How the replica of the Italy’s Northeast Industrial District model failed in Timisoara (Romania)

: Italy’s industrial districts of the Northeast are oen cited as beings models of success due to their economic organisation, thus they have almost become an archetypal myth. One of the reasons for the Northeastern ﬁrms’ success is proximity (both geographical and relational), particularly with regard to the creation and development of the districts and their innovative and competitive capacity on a global scale. Our research compares the structures of proximity in Montebelluna, an industrial district in Northeast Italy, with Timisoara (in Romania), which has been one of the more favoured areas of delocalisation for Veneto ﬁrms. is comparison made it possible to focus on the non-economic categories (such as trust, face-to-face relations, cooperative attitudes, local context…) that have contributed to a large part of their success, although, in reality, they are oen overvalued. Above all, the analysis reveals how these Italian industrial districts have concealed an unequal distribution of power, not only inside the ﬁrms and between the ﬁrms, but also between the ﬁrms and their delocalised territories. Today the “Northeast” model is questioned and debated; however a rethinking of regional dynamics is only possible if these hidden dimensions of district development are taken into consideration.


Introduction
According to the literature, the success of the industrial district (ID) model relies mainly on the virtuous dynamics of proximity that facilitated innovation processes: trust and reciprocity based on familiarity, faceto-face relations, cooperative attitudes, embedded routines, habits and norms, local conventions of communication and interaction, and institutional thickness (Amin and ri, 1992;Aydalot, 1986;Bagnasco, 1977;Becattini, 1987Becattini, , 2007;;Camagni, 1991;Cooke, 1988;Morgan, 1997;Scott and Storper, 1988;Sforzi, 2002;Storper, 1997).e present paper contends that some of these non-economic categories have not been so crucial.rough an exploratory research precisely focused on the analysis of proximity configurations, this paper demonstrates that -in the cases examined-the virtuosity is more mythical than real.Qualitative field research was conducted in the north-eastern districts of Veneto, an area where the IDs model attained one of its major successes.It was a socio-productive-spatial model of such accomplishment that it became a "myth".e second case study is in Romania where many Veneto district firms have delocalised since the 1990s.
For what concerns Italy's IDs, Becattini strongly emphasises the fundamental role of the endogenous and socio-cultural factors which bring this aspect of his theory well "beyond the Marshallian externalities agglomeration theory" (Celata and Rossi, 2009, p. 390;Becattini, 1989;Capello, 2004;Agnoletti, Iommi and Lattarulo, 2015).Because of it being locally-rooted and "socially-cohesive" and marked by "diffuse intelligence" (Rullani, 2006, p. 18;Becattini, 2009), the Northeast district model was for a long time regarded as a third way between industrial giants on the one side and one-person enterprises on the other (Conti and Julien, 1991;Conti, 1997).It strongly contributes to the consolidation of the "geographic imaginary and mythology associated with the rise of the ird Italy as a socio-spatial paradigm in post-Fordism capitalism" (Celata and Rossi, 2009, p. 390).
Major attention was placed on some of the non-economic categories conventionally used to explain the success of the industrial district model.rough our fieldwork, an attempt was made to verify and provide empirical evidence to the critics.e research has a dual focus.On the one hand, the concept of proximity is used to analyse the socio-spatialproductive nature of the district model.Proximity is an inevitable key for interpreting the district model.Moreover, the dynamics of proximity is reflected in the power relationships between actors and between actors and territories (Oinas, 1999;Rallet and Torre, 2000;Boschma, 2005;Torre and Rallet, 2005).On the other, the proximity of the original industrial district firm is compared with the one that was delocalised.e goal was to understand the specific nature -the depth and strength-of the building blocks of the districts: the non-economic factors.In this regard, the analysis presented here is innovative for two reasons.
First of all, it does not consider only the places where the district model originated or only the places where the model delocalised.e paper focuses on the evolution of the proximities that had been built in the two places and between the two places, and on the observations that such comparisons stimulate concerning the non-economic factors.e new situations can be used as instruments for the dismantling of the "myth" of the Italian Northeast productive model (Stella, 2000, p. 26), in addition to an analysis of the failure of those very non-economic factors that should have been a solid guarantee of this model.Hadjimichalis's (2006a;2006b) criticism of the key role of non-economic factors -cultural, social, institutional, territorial-in the development and success of the industrial district model provides the starting point for this analysis.What Hadjimichalis clearly criticises is the role of "autonomous forces in shaping development" (2006a, p. 693) of the non-economic factors cited.He refuses "to accept its prioritization" (2006a, p. 699) with respect to other factors such as, for example, economic ones.Hadjimichalis (2006aHadjimichalis ( , 2006b) ) refers to the IDs in general and focuses on the ird Italy and the Veneto districts in his studies.He stresses the importance of empirical research in order to avoid the risk of "poetic romantic views" (2006a, p. 696) when speaking about the success of the IDs models.Following this same line of thought and through our fieldwork confirms the fact that the "myth" of the industrial district is based on a "strong" representation that conceals complex spatial relations.Over time, the Northeast's echoes and fame shaped the north-eastern people's attitudes and behaviour.As researchers in Veneto, we have also been "victims" of this trap.For this reason, the fieldwork done outside of the Northeast was determinant in order to subvert our awareness about the "expected way" to observe the industrial district spirit displayed in the delocalised spaces.It was predictable to find those fundamental non-economic factors within the socioeconomic model naturally reproduced abroad, and, of course, well rooted internally.But, deep doubts surfaced during the analysis of the new proximities that evolved in the delocalisation process.Reciprocal trust between entrepreneurs was non-existent; there was no exchange of knowledge; and they did not believe that interpersonal relations could create mutual or personal benefits.In other words, all the aspects that characterised the success of the north-eastern districts were absent: the only thing that mattered was the unbounded, scalar spatial systems where local and global economic dynamics met and clashed.Playing with the dynamics of proximity, evidence of a new reality was acquired: because our fieldwork placed some distance between us and that of the successful Northeast model, it was possible to observe first-hand the inconsistency of such a socio-productive model.It was necessary to consider this "situated production of knowledge" (Rose, 1997) (the delocalised places) not as a way to confirm the data, but rather as a means of highlighting what the available data was not able to show (Leslie, 1999;Baxter and Eyles, 1997).is does not deny the role that non-economic factors played in district development, nor does it deny the significance that it had for the districts themselves in the economic growth of Italy's Northeast.Yet by observing how the Veneto entrepreneurs operated in the areas of delocalization, we gained a better understanding of how much the non-economic factors hid power dynamics.Our thesis argues that within the context of the "northeastern model" these power relations were, on the one hand, less evident thanks to the redistribution of wealth provided by its initial success, and, on the other, they were still bound by strong social and cultural ties which held back excesses or better, at least in some cases, hid them effectively.In the areas of delocalization, the great disparity of power between the actors involved and the lack of "social control", combined with the loss of public visibility of what was happening "out of sight", exalted the most difficult, sometimes cruel, dimensions that remained unmentioned or hidden in their homeland.
is article focuses on one of the places that symbolically represents the early establishment and subsequent crises of the industrial districts.is is the so-called Sportsystem (clothes and footwear) of Montebelluna (in Treviso) that is not only in its field the district leader at national level, but also plays a decisive role internationally.It is located close to the centre of the heart of the Northeast: the triangle of Treviso, Vicenza, and Padua (see fig. 1).e Northeast model is better known as the Veneto model.e main city of the industrial district is Montebelluna whose origin dates back to a guild of shoemakers and cobblers ("calegheri" and "zavattieri" in the local dialect) that had already published their first charter by 1271 (Durante, 2006, p. 3).In the past, the district used to be described as "the liveliest production district in Italy": 400 firms and about 8,000 workers.By the end of the 1970s, Montebelluna was one of the richest municipalities in Italy thanks to its ski and aer-ski boots production.If at the close of the 1980s the district was responsible for 75% of the world production of ski boots, aer-ski boots and footwear for cross-country skiing, today ski footwear only accounts for 10% of the district's income.As Aldo Durante, the former Montebelluna Shoes Museum director stated: "Montebelluna has disappeared from the map as a production district".
e decline continued into the next century.By 2006 almost 29% of the firms of the Montebelluna district had delocalised.Among these outsourcing firms, 61% declared they had their production units in Romania, and more than 30% in China (Durante, 2009).For this reason, the decision was made to analyse delocalisation in Timisoara.Many firms from the Northeast, particularly from Montebelluna, delocalised to Timisoara: it gave the impression that they were indeed trying to recreate the district abroad.In 2008, fiy-nine per cent of the firms of the Sportsystem district of Montebelluna delocalised (Durante, 2009).Of particular significance, the former Montebelluna Shoes Museum director states in 2013: Our district is living difficult moments and has failed to find the right collaborations that could have allowed it to face the crisis.Every company has its own story, everyone goes on by themselves and nowadays the district counts for nothing (La Tribuna di Treviso, 2013).
People from Veneto and Romanians have been connected for a long time.In the mid-nineteenth century Romania was referred to as "the Oriental Latin sister", "another Italy", "Italy's younger sister" (Scagno, 2008, p. 5-13).e Romanian language, which is not very difficult to understand for Italian speakers, was one of the main reasons why the Northeast chose to outsource to Romania.Timisoara could also offer useful know-how: the first footwear firm (Fabrica Turul) opened in 1900 and it eventually became one the largest in Europe.Such expertise became stronger during the Communist era.Romanian national factories worked for Western companies throughout the 1970s and 1980s (Surubaru, 2008).Infrastructure also plays a key role; it takes only eight hours to drive from Treviso to Timisoara.Production can thus be directly controlled without being away from "home" for too long and products can be ready for market in few days.All of these factors contributed to the earlier decision for delocalisation in Timisoara, however, it is clear that the main reason behind the choice was that it offered economic advantages and filled a missing gap: producing in Romania was cheaper and more competitive.is further explains why in 2007, when Romania became a member of the European Community, many Italian firms le the country.
e main drivers of delocalization were, on the one hand, the redefinition of global value chains which redistributed the roles of each country in production activities, and, on the other, the impact of the financial crisis which contracted the markets and reduced the financing opportunities of companies, forcing them to move towards areas with competitive and/or commercial advantages.

Research Methodology
is research is mostly based on fieldwork and qualitative analysis (participating observation, shadowing, 123 in-depth interviews 5 , collection of life stories, semi-structured questionnaires).e qualitative empirical research not only intended to listen to or giving voice, even if, according to Wilton (1999) it is not oen so easy collecting some voices (small entrepreneurs or employees in this case).
Different teams within the research group conducted the work: three Ph.D. students, four researchers and three professors.e interviews took place in Veneto region (Montebelluna district) and in several areas in Timisoara Department from 2007 to 2012 during different fieldwork periods ranging from fieen days to eight months.
Interview questions concentrated on interpersonal relations: among entrepreneurs and between entrepreneurs and employees; between entrepreneurs and official institution and unions; feelings regarding how entrepreneurs live as expatriates in the delocalised spaces; on production chains and market flows; on labour market features and the relations between small and large firms; and in the end, on types of development "transferred" to different Romanian locations.
e innovative aspect of the analysis further became evident because the team's field research required extensive qualitative research on the micro, small, and medium-sized firms whose data are not usually considered in official surveys.In some cases, the area was also "physically" explored in search of the firms.Some companies concealed themselves and changed their names.Many of them were hidden in the countryside.For example, in Timisoara the administrative offices and governmental bodies representing Italy did not have precise information and did not know exactly how many Italian entrepreneurs were in the area.Unfortunately, data collected previously with fieldwork did not help much as they were not updated and mainly referred to just the last few years.In many cases, the sectors surveyed were undifferentiated and oen there was no correspondence with reality: non-existent, bankrupt or closed firms; 'fake' firms that were only opened in order to buy some land; and non-existent or non-corresponding addresses.Indeed, it appears that all of these conditions were a frequent occurrence.It was also discovered that some firms split into two different companies in order to have tax reductions.Moreover, Italian firms were reluctant to declare the geographical region they came from or the name of their firm in Italy in order either to avoid reporting their taxes to the Italian revenue service, or to be able to continue to be qualified as "Made in Italy".To complicate identification of firms even more was the fact that some firms took advantage of tax reductions exclusively reserved for off-shoring companies.Since reductions are only applicable for the first ten years, many firms pretended to close down, only to immediately re-open for business again under a new name.All of these examples cited above underline the importance of the qualitative approach which stresses the importance of observing entrepreneurial dynamics through lived experiences, assumptions and codes of behaviour (Crang, 2000;Agnes, 2000;Oinas, 1999).
Two methodological techniques were used.e first is individualised comparison where "the focus is on a particular case study, exploring its specific characteristics.e emphasis is on local details" (Ward, 2010, p. 474).On the other hand, work also concentrated on the variationfinding comparison which seeks to "establish a principle of variation in the character or intensity of a phenomenon by examining systematic differences amongst instances" (Tilly, 1984, p. 83) to understand why difference or similarity persists.Identifying empirical regularities, however, does not necessarily establish cause-effect relations or define counter-theories.e perspective was that of a relational comparative approach "that refuses to measure cases against a universal yardstick", but observes: How they are formed in relation [to] one another.In this conception, particularities or specificities arise through interrelations between objects, events, places and identities; and it is through clarifying how these relations are produced and changed in practice that close study of a particular part can illuminate the whole.(Hart, 2002, p. 14-15).e paper will proceed to describe the "crumbling" of the Northeast IDs model.A focus will be placed on the present situation of the model, followed by an analysis of some of the proximities designs that were observed in the Northeast industrial district and in the areas where they delocalised.In particular, a comparison will be made regarding geographical, social, institutional, cognitive and organisational configurations of proximity between Montebelluna and Timisoara.However, cracks in "the entrepreneurial success" began to show back in the 1990s and a domino effect soon affected both the "clusters of enterprises" and the apparently well-rooted "socio-territorial project of the district" (Corò and Micelli, 2006).

e Northeast IDs: the crumbling of an expectation
Undoubtedly, the pressure caused by the reconfiguration of value chains on a global level has led to the loss of some fundamental competitive elements that, up to that time, had fueled the success of the industrial districts.One of the expressions of such a crisis has been delocalisation.By the end of 2007, the phenomenon involved 30% of the companies based in Veneto and in Friuli Venezia Giulia (Marini, 2013).In 2013, 44.5% of a sample of companies from Veneto were active abroad and 31.5% of the companies whose customers delocalised had to reduce their staff.A further 46.8% also suffered a significant decline in their revenue, while a third decided to relocate their production to new areas (Marini, 2013).
e official reasons why the model failed have been mentioned many times.e crisis further deteriorated because of the financial disorientation that resulted from the suspension of the frequent devaluations of the lira, which had constituted a crucial comparative advantage on the market, and by the quality of Northeast products, which could be easily imitated and sold at a cheaper price by newly industrialised countries (Fiorentini, Tattara and Volpe, 2007).e competitiveness of production reached a negative peak in 2003 when Chinese competition reduced the prices of footwear by 15%.e same happened to the textile and clothing industry in 2005.e end of the "Accordo Multifibre" indeed liberalised the import market.e ensuing downturn and decision to move in search of -increasingly narrowing -margins of competitiveness are seen as inevitable (Crestanello, 2008;Turato, 2008).
If the constitutive elements of the "myth" and their consistencies are very well known and shared (Becattini, 1989(Becattini, , 2016;;Putnam, 1993;Storper, 1997;Morgan, 1997;Porter, 1998;Amin and Wilkinson, 1999;Dei Ottati, 1994;Sforzi, 2002;Celata and Rossi, 2009), a conviction that is less recognised and shared is that there are, however, many partial truths and important missing issues in the discourse on Italian industrial districts and some important post-2000 developments seem to contradict the dominant mythology around them.Among these developments -mergers and acquisitions -de-localisation (…) seems to mark a new phase in their history, questioning the celebrated characteristics of the past.(Hadjimichalis, 2006b, p. 83).
For these reasons, Hadjimichalis's conclusions are that "If there is a 'third way' or any other alternative to economic and social development, this cannot come from universally applicable academic models and definitely not from the constructed myths on ird Italy" (2006a, p. 700).What kind of mythology was Hadjimichalis referring to? e criticism of the geographer stressed some issues such as power and inequalities within IDs, the limitation of networking, what co-operation, reciprocity and 'social capital' really mean, the informal economy, gender and ethnicity, the generational gap, wages and working conditions, the role of the state and uneven development -to mention but a few-remain in the dark.(Hadjimichalis, 2006b, p. 83).
It has been 50 years since the development of the Northeast industrial district model and 20 years from when its dissolution started -despite the well-known criticisms that questioned the role of some non-economic factors as criterion to its success (Hadjimichalis, 2006a;2006b;Bourdieu, 2005;Blim, 1990;Paci, 1992;Mingione, 1998;Sayer, 2000;Fine, 2001;Koniordos, 2005;Bagnasco, 1994;Mouritsen, 2003)-only recently have some scholars, local actors and stakeholders begun to strongly admit that the causes of the crisis and failures were due not only to the economical global crisis aer 2008, that was very burdensome, but also to internal inconsistencies.
While still recognising internal fragilities, some other scholars support the fact that past conditions were not suitable to tackle the new challenges brought by the globalisation and the economic crisis (Rabelotti, Carabelli and Hirsch, 2009;Coltorti, 2013;Omiccioli, 2013).
By contrast, according to the President of the Regional Industrial Association, the Northeast nowadays needs "a bath of humbleness" and "it is no longer possible to continue thinking to be better than "others" (Peghin, 2016, p. 9).e Director of the Northeast Foundation (a research centre created in 2000 with the aim of studying "the new socio-economic miracle" known as Northeast industrial model) strongly points out: the intrinsic limits of this economic system in setting up and implementing adequate institutions and territorial features able to face the challenges of the global economy.e crisis of two of the major banks (Banca Popolare di Vicenza and Veneto Banca) beside a number of small banks has provided great evidence of the inability of the Northeast territorial governance and of the inadequateness of the ruling class.(Micelli, 2016, p. 20).
e crisis of the Veneto banking system is not an isolated episode, rather it is the result of years of failures concerned with institutions, economics, research policies and infrastructural projects (the "Mose" project, a flood barrier to protect Venice from floods, and the High Speed Railways Project or the Pedemontana highway are cases in point).What is particularly significant regarding the bank crisis is that the consequences directly affected family savings (the losses amounted to 10 billion euro in a very circumscribed space and involved 180,000 savers and local firms).In effect, for over twenty years the leaders of the two banks significantly affected the economic development of the territory and their authority was never questioned.e executives were local figures from trade associations, chambers of commerce, universities and well-established business contexts who clearly abused their role (Micelli, 2016).Even the former Governor of the Veneto, Giancarlo Galan, author of the book "I am the Northeast", was convicted in various legal proceedings, together with other politicians and administrators.

Geographical Proximity
When discussing geographical proximity, "physical" distance -i.e.how close or far places are from each other-may be relative.What counts is "how" such distance can be perceived in order to interpret the relational geographies that result from it (Boschma, 2005).In the Northeast, however, it seems that physical distance should have played an important role: small and medium-sized firms (especially in manufacturing) were built next to, above, and sometimes inside homes and near one another (an example of a very close geographical proximity).In this context closeness should have been an essential condition for exchange of information and knowledge, both for experimentation in the production process and for rapid testing of products.However, in the 1990s, these characteristics of very close proximity began to collapse when the distance between local and global rapidly decreased.In a short time, new proximities emerged as the district was taken over by new actors, larger companies and multinational corporations "in search of those context-specific competences and local information networks that are real intangible assets difficult to reproduce outside the area" (Corò, Gurisatti and Rossi, 1998, p. 132).According to Hadjmichalis (2006b), "the processes of capital concentration via mergers, acquisition and formation of vertical industrial groups (…) seem to be a major turning point" (p.90).ere was no way out for small-scale firms.ey lacked the expertise on how to move so easily in these new dimensions.For many artisans there were only two possibilities: delocalising or closing down their small businesses.e few that remained, if they wanted to stay in business, were physically and economically bound to "orbit" around the bigger firms to take orders, consign goods, and whatever work they could find, yet only as suppliers or sub-suppliers.As one of the entrepreneurs interviewed explained during fieldwork in Montebelluna: "Working as third party manufacturers somehow seems to imply losing entrepreneurial autonomy and the sense of enterprise as an adventure in which people can prove themselves and their skills.It is a kind of failure of the original plan" (Pasquato, 2010, p. 224).Consequently, geographical proximity assumed the form of territorial lock in (Amin and Wilkinson, 1999;Torre and Rallet, 2005;Hadjimichalis, 2006a;Boschma, 2005).

Social and Institutional Proximity
e concept of social proximity is developed in relation to the level of embeddedness of economic actions, that is, in order to understand how and how much economic actors, as with companies in this case, depend on the specific (and evolving) social contexts in which they are located (Polanyi, 1967;Granovetter, 1973Granovetter, , 1983)).Social proximity is first of all connected to the micro-social context, i.e. the relationships among individuals inside a small group ("strong ties").According to Granovetter (1973), analysing these interpersonal relationships can help us to understand relations at the macro level where proximity dynamics become "institutional" (Boschma, 2005).However, the actors' multistability and their opportunities to develop depend on their ability to create "weak ties", i.e. to build bridges connecting them with the outside world and with what is different from their own small group or family (Granovetter, 1973).Socially speaking, Veneto's entrepreneurship is recognized as a symbol of "strong ties".It is shaped around the structure of the traditional, conservative, intensely Catholic, extended family and has been defined as "the largest part of the Italian vestry" (Covre, 2006, p. 81) whose common cultural habits include social conformism, the spirit of obedience and attention to appearance, something that could be expected from the parochial countryside of the "white" Veneto (Brusco, 1994;Rullani, 2006).
In the beginning, the entrepreneurs-to-be of the Northeast model started out as artisans and/or "metalmezzadri", i.e. tenant farmers who supplemented their income by working in factories (Roverato, 1984;Ginsborg, 1989;Corò, et al., 1998;Diamanti, 2006).e stereotype of the self-made man was an emblem of this essentially male entrepreneurship.e human-family resources were the locomotive of the enterprise and its main investment: they were cheap, flexible and reliable (as the tie between family members was both affective and productive).Family, ownership and control of the enterprise all became one and the same and family rules became the rules of the enterprise (Hadijmichalis, 2006a).Such supposed "omnipotence" is the reason why the district's entrepreneurs did not see the challenges posed by the new globalised context as a "common destiny".e family-firm believed it had everything it needed within itself (as it was in the beginning), and so it consequently underestimated the value of the changes that needed to be put into place by the narrowing distance between local and global.e former family "strong ties" became a safe haven and a trap at the same time.
With respect to institutional proximity, local governance had been unable to handle the explosion of micro industrialisation and industrial districts and, in the same way, it was unable to interact with the localglobal dialectic or to protect weaker actors who were the small artisans that had actually created the Northeast model (Messina, 2008).is is why many firms did not identify with the districts.In the words of an entrepreneur in Montebelluna: "e interests that the districts pursue are those of the medium and large companies that have ruled over the territory for generations" (Pasquato, 2010, p. 255).In 2006, the report made by Socio-Economic Observatory of the Sportsystem district of Montebelluna (OSEM) surveyed 365 firms: only 150 had been formally registered in the Sportsystem District, which had only been officially institutionalized in 2003 (Durante, 2006).e latest data collected from these firms dates back to 2008.In 2009 the report stopped being funded.In 2013 the Shoes Museum and the Sportsystem Foundation of Montebelluna, "witnesses" and "symbols" of the historical past of the district, avoided closure by being absorbed by the Montebelluna Culture Institution, a state entity.e local territory that created it did not have the resources necessary to guarantee its survival or its role.Districts in other regions organized differently.In Tuscany or in Emilia Romagna, where inter-firm cooperation and interdependence were "strongly supported by the local and the regional government (…), has created the institutional thickness on whose basis industrial districts have grown up and flourished in the region" (Celata and Rossi, 2009, p. 390).In Veneto, the present bank crisis demonstrates how institutional proximity cannot only be "a matter of voluntary civic engagement, without taking into account the inherent conflicting parameters characterizing politics" (Hadjmichalis, 2006a, p. 690).

Cognitive and Organisational Proximity
e main reference for cognitive proximity is connectedness, i.e. access to information sources.However, for actual knowledge transfer and interactive learning to take place, being connected is not enough.Receptivity is also needed (Antonelli, 2000).Torre and Rallet (2005) define organisational proximity as "the ability of an organisation to make its members interact" (p.49).Two aspects need to be considered at the same time: the stabilisation of internal routine practices and the resulting dynamics thanks to inside-outside relations that allow people to learn new things.In the areas where the Northeast model developed, individually acquired crasmanship -whose most precious material and immaterial secrets were held within the family-firm system-was consolidated through reciprocal "monitoring and comparison" (Bathelt, Malmberg and Maskell, 2004, p. 36) of knowledge and experiences: the so-called "local buzz", context-specific or tacit knowledge (Polanyi, 1967;Becattini and Rullani, 1993;Bathelt, et al., 2004).However, when mainly multinationals and some local firms -the strongest, best-organised, and most alert to global economic changes-understood the importance of reorganising the knowledge that was fragmented among many actors, they absorbed artisan know-how, put together certain specifics and turned them into technological innovation by investing in research.As a result, small entrepreneurs completely lost control of the organisational, productive and cognitive processes.
By buying local brands and imposing their production logics and mass marketing, large companies dealt a fatal blow to local small producers that became subordinate to large companies from several points of view: cognitively (research and innovation were carried out by the large companies and the small ones could only imitate) (Corò, et al., 1998;Rullani, 2006); entrepreneurially, in that small entrepreneurs found it more difficult than larger ones to fill the gap between available (contextspecific) and increasable (codified) knowledge and which, consequently, decreased the possibilities to acquire innovative technology and enter new markets; and economically (only the strongest companies could invest in technology while the others, with no access to credit, fell behind).
Today the Montebelluna district is made up of small firms that have delocalised 90% of their production abroad.As a consequence, the internationalised companies that are in the territory of Montebelluna produce for the most part an immaterial economy (Rullani, 2006).e added value of their physical proximity concerns what districts used to be in the past, such as in heritage marketing (Toninelli, 2006;Montemaggi and Severino, 2007), but not of the present (Pasquato, 2010).Local productivity and know-how is being lost and, consequently, firms are headed towards an increasingly a-territorial cognitive proximity.
Furthermore, the research proved that the very non-economic factors that were considered the strengths of the district model were not actually deeply rooted.Interdependence between production organisations had not created symmetrical power relations, so that some firms were dependent rather that inter-dependent on others.e model had grown under the influence of power mechanisms that had been strengthened by layers of socio-cultural stereotypes.e family model, which eventually became the business model, was built on a division of roles based on gender and social classes and a religious culture whose main values were obedience, sacrifice and respectability.
ese first attempts at self-criticism that are now taking place in Veneto are more dramatically demonstrated when they are confronted with the socio-productive relations within the delocalised spaces.

Geographical Proximity
New jargon has evolved with delocalisation such as "Trevisoara" (short for Treviso and Timisoara) or "Timisoara is like Veneto" (Redini, 2007).Western Romania is even considered the fourth region of the Northeast (Sarcina, 1997;Iordache, 2004;Rullani, 2006) and Timisoara the eighth province of Veneto.
However, in regard to their relationship, the distance between the firms in the Northeast and the Timisoara territories is, in fact, enormous.
Firms and plants are "islands" in which entrepreneurs move.e Northeast is always closer than the space outside the gates of the factory in Timisoara.Different production managers and technicians of delocalised companies frequently stated: "I have worked here in Timisoara for three years (…).It takes only a one-hour flight to be home.I can go home if there is a problem"."So you don't live here, do you?""No, I don't live here, I live in Italy.I usually leave on ursday and go back on Sunday" (Interview, 03.03.2008).

Social and Institutional Proximity
Consistent with their individualistic business style, for many years Veneto entrepreneurs delocalised on their own.ey decided where and how to start and develop their business without any assistance (Corò, et al., 1998).Many of them felt humiliated when their businesses failed (Scroccaro, 2012).Bankruptcy generated a sense of "shame"; the pain of having to close down the "family" business is still very much alive in the entrepreneurs' interviews.Even today, "shame" is one of the main causes of suicide among Veneto entrepreneurs.A technician working in a delocalised firm says: "My family has been making shoes for three generations, I was born making shoes, but with me everything finishes.I went bankrupt.Twenty years have passed but in Italy even today if you go bankrupt you can starve to death" (Interview, 11.03.2008).Entrepreneurs clearly do not have that background of a strong, effective social proximity that was one of the main declared values of the industrial districts.ose working in the same sector oen did not talk to each other and obviously did not exchange information to work as a team.Many of them even ended up "stealing" workers by offering a few euros more than their Italian neighbours, especially when demand for labour was high and deadlines tight.Such practices caused several bankruptcies because of the shortage of workers and the impossibility of producing in the due time (Scroccaro, 2012).Veneto entrepreneurs had little to do with one another, and they did not want to develop relations with each other.Of course, close relationships with Romanian workers were also normally avoided.Exploitation of the workforce, especially women, in the areas of delocalisation could be described as "primitive (or bloody) taylorism" (Lipietz, 1997, p. 10).During this first round of delocalisation, entrepreneurs confessed that "workers were kept in inhuman or pseudo-human conditions, in the cold, with little light" (Interview, 29.01.2008).Compared to that, the conditions of the Turul Factory (in Timisoara) were much better.When the plant was established there were lamplights, several canteens and kitchens where workers could heat up their food (Patac and Cazan, 2001).e idea of "working time" is another appropriate parameter that further confirms the lack of social proximity between Romanians and Italians.e entrepreneurs from Veneto believe in a labour intensive model that produces a culture and a "cult" of tireless work, further characterized by an "inflexible temporal flexibility".e local workers only chance to challenge this model was to boycott it."Delays", frequent breaks (for example to use the toilette), malfunctioning of machines, the of shoes or primary materials, worker absenteeism and the continual turnover of workers from one firm to another, are typical examples of some attempts to redress the balance of power in the relations involved, thus slowing down production and creating problems for the company and the owners.
Passing from the interpersonal relational aspects at the micro-social level to the institutional relations or macro-level in Timisoara, as had occurred in Veneto, the organisations of entrepreneurs and formal political and economic institutions intervened only aerwards in an attempt to organise, consolidate and, in some cases, justify what the entrepreneurs had already realised autonomously on the field.When Italian entrepreneurs arrived in Romania for the first time in the 1990s, their only point of reference was the Italian Embassy in Bucharest and the Italian-Romanian Bank which was founded in 1980: the first joint venture between Italy and Romania in the banking sector.In 2000 Veneto Banca (one of the present investigated banks) assumed control.In 1994, some representatives of Italian entrepreneurs in Romania established the Italian Chamber of Commerce in Bucharest.All these institutions were concentrated in the capital of Romania, while the majority of the small and medium Italian companies were located in the west of the country.Only ten years aer the beginning of delocalisation, and only when medium and large companies arrived and put into place production and logistics platforms, were some functional bodies set up to help foreign investors.It was not until 2000, aer the establishment of the Consulate of Italy in Timisoara, that three institutional organisations were established in the western Romanian region: the ICE office, the Italian Trade Promotion Agency, a governmental organisation that promotes the internationalization of Italian companies, in line with the strategies of the Ministry for Economic Development; "Unimpresa Romania", an entrepreneurial organisation, now called "Confindustria Romania", which supports and defends its members when they have to deal with the Romanian Government and Romanian labour unions; and "Antenna Veneto" established in 2002 in Timisoara by the Foreign Trade Centre of the Venetian Chambers of Commerce, now "Veneto Promozione".Due to the fact that their services oen overlapped, the entrepreneurs oen became disoriented and suspicious about the services that were offered (Scroccaro, 2012).Consequently, most of the entrepreneurs who delocalised have never contacted Italian authorities abroad and have no intention of doing so.For that matter, as Diamanti (2006) has written, a well-informed expert of the Northeast territory, that it is indeed known to be "far from the State and suspicious of everything that is public" (p.168).Even today, not only for this reason, some of these institutions, such as the ICE offices and the Consulate of Italy in Timisoara, have since closed down leaving it once again to the entrepreneurs to manage the difficult evolution of delocalisation in times of a global economic crisis.

Cognitive and Organisational Proximity
e quality of social and geographical proximity obviously influences cognitive proximity even if geographical proximity does not necessarily enhance cognitive proximity (Boschma, 2005;Torre and Rallet, 2005).To shed light on this aspect, the entrepreneurs were asked: "How are relationships with other firms?".e most frequent answers were all practically the same: no collaboration among the companies.As time went on, some codified knowledge certainly passed from the firms' technicians to the workers, but, as a rule, there were rigid controls on the workers to make sure that they only acquired the skills that were necessary to perform their specific tasks.Italian technicians followed this rule very strictly, particularly since aer having accepted to work far from home they were afraid to lose their jobs to Romanian technicians.
In a more advanced stage of delocalisation, other segments of production were outsourced, such as finishing and then, more recently, innovation and research: 29% of the firms interviewed transferred to Timisoara mainly for the construction of models and 18% for design and marketing (Scroccaro, 2012).e spill-over effect that resulted from this was the first step towards the emergence of Romanian entrepreneurs.In some cases, locals started to run satellite branches or firms delocalised to neighbouring countries but never where the most valuable knowledge was concentrated.An asymmetrical distribution of power was still very visible.e situation mirrored what had happened in Veneto in the districts between medium/large companies and small local firms.
Also in regard to organisational proximity, essentially two configurations also mirrored what used to happen in the north-eastern districts.e first is individual delocalisation, in tune with the northeastern entrepreneurs' isolated spirit (limited to family).As mentioned earlier, for most of the enterprises delocalisation happened in an individualistic and uncoordinated way that can at best be defined as a process of imitation.ese enterprises outsourced a production segment but were oen sub-suppliers for other firms.e second configuration of organisational proximity occurs when a strong actor, such as a leading firm, delocalises and conditions its suppliers (third-party manufacturing, supplying, and sub-supplying, [Dicken, 2011]) to follow it in delocalizing (Crestanello, 2008;Pitingaro and Sartor, 2008).ese weak actors are drawn by the "weight" of the medium/large firm: they foresee more certainty for their businesses in its network of relations.It has already been highlighted how fragile this situation of dependence could be (Andreff, Andreff and Boudier-Bensebaa, 2001;Bioteau, Boulineau, Cristescu and Michalon, 2008;Crestanello and Tattara, 2010;Kurko, 2009;Labrianidis, 2008).
To conclude, the same values described above for the Northeast model drove the organisational proximity of the delocalised enterprises: the owner is the only source of authority.

Discussion
As described in the previous sections, some configurations of proximity that have helped to create the success of the NE, reveal significant contradictions.Some of the values of the district model, which have been listed by Putnam (1993) as "trust, norms and networks that facilitate cooperation for mutual benefit" (p.167) were not so easy to recognize.A possible explanation lies in Hadjmichalis's (2006a) words: Social capital, reciprocity and trust have become very poetic, romantic views of human economic cooperation in those studies of successful regions, fostering an image of altruism and self-sacrifice, as opposed to opportunism.Remarkably, the trust theorists fail to note how trust depends on trustworthiness, which in turn is oen more a result of submission to domination than a free moral choice.(p.696).
ese aspects are difficult to detect in the territories of the districts because it is not immediately possible to quantify and, therefore, objectify them.ey are the product of long-established power relations whose mechanisms are oen difficult to grasp even by the actors that are involved.ere are power asymmetries that are oen perceived as "normal" and are therefore not questioned.
Moreover, it is shown how these contradictions emerge more clearly when observed within the delocalised context.In fact, the qualitative study of new phenomena of proximity produced through delocalisation revealed that in Timisoara (Romania), Veneto entrepreneurs do not trust each other, and they are not interested in exchanging knowledge or aware of the fact that collective action may benefit them personally.A new perspective, which has thus far been ignored by literature, clearly emerges from the interviews conducted with local actors.e former managing director of a large north-eastern company who delocalised in Romania explains: Who were the entrepreneurs of Northeast districts?ey were former workers of manufacturing companies.ey knew how to make certain products and they started building their own tools and to work like dogs.ey used to work on Saturday and Sunday with their wives and children.ey did not write any handbook and were not able to pass their knowhow on to other people, they did not create a model.As a consequence, the companies that have delocalised here worked well as long as the entrepreneurs, or somebody managing the company on their behalf, were there.When entrepreneurs die, factories die.(Interview, 29.01.2008).
In the areas of delocalisation, many entrepreneurs agree that in Romania things work in the same way as in Treviso, at least for what is happening today, aer the impact of global competition: In the Treviso district all firms compete with one another.Everyone does their own thing.e district was created by the large companies.For everything else, everybody manages their own business, they do not have the mentality of working together and everyone searches for work without any help (especially in the footwear industry).In the delocalised district there are no horizontal relationships, only vertical relationships.Large companies move and take suppliers and services with them.Small and medium-sized firms do not collaborate.(Interview, 26.07.2007).
e research conducted by Popescu (2013) on the Italian footwear and textile-garments producers in Timisoara shows similar results.According to her study, medium-sized and big Italian firms are poorly embedded in the territory, either because they exploit lower labour cost or take advantage of the international trade opportunities.One of the consequences is that each company promotes its own vision rather than share a common vision of the cluster.e companies do not show any interest in constructing an identity and creating a sense of belonging to the same business community.Furthermore, they cooperate and form business networks but they do not involve in social processes and networks.ey are weakly related to the educational and researchdevelopment infrastructure and the technological knowledge they have does not nourish the local economy.e innovative and knowledge management capacity at company level is at its lowest.(Popescu, 2013, p. 21).
All these observations of the contexts of delocalisation helped to reveal contradictions in the myth of the district model.Geographical proximities may be almost a trap rather than an opportunity: knowledge is exchanged in a way that is disadvantageous for the weaker actors and large companies do not hesitate to dismiss local artisan know-how and to hire cheaper sub-suppliers outside of the district area or by using illegal sweatshops in the district territories.Besides, even in the areas of delocalisation, firms easily do away with the sub-suppliers they had previously dragged with them abroad (aer having declared they felt "responsible" for them) when they think it may be cheaper to delocalise again.

Conclusions
In the area of delocalisation under consideration, the characteristics of the Northeast model appear without the facade of the "domestic" cultural models.Far from the "sets of common habits, routines, established practices, rules, or laws that regulate the relations and interactions between individuals and groups" (Edquist and Johnson, 1997, p. 46), the involved actors unveiled their true colours in an individualistic fight for survival.e interviews collected demonstrate that there was no search for geographical proximity that would have promoted exchange of knowledge; no reciprocal mutual trust; nor were organisational structures put into place that could go beyond a mere economic advantage; and lastly, institutional proximity was not sought aer.e Northeast model seen through the eyes of the delocalisation process has turned out to be an unpleasant surprise.e proximities active in the district of Montebelluna were not able to reproduce abroad as expected.
Veneto IDs have suffered through several crises, yet because of their capacity for flexibility and innovation, they will, in all probability, overcome the current crisis.What is certain is that its future characteristics will be completely different from those used to construct and celebrate the archetypal mythology.e Veneto Northeast model, although it undoubtedly constituted a powerful mechanism of economic success during its phases of expansion, will never be what it was alleged to be.Nowadays, entrepreneurs and opinion makers no longer want to talk about the area as an economic model different from the rest of Italy ( VeneziePost, 2013).e Northeast Report 2016 declares: this is perhaps the first time that the Northeast is self-processing publicly.e territory has ceased to be a virtuous place par excellence.e local ruling class has not been able to take advantage of this favourable moment and merely reproduces in the local area the old national vices.It is the end of the ideology of the 'territory' (Micelli, 2016, p. 23).
Clearly, the proximities discussed in this paper can no longer be regarded as necessary pre-conditions for rethinking the new phase for social and economic ventures in the Northeast.
From 1997 to 2008, the number of firms and workers in the Sportsystem district steadily decreased.It appears that delocalisation was the indicator for the decline, and Timisoara (Romania) was the chosen destination.In 2002 there were 393 enterprises from Veneto in Timisoara and by 2005 there were 501.In 2005 they represented 30.35% of Italian firms ( Istituto nazionale Commercio Estero [ICE], 2010; Antenna Veneto, 2005).In 2005, Timisoara represented 57 footwear, textile and clothes companies from Veneto and 7% of all the Italian footwear, textile and clothes in Romania ( Antenna Veneto, 2005) (see fig. 2).

Figure 2
Figure 2 e localization of the Italian shoe factories in the province of Timisoara (Romania) Source: National Trade Register Office, Timisoara, 2008.Own elaboration.

e
Northeast model is made up of 30 districts ( Osservatorio Nazionale Distretti Italiani, 2014).In Veneto alone, there are 22 districts, 56,105 manufacturers and about 2 million people involved.Forty-five per cent of the labour population of Veneto work in the local districts ( Centro Studi Unioncamere, 2014).