Local manufacturing systems and global economy: are they compatible?

Journal of Operations Management 21 (2003) 151–171

Local manufacturing systems and global economy: are they compatible? The case of the Italian eyewear district Guido Nassimbeni∗ Dipartimento di Ingegeneria Elettrica, Gestionale e Meccanica, Università Degli Studi di Udine, Viale delle Scienze 208, 33100 Udine, Italy Received 12 October 2001; accepted 21 May 2002

Abstract In the era of global economy, are local manufacturing systems still playing a role or are they destined to fade away? No unequivocal answer to this question can be found in the literature. Some think that globalization exposes all the limits of local industrial systems. Others consider the contraposition between global and local to be only apparent. There could be, however, a clear relation between the firm’s competitiveness on the international market and its territorial roots. The present paper focuses on one of the most important local systems in Italy, i.e. the eyewear district. Based on empirical research, the study analyzes the evolution of the eyewear industry, the strategies used by the larger producers to face the current competitive challenges and the consequences of such actions on the local supply system. It is shown that the district has lost most of its traditional features and inter-organizational relations have significantly changed. Opportunities arising from new technologies and global economy seem to favor better equipped, large-sized enterprises rather than small suppliers, or in any case, single units or groups rather than the system. Finally, some suggestions are provided for entrepreneurs or managers operating in industrial districts, concerning vertical integration and sourcing choices (main-contracting firms), and technological and managerial levers that could better qualify the supply offer (sub-contracting firms). © 2002 Elsevier Science B.V. All rights reserved. Keywords: Operation strategy; Supplier management; International issue; Case study research

1. Introduction The debate regarding local manufacturing systems, industrial districts and networks of small firms was topical during the 1980s, when the concept of “flexible specialization” (FS) was developed. Piore and Sabel (1984) made the first important contribution to these studies. They argue that the history of industrialization has kept one major alternative to the system of ∗ Tel.: +39-432-558333; fax: +39-432-558251. E-mail address: [email protected] (G. Nassimbeni).

mass production open, namely clusters of small firms, connected by horizontal and vertical competitive and co-operative relationships that can be achieve through strong inter-firm division of labor, greater collective efficiency than larger-scale Fordist enterprises. Piore and Sabel has taken Japan, West Germany and Italy as examples, where flexible specialized firms are often clustered together. In these systems, producers increase their own capabilities by specializing, while having close, but not exclusive, relations with other specialists. Network systems flourish in regional agglomerations where repeated interaction fosters shared

0272-6963/02/$ – see front matter © 2002 Elsevier Science B.V. All rights reserved. PII: S 0 2 7 2 - 6 9 6 3 ( 0 2 ) 0 0 0 9 0 - 6

152

G. Nassimbeni / Journal of Operations Management 21 (2003) 151–171

identities and mutual trust while at the same time intensifying competitive rivalries. It appears that the flexible specialization model is one of the consequences of overtaking Fordism– Taylorism. However, for some time the future of local systems and their real effectiveness as an alternative to Fordism has been questioned. In the era of global economy, are local systems still playing a role or are they destined to fade away? The issue is crucial, especially for a country like Italy where, local systems have been and still are a fundamental part of the industrial system. However, no unequivocal answer to this question can be found in the literature. Some scholars believe that globalization reveals all the limits of local industrial systems. These limits could be due to the district enterprise, for example, because of general under-sizing, it is difficult for it to obtain adequate technology and financial aid, consequently there is a delay in the introduction of new technologies and other radical innovations, and the sales marketing systems are weak. Other limits are vice versa attributable to the network, the agglomeration of these units, whose external economies seem to be progressively weakening. Many local systems blame the excessive dispersion of added value and the difficulty in developing combined and coherent strategic moves. Globalization (of the markets, competition, work distribution, and informative networks) urges district enterprises to pass beyond the local borders, recommending collaboration over an extended geographical area and modifying the district values and rules. Other researchers consider the contraposition between global and local only apparent. There could be a clear relation between the firm’s competitiveness on the international market and its territorial roots, since the local system constitutes an irreplaceable sourcing area for new competencies, creativity, and variety. Becattini (1999), one of the most influential Italian scholars, writes about “re-blooming” of industrial districts, whose number and profit rate seem to have grown in the last few decades. A series of studies and empirical evidence supports this perspective (Porter, 1990, 1998; Signorini, 1994; Fabiani et al., 1998). The future of industrial districts, or more generally, the aspects and problematics of local manufacturing systems, though central to the subject, have rarely been discussed by Operations Management (OM)

researchers. Industrial districts are specific manufacturing systems that are gradually being transformed by current changes in competitive environments. In this paper, first the literature concerning the evolution and prospects of local manufacturing systems is reviewed. Next, according to the OM literature on manufacturing strategies and typologies, the industrial districts are characterized. The challenges posed by globalization to the assumptions underlying the flexible specialization model are pointed out. Then the study focuses on one of the most important local systems in Italy, i.e. the eyewear district. It produces more than 80% of the eyewear, which is one of the leading sectors in the Italian “fashion” system with an output whose value is approximately 1.75 billion. This district is passing through a phase of transformation. In the past, the district formula enabled Italian producers to achieve cost advantages, and mix and volume flexibility. Today it is difficult for such a formula to satisfy the global market which requires the producers to shorten product development time, produce new models faster and more frequently, and better differentiate the product (especially quality and design) from that of international competitors, who are stronger in terms of innovation (Japan) and costs (south-east Asia). The local system is often inadequate to such an extent that various producers are internalizing some processes or rationalizing the supplier base, so as to achieve greater control of quality, times and costs. As a result, relations in the district, and kind and type of local sub-contracting firms involvement are gradually changing. Based on an empirical research, the evolution of the eyewear industry and the strategic response of large producers to present-day changes are analyzed. Particular attention is given to supply policies and their effects on the local sub-contracting system. In addition, the characteristics, technological and managerial profiles of the local supply system are examined to determine weaknesses. Finally, entrepreneurs or managers operating in the industrial districts are given some indications about vertical integration and sourcing choices (maincontracting firms), and technological and managerial tools and practices (sub-contracting firms). The paper is arranged as follows: Section 2 reviews some of the main issues arising from the debate on local manufacturing systems and their evolution.

G. Nassimbeni / Journal of Operations Management 21 (2003) 151–171

A specific subsection is devoted entirely to the key OM aspects that characterize local manufacturing systems. Then the research objectives are introduced (Section 4), followed by the methodology (Section 5) and the results of the empirical research on the main-contractors’ (Section 6) and sub-contractors’ (Section 7) sides. The last two sections contain an interpretative model of the temporal evolution of the district, hypotheses on future developments, the implications for entrepreneurs and managers, and concluding remarks.

2. Local manufacturing system: some debatable issues The first and most important work in this field was carried out by Marshall (1920) on English industrial districts at the end of the nineteenth century. He describes districts as a territorial concentration of numerous (small) enterprises characterized by high vertical and horizontal specialization, a dense network of social and economic, competitive and co-operative relations. The exchange of commercial and technical information including any innovations is prompt but informal. Marshall ascribes their competitive advantage to the presence of external economies, i.e. economies linked to the environment in which the enterprise operates. These include faster dissemination of new ideas, experience and know how thanks to geographical propinquity, cultural homogeneity, common manufacturing traditions, reduced cost of transport (and of transactions in general), and the easier access to complementary services or capabilities. This seminal work sparked a series of studies on districts. Local clusters of numerous, mostly small enterprises which alternately compete and co-operate with one another and specialize in particular aspects and phases of production are recognized and analyzed in various contexts (Brusco, 1982; Piore and Sabel, 1984; Scott, 1988; Pyke et al., 1990; Saxenian, 1991; Harrison, 1994, to mention just a few). Table 1 summarizes the main characteristics of industrial districts. The debate on industrial districts reached its height during the 1980s, when the consequences of post-Fordism came under fire. The most influential work is probably Piore and Sabel’s (1984) ‘The Second Industrial Divide’. They carried out a meta-

153

Table 1 Main features of industrial districts High proportion of small and very small firms Clustering of firms in a geographical location Firms engaged at various stages of production—intense specialization Dense networks of a social and economic nature Blend of competition and co-operation between firms Rapid and mainly informal diffusion of information, new ideas, experiences and know how Adaptability and flexibility

historical analysis of the late nineteenth and twentieth century capitalist mode of production. The fading epoch of mass production, characterized by the manufacture of standardized products using specialized machinery and predominantly unskilled labor, has given way to the small firm economy. A strong inter-firm division of labor among clusters of small firms, connected by horizontal and vertical competitive and co-operative relationships, can lead to greater collective efficiency than that of larger-scaled Fordist enterprises. They cite industrial districts in central and northern Italy, where co-operation is promoted “by establishing an ethos of interdependence among producers in the same market” (Piore and Sabel, 1984, p. 272), while competition is encouraged but controlled by mechanisms of social cohesion within the local community. Piore and Sabel’s view has been further developed by other researchers. Hirst and Zeitlin (1991), for example, distinguish between two types of industrial districts. The first is composed of “geographically-localized networks of small firms that sub-contract between one another and share services that are beyond the economy or productive capacity of the single firm”. The second arises from the break up of firms into decentralized autonomous productive units, which continue to benefit from the research, marketing and finances of other divisions of the parent company. According to Best (1990) industrial districts can continuously adjust and improve their economic performances through quasi-spontaneous co-ordination mechanisms, when entrepreneurial attitudes facilitate innovation. All these studies emphasize the dynamism of small firms and their key role in modern economy. Many cite the Italian experience, particularly that of the so-called “Third Italy”, an area of north-eastern Italy centered

154

G. Nassimbeni / Journal of Operations Management 21 (2003) 151–171

in the region of Emilia-Romagna (for example, Sabel, 1982; Lazerson, 1988; Best, 1990; Pyke et al., 1990; Storper and Scott, 1992). Day et al. (2000) observe that “the various regions of the so-called “Third Italy” are heralded as bastions of flexible specialization”. In reality, the lightly used term “industrial district” conceals several forms of labor organization and socio-cultural structures (Pyke et al., 1990). On the basis of a thorough empirical comparison of 24 districts, Paniccia (1998, 1999) demonstrates that only some exhibit the canonical (Marshallian) characteristics of districts. The models of labor division vary from place to place, being influenced by historical and technical factors. Social variables (embeddedness, trust, familiarity, and shared values) generally assume an importance, which is far below that of the canonical description. Similar conclusions can also be found in Staber (1988). Thus, strong differences exist between local systems, often hastily labeled as “districts”, and some of the most characteristic variables, particularly those of a social nature, seem to be more significant in theory than in practice. In the recent years, technological developments and market globalization have shown up all the limits and difficulties of the traditional district formula, such as acquiring adequate technology and financial resources because district enterprises are generally under-sized. Consequently, radical innovations are postponed, marketing and sales systems become embrittled, and there is excessive dispersion of added value (Amin and Robins, 1990; Grabher, 1993; Garofoli and Mazzoni, 1994; Gottardi, 1996; Varaldo and Ferrucci, 1996). Several of these limits can be ascribed to the district enterprise, others to the network, the district system, which is often unable to muster strategic actions in the face of new challenges. These difficulties are, to some extent, linked to the nature of the districts themselves. As they are spontaneous, self-organizing structures formed by distributed intelligence and decisional capabilities they often lack the ability to organize joint planning, in other words, govern the change together (Corò and Micelli, 1999). Other studies seem to indicate that we are passing through a period when districts are “re-blooming” (Fabiani et al., 1998; Becattini, 1999; Signorini, 1994, 2000). Porter’s work (1990, 1998) merits particularly mentioned. He observes that a geographic

concentration of producers can increase productivity by enhancing access to knowledge and other factors involved in production. These contributions are part of a stream of studies that claims that global economy does not necessarily involve the destruction of local systems (Vaccà, 1994; Becattini and Rullani, 1996; Rullani, 1998). This theory is supported by two convictions. First, globalization will not mean uniformity and homologation, i.e. standardization of products, processes and knowledge. On the contrary, it will reward differences, variety, and specificity. Second, a firm’s competitive advantage is rooted in its territorial environment, that irrenounceable place, where knowledge, relations and interactions regarding production processes can be exchanged (March, 1990). They are, after all, the places where the value and variety are generated. According to this perspective, global economy will favor greater geographical division of labor and additional specialization, offering opportunities for the development of local systems that constitute the irreplaceable basin of specific, and non-transferable resources and competencies. Thus, various positions regarding the prospects of local systems in the new competitive context can be found. It is a widely shared opinion, though, that global economy is modifying the traditional territorial forms, changing inner ties and strengthening the need to open up to the outside (Amin, 1993; Rullani and Romano, 1998). In conclusion, a close look at the evolution of many territorial systems shows how far concepts regarding Marshallian districts and flexible specialization are removed from reality. At times the survival of the district itself is queried. This question is often asked (Amin, 1993; Harrison, 1994; Varaldo and Ferrucci, 1996; Becattini and Rullani, 1996; Asheim, 1996): “In the era of globalization is the district formula maintaining its vitality, beginning to decline, or just changing shape?”

3. The operations management perspective The above question is mostly found in economic literature. The majority of the authors cited in the previous section—from Marshall to Piore and Sabel—come from this background. Only rarely have OM scholars

G. Nassimbeni / Journal of Operations Management 21 (2003) 151–171

analyzed the problematics connected with industrial districts: • A branch of OM studies put forward the idea that new production paradigms, as opposed to the consolidated practices of Fordism and Taylorism, are emerging. Bartezzaghi (1999) observes that the process of surpassing Fordism–Taylorism has not produced one single model, but rather a whole range of solutions. These latter are: (a) Japanese lean production; (b) German diversified quality production; (c) Swedish reflective production; (d) flexible specialization. Spina (1998) has reached a similar conclusion. Thus, the flexible production model is described as one of the various post-Fordism experiences, but is not characterized in OM terms and discussed in relation to current changes in the competitive context, in particular globalization. • Some works analyze specific supply chain dimensions in industrial districts. Carbonara et al. (2002) propose a framework to describe supply chains within industrial districts in terms of physical, technological, strategic and organizational variables. De Toni and Nassimbeni (1995) compare the evolution of two industrial districts, analyzing the variables that determine or impede the formation of strongly linked buyer–supplier systems and fashion logistic pipeline mapping and management. In these papers, the local manufacturing system is considered as a given system, without first characterizing its OM specificity. The fact that the OM perspective is somewhat neglected is surprising: industrial districts are specific manufacturing systems, i.e. systems characterized by a particular combination of manufacturing tasks and choices. Thus, the objectives of this section are to: (a) characterize industrial districts and the flexible specialization model emphasizing the OM aspects; (b) highlight the challenges globalization poses to this model. The most relevant OM aspects are defined according to the literature on manufacturing strategies and typologies. Miller and Roth (1994) describe two core elements that are essential for the definition of a manufacturing strategy. The first is the task, i.e. capabilities the manufacturing units must possess in order to compete. Lists of critical competitive capabilities typically include quality, cost/efficiency, delivery/responsiveness,

155

flexibility (Hayes and Wheelwright, 1984; Ferdows and DeMeyer, 1990; Roth and Van Der Velde, 1991). The second is the pattern of manufacturing choices. Tasks and choices should be congruent: the central theme of papers on manufacturing strategy is the fit between these two elements. At this point a methodological problem arises. Industrial districts are, by definition, composed of a number of units. So, to which unit of analysis should the manufacturing tasks and choices be referred? In our opinion, to the leading firms of the district. In fact, these firms—usually located at the end of the value chain (main-contractors)—are the units that decide how the work should be distributed and select the external sources. Therefore, the flexible specialization model can be characterized by the following combination of tasks and choices (Table 2): • manufacturing task. The main task of a locally distributed manufacturing system is flexibility, as the term “flexible specialization model” indicates. There are a number of definitions and classifications of manufacturing flexibility. D’Souza and Williams (2000) recently analyzed this literature, and found that most definitions describe manufacturing flexibility as the ability of the manufacturing function to make the necessary adjustments in response to environmental changes, without significant sacrifices to performance. In addition, they identified and carried out an empirically test on the main dimensions of this construct: volume, variety, process and material handling flexibility. Literature on industrial districts shows that, historically, local manufacturing systems performed well in at least three of these dimensions. They could respond quickly to changes in the level of output (volume flexibility), produce a number of different products (mix flexibility), and adjust to and accommodate changes or disruptions in the manufacturing processes (process flexibility). These capabilities are connected to: ◦ a manufacturing capacity higher than that of an integrated enterprise. This production capacity is governed by mostly small firms, where the work-force is more flexible and polyvalent, i.e. in terms of duties and working hours, than in large-sized enterprises;

156

G. Nassimbeni / Journal of Operations Management 21 (2003) 151–171

Table 2 Key characteristics of the flexible specialization model Key dimensions of the manufacturing strategy

Underlying hypotheses

Challenges

Task

Flexibility, cost

End-producers located in developing countries may achieve a better flexibility/cost ratio

Choices

Vertical integration: low (extensive outsourcing)

Cluster of small firms can achieve greater flexibility and collective efficiency than a large-scaled integrated enterprise Specialization on a few value chain activities provides advantages such as economies of experience and scale Local sourcing provides advantages associated with the presence of external economies (lower transaction costs, institutional support, infrastructures for training and financial support, etc.)

Geographical dispersion of operations: local facilities and sourcing

◦ a wide mix of specialist capabilities. Localized manufacturing systems are based on specialization: each unit focuses on a single phase or technology, where it develops scale and experience economies. The units of a local system are legally and economically independent, therefore, their entrepreneurial attitude, motivation and levels of control are better than those of production departments in a large-sized integrated enterprise. They can often have significant fiscal advantages too. Thanks to these characteristics flexibility in local systems is achieved at relatively low costs, indeed cost can be considered the second main task of this type of manufacturing system. Recently, Adler et al. (1999) reviewed the literature on the trade-off between flexibility and efficiency. Among the instruments for shifting this trade-off, they number “partitioning”, that is, the division of the “structure into separated units that can be staffed by specialized personnel and structured and managed differently to assure their optimal performance”. The resulting specialization enables subunits to refine their capabilities in each activity, gaining cost-advantages. • Manufacturing choices. Manufacturing strategy should focus on a consistent pattern of decision making within key manufacturing resources categories (Hayes and Wheelwright, 1984; Buffa, 1984; Hill, 1989). Different manufacturing typologies

The dispersion of added value lengthens the time to market and hinders product and process innovation Global sourcing can bring benefits to firms through improved price, quality, and innovation

originate from a combination of different choices. OM literature proposes various typologies, based on a number of dimensions, i.e. the complexity of manufacturing technology (Woodward, 1965), the maturity of the process (Abernathy and Townsend, 1975), the sophistication of the product and the production method (Miller, 1988). Hayes and Wheelwright (1984) argue that many characteristics of productive units are functions of two primary dimensions with complimentary life cycles—process and product structures. Their work has been extended by Kotha and Orne (1989). They point out three-dimensions characterizing manufacturing typologies: process structure complexity, product line complexity, and organizational scope. Recently, Devaraj et al. (2001) carried out an empirical comparison of Hayes and Wheelwright (1984) and Kotha and Orne (1989) models. They found that including the “organizational scope” in the two-dimensions of the product–process matrix as proposed by Kotha and Orne (1989) adds explanatory power when predicting a broad array of manufacturing performances. Therefore, we chose Kotha and Orne’s approach to characterize industrial districts in operational terms. Among their three-dimensions only one, organizational scope, seemed to be suitable. The Italian experience and that of other regions demonstrates that production localized in industrial districts covers a wide range of products (machine tools, micro-mechanics, furnishings, clothing

G. Nassimbeni / Journal of Operations Management 21 (2003) 151–171

elements and accessories (dresses, eyewear, jewellery, footwear), agro-industrial products, ceramics, etc.) and processes. Using Kotha and Orne’s categorization, we can say that the end-product of industrial districts differs in complexity, variety, maturity, and volume, and equally in process mechanization, systemization and interconnection. Within the dimension “organizational scope” the relevant variables are: geographic manufacturing scope, geographic market focus, vertical integration, customer–market scope, and scales (Kotha and Orne, 1989). Among these variables, two in particular can be used to characterize the flexible specialization model: • vertical integration, that is the number of process stages coordinated within the firm. By definition local manufacturing systems originate from an intense process of work distribution among a number of specialized units. These systems are, therefore, characterized by low vertical integration; • geographic manufacturing scope, that is the geographic extension of the firm’s manufacturing structure. Industrial districts are characterized by a local distribution of production, i.e. a manufacturing scope restricted to a limited geographic area. Consequently, two assumptions underlie the flexible specialization model. First, the division of labor (specialization) and the resulting focalization on a particular step in the process leads to economies of experience and scale in each step, as well as the various forms of flexibility previously introduced. Second, local sourcing is preferable to in-sourcing or extra-local out-sourcing, since firms benefit from external economies (the presence of agencies that offer specific services, a specialized labor market, support infrastructures, lower transport cost, greater facilities for interaction thanks to geographic proximity and common cultural identities, etc.). According to the flexible specialization model, the combined effect of specialization and localization advantages justify the development of local industries. Global economy questions these assumptions, at least in some districts. As far as the manufacturing task is concerned, end-producers localized in developing countries can offer a better flexibility/cost ratio owing to less restrictive environmental regulations, lower site costs, accessibility to certain natural resources,

157

and lower wages (Prasad and Babbar, 2000). Leading firms of Western districts may be solicited to differentiate their offer in terms of quality, innovation, and delivery/responsiveness. However, as already observed, local systems have frequently shown marked weaknesses in regard to these capabilities. For example, the dispersion of added value among a number of (mostly) small firms can lengthen the time to market and limit product or process innovation, especially when innovation spans several production stages (Robertson and Laglois, 1995). Sometimes the generally small-sized units of a local system do not possess adequate financial and managerial resources to introduce important innovations. As a result leading firms may prefer internalization of production or even extra-local sourcing. Prasad and Babbar (2000) recently published a review of articles about facility location decisions and global sourcing. They observe that firms increasingly strive to gain competitive advantage by sourcing overseas. Such undertakings have been reported extensively in global investigations of purchasing patterns. Global sourcing can benefit firms by improving price, quality, and innovation. In addition, world-wide diffusion of modern information and communication technologies makes firms less dependent on the local context and encourages them to extend their sourcing areas. In conclusion, the initial question can be rephrased in OM terms: “Is the flexible specialization model, characterized by wide outsourcing from local and strongly specialized enterprises, still valid?”

4. Objectives This article presents the results of an empirical research carried out on one of north-eastern Italy’s most important districts: eyewear. Italy is a world-class producer of eyewear: the first four world-wide leaders are Italians. As a consequence this sector represents a significant part of the Italian “fashion” industry. In 2000, the Italian eyewear sector registered sales valuing around 1.75 billion (85% exported). The production value of the districts is 1.5 billion (more than 80% of the national production), with 13,700 employees in 820 companies (ANFAO, 2000, 2001). Indeed, the industry has been in this district for more than a century.

158

G. Nassimbeni / Journal of Operations Management 21 (2003) 151–171

The district is characterized by the presence of a limited number of medium and large-sized companies and a dense network of small enterprises (usually sub-contractors). The former expanded in recent years with growth rates of around 10% per year, while the latter are finding it difficult to defend their sales levels, when not in fact fighting to survive. During the period of 1992–1996, the index of concentration (market share of the top four enterprises in the Italian market) shifted from 57.3 to 60.1% (Databank, 1997). In 1999, the small enterprises declared an average decrease of 25% in sales to foreign markets, while the first seven Italian companies recorded an increase of 65% (ANFAO, 2000). The district is changing, particularly in regard to inter-firm relations between the leading companies (main-contractors) and sub-contracting units. Core-firms are drastically re-thinking their sourcing choices, the supply chain and the involvement of the local sub-contracting system. The objectives of this work are to: • analyze the evolutionary dynamics of the eyewear industry, with particular reference to the impact of globalization and the consequent strategies adopted by large producers; • analyze the large producers’ supply policies and the effects these polices have had on the local sub-contracting system; • analyze characteristics, technological and managerial profiles of the local supply system, and determine its weaknesses. 5. Methodology 5.1. The industry The eyewear industry and the district considered are the appropriate setting for research on changing local systems. First of all, the district model has long been successful in this type of manufacturing. Eyewear production favored inter-firm work distribution, since a broad range of modularity could be achieved and the economies of scale varied according to the manufacturing steps. In addition, out-sourcing from the local market was cheaper as transaction costs were lower. These conditions fostered district development. In turn, the district was the main distinctive

resource of Italian producers (besides the excellent design features of Italian products), guaranteeing world-wide leadership. However, internationalization has gradually modified those aspects (optimal dimension of scale, specialization advantages, and costs of local transactions) that constituted the competitive advantage of the district. Consequently it is an optimal environment in which to analyze a changing local system. 5.2. Research design The investigation developed along two lines: the first, considered main-contractors and, the second, local sub-contractors. This way the phenomenon could be analyzed from both the buyers’ (large producers) and local suppliers’ points of views. Two different approaches were chosen, as combining more than one can give better results (McCutcheon and Meredith, 1993). A multiple case-study analysis (main-contractors side) was used in conjunction with a survey (sub-contractors side). On the main-contractors side mainly qualitative information was collected (i.e. the evolutionary dynamics of the eyewear industry and the strategic response of large producers): the multiple case-study approach is appropriate when a detailed analysis is required of a phenomenon that can be described mostly in qualitative terms (Flynn et al., 1990). When the aim of a research project is mainly to describe events and outcomes in order to understand the process and the environment, case-study research is commonly applied (McCutcheon and Meredith, 1993). What is more, a standard survey instrument was not suitable for most of the data considered (e.g. product, market, and district evolution, buyers’ production and sourcing policies). In contrast, a survey was carried out among the sub-contractors to collect mostly quantitative data (size and trend of the sub-contracting flow, technological characteristics of firms, etc.). Here a larger sample of firms was required, since the population of sub-contractors in the district is far greater than that of main-contractors. As specialised interviewers carried out the survey, they were able to collect qualitative information and subjective opinions on the evolution of the district and the prospects of small local units.

G. Nassimbeni / Journal of Operations Management 21 (2003) 151–171

5.3. Sample

Table 3 The sample of sub-contracting units

Five main-contractors (buyers) and 49 local subcontractors were involved. Main-contractors were selected according to their size and importance to local sub-contracting systems. These firms were identified using databases, personal contact and institutional sources of information. The main-contractors approached are the top eight producers in the district. Five of them agreed to co-operate; two of these are the global leaders in the frame sector. Both, in terms of sales (more than 60% of overall district sales) and product typology, this sample can be considered satisfactorily representative. The next step was to identify a sample of their small local sub-contracting units. An effort was made to select suppliers covering almost all purchasing typologies, and characterized by a wide range of performance levels and sizes. We focused on “small” suppliers, i.e. enterprises with less than 22 employees (Italian regulations), that are by far dominant within the district. First, the entrepreneurs of sub-contracting units were contacted by letter. The aims of the research project were explained and then they were asked whether they were willing to co-operate. This contact strategy, together with the sponsorship of main-contractors and the local “Agency for District Development”, resulted in a high response rate (75%), corresponding to 49 companies. The characteristics of the sample are reported in Table 3. In order to assess non-response bias, our sampled units were compared with the entire sub-contracting population of the district in terms of size (the only parameter available from institutions). The average size of our sample was slightly below the average of the population (8.92 versus 11.8 employees, ANFAO, 2000). Therefore, some empirical evidence—in particular, that related to structural weaknesses of small units—could be overestimated.

Size No. of employees Turnover (in millions of

5.4. Data collection By means of structured interviews, case-studies of the main-contractors were developed, gathering information on each company’s history, product, and market characteristics, procurement policies and outsourcing choices, the evolution of the local subcontracting network and geographical extension of the

)

Market Percentage sales in the district on total sales Percentage sales in the rest of Italy to total sales Percentage export on total sales Customers Percentage sales to medium and large-sized industrial firms Percentage sales to small-sized industrial firms Percentage sales to commercial agents or firms Kind of products Percentage of products/components made on forecasts Percentage of products/components designed and made-to-order Percentage of products/components made on own design Percentage subcontracted workings or services

159

8.92 0.60 89.89 3.26 6.86 36.28 50.81 12.91 1.22 10.91 6.12 81.73

n = 49.

supply base. When possible the individuals directly involved in purchasing and actively managing external relations were interviewed. These interviews were supplemented with documents and records that provided extensive information on each firm’s evolution, problems, and relationships with local enterprises. This was followed by a survey of 49 sub-contracting units in order to gather data on their demographic, managerial and technological profile, innovations and amounts and kinds of investments. The questionnaire used in the survey was developed by the research group partially using or re-elaborating measurements proposed in the literature (Lefebvre and Lefebvre, 1992; Sakakibara et al., 1993). Other measures had been developed and validated in previous research work by the author. The questions are mainly objective but certain perceptive measurements of the Likert type were used to evaluate the use of some management practices. Appendix A gives a brief description of the questionnaire. A pilot test of the questionnaire was conducted in two firms. Specialized interviewers carried out the survey and the questions were addressed to the owner of the firm or a partner. These interviews also supplied qualitative information on the trends of the sub-contracting flow and relations with the customers.

160

G. Nassimbeni / Journal of Operations Management 21 (2003) 151–171

Moreover, interviewers explained and clarified how to compile the questionnaire, so reducing the amount of missing data and obtaining more reliable answers. 5.5. Data analysis Similarities and differences among the case studies of the main-contractors analyzed were noted. Although some strategic choices (make or buy decisions, direction and entity of investments) of the analyzed companies differed, views on the evolutionary dynamics of the eyewear industry and district generally converged. Simple descriptive statistical analyses were carried out on the sub-contractors’ side. The sample was divided into two groups: “preferred sub-contractors”— those firms that over the last 3 years had recorded an increase in turnover, i.e. an increase in the subcontracting flow—and “marginal sub-contractors”— the remaining enterprises. The quantitative variation in the turnover in the last 3 years was shown to be strongly correlated to two perceptive measures: future expectations of an increase in the turnover, and future expectations of an increase in the number of employees. Therefore, it appears that this sample partition can discriminate well between the firms characterized by favorable market prospects and the remainder. Then these two sub-samples were compared in relation to a variety of demographic, technological and managerial factors. One-way analysis of variance (ANOVA) for metric variables was used, which corresponds to the common Students t-test for two independent samples. The variables used in this study are both discrete (i.e. number of employees) and percentages (i.e. percentage of sales in Italy), so that their distribution can shift from the normal distribution required by the ANOVA. However, the well-known robustness of the F-test used in ANOVA is adequately reliable for a study of this type. In the presence of dichotomous variables (i.e. use of titanium alloys in the production process), that is, variables that can only assume two values (0, 1), the χ 2 -test was employed. Here, the “independence” hypothesis has to be verified in the 2 × 2 contingency table, which corresponds to the hypothesis of “equality” of percentages of membership in the two groups (“preferred” and “marginal” sub-contractors) for each of the two sub-samples.

6. Results: the main-contractors’ side Managers of the main-contracting firms were asked about the characteristics of the international offer, market changes, and critical competitive factors. Then the procurement policies and outsourcing choices were investigated, focusing on supplier selection criteria and the re-organization of the supply chain. Information received from the main-contractors is presented in this section, as well as data supplied by sources, such as the National Association of Factories of Optical Articles (ANFAO, 2000, 2001), the Local Manufacturers Association (Assindustria Belluno, 2001), and former research work (Databank, 1997, 1998; Camuffo, 2002). 6.1. Eyewear producing countries Purchasing managers of the maincontractor firms were asked how the eyewear production and sourcing areas developed over time. Traditionally, the production of frames was concentrated in Europe (mainly Italy, France and Germany) and the far east (Japan, Korea, Hong Kong, Taiwan and Indonesia). Europe and the far east differed in regard to the type of product. That from the far east was cheaper and inferior in quality (with the exception of Japan), while more expensive and better-differentiated frames (in terms of quality, design and color) came from Europe. Although Italy still maintains its leadership (Italian producers claim 54% of the world market (ANFAO, 2001)), all the managers interviewed agreed that Asia has become more competitive, since the quality/price ratio of its products has improved considerably. The Italian association of factories for optical articles estimates that in China alone around ninety thousand employees work in this industry (ANFAO, 2001). The Italian import of Chinese eyewear components and products has quintuplicated in the period 1998–2000 (ANFAO, 2001). European producers have found that the cost of labor is much more favorable than at home. ANFAO (2001) estimates that the wages of Chinese workers are about 20 times lower than those of Italian workers. The wage factor has convinced several companies to transfer some of their production to these countries. For example, Luxottica—one of the sampled enterprises—recently acquired a plant in

G. Nassimbeni / Journal of Operations Management 21 (2003) 151–171

Hong Kong for manufacturing crystal lens. Luxottica together with a Japanese partner have opened a joint venture in China for manufacturing metal frames. Similarly Safilo—another of the sampled enterprises— has gone into partnership with a Turkish producer for manufacturing frames in Istanbul. As a consequence, the sub-contracting flow directed to the local system has decreased. According to the local Manufacturers Association (Assindustria Belluno, 2001), the population of district enterprises has been reduced by 25% in the period 1995–2000. The vanishing firms are mostly smallsized. 6.2. The role of fashion All main-contractors interviewed pointed out that nowadays eyewear sales are strongly influenced by fashion trends. In fact, the role of the product has changed. In the past, the consumer’s purchasing behavior was mainly governed by the “functional” use of the product, i.e. correction of a visual pathology. Therefore, consumers considered mainly the technical features such as practicality and comfort. Subsequently, attention shifted to aesthetic and intrinsic qualities. Producers then began to think of the product also in terms of style, design and color (ANFAO, 2000, 2001). The strong influence of fashion is demonstrated by the high number of famous stylists and eyewear producers that have begun to co-operate in the recent times. Luxottica, the largest maincontractor analyzed, is linked to stylists such as: Giorgio Armani, Moschino, Valentino, and Yves Saint Laurent. Safilo co-operates with Pierre Cardin, Gucci, Biagiotti, and Missoni, among others. The most famous fashion stylists and designers carefully choose eyewear producers to whom they will entrust their brand name, privileging those equipped with greater productive and commercial solidity. The trademark has become one of the most important competitive weapons capable of profoundly influencing sales and consequently marketing and production polices. Good management of the brand calls for a sound commercial structure. In the sampled enterprises from 8 to 12% of all investments are in sales promotion. Collaboration between stylist/designer and producer is no longer occasional but has become a medium to long-term investment.

161

6.3. Innovation All the main-contractors interviewed agreed that the ability to innovate processes (i.e. new manufacturing technologies) and products (i.e. testing lighter, more resistant materials—“nickel-free” alloys, titanium, etc.) has become a critical competitive factor. This confirms the research carried out by Databank (1998) on the key competitive factors in the eyewear industry. Product innovation immediately follows “distribution capability” and “trademark” as an important factor. Over the last few years, this has radically modified the product and its constituent parts, including lenses for both sun and prescription glasses. Developments in product treatment and coloration are also underway. Finally, the need for quality control is growing. This is no longer based only on visual tests, but also on chemical, physical and mechanical analyses. Product and process innovations are costly and constant monitoring of the technological available is required. 6.4. The product life cycle The heightened sensitivity of the product to fashion trends has shortened its life cycle. The main-contractors in our sample completely renew their collections on average every 2.5 years, that is they change 40% of their products every year. This reduced life has enhanced the complexity of forecasting, product design, manufacturing, planning and control phases. In our sample the time to market is approximately 6–8 months, data confirming that of ANFAO (2001). When the commercialization of the latest model begins, the product development cycle of the subsequent model should be already under way. Moreover, the eyewear market has undergone a process of segmentation. For example, Luxottica makes approximately 2100 different product models, with more than 20,000 size and color variants. As a result, small–lot productions are required and all producers are actively seeking mix and volume elasticity. 6.5. Sales and distribution There is general agreement that the competitive focus has moved from production to sales/distribution, so critical activities in value generation and higher profits are located in the terminal phases of the value

162

G. Nassimbeni / Journal of Operations Management 21 (2003) 151–171

chain (ANFAO, 2000, 2001; Camuffo, 2002). In fact, the average mark-up of the distribution chain is around 50%, however, it can reach 70–80% of the final sale price (Databank, 1998). Our data confirm this trend: all the five main-contractors analyzed have significantly invested in sales and distribution activities and, in the two cases, they have carried out a clear strategy of forward integration. For example, Luxottica has acquired two American distribution chains: “Lens Crafters” (860 stores in the USA) and “Sungluss Hut International” (the largest eyewear distribution chain in the world, 1730 sales points in North America) in order to penetrate the US market, which alone accounts for approximately 40% of world-wide sales (ANFAO, 2000). Not only by chance has Luxottica been quoted since 1990 on the New York Stock Exchange, together with the third largest Italian company, De Rigo. Similarly, Safilo has acquired Smith Sport Optics, the leading company in the US for sports eyewear and ski goggles. The smaller enterprises and especially those trying to enter international markets, encounter several difficulties at this stage. In terms of marketing competencies, product mix, delivery time and technical assistance to the final distributors they are clearly less competitive than the larger firms. In addition, they suffer from scale disadvantages. 6.6. Sourcing policies Main-contractors must nowadays manage a design– production–distribution chain whose lead times have been drastically reduced while quality and mix requirements have been proportionally increased. These new competitive requirements have underscored all the limits of the district model. The latter, in fact, has serious problems of network co-ordination. The difficulties include satisfying qualitative levels, respecting delivery times and production cycle synchronization. According to the managers interviewed many small district companies cannot carry out the managerial and qualitative improvements required. Two main-contractors analyzed have internalized several production phases so they can be more effectively controlled and co-ordinate. The remaining leading firms have rationalized their supply base, greatly modifying customer-supplier relations and the supply chain configuration. However, there is still significant resort to local production. Main-contractors prefer to

maintain relations with a reduced number of suppliers and establish medium to long-term links with selected sources. The sourcing area is generally international and purchases from the far east are increasing (usually lower value and labor intensive components). Local sourcing remains dominant, though all main-contractors analyzed deplore the technical and managerial weaknesses of several local suppliers. Managers were asked to rank the vendor selection criteria. They agreed that the crucial criteria are: “timeliness and reliability of deliveries” (essential for reducing time-to-market), “volume elasticity” (the “fashion” factor requires smaller lot sizes), and “quality” (sub-contractors must guarantee certified quality). “Price” and “terms of payment”, though important, are relegated to second place. Since the eyewear environment lays particular importance on reservedness (imitation is frequent), “exclusivity” is very highly appreciated. Criteria such as “technological profile” and “specialization on specific workings” are important for some items. Most of the main-contractors interviewed consider “design capability” a relevant feature. “Geographical proximity” is almost irrelevant. Buyers are willing to cross domestic borders in search of a more favorable offer (the incidence of transport costs of eyewear components is low). As already pointed out, all the main-contractors interviewed reported that the quality, technological profile and design capabilities of many local sub-contractors are not adequate.

7. Results: sub-contractors’ side The survey involved 49 sub-contractors. A variety of aspects relating to the demographic profile, technology, management, innovation and investments of the sub-contracting units were investigated in order to obtain a complete picture of the strengths and weaknesses of these enterprises, and consequently their sub-contracting offer. The main empirical results are summarized as follows. 7.1. Demographic profile Table 3 reports the main demographic characteristics of the sample. Its productive structure is modest

G. Nassimbeni / Journal of Operations Management 21 (2003) 151–171

in size (8.92 employees per enterprise) and the small handicraft unit was found to be dominant (the volume of sales is lower than 0.25 billion in more than 50% of the enterprises). In 64% of the cases, the entrepreneur worked in the past the same industry having served, on average, 7 years apprenticeship. The companies are firmly labor intensive. The cost of the work-force accounts for 54% of the turnover. Sub-contracting of single components (such as hinges, noses, and bridges) or workings (such as pickling, galvanization, tumbling, and buffing) prevails (81.75%), while few firms manufacture/assemble phase products (a group of components, such as rods) or end products. The high percentage of sales absorbed by the first customer (70%) indicates nearly total commercial dependency on a few customers, for the most part located in the district (90%). Commercialization of the products is almost entirely carried out by the entrepreneur. Resorting to commercial agents is rare. The greatest problems perceived by the sampled units regard access to distribution channels and the cost of distribution.

163

to TQM practices, the formalized definition of quality objectives and procedures is fairly widespread, as is the use of quality performance indicators on entry and exit flows. On the other hand adoption of management techniques to study defects (Pareto diagrams, cause–effect diagrams, control charts, etc.) is rare. 7.3. Innovation and investments Investments are mainly in production machinery, which is substantial (11.6 manual machines, 2.27 logical programmable control (PLC) and 0.23 computerized numerical control (CNC) machines per company) and quite modern (average age respectively: 3.11, 2.54 and 2.8 years). Expenditure for external consultancy, hardware and software equipments, instead, has not received particular attention. Investments in sales promotion, work-force training and technical patents or licenses are neglected. Obstacles hindering the introduction of innovation, in the opinion of the interviewed entrepreneurs, are mainly financial.

7.2. Technology and management 7.4. Market prospects Among the various technological classes examined, productive technologies are by far the most important, especially in companies that work innovative materials (nickel-free and copper–beryllium alloys, titanium, etc.) and are, therefore, forced to replace obsolete machinery with modern equipment. The use of complementary technologies (for planning, quality control, handling, and storing) is limited. There are probably several reasons for this: (1) the level of investment, judged by the entrepreneurs to be too high with respect to the volume of products manufactured; (2) the slight importance that entrepreneurs attribute to areas complementary to production; (3) the difficulties they have in understanding the potentials of such technologies. Few units (14%) use computer aided design (CAD) instruments. Therefore, only a small number of units is aware of the fact that advanced design instruments can in the long run provide better market opportunities. Some companies (31%) use management software. This is mainly true in the area of administration (24.5%) and commerce (20.4%). However, the use of software for production planning and control (14.3%) and for quality control (18.4%) is limited. With regard

The data summarized above depict a typical subcontracting unit which possesses a considerable amount of productive equipment, but is weak in complementary areas (such as planning and control, design, quality management, and sales). Market prospects, estimated in terms of variations in the turnover forecasted by the interviewed entrepreneurs, are generally stationary or negative. Many of the sampled companies attribute their current difficulties to a decrease in the volume of sales to the final market. Actually, in the last few years this Italian industry has increased its volume of sales (ANFAO, 2000). Therefore, what has dropped is not primary demand, but the supply volumes as a consequence of the internalization or supply base rationalization made by main-contractors. However, the survey has shown that these uncertain market prospects do not involve the entire sample of sub-contractors. Some companies have positive trends and favorable future prospects. Thus, an attempt was made to pinpoint which factors differentiate the companies with favorable prospects from the rest. The sample was divided

164

G. Nassimbeni / Journal of Operations Management 21 (2003) 151–171

Table 4 Comparison between “preferred” and “marginal” suppliers Variables

“Preferred” suppliers (n = 19)

“Marginal” suppliers (n = 30)

Statistic

P (significance)

Technology and management Average no. of PLC machines Use of CAD instrument (present or forecasted) Use of software for production planning and control Use of software in purchasing/materials management Quality certification (present or forecasted) Importance of quality control technologiesa Statistical process control

4.94 55% 27% 39% 83% 4.25 29%

0.83 26% 7% 7% 54% 2.80 7%

F = 4.157 χ 2 = 4.828 χ 2 = 3.982 χ 2 = 2.674 χ 2 = 2.598 F = 5.893 χ 2 = 5.246

0.047 0.028 0.046 0.102 0.107 0.019 0.022

Innovation and investments Level of future investments in technologya Product material: titanium Product material: nickel-free alloys Product material: copper–beryllium alloys

3.06 50% 44% 67%

2.11 13% 20% 30%

F = 6.115 χ 2 = 7.550 χ 2 = 2.911 χ 2 = 5.916

0.017 0.006 0.088 0.015

a

Five-point Likert scale.

into two groups: “preferred” suppliers and “marginal” suppliers, according to the design presented in Section 5. Using ANOVA and the χ 2 -test, these two sub-samples were compared in terms of demographic profile, technology and management, and investment and innovation capabilities. Table 4 indicates only those factors that discriminate between the two subsamples. None of the demographic factors investigated (size, production typology, and customer typology) was found to be discriminating. The factors that distinguish “preferred” from “marginal” suppliers are mainly technological and managerial. The number of PLC (4.94 versus 0.83) is considerably higher in the “preferred”. Significant differences (55% of “preferred” and 26% of “marginal”) emerge in the use (present or forecasted) of CAD systems. Possession of advanced instruments for design activities is of growing importance when selecting sub-contractors. Buyers look for partners with advanced and compatible systems for defining and exchanging product and process specifications. As far as management software is concerned, the analysis highlights some significant differences between the two sub-samples. The enterprises with better prospects make greater use of software in purchasing and materials management (39% versus 7%) and production planning and control (27% versus 7%) areas. They attribute more importance to

quality certification (actually present or forecasted in 83% of “preferred” versus 54% of “marginal” firms) and to statistical process control (29% versus 7%). Indeed main-contractors are urging their sources to obtain qualitative improvement by formalizing their definition of objectives and procedures. They also insist on quality control of all incoming and outgoing parts and materials and the use of defective classification techniques. Furthermore the percentage of companies that use innovative materials (titanium, nickel-free and copper–beryllium alloys) is much higher within “preferred” suppliers. Finally, main contractors privilege those sources that are able to supply a phase (a complete group of components, i.e. the headpiece completed with nosepiece, rings and bridge) or finished products rather than single items. This way buyers reduce the number of entry channels, limit the administrative costs and the supply lead time, have prompt control of materials, and simplify the production planning activities.

8. Discussion First the main reasons for the transformation of the district analyzed are pointed out and its historical evolution described (Section 8.1). Next future developments of the district are hypothesized (Section 8.2).

G. Nassimbeni / Journal of Operations Management 21 (2003) 151–171

165

financial, training, and technological support). Eyewear manufacturing took root quickly in a territory that possessed a patrimony of handicraft expertise and practical knowledge. This was both an initial advantage and the factor leading to conglomeration. The first phase was characterized by the gradual development of external economies, while the advantages of specialization still had to be explored. Almost all the phases of the process (from design to product sale) were frequently governed by small to medium sized enterprises.

The scheme reported in Fig. 1 summarizes the historical evolution of the district. Two variables are used (a) the advantages of specialization, and (b) the advantages of localization (Fig. 1). These two variables were introduced in Section 3: they best characterize the manufacturing typology typical of local manufacturing systems. Marshall (1920) explains the genesis of districts in terms of a combination of localization and scale economies. A (canonical) district develops when advantages are gained by distributing work, and at the same time the outsourcing choice favors the local system. Fig. 1 reports the three evolutionary steps of the district.

8.1.2. Hierarchization Later on, as the competition became fiercer and extended to international markets, enterprises emerged that were better equipped, in terms of managerial and financial resources, to confront international competition and enter foreign markets. Functions and competencies that had become critical in the changed competitive scenario, i.e. design, promotion, and distribution capabilities had to be carefully developed and controlled. The ability to innovate product design and develop and test new materials and process technologies became essential. Sales promotion capabilities were needed when it became important to join with famous stylists (“griffe”) as a prestigious brand name constituted the main selling vehicle. Distribution capabilities were required because sufficient financial solidity, adequate volumes, and a strong image had to be presented to distributive chains and sale points, in order to find space for their own models. The exigency to control these critical functions favored an increase in size and selected leaders, who focused on greater added value activities and depended on the local sub-contracting system for flexible and cheap manufacturing. The advantages of localization and specialization fully unfolded during this phase.

8.1.1. Birth and development Around the beginning of the 1970s there was a rapid proliferation of eyewear enterprises, often through processes of affiliation. This industry already had a solid tradition within the geographical territory considered, but was confined to a small number of units. A virtuous circle was triggered. The proliferation of the companies generated and justified the development of initiatives to promote the local industry (i.e.

8.1.3. Internalization or selective resort to the outside The present situation is characterized by the difficulties of the district model to meet customers’ demands for greater quality, product variety, and shorter lead times. On one hand fragmentation of the production tasks has led to problems in co-ordination and control of flow. On the other, several sub-contracting units are incapable of collaborating with the leaders in those processes and innovations that the market

Fig. 1. The interpretative scheme.

Finally the significance of the findings for managers and entrepreneurs is highlighted (Section 8.3). 8.1. Three evolutionary steps of the district

166

G. Nassimbeni / Journal of Operations Management 21 (2003) 151–171

demands. The main-contractors’ reaction to this is either to internalize production tasks and, so obtain direct control of the flow or rationalize the supplier base by forming a mutually more binding and demanding relationship with a small select group. In other words, in the supply relationship there is a transition from “capacity” to “speciality” sub-contracting, from “quantity” to “quality”. Outsourcing, in these cases, does not necessarily mean local (district) sourcing. Main-contractors are now willing to cross the district borders to reach more economic (far east) or better qualified suppliers. Up to now, this has regarded mostly low value added items, selected primarily on price. In a few cases, it has been justified by the need to contact sources able to guarantee high quality or technical capabilities, not possessed by district suppliers. Apart from the entity of international sourcing, suppliers selection criteria show that geographical proximity does not condition sourcing choices. As far as the advantages of specialization are concerned, no clear verdict has yet been reached. Some enterprises choose to resort to outside production while others opt for internalization. The main characteristics of each evolutionary step are summarized in Table 5. 8.2. The future of the district Is the eyewear district destined to decline? It is difficult to answer this question. There are some indications that the current district structure will not survive the challenges of global competition. In the

first place, leading firms are re-organizing processes and sourcing choices in an attempt to keep core competencies and key abilities inside. Consequently there is a risk that small sub-contractors will loose their distinctive abilities, and be transformed into satellite units or become productive appendices strongly subordinated to customers. In some cases, as when main-contractors decide to internalize production, there is the danger that the units will not survive at all. The district of the past, composed of a large number of vital, closely interconnected small enterprises each with a cultural identity and specific patrimony of technical expertise, will gradually transform into a small group of leaders surrounded by their satellite units. Indeed the more critical activities, the distinctive capabilities, are no longer technical-productive but have shifted to the ambit of design and sales distribution. Nowadays, the value of eyewear is largely intangible, linked to its image, brand, and the consumers’ perception. Localized knowledge (of a technical-handicraft character) has gradually lost the importance it had when manufacturing aspects played the most critical role. In the second place, prevailing customer-supplier interactions have modified some cognitive characteristics of the supply relationship. Innocenzi (1998) and Albino et al. (1999) observe that in other Italian districts there has also been an increment in codified knowledge to support the current requirements of district relations. This is because new forms of planning/synchronization of the production flow demand adequate information infrastructures, interactive technologies for product development (co-design through

Table 5 The main characteristics of the evolutionary steps of the district Internalization Early development

Hierarchization of the district Selective out-sourcing

Rapid proliferation of enterprises

Emergence of leading firms as a way to face international competition

Institutional initiatives for local industrial development

Design, sales promotion, distribution capabilities required of leading firms

Enterprises usually govern all the phases of the process Development of external economies

Local sub-contractors for flexible and efficient manufacturing District as a closed system

Market requires differentiation: leading enterprises must be able to guarantee high quality and technical capabilities Supplier base rationalization: transition from “capacity” to “specialty” sub-contracting Internalization in order to better coordinate the flow Sourcing extends beyond district borders

G. Nassimbeni / Journal of Operations Management 21 (2003) 151–171

CAD–CAM instruments), and quality certification. The tacit knowledge that was specific to the district has become progressively more explicit, translated into software instructions that are manageable by the main-contractors’ design and production planning computer packages. Without much trouble the leading firm can now co-ordinate the flow and control the product. Moreover, when knowledge becomes explicit (codified) and the interaction formalized, then it can be more easily transplanted elsewhere. Transfer of production to countries with a lower cost of labor is still at the experimental stage but it has been found that the district manufacturing experience can be transplanted to other contexts. It is often emphasized that globalization and advancements in information and communication technologies create not only threats but also opportunities for local systems. The empirical evidence shows that the eyewear district has lost most of its traditional features and inter-organizational relations have significantly changed. If the trend in the supply flow directed towards the local sub-contracting units is considered, the future prospects of many are not very bright. The opportunities arising from new technologies and global economy seem to favor better equipped, large-sized enterprises rather than small suppliers, or at any rate, single units or groups rather than the system. 8.3. Implications for entrepreneurs and managers As this study focuses on a single industrial district the indications provided cannot be rigidly generalized. However, entrepreneurs and managers of main-contracting and sub-contracting firms operating in industrial districts, are offered several important insights. 8.3.1. Main-contracting firms According to the flexible specialization model, a firm opts for a low level of vertical integration and ample local outsourcing so it can benefit from the advantages of specialization and localization offered by the local environment. The underlying assumption of this model is that the strong inter-firm division of labor in local clusters of small firms can achieve a greater flexibility and collective efficiency than larger-scaled, vertically integrated enterprises

167

(Table 2). The documented competitive advantage of several districts shows that these assumptions used to be valid in many industrial and territorial contexts. However, current changes in the competitive environment, in particular globalization, make this model debatable. Local manufacturing systems could lose their cost competitiveness, especially in laborintensive industries where producers operating in developing countries possibly have significantly lower workforce costs. Local systems may lose quality and innovation competitiveness, because limited financial and managerial resources hinder the introduction of product or process innovations. They can lose time/response competitiveness, when excessive dispersion of added value causes problems in the coordination and control of flows. When rationalizing the local supply base only the better equipped suppliers are selected and medium to long-term programs planned with them. However, it is becoming more unlikely that local basins can completely satisfy the demands for innovation, quality, design, and manufacturing services that modern industrial enterprises expect. Consequently maincontracting firms must extend their sourcing area, selecting suppliers from outside the district. Finally, the study suggests that managers should re-examine their internalization choices as the internal coordination of processes could reduce manufacturing lead times and time to market, while direct control of the flow may enhance product quality. 8.3.2. Sub-contracting firms The study indicates the characteristics of district enterprises (“preferred sub-contractors”) that were able to increase their sub-contracting flow and develop more direct and stable relations with important customers. These mainly refer to technological and managerial profiles and innovation capabilities. The prospects are still good for small district sub-contractors that can sustain an advanced technical dialogue with the main-contractor (using, for example, CAD–CAM systems), use innovative materials (such as titanium, nickel-free and copper–beryllium alloys), and connect up to the buyer’s operational systems (e.g. use production planning and control systems compatible with those adopted by the buyers).

168

G. Nassimbeni / Journal of Operations Management 21 (2003) 151–171

Some of these innovations, in particular CAD and software instruments for design and production/materials management, require a financial resources that is well within the reach of even very small enterprises. It is essential for managers to re-discover the benefits that can be gained from investment in items other than traditional manufacturing machinery.

9. Conclusions The eyewear industry has undergone major changes in the recent years. These have generally aided the larger companies with sufficient strategic planning capabilities, financial resources and managerial expertise to face present challenges. The transformations have, however, shown up all the limits of the smaller district enterprises. Leading firms are re-engineering the main processes so that they can offer better quality and time to market. This restructuring means, either internalization of production, or rationalization and selection of sub-contracting relationships. Main-contractors require a pool of suppliers with advanced design and manufacturing equipment and tools suited for the synchronized management of the production. This investigation demonstrates that “preferred” suppliers are superior in terms of technological profile, design expertise, instruments, and investment capability. These characteristics are lacking in most local suppliers. The transformations taking place in the eyewear districts are paralleled in other Italian districts, for example the hierarchization process, subsequent development through internal or external lines, and gradual impoverishment of the local system (Carminucci and Casucci, 1997). Similarities can be found also in some Italian enterprises that in the past made wide use of the local production network, such as Benetton (Camuffo et al., 2001). Is the district formula destined to decline? The industry and the district are still changing so a verdict is not possible. However, the transfer of production to the countries with a lower work-force cost, even if still at the experimental stage, has given dubious results. And although a certain volume of supplies comes from outside the district (far east), this mainly concerns non-critical parts and components of cheaper product

lines. It is still possible that technical-productive traditions and geographic proximity between the leading firms and their local sub-contractors will play a role. Geographic proximity also means cultural affinity, a common language, and more frequent and direct interactions. However, it is beyond doubt that the inter-firm relations currently developing in the district are distant from the “Marshallian” concepts. The relational “atmosphere” and co-operative spirit that cemented the “canonical” district seem to have been surpassed by vigorous competitive attitudes, and crushed by the rigid supplier selection criteria of the leading firms. Nevertheless the best sub-contractors still complement the operations of the leading firm thereby perpetuating a territorial concentration of eyewear manufacturers. On the other hand, concentration alone is not enough to qualify a group of enterprises as an industrial district, at least not in the “canonical” sense of the term.

Appendix A The questionnaire addressed to the sample of sub-contractors is divided in the following sections: • The firm’s characteristics. We used the traditional indicators for “size” (number of employees (aggregated or separated according to their qualifications), sales) and “age” (number of years since the firm had been founded). We evaluated how firms respond to the market through the percentage of make-to-stock, make-to-order, engineer-to-order production. In order to characterize the client profile, data were collected on: (a) the percentage of sales directed to industrial versus commercial and to large versus medium or small sized firms; (b) the percentage of sales to customers located in the district, in the rest of Italy or in foreign countries; (c) the percentage of sales absorbed by the first three customers. Other questions relate to the incidence of the cost of the workforce and the kind of components or services supplied. • Technology. The importance and up-to-dateness of the technology adopted in the main areas (manufacturing, quality control, management, design, communication, handling, storage) were evaluated through perceptive five-point Likert scales. In

G. Nassimbeni / Journal of Operations Management 21 (2003) 151–171

addition, we used the following quantitative indicators of the presence, number and type of tools used: • production technologies: number and age of logical programmable control (LPC), non-computerized numerical control (NC), computerized numerical control (CNC), direct numerical control (DNC) machines, and flexible manufacturing systems (FMS); • design technologies: number of CAD and computer aided manufacturing (CAM) tools; • information technologies: presence of software packages for client order management, administrative activities, purchasing/material management, production planning and control;

169

• communication technologies: presence of informatic linkages, both to public (i.e. through internet) and private (i.e. through electronic data interchange) networks. • Management. The following Likert scales were used to measure just-in-time, concurrent engineering, and quality management practices. Some of them were adapted from Sakakibara et al. (1993). Please circle the number which indicates the extent to which you agree or disagree with each statement (1 = completely disagree . . . 5 = completely agree).

Just-in-time practices JIT delivery: “Our main suppliers deliver to us on a Just-in-time basis”. Plant lay-out rationalization: “The layout of our factory facilitates fast throughput”. Set-up time reduction: “We are actively working to lower setup times in our plant”. Small lot-size production: “Our lot size is equal to present requirements”. Concurrent engineering practices Co-design with clients: “We are actively involved in the design activities of our customers”. DFM/DFA: “We use design for manufacturing/design for assembly techniques”. Component reduction: “Component reduction is very important for us and we focus on it”. Component standardization: “We make every effort to use standard components”. Product modularization: “Product modularization is very important for us and we focus on it”. Total quality control practices Procedures and norms: “Our firm has established and formalized procedures and norms for total quality achievement”. Use of defect analysis technique: “We use a defect analysis technique (such as Pareto diagrams, ABC diagrams, etc.” Quality control on entry and exit flows: “We have defined measures to control the flow coming from main suppliers”, “we have defined measures to control the flow directed to main customers”. Preventive maintenance. “We invest in preventive maintenance”. Use of statistical process control. “We use statistical process control to monitor the process”.

1 1

2 2

3 3

4 4

5 5

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5

1 1

2 2

3 3

4 4

5 5

170

G. Nassimbeni / Journal of Operations Management 21 (2003) 151–171

In addition, a dichotomous variable assesses whether the firm has been certified by an institutional agency. • Innovation. Four Likert scales were used to measure the level of the firm’s past and future investments in new materials, product functions, production technology, and working methods. The questions are reported below. Please indicate the level of resources your firm has invested in the last 5 years and intends to invest in the next 5 years in the following areas: Last 5 years New materials New product functions New production technology New working methods

Next 5 years

1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5

In addition, dichotomous variables were used to capture the firm’s resort to outside services (technological and management consultancy) and the kind of materials used (nickel–silver, stainless, brass, titanium, nickel-free, copper–beryllium, and carbonium alloys).

References Abernathy, W.J., Townsend, P.L., 1975. Technology, productivity and process changes. Technological Forecasting and Social Change 7, 379–396. Adler, P.S., Goldoftas, B., Levine, D.I., 1999. Flexibility versus efficiency? A case study of model changeovers in the Toyota production system. Organization Science 10 (1), 43–68. Albino, V., Garavelli, A.C., Schiuma, G., 1999. Knowledge transfer and inter-firm relationships in industrial districts: the role of the leader firm. Technovation 19, 53–63. Amin, A., Robins, K., 1990. Industrial districts and regional development, limits and possibilities. In: Pyke, F., Becattini, G., Segenberger, W. (Eds.), Industrial Districts and Inter-Firm Co-operation in Italy. International Institute for Labour Studies, Geneva, pp. 51–72. Amin, A., 1993. The globalization of the economy: an erosion of regional networks? In: Grabher, G. (Ed.), The Embedded Firm. On Socioeconomics of Industrial Networks. Routledge, London, pp. 278–295. ANFAO (National Association of Factories of Optical Articles), 2000. L’occhialeria Italiana. ANFAO, Italy.

ANFAO (National Association of Factories of Optical Articles), 2001. L’occhialeria Italiana. ANFAO, Italy. Asheim, B.T., 1996. Industrial districts as “learning regions”: a condition for prosperity? European Planning Studies 4 (4), 379– 401. Assindustria Belluno, 2001. Osservatorio sulle dinamiche del distretto dell’occhiale, 2001. Bartezzaghi, E., 1999. The evolution of production models: is a new paradigm emerging? International Journal of Operations and Production Management 19 (2), 229–250. Becattini, G., 1999. La fioritura della piccola impresa e il ritorno dei distretti industriali. Economia e Politica Industriale 103, 5–16. Becattini, G., Rullani, E., 1996. Local systems and global connections: the role of knowledge. In: Cossettino, F., Pyke, F., Sengenberger, W. (Eds.), Local and Regional Response to Global Pressures, IILS, Geneve, pp. 159–174. Best, M.H., 1990. The New Competition. Harvard University Press, Cambridge. Brusco, S., 1982. The Emilian model: productive disintegration and social integration. Cambridge Journal of Economics 6 (2), 167–184. Buffa, E., 1984. Meeting the Competitive Challenge. Irwin, Homewood, IL. Camuffo, A., 2002. Duelli organizzativi: distretti e imprese nell’occhialeria italiana. 3◦ Workshop dei Docenti e Ricercatori di Organizzazione Aziendale, Università di Genova, Genova. Camuffo, A., Romano, R., Vinelli, A., 2001. Back to the future: Benetton transforms its global network. Sloan Management Review 43 (1), 46–52. Carbonara, N., Giannoccaro, I., Potrandolfo, P., 2002. Supply chains within industrial districts: a theoretical framework. International Journal of Production Economics 76, 159–176. Carminucci, C., Casucci, S., 1997. Il ciclo di vita dei distretti industriali: ipotesi teoriche ed evidenze empiriche. L’Industria 23 (2), 283–315. Corò, G., Micelli, S., 1999. Distretti industriali e imprese transnazionali: modelli alternativi o convergenti? Sviluppo locale 6 (10), 16–40. D’Souza, D.E., Williams, F.D., 2000. Toward a taxonomy of manufacturing flexibility dimensions. Journal of Operations Management 18, 577–593. Databank, 1997. Montature e Occhiali. Databank. Databank, 1998. Montature e Occhiali. Databank. Day, M., Burnett, J., Forrester, P.L., Hassard, J., 2000. Britain’s last industrial district? A case study of ceramics production. International Journal of Production Economics 65, 5–15. De Toni, A., Nassimbeni, G., 1995. Supply networks: genesis, stability and logistics implications. OMEGA 23 (4), 403–418. Devaraj, S., Hollingworth, D.G., Schroeder, R.G., 2001. Generic manufacturing strategies: an empirical test of two configurational typologies. Journal of Operations Management 19, 427–452. Fabiani, S., Pellegrini, G., Romagnano, E., Signorini, L.F., 1998. L’efficienza delle imprese distrettuali italiane. Sviluppo Locale 5 (9), 42–73.

G. Nassimbeni / Journal of Operations Management 21 (2003) 151–171 Ferdows, K., DeMeyer, A., 1990. Lasting improvements in manufacturing performance: in search of a new theory. Journal of Operations Management 9 (2), 168–184. Flynn, B.B., Sakakibara, S., Schroeder, R.G., Bates, K.A., Flynn, E.J., 1990. Empirical research methods in operations management. Journal of Operations Management 9 (2), 250– 284. Garofoli, G., Mazzoni, R., 1994. Sistemi produttivi locali: struttura e trasformazione. FrancoAngeli, Milano. Gottardi, G., 1996. Technology strategies, innovation without R & D and the creation of knowledge within industrial districts. Journal of Industry Studies 3 (2), 119–134. Grabher, G., 1993. The weakness of strong ties: the lock-in of regional development in the Ruhr Area. In: Grabher, G. (Ed.), The Embedded Firm. On the Socioeconomic of Industrial Networks. Routledge, London, pp. 255–277. Harrison, B., 1994. Lean and mean. The changing landscape of corporate power in the age of flexibility. Basic Books, New York. Hayes, R.H., Wheelwright, S.C., 1984. Restoring Our Competitive Edge. Wiley, New York. Hill, T., 1989. Manufacturing Strategy: Text and Cases. Irwin, Homewood, IL. Hirst, P., Zeitlin, J., 1991. Flexible specialisation versus postFordism: theory, evidence and policy implication. Economy and Society 20 (1), 1–56. Innocenzi, A., 1998. Gerarchia e contratti. Il ruolo dei rapporti di subfornitura tra piccole imprese nell’evoluzione dei distretti industriali. L’industria, aprile-giugno, pp 391–415. Kotha, S., Orne, D., 1989. Generic manufacturing strategies: a conceptual synthesis. Strategic Management Journal 10, 211– 231. Lazerson, M.H., 1988. Organizational growth of small firms: an outcome of markets and hierarchies? American Sociological Review 53, 330–342. Lefebvre, E., Lefebvre, L.A., 1992. Firms innovativeness and CEO characteristics in small manufacturing firms. Journal of Engineering and Technology Management 9, 243–277. March, J., 1990. Decision and Organizations. Blackwell, Oxford. Marshall, A., 1920. Principles of Economics. Macmillan, London. McCutcheon, D.M., Meredith, J.R., 1993. Conducting case study research in operations management. Journal of Operations Management 11, 239–256. Miller, D., 1988. Relating Porter’s business strategy to environment and structure: analysis and performance implications. Academy of Management Journal 31 (2), 280–308. Miller, J.G., Roth, A.V., 1994. A taxonomy of manufacturing strategies. Management Science 40 (3), 285–304. Paniccia, I., 1998. One, a hundred, thousands of industrial districts. Organizational variety in local networks of small and

171

medium-sized enterprises. Organization Studies 19 (4), 667– 669. Paniccia, I., 1999. The performance of IDs. Some insights from the Italian case. Human System Management 18 (2), 141–159. Piore, M., Sabel, C., 1984. The Second Industrial Divide. Basic Books, New York. Porter, M., 1990. The Competitive Advantage of Nations. The Free Press, New York. Porter, M., 1998. On Competition. Harvard Business School Press, Boston. Prasad, S., Babbar, S., 2000. International operations management research. Journal of Operations Management 18, 209–247. Pyke, F., Becattini, G., Segenberger, W. (Eds.), 1990. Industrial Districts and Inter-Firm Co-operation in Italy. International Institute for Labour Studies, Geneva. Robertson, P.L., Laglois, R.N., 1995. Innovation, networks, and vertical integration. Research Policy 24, 543–562. Roth, A.V., Van Der Velde, M., 1991. Operations as marketing: a competitive service strategy. Journal of Operations Management 10 (3), 303–327. Rullani, E., 1998. Riforma delle istituzioni e sviluppo locale. Sviluppo Locale 5 (8), 5–46. Rullani, E., Romano, L., 1998. Il postfordismo. ETAS. Sabel, C.F., 1982. Work and Politics. The Division of Labor in Industry. Cambridge University Press, Cambridge. Sakakibara, S., Flynn, B.B., Schroeder, R.G., 1993. A framework and measurement instrument for just-in-time manufacturing. Production and Operations Management 2 (3), 177–194. Saxenian, A., 1991. The origins and dynamics of production networks in Silicon Valley. Research Policy 20, 423–437. Scott, A.J., 1988. Flexible production systems and regional development: the rise of new industrial spaces in North America and western Europe. International Journal of Urban and Regional Research 12, 171–186. Signorini, L.F., 1994. Una verifica quantitativa dell’effetto distretto. Sviluppo Locale 1 (1), 31–70. Signorini, L.F. (Ed.), 2000. Lo sviluppo locale. Meridiana libri. Spina, G., 1998. Manufacturing paradigms versus strategic approaches: a misleading contrast. International Journal of Operations and Production Management 18 (8), 684–709. Staber, U., 1988. Inter-firm co-operation and competition in industrial districts. Organization Studies 19 (4), 701–724. Storper, M., Scott, A., 1992. Pathways to Industrialization and Regional Development. Routledge, London. Vaccà, S., 1994. Le imprese transnazionali tra sistemi locali e sistemi globali. Economia e Politica Industriale 84, 8–21. Varaldo, R., Ferrucci, L., 1996. The evolutionary nature of the firm within industrial districts. European Planning Studies 4 (1), 27–34. Woodward, J., 1965. Industrial Organization. Oxford University Press, Oxford.