Sergio Conti

University of Turin,

Italy

1. THE ITALIAN AUTOMOBILE INDUSTRY: SOME STRUCTURAL FEATURES

The Fiat group currently represents the largest Italian industrial group. A few figures will give an idea of its significance to the Italian economy: revenue represents 3.5% of GDP, employees are 4.2% of those in employment, investment is 2.1% of the total, and research and development is 13.1%. Fiat Spa is a holding company with diversified activities (from cars and industrial vehicles to agricultural machines, telecommunications, publishing and civil engineering) but fully adopted this structure only at the end of the 1970s (Mosconi and Rullani, 1978). Before this, it was organised on functional lines, with a single profit centre and real decision-making concentrated in the person of Vittorio Valletta. The passage to the holding company structure began in 1973 with the constitution of Fiat Engineering and ended in 1979 with the constitution of Fiat Auto.

Despite the diversification, however, company resources and skills are concentrated around the core business of the vehicle industry (Enrietti, 1995), to which the following Fiat sectors belong: automobiles, industrial vehicles, metallurgical products, vehicle components, industrial components, accumulators, means and systems of production. In the mid nineties these accounted for 82% of overall revenue. The automobile sector alone generated 48%. This bias towards the automobile industry deepened further during the 1980s through various acquisitions (Table 1).

Fiat Auto is the most important company in the group and effectively holds a monopoly position in the production of vehicles in Italy (excepting only a few bodywork specialists such as Pininfarina and Bertone, and a few producers of sports cars, such as Lamborghini). This dominance is the product of a history of continual takeovers and acquisitions. Without going back to the period between the two world wars, it is enough to note the acquisitions in the last twenty years: Lancia in 1968, Ferrari in 1974, Alfa Romeo in 1986, Maserati and Innocenti in 1990.

This process has blocked all attempts by foreign manufacturers to set up their own plant in Italy: the acquisition of Alfa Romeo shut out Ford and that of Innocenti stopped potential Japanese competitors. The defence of its own national territory has also meant that, despite the fact that Fiat possesses subsidiary producers and licensed manufacturers in various continents, Italy remains the production heart of Fiat's auto production system: in 1991, only 25% of total production (i.e. 638,000 cars) was manufactured by sister or licensed companies abroad.

There is a similar concentration of sales corresponding to that of production: 60% of the cars sold in Europe were sold in Italy, keeping in mind that Europe absorbs 94% of Fiat's exports. This dependence on the Italian market is the fruit of a strategic decision taken at the beginning of the 1980s when Fiat Auto was confronted with a series of critical problems: high debt and low capitalisation, delays in updating its range of models, an over-extended international presence, specialisation in low horsepower cars, rigidity in industrial relations and significant losses of market share in Italy (Volpato, 1996). The strategy chosen was thus that of a drastic concentration of its efforts in the European, and especially Italian, market. In effect, starting from 1980, Italy has absorbed between 68% and 72% of sales in Europe (against 63% in 1979) giving Fiat the opportunity to set prices and achieve a greater rate of profit than elsewhere in Europe. The dependence on the European market has at the same time reinforced the company's specialisation in the lower segments of the market, focused on smaller and cheaper cars, that is those most in demand in the domestic market.

There is an important qualification to this leading position, however, as Fiat models have high shares where there are barriers to Japanese cars: Fiat holds 21.1% of their market, against Peugeot's 17.1% and Volkswagen's 10.1%. In contrast, its position is much lower in countries where no barriers exist (only 4.8% against Fiat's European average of 14.8% (Mitsubishi Research Institute, 1990; Camuffo and Volpato, 1997).

Compared to its competitors, Fiat is distinctive in possessing one specific feature, that of being able to depend on an important series of supply companies belonging to the Group itself. The company took a strategic decision over these activities in the late 1970s (Enrietti and Fornengo, 1989). It transformed several of its own plants into independent companies. A first phase of rationalisation was followed by a second in which greater importance was given to innovation and diversification in the automobile market.

We shall now look directly at aspects of the territorial organisation of the company, going back to the fundamental bond that exists between strategic behaviour, industrial policies and the spatial dimension of development. The result that will emerge is a model of the production system within which it will be possible to reorganise the wide variety of questions which have been discussed in the previous section and to open up new discussion. The territorial approach to the problems of transformation of an industrial system is in fact a way of proceeding which is aimed, in other words, at understanding how the system "works" as a whole through the relations which link economic, social and institutional actors.

The years immediately following the second World War were crucial for the future development of Fiat. Working in a rapidly expanding market like the Italian one, without any significant import penetration and with few local competitors, Fiat focused its market strategy on small and medium-sized models, It also organised production in a way which paid little attention to R&D, planning or scientific activity. Fiat's entire strategy war directed towards quantitative expansion, both in terms of employment and productive capacity, paying little attention to its internal organisation.

These technological and product decisions were accompanied by a spatial strategy which deliberately enhanced the relative advantages of concentration and mass production. This had two main features: the concentration in the north-west of the country and in the city of Turin in particular; and concentration of production in a few large, vertically integrated, factories. In the Turin area, Fiat has in fact found, produced or has seen others produce (the public administration) a large part of those "territorial conditions" which marked the establishment of the system of mass production (Castronovo, 1971; Gabetti, 1977).

Turin is rightly seen as representing the model of the factory town, albeit with its own specific features and connotations. It comes as no surprise that even in international literature Turin has been a favoured area of study for the relationship between industry and the city in the era of the second industrial revolution (Gabert, 1964; Jalabert and Gregoris, 1987; Sallez and Schlegel, 1963).

This technical and location decision defined a company strategy which was not to change right to the late sixties. With a few exceptions and timid processes of internationalisation, Fiat's industrial investment was identified with the boundaries of the Turin agglomeration: this area not only contained the entire car production cycle (in 1968, about 1,300,000 out of 1,550,000 cars came from the Turin area), but Fiat also participated in the whole range of production based on the internal combustion engine, involving a close-knit network of small and very small supply companies, often completely dependent upon Fiat.

While at the end of the 1960s, there were over 125,000 employees working in Fiat's Turin factories, at least an equal number worked in production units which directly or indirectly were part of its network of subcontractors. In reality, it has never been possible to delimit exactly the boundaries of this network because of its composition is in a state of constant flux, especially as regards the smaller suppliers. Fiat drew on around 1,200 direct contractors, about a third of the 3,500 units (often small and very small) linked in some way to the automobile industry (Rossignolo, 1971). Under these conditions, despite the high internal integration, during the 1960s, Fiat acquired on average over 50% of its total turnover from external companies, of which about half were located in the Turin agglomeration.

As a result of the automobile industry's strategy towards indefinite output expansion and a consequent, and sometimes uncontrolled, enlargement of the production capacity, the Turin area became one of the most sectorally specialised regions in Europe, comparable to just a few other international examples such as Detroit and the West Midlands. In Turin, in contrast to these other areas, all these activities were concentrated within a single corporation, while the company's control extended deeply into the labour market and was at the centre of intense interlinking of political and economic roles (Borlenghi and Dematteis, 1982; Conti, 1986).

The first great wave of reorganisation of the Italian automobile industry - and of Fiat in particular - started in the 1970s and can be largely explained by the environmental crisis produced by the location and organisational "model" followed until that time. The rigidity of the connections between the city and the company created conditions which challenged the network of co-operation which had previously ensured successful accumulation. It developed, in fact, an entropy of the social environment which made the process of change triggered by the company increasingly difficult to foresee and control.

For Fiat, this form of production rapidly became impracticable for reasons of a social and political nature. The reorganisation of the sector thus implied a profound change in the economic and territorial development model, expressed in:

a) a profound internal reorganisation of the conglomerate, involving the transformation of Fiat from a traditionally integrated company, with a very rigid and pyramid managerial structure, into a "divisional" structure. Initially, in 1972, it was split into three operational sectors (cars, industrial vehicles and tractors, miscellaneous activities) and then in 1976 into eleven sectors, each headed by a holding company, in turn controlled by a single central holding company.

b) in constant attempts to regain control of the production process through the introduction, in the "old" Turin factories, of electronic technologies, accompanied initially by considerable falls in employment.

c) in a changed policy of industrial location, put into practice in 1970 with a two-year investment plan for the construction of nine factories in southern regions (for a total of about 17,000 employees, a figure which was to double by 1981, mainly financed by regional policy legislation in favour of the Mezzogiorno). The other element of the new locational policy involved even more substantial projects on an international level. During this period construction was completed of a car factory in the Soviet Union, and a share of Citroen was purchased. In 1971 an agreement was signed to build a new production plant in Poland, and a massive plan for investment in Latin America (mainly Brazil and Argentina) got underway. In Western Europe, most production remained in Italy and Spain (Seat) with significant exceptions in Ireland and Portugal.

The overall goal of reorganisation was to seek maximum flexibility in production methods and locations, which in turn implied operational objectives: first, to move into a new market area and, later, to structure the decentralised plants as a function of the whole system. The new organisation of production was thus not limited to the mere duplication of plant, but aimed at the decentralisation of specific stages of production to dispersed medium-sized and specialised factories which were functionally and strategically connected (Amin, 1986. The strategy was, in other words, to "lighten" the Turin area - accompanied by the standardisation of some manufacturing processes and some intermediate products without undermining the continuing technological and organisational "centrality" of Turin. which does not in itself deny the continuing technological and organisational "centrality" of Turin.

At the beginning of the 1980s Fiat Auto found itself in difficult conditions in its national market. Domestic demand for cars was slower than in other European countries (only in 1979 did Italy return to the pre-crisis sales levels of 1973, while in the rest of Europe this happened in 1976). To this must be added low productivity and difficulty in managing labour, inadequacy of the components supply industry, absence of an industrial policy and high inflation.

In order to regain operating conditions comparable to its competitors, a complex defensive strategy aimed at restructuring was designed (Balliano, 1986). Apart from a decisive cost-cutting policy (the company reduced its workforce by over 40% in seven years, from 134,621 in 1980 to 77,910 in 1986, and productivity doubled in the same period), the main elements of which were: a) plant reorganisation, b) technological strategy, c) restructuring of supply.

The sudden introduction of technology necessarily had to be accompanied by technical reorganisation: the almost immediate closure of three engineering factories was followed by the decision to allocate highly innovative investments to the South (automated and robotised manufacture and assembly in Termoli and automated vehicle assembly in Cassino).

On the whole, this reorganisation of the production structure led to a reduction in the number of active factories, to a growth in the degree of saturation in the remaining ones and, above all, to the reduction in the break-even point from the more than 1.5 million cars of the early eighties to the 1.2 million at the end of the decade (Scott, 1991, 258). In addition, the decision rather than in Piedmont has reinforced the move of the centre of production towards the South, also taking into account the closure of two plants in the North in 1992, those of Desio (Milan) and Chivasso (Turin). In any case, the central company functions (top management, research, management training, purchasing) have stayed concentrated in the North; it was only in the nineties that some segments of research began to be moved to the South.

The overall rationalisation of the system would thus seem to be going in the direction of a production structure based on new forms of "polarisation" around a few highly integrated plants: Mirafiori and Rivalta in the Turin area, Arese in Lombardy (formerly Alfa Romeo), which, with the southern factories of Cassino, Termoli, Melfi, Pratola Serra and Pomigliano d'Arco (again formerly Alfa Romeo), will constitute the key nodes in the Italian automobile production system. The expected effect is therefore that of creating a strategy of systemic integration between the various Italian regions and within the Mezzogiorno itself by creating greater uniformity in the distribution of the phases of work throughout the territory. In conclusion, more detailed analysis of the production stages in each plant lends support to these claims: as far as the stages of bodywork and sheet steel pressing are concerned, there has, over time, been a process of replacement of the plants located in the North with ones in the South. Engineering is the only activity in which the number has actually increased with the new factories in the South. It follows that in 1996 there was an equal number of engineering plants in the two parts of the country, while there were more bodywork plants in the South.

In conclusion, the progressive location of factories in the South, with the consequence that from 1994 more than 60% of Italian production come from this area, expresses "discontinuity" in Fiat's strategy. The position may be summarised as follows.

First, the policy of relocation towards the South represents a break with the past compared to Fiat's classic location policy, an organisational structure centralised and concentrated in Piedmont. Production is now scattered in factories distributed throughout many regions.

Secondly, starting in the 1980s, the southern factories, both old and new, have been the places where Fiat Auto has experimented and introduced not only process innovations, but also new models of production organisation and of industrial relations.

Thirdly, until the 1970s, Fiat's production location was polarised, with the North which constituted as an integrated system and the South specialised in only a few functions. In the 1990s, there is a single integrated model nationwide, with "sub-integration" in North and South.

Fourthly, in the context of the changes just described, continuity is represented by the company's central functions (top-level management, research, management training, purchasing) remaining concentrated in the North. In the "historical" region of Turin, there is a strengthening of the strategic metropolitan role (management functions, R&D, marketing) and the trend towards specialisation in production with a highly innovative content. This can be deduced from analysis of investment plans made public by the company for 1992-1996: investment aimed at process innovation and for maintenance of production efficiency were concentrated in Piedmont with, respectively, 36% and 52%, of the company’s total investments. A much higher investment share (about 80%) went to Piedmont for new products (Figure 1).

Technological transition was ensured by an intense process of fixed investment. The number of robots in use rose from 225 in 1980 to 2,500 in 1992; in some factories and some production segments automation was almost total.

With the gradual rise in the number of production plants in the South, this area has become increasingly important in the dynamics of the technological and organisational development of Fiat Auto. During the 1980s, with the factories in Termoli and Cassino, the Highly Automated Factory (HAF) was introduced. This marked the passage from the traditionally rigid automation to a flexible form.

The reorganisation of these plants was in fact in the direction of obtaining high flexibility, i.e. the possibility of producing more models and versions in the same factory at the same time, increasing the degree of differentiation of the various models, adapting them to the needs of specific market segments. Thus, at Cassino, flexibility, in terms of the possibility of alternating production of different models during the same day, at the end of the 1980s was almost double compared to the "old" plants in the Turin metropolitan area, where the same car models were produced. Unsurprisingly, the productivity analysis carried out in 1988 on a sample of 38 world car production plants (Krafcik, 1988) showed that Fiat had the best performance in Europe, even though there was still a gap between it and Japanese manufacturers.

With the end of the 1980s, however, the HAF model began to be questioned, as it was realised that Total Quality, Fiat Auto's strategic objective for the nineties, "cannot be the result simply of high technology, but also [must be] the fruit of intelligent and responsible human work" (Bonazzi, 1993, 77). A new technological and organisational strategy was thus devised. This was the Integrated Factory (Cerutti and Reiser, 1991; Bonazzi, 1993), in which the adjective integrated underlines the project's main feature and aim: to integrate the functions present in the factory by process.

In effect, in the traditional production structure, each Fiat plant is divided into three parts: manufacturing, which controls production and maintenance; technical services, including the planning of maintenance and technology link-ups; production services, including logistics.

With the Integrated Factory, the plants are re-organised around just two operating units: the "common services and plants" units for activities which cannot be decentralised, such as energy production, and the operational production units, which are independent from the technical and management points of view. Production units are, in turn, divided into "Production", responsible for manufacturing itself and for materials management and planning, and "Production Engineering", which ensures the effectiveness of the technical system through maintenance men and specialists and also manages the evolution of products and processes.

The basic operating structure of the production unit is the Elementary Technical Unit (ETU), which is entrusted with the government of elementary technological sub-systems, characterised by a homogeneous process and/or product (for example, in the case of engine production, the ETU manage, in addition to assembly, the production of the engine block, the driving shaft, the distributing shaft, the piston rod and the cylinder head), with the aim of optimising production processes, to improve results in terms of competitiveness, quality, costs, mix and quantity. From the organisational point of view, within the ETU the number of hierarchical levels is reduced, non-hierarchical forms of organisation of work are present, such as the "technological team" (made up of the elementary technical unit manager, the line technologist, the maintenance man and operators), and for workers, the multi-functional, regulation and process micro-management aspects are increased within a model which demands co-operation.

Drawn up at the end of the 1980s, it was not by chance that the IF project, was again first introduced experimentally in the plants of Termoli and Cassino, in which the HAF had already been introduced, because this was an organisational model closer to the IF than the traditional northern Italian plants.

A further problem was the generalised reorganisation of the supply system, given that Fiat currently has a level of vertical integration of about 45 per cent of turnover. It purchased 13 % of requirements directly abroad, another 25% from Italian branches of foreign multinationals, another 35% from by independent Italian manufacturers and 27% from elsewhere within the Fiat group.

This is perhaps the most visible aspect of the eighties and nineties, and thus explains some of the most visible transformations in the Turin production system.. It also represents an extreme case in the map of European car manufacturers. It is for this reason that it is necessary to look at this in detail.

The changes in the relationship between Fiat and components suppliers involved a great selection in their numbers and the pursuit of increasing co-operation and partnership between the two parties, attributing broader design functions to the suppliers (Camuffo and Volpato, 1997; Pulignano, 1997). Some phenomena, in particular, are capable of explaining together the fundamental processes underway.

1. Processes of selection. At the beginning of the eighties, there were about 1,200 direct Fiat suppliers. In the early years of the decade, a process of selection began based on the capacity to ensure innovation, competitive prices and reliability, with the result that in 1987 about 350 companies had already disappeared. From that year on, the fall in the number of suppliers increased notably, falling to a little less than 400 in 1997. At the same time, an intense process of concentrating on a limited number of suppliers was underway, within which a limited number of companies - 138 in 1994 - provided 90% of supplies (83 of these companies are from Piedmont).

They make up 60% of the overall number of suppliers, and provide 40% of purchases. Furthermore, they are heavily concentrated in the Province of Turin (about 90%) and represent a strong financial concentration: 75 of them (90%) belong to 57 groups, and only 8 companies, in any case characterised by fairly low levels of sales to Fiat, are independent (Table 2).

1994 1997

The economic concentration of supplies is also high: the first company covers 15% of total supplies, the first four 33%, and the first ten 51%.

For the purposes of examining the Turin vehicle cluster, the selection process that has involved the sector in the past fifteen years, is therefore of fundamental importance. Its rationale can be explained by the following factors:

1. Reduction in the number of supplier companies

2. Increase in production volumes for the remaining companies

3. Growing use of economies of scale

4. Increase in productivity

5. Reduction in costs

6. Reduction in component costs

7. Reduction in costs for the final producer.

2. Processes of deverticalisation. The process of selecting component manufacturers is closely related to the reduction in the level of vertical integration of Fiat Auto (Table 3) and thus the increase in outside purchases.

1987 1992 1997

Make inside 38 35 30

Fiat suppliers 14 18 17

Outside purchases 48 47 53

Total 100 100 100

It is necessary to specify, however, that a considerable part of the components produced by suppliers is still designed inside Fiat, which assigns contracts with its own designs, and in some cases even with its own equipment, such as dies. In fact, if we consider the division of the value of the parts designed, it can be seen how Fiat’s level of vertical integration was still very high at the start of the nineties, and has fallen decidedly in recent years. For example, the share of components designed outside Fiat for the Uno (1983) was 30%, going up to 55% for the Punto, and to 60% for the Lancia K (Table 4).

1991 92 93 94 95 96

Internal 76 70 60 50 40 30

External 24 30 40 50 60 70

The objective of selection is to reach the situation of only one supplier, or at most two, for each product line, in other words, the extension of the relationship of mono-supply: in the case of the Punto, for instance, 88% of components purchased were done so in this form of supply.

3. Complex supply. From the technological point of view, the selection is linked to the transformation of the components manufacturers left in suppliers of complex systems: a single supplier, acting as main contractor, thus unites the functions previously played by several companies. This passage also implies that the Fiat/supplier relationship is becoming increasingly co-operative, seen in the attribution to suppliers of advanced tertiary functions, such as research and development and design.

The first level suppliers were also entrusted with the task of co-ordinating the sub-contractors: all sub-contracting thus ended up converging on the companies, who had responsibility for assembly and testing before delivering the component to Fiat.

From the standpoint of the supplier, however, the situation of the sub-contractors themselves changed: these were not just small companies, without any particular design capacity, but even leading firms in their field, whose product was inserted into a system supplied to Fiat by another company. This has stimulated, and will continue to stimulate even more, companies to move towards establishing groups and towards a policy of technology and research agreements .

The size of sub-contracting (there are 14,900 sub-contractors of the first 140 Italian suppliers of Fiat Auto) and its importance in the filière make this segment of companies a field of intervention for Fiat itself from the perspective of rationalisation and cost reduction: in effect, the dispersion is very high (only 12% of sub-contractors’ sales is made through relations within the top 140 Fiat suppliers, and only another 12% concerns purchases from suppliers shared with Fiat Auto), encouraging action to concentrate sub-contracting in a smaller number of companies, to achieve economies of scale both in production and design. It is worth underlining that a "spontaneous" process of selection of sub-contractors is already underway: in the last five years, on average the first level suppliers have reduced the number of their suppliers by 23%.

To summarise, what Fiat has asked its suppliers has been: a) improvement in levels of quality, promptness and reliability; b) increase in their design capacities; c) higher capacity of co-ordination with Fiat technicians, both in times and methods; d) the consequent development of investment capacity in machinery, technology, research and development.

The final image is not dissimilar to the arrangement that its competitors have adopted: from the hierarchical filière of ten to fifteen years ago, to a structure that can be broken down into at least three levels, with the formation of a fabric of large first level suppliers (almost equal interlocutors with the final producer) to whom management of sub-contracting is delegated.

For suppliers, the need to sustain high levels of investment (machines, research and development, property, organisation, information system, training) thus becomes crucial, resulting in a process of selection on the basis of which it is it is likely that only the most financially and economically solid first level suppliers will survive. In terms of size, this tends to be translated into the marginalisation of small and medium-small companies from the direct relationship with the manufacturer and their relegation to a second level of supply.

Starting from company behaviour and strategies, we have thus highlighted some dynamics typical of the Turin system. The story is not yet finished, however. When faced with the employment crisis, the processes of restructuring and upgrading the system, the problem is to understand the more complex scenarios that are appearing on the horizon for a region of old industrialisation. It is obvious that these cannot be comprehended without understanding the technological, social and institutional legacy that has historically been dependent on the automobile and its dominant corporation. Two phenomena regarding this are of essential importance.

1. In the mid eighties, for the first time, the presence became apparent in the Turin metropolitan area of a major concentration of companies (about 200) operating in a series of activities connected to the automation of industrial processes. More recently, other companies working in robotics and electronics components have established themselves. This whole set of entrepreneurial activities goes back mainly to the seventies and eighties at the time of the introduction of numerical control and robots. The Fiat Group and its suppliers have played an important role in this evolution, not only because they represented an important area of demand, which attracted from outside purely commercial initiatives (sales representatives, commercial offices both for machine tools and above all for electronic and computer products and components like CAD stations), but also as a technological incubator, in the sense that many neo-entrepreneurs are technicians or workers who started out at Fiat or in its supplier companies.

The role of the Fiat Group has gradually been reduced, however, both because its strictly industrial activity in the Turin area has diminished, and because of the emergence of Comau, which is now Europe’s largest company in this sector.

Seen as a whole, these activities differ greatly from one another. Despite the fact that this concentration is unrivalled in Europe, its visibility is relatively restricted because of the limited size of many companies and because of their location within a metropolis where other industrial activities of often significant size are present.

Leaving aside quantitative aspects, a number of interesting considerations can be made from the qualitative point of view:

i. mechanical engineering is a major factor of Turin production;

ii. although there is good coverage of all types of machine tools, milling and grinding machines remain important;

iii. dependency on the local market (and on Fiat in particular) has diminished in favour of international markets;

iv. the solidity of mechanical and machine tool suppliers with significant international positions has increased;

v. the leadership positions held by some electronics and robotics companies has been reinforced;

vi. sub-contractors capable of covering the most varied needs have strengthened and diversified (in engineering, electronics and computers);

vii. Turin has become the Italian centre of industrial metrology, with private companies operating alongside public and university research institutions.

2. A cluster of industrial design and engineering has been established definitively, revealing itself to be fundamental in the recent dynamic of Turin’s economic system. Its roots lie in the city’s manufacturing tradition, the widespread know-how that can be traced back traditionally to vehicle and machine tool production. More in particular 104 companies of fairly varying size have been identified in the area which, effectively, have no international competitors. It is not stretching things to say that almost the entire design activity external to the style centres of the main vehicle producers happens in Turin (Bertone, Italdesign, Pininfarina, Stola, Idea). It is concentrated not only in the stage of conception, but also covers a series of practical aspects that demand certain technical capacities and close relations with production.

As far as design support activities are concerned, the quantification is rather difficult as these are often small studios or individual professionals. There is, nonetheless, a confirmation of the importance and quality of the sector in the Turin area, considering the fact that the activities connected to design (CAD, modelling, prototypes, control etc.) are effectively contained in the local Turin system, as they are run inside the companies or delegated to local suppliers.

Automobile design, in particular, is split into two levels: on the one hand, the Turin system includes the major designers that work for the world’s leading manufacturers; on the other hand, a significant number of subcontractor small and medium size designers has developed that works for the big designers and for first level Fiat component suppliers. In the 1980s and 1990s there was a considerable expansion of both levels, with the development and embedding of high level skills.

The growth of the automobile component of the cluster is closely linked to changes in the relations between the final vehicle manufacturer and its suppliers: in fact, the involvement of suppliers in the design of the component and the consequent increase in investment has meant that specific skills once developed inside the final manufacturer have spread and been embedded locally, thus laying the foundations for the independent existence of a design and engineering cluster.

One extremely delicate aspect, but of great interest, is education, which, in the Turin system, is the responsibility of three main bodies: the Architecture Faculty of the Turin Polytechnic, the School of Applied Art and Design, founded in 1978, and the European Design Institute, whose teaching centre in Turin was established in 1989. Observing the history of these teaching programmes the strong bond between Turin design and the automobile and machine tool clusters appears evident.

The situation is different for technical design staff, an activity that is done almost exclusively nowadays with CAD (Computer Aided Design) programmes: in this field, training is mainly run through vocational courses, often organised in the framework of training/employment programmes. One of the main issues for the development of the cluster in the coming years is linked to the specialisation of training courses for designers: in fact, while the education of the designers reflects the domination of the automobile cluster in Turin manufacturing, the very theoretical and generic training of technical designers does not seem to respond to the real needs of design studies.

These three clusters together - automobiles, machine tools and design - in the Turin that looks to the new century represent a full-blown strategic system of value. It is founded explicitly on localised and active conditions of competitiveness (i.e. intensely exploited by existing companies), and also on latent conditions, in other words, not exploited but potentially capable of being activated by coherent industrial policy action. In Figure 2, the colours corresponding to the three strategic clusters, and the conditions of competitiveness are found in the overlapping areas. In brief, these emerged from a survey conducted by questionnaire with over 300 companies, on which I shall not spend more time here.

1. Active conditions. Among the first, we obviously find the major automobile company (the reasons for its impact on the other systems that have formed does not need to be underlined further at this point) and highly skilled workforce, itself the expression of the manufacturing history and the engineering culture.

But there are other components of great importance for establishing a fairly integrated production system: entrepreneurial embeddedness, the presence of a diffuse fabric of small and medium size engineering companies. Other conditions are, instead, factors of specific competitiveness, i.e. key elements in the production and competitive capacity of the individual clusters (infrastructure accessibility for Vehicles, consolidated intercompany relations for Design and engineering, logistics for Machine tools).

2. Latent conditions. Among the second, are the conditions whose activation is held possible, and which would constitute a vital strategic outlook for the economy of the region when faced with future competitive challenges. However, systematic strategies of local industrial policy are required for these. Again, some factors - such as specialisation of vocational education and materials innovation - have a general value, providing support shared by all three clusters involved and to all the support companies operating in the area. Other conditions, for which spin-off effects for the entire system should not be ruled out, of course, assume specific meanings (they are cluster-oriented, in other words): support for electronics research and support for synergetic relations between enterprise and university are essential for the strengthening of the Machine tools, robotics and industrial automation cluster; services to assist product development appear vital for the Design and engineering cluster.

In the last twenty years, the functional regeneration of the regional economy, while respecting its historical legacy, has occurred in a fairly spontaneous manner, and has been insufficient to block the loss of jobs in manufacturing, and only partially successful in creating the conditions for an effective functional differentiation of Turin’s production structure. A medium to long term strategic design needs to be defined to "accompany" the changes already underway in the system, providing support for them.

The regeneration of production structures must not be approached by trying to attract what are conventionally defined as high-tech industries (biotechnologies, semiconductors, aeronautics, software etc.). Obviously, this does not mean not pursuing a technology policy. However, and this is the important aspect, it must be directed to the use and "regeneration" of technological resources historically embedded in the region’s economy.

This means that the regeneration of the regional economy cannot happen by promoting unlikely new activities, but by pursuing the relaunching of manufacturing. If the region already possesses an integrated industrial structure of varying degrees of complexity, the solutions are to be sought in the respect of (as well as support for and regeneration of) these production situations, promoting specialisation and functional differentiation together. The road to be followed, in this case, is a network strategy aimed at encouraging and supporting interaction between actors (between companies, and between companies and others).

Mosconi and Rullani, E. (1978)

Enrietti, A., Follis, and Fornengo, G. (1988)

Silva, Grillo and Prati (1982)

Volpato (1996)

Enrietti and Fornengo (1989)

Castronovo, V. (1971)

Gabetti, (1977)

Gabert, (1964)

Jalabert, G. and Gregoris (1987)

Sallez and Schleged (1963)

Rossignolo (1971)

Borlenghi, E. and Dematteis, G. (1982)

Conti, S. (1986)

Amin, A. (1986)

Balliano (1986)

Scott, A. (1991)

Krafcik (1988)

Bonazzi (1993)

Cerutti and Reiser (1991)

First International Humboldt Meeting. Buenos Aires, Argentina. November, 1999.