterprises are heterogeneous and have complementary knowledge and capabilities. In enterprise cooperation, knowledge spillovers caused by technological externalities have costs. Enterprises' utilization of knowledge spillovers depends on their own absorptive capacity, and absorptive capacity is positively related to the enterprise's own knowledge stock and R&D investment. C J Caniels and A Romijn (2001) studied the role of the accumulation of technological capabilities in the regional development of small and medium-sized enterprises in the context of economic liberalization and international economic integration, and established a conceptual framework to analyze the mechanism by which geographical agglomeration affects the accumulation of technological capabilities, from which The advantages of agglomeration are analyzed from both macro and micro levels, and corresponding industrial cluster policies are proposed.
NicolaiJ Foss (1999) discussed the role of leadership in coordination games from the perspective of game theory, and believed that different knowledge concepts are very important for understanding leadership. NicolaiJ Foss (1999) compared capability theory and organizational economics theory, which are traditionally used to explain inter-firm relationships, and analyzed the advantages and disadvantages of capability theory and economics in explaining inter-firm relationships. Although capability theory can explain cooperation between enterprises, it lacks a theoretical foundation. Carlos Quandt (2000) believes that innovation groups and cooperation networks are the main tools to promote regional development, enhance innovation capabilities and regional competitive advantages, and reduce space and social imbalance. DirkMessner and Meyer Stamer (2000) discussed what is a network and how to understand it, and studied network governance logic from three aspects (namely, interest groups and decision-making styles, network social functional logic, and seven network issues); Finally, the role of network governance on industrial clusters and value chains is studied. Jorge Britto analyzed the network form of cooperation between enterprises, introduced factors related to network structural characteristics, and discussed the determinants of network competition. He divided the forms of cooperation between enterprises into four types: traditional networks, technology structure networks, complex technology networks and technology-based networks. Meyer Stamer (2002) analyzed the mode of enterprise cooperation within industrial clusters, studied the typical obstacles to enterprise cooperation, discussed how to overcome the adverse impact of culture on cooperation, and finally proposed to create an innovative environment through enterprise cooperation, thereby improving the efficiency of industrial clusters. Innovation capabilities and pathways to competitive advantage.
MarkLorenzen (1998) discussed the information cost based on trust and believed that there are different ways of obtaining information in different environments, and different types of trust have different information costs. Therefore, in different environments, different types of trust have corresponding dominance. Through empirical analysis, he studied the information cost characteristics of industrial cluster enterprises, explained the reasons for the existence of different trusts in different industrial clusters and the relationship between geographical proximity and information costs.
4 Overview of foreign industrial cluster research
5. Empirical analysis and industrial policy suggestions on industrial clusters
Michael Peneder (1997) conducted a study on Australian industrial cluster policy After conducting research, it is believed that the group analysis method can help determine the optimal policy tool, is very sensitive to the needs reflected by the micro-level system feedback mechanism, and emphasizes the importance of eliminating institutional obstacles and institutional distortions. Mike E. Porter (1998) believes that clusters are conducive to regions and countries gaining competitive advantages, emphasizing the advantages of clusters in obtaining employees and suppliers, specialized information, complementarity, and obtaining public goods, and discusses location The role of selection, local participation, cluster upgrading and collective collaboration in improving cluster competitiveness was reviewed, and the shortcomings of traditional industrial policies were reviewed, and new industrial policy design ideas based on industrial clusters were proposed. ShoheiKaibori introduced the current status and future development direction of Japanese industrial clusters.
Khalid Nadvi and Gerhard Halder (2002) conducted an empirical analysis of surgical instrument clusters in Sialkot, Pakistan, and Tuttlingen, Germany. The two industrial clusters are located in developing and developed countries respectively, and are at the high and low ends respectively from a technical perspective. However, they have considerable connections in production and technology. They also face the challenges of quality upgrading, low-cost competition and medical technology development. challenge. They used clusters and value chains as analytical methods to conduct an empirical analysis of the relationship between local industrial clusters and global value chains, and also studied the relationship between industrial clusters in developing and developed countries.
OECD’s research on industrial clusters is based on research on national innovation systems. The OECD's research on national innovation systems is divided into two stages. The first stage mainly studies the knowledge distribution capability evaluation system in the national innovation system. In the second stage, different research groups were formed to further deepen the research on the national innovation system. The research themes of the second phase are: institutional relevance; human resource flows; innovative enterprise behavior; innovation systems in developing countries; and industrial clusters. Industrial clusters are the subject of research, and their research scope includes: the definition of clusters, cluster innovation methods, innovation styles, performance research and difference analysis of the same clusters in different countries, the policy significance of industrial clusters, and the principles of industrial cluster policy design. The OECD conducted an empirical analysis of industrial clusters in the following countries: Denmark, Finland, Sweden, Belgium, the United States, the United Kingdom, and the Netherlands. On this basis, it raised questions that need to be deepened: how to make industrial clusters more competitive; the analysis of important knowledge issues Confirmation; design of industrial upgrading and optimization strategies; how to move from traditional competition to strategic collaboration and differentiated competition.
6. Research levels, methods and conclusions of clusters
1. Analysis levels and analysis techniques of clusters.
J A Robelandt and PimdenHertog divided the analysis levels of clusters into three levels: macro, meso and micro (see Table 3 for details).
Table 3 Different cluster analysis methods Analysis level Cluster concept analysis focus
Macro level (country) Industrial correlation in the economic structure National/regional specialization model; a large number of products and Process upgrading and innovation
Meso level (industry) Industry SWOT analysis and benchmark analysis of intra-industry and inter-industry correlations at different stages in the product chain of similar final products; exploring the need for innovation
Micro level (enterprise) inter-enterprise association: professional suppliers are concentrated around one or several core enterprises for strategic business development; chain analysis and management; cooperative innovation project development
Commonly used abroad The main methods for researching industrial clusters include: (1) input-output analysis method; (2) graph analysis; (3) consistency analysis; (4) case study method.
2 Conclusions of cluster research. To sum up, the logical relationship of foreign research on industrial clusters can be represented by the picture on the left: Foreign cluster research mainly focuses on the mechanism of industrial clusters, technological innovation, organizational innovation, social capital and the relationship between economic growth and industrial clusters, based on industry Industrial policy and empirical research aspects of clusters. Foreign scholars have studied industrial clusters from different aspects, but they still have not formed a systematic theoretical system. Foreign research focuses on empirical analysis and induction on this basis. Industrial clusters have attracted the attention of researchers from different disciplines and have become a hot spot in foreign theoretical research. Most of the foreign research on clusters appears in the form of research papers, and there is a lack of systematic research monographs. Regarding the research on industrial clusters, current theory still lags behind practice. Despite this, the research conclusions on industrial clusters have become the basis for many countries to formulate industrial policies and have achieved very good economic performance.
The revitalization of old industrial bases is a worldwide topic. Almost all traditional old industrial areas in developed countries have gone through a cycle of creation, development, prosperity, decline, and transformation. World-famous old industrial areas such as the Ruhr Industrial Area in Germany, the Great Lakes Industrial Area in the United States, and the Lorraine Industrial Area in France have all been transformed one after another and achieved results. The author selected Shenyang Tiexi Industrial Zone to conduct an empirical study on the revitalization of old industrial bases.
Industrial Clusters and Transformation of Old Industrial Bases Porter proposed the concept of "Industrial Cluster" in "National Competitiveness Advantage", that is, in a specific field (usually with a leading industry as the core), a large number of industries are closely related. Enterprises and related support institutions gather in space and form a phenomenon with strong and sustainable competitive advantages.