1 Abstract Drawing from literature on immigrant entrepreneurship, local production systems and the concept of embeddedness, this paper aims to explore the role of small and medium sized firms established by Chinese immigrants in some Italian industrial districts. In particular, it highlights how this ethnic group is strictly involved and active in current changes into the district supply chain of the garment sector, notwithstanding the classical culturalist interpretation of immigrant s firms which depicts Chinese businesses as self-referential and endogenous ethnic economies. Results indicate that Chinese firms do both create independent ethnic supply chains and collaborate with local firms, helping them to face today global competition through a sort of in-loco internationalization. In this case, actual inter-firm and social relations' patterns seem to be a consequence of structural endogenous and exogenous changes in district labor markets and value chains. Sintesi Dopo un analisi della letteratura socio-economica riguardante l imprenditorialità immigrata, i sistemi di produzione locale ed il concesso di embeddedness, il lavoro qui pubblicato esamina il ruolo delle piccole e medie imprese costituite dagli immigrati cinesi in alcuni distretti industriali italiani. In particolare, si evidenzia come questo gruppo etnico sia direttamente coinvolto negli attuali cambiamenti che stanno caratterizzando il settore tessile-abbigliamento, a dispetto della classica interpretazione di matrice culturale del fenomeno migratorio che descrive le imprese gestite da imprenditori cinesi come organizzazioni chiuse ed autoreferenziali. I risultati indicano che gli imprenditori cinesi, oltre a creare delle vere e proprie catene di fornitura indipendenti, gestite solo da connazionali, collaborano attivamente con le imprese locali, svolgendo diverse fasi produttive in tempi rapidi ed a basso costo, aiutando così gli imprenditori italiani a fronteggiare le sfide della competizione globale attraverso una sorta di internazionalizzazione locale. Università degli Studi di Urbino Carlo Bo FACOLTÀ DI ECONOMIA THE ROLE OF CHINESE SMES IN ITALIAN INDUSTRIAL DISTRICTS EDUARDO BARBERIS, SELENA AURELI Selena Aureli è Ricercatore in Economia Aziendale presso la Facoltà di Economia dell Università degli Studi di Urbino Carlo Bo Eudoardo Barbersi è Assegnista di Ricerca presso la Facoltà di Sociologia dell Università degli Studi di Urbino Carlo Bo Quaderni di Economia Aziendale n. 14
2 Università degli Studi di Urbino Carlo Bo FACOLTÀ DI ECONOMIA THE ROLE OF CHINESE SMES IN ITALIAN INDUSTRIAL DISTRICTS COMITATO SCIENTIFICO/SCIENTIFIC BOARD Francesca Maria Cesaroni Massimo Ciambotti Paola Demartini Fabio Giulio Grandis Mauro Paoloni SEGRETERIA DI REDAZIONE EDITORIALE/ EDITORIAL OFFICE Dipartimento di Studi Aziendali e Giuridici (DISAG) Facoltà di Economia/Faculty of Economics Università degli Studi di Urbino Carlo Bo /University of Urbino Via Saffi, 42, URBINO (PU) - ITALY Tel Fax: EDUARDO BARBERSI, SELENA AURELI I contributi pubblicati sono sottoposti a doppio referaggio/ Papers have been double-blind peer reviewed before final submission ISBN Quaderni di Economia Aziendale n. 14 Stampato presso il Dipartimento di Sudi Aziendali e Giuridici (DISAG) dell Università degli Studi di Urbino Carlo Bo nel mese di ottobre 2010 Per la presente pubblicazione sono stati adempiuti gli obblighi previsti ai sensi della L n. 106, del D.P.R n. 352 e del DM MiBAC
3 I Quaderni di Economia Aziendale della Facoltà di Economia di Urbino Quaderno n. 13 Selena Aureli, Paola Demartini (2010) INTERNATIONALIZATION OF CHINESE FIRMS IN EUROPE: THE ROLE OF CULTURAL DIFFERENCES IN THE FUNCTIONING OF A M&A IN THE AUTOMOTIVE INDUSTRY Quaderno n. 12 Mara Del Baldo (2006) LA GESTIONE MANAGERIALE DEL PATRIMONIO ARTISTICO E CULTURALE DELLA CHIESA. L ESPERIENZA DEI MUSEI DIOCESANI NELLE MARCHE Quaderno n. 11 S. Aureli - F.M. Cesaroni - P. Demartini - P. Paoloni SMES FINANCIAL MANAGEMENT INFORMATION AND WEBSITES Quaderno n. 10 Selena Aureli (2006) IMMIGRANT ENTREPRENEURS IN ITALY. DOES NATIONAL CULTURE MATTER? Quaderno n. 9 P. Demartini F. M. Cesaroni S. Aureli (2005) INTERNATIONALISATION AND FINANCIAL COMMUNICATION SKILLS FOR SMALL BUSINESSES Quaderno n. 8 Paola Paoloni (2004) L APPLICAZIONE DEI PRINCIPI CONTABILI INTERNAZIONALI ALLE PICCOLE IMPRESE Un analisi dell utilità Quaderno n. 7 M. Paoloni F.M. Cesaroni - P. Demartini (2003) UPGRADING SME INFORMATION SISTEMS: THE ROLE OF VENTURE CAPITALIST INVESTORS
4 THE ROLE OF CHINESE SMES IN ITALIAN INDUSTRIAL DISTRICTS 1 1. Introduction 2 This paper is intended as an outlook on Chinese migrants' participation into Italian industrial districts (IDs), grounded in a new economic sociology approach focussing on intergroup links and an interpretation of social embeddedness of economic action (Granovetter, 1985) not limited to in-group relationships. In particular, this research attempts to answer the following questions: are Chinese immigrants firms cooperating with local firms held by Italians? Are they entering Italian IDs as substitutes of pre-existing artisan businesses, or are they creating their own parallel supply chain inside an ethnic enclave? Chinese migrants' participation in IDs is interesting for different reasons: first of all, the number of Chinese small enterprises is rising and sometimes countering the downfall of Italian family businesses; secondly, there is a lively rhetoric on the exceptionalism of Italian IDs, based on traditions and community networks making them a place-specific cluster protected by external influence (Pyke, Becattini and Sengenberger, 1990) and thus theoretically resistant to the entrance and integration of 1 The paper is part of a common research project developed by the two authors. Aureli wrote par. 1, 2.1. and 2.2. while Barberis wrote par. 2.3., 3, 4, 5 and 6. 2 A previous version of the paper has been accepted and presented at the 2 nd ASIAN MANAGEMENT AND ENTREPRENEURSHIP WORKSHOP held in Brussels (Belgium), 30 November- 1 st December 2009.
5 4 S. Aureli, E. Barberis foreign operators... though, more and more immigrants are working there as employees and entrepreneurs; lastly, we can identify similarities between two types of ethnoindustrialization (Piore 1990), the IDs themselves and the ethnic economies. In order to understand Chinese economic role, the main assumption is that the supposed social closure of Chinese migrant groups cannot be analysed separately from socio-economic features of destination areas, where inter-group networking processes involves both local and foreigner actors in a sort of coevolution process with mutual consequences. Hence, the usual focus available in the literature on Chinese ingroup closure and communitarian bonds is not so fit for our case-study. It is assumed here that a structural turn is needed to better understand closures usually analysed only from a culturalist point of view. The paper is organized as follow. After a theoretical overview of main concepts and theories at stake useful to frame the issues at stake, we will account empirical findings coming from field research on different IDs, with a special focus on the case of Carpi clothing district. Then, final paragraphs will deal with the consequences of Chinese insertion in Italian IDs and the possible future chances of inter-group relationships between Chinese and locals. 2. Review of the literature The above-mentioned goals have to be achieved through a multifaceted approach, requiring clarifications of different concepts and theories. On the one hand, we have arguments on immigrants' and minorities' economic participation (underlining the role of in-group ethnic resources in fostering immigrant entrepreneurship); on the other hand, we have arguments on local production systems, in particular IDs.
6 Chinese SMEs in Italian industrial districts Ethnic economy and immigrant entrepreneurship In the last thirty years, immigrant entrepreneurship gained momentum in different social sciences, so that some years ago a huge number of research products (over 1700) on this subject were counted (Kloosterman & Rath, 2003). Main theories focussing on immigration and minority business usually refer to some basic issues (Engelen, 2001): the role, so to say, of the ethnic specificity and of cultural factors; the importance of the supply-side in the structuring of a market for immigrant entrepreneurs. Actually, it is maintained that people belonging to a specific immigrant group have cultural features and behaviours helping self-employment, or that members of an ethnic group can access collective resource supporting their entrepreneurial choice. References to a so-called ethnic economy are focal in this literature branch, mainly based on the account of cultural specificities turning a group toward particular sectors and business choices: human (employees, helpers) and relational (information, contracting, connections, etc.) resources are to be found within the supposed ethnic community itself. First theories by Portes ( ethnic enclave ) and Waldinger ( ethnic niche ) usually consider the above-mentioned variables as discriminant, since structural, class features helping business choice are maintained to be more or less the same everywhere, based on bourgeois values not so interesting as a study object, since they are considered as invariable facts (Wilson and Portes, 1980; Light, 1984; Waldinger, 1994). Though, these approaches are somehow weak in explaining how different groups access entrepreneurship and/or build up ethnic advantages. Moreover, in-group relationships of the ethnic group are taken for granted (there is a static idea of culture) instead of being empirically proved in their existence and extension (Baumann, 1999) and inter-group relationships and structural
7 6 S. Aureli, E. Barberis factors framing action (and affecting in- and inter-group interactions) are undervalued (Engelen, 2001; Rath, 2001; Werbner, 2001). Thus, from the 1990s onwards, previous studies were increasingly challenged by more dynamic interpretations: Waldinger formulates his interactive model theory (Waldinger, Aldrich and Ward, 1990), taking into account the opportunity structure, while Portes, getting into the new economic sociology debate, studies relational and structural embeddedness (Portes and Sensenbrenner, 1993). These approaches draw a bigger attention to structural factors, but set them consequently with previous interest for cultural variables, i.e. analysing mainly supply-side factors, whereas contexts where entrepreneurial actions are embedded are considered a background. So, Waldinger s structure of opportunity is just a resource pattern which given ethnic predisposing factors take more or less advantage of; similarly, Portes structural embeddedness is a fait accompli less interesting than mainly in-group relational embeddedness (Portes and Borocz, 1989; Waldinger, 1989). Within this model, Chinese immigration has been a theoretically relevant and much studied case, due to its strong self-employment rate consistent with the idea of a self-referential community (Rath, 2002). In fact, Chinese nationals abroad are usually considered one of the migrant / minority communities characterized by the highest degree of social closure and communitarization, largely relying on in-group relationships, reciprocity and bounded solidarity (Basu and Goswami, 1999; Wong and Ng, 1998 and 2002). At the same time, also some strongly structuralist approaches developed, both in economics and sociology. These approaches use to focus on macro-dimensions such as the political and institutional frame ruling economy and immigration policies; changes in the mode of production; economic circumstances. On the one side, this started an attention on space- and time-framing of opportunities, but on the other side an actor-centred perspective
8 Chinese SMEs in Italian industrial districts 7 was lost. Dual markets, secondary sector, blocked mobility, shelter firm theories (Piore, 1979; Sassen, 1991), even the interesting essay by Engelen (2001) on the economic factors making an immigrant enterprise successful, underestimate individual and relational resources affecting the impact of macrovariables. Also in this case the key explaining factor - context - is mainly conceived as a static element. To summarize we can see a structuralist interpretation facing a culturalist one, both with an atomized view of the actor, both with a problem in representing local structure and collective action as hypostatic. The first stream narrows down its analysis on the local context just from an institutional point of view, while the second one discusses immigration issues just from an in-group community point of view. Hence, it is maintained here that the main problem is the asymmetry in the study of agency and structure: usually agency being taken into account only as an in-group fact, and the structure as an out-group issue. In other words, agency is given just to studied minorities, while other relevant actors in the locale, affecting also immigrant entrepreneurs networks, are neglected. Moreover, in our concern, the poor attention given to the building and transformation of networks linking migrants and mainstream social actors is the hint of the need for an encompassing theory of society. Thus, it is worth approaching both sides (local and immigrant people) with similar tools and methods, so to prevent essentialism and an asymmetric understanding of social bonds: kinship networks are not just an immigrant affaire and can be cross-identity, while institutional settings are not exclusively defined by mainstream people Industrial districts Since the 1970s, the Italian district model has been subject of many national and international studies, particularly following Giacomo Becattini s re-elaboration of Alfred Marshall s thought (Becattini, 1979 and 1990). Becattini described a new kind of
9 8 S. Aureli, E. Barberis economic organisation based on a territorially bounded community (of people and firms) sharing a common knowledge, personal motivations and collective values. In his idea, the local production system is a place-based collective actor including territorial relationships (Becattini, 2004; Crouch et al., 2001). From the point of view of the organization of production, they have been usually defined as closed or captive systems characterized by highly localized inter-firm linkages, while economic relationships with the external environment were limited to the few companies devoted to the commercialization of the final products. IDs formation depends on some territorial as well as organizational and cultural factors, while they derive their strength from a specific division of labour, flexible specialisation and a competitive-cooperative climate, which stems from a class structure based on artisans and a protected working class (Bagnasco, 1981; Piore and Sabel, 1984; Trigilia, 1986; Brusco, 1989; Bagnasco and Trigilia, 1993). For almost twenty years now, there has been a debate on changes affecting these local systems (e.g. global competition, consumer behaviours, technology, etc.), and scholars wonder if there is a transformation leading to new scenarios or just a decline (Garofoli, 2003; Giuliani et al., 2005; Biggiero, 2006). A sharp look over these changes should start from the evolution of both the internal and international context. The increase of import-export represents a relevant internationalization process involving both final and phase companies in local economies. Moreover, there is an increase in the outsourcing of most labourintensive production activities, while several foreign multinationals have been entering district supply chains. The result is that manufacturing districts are loosing their captive quality and the production chain is no longer located completely within the district (Corò and Grandinetti, 2001; Aureli et al., 2008). Last, but not least, we notice a growing role of immigrant labour, both as employees and self-employed (Ambrosini 2005).
10 Chinese SMEs in Italian industrial districts 9 In particular, migration challenges interpretations based on endogenous advantages, and asks for deeper analyses on change paths characterizing district societies and economies in the last years. The appearance of small firms held by immigrants shows that we have to get over interpretations related to the supposed consumption of aboriginal social capital, in order to analyse the re-embedding processes that (on the one side) keep continuity with historical district configuration and (on the other side) engender new and undervalued developments A common ethnoindustrialization. putting together ids and ethnic economies In a way, we can see a common weakness in theories about IDs and ethnic economies: a static idea of local and migrant communities focussed more on cultural bounds, shared pasts and in-group closure than on their ability to manage internal and external relations. Thus, it is appropriate to re-frame these two major scientific backgrounds with a common and consistent approach. It is maintained here that New Economic Sociology and the social embeddedness of economic action (Polanyi, 1957; Granovetter, 1985; Granovetter and Swedberg, 2001) can be inspiring in overcoming these interpretative problems. First, as for immigration and immigrant entrepreneurship, there are some pieces of literature making reference to the embeddedness theory (Portes and Sensenbrenner, 1993): in the last decade, it has been enhanced by the "mixed embeddedness" approach (Kloosterman and Rath, 2001), underlining that immigrant firms are rooted both in normative and institutional opportunity windows and in personal and group resources. Therefore, there is an attention for the working and structuring of markets, to be seen as sets of rules and social relations in which immigrant entrepreneurs need to be embedded (Rath, 2002). As a further contribution to this approach, it is maintained here that the social construction of markets, with an attention on the
11 10 S. Aureli, E. Barberis making/changing of structures themselves (as cause and consequence of immigrant market participation) needs to be seen through the analysis of place-based inter-group relationships (underlining the agency of local and immigrant actors in building up their social and economic participation) (Barberis, 2008) 3. Studying agency in a cross-group perspective allows the use of research tools and approaches highlighting undervalued ties. The embeddedness of immigrant firms should be placed within a wider context, including also majority and mainstream bonds, since trust, networks, and their role in engaging staff or setting inter-firm ties, the sharing of an industrial atmosphere reducing transaction costs, the density of cooperative-competitive relations within a large number of small firms in a single locale are not issues related only to immigrant business or to district economies, but cross-cutting them. Hence, foreign entrepreneurs should be seen as tied with several actors (individuals and institutions, autochthonous and migrants), framed in a single network (though frayed and differently dense) made up by local-based economic and social bonds, allowing and intermingling with the economic and social actions of immigrant entrepreneurs themselves. To simplify, we can imagine just two reference groups in a local context, both characterized by different degrees of embeddedness (Burt, 1992; Grabher, 1993; Uzzi, 1996 and 1997): i.e., every group can have networks more or less overembedded (when interfirm ties happen almost exclusively within a thick kinship clique), more or less underembedded (when ties are mainly spot and market-based, with a poor role of in-group bonds). The inclusion of two groups within the same socio-economic context produce embeddedness patterns based also on the mutual relationship setting, changing developments and paths. In short, the embeddedness of group A is interwoven with and co-dependent 3 Thus, the focus will be on inter-group relations and relative positioning in the local system, not considering in-group closure as taken for granted: the existence of both the majority and the minority community and their bounded solidarity should be verified (Baumann, 1999).
12 Chinese SMEs in Italian industrial districts 11 from the embeddedness of group B, according to the role played by the relationships between A and B (as far as density, homogeneity, redundancy, and especially power configurations are concerned). In our case, one group is given by immigrants accessing entrepreneurship (they-group) and the other by autochthonous dwellers (we-group). According to this intersection between embeddedness paths, we can theoretically draw up a set of possible configurations. Fig. 1. Inter-group relationship and embeddedness model In short we can identify the following configurations: II: Overembedded we-group / Overembedded they-group. In this case, two groups insist formally on the same territory, but having poor socio-economic exchanges, so that a balance is kept by segregation (Sofer & Schnell 2001). IV: Underembedded we-group / Underembedded theygroup. In this case in-group bounded solidarity is weak for
13 12 S. Aureli, E. Barberis both the groups (even, they can hardly be defined groups ), and everyone is similarly exposed to free market forces. III: Overembedded we-group / Underembedded they-group. The we-group is locked-in and can hardly produce innovation to answer to external stimula; thus risks are handed off to the they-group, to reduce in-group conflicts. Flexibility and adaptation is granted by asymmetry between two hierarchically different social groups. Low density of links between We- and They-groups allows also exploitation ans systematic violations of agreements, because negative behaviours toward They-group have no immediate return effect on We-group membership. Furthermore, the Theygroup can hardly resist, since it is weakly organized and widely exposed to external forces. Anyway, this hierarchical configuration erodes the trust in the system as a whole. Inclusion in the core and social mobility are hindered by the strong closure of the more powerful group, so that the ingroup itself becomes weaker due to lack of transformation and adaptability skills. I: Underembedded we-group / Overembedded they-group. This is the configuration better describing the ethnic niche economy, where a minority group can find its place in a socio-economic fabric. Advantages are given by bounded solidarity acting as a last resort network, but there are also disadvantages tied with lock-in risks. V: Embedded we-group / Embedded they-group. Ideal configuration where the two groups are structurally coupled and interdependent on an equal footing (no group has a gain in opting out), with reciprocal advantages and externalities. 3. Methodology In order to understand Chinese s involvement in current changes into the district supply chain, considering both their in-group relationships and possible interactions between immigrant and
14 Chinese SMEs in Italian industrial districts 13 local firms, we use empirical data derived from Barberis Ph.D. research project carried out between 2005 and 2007 (Barberis, 2008). The study is based on both quantitative and qualitative data as follows: a context analysis of local production systems (identified on the basis of commuting areas, as defined by the Italian Institute of Statistics ISTAT) in two Italian provinces (Vicenza and Modena) located in North-Eastern regions of the country and characterized by a strong entrepreneurial vocation and high immigration rates; data processing and analysis of about 11,000 immigrants' firms located in those two provinces; 22 interviews with key informants (immigrant entrepreneurs; local stakeholders; business associations' experts and leaders; politicians; civil servants, etc.). Information will help us to verify if Chinese insertion in local economies is not only related to in-group links, but also to the inter-group and out-group ones and to associated negotiations. In particular, we will analyse territorial and niche specializations of both Italian and Chinese firms, since they can be seen as a proxy of local interaction (territorial bonds that are a typical characteristic of IDs). Actually, localization cannot be explained just by in-group resources: IDs can have a relevant role in structuring Chinese nationals' settlement patterns and labour insertion. 4. Data analysis 4.1.Chinese migrants in Italian industrial districts In general terms, we can observe that migration in Italy is widespread in Central and Northern Italy and due to the polycentrism of the Italian urban system quite scattered in
15 14 S. Aureli, E. Barberis medium- and small-sized towns (even though the two urban areas of Rome and Milan sum up almost 20% of migrants and 13% of the whole population) (Istat, 2008 and 2009). Compared to the biggest foreign national groups like Romanians, Moroccans and Albanians, Chinese migration is much more centred in few areas; and compared to other Asian migrations (e.g. from Bangladesh, Pakistan, India, Philippines), it is less centred in large cities. Actually, Chinese migration is strongly linked with district areas manufacturing classic made in Italy products (light industrialization in mature markets with labourintensive production processes), mainly in the sectors of leather goods, footwear, textile and clothing (see Fig. 2). Moreover, Chinese nationals have a high entrepreneurship rate. They create small businesses in a contest where small and medium enterprises (SMEs) are focal for the economic system and, moreover, they create firms that do not just replace disappearing street-corner shops, but enter in a competitivecooperative game with Italian firms (Guercini, 1999). Data indicates that settlement patterns of Chinese in Italy widely involve some of the most traditional and well-known industrial districts and wholesaling centres of the country (Ceccagno 2003). Thus, Chinese socio-economic participation can be considered as quite unique. As a consequence, the focus on Chinese migrants is theoretically meaningful, in order to understand participation into major Italian local economies (including IDs), in the frame of wider socio-economic changes (Carchedi & Ferri 1998; Cologna 2005; Pieke et al. 2004). So, it is possible to see a range of specializations and microspecializations, usually connected with textile, clothing and leather goods production. Though, it is important to notice that not all the Chinese nationals working in the manufacturing sector are employed there (e.g. some can be found in biomedical, metalwork, tile and furniture districts), while on the other hand Chinese firms and workers do not get into every textile, clothing and leather goods district. There is a selectivity to be analysed, in
16 Chinese SMEs in Italian industrial districts 15 terms of constraints and opportunities, entry barriers and facilities. Fig. 2. Chinese citizens per municipality (data 01/01/2008; share per 1000 residents) and localisation of local production systems 4 Source: processing on ISTAT data (ISTAT 2009) 4 Local production systems are labelled as follows: 1) Barge (stone industry); 2) Carpi / Reggio Emilia (textile and clothing knitwear/ ladies' wear); 3) Mantua (textile and clothing hosiery); 4) Rovigo (textile and clothing); 5) Montebelluna (footwear sportswear / technical apparel); 6) Belluno (eyewear); 7) Prato- Firenze (textile / leather goods); 8) Forlì (forniture couches and armchairs); 9) Casentino / Sant'Angelo in Vado (textile and clothing wool / denim); 10) Fermo / Monte San Giusto (footwear); 11) Ascoli Piceno / Teramo (textile and clothing; leather goods and footwear subcontracting); 12) San Giuseppe Vesuviano / Solafra ( textile and clothing; leather goods). Also the cities of Milan (MI) and Rome (RM) are higlighted.
17 16 S. Aureli, E. Barberis 4.2. Chinese nationals in Modena and Vicenza industrial districts We chose two Provinces, Modena and Vicenza, because they are both characterized by a high level of Chinese residents and a significant entrepreneurial attitude stemming from this national group. In 2008, in the Province of Vicenza there were some foreign nationals (10,5% of resident population), around from China. In the Province of Modena, they were some (11,1% of resident population), whose around were Chinese. With reference to Chinese firms, we can notice that in both local areas they are active in manufacturing sectors and the entrepreneurship rate is close to 20% (see Table 1). In 2005, Chinese firms in textile and clothing sector were 417 in the province of Modena and 142 in the province of Vicenza. They accounted for 2/3 of Chinese firms in Modena, and half in Vicenza, and in both areas 9 firms out of 10 led by foreign nationals in the sector were Chinese. However, if we compare the two cases, we can see some contextrelated differences. Actually, beside the apparent concentration of Chinese firms in both Provinces, there is a diverse distribution of Chinese firms in ID cores and their outskirts. A more detailed mapping of the Modena Province shows that 58% of Chinese apparel firms lie some 20 km far from the core town of the ID (Carpi and the surrounding municipalities 5 ) - just beyond the boundaries of the district as designed by commuting catching areas; 23% are in the core area itself. Thus, more than 80% of Chinese apparel firms are tied with Carpi ID, even though somehow in its peripheral area (mainly in the small town of 5 Carpi knitwear ID has some firms and employees. Around 34% of the whole active population is involved in the ID activities. ID turnover is around millions Euro (1.600 millions USD) 4% of Italian turnover in this sector and the export accounts for 36% of the total. Among the most known brands based in Carpi, we can list Champion, Fujiko, Gaudì, Blumarine, Denny Rose. Furthermore, Carpi contractors and subcontractors work for other famous Made in Italy brands.
18 Chinese SMEs in Italian industrial districts 17 Mirandola 6 ). Thus, we can see a kind of district effect (e.g. the territorial aggregation of firms in the same supply chain), mixed with a kinship effect (e.g. the territorial aggregation of firms and entrepreneurs with similar socio-cultural features). Although Mirandola is specialized in metalwork production, here we can find a strong concentration of immigrants businesses in the garment sector: among the 300 manufacturing firms run by foreign nationals (data 2005), more than 60% are in the tailoring industry, and a good part of the remaining 40% is active in connected productions, e.g. knitwear or apparel and textile finishing. This specialization is very ethnicized: Chinese nationals are almost 100% of non Italian entrepreneurs in the sector a share much higher than in the core area of the Carpi ID. Furthermore, they are active in almost every niche in the sector and related works: weaving; making of pullovers, cardigan sweaters, hosiery, stitching and basting, till the production of clothes boxes and the repair of knitwear machines. So, at a first glance it can seem an ethnic enclave with a full supply chain, a Chinese self-sufficient parallel district. Nevertheless, a deeper analysis of data from the Chamber of Commerce and from key informants show a different picture. Chinese firms are mainly subcontractors (and subcontracting is declared to be the sole activity by some half of them), often working in very simple and labour-intensive processes, so that often they are the last (and often the weakest) link in the subcontracting chain 7. 6 Mirandola has inhabitants, whose more than are foreigners (13,3%). Regular Chinese residents are some This is one of the biggest differences with Prato the well-known Italian textile ID involving a huge number of Chinese nationals, making up the second biggest Chinese community in Europe. In Carpi and surroundings Chinese firms do not (yet?) build up their own Made in Italy with a more extensively ethnicized production chain, for many reasons including: the different duration and size of the Chinese community; the different specialization of the district (Prato is more textile, Carpi more knitwear) needing different expertise; the different economic stage of the two districts as for duration, buy-out and externalization processes, size of leading enterprises (Nomisma 2008).
19 18 S. Aureli, E. Barberis So, we have here a special kind of district effect: these firms are minority in the economic specialization of Mirandola local production system, but at the same time they are strongly linked with the nearby Carpi ID. Thus, immigrant business find a room deflected from the core ID area, and reproducing its district mechanisms just some kilometres away. This creates a new mix of dependency by Carpi buyers and autonomy of an ethnicized market out of sight from district formal and informal rules. Selfsufficiency is not the case, here; nevertheless in this area there's more room for subcontracting chains among Chinese (e.g. poor sub-subcontracts, in which Chinese subcontract to other Chinese), which is instead more infrequent in the Carpi core. With reference to the core area of the Carpi ID, we can see some relevant differences with Mirandola s textile and clothing business: the interdependency between local and immigrant entrepreneurs is here somehow different, with a largest share of immigrants firms operating in finishing services (application of buttons, pins and sequins; labelling; ironing and packaging) or working as subcontractors for Italian firms (subcontracting is declared to be the most part of production activities by 3 immigrants' apparel firms out of 4). Furthermore, the distribution by citizenship is a bit more complex, since Chinese (which still represent a large majority) are side by side with other Asians (especially Pakistanis) and Eastern Europeans. The joint analysis of Mirandola and Carpi local production systems weakens the idea of an ethnicized parallel district and intra-district inter-group networks appear as important gatekeepers to access the district core. The result it that just some foreign nationals' enterprises are able to get into the district core: the most organized, established and networked. On the other side, the weakest, poorest, most marginal (also as far as legal requirements are concerned) and isolated ones are stuck in the periphery of the district 8. 8 This engenders a difference in the survival rate (Hazard ratio Cox Regression) of foreign nationals' firms. On average, in Carpi they have a closing-down risk lower than in Mirandola, and even lower than in the rest of the Province.
20 Chinese SMEs in Italian industrial districts 19 On the contrary, in the Province of Vicenza the district effect is affecting much less migrants' firms, and the number of Chinese garment entrepreneurs is limited even in comparison with surrounding areas in the same region. This fact has to be contextualized: even though there are some very important textile and clothing IDs and producers in this province 9, territorial bounds are less visible than in Modena, due to an advanced process of unmaking of the district, strongly hit by outsourcing, internationalization, buy-outs that weakened the ties with the local economy. In that area, ID territorial bound was lost several years ago, due to a peculiar subcontracting system less integrated and more based on medium-large sized buyers. As a consequence top contractors had quite asymmetric relations with local subcontractors, and were able to strongly redesign their territorial bound. Actually, business groups from the Province of Vicenza have been among the first in Italy to relocate production branches abroad (also thanks to nearby Eastern European countries with lower labour costs and more flexible legal standards). As a consequence, internationalization has been a substitute for precarious immigrants' subcontracting, and the decrease of employees and firms in Vicenza garment industry was very sharp (Corò and Volpe, 2006) 10. Thus, Chinese entrepreneurs tried to gain room in the Vicenza garment sector, but the saturation of the market has been very fast: after a growth of foreign nationals' firms in the late Nineties, Comparing just the garment ones, the difference in the closing-down risk is very large (own elaboration drawn form Infocamere data). 9 Valdagno, Thiene, Schio and other towns in the surroundings of Vicenza are the birthplace of brands like Diesel Jeans and Marzotto (heading a group including Lanerossi, Bassetti, Hugo Boss and Valentino Fashion Group). In the nearby provinces of Verona and Treviso there are other important brands like Benetton, Stefanel, Lotto, Diadora, ecc. 10 Just to give an example, the Marzotto group strongly reduced territorial bounds by moving its HQ from Valdagno to Milan, and outsourcing different branches (not only the lower tiers of production, but also some of the R&D and capital intensive activities) in foreign countries, with no room for a local district revival.