Industrial and Regional Clusters:

Concepts and Comparative Applications

Edward M. Bergman and Edward J. Feser

TABLE OF CONTENTS

CHAPTER ONE

Introduction

Intended Audience

Organization

CHAPTER TWO

2.1 Introduction

2.2 What are Industry Clusters?

2.3 Explaining Competitive Advantage

2.4 Theoretical Foundations

2.4.1 External Economies

2.4.2 The Innovation Environment

2.4.3 Cooperative Competition

2.4.4 Rivalry

2.4.5 Path Dependence

2.5 Summary

CHAPTER THREE

3.1 Introduction

3.2 Micro-oriented Cluster Applications

3.3 Methods of Meso Industry Cluster Analysis

3.3.1 Expert Opinion

3.3.2 Location Quotients

3.3.3 Identifying Clusters via Input-Output

3.3.4 Network Analysis

3.3.5 Surveys

3.4 Summary

CHAPTER FOUR

4.1 Introduction

4.2 Cluster Taxonomy Research and Analysis Applications

4.2.1 International Comparison of Value-chain Clusters

4.2.2 Spatial Bunching in Cluster vs. All other Firms

4.3 Regional Policy Development Frameworks

4.3.1 Modernization of North Carolina Industrial Base

4.3.2 Technology Composition of Regional Clusters

4.3.3 Visualizing Intercluster Trade and Technology Flow Networks

4.3.4 Visualizing Regional Cluster Portfolios

4.3.5 Policies for Small and Independent Firm in Clusters

4.3.6 Cluster Targeting and Tradeoff Policies

Appendix 1

Appendix 2

Bibliography

Glossary of Terms

Industrial and Regional Clusters: Concepts and Comparative Applications

Edward M. Bergman and Edward J. Feser



CHAPTER ONE



Introduction

Industry clusters refer to the tight connections that bind certain firms and industries together in various aspects of common behavior, e.g., geographic location, sources of innovation, shared suppliers and factors of production, and so forth. Industry cluster concepts date from the last century, but they have captured the imagination of active policymakers and the serious attention of scholars only in the last decade of this century. Because clustering behavior is such a pervasive aspect of modern economies and global trade, it draws the attention of many different disciplines and benefits from their scholarship. Although a consideration of research on this topic might alone justify book-length treatment, industry cluster concepts are also powerful metaphors that are used routinely to guide industrial and regional development planning throughout the world.

Drawing on classic materials and the recent burst of scholarly and policy activity, this monograph examines and demonstrates the use of industry cluster ideas as a means of understanding and shaping regional economies. The highly fluid nature of the rapidly developing body of literature and research on clusters is ideally suited to a web-based presentation, and we attempt to take full advantage of the medium here. Given the limits of a monograph presentation and the highly fluid nature of the literature, however, we focus most attention on areas where the advances have been somewhat more stable, incremental, and generally cumulative. Those concern mainly operational concepts of industry and regional clusters that have benefitted from alternative analytic approaches, and that generate relevant information for regional development policymakers. Each chapter represents our best effort to sort through, clarify and perhaps codify important concepts, definitions, labels, and methods. Occasional self-contained discussions of closely related sub-topics, explanatory sidebars, and short glossaries of terms provide additional elaboration.

Intended Audience

The monograph is most appropriate for students of regional science, regional studies, economic geography, and related fields such as regional planning and development. We often refer to concepts, without full elaboration, that are common and well-known to those closely connected fields. We provide more details when less-familiar concepts are borrowed from more distant sister fields. Industrial clusters are also of interest to students of business, development studies, international development, industrial innovation and strategy, and international trade.

At the same time, an integrated treatment of industry and regional clusters should be of interest to local and regional policy makers in all parts of the world. Other audiences are likely to include consultants and industry analysts. Indeed, it is the latter that are responsible for many existing industry and regional cluster studies.

Organization

In Chapter 2, we open by asking why we should study industry clusters and supply a working definition of what we mean by the term. The chapter continues by presenting overviews of the major theories and models that provide the underlying foundation for cluster concepts, with particular reference to regional analysis and policy applications. We make a key distinction between clusters in economic space and clusters in geographic space. Related concepts from growth poles to agglomeration economies are reviewed and related to recent contributions to the cluster literature.

In Chapter 3 we describe techniques for identifying industry clusters, including the use of expert opinion, measures of specialization, input-output methods, network analysis techniques, and surveys. Particular emphasis is placed on input-output based methods (the identification of value chain-based industry clusters ). We then present an application of an input-output based approach as an illustration. The application serves as the baseline for subsequent analysis at the regional level in Chapter 4.

Chapter 4 argues that industry cluster concepts are of greatest use to regional development scholars and practitioners when they permit a unified view of a regional economy and insight into its evolving structure. The chapter offers a sampling of how industry cluster ideas can be effectively applied to better understand–and draw policy guidance for--regional economies. Research applications extend or build upon industrial cluster concepts to investigate related development concepts, while policy applications apply industrial concepts directly to guide development decisions and policy efforts at the regional level. Efforts are made to incorporate or compare others’ work where similarity of application or sufficient detail of results permit. This section may be the most subject to amendment and expansion in future editions. If so, it may also stimulate valuable documentation of analytic approaches underlying new applications that would further enrich the other three sections as well.

CHAPTER TWO

Basic Operational Concepts and Supporting Theoretical Frameworks

2.1 Introduction

Industry clusters have become one of the most popular concepts in local and regional development research and practice. Even a cursory Internet search will turn up numerous dedicated web sites by research institutes, industry associations, consultants, and cities, states, and regions reporting cluster studies for particular localities or offering perfunctory guides to industry cluster concepts. Arizona, California, Connecticut, Florida, Minnesota, North Carolina, Ohio, Oregon, and Washington are among many states that have designed and/or implemented cluster-based economic development strategies.1 Hundreds of U.S. cities and regions have also developed cluster strategies, from Monterey Bay, California to Jacksonville, Florida. In 1999, two major U.S. associations of development practitioners (the National Association of State Development Agencies and the State Science and Technology Institute) held national day-long workshops on cluster analysis and practice.

European cities and regions have embraced the cluster concept with even more enthusiasm. Not surprising since much of the research that informs industry cluster studies originates in case studies of European regions in northern Italy, southern Germany, Great Britain, and Denmark. Industry cluster analysis informs National Innovation Systems (NIS) planning at the OECD and most members of the European Union have conducted cluster studies at the national and or industry-level (see Roelandt and den Hertog, 1999).

It seems that the question "Why study industry clusters?" could be answered simply by noting the popularity of the concept in development planning; it is clear that one cannot fully understand regional development policymaking without some knowledge, perhaps even direct experience, with industry cluster applications. But are industry clusters a passing fad, the latest craze in a field prone to embrace miracle solutions only until a more fashionable idea emerges? Certainly, at issue among some regional scholars is whether there is actually anything new or innovative about industry clusters (Enright 1996; Feser 1998).

In this monograph, we argue that the greatest value in the industry cluster concept is its capacity to help both the analyst and policymaker "see the regional economy whole." That is, industry cluster analysis is not so much an innovation in regional theory or methods, as it is a comprehensive approach for understanding regional economic conditions and trends, as well as the policy challenges and opportunities those conditions and trends portend. In large measure, industry cluster analyses and policies may be viewed as applications of a set of well-worn but rejuvenated theories of how geography helps drive economic growth and change. Industry cluster analysis can help exploit the growing wealth of regional economic data, provide a means of thinking effectively about industrial interdependence, and generate unique pictures of a regional economy that reveal more effective policy options.

Boekholt (1997, p. 1) writes that the "multitude of cluster initiatives has led to a wide spread confusion of what clusters really are, and in what way they differ from related phenomenon, such as industrial districts, technopoles, networks, and industry-research collaborations." Held (1996, p. 249) notes: "Sadly, in the rush by various governments to employ clusters, some fundamental issues have been slighted, including appropriate research methods and even the definition of the cluster itself." In this chapter, we examine the rich literature in geography, regional science, urban planning, economics and other related fields that explores core theoretical concepts on which applied and scientific industry cluster studies are based. We begin by laying out some working definitions.

2.2 What are industry clusters?

An industry clusters [link to Exhibit 2.1] may be defined very generally as a group of business enterprises and non-business organizations for whom membership within the group is an important element of each member firm’s individual competitiveness. Binding the cluster together are "buyer-supplier relationships, or common technologies, common buyers or distribution channels, or common labor pools (Enright 1996, p. 191)." Competitive firms make a competitive cluster, and, as Enright (1996) notes, economic self-interest is ultimately the glue that binds the cluster together. Though many scholars emphasize the role of trust and cooperation among cluster firms (and their role in solving joint problems and generating other benefits for cluster enterprises), in the end, a "cluster" comprised of enterprises that gain no real economic advantage from their presence in the group loses all conceptual meaning from a theoretical and policymaking perspective. Non-business organizations may include industry associations, technical and community colleges with specialized industry programs, universities, government industrial extension programs, network brokers, and the like. Such entities are frequently defined in the literature on clusters as "related and supporting institutions"; they are often a critical element in the success of the cluster.

In application, defining an industry cluster can become exceptionally difficult, particularly as competing policy objectives come into play. On the one hand, both space and time are relevant dimensions, such that the basic characteristics of the policy-relevant cluster vary widely between applications. On the other hand, data and methodological constraints may partially dictate cluster definitions. The latter is not necessarily a limitation if recognized explicitly by the analyst and policy conclusions are determined accordingly. However, if clusters are defined one way and measured another, resulting policy conclusions will clearly be tenuous.

Industry clusters may be more or less geographically concentrated. As we outline below, early regional development theories explicitly recognized that interdependence between enterprises may be distance-sensitive to varying degrees. It is entirely possible that the "binding ties" among a localized group of enterprises may very well be between a firm or firms located in a distant region. A sizable number of southern Ohio and northern Kentucky automotive components manufacturers that sell to final market assemblers in Michigan and South Carolina is one example. At one scale, an automotive cluster may appear to be concentrated in the southern Ohio-northern Kentucky region. Another level of analysis, however, might reveal an automotive cluster along the north-south axis between the traditional vehicle heartland in Michigan, concentrations of suppliers in Kentucky, Ohio, the Carolinas and Georgia, and the new southern automobile manufacturing regions in South Carolina, Alabama and Tennessee. Development officials in northern Kentucky cannot afford to ignore key linkages of local firms to enterprises farther north and south, at least if they truly want to understand what drives the competitiveness of their local industries.

Exhibit 2.2 summarizes a working set of definitions. Regional industry clusters are industry clusters that are concentrated geographically, normally within a region that constitutes a metropolitan area, labor market shed, or other functional economic unit. Regional clusters are similar, in varying respects, to Italianate industrial districts , business networks , industrial complexes of the early regional scientists (Czamanski 1976), and Maillat’s (1991, Maillat and Vasserot 1988) concept of the innovative milieu (see Enright 1996). All of those concepts hold in common the notion that geographic proximity between member enterprises lends certain competitive advantages, though the specific nature of the advantages varies slightly from concept to concept.2

Some clusters exist at the present time. Vehicle production in Detroit, computers in Silicon Valley, and flowers in The Netherlands are examples. Others may be more accurately characterized as emerging or potential . Biotechnology as a cluster is only now emerging in a limited number of regions worldwide, as advances in medicine, biology, and chemistry make it possible create entirely new products and new associations between firms, industries, universities, and other economic agents. From a policy point of view, knowing what could become a cluster (perhaps with proper policy stimulation) is frequently more critical than knowing what is a cluster. Indeed, the latter may be obvious more often than not.

Measurement issues also play some role in defining clusters. One of the only consistent and detailed sources of data on cross-industry linkages are input-output tables. Analysis of input-output patterns to identify "clusters" had its beginnings in the 1960s (see Czamanski and de Ablas, 1979, for a review), fell off in the late 1970s and 1980s, and has seen a resurgence with the recent policy interest in industry clusters. What has effectively occurred is a merging of traditional regional science methodological techniques with a conceptual framework based largely in strategic management, industrial organization, and economics proper.

This blending of perspectives and research interests has both advantages and disadvantages. On the one hand, analytical methods developed by early regional scientists and recently advanced in cluster applications lend analytical rigor to more the qualitative and pragmatic research approaches common to the strategic management and industrial organization literatures. On the other hand, partly as a result of conceptual differences and partly because of data and definitional limitations, input-output based derivations cannot fully capture the set of interrelationships specified in the modern industry cluster concept. Thus, it is important that input-output based clusters be clearly distinguished.

In this monograph, we identify value-chain industry clusters as those defined primarily on the basis of trading patterns among member enterprises.3 Trade between enterprises need not be direct (it might be indirect through tertiary partners). Moreover, it is possible (as we argue in Chapter 3), that input-output methods can identify a set of enterprises and industries that constitute the most likely candidates (or "suspects") for non-trade-based dependencies (i.e., linkages based not on trade, but rather similarities of technology or shared labor pools).

2.3 Explaining Competitive Advantage 4

Real policy interest in regional industry clusters has its origins in Michael Porter’s The Competitive Advantage of Nations, published in 1990. Porter’s readable account of the sources of national competitive advantage, which includes a key role for geographic proximity, is largely consistent with a growing body of literature on how interdependence between firms, industries, and public and quasi-public institutions affects innovation and growth in regional agglomerations. Porters’ work, which included case studies of competitiveness in multiple nations, also offered some anecdotal verification of highly theoretical research on the role of business externalities and spillovers in driving growth and innovation. Nearly every analysis of industry clusters begins with--or at least makes some mention of--Porter’s "diamond," a characterization of his four key drivers of competitiveness.

[Exhibit 2.3]

For Porter, industries’ success in international markets are the primary barometer of the competitive strength of a nation. The success of any given firm can be traced to four major factors: 1) the nature of firm strategy, structure and rivalry in the country, including attitudes toward competition, market institutions, the degree of local competition, and other cultural and historical factors affecting how firms do business with each other, their workers, and the government; 2) factor conditions, or the basic endowments or conditions on which the firm seeks to compete (e.g., cost-related basic factors such as ready supplies of natural resources or inexpensive, unskilled labor versus knowledge and/or technology related advanced factors); 3) demand conditions or the nature of local demand (e.g., the needs and wants of the consumer for foreign and domestic goods as well as the existence of local industrial demand for related intermediate goods); 4) the presence of related and supporting industries, including suppliers and successful competitors (both to stimulate cooperation, the latter to also stimulate rivalry).

Competitive companies must depend, to a degree, on the competitiveness of their intermediate input suppliers, who must depend on the capabilities of their suppliers, and so on, back through all links in the value chain. But such companies also depend on service providers (management, marketing, financing, legal, etc.), sources of basic and applied R&D (e.g., universities and/or contract research organizations), capital goods suppliers, wholesalers and distributors, and suppliers of trained workers (again, universities and colleges). Even competitors are important, including direct competitors to the company as well as competitors to the company’s suppliers, since their presence maintains pressure to continually upgrade processes and techniques and to seek new opportunities. Competitors also provide opportunities for cooperation in solving joint problems or addressing industry-wide issues (see also Best 1990).

Thus, the success of an individual company may be partly traced to the size, depth, and nature of the cluster of related and supporting enterprises--both private and public--of which it is a part. Much of Porter’s analysis focuses on outlining the basic conditions determining cluster competitiveness. His framework leads naturally to a focus on end-market sectors as the point of departure for studying clusters. But such end-market industries should not be studied in isolation; the critical function of interdependence in the process of economic growth and change–not just in terms of how it has traditionally been viewed, i.e., in technical or input-output terms--is the guiding principle in his study.

Porter does not argue that the dynamics that characterize industry clusters are necessarily localized in scope, though he does believe that clusters tend to be geographically concentrated. To complicate matters, the degree of economic and geographic clustering one observes for a particular end-market industry is relative to space, time, and scale. An end-market sector clustered geographically from a national perspective may not be spatially clustered from a regional or local one (or vice versa). Moreover, a given sector becomes more economically clustered as vertical, horizontal, and lateral linkages and relationships expand and deepen (with growth in related and supporting industries and/or the establishment of stronger ties or networking with existing enterprises). And there is no reason, a priori, to assume that clustering along either dimension need only increase over time, even with economic growth and nominal increases in various elements of the cluster. Changes in the social, cultural, or political environment could lead to altered relations between cluster firms such that the positive synergies described by Porter are reduced. Alternatively, improvements in the transportation or communication infrastructure may lead to some spatial dispersal of cluster firms and a reduction in geographic clustering. Finally, there is the element of scale. True clusters are probably large, perhaps exceeding some threshold, but size alone does not guarantee clustering.

Porter’s ideas are not without important antecedents. In the 1950s, Francois Perroux argued that to understand economic growth and change, analysts need to focus on the role of propulsive industries, those industries that dominate other sectors because of their large size, considerable market power, and/or role as lead innovators. Propulsive industries (or even individual firms) represent poles of growth which attract, focus, and direct other economic resources (Darwent 1969). Such constellations of producers, suppliers, and other economic actors sound surprisingly like clusters.

Perroux (1950) viewed economic space as the non-spatial sphere in which relations between firms and their buyers and suppliers (as well as other key economic institutions) take place. For Perroux, there is no reason why physical space should necessarily bear any relationship to economic space; enterprise linkages will extend without spatial limit throughout the globe, at least where they are economically justified. Directing one’s analysis to particular regions will only provide a distorted picture of the growth and development process (geographic space as ‘banal’).

The similarities between the cluster concept and Perroux’s theory of growth poles are readily apparent. The cluster focus on how end-market industries drive the deep and broad value chains of which they are a leading part is consistent with propulsive industries as dominant economic actors. End-market industries in given clusters transmit growth pulses through the cluster through demand for intermediate and capital goods. In addition, because they are composed of internationally competitive, best practice firms, they may play an important role as diffusers of process and product innovations. To the degree, for example, that large original equipment manufacturers (OEMs) can use their market power to dictate (or perhaps strongly encourage, even assist with) technology upgrades and improved manufacturing strategies to their suppliers, such end-market industries might be said to drive, at least in part, overall cluster competitiveness. On the other hand, one can also conceive of market power among some cluster members as exerting a detrimental influence on the overall cluster. For example, short-term, least cost-focused contracting practices of OEMs with their suppliers may actually discourage strategic thinking and investment.

Perroux’s ideas, and their extensions, are relevant for the industry cluster concept in another respect: his theory gave rise to a related regional development strategy (growth centers) that enjoyed a meteoric rise in popularity in policy circles only to eventually prove a dismal failure. While it is too soon to tell whether industry cluster policies will be similarly ineffective, the rise in their stock appears nearly as dramatic.

The failure of growth center policy is described in detail elsewhere (Higgins 1983, Higgins and Savoie 1995). But one of the most important reasons why such strategies misfired more often than they succeeded is that too little attention was paid to the economic and social pre-requisites that are necessary–at least as hypothesized in the vast theoretical literature–for growth centers to work (Malizia and Feser 1998). Most applications focused on the role that backward and forward linkages to strategic and favored sectors could play to leverage regional growth, particularly in underdeveloped areas (Hirschman 1958). But political and equity considerations often dictated, through a criterion of need rather than potential, the designation of very small and peripheral towns as "growth centers." Linkages were regarded almost mechanically, as if localized interindustry trade would automatically flow in a form resembling the average patterns observed in input-output accounts. A critical difference between growth centers and the industry cluster concept is that the latter emphasizes why businesses choose to align themselves or partner in trade (where location is only one possible reason), while the former effectively assumed such partnerships were inevitable.

Still, there is no question that industry clusters identified in practice often bear little resemblance to Porter’s ideal type. It is not uncommon for local and regional agencies to designate clusters for policy attention that are actually very poorly developed or that constitute the only viable industry in the given region (designations motivated more by limited choice sets or politics than any economic rationale; Held 1996). While most of the current cluster debate is taking place in industrialized countries with already diverse economies and relatively strong effective demand (domestic and/or international), in less developed regions a policy decision to concentrate resources on key industries, instead of more general infrastructure needs or other strategies that would serve best a broad array of industries, brings with it significant risks against which the gains remain unverified.

If one thing is clear, it is that Porter’s eloquent and convincing account of economic interdependence, geography, and competitiveness is short on specifics. As a result, most of the literature takes his concepts only as a point of departure, tapping instead a wide range of more developed ideas to explain the origins of industry clusters, the dynamics of cluster growth and change, and advantages to using clusters as a basis for regional policy. The following section outlines a set of core concepts that are frequently cited or called upon to either explain cluster dynamics or legitimate cluster-based policies.

2.4 Theoretical Foundations

Some analysts of the behavioral phenomena behind industry clusters emphasize explanations for observed spatial clustering of business enterprises (e.g., Enright 1990, 1996), with theories of business externalities, agglomeration economies, labor pooling, and knowledge spillovers the main focus. Others stress the link between innovation and clustering, drawing on theories of growth poles, development blocks, and Schumpeterian entrepreneurship (DeBresson 1996). Doeringer and Terkla (1997) specify three major drivers of industry clustering: 1) strategic business opportunities derived from specific kinds of interfirm alliances ; 2) traditional regional factor market advantages (labor pools and localized knowledge spillovers); and 3) the role of non-business institutions such as universities, colleges, trade unions, and associations. Rivalry, just-in-time trends, niche marketing, and civic capacity are among the concepts in their survey.

In the end, there is no obvious organizational scheme for laying out and drawing connections between relevant theories. Our own reading of the literature suggests that five core theoretical concepts underpin the literature on regional industry clusters: external economies, the innovation environment, cooperative competition, interfirm rivalry, and path dependence. We make no attempt to be comprehensive and exhaustive in our discussion; readers should consult the many citations to flesh out ideas in more detail.

2.4.1 External Economies

Regional scientists and geographers are keenly interested in how and why enterprises cluster in geographic space, and particularly how such clustering influences regional development paths. Two basic conceptual approaches to understanding benefits to concentration dominate the literature: industrial location theory that builds on Weber and Hoover (1937), where the benefits are called agglomeration economies, and the Marshallian perspective that takes as its point of departure Marshall’s ([1890] 1961) analysis of external scale economies and their presence in "industrial districts." In both cases, various, often anecdotal, types of externalities (or, more appropriately, sources of externalities) are cited as the reason why firms co-locate. The literatures differ somewhat in their relative emphasis on static versus dynamic externalities, while neither perspective is particularly concerned with distinguishing between pecuniary and technological externalities.

Industrial Location Theory. Weber (1929) identifies agglomeration economies–defined as cost savings firms enjoy as a result of increased spatial concentration–as one of three primary causes of spatial clustering or agglomeration. But Weber is not particularly concerned with why such agglomeration economies arise, preferring to suggest that they are simply external varieties of internal scale economies (see Weber, 1929, p. 127). In point of fact, his primary aim was to model how such economies might lead to agglomeration (rather than identify what explains the economies themselves). It was a theoretical approach and methodological emphasis that eventually became the traditional regional science/urban economics approach to the study of externalities.

Though Hoover is nearly as vague as Weber, he does introduce the now accepted distinction between urbanization and localization economies. In the cluster literature, the focus is mainly on externalities related to proximity among business enterprises (localization economies), rather than externalities associated with general urban advantages (urbanization economies). Other researchers cite particular advantages of proximity between firms, including increased market power through brokered buying and selling, the better availability and use of specialized repair facilities, shared infrastructure, reduced risk and uncertainty for aspiring entrepreneurs, and better information (Isard 1956, Lichtenberg 1960, Vernon 1960, Carlino 1979). In a recent monograph intended to guide practitioners, Rosenfeld (1995a, p. 20) cites "tailored infrastructure" as a key advantage firms in regional clusters enjoy. He uses a scale economy logic: "As industry concentration increases, individual businesses benefit from the development of sophisticated institutional and physical infrastructure tailored to the needs of specific industry." Such infrastructure includes "local product showrooms, foreign sales offices or distribution centers, supply centers, common waste treatment facilities."

Marshallian Theory. Marshall ([1890] 1961) defines external scale economies as cost savings accruing to the firm because of size or growth of output in industry generally. Such economies contrast directly with internal scale economies, which are the source of increasing returns from growth in the size of plant. Such external economies are essentially spatial externalities, which may be defined generally as economic side-effects of proximity between economic actors. They can be either negative or positive, static or dynamic, pecuniary or technological. The static variety are reversible, whereas dynamic externalities are those associated with the technological advances, increased specialization, and division of labor that accompanies and/or drives growth and development (Young 1928).

For the most part, regional scientists are interested in dynamic external economies, though this is not made explicit. A static external economy enjoyed by a firm in a given industrial district might be the lower costs it enjoys for intermediate inputs because of proximity to its suppliers (e.g., as a result of reduced shipping costs). That economy is also pecuniary and imposes no market failure since it is fully reflected in the price mechanism. There is certainly no role for government to encourage geographic clustering in this context. To the degree that such benefits outweigh any costs associated with agglomeration (congestion), enterprises will be inclined to cluster on their own.

Of most relevance for understanding industry clusters are dynamic external economies associated with learning, innovation, and increased specialization. Marshall illustrates the workings of (largely dynamic) external economies with reference to concentrated industrial districts, places where firms enjoy the benefits of large, skilled pools of labor, greater opportunities for intensive specialization (a finer social division of labor), and heightened diffusion of industry-specific knowledge and information (knowledge spillovers). Behind those dynamics is not just the size of the district alone, but social, cultural and political factors, including trust, business customs, social ties, and other institutional considerations (Bellandi 1989). Much of Marshall’s analysis is relevant to Porter’s (1990) discussion of firm structure, strategy and rivalry as one of the four determinants of competitiveness (Peneder 1995), about which we have more to say below. In effect, Marshall provides some of the first hints as to how micro-level business relationships might influence regional growth and development.5

2.4.2 The Innovation Environment

Just as enterprises do not conduct business in isolation, they do not innovate in isolation. The innovation environment constitutes research that attempts to the characteristics of "learning economies," economies that help sustain the perpetual research and innovation necessary to continually generate new products and open new markets. According to Roelandt and den Hertog (1999, p. 1):

In modern innovation theory the strategic behaviour and alliances of firms, as well as the interaction and knowledge exchange between firms, research institutes, universities and other institutions, are at the heart of the analysis of innovation processes. Innovation and the upgrading of productive capacity is seen as a dynamic social process that evolves most successfully in a network in which intensive interaction exists between those ‘producing’ and those ‘purchasing and using’ knowledge.

Industry clusters are regarded as an important tool in policies related to national innovation systems (NIS), an important theoretical framework in European national and regional policy circles. Lundvall (1992) defines NIS as "the elements and relationships which interact in the production, diffusion and use of new, and economically useful knowledge. . .that are either located within or rooted inside the borders of a nationstate (quoted in Roelandt and den Hertog 1999, p. 2). In a more recent contribution, Lundvall (1999, p. 2) argues that increases in the rate of innovation dictate important changes in national and regional policy, with particular emphasis on contributions to "the learning capability of firms, knowledge institutions and people."

Industry clusters and networks can serve as mechanisms whereby firms exchange knowledge and information that cannot be codified. Such tacit forms of knowledge are viewed as increasingly important given the rapidly changing global economic environment. Tacit knowledge must also be exchanged between individuals, not business entities, reinforcing advantages to spatial clustering (Lundvall 1999). Such advantages are likely to be strongest for technology-intensive firms, yet even traditional, design-oriented sectors such as furniture and apparel, may seek to improve flexibility and ability to innovate by clustering in particular regions. Characteristics of the regional environment may also play a role in helping firms innovate. Saxenian (1994), for example, highlights even land use and design issues in describing the unique capacity of Silicon Valley to promote innovation.

Another perspective is that of the "innovative milieu." According to Maillat (1991, p. 113):

The milieu must be envisaged in such a manner that it has a significant action on the manner of giving life to the innovation process. The milieu is not a warehouse from which one obtains supplies, it is a complex which is capable of initiating a synergetic process. From this point of view the milieu cannot be defined merely as a geographical area, it must be envisaged as an organization, a complex system made up of economic and technological interdependencies.

Like research on industrial districts, literature on the milieu focuses on the specific nature and quality of transactions, alliances and partnerships between enterprises. But the focus is less on bilateral ties than the degree to which they support a collective environment for innovation (Malecki 1997).

2.4.3 Cooperative Competition

One of the predominant themes in the industry cluster literature is "cooperative competition," the notion that the most competitive firms find ways to work together even as they go head to head in the development of new products and the battle for markets. Out is the notion that companies minimize risks and maximize their competitive position by strictly regulating any information exchange with direct competitors. Modes of cooperation based on trust, familial ties, and tradition are most often described for industrial districts in Third Italy, where they are believed to be one means by which small and medium-sized enterprises seek to counter internal scale economies enjoyed by their larger competitors (see Asheim 1997, Park and Markusen 1995, Park 1997). The new industrial districts literature, in turn, draws on theories of flexible specialization (Isaksen 1997, Asheim and Isaksen 1997, Heidenreich 1996), though the latters’ focus on substantiating a basic sea-change in the organization of production is less important for understanding the specific, micro-level relationships that undergird regional industry clusters.

Of more significance are recent efforts to clarify the general relationships between scale and scope economies (e.g., Bellandi 1996), as well as the many case studies of particular industrial districts that identify not only basic economic trends in agglomerations of smaller firms, but also social and cultural behavioral codes that govern relationships between firms in those dynamic regions (see Humphrey 1995 and related articles in World Development 23, 1). The study of the ‘social embeddedness’ of economic transactions constitutes a principle contribution of the new industrial district literature (Harrison 1992), and holds promise for making clearer the broad institutional factors Porter cites in his work.

Outside of the industrial districts literature, however, examples of cooperation between enterprises in given clusters are relatively few. Many of the characteristics of firm interdepence in Italy are culturally specific; modes of business behavior in the United States and many other industrialized countries are very different. Doeringer and Terkla (1997) offer two circumstances in which cooperation among co-located firms can payoff. The first is when just-in-time inventory and delivery systems are used. They cite the joint location choices of Japanese manufacturers and their suppliers, which is often necessary to make JIT truly work, as evidence of how cooperation drives regional industry clustering. The second example is a function of the speed and frequency of interactions between companies in a regional industry cluster. The more frequent and rapid the interaction between suppliers, the more likely companies are to identify niche markets and new specialized products. They characterize such dynamics as "collaboration economies" or "the ability to participate in, and respond rapidly to, changing design and manufacturing practices among firms that buy and sell from one another (1997, p. 182)."

The problem with both of those examples is that they apply primarily to end market producers and their suppliers, rather than between competing end market producers. As Enright (1996, p. 199) notes, the distinction between vertical and horizontal types of cooperation is important, since the potential costs and benefits of each type vary significantly. He cites lobbying, foreign market research, joint export promotion, trade fairs, and specialized infrastructure investments as typical areas in which competing producers might cooperate. On the other hand, they tend to compete in the areas of marketing, production, sales, new product development and process improvements. Contrast this view with writing on industrial districts, which focuses much attention on cooperation in production (particularly collective efforts to solve joint production problems). Ultimately, the "social embeddedness" of firm relationships means that internal dynamics of regional industry clusters are likely to vary widely between countries, and often within them.

2.4.4 Rivalry

On the face of it, the emphasis on interfirm rivalry in Porter’s analysis would seem to contradict the view that clusters are imbued with a spirit of cooperative competition. Porter adopts the traditional neoclassical view in arguing that a competitive industry structure–i.e., multiple companies competing on the same playing field–ensures continued pressure to upgrade technologies, minimize costs, innovate, and so forth. But a simple industrial concentration index is not an adequate barometer of the degree of rivalry among firms in a given industry or region. More important is the competitive ethos of the industry. Also, rivalry will likely be stronger among competing firms are geographically concentrated in a particular area. In such a case, the dimensions of competition multiply. Firms in the same region compete not just for customers, but also for labor, capital, publicity, and political support.

An early analysis of the link between market structure and geographic concentration is Chinitz’s (1961) paper on market structure as a key determinant of agglomeration economies. In a brief but rich discussion that essentially anticipates the present-day focus on how firm and industry organization influences regional development paths, Chinitz essentially draws a direct link between firm structure and rivalry and regional economic fortunes. Critiquing the agglomeration economies literature’s focus on urban and industry size, Chinitz argues that industrial structure particularly influences learning, innovation, and entrepreneurship, giving diverse, and small-firm rich places like New York a leg up over large-firm, single-industry towns like Pittsburgh. This has become an important theme in the Marshallian new industrial district theory as well.

2.4.5 Path Dependence

Polarization, core-periphery, and cumulative causation models all refer to the tendency for regional growth or decline to reinforce itself (Myrdal 1957, Friedmann 1966, Kaldor 1970). While such models emphasize disequilibrium in the space economy, with some regions establishing dominant positions vis-a-vis peripheral regions, neoclassical regional growth theory predicts that natural market mechanisms tend to gradually eliminate interregional economic disparities. The latter result is based on a constant returns world that admits no role for externalities. Neoclassical theory tended to dominate mainstream views of regional growth through the 1980s.

The debate between equilibrium and disequilibrium views of regional growth was renewed in the 1990s with recent contributions in mainstream economic growth theory that highlight the role of increasing returns. According to Krugman (1995), what accounts for the new interest in increasing returns among mainstream economists are modeling advances that permit their more rigorous and consistent treatment. New growth theory suggests that a comparative advantage established in a given region or country, perhaps by accident, chance, the distribution of natural resources, or other non-behavioral phenomena is likely to strengthen as a result of external scale economies (usually described in Marshallian terms rather than in the language of agglomeration theory).

Like the new growth theory, ‘new international economics’ also holds important implications for regional analysis. It is not that trade theory now admits a geographic dimension; trade theory has always been spatial theory (Ohlin 1933, Krugman 1991). Rather, the incorporation of increasing returns in models of trade implies the prospect of a highly concentrated geographic pattern of development (Krugman 1990), including sustained disparities in regional income and employment. Again, the focus is on knowledge-related externalities as sources of increasing returns, particularly in advanced technology industries (Krugman 1996). The process of cumulative advance in regions whose industries have established a competitive lead in given markets has been described as an example of a ‘lock-in effect’ (Arthur 1989, 1990a, 1990b). In principle, the initial lead may be as much a result of luck or historical accident as business acumen. But either way, particular ‘locational clusters’ may be able to establish a type of monopoly advantage over industries in other places. How likely or sustained such a process would be is an empirical matter (Krugman 1996).

Path dependence refers to the general notion that technological choices–even seemingly inefficient, inferior, or suboptimal ones–can assume a dominant lead over alternatives and be self-reinforcing, though not necessarily irreversible given a significant enough shock. David’s (1985) discussion of the modern keyboard is the classic example. Path dependence can have clear geographical implications by virtue of the fact that businesses, as a general rule, cluster in space. Krugman (1991, p. 60) cites the carpet industry in Dalton, Georgia. From a geographers point of view, it was certainly by chance that tufting technology was essentially invented there:

. . .in 1895 the teenaged Miss Evans made a bedspread as a gift. The recipients and their neighbors were delighted with the gift, and over the next few years Miss Evans made a number of tufted items, discovering in 1900 a trick of locking the tufts into the backing. She now began to sell the bedspreads, and she and her friends and neighbors launched a local handicraft industry that began selling items well beyond the immediate vicinity.

There was no carpet technology institute at the local university, no cluster of carpet producers in the region, and no history of carpet making among local workers. Yet Dalton became a leader in carpet production (indeed, a carpet industry cluster), scale economies and externalities reinforced its lead, and the rest is history. Because technology can be path dependent, regional development trajectories can become path dependent (see also Meyer-Stamer 1998). And stories like the one above suggest that being the first-mover can be critical to development success.

2.5 Summary

Industry clusters have become an extremely popular concept in development policy circles. This chapter presented a set of working definitions, a brief summary of Porter’s (1990) important contribution, and a discussion of five core theoretical concepts that are frequently cited in the literature as forces driving cluster change or as justifications for cluster policies.

At the present, industry cluster initiatives have seen relatively little criticism. Yet they raise fundamental empirical and policy questions (Feser 1998). On the one hand, very little evaluation of cluster-based policies has been conducted. Above we note the failure of related growth center applications. Though industry cluster policies are based on different theoretical principles in many respects, there is still evidence that the concept is often misapplied either as a sector-based approach or as wishful thinking in an underdeveloped area.

On the other hand, regional cluster initiatives, by definition, imply a policy-led attempt to strengthen regional concentrations. If industry is most competitive when geographically clustered, this may make good sense. But a traditional goal of regional policy has been to minimize regional disparities in growth and income. Unlike growth pole/growth center concepts, which at least attempted to address links between core and peripheral regions, the industry cluster theories speak very little to the spatial diffusion of growth. European unification, the North American Free Trade Agreement, and other attempts and common market creation and economic integration are bringing with them renewed focus on development imbalances. Moreover, industry clusters policies contract traditional wisdom of regional industrial diversification. While it is true that the largest places will develop multiple clusters, or specializations, the vast majority of cities and regions have little prospect of developing more than one or two viable clusters. Such issues must become central to the industry cluster debate.

End Notes

1. A recent study funded by the U.S. Economic Development Administration examined seventeen cluster initiatives across the U.S. See Gollub (1997).

2. Rosenfeld defines "business clusters" as a "geographically bounded concentration of similar, related, or complementary businesses, with active channels for business transactions, communications and dialogue, that share specialized infrastructure, labor markets and services, and that are faced with common opportunities and threats" (Rosenfeld 1995a. p. 13). The definition is consistent with "regional industry clusters" as defined here.

3. Roelandt and den Hertog (1999, p. 1) define clusters as "networks of production of strongly interdependent firms (including specialized suppliers) linked to each other in a value-adding production chain. In some cases clusters also encompass strategic alliances with universities, research institutes, knowledge intensive business services, bridging institutions (brokers, consultants) and customers." The Roelandt and den Hertog definition is most consistent with an input-output based measurement approach.

4. Parts of this section and the next draw heavily on Feser (1998).

5. Marshall, also emphasizes how important industrial districts are for small firms, which, through a social division of labor, may enjoy the same types of benefits large firms earn through internal scale.

CHAPTER THREE



Cluster Morphology of Regions: Analytic Options

3.1 Introduction

In this chapter, we examine a set of methods for identifying and analyzing industry clusters. There are a variety of tools available for the task, from simple measures of specialization (location quotients) to input-output based techniques. We begin by making a distinction between highly stylized studies of pre-determined sectors (often in the Porterian tradition) and studies that attempt to infer the identity of clusters embedded within a very diverse and reasonably comprehensive set of regional industries. The first kind, what we label "micro-level cluster applications," are typically driven by specific regional interests or policy concerns. In micro-level applications, clusters are defined as a group of firms that produce similar products (i.e., industries), but that hold key complementary informal and formal ties. The clusters may include some limited supplier chain characteristics, but in such studies, explicating value-chains is less important than characterizing ties between similar producers. Such industry-focused, firm-level studies are likely the most well-known type of industry cluster application (the many studies of industrial districts around the world are of this variety).

Most regions interested in pursuing industry cluster analysis fall into one of three categories: 1) they have become aware of their leading industries but desire an understanding of how ties among firms within those industries might be strengthened and turned to competitive advantage; 2) they are aware of their principal industries, but want to identify unseen complementarities and potential strategic alliances between those and wholly different--or perhaps as yet undeveloped-- regional industries; 3) they have little knowledge of their core regional strengths and potentials, apart from what can be gleaned from single-sector trends. Micro-level studies pursued singly (i.e., not in concert with other methods) apply most readily to cases in the first category.

For the second and third categories, techniques that permit a comprehensive investigation of virtually all sectors in the regional economy are needed. We label analysis based on such techniques "meso-level cluster applications," following terminology adopted by the OECD. Meso-level applications may very well be followed by intensive micro-level analyses of relationships between firms in identified clusters. Indeed, a two-stage industry cluster analysis is probably ideal, resources permitting. Nevertheless, meso-level cluster studies even in absence of micro analysis can generate unique and policy-relevant intelligence about the regional economy.

3.2 Micro-oriented Cluster Applications

Linking the theory (of Chapter 2) and application (as discussed here) are the motives and policy interests that drive inquiries and support regional studies of any kind. While a strong epistemological base is necessary in policymaking as a legitimate foundation for conducting empirical research, core policy interests often play a strong role in determining the nature and quality of analysis. What this means is that many cluster studies–both the definitions of clusters and the methods used to identify them–are based on political concerns or pre-determined policy options rather than established theoretical models.

Such "interest-based" empirical applications have always been the case in the field of economic development. Witness early state-level pursuit of exogenous policy levers such as "growth poles," "counter-cyclical industrial portfolios," "industrial targeting and recruiting," and the wide range of related initiatives designed to propel peripheral areas into prosperity or stave off decline in more developed regions. Such approaches invariably reflected core local interests, usually some representative derivative of basic local production factors (labor and capital) and the nation-state. As the tides have turned toward more endogenous views of regional development (e.g., the creation of local state and development partnerships, business entrepreneurship strategies, incubators, programs to build social capital, human capital and technology initiatives, and industry clusters) to cope with global risks and opportunities, different political interests, as well as communities of scholars, seek different kinds of empirical applications.

North American regional development policy as a supporting interest comes relatively late and comparatively uninformed to the strategic consideration of industry clusters. More advanced are regional bodies in which industry clusters (or the industrial district variant) have been supported and studied longer (e.g., Alpine-Adriatic Europe, particularly northern Italy). Firms and industries (particularly associations), including increasingly those in the U.S., that seek agility in a turbulent global economy, a keen understanding of core competencies, and greater advantage from localized technological spillovers have shown considerable interest in the industry cluster concept.

Such interests were clearly stimulated by early forays during late 1980s into the topic by management strategist Michael Porter (1990) and his many emulators (see Chapter 2). Porter’s first-to-market success in showing how clusters support firms’ effective strategic options blazed a simple analytic approach that, using another newly popular concept, very nearly became the "path dependent" default method of analysis. This despite the fact that Porter’s analytical methods were opaque at best (Enright 1997). The upshot is that regional development policymakers coming late to the concept usually encounter a business or industry flavored- approach to identifying and analyzing clusters that we title micro.

Micro-level studies begin with some of the same theoretical insights presented in Chapter 2 of why firms successfully co-locate with other firms in industry clusters.1 These concepts are presented in somewhat stylized form, permitting greater focus to be placed on how similar-sector firms cooperatively share production capacities, markets, labor and technologies, reserving for such Italianate arrangements the term ‘cluster.’ The underlying cooperative behavior is seen as a current that follows barely-visible local channels, such that:

The "current" of a working production system [is] less easily detected and is often embedded in trade, professional, . . .and civic associations, and in informal socialization processes. . .[such]. . .that a cluster is a "geographically bounded concentration of interdependent businesses with active channels for business transactions, dialogue, and communications, and that collectively shares common opportunities and threats (Rosenfeld 1997, p. 10)."

Rosenfeld also describes this collectivity as a ". . .a critical mass of firms in a region of the same, closely related or complementary sectors (emphasis added)." The relevant point here is that such clusters typically consist of very similar types of firms selling similar consumer or household design-intensive products. In other words, single-industry clusters set the standard for studies under consideration by development policy officials that face a large portfolio of very different, interacting industries.

Italianate industry clusters often consist of commodity or raw material inputs that are transformed by cooperating producers employing similar production technologies and cooperative cultures. The relatively short supply chains are of comparatively less importance to this definition of clusters than the factors presented by Rosenfeld. Less significant too for the success of such clusters is the underlying technological system that supports these highly effective production regimes, or the appreciation that the technological origins of production methods that support Italian consumer good clusters differ radically from the producer good clusters to be found elsewhere in the Italian economy (Debresson 1996).

The richly detailed accounts of these uniquely successful industrial groupings are instantly familiar and compelling,2 particularly to politicians and policymakers desperately seeking immediate solutions to regional economic problems, while they are also of occasional use to theorists who wish to illustrate far more complex concepts. 3 As in other ethnographic inquiries, the studies are so uniquely etched that enduring lessons and generalizations prove difficult to distill or to apply in other regional economies.

Further, such studies, by definition, limit attention to physically detectable evidence of "currents" flowing among similar sector firms that are best uncovered up close and at fairly small geographic scales by labor-intensive investigations (e.g., on-site interviews, Delphi techniques, or focus groups). Not surprisingly, this approach restricts its view to a single visible collection of similar sector firms, thereby overlooking linkages that some of its members may have with regionally co-located firms from very different sectors, or the robust clustering of other sectors. A micro-level study then tends to document one cluster per region, usually that of its policy client. Apparent indifference to the presence of additional clusters, particularly those based on alternate criteria or detectable only from a wider spatial view or from data-intensive sources, is due mainly to micro-oriented investigations of an a priori cluster definition. An implication is that significant instances of region-wide industrial clustering go unrecognized by micro studies. At the same time, the labor-intensive method of study all but precludes a region-wide investigation of all industrial clusters that might form the basis for "seeing regional economies whole."

Recognizing that regional development interests are eager to learn about all components of a local economy for which they are responsible, micro study analysts sometimes precede or accompany their proposals for detailed study of single industries by employing certain simple single-industry techniques drawn from regional analysis, which are then applied repetitively to commonly available multi-industry data. Location quotients are the most frequently applied method to identify unusually high relative concentrations of industrial activity, which in these studies are taken as evidence of "industrial clusters." The cluster studies that employ this simple technique to widely available employment data are generally indifferent to the fact that high concentrations are–in the hands of other analysts–interpreted as inferential evidence of local export production (economic base theory). Worse, and somewhat perversely, such studies often appear completely unaware that employment concentrations per se are indistinguishable proxies for total industry output, regardless of whether that production is concentrated in one huge branch plant or distributed within a "cluster" of cooperating establishments and firms.

Micro studies also tend to revolve around the needs of the focal industries to survive or thrive in their settings, and study designs are therefore geared to learning what is needed for members to act decisively in their specific economic and regional environments. These studies attempt to provide useful specificity, detail, and subtlety of how connections are made, networks are maintained, and interpersonal assets are translated into cluster advantages of utmost importance to the sponsoring clients. These interests may align or be at odds with host regions that wish to restructure their economies away from the most vulnerable to the most promising clusters. At the same time, an uninformed application of standard techniques drawn uncritically from the regional scientist’s toolbox offers little in the way of improvements that would benefit an overall regional perspective.

Micro-oriented studies of regional industry clusters are appropriate in some circumstances. When an analyst is beginning with a definitive set of industries that constitute the policy interest, the kinds of qualitative and labor-intensive research needed to truly identify evidence of clustering behavior are called for. There virtually no secondary sources of information on cooperative relationships between local companies; input-output data can only provide hints of such relationships, or perhaps the most likely suspects among which such relationships might be organized.

The following section focuses on methods designed to distill the industrial complexity of a given region in such a manner as to identify regional clusters or potential regional industry clusters. The techniques are quantitative and, for the most part, data intensive. This kind of analysis may very well be followed by a qualitative examination of specific identified clusters. Indeed, it probably makes most sense to conceive of regional cluster analysis as a two-stage process: 1) an initial scan of the regional economy, using detailed quantitative sources; 2) then a detailed, perhaps painstaking, investigation of specific industrial features/groupings identified in the scan. The two-part approach implies that the analyst is beginning with a "clean slate," that is, no restrictions or a priori predilections of the sectors that are of most import.

3.3 Methods of Meso Industry Cluster Analysis

This section identifies several ways of identifying industry clusters, with most of the detailed focus placed on input-output based methodologies. The discussion is presented from the perspective of an analyst considering issues of study design and methods. For a discussion of general cluster approaches from the perspective of the policy maker considering whether to commission a cluster study, click here.

Exhibit 3.1 lists six basic analytical approaches, ordered roughly in terms of how commonly they have been used: expert opinion, location quotients, trade-based input-output analysis, innovation-based input-output analysis, network analysis, and surveys. The following sections summarize each approach, save innovation-based input-output analysis. The latter is based on innovation survey data available in only a few countries.

3.3.1 Expert Opinion

Probably the most common approach to identifying regional clusters is the use of interviews, focus groups, Delphi survey techniques, and other means of gathering key informant information. Regional experts--industry leaders, public officials, and other key decision makers--are important sources of information about regional economic trends, characteristics, strengths and weaknesses; they are the "agents who know the region’s industries in terms of basic practice, supply chains, current investment patterns and potential opportunities for new products. . .(Stough, Stimson and Roberts 1997, p. 2)." Industry association reports, newspaper articles, and other published documents that are anecdotal or otherwise not based on systematic empirical analysis also fall under the category of "expert opinion."

While gathering expert opinion data can be relatively cost and time effective, as well as yield rich contextual information about the region’s economy, it is rarely done systematically enough that findings can be generalized. It is easy for researchers to overestimate the accuracy of strongly held opinions among key stakeholders and to forget the multitude of potential biases affecting each expert’s views, as well as each expert’s limited field of experience within the broader economy. Moreover, there have been few attempts to use expert opinion in comprehensive assessments of the regional economy (the meso-analytic approach).

Expert opinion is most commonly used in the kinds of micro studies described in section 3.2. There the threat of bias is particularly strong since the researcher is embarking on the analysis with a pre-determined sense of the most important regional sectors, actors, and relationships. Unfortunately, the literature on clusters pays scant attention to valid expert data collection techniques. There has also been comparatively little research on ways to marry expert opinion data with secondary economic data, an important feature for meso-level cluster studies. For example, if we envision a two-stage cluster analysis with a quantitative regional "scan" preceding a qualitative investigation (including the collection of expert opinions), how does one effectively merge findings from the two stages in a way that generates insight greater than the sum of the parts?

Among the few to take up that question, as well as to design an approach for scanning a range of sectors using expert opinion data, are Roberts and Stimson (1998). They describe a tool, which they title multi-sectoral qualitative analysis (MSQA), for helping identify "core competencies, economic possibilities, strategic markets, and economic risk (1998, p. 470)." The method entails a simple categorical scoring of regional sectors along on a set of performance criteria (a total of 34 in their application to Far North Queensland, Australia). The ranking of each sector as "strong," "average," or "weak" was based on "I/O table data, focus and industry leader group discussions, reviews of 30 economic reports and studies of the FNQ region, and local knowledge (1998, p. 476)." The performance of each sector is then compared by attaching weights to the scores and summing them. Roberts and Stimson suggest several different indexes that can generated from the results.

The potential of the MSQA approach for utilizing expert opinion in cluster analyses is revealed more clearly in Stough, Stimson and Roberts (1997). In an application to Northern Virginia, the authors utilized a survey of regional experts (". . .selected from industrial directories and from economic development agency bases to ensure that they represented senior officials from the region’s major industries (1997, p. 6)." Respondents evaluated the region’s competitiveness on 35 dimensions from their own firms’ perspective and from the point of view of any general regional business. Small group meetings were then held where respondents were first asked to interpret, elaborate on, or modify findings from the survey. Participants then "identified new business opportunities for the future of their sectors and then assessed the risk associated with developing these options. Out of this exercise it was possible to create alternative proposals for deepening, and stretching and leveraging the sectors (1997, p. 6)." Stough, Stimson and Roberts identify a set of future Northern Virginia industry clusters from the results.

It should be emphasized that Stough, Stimson and Roberts’ cluster findings are more consistent with a single-industry definition of clusters (as in micro studies) rather than broader a value-chain definition. Nevertheless, the MSQA technique is suggestive of ways that more systematically collected expert opinion can be incorporated in meso-level cluster analysis.

3.3.2 Location Quotients

A very common, though limited and misunderstood, means of identifying regional industry clusters is the location quotient (LQ). The location quotient is simply a ratio of employment shares: regional industry i’s share of total regional employment over national industry i’s share of total national employment. An LQ of 1.0 indicates that the regional economy has the same share of employment in industry i as the nation as a whole.4 (Note that any other measure of economic activity and/or reference area could be used depending on the analysis.) Location quotients exceeding 1.25 are usually taken as initial evidence of a regional specialization in a given sector. The many potential conceptual and measurement pitfalls in using location quotients have been described in detail by others (see, for example, Isard et al. 1998, pp. 24-6).5 Here we focus on the value they have for industry cluster analysis.

Applied in the traditional manner, location quotients say absolutely nothing about regional industry clusters. They are an industry-based technique and therefore offer no insight on interdependencies between sectors. Industry cluster studies that rely solely on location quotients to identify clusters are simply sector studies in disguise. Location quotients in concert with other techniques may contribute to a meso-level cluster analysis however.

Top-down Versus Bottom-up Industry Cluster Analysis. There are two basic types of meso-level industry cluster analyses: top-down and bottom-up (see Exhibits 3.2 and 3.3). In the bottom-up approach, the analyst seeks to identify industry clusters by beginning with individual sectors and then finding linkages with other industries and related non-business institutions. In essence, the analyst builds a picture of regional industrial interdependence from the ground up, one sector at a time. The bottom-up approach is particularly appropriate in small regions with only a few industries, or in those places with only a few sectors with non-trivial employment. Top-down industry cluster methods attempt to identify industry clusters through various data reduction techniques (statistical cluster analysis, factor analysis, and the like). They are appropriate when there is sufficient industrial diversity in the regional economy to preclude a sector-by-sector "piecing together" of the picture of regional economic interdependence. What top-down method surrender in terms of control over the analysis they gain in terms of their capacity to make sense of complexity.

Location quotients can be used in bottom-up analyses as one of several simple measures of sector performance. The full set of regional industries might be ordered alternatively by size (measured in employment, value-added, income, or other terms), number of establishments, growth rates, specialization (location quotients), change in specialization (rate of change in the location quotient), share of total regional activity, share of total national activity, change in regional and national shares, and so on. Several categories of sectors might then be selected to begin the analysis, e.g., largest sectors, major specializations, growth industries (or combinations, such as growing specializations). Input-output data (see below) or other data on formal and informal linkages may then be used to map out value chains (suppliers and buyers of the target sectors).

Ultimately, location quotients are only useful in concert with methods that utilize, in some form, information on industrial interdependence. Even then, they can only play a minor role in identifying clusters. Spatial and economic interdependence are the two key features of the regional industry cluster concept. We now turn to the principal means of studying industrial interdependence: input-output techniques.

3.3.3 Identifying Clusters via Input-Output

Regional scientists have long used a range of methodologies, including graph theory, triangularization, and factor/principal components analysis for sorting industries into groups based on input-output (IO) linkages. Czamanski and Ablas (1979) provide a useful review of early contributions. A more recent study uses statistical cluster analysis to group sectors for Alberta, Canada (Roberts 1992). U.S. Census researchers also recently used statistical cluster analysis to combine SIC sectors into groups that presumably shared the same production technologies (Abbott and Andrews 1990). Feser and Bergman (1999) use factor analysis of the U.S. input-output table to construct U.S. value-chain "templates" for use in the descriptive analysis of potential trading patterns in North Carolina (discussed in more detail below; see also Bergman 1998). Other examples of input-output based applications include Scott and Bergman (1997), Hewings et al. (1998), and Roelandt and den Hertog (1999).

An important input-output approach applied in a number of OECD countries is based on analysis of innovation interaction matrices rather than (or sometimes in concert with) traditional production flow matrices. Debresson (1996) offers a comprehensive source for techniques and examples of such analyses. Innovation matrices, derived from surveys (e.g., the Community Innovation Survey of Eurostat), describe flows of innovations between innovation-producers and innovation-users. As noted by Roelandt and den Hertog (1999, p. 5), the principal advantage of innovation matrices is "their focus on actual innovation interdependency and actual interaction between industry groups when innovating." Disadvantages are the costliness of data collection and conceptual difficulties in survey design. A survey similar to Eurostat’s Community Innovation Survey has not been conducted for the United States.

Acknowledging the considerable advances made by the innovation survey approach, we concentrate here on the analysis of production flows. We begin by describing a set of general steps in input-output cluster analyses, and particularly conceptual decisions that have to be made along the way. We then provide an example of an input-output industry cluster analysis, our own study of potential clusters in North Carolina. We then briefly contrast our approach with that of several others, mainly to highlight major methodological differences.

Analytical Steps. There are five major steps to conducting an input-output based industry cluster analysis:

1. Define industry clusters (existing or potential/emerging, localized or non-localized);

2. Determine whether a top-down or bottom-up method is appropriate;

3. If top-down, identify an analytical method (statistical cluster analysis, factor analysis, other);

4. Collect data;

5. Apply and interpret analysis.

The first step essentially entails framing the policy issue (or set of issues) the cluster analysis is intended to inform. In Chapter 2 we make a distinction between potential (possibly emerging) and existing clusters. We also emphasize that industry clusters may manifest themselves at different spatial scales. Choices regarding existing/potential and spatial scale may determine the kind of input-output data that are most appropriate for the analysis.

Whether or not an analyst should use a regional or national input-output table to identify regional clusters is usually regarded as obvious: a regional table should be used since only it provides information about regional trading patterns. But, in actuality, the decision is not so simple. It is true that only regional input-output tables provide information about existing trading patterns between sectors currently in the region (the same is the case of regionalized national input-output tables). But because such tables provide no insight regarding interdependence of industries absent in the study area, they cannot be used to explicate possible development paths or avenues for regional diversification. For that purpose, a national table must be used, or, if such existed, a "global" input-output table. Using a "global" table, one could identify industrial interdependency among sectors regardless of location and then investigate, perhaps with the help of a regionalized table, possible linkages between and among those sectors in the region. Since there is no such thing as a global table, a national table (particularly in highly diverse economies such as the United States) constitutes a workable substitute.

Once a decision regarding regional- or national-level analysis (or perhaps a combination) is reached, the analyst must decide whether to utilize a top-down or bottom-up methodology. Some regions are so small or contain so few sectors that use of a data-reduction technique is unwarranted. Connections between sectors can be identified by constructing simple measures of input usage and sales (several are defined below). In section 3.3.4, we briefly summarize some graphical network analysis techniques that are particularly appropriate for bottom-up applications. They permit the visual description of cross-sectoral linkages and can be combined (using a variety of visual dimensions) with descriptive data on regional industries to effectively "overlay" information on interdependence with indicators of regional industry performance.

Step three involves identifying a data reduction method (for top-down applications). The two most common in industry cluster studies are statistical cluster analysis and factor analysis. A principal difference between the two is that the former yields mutually exclusive groups of industries. Though this aids interpretation, it is frequently unrealistic. Due to complex trading patterns, industries tend to trade with sectors that belong to multiple clusters (though their links to each cluster vary in strength). Factor analysis can accommodate, and even provide ways to explore, this complexity. All data reduction techniques, which are themselves primarily exploratory methods, involve numerous user-defined assumptions. With today’s user-friendly statistical software, it is easy to produce a cluster or factor analysis in seconds with minimal user input other than the base data. However, default assumptions embedded in canned software routines should be carefully examined and modified as appropriate.

Procedures involved in data collection and analysis/interpretation obviously vary from case to case. Definitional considerations and data collection issues in input-output analysis, particularly for the U.S. case, are reviewed in Miller and Blair (1985).

A Note on Data Sources. The principal source of input-output data in the United States are the Benchmark Input-Output Accounts of the United States, produced twice every decade in years ending in 2 and 7. The latest table available at this writing was for 1992; 1997 is scheduled to be released in 2000. Regionalized tables for the U.S. are available from the Bureau of Economic Analysis, or from several proprietary sources. Minnesota Implan Group, Inc., for example, produces relatively inexpensive economic impact analysis software from which regionalized tables can be extracted. Regionalization techniques used in Implan software, or by any other vendor of regional analysis software (e.g., Regional Economic Models, Inc.), are well-known and can be replicated given the necessary data. Miller and Blair (1985) and Isard et al. (1998) outline various methods for regionalizing national IO tables in detail. Survey-based tables for specific regions in the U.S. are very rare. A very recent description of socioeconomic data series useful in regional analysis (including IO) is Cortright and Reamer (1998) .

Example. Here we illustrate a top-down meso-level analysis designed to identify potential clusters and sectoral interdependencies. The study was initially conducted in support of a technology diffusion program at the state-level and is reported in detail in Bergman, Feser, and Sweeney (1996), Feser and Bergman (1999), and Bergman (1998). The policy agency wanted to target specific manufacturing sectors for technology adoption assistance such that within industry value-chains, internal pressures for the diffusion of advanced production technologies would be created. The agency was also interested in identifying elements of value-chains that could be singled out for a variety of industrial development strategies [link to Appendix 1]. With those considerations in mind, we first analyzed U.S. input-output patterns to identify a set of industry cluster "templates," national-level manufacturing value chains. We then used the chains in combination with confidential establishment-level employment and wage data to characterize the presence of the chains in the state (North Carolina). Sub-state level-analyses and simple mapping of establishments in each cluster gave some indication of regional clustering patterns. Chapter 4 uses findings from the study to illustrate a range of techniques and exploratory methods for further analyzing regional industrial interdependence.

Our methodological approach uses principal components analysis on a matrix of national interindustry linkages (derived from the 1987 U.S. IO table) as the basic methodology to derive clusters. Principal components factor analysis exploits the common statistical variation among multiple variables to generate a reduced number of "principal components" that represent linear combinations of the original set of variables. Measures of interindustry direct and indirect linkages computed from the input-output accounts for each sector are treated as variables. The derived components are then rotated to a varimax solution to facilitate interpretation. The methodological details behind factor analysis are beyond the scope of this monograph; Tinsley and Tinsley (1987) provide a summary introduction.

The input into the factor analysis is a matrix of interindustry linkages between all sectors in the U.S. manufacturing economy. There are a variety of ways such matrices can be developed. As an initial approach, one can group only those industries with non-zero employment in the study region based on those sectors’ estimated patterns of commodity use and production, as revealed by the U.S. make and use tables. This involves scaling the use and make tables with study area wage data, followed by conducting a factor analysis on the resulting matrices. Note that no assumptions are made regarding where, in geographic terms, study region industries purchase their inputs or sell their outputs.

The 1987 478 x 519 U.S. use matrix (U) reports the dollar value of each of 519 commodities used by each of 478 producing U.S. I-O industries.6 To focus only on manufacturing, U can be reduced to a 362 x 519 manufacturing use matrix (UM). Given 362 x 1 vectors of total manufacturing wages by industry for the U.S. (wUS,M) and study region (wNC,M), a 362 x 519 scaled use matrix (UNC) can be derived that reports the estimated dollar value of 519 commodities used by 362 study region I-O industries:

Each cell entry in UM,W is the ratio of output of commodity i purchased by U.S. I-O industry j to the total wages paid by industry j. Applying factor analysis to the resulting n x 519 data matrix clusters industries based on commodity use patterns. The reduced 328 x 519 UNC matrix is identical, in terms of the factor analysis, to a 328 x 519 UM matrix (where the industries without a presence in the study region are removed); the use of study region wages to adjust the use matrix provides a simple means of performing this basic adjustment. Repeating similar matrix operations and factor analysis for the make matrix generates clusters based on commodity production patterns.

While such an approach reveals differences in clustering based on commodity use and production patterns, it provides no means of jointly evaluating interindustry linkages to derive one set of clusters. Thus it makes both the final derivation of clusters considerably more complicated and the interpretation of any final result more difficult. Roepke, Adams, and Wiseman (1974) suggest a different approach. First, a standard 478 x 478 interindustry transactions matrix (T) is derived from an adjusted use matrix UA, a 516 x 1 vector of

commodity outputs (OC), and a 516 x 478 commodity by industry make matrix (M):7

Each cell (aij), in T gives the dollar value of goods and services sold by row industry i to column industry j. Since industries may be related by both input and output patterns, a symmetric matrix LT is derived from T such that,

Each column in LT gives the pattern of total (input and output) linkage between the given column industry and every other (row) industry. Eliminating non-manufacturing industries from the columns of and rows of LT and subjecting to the resulting data matrix to the factor analysis generates a set of industry clusters.

The drawback of Roepke, Adams and Wiseman approach is that evidence of indirect linkages, e.g. relationships between sectors based on links between second and third tier buyers and suppliers, will be largely absent from the groupings. The third approach employs a slightly different interindustry linkage measure. Czamanski (1974) demonstrates that given, for each industry, total intermediate good purchases (p) and sales (s), the type of functional relationship between any two industries, i and j, may be expressed in terms of four coefficients (where a is defined as above):

Each coefficient is an indicator of dependence between i and j, in terms of relative purchasing and sales links:

Selecting the largest of the four coefficients for each pair of manufacturing industries yields a symmetric data matrix LU, which, when subjected to principal components analysis, generates clusters that at least partially capture indirect linkages between industries.

In this case, functional linkage between pairs of industries in isolation are investigated. Correlation analysis permits the assessment of linkages between pairs of industries based on their total patterns of sales and purchases across multiple industries. Each column (x) in a matrix of x’s, X, gives the intermediate input purchasing pattern of the column industry. Each column (y) in a matrix of y’s, Y, gives the intermediate output sales pattern of the column industry. Four correlations describe the similarities in input-output structure between two industries l and m:

When working with a reduced set of industries (e.g., only manufacturing sectors), the four correlations can be calculated for each pair of industries using alternative specifications of X and Y. One specification consists of buying and selling patterns for each member of the reduced set of industries across all other industries in the reduced set itself. Another specification consists of buying and selling patterns for each member of the reduced set of industries across all other industries, both in and out of the reduced set. In the case of an analysis of the manufacturing sector alone, interindustry correlations calculated using the second specification of X and Y also account for similarities in manufacturing industries’ sales/purchase patterns to/from non-manufacturing industries (e.g. construction, wholesaling, services).

Deriving the correlations from the first set of X and Y matrices and selecting the largest of the four between each pair of industries yields a symmetric matrix, LV. Each column of LV describes the pattern of linkage between the column industry and all other industries in the study set. Factor analysis can then be used to identify groups of related industries.

For each factor (group of industries), the analysis generates a set of loadings, which represent the correlations of the variables with the factor. The loadings provide a measure of the relative strength of the linkage between a given industry and a derived factor, where the highest loading industries on a given factor are treated as members of an industrial cluster. It is often regarded as standard procedure in factor analysis to regard only loadings greater than 0.5 (in absolute value terms) as significant or worthy of interpretation. This approach, however, does not provide a means of interpreting gradations in loadings. For example, industries with loadings exceeding 0.75 on a given cluster might be regarded as closely linked to that cluster, while industries with loadings from 0.5 to 0.75 and from 0.35 to 0.50 may be viewed as only moderately and weakly linked, respectively. For the reasons described below, analysts should adopt a combination of rules of this type. Because any approach to delineating cluster industries from factor analysis output is necessarily partially arbitrary, loadings should also be reported to allow study users to draw their own conclusions.

In interpreting the factor analytic results to identify specific industrial clusters, analysts typically face several competing objectives. First, they want to derive a set of clusters based on the most significant linkages as revealed in the IO data matrix. According to that objective, the concern is to identify the industries with the tightest linkages to each cluster (i.e., the highest loading industries for each factor), regardless of whether or not some of those industries are also tightly linked to another cluster. Frequently a second objective is to identify, to the degree possible, a set of mutually exclusive clusters in the sense that each sector would be assigned to only one cluster. Such a result facilitates cross-cluster comparisons of size and growth rates using regional economic data sources. A common third objective is to investigate the linkages both between clusters as well as between industries within each cluster. Such linkages are sometimes revealed by an examination of sectors that are only moderately or weakly related to each cluster, thus competing with the first objective.

Such multiple objectives can be met, at least partially, by distinguishing membership in each cluster according to the strength of linkage as suggested by the loading. We derived, for example, a set of "primary" and "secondary" industries. Although there are alternative means of doing this, we suggest the following definitions based on our experience. Primary industries for a given cluster are those sectors that achieve their highest loading on that factor and whose highest loading is 0.60 or higher. Secondary industries for a given cluster are those sectors that achieved loadings on the cluster equivalent to or greater than 0.35 but less than 0.60. For some clusters, the set of secondary industries will include industries with loadings exceeding 0.60 but that achieved their highest loading on a different cluster.

Based on those definitions, as a general rule, primary industries are those that are most tightly linked to a given cluster while secondary industries are those that are less-tightly or moderately linked. Considering only primary industries yields a set of mutually exclusive industrial clusters that can be used for cross-comparison purposes. But some caution should still be exercised in interpreting the clusters derived on this basis since some "secondary" industries will actually be more tightly linked to a given cluster than a few of the primary industries in the same cluster. Often the advantages of deriving a set of mutually exclusive clusters will be viewed as significant enough to warrant the pragmatic approach.

Our analysis identified 23 clusters in the U.S. manufacturing sector [see Exhibit 3.4]. Basic summary data on the 23 clusters identified in the U.S. manufacturing economy are provided in Exhibits 3.5 and 3.6 . Exhibit 3.5 represents the breakdown of the clusters when both primary and secondary sectors are included in the cluster definition; the clusters in Exhibit 3.6 are constituted solely of primary sectors. The clusters consist of heavy manufacturing (e.g., metalworking, vehicle manufacturing, chemicals and rubber, nonferrous metals), light manufacturing (e.g., electronics and computers, knitted goods, fabricated textiles, wood products, leather goods, printing and publishing), five separate food-related clusters, and several clusters closely related to other major clusters (e.g., brake and wheel products and platemaking and typesetting). With the exception of the growth in importance of key high tech clusters (electronics and computers and aerospace), the set of clusters is roughly similar to results found in earlier cluster studies conducted using input-output data from the 1960s and 1970s. Also reported in the tables is the number of 3- and 4-digit SIC sectors that make up each cluster (column 3 in each exhibit), as well number of different 2-digit SIC sectors represented (column 4).

In addition to relative size, the exhibits highlight two key features of the clusters. First, the number of component sectors in each cluster varies dramatically from 116 in the metalworking cluster to just 4 in the tobacco products cluster (when both primary and secondary industries are included in the cluster definitions). Clusters with the largest number of component sectors sometimes include multiple final market product chains, whereas smaller clusters (tobacco, dairy products, meat products, etc.) generally describe only a single major final market product chain. Second, most clusters are composed of sectors from a variety of 2-digit level SIC industries. Sectors from 10 different 2-digit SIC industries are represented in the metalworking cluster, for example; sectors from 16 different 2-digit SIC categories make up the vehicle manufacturing cluster. Therefore, although the 23 clusters are similar in number to the 20 official 2-digit SIC classifications, they are, in fact, very different in composition. Template clusters defined on the basis of interindustry linkages generate a unique picture of the manufacturing economy when used in subsequent economic analyses. See Bergman, Feser and Sweeney (1996) and Feser and Bergman (1999) for a description of the basic makeup and characteristics of the largest of the 23 U.S. clusters.

Exhibit 3.7 provides the detailed sectoral makeup of the 23 clusters. The columns labeled Cluster ID provide a rough indication of some of the linkages between the vehicle manufacturing cluster and the remaining 22 clusters, though a complete analysis is possible only with primary input-output data and detailed intersectoral comparisons. The cluster in which a given sector is most tightly linked is given in column L1. L2 and L3 report additional clusters, if any, in which the sector is also moderately linked based on our criteria.

For example, as might be expected from the high metal content of most transportation equipment industries, 20 of 58 total primary and secondary industries in the vehicle manufacturing cluster are also members of the metalworking cluster. Other sectors are members of an additional 10 clusters, with the chemicals and rubber (including plastics), printing and publishing, fabricated textile products, and electronics and computers clusters the most significant in terms of number of cross-cluster linkages. Not surprisingly, the vehicle manufacturing cluster is also closely linked to the brake and wheel products cluster, which itself shares most of its component industries with the former as well as the metalworking cluster.

For 44 of the 362 manufacturing sectors, sectoral interdependencies are too weak to qualify them as a primary industry in any cluster. Therefore, another category of industries remains that requires attention here. The last row of Exhibit 3.6 reports the total number of U.S. companies, establishments, employees, and value-added represented by such industries in 1992. At over 11 percent of total manufacturing value-added, these "independent" industries constitute a significant share of U.S. manufacturing production. Exhibit 3.8 lists the industries that failed to load as a primary industry on any cluster along with their maximum factor loading and the cluster on which this loading was achieved. 8 The most significant of the independent industries are pharmaceuticals (SIC 283), paper and paperboard mills (262-3), photographic equipment and supplies (386), and toilet preparations (2844).

Additional Points and Clarifications. In Chapter 4, we demonstrate how the cluster templates can be used to "see regional economies whole." Our example is specific to the policy needs of the technology agency that commissioned it. Nevertheless, the national templates can be used in for studies in any U.S. region, where knowledge of actual local trading patterns is not the over-riding concern but instead a means of identifying potential cluster firms is of interest. They also can be used in conjunction with bottom-up methods. Exhibit 3.9 maps out supplier linkages to the non-upholstered household furniture sector, and, using the templates, illustrates how different industries in the chain are linked to different manufacturing clusters. For a comparison of the input-output application with a micro-level approach, click here.

A number of clarifying points are in order regarding top-down, input-output illustration. First, although the use of the national table yields clusters with very specific uses, the basic techniques to derive the clusters (measures of interindustry linkages and factor analysis) can be employed in a variety of circumstances (e.g., with regional input-output tables).

Second, although the derived industry clusters are obviously based on formal trading patterns, the construction of the linkage measures in combination with the factor analysis means that many indirect trading patterns are considered. The clusters may be viewed, in one sense, as an excellent first guess of what sectors are likely to engage in both formal and informal kinds of cooperative behavior, that is, if we believe cooperative relationships are most likely to occur between firms in sectors with rough technological affinities. This is another instance when IO based approaches can provide support to micro or more qualitative analyses.

Third, early regional science research on industrial complexes (see definitions in Chapter 2) has already demonstrated that it is a mistake to attempt to replicate the national industrial mix at the regional level. The templates do not provide a blueprint for how any region should develop, but rather serve as an analytical device to further analyze regional industrial interdependence. This will become clearer in Section 4.

3.3.4 Network Analysis

A relatively novel way of identifying industry clusters is through network analysis of linkages between firms or sectors. The most obvious data sources are trade or innovation-based input-output tables, however surveys of regional experts or other qualitative sources of connections between regional industries can also be used. Indeed, qualitative analysis of industry clusters using techniques perfected in the social network analysis literature (see Wasserman and Faust 1994) is promising though has not been attempted to our knowledge. Debresson (1996, pp. 167-173) provides a short discussion of techniques for identifying clusters by directed graph (see also Debresson and Hu 1999).

An example of the power of even simple descriptive network techniques can be illustrated using vehicle manufacturing template from Section 3.3.3. To completely analyze linkages among the sectors that comprise the cluster, one could examine the base correlation matrices used in the factor analysis. Although this would provide the most comprehensive picture, the detail involved in summarizing relationships among 58 sectors precludes such an approach (there are 6,728 distinct linkages in total). Another alternative is to use the indicators of dependence defined above (xij, xji, yij, yji) to identify the major relationships tying the cluster together. We used simple network graphing software to diagram key intracluster purchasing linkages in the vehicle manufacturing clusters.

Exhibit 3.10 is the result. Arrows are drawn between significant trading partners (i.e., the direction of an arrow between sectors i and j indicates that sector j purchases a significant share of its inputs from industry i, where "significant" is defined as exceeding a threshold based on the distribution of linkages between all sectors in the cluster). (SIC codes are defined according to the 1987 SIC system.) What the figure highlights is the core role of SIC 308, miscellaneous plastics products, in the U.S. vehicle manufacturing value chain. Also indicated are other sectors that serve as suppliers to multiple cluster industries.

The principal challenge of graphical network analysis techniques for identifying regional industry clusters is finding ways to interpret the revealed complexity. Software for the purpose is still limited. What is available is geared toward social network analysis, though even sociologists suffer from a lack of good software. Freeman (1999), for example, provides a recent review of molecular modeling software that can be used–imperfectly–to generate images of social networks [can be linked to at eclectic.ss.uci.edu/~lin/chem.html]. Developing better graphical techniques and associated software is a potential area of research for industry cluster analysts.

3.3.5 Surveys

In principle, one could survey regional firms to identify local and non-local trading patterns, cooperative alliances, and so on. Not surprisingly, however, survey-based methods for analyzing industry clusters are very rare. Surveys are expensive and the level of detail required in the survey instrument in order to fully explicate cross firm trading patterns and informal linkages is almost always prohibitive. There does seem to be potential for marrying limited surveying with other quantitative methods. To our knowledge, there have been few if any attempts to do this.

3.4 Summary

This chapter summarizes a range of techniques for identifying regional industry clusters. We began by characterizing micro-level cluster analyses, usually of the industrial district variety, that labor-intensively examine cooperative behavior between firms in the same or closely similar industries. We then focused most attention on methods that attempt to identify clusters from a comprehensive analysis of the regional economy. Such approaches we labeled "meso-level analyses."

Industry cluster analysis is a relatively new trade, despite its modern origins in regional science in the 1960s and 1970s. Only since the early 1990s have industry cluster applications become numerous enough to begin to discern trends in methods and approaches. Yet most cluster studies retain a highly idiosyncratic element, often dictated as they are by place-specific policy concerns, resource constraints, data limitations, and varying interpretations of the theoretical literature. Over time, a more systematic and widely-held set of definitions and analytic techniques will probably emerge. Until then, would-be industry cluster analysts should acquaint themselves with the literature. The many citations contained in this chapter are a good start.

End Notes

1. This sub-section draws upon work previously published in Bergman (1998).

2. Unsuccessful groupings of similar industries, lacking inherent interest to study sponsors, remain relatively unresearched, therefore leading to selection bias in available scholarship. Absent studies that investigate why certain firm clusters are unsuccessful, we cannot be confident of which factors are responsible for cluster success and which are simply result from clusters everywhere. The restricted study of successful clusters is due in part to Porterian-type analyses that were specifically intended to identify the factors most closely associated with "competitive clusters."

3. "As I mentioned at the beginning of this lecture, in 1895 the teenaged Miss Evans made a bedspread as a gift. The recipients and their neighbors were delighted with the gift, and over the next few years Miss Evans made a number of tufted items, discovering in 1900 a trick of locking the tufts into the backing. . .[two paragraph expansion traces origins of carpet cluster]. . .And so the little Georgia City (of Dalton) emerged as America’s carpet capital" (Krugman, 1991, pp. 60-61).

4. Isard et al. (1998, pp. 26-30) also review two related measures of specialization/localization: the coefficient of localization and the localization curve.

5. There are also policy pitfalls: "We find in the regional literature suggestions that those industries with location quotients greater than unity represent areas of strength within a region and ought, therefore, to be further developed; and, in somewhat contradictory fashion, that those industries with location quotients less than unity ought to be encouraged in order to reduce the drain of imports" (Isard, 1960, p. 494, as quoted in Higgins and Savoie, 1995, p. 156).

6. One of the "industries" in the use table is an inventory valuation adjustment (I-O code 85.0000) and three "commodities" are not directly produced by business enterprises (noncomparable imports--I-O 80.0000, used and secondhand goods--I-O 81.0002, and rest of the world adjustment to final uses--I-O 83.0001).

7. This operation invokes the "industry-based technology assumption," which assumes that the total output of a given commodity is provided by industries in fixed proportions. See Miller and Blair (1985). UA is U with noncomparable imports, secondhand goods, and rest of the world adjustment to final uses removed. Those "commodities" are not reported in the make matrix since they are not produced goods.

8. Note that all of the independent sectors are classified as secondary industries in one or more clusters.

CHAPTER FOUR

4.1 Introduction

The policy literatures and regional development conference buzzwords resonate with affirmations of how and why industrial or business clusters are relevant to understanding nearly every regional development issue. Industrial clusters are seen as permitting possibilities of linking together several strands of regional policy interest into a single framework: technology, regional productivity advantages, growing vs. declining sector balancing, etc. At the same time, academic conferences and journals flourish with newly found or refashioned evidence of clustering behavior in bewildering varieties and regional contexts. Chapter 2 describes the complexity and currency of these debates. Even after accounting for the inevitable half-life to which every emergent approach to regional development is subject, it is clear that industry cluster concepts are likely to survive in some recognizable form for a considerable time, and for good reasons.

Reasons begin, but do not end, with scholarly ambitions to reconcile and possibly integrate a wide array of existing regional development theories, using industry cluster concepts as a unifying theme. Regional scientists or other academic investigators base their cluster studies on testing and refining more robust conceptual regional development frameworks and theories of the type reviewed earlier.

At the same time, there is keen interest to adopt policies and approaches that give advantage to ‘competitive clusters,’ which is the main interest of firms or industry groups seeking competitive advantages, and is often the interest of host regions. While the competitiveness of local sectors is an important objective when facing global markets and widely traded goods, the hallmark of regional policymaking is a balancing of regional interests, mobilizing all potential resources, marshaling consensus, offsetting economic losses with gains, building on past assets to seize future opportunities, and so on. This requires knowledge of more than which clusters and cluster fragments seek advantage in a region.

Fortunately, industry cluster concepts are sufficiently well-understood among public and private sector members of regional development partnerships that possibilities of regional policy implementation are much enhanced. Common nomenclature, similar concepts, and decision implications are even more familiar to all decision sectors if framed in terms of value-chain clusters at the regional level. This is no small advantage when joint investment decisions must be coordinated, when new policies are under active consideration, or when existing approaches deserve reconsideration in light of ‘seeing regional economies whole.’ 1

Seeing regional economies whole is perhaps one of the greatest advantages permitted by use of regional value-chain clusters, which is an approach that cannot be supported with micro-based cluster studies. A regional mapping of the economy’s many interrelated sectors offers strong visual reinforcement of existing and possible connections affected by the local mix of policies and practices. These interrelationships also help understand how a region’s key sectors and clusters are linked to internal and external threats and opportunities, or how they are mutually buffered and advantaged by the region’s unique portfolio of assets.

We argue that the prevailing micro approaches to industry cluster analysis are based heavily upon the apparent needs of specific firms and industries. These needs may partially overlap the differing needs of other sectors for which regional development officials are responsible, but these interests of firms may more frequently differ markedly for many of the reasons already mentioned. Depending upon the client who commissions the cluster analysis [link to Appendix 1], there is a discernible bias toward what precisely is examined, how studies should be organized and conducted, and the range of possible uses to which the results can be applied.

This section will offer a sampling of several applications of industry clusters drawn mainly from the authors’ expansions of the approach outlined in earlier passages and subsections. The risk of immodest self-reference is taken in the interests of more certain knowledge of the details and documentation of most salient applications now available using this approach. At the same time, efforts are made here to incorporate or compare others’ work where similarity of application or sufficient detail of results permit. This is perhaps the section that is most subject to amendment and expansion in future editions. If so, it may also stimulate valuable documentation of analytic approaches underlying new applications that would further enrich the previous sections. Discussions will draw upon extensive sources for some applications in the text, while other more tightly focused issues might be treated entirely within sidebars.

The discussion will be organized in two major subsections. The analytic uses of value-chain clusters at state and regional levels will be discussed in Section 4.2. Much of this analytic potential is based on ‘sector-cluster taxonomies,’ which result from the value-chain partitioning of all pre-classified (SIC or I/0-based) sectors of an economy in ways that reveal their interdependent structure. The taxonomies are of inherent interest in terms of their methodological derivation, as explored in the previous section. This section explores their principal utility as an ‘open application architecture’ that permits various tests of propositions or concepts.

We will first illustrate this application potential for research in general, and then demonstrate specifically how spatial proximity can be shown to differ for several clusters in a given economy or how taxonomies of similar clusters might be tested in ways that reveal the effects of alternate methods of study or country of application.

Second, we will discuss and illustrate in Section 4.3 how the partitioned taxonomies might be viewed as ‘cluster templates,’ which are more stylized and accessible policy frameworks for applying value-chain clusters to development issues. Our experience indicates that cluster templates are more useful way of evaluating potential regional policy applications, than by expressing equivalent concepts with tabular data or symbolic abstractions. Cluster templates also reveal more of the implicit meaning inherent in performance measures available from regionally specific structural and dynamic micro data. Implicit meanings and development potentials drawn from template-framed data are further enhanced when visualized through graphic or mapping techniques that reflect the organizing properties embedded in the idea of templates.

4.2 Cluster Taxonomy Research and Analysis Applications

Cluster taxonomies partition detailed sectors for which widely available data can be aggregated and analyzed from a value-chain logic. Each cluster identified by the overall taxonomy consists mainly of primary sectors that trade among themselves far more than with others. Such primary trading groups could be termed a ‘clique’ by directed graph theorists who study innovation clusters, while clusters involving secondary sectors that trade at lower frequencies link them and primary sectors together in a series of interesting configurations: ‘non-standard cycles,’ ‘technological complexes,’ and ‘simple agglomerations’ (Debresson, Sirilli, et. al., 1996, p. 170-172). Whatever the configuration a particular cluster may take, the vast majority of industrial production occurs within one or more of them. Thus, most national or regional production data can be analyzed for meaningful groups of linked sectors that have been distinguished elsewhere as industrial value-chain clusters.

Partitioning a production economy into distinct groups of logically linked clusters provides an additionally useful conceptual taxonomy generally absent from SIC or other sectoral classifications. The general value of this taxonomy can be appreciated by demonstrating how industrial value-chain clusters yield quite different interpretations of the strength and complexity of North Carolina’s industrial base, particularly its motor vehicle industry (Feser and Bergman, 1999).

Traditionally measured,2 the North Carolina manufacturing economy appears dominated by textiles and tobacco, followed by smaller but significant concentrations of activity in furniture, apparel, and heavy industrial machinery. Rounding out the top five manufacturing industries in employment terms are furniture, apparel, industrial machinery, and electronic equipment. While pharmaceuticals and industrial chemicals sectors are growing rapidly, they still constitute less than five percent of total manufacturing employment. Exhibit 4.1 illustrates the composition--using standard industrial categories--of the state’s manufacturing sector in terms of value-added and employment.3

However, Exhibit 4.2 reveals a very different picture of the NC manufacturing after its detailed sectors are re-partitioned into the value-chain cluster taxonomy. For example, when transportation equipment is identified by its most general Standard Industrial Classification (37), it appears relatively inconsequential to the economy, since it accounts for less than 3 percent of manufacturing value-added and employment (see Exhibit 4.1). After assigning industries to their value-chain cluster and re-calculating the aggregate figures, our new taxonomy puts metalworking, chemicals and rubber, and vehicle manufacturing among the largest clusters in North Carolina, next to knitted goods and fabricated textile products. Only by considering the many industries that typically supply transportation equipment manufacturers does the potential significance of the vehicle industry for the state’s economy become apparent. In terms of primary cluster industries only, the vehicle manufacturing input-output chain accounted for 15 percent of total North Carolina manufacturing employment in 1994. Together, manufacturers associated with the vehicle manufacturing and knitted goods clusters account for 37 percent of statewide manufacturing employment in 1994.

As this simple comparison clearly illustrates, the availability of taxonomically rigorous value-chain clusters permits one to re-frame research and analytic approaches, including the consideration of more complicated questions by such means as pre-sorting available secondary data into variables or specifying cases for further analysis. Two additional examples discussed below illustrate the potential of the taxonomy: the first tests the utility of using similar or identical cluster taxonomies to characterize restructuring underway in economies of widely varying countries and regions, and the second examines relative spatial distributions of firms according to their taxonomic cluster membership.

4.2.1 International Comparison of Value-chain Clusters

Work presently underway within OECD’s cluster working group to apply a common value-chain cluster estimating procedure to several member countries will provide a commonly-available means of refining procedures and comparing cross-country results (Roelandt and den Hertog, 1999). Similarly, another study presently underway will apply to the newly released Austrian I/O table the value-chain clustering techniques used and reported by the authors in this monograph, in addition to analytic procedures being tested by OECD team (Bergman, Maier, and Lehner, 1998-99). In the absence of results from these more ambitious comparisons, we illustrate below an international comparison by applying a single value-chain cluster based on the U.S. taxonomy to two U.S. and Austrian regions.

In this instance, the comparison is used to illustrate restructuring dynamics for the same cluster. The comparison is somewhat complicated because the original industrial classifications of the countries differ: two sets of concordances were used to convert sectoral data originally recorded under two co-existing Austrian classifications (ÖNACE and BS68) to the U.S. SICs in which industry cluster taxonomies were originally expressed (Bergman and Lehner, 1998b). Complications also arise because the comparison includes two different time periods during which clusters were observed to have restructured.

Since the basic value-chain cluster definition was derived from I/O trading behavior within U.S. industrial value-chain clusters of North Carolina, we selected its largest region. The ‘Carolinas’ region is so-named because it borders and influences heavily its South Carolina neighbor; it is home to several small cities and the city of Charlotte, North Carolina’s largest city, which is now one of the nation’s largest financial centers, although the regional economy was historically based upon apparel, textiles and furniture production, and still reveals strong concentrations in these clusters. The other region is Upper Austria, home to many small cities and the Danube-straddling city of Linz, one of the country’s inter-war centers of heavy industry and manufacturing, although furniture, textiles, ceramics and other industry clusters are also present in the region.

Choice of these regions has two implications that require immediate comment. First is data availability. The single consistent measure of sectoral activity available in both regions is employment, although not for consistent periods. This is less a problem for our attempt to illustrate restructuring than it first appears, since the 1981-91 period for Austria captures quite well a significant period in which the country steadily shed its state-sectors, opened more of its industries to privatization and global trade, and began large cross-border investments permitted by the 1989 opening of the east. This decade included dramatic adjustments in the industrial reorganization of production and the internationalization of investment.

For North Carolina, a more recent five year period from 1989-94 was possible to select, during which the economy began its post-recession (and post-restructuring) boom that has continued to propel many of its remaining core industries to new heights. This was also a period following which a substantial share of motor vehicle production had consolidated in the mid-South along its key transport corridors shared by North Carolina, including the recent BMW investments just inside South Carolina’s border. In 1994, about 60,000 of a total 400,000 manufacturing employees were counted in the Carolinas region’s vehicle manufacturing cluster, thereby accounting for some 15 percent of regional manufacturing employment. In contrast, about 58,000 worked during 1991 in the same cluster of Upper Austria, which comprised some 11 percent of its total regional employment (nearly 508,000).

The second implication is largely technical: cluster definitions and NC data are re-partitioned SIC sectors, which must be concorded to permit the use of sectoral data organized according to the Austrian industrial classification systems. As in North America, Europeans are now harmonizing their common industrial classification system among all continental trading partners, although an earlier Austrian classification system applies to information collected for these particular dates.

As a consequence, a considerable amount of cross-coding from industrial concordances was necessary, and this resulted in a slightly lower overall resolution of industrial detail for our comparisons, simply because certain sectors lack one-to-one correspondence in both classifications. The full task of concordance revision undertaken here is onerous and will be unnecessary for future data classified according to ÖNACE, so only one familiar industrial value-chain cluster was selected with which to illustrate our templates: motor vehicles.

Upper Austria lost about 1 percent of its vehicle manufacturing cluster employment in the full 1981-91 decade, while the Carolinas region cluster gained at about 1 percent over the shorter, more recent half-decade period. While the Carolinas regional cluster expanded and Upper Austria contracted in roughly similar proportions, coefficients of sectoral variation within the motor vehicle clusters of both regions increased by some 10 percent, leaving the Carolinas region with slightly more sectoral variation (1.62 in ’94 vs. 1.34 in ’91).4

Cluster graphics are organized in this illustration by declining size of sectoral employment (see Exhibits 4.3 and 4.4). The largest sector of the Carolinas’ motor vehicle cluster is 3714 (motor vehicle parts and accessories), while the largest sector in Upper Austria is sector 3711 (motor vehicles and car bodies). As both regional templates indicate, the remaining sectors drop off dramatically in size and number, and large portions of the total cluster graphic of both regions are totally uninhabited. Sectoral representation of Upper Austria’s cluster might be somewhat affected by concordance artifacts that arose when translating two national industrial classification schemes, but it is far likelier that our depiction is generally accurate in both regions, particularly their depiction of heavy concentrations in very few sectors, a minor presence in several, and absence of many others.

Upper Austria lost significant employment shares in the vehicle manufacturing sector (SIC 3711, 3716) and engine components (carburetors, pistons, rings, valves: SIC3952), whereas its vehicle parts and accessories (3714) production gained employment.5 In the Carolinas Region, the sectors most closely tied to this cluster grew strongest from 1989 to 1994, including the secondary sectors producing technology and equipment (welding and soldering equipment, machine tools, and metal cutting: SICs 3548, 3541) used in vehicle and parts production.

The clusters differ quite obviously in their composition, and their host regions differ markedly in overall economic structure as well. But the regional templates yield even stronger hints about the formation processes taking place within each region. The Carolinas region template indicates that its vehicle manufacturing cluster is expanding in nearly all its 1989 sectors, with more absolute growth in the largest. Its vehicle manufacturing cluster seems to have reached an optimal growth composition in ’89 and expanded in the following five years along, perhaps, an increasing returns trajectory.

The template suggests a quite different growth process for Upper Austria: sectors described above expanded dramatically, while others, even very large sectors, contracted equally dramatically. Both interpretations offered earlier imply considerable restructuring underway in Upper Austria’s motor vehicle cluster over the ten year period. It is possible that Upper Austria’s remaining cluster segments may repeat in the next decade some version of the story told from by the Carolinas’ changing regional template, particularly if the remaining sectors are well niched into its regional economy in ways that permit them to cross-trade competitively with EU and other regions to the east, yet produce efficiently in Upper Austria.

This illustration demonstrates the potential for applying commonly defined cluster taxonomies to very different regions as a means of comparing their underlying processes. This implies that an OECD or EU-based set of cluster taxonomies might become a very valuable analytic tool, particularly when applied to smaller, open economies that host only some of all potential sectors associated with a full value-chain cluster.

4.2.2 Spatial Bunching in Cluster vs. All other Firms

Can cluster member firms be convincingly shown to ‘bunch’ together more (or less) tightly than they bunch together with average firms (Bergman and Feser, 1999) Clusters based on value- or supplier-chain criteria may consist of firms whose trading behavior is accompanied by co-locational tendencies, perhaps due to JIT transactions or to capture technological and other spillovers present in the same region. This is an interesting question, since it asks whether firms that are ‘close’ in their value-chains are also located physically close in space, a question that requires evidence of both kinds of proximity and a method by which to make the comparison. In short, how closely located are the value-chain cluster firms?

Results from industrial value-chain cluster analyses are the base that permit a ‘spatial-economic test’ of this question (Feser and Sweeney, 1999). The test involves a use a case-control design to test whether certain types of manufacturing firms (i.e., cluster firms) are more spatially concentrated than might be expected, given the general geographic pattern of all manufacturing firms in the state. All plants associated with a given industry cluster are used as cases and a matched sample of all other manufacturing firms is drawn as comparison case. The difference in concentration between the two is measured by using a D statistic derived from two K functions (a standard statistical geography technique; see Feser and Sweeney, 1999, for details), thereby providing evidence of spatial concentration or dispersion at different spatial scales for the firms in the economic cluster. A positive (negative) value of D outside defined confidence bands implies statistically significant spatial clustering (dispersion).

Findings for three regional clusters with distinct degrees and types of spatial tightness are particularly illustrative: vehicle manufacturing, printing and publishing, and wood products (see Exhibits 4.5, 4.6 and 4.7). In the case of the vehicle manufacturing cluster, its firms are more tightly concentrated at all spatial scales shown, although spatial clustering is most significant at scales of two to six kilometer radius. JIT practices known to characterize this cluster’s supply chain imply greater than average spatial tightness over a wide range of distances. Very gradual convergence toward average spatial concentration over ever longer distances may result from the many different sectors that comprise this cluster, ranging from highly urban, skill-intensive sectors to fairly rural, standardized production sites, which are spread widely along connecting interstate highways and major transportation corridors.

Similar in pattern, yet still unique, is the printing and publishing cluster. It also begins higher than average spatial concentration, but relative concentration peaks earlier at about a 12km radius, and converges rapidly toward average concentration after 50km. This is clearly a highly urban industry, where shorter radial distances are the rule, often with face-to-face contacts necessitated by frequent design or delivery requirements.

Wholly distinct is the wood products cluster. Its pattern shows that cluster members are far more dispersed relative to each other than they are to non-cluster firms. From 7km onwards, wood cluster firms become increasingly more dispersed (relative to the average). This is one practical consequence of wood products being a natural resource-based industry cluster, where proximity to high-weight, moderate-value inputs automatically disperses its firms to remote places of resource availability.6

From these findings, it is evident that firms in some clusters are indeed far more closely co-located with each other than with other non-cluster firms. This suggests that cluster externalities and advantages exceed those available to all other firms that enjoy available urbanization externalities. Greater spatial tightness also implies stronger face-to-face possibilities, and the diffusion of technology, knowledge, and general learning that is possible through such spatially-permitted contacts. Findings for the motor vehicle and printing and publishing clusters confirm industrial folklore about the role of localized suppliers and machinery vendors in many industries, particularly the needle trades industries, but it also gives support for certain spatially-centered localization economies based on commonly provided services to firms in such clusters.

4.3 Regional Policy Development Frameworks

As argued above, value-chain cluster definitions and their detailed taxonomies permit researchers to incorporate available data and related concepts in ways that reveal more about broader issues of regional development and aid the empirical research that helps test and build regional development theory. The same quality is valuable for policy applications in specific regions because clusters logically organize large bodies of regional data in more concise and analytically useful categories. To help organize and present such data in a more convenient and less formal framework, we extend the idea of ‘cluster templates.’

These are simply another way of ‘seeing’ the same cluster concept, but the emphasis shifts to the idea that the sectors of a particular region can be seen to fit into two or more representative cluster templates and secondary data can then be used to characterize key components of the overall regional economy.7 Consistent with this approach is the further introduction of graphic and mapped visualizations of templates, which are illustrated for several regions.

The utility of these approaches is based directly on the types of policy questions that can be addressed or policy answers that have been sought. It is therefore appropriate to begin our discussion of applications with the main policy question to which our value-chain approach was first applied.

4.3.1 Modernization of North Carolina Industrial Base

In 1996, NCACTS wanted to identify the principal channels through which modern production technologies tended to spread and diffuse in North Carolina. The agency was particularly concerned with a specific policy problem: how to diffuse advanced production technologies efficiently among businesses in a manufacturing economy traditionally dominated by a least-cost competitive ethos . North Carolina’s rapid growth from the 1960s through the 1980s was fueled initially by the re-location of branch plant facilities from high wage, unionized locations in the industrial mid-western and northeastern parts of the country. Although the state has gradually established a solid high technology base (principally centered upon Research Triangle Park) and banking presence (in the Charlotte region), its economy remained disproportionately specialized in traditional sectors under unrelenting pressure from low-cost, overseas producers (e.g., textiles and apparel). In this environment, encouraging producers to invest in, adopt and utilize best practice production technologies can be an exceptional challenge.

In an earlier study of technology adoption practices among producers in the state’s nominal automotive supply chain, the authors found that smaller and often more rural producers tended to be less aggressive in adopting new manufacturing techniques (Bergman et al., 1995; Bergman, Feser and Kaufmann, forthcoming). Reasons cited included lack of information about advanced technologies and inadequate access to sources of capital that do not dilute control over the firm. More passive or traditionally-oriented firms appeared satisfied with the existing market, and not interested in pursuing an aggressive growth strategy through investment in technologies that could open new and protect old markets, even though such complacency is surely fatal in certain traded industries.

The authors also found that producers presently in the NC vehicle supply chain tend to adopt and use technologies at a significantly higher rate. Consistent with other research, study evidence suggested this results partly because final market vehicle assemblers were essentially ‘forcing’ adoption of new methods by their suppliers as well as serving as a source of information about best practice techniques. Also important were increasingly strict international certification requirements (e.g., ISO 9000) that maintain buyer confidence.

There is considerable evidence of powerful diffusion effects that spread competitive production technologies through the supplier or value-chains, a well-known view that continues to receive considerable support from the growing research on buyer-supplier relations. Indeed, Roelandt and den Hertog (1999) make an even broader case that value-chain clusters are actually industry or region-specific ensembles of the larger ‘national innovation system.’ Based on either viewpoint, industrial modernization policies [www.cherry.gatech.edu/mod99/index.htm] might be coherently designed and implemented for the supply-linked firms of certain clusters considered important to state and regional economies.

4.3.2 Technology Composition of Regional Clusters

Regional development authorities responding to statewide policy initiatives of the type pursued by NCACTS would surely need to know how its clusters might be affected. At the same time, a region has many related policy considerations that might depend on the technological level at which its key clusters presumably function, e.g., education programs, public services, or infrastructure.8 Therefore, it is important to develop policies with at least a primitive understanding of its situation and that of other regions.

Gaining this understanding can be illustrated below by comparing ‘high technology’ intensity indices that were calculated for industrial value-chain clusters of both the U.S. and North Carolina production economies (Bergman and Feser, 1999). High technology shares of output and employment of industrial value-chain clusters reported below for both the U.S. as the ‘reference area’ and North Carolina are simply calculations possible for any region. Output in several U.S. industrial value-chain clusters is predominantly in sectors that are characterized as high tech at some level.

When their primary industries alone are considered (second column, Exhibit 4.8 ), the share of output in sectors classified as high technology meets or exceeds 80 percent in the petroleum, aerospace, chemicals and rubber, electronics and computers, and aluminum clusters. Several other clusters range from low to moderate shares of high tech output: vehicle manufacturing (63 percent), platemaking and typesetting (35 percent), metalworking (36 percent), and fabricated textile products (23 percent). Fourteen of twenty-three clusters, including the five food products clusters, knitted goods, nonferrous metals, wood products, printing and publishing, tobacco, cement and brick, brake products, and earthenware products produce very little or no high technology output.

A comparison of cluster output in North Carolina versus the U.S. suggests some under- and over-representation of high technology sectors in the state’s industrial value-chain clusters. The ratio of high-to-standard technology production in the North Carolina chemicals, electronics and computers, and aluminum clusters equals or exceeds (in the case of aluminum), the ratio for the U.S. as a whole. Confirmation of high-tech intensities is impossible without a detailed look at the component sectors in each cluster, so one should interpret aggregate profile indicators as tentative evidence that at least some of the critical high technology links are present in the state’s extended buyer-supplier chains, including the strong probability that high-tech links are more concentrated in certain regions than in others.

North Carolina‘s most traditional manufacturing base operates at generally lower levels of technology, although specific industrial value-chain clusters or product chains contain very high concentrations of high technology sectors. The percentage of high technology production in North Carolina’s metalworking cluster, for example, well exceeds its U.S. benchmark. As shown below, the majority of statewide activity in this cluster is in the higher tech, higher wage industrial machinery sectors, rather than basic metals production and fabrication.

Conversely, the share of high tech production in the comparatively very small NC aerospace and petroleum clusters falls well below U.S. averages; the few establishments in these clusters are producing largely standard rather than high technology output. Other value-chain clusters that contain moderate shares of high tech activity at the national level (vehicle manufacturing, fabricated textiles, and platemaking and typesetting clusters) consist in NC of sectors that contain significantly lower relative shares of high tech production. To the degree that supply chain diffusion strongly influences technology adoption, lower overall levels of technology in sectors now linking these chains could limit technology upgrading possibilities, particularly among its neediest cluster members.

4.3.3 Visualizing Intercluster Trade and Technology Flow Networks

Tentative conclusions reached above about the implications of limited technology flows among cluster members may seem obvious, but for many development officials these conceptual insights would be far stronger if reinforced through visualization methods.9 If actions are expected of those for whom unfamiliar data and obscure inferential methods are misunderstood or mistrusted, then it is vitally important to demonstrate clearly the inherent consequences of value-chain trade and clusters. To do so, we add visual value to our data templates through the use of network plotting applications.

We have shown earlier that one can identify with which clusters a listed sector is likely to be linked as its first, second and third most important trading partners. Purely secondary sectors often have weaker internal links with any single cluster when they buy or sell interindustry goods to several sectors that are core members of different clusters. While it is these primary sectors that essentially define value-chain clusters because of their intense internal trading patterns, this conversely implies that it is the so-called secondary sectors’ multiple points of cluster contact that serve as the main technology transmission channels between the clusters of any region. Multiple points of contact and linkage are very difficult to explain or grasp unless visualized properly, the possibility of which is illustrated below.

To do so, we will represent these linking relations in visualized template form (Krackplot™ or similar) for two North Carolina Regions (Exhibits 4.9 and 4.10). The template illustrated here identifies which secondary sectors trade with two or more of a region’s clusters. A graphically depicted set of trade linkages is represented by specific sectors (ovals) that simultaneously trade across two or more clusters (boxes).

The information value of any graphic is greatest when regionally relevant data are embedded within the visual templates. The linking elements might be scaled by an essential channel characteristic, which is the case shown here. Each sector is known to have some characteristic technology, or what we might call technology density. This density is indicated on the graph by different width lines for secondary sectors that trade between various regional clusters. Sectors with the greatest density have higher probabilities of transmitting technology flows between their linked clusters. This permits regional development officials to identify important key sectors, based on their role as potential technology channels.

To illustrate these ideas, we compare here the technological linking of three clusters in two North Carolina Regions: Research Triangle Park Region and Southeast Region. The relative technological ‘carrying capacity’ of sectoral linkages, as indicated by their widths, shows that the RTP Region has more and technologically denser sectors linking its Electronics, Aerospace, and Metalwork clusters than do Southeast Region linking sectors. Similar differences are also visible for other combinations of clusters and sectors. Note also that secondary linking sectors present in one are not necessarily in the other region, and many are absent in both regions.

Regional value-chain clusters are the locally adapted sectoral ensembles that presumably enjoy shared production advantages. The links that connect secondary sectors and their multiple-cluster memberships permit us to grasp another dimension of a region’s coherence. The high potential for intraregional trade between them establishes a provisional understanding of the full ensemble, and offers a framework for seeing a regional economy as a coherently interconnected whole.

4.3.4 Visualizing Regional Cluster Portfolios

The coherent wholeness of a region revealed by graphic representation of linked secondary sectors necessarily omits details concerning the composition of primary sectors that define a region’s main clusters, representing them only by titled boxes. These characteristics of value-chain clusters can be expressed by other software-assisted visualizations to help reinforce understanding of their composition. A consistent graphical framework also provides a common visual vocabulary with which to discuss their meaning and consequence. As we shall demonstrate, there is much room to better understand how the underlying concepts of industrial value-chain cluster templates shed new light on a region’s constituent sectors, using simple spreadsheet graphics.

Direct comparisons of relative specialization in the particular sectors that comprise each region’s clusters are illustrated in Exhibits 4.11 and 4.12. Selected value-chain cluster templates are illustrated for the Research Triangle and Southeast regions. A template is configured for all possible sectors in the value-added cluster, starting with its innermost core sectors positioned at 12 o'clock, all others spiraling clockwise in declining order of ‘trading tightness,’ as measured by correlations with the overall cluster . In other words, we can visualize first the overall distribution of sectors, while those present in the region are indicated by employment vectors measured radially from the center.

Regional employment data are embedded for all sectors along their appropriate radians. A cursory glance at the templates shows each region has only a subset of sectors for a given cluster, and they are typically quite different sectors. The selective presence of sectors reveals how internally specialized the value-chain clusters in regional economies within the same state can become and why regionally-specific cluster policies are essential.

A ‘bulls-eye’ size and share graphic is also located next to each template to gauge the relative importance of the cluster to its region’s economy. The full circle represents 100 percent of each region’s total manufacturing (the Research Triangle’s larger full circle indicates proportionately more total manufacturing employment in all its clusters). The white inner circle represents a cluster’s share of total regional cluster employment (note the Triangle’s relatively much larger electronics cluster), while the black inner circle represents the most tightly trading sectors within that cluster. The white halo represents the relative size of the main linking sectors that trade with more than one cluster, as plotted in Section 4.3.2 above. The Research Triangle region’s relatively small textile cluster consists of higher proportions of core sectors that trade among themselves than is true for the Southeast region’s relatively larger textile cluster. However, the larger Southeast textile cluster also includes higher shares (white halo) of cross-trading sectors, thereby indicating that it is more strongly inter-linked with other clusters in its region or elsewhere.

Other types of regionally available data can also be embedded in value-chain templates, although the data shown here illustrate the potential clearly. Considered as a whole, these visualized templates compress a considerable amount of regionally relevant information into a very tightly organized and easily compared image that helps a policy maker assess important key features and detect proportionate relationships that may escape attention when embedded deep in some data table.

4.3.5 Policies for Small and Independent Firm in Clusters

The small business revolution in regional development thinking shifted attention away from large, exogenous investment development projects pursued under growth center strategies (link to Chapter 2) and more toward indigenous firm incubation and support policies that began in the 1980s. Small and medium enterprise strategies also focused attention on the adequacy of a service industry base seen as necessary to supply key enterprise and producer services. Early industry cluster advocates have repeatedly stressed that the very essence of clusters revolves around the flexibility and shared resources of such firms, usually by drawing attention to the success of Italianate clusters. More recent formulations of industry clusters based on value-chains include firms of all sizes and ownership, although no clear typology has been accepted by which to differentiate all the various concepts (Harrison, 1992; Markusen, 1996; Polenske, 1997).

What is clear is that regional development officials often give due consideration to the firm size and ownership distribution when formulating various policies. This may be even more true if such policies are intended to help support and promote the success of a region’s portfolio of value-chain clusters. The importance of these considerations can be illustrated clearly in the case of industrial modernization policies.

Size and branch plant status have consistently proven key indicators of the level and rate of advanced process technology adoption among manufacturing plants in scientific studies. Numerous surveys have found that large branch plants in nearly every major manufacturing industry adopt new technologies faster and to a greater degree than their smaller independent counterparts (Bergman, Feser and Scharer 1995).10 Smaller producers have fewer of the necessary resources, both financial and human, to effectively integrate complicated new technologies into their production regimes.

Alternatively, the owners of some smaller businesses show reluctance to invest in technology upgrading if the financing of such investment requires dilution of their equity in and control of the firm. Identifying those sectors with a predominance of smaller manufacturers, particularly those at the smallest end of the size scale, is thus one preliminary means of narrowing down areas of potential demand and need for competitiveness initiatives. Size, in effect, serves as a very rough proxy for level of modernization, and indirectly, of a need for some form of technology assistance.

Exhibit 4.13 lists the shares of both small and single (versus branch plant) establishments in each cluster. Among the largest North Carolina clusters, the wood products, printing and publishing, and metalworking clusters are each made up predominantly of very small firms and establishments. In each case, close to 80 percent of businesses employ fewer than 50 workers. With the average share of branch plants at just 12 percent, these clusters are also largely composed of single-establishment enterprises. The clusters with the lowest shares of small plants are knitted goods (41 percent), packaged foods (52 percent), fabricated textile products (55 percent), chemicals and rubber (55 percent), and vehicle manufacturing (58 percent). With the exception of vehicle manufacturing, close to one-third of the establishments in each of these clusters are branch locations of multi-location firms.

4.3.6 Cluster Targeting and Tradeoff Policies

Industrial clusters can be used to detect the presence of a critical mass of value-chain sectoral activity that might benefit from the strategic application of targeted development policies. Scott and Bergman (1996), for example, examined the prospects for developing a ground transportation manufacturing complex in southern California, after mapping important input-output linkages of key sectors. North Carolina appears similarly poised to take advantage of the continuing southward shift in the geographic center of vehicle production in the United States (Klier 1994, 1999). At present, there are no automotive assembly plants in the state, although trucks, school buses and specialty transportation equipment are now produced. Moreover, the recent location of production facilities of several major automakers to the south and west, including BMW in South Carolina, Saturn in Tennessee, and Mercedes in Alabama constitute potential markets for suppliers based in North Carolina. Consistent with these trends, the vehicle manufacturing cluster within the state appears to be shifting westward. The international comparison of motor vehicle cluster restructuring discussed in Section 4.2.1 showed increasing concentration of this cluster in the Carolinas Region, which is along the southwest border of the state.

Though it cannot be known from value-chain clusters alone which local firms in the vehicle manufacturing cluster produce goods related directly to automaking or the production of trucks and busses, there is nevertheless strong and visible potential in the state for the further development of its vehicle manufacturing chain, including the possible recruitment of a major final assembly plant. But in establishing policies that target one cluster, there is always the question of how to consider other potential clusters.

At the other extreme, consider the knitted goods cluster, which is this region’s largest (19 percent) and perhaps most threatened cluster (wood products (18 percent), vehicle manufacturing (15 percent), and fabricated textile product (11 percent) clusters also account for more than 10 percent of total primary cluster employment). All are significant components of the regional economy whose combined needs require a portfolio of suitable policies. Larger regions or cities typically adopt policies suited to a continually changing mix of industries distributed across segments of several value-chain clusters, rather than to only one of 23 total possibilities. The four larger clusters mentioned above are joined at some threshold presence by electronics and computers (4 percent), printing and publishing (3 percent), chemicals and rubber (2 percent), plus slight traces of ten other clusters.

Any of these may contain the seed of quite dramatic economic transformations in future decades, or the linking agents that connect important common elements of larger clusters now active in a region. Accordingly, regional development officials and company managers who are committed to making interdependent investment decisions concerning the full regional economy may proceed with greater confidence when informed by readily envisioned clusterings and potential reconfigurations of a region’s many firms and industries.

To appreciate the possibility of improved regional strategies that account for more than one cluster at a time, consider the following evidence for the Western Economic Development Partnership Region. First, the spatial concentration of the knitted goods cluster, as revealed by common GIS mapping procedures in Exhibit 4.14 clearly distinguishes its spatial intensity, major interstate highway systems, and proximity to nearby cluster concentrations of other regions. The spatial pattern of all firms in this and other significant clusters permits better alignment with existing or planned features of infrastructure in the region and neighboring regions. Transportation improvements, essential public utilities, key public service areas, education and training facilities, and similar development policies may be fully reconsidered in light of each cluster’s pattern.

Joint spatial patterns of a region’s principal clusters reveal possibilities for designing comprehensive policies, particularly since the mountainous landscape and widely dispersed towns and production centers of these regions require very careful infrastructure planning. The same logic also applies at higher levels of policy responsibility, e.g. statewide policies, where spatially networked clusters of several regions are involved. For example, North Carolina’s declared interest in promoting the vehicle manufacturing cluster and the present strength of this cluster in neighboring regions (Exhibit 4.15) readily suggests a strategic restructuring of the WEDP region away from its primary dependence on the knitted goods cluster.

Each region has its own cluster targets and tradeoffs to consider in formulating reasonable development policies. It is therefore essential that the kind of information and analysis described here be used to supplement the usual sources and frameworks available to policy officials.

Because of this need, these and many other regionally specific data were drawn upon to prepare a series of separate policy framework documents (50+/- pages) to guide ongoing policy discussions in each of North Carolina’s seven economic development partnerships. It is important to note that these reports are based totally upon the analysis and presentation of data that adopt value-chain clusters as the organizing framework.

Notes

1. These broader concerns were first advanced elsewhere (Bergman 1998), some points of which are emphasized and sharpened in this section. Author self-sources are repeatedly drawn upon in this section to provide a range of applications, each of which will cited at relevant discussion points.

2. Traditional sectors at have been renamed ‘industry clusters’ in many recent (mis)applications of the broader concept for states and regions. Some of this renaming is merely the marketing residue of quick and dirty sectoral studies, even though these efforts often result from a genuinepolicy interest in understanding and harnessing the forces that now sweep through regional economies. http://www.credc.com.au/cluster.htm, http://www.users.uswest.net/%7Egaryboydston/index.html, http://www.commerce.ca.gov/california/economy/neweconomy/addendum2/index.html

3. Value-added is estimated by applying industry specific wage/output ratios derived from input-output tables to wage data from ES202 files. Employment data also taken from ES202 files.

4. Calculated net of large wood processing sectors that trade more heavily with other key clusters in both regions. As the wood processing sector is a major component of both regional economies, the result would be biased in favor of more variation, leaving the overall picture, however, the same.

5. These shifts could not have occurred because of simple classification artifacts, as the classification remained stable in both years. A decade-long restructuring away from larger or state-owned firms of a dominant classification into smaller firms of different but more precise classification in 1991 is more likely responsible, even though it is impossible to know if this happened or whether such a case would imply a true shift in the types of goods produced.

6. See discussion of policy issues facing firms of this cluster for California. http://www.commerce.ca.gov/california/economy/neweconomy/addendum2/ad250.html

7. Data is generally available at highest level within which supply chains deliver interindustry goods unimpeded, such as a nation or customs union (NAFTA, EU), that can be aggregatively transformed into useful attributes of industrial value-chain clusters. As demonstrated elsewhere, the most obvious imputed detail is output data (value of output, value-added, etc). However, basic factor input data (labor, capital or resources) and a wide range of supplemental data collected for constituent sectors only at high levels of aggregation (e.g., technology levels, productivity, production residuals, energy consumption, etc.) permit other ways to characterize and compare economies. Weighted proportions of sectoral presence in each cluster can be employed to calculate a central tendency measure of its constituent sectors. There are many opportunities afforded by various secondary data sets that can be incorporated to characterize value-chain clusters in greater detail or to capture more finely-calibrated distinctions, as earlier analytic discussions and schematics have demonstrated. Certain possibilities are illustrated in the text by characterizing a cluster in terms of the average technology or types of establishments that comprise its constituent sectors, as revealed by their size and organizational structure.

8. Arizona, for example, uses high-technology cluster needs as a useful rationale for organizing and focusing various education programs. http://aztec.asu.edu/k12/htic/

9. A shortcoming common to many analytically-based studies is the difficulty of demonstrating how to draw inferences or detect meaningful relationships from the analyses that inform options and provide good overall policy perspectives. This is part of the challenge of ‘seeing local economies whole.’ Wholeness will not matter much if it cannot beheld firmly within the mind’s-eye as alternatives are posed and considered. The very ideas embedded in value-chain clusters and their inherent internal and external linkages drift easily from view in the welter of tables, graphs and statistics supporting analytically sound but opaque concepts that such results often contain. Edward Tufte built a wide reputation by publishing his own series of books about the utter importance and utility of visual representation http://www.amazon.com/exec/obidos/subst/features/t/tufte/tufte-edward-interview.html. Tufte has also noted the equally important point that analysts often don’t fully understand their evidence without such assistance.

10. A value-chain cluster can serve as the sampling frame for in-depth survey or interview methods of micro investigation. The sampling-frame approach permitted by value-chain cluster definitions has been used with good success by authors in micro studies of one cluster in North Carolina (Bergman, Feser, and Scharer, 1995) and of four clusters in four Austrian regions (Bergman and Lehner, 1998a). http://iuwhp1.wu-wien.ac.at/iir-clusters/