Cities as development hubs in Italy an analytical approach

Abstract Italy, the country where the modern cities were born, suffer from the early ’90 of past century for a lack of vision and political address in terms of adoption of its urban agenda. The public debate about cities is focused only on administrative and insti aspects; not enough attention is dedicated to the problems of the Italian cities as development hubs. As a result, the law of 2014 that introduced the formal regulation of new metropolitan authorities lies unapplied. The paper gives some evidence about the economic specialization of the most important Italian metropolitan cities and about their relative room for innovation, using an hirschmanian index.


PUBLIC INTEREST STATEMENT
In the context of globalization, the role of cities in national socio-economic systems is becoming more relevant. International value chains have of placed the most important parts of the productive processes in the more innovative cities of the advanced countries, because cities have the most of the material and immaterial capital of each country, give the chances to innovate and overcome problems and challenges of modern societies.
In this process, it becomes important to adopt economic models to describe and analyze the status and the potentialities of cities. The paper aims at offering a framework to understand the role of different businesses in urban areas, adopting the hirschmanian "network development view", showing that growth in territorial contexts come from a diversified entrepreneurial environment.
In particular, the paper shows the case of Italy, a country with a secular history of urbanization at its back, that has missed the modern trends of urban dimension of national economic growth.
At the end of the '80, it was clear in the public debate that Italy needed a formal urban agenda, and that a Minister in charge of the process could help to develop a common view on some measure that could improve the performance, on an institutional and economic level, of the major cities.
Unfortunately, that awareness was overwhelmed with the entire institutional and political system in early '90. All the public debate was addressed to the problem of corruption of the political parties, an entire leading class of the parties born with the republican Constitution after the II world war was forced to give up its political careers.
During the '90 a number of important institutional and political issues were, in a certain sense, forgotten; a new political debate raised, focused on the level of taxation, with a strong accent on an anti-institutional rhetoric.
In 1990, with the law n. 142, there was a last significant effort of the Parliament to renew the local government logic and rules. In this law, for example, some rules were introduced to lead to the building of new metropolitan authorities, to facilitate the collaboration of small municipalities, to reduce costs of local government.
Despite the value of some of these new rules, the law n. 142 of 1990 remained largely unapplied. In the following years, the raise of the new political regional party called "Lega" moved all the attention of politicians and citizens (but of scholars and cultural institutions too) to the problem of a more effective regional autonomy. The urban issue, the aim to a national urban agenda passed in a second row and were, little by little, totally neglected.
This shift of attention from urban to regional issue is in some way understandable but not acceptable. All the international literature, major practical reforms in leading European countries as Germany, Spain and recently France should have suggested more attention in the political koine. More, in the first years of the new century the birth of a proper European urban agenda, and the important moment of adoption of the brilliant document of the Commission on "Cities of tomorrow" (EU Commission, 2011) pushed even forward the attention of leading classes all around Europe.
In Italy, on the contrary, the debate was weak and politically irrelevant.
Only in recent years, the needs of public finances pushed a new attention to the urban dimension of public institutions. In 2014 a new important law was passed by the parliament on local government, after a long and bumpy parliamentary road. The new law n. 56 is a very disputed one, showing a prevalent accent on political and monetary aspects. Some notions coming from the international evidences entered the law; but in a very rhetoric and uneffective way.
The law changes the provinces from directly elected large areas authorities to second level institutions dedicated to collaboration among majors; introduces for the second time, after the failed attempt of 1990, metropolitan authorities, but without giving them a precise policy role and the correct territorial definition; tries anew to push small municipalities to get together to give citizens important services and functions.
More than two years after the adoption of Law No. 56/2014, the fundamental issue that those very norms had posed remains unsolved, or even evaded: was there a need for a "bureaucratic reform" of provinces and municipalities (at least to ensure some degree of expenditure saving)? Or was the urgency of the reform functional, that is to say the very reason of being of cities in the context of a globalised economy? In other terms, was there an urgency to overcome the straightjacket of former laws on local bodies?
The question lends itself to both answers; what is unacceptable is for it to remain circumvented.
Anywhere in the world cities have become the tow of national systems from both the economic and development point of view. 1 Not surprisingly so: cities are where the maximum material and immaterial concentration of capital of every country is located, they allow contiguity and spill-over effects that would be impossible in more rarefied anthropic contexts, they tend to attract the finest and most valuable segments of great multinationals' sophisticated chains of production worldwide (R&D, command and control, finance, creativity and design, systems of knowledge and development of know how, etc.). 2 Therefore, "global cities" (Sassen, 1991) produce by now the major part of added value and thus of wealth.
In this planetary trend Italy is in a frightful delay, namely a delay in awareness and vision that pre-exist the one in the definition of concrete policies; in other words, it seems that what lacks to the koinè of national public decision makers (and not exclusively to the battered politicians 3 ) is even the awareness of the processes under way and of the necessities of the country.
Italian cities, in an economic context where cities take on the function of leading development hubs everywhere (in the double meaning of economic growth and of the maintenance of sustainable systems of social inclusion), are to a large extent missing the opportunity of taking the new role that modernity is assigning them (Consiglio Italiano per le Scienze Sociali [CSS], 2005).
They make a humble contribution to the GDP and to the modernisation of the country: according to recent data, "Italy covers the second to last position among OECD countries for the contribution made by metropolitan areas to the overall growth of employed people in Italy between 2000 and 2012 […] the contribution made by metropolitan areas to the national GDP growth is below the OECD average". 4 It thus becomes inevitable to trace back this systemic deficit to a delay in the definition of policies aimed at creating favourable conditions to economic development in urban territorial settings, according to Hirschman's economic lesson; these missing policies comprehend the adjustment of institutional systems, in order to exploit the added value that the single territories can give to entrepreneurial systems. Not without reason, Italy is still, for the reasons already recapped, nearly the only European country that does not have a real "national urban policy agenda", 5 coherent with the one existing at the European level mentioned above (EU Commission, 2011).
It is true that the reason for this delay is to be partly traced back to the very nature of the directions of development that the country has taken. Italy is a polycentric country, with various medium-sized cities and almost a quarter of the population still residing in intermediary or marginal areas, and with many twentieth-century industries which have been established in nonurban and even peripheral areas (Dipartimento Politiche di sviluppo e coesione, 2013). But this path dependence is not sufficiently explanatory: since the end of the Seventies of the twentiethcentury the framework of what is produced, how it is produced and mostly where it is produced has changed drastically; and it is exactly as a consequence of this that cities all around the world have reacquired that role of tow of national systems that was mentioned before, exactly for this purposes, adjusting, among wide, long and ambitious redesigning programs of territorial administration, their own institutional composition. 6 It is not easy to gather from the norms whether or not the legislator was aware of all these phenomena; there are indications in both senses in the new discipline, starting from the uncertain and even contradictory design of the new metropolitan authorities (but, actually, really too similar to the pre-existent wide areas administration, the Napoleonic "province").
Perhaps for this reason the attention of both operators and politicians has been focused so far on aspects of juridical rationality, of reconstruction of a framework of statistical knowledge in urban contexts (at best), and occasionally only to aspects of reorganisation of territorial power dynamics (on the edge of the worst gerrymandering).
Vice versa, there is still a lack of a set of shared instruments of analysis of economic potentials and of development of our cities.
The statistical data on production, specialisation and added value that the Italian national statistics authority (Istat) makes available is even put together at a provincial level, not a municipal one; as a consequence, it engenders an extreme difficulty in putting together, in the analysis, what is actually urban and metropolitan in the wider setting of ex-provincial areas (not just at the level of bigger cities, but also of medium-sized cities and small dynamic centres).
The brief notes that follow do not aim at rectifying this serious gap. They rather constitute a first attempt to offer an instrument to understand the economic urban reality, so as to provide a stimulus for others to submit additional and more refined attempts.
The aim is to demonstrate that the concrete economic potential of Italian cities as development hubs would need more flexibility in the identification of their borders, to develop more the metropolitan functions to attract and support businesses, to dismiss the political and bureaucratic approach to the urban issue prevalent in the national decision-making koine.

Theoretical context of reference
Focusing on the economic aspect of urban realities, it is necessary to identify an adequate theoretical framework in order to represent the spatial dimension of development, with the aim of grasping the growth trajectories that characterise each territorial context. Economic development, in its evolution, affects the perimeter and the extension of the city, in a process that continuously redefines the borders, along lines that are difficult to decipher without an analysis of economic and social determiners of production levels.
The historic and economic context in which one must frame the most recent transformations of Italian cities is characterised by a fundamental transformation. On the one hand, the sunset of the Keynesian mixed economy, based on the pivotal role given to the great public enterprise and on a balanced management of distributive tensions; all elements that determined a continuous stimulus to the aggregated demand in the form of public investments and private consumption, fed by growing wages. On the other hand, there is a transition taking place-accelerated by the increasing degree of the European integration of Italy-towards a market economy exposed to international competition and unequipped of meaningful limits to the widening of inequalities, with the restraint of the role of the State and of all those political-institutional factors that constituted an obstacle to the continuous pressures to decrease the wages of workers and, by means of this, represented Keynesian antidotes to the risk of a lack of aggregated demand, that could result in a penalisation of the production.
In the current context, which has dealt with a long recessive crisis, those factors, linked to the domestic demand, that had determined the growth and the industrialisation of post-war Italy in the Eighties have turned out weakened, whereas the productive layer has undertaken evolutionary paths unknown before in our country.
Outside of the standards of the mixed economy, the private enterprise has shown in many cases an ability to compete on global markets, through a continuous renovation of processes and products, occupying niches with an elevated added value and highly elevated technological and qualitative content. These brief notes will be an attempt to draw a theoretical framework apt to favour the emergence of the specific traits of these transformations of the very morphology of the Italian productive layer, tracking down in the growth led by innovations and connection a possibility of development for cities that, keeping together virtuous examples of industrial vitality and social cohesion, seems to represent the best alternative to the mere application of the wage austerity paradigms and of the containment of expenditures, which characterises the European economic politics landscape of the past few years, as a sign of the exclusive reference (and at times obsessive) to the model of the economy hauled by productivity. This latter paradigm, coherent with an anti-Keynesian approach to the by now triumphant economic reality, 7 has found an almost unanimous application in Italy.
Along the traditional lines of the neoclassic theory, the decline of the Italian economy has been often explained as the inevitable corollary of a productive layer primarily constituted by small and tiny enterprises: reduced dimensions would be associated to low productivity determining, on the supply side, limits to the growth and to the ability to compete on markets.
This interpretative scheme presents two fundamental limitations. From the empirical point of view, it clashes with the reality of a decline of the great Italian enterprise, precisely when the best examples of industrial vitality arise from the main role exercised by small and medium-sized productive realities (Cersosimo & Viesti, 2012). From the theoretical point of view, the neoclassic scheme is limited to the hypothesis of the full employment of resources (an assumption that appears unacceptable in the current recessive context, with a workforce structurally underused) and focuses itself solely on the measure of the relocations of a given productive function, placing itself in the impossibility of grasping those changes and those innovations that are the mutation of the composition of the product (in primis product innovations), namely the very transformations that are at the centre of the current evolutionary horizon of the Italian productive system (Ginzburg, 2012).
When going into an analysis of urban contexts, there is an attempt to take some distance from the traditional vision of economic development, flattened on the dynamics of productivity, and to take the wider point of view offered by the so-called "network view of development".
Recent studies on technological innovation stress the role of connections in the development of new products, new functions and new processes. Starting from Hirschman's works (1958Hirschman's works ( , 1983; Haussmann, Hausmann and Hidalgo et al. (2014) show how the space for innovation, and thus for economic growth, originates from proximity relations between products, contiguity between industrial specialisations, diffusion of networks of enterprises and interaction between information on territories. The proximity multiplies the probabilities that a "discover" produces positive effects in different sectors than the one from which it originates, evolving in the passage from one sector to the other and thus perfecting itself. The evolutionary paths so designed can also engender further innovations, that spring from the application of a technique in a different field than the one in which this technique has been initially developed. (Lane, 2009) These intuitions amount to a complex vision of technological innovation, different from the stylised imagine offered by the traditional theory, which presupposes the production of one homogeneous (composite) good in every economy and associates to the idea of innovation solely the increase in productivity of that given good, excluding the possibility that innovation proceeds while modifying the composition of the social product, as it happens when there is the conception and the realisation of a new product. (Hausmann, Hwang, & Rodrik, 2007) The "network view of development" admits instead the inherent complexity of the technological innovation process and ascribes to the existence of territorial competences networks the ability of a certain layer of production to excel in international markets not by means of containing the costs of production but rather by means of researching the quality of products: around products there is the building of social relations that convey the knowledge eradicated in the territory towards the best opportunities of development, in an iterative process that can be reinforced by public action and oriented towards new forms of industrial politics: as stated by Ginzburg (2012) within this perspective "the development thus does not consist in the accumulation of capital (physical, social, human, etc.) but in the organisation and coordination, around heterogeneous products, of a plurality of heterogeneous competences".
It must be noted that such a theoretical framework brings to the core of the analysis the relevance of public intervention in the planning and coordination of local development: the function of directing and planning, given by the legislator to the metropolitan city, finds in this interpretative frame a valid operative interface capable of grasping the current dynamics of local development and innovation. And the same faculty (not duty) to the new "aree vaste" (wide areas), that have substituted ex lege the old provinces, of producing a strategic plan, offers a framework of reference, to this aim, to medium-sized cities as well.
An instance of the operative potential of the "network view of development" can be given by the complex mapping of the connection, the "product space", elaborated by Hausmann and Hidalgo et al. (2014) with the purpose to illustrate the variety of proximity relations between different segments of the market and, at the same time, the different collocation, central or peripheral, of single productive segments (Figure 1). The different products, divided according to commodityrelated categories, are represented by trade and travel hubs whose dimension reflects the relevance of the good in the international market. The connections have been deduced starting from an analysis of the productive specialisation: a country is specialised in the production of a good if it presents an index of specialisation of the export greater than the unit, and two goods result connected if a country appears specialised in the export of both. 8 It is thus defined a product space that shows the proximity relations between different productions, with the idea that from these relations innovation is going to spring: the possibilities of ideating new products or new processed increases with the increase of connections, real channels of innovation. Within this scheme, the lesser is the number of connections that a sector manifests, the more peripheral its position on the map will be, whereas the sectors characterised by the more elevated number of connections will appear at the centre of the graphic representation. The basic intuition is expressed by the idea that economic development represents a process of extension of productive specialisation from the periphery towards the centre of the map: to validate this, Hausmann and Hidalgo et al. (2014) show that developing countries register an elevated specialisation in peripheral sectors, where they remain confined, whereas more economically advanced countries arrive to occupy the central areas of the map, and thus to multiply the possibilities of innovation and growth (Hidalgo, Klinger, Barabasi, & Hausmann, 2007).
The present work seeks to apply this interpretative grid, conceived by its authors with reference to the wider context of global economy, to the level of metropolitan cities, by means of the computation of indexes of specialisation of exports at the provincial level.
The Balassa index is constructed in this case comparing the data on provincial export 9 with those of national export, from the Istat source, year 2014. 10 This methodological choice appeared valid with reference to the theoretical approach, the "network view of development", that identifies precisely in the territorial proximity relations between producers the main incentive to innovation and development: the intuition, peculiar of the school, to associate the development to the relations of similarity between products and between techniques to the dialectic between neighbouring productive units (Lane, 2009) finds, in our opinion, an effective application in the analysis of single provincial territories, suitable to grasp exactly those connection that are at the centre of Hirschman's reflection and of the authors that have developed his thought as well. The analytical representation of local productive vocations acquires all the more sense the more it is possible to circumscribe the territory that expresses it: with the passage from the global dimension (global market) to the local dimension (provincial market) the degree of detail of the investigation improves.
The application of this model to Italian metropolitan areas allows the emergence of the relevance of some industrial districts that have been able to resist to the pronounced decline in demand thanks to a strong ability to export their products on international markets. The quality of these areas appears linked to the productive vocation of the territories they are located in, but also to the long process of development and settling of competences and techniques that risks to be scattered with the ongoing of the crisis. For this reason, the results of the present analysis show all the potential expansive effects, both in the short and long term, of a renewed main role of an industrial politics that knows how to identify and support those segments of the industry that nowadays grant the best perspectives of development for the territories they are installed in. In the short run, a more efficient governance for industrial districts would attract foreign demand through the quality of the products rather than through wage deflation, thus stimulating the economy on the demand side. In the long run, the development of the industrial districts will foster the competitiveness of Italian commodities in the international markets, thus improving the supply side of the economy and connecting the national industry to those segments of the world market characterized by high prices and low demand volatility.

A few empirical evidences
The outcome of the application of these reasoning to the five metropolitan Italian cities examined is constituted by some graphical representations that picture the different possible trajectories of development of these areas. 11 In the case of Bari, mechatronics (represented in Figure 2 by the gearwheels) is considered as having a pivotal role in the product space, holding relevant opportunities for innovation; meanwhile, taking under consideration the connections of this productive segment with several specialisations and excellences both in the Bari area and that of the surrounding provinces, the productive layer of Bari represents the nerve centre of the development trajectories of both the local and extra-provincial area.
As far as the ex-province of Florence is concerned, whose area has been examined with those of the provinces of Prato and Pistoia, welcoming the "tri-polar redefinition" that the OECD has given in that "urban" area (Calafati, 2014b) what emerges is an integrated productive layer that presents an excellence in the textile, leather, footwear and clothing sectors pertaining to the "fashion system" that, however, already belonged to each of the three provinces ( Figure 3). Nevertheless, the consideration of the tri-polar area gives an opportunity to the other two provinces as well to avail themselves of Florence specialisations with a higher technological content, enlarging the  network of knowledge and allowing the shared development of new products and competences as well.
As regards Turin, the analysis has revealed the excellence of the automotive and an elevated specialization in the aerospace sector and in the mechanical one; the latter is located at the centre of the map and thus engenders numerous connections and development opportunities. Indeed, the sectors with the highest technological content revolve around this sector and the excellence of the automotive industry, as Figure 4 shows.
If we restrict the analysis to the two fundamental sectors of Turin's exports (automotive and aerospace), we observe that they constitute sectors of excellence in the neighbouring province of Cuneo, Asti and Aosta as well; recalculating the Balassa index taking into account these provinces, we obtain higher values without, however, affecting the specialisation map.
Hence, it would be reasonable to imagine development trajectories extended to the mentioned provinces as well (see Figure 5 below).
The study conducted in the metropolitan area of the city of Naples and on the eventual interactions with the other regional provinces has shown the existence in the metropolitan city of an excellence of the aerospace sector, in particular in the realisation of aircrafts and their components that is located in the peripheral part of the map differently from the other sectors that are necessary to the production of capital goods, such as the mechanical, plastic and electronics one, and therefore pivotal to the products universe ( Figure 6).
Anyway, the inspection carried out on the other productive specialisations present in the metropolitan city reveal, aside from the aforementioned excellence in aerospace, specialisations that are located in the peripheral parts of the maps (Figure 7). The reiteration of the analysis for the other provinces and the recalculation of the Balassa index for a wider area than that of the ex-province of Naples has shown that, even obtaining higher values than the initial ones, the best possible outcome is obtained by drawing together the specialisations    The metropolitan city of Rome presents two excellences (high-tech and chemistry) and four specialisations (aerospace, pharmaceutical, paper industry and petrochemical) in sectors with high and sky-high technological content ( Figure 10) but with a contained space for innovation, see the peripheral position in the map ( Figure 11).
Nonetheless, if the scope were to be extended to the provinces of Latina, Rieti and Frosinone, passing the regional borders to arrive to Aquila (Figure 12), the central positions of the map are "conquered" and specialisations like the pharmaceutical one become excellences, the same happens for the advanced mechanical sectors and the aerospace and electronics ones.
Lastly, as far as the former province of Milan is concerned (Figure 13), two excellences emerge from the analysis of the revealed comparative advantages: these are two particular branches of the electronics, namely the production of recoded medias, and the manufacturing of artistic-sportentertainment requirements. Moreover, the Milan area records seven sectors showing  a specialisation: textile productions, clothes, leather items, chemical-pharmaceutical goods, ICToptical equipment, ICT services and recycling products.
According to the product space, however, some limits of the Milanese productive system emerge since the most export-performing sectors lie on the outskirts of the map, where limited innovation opportunities can be exploited. Nevertheless, it is enough to enlarge our analysis to the bordering provinces to discover that the product space is almost totally covered, as the interland is remarkably featured by excellences and specialisation which are complementary to the Milano area. Particularly, we refer to the "core" productions of mechanics, the rubber and the paper industry, and a broader chemical district ( Figure 14). All in all, we can argue that, according to the products map, within the Milan area the "last" ring of the value chain is coined: here, the ultimate productions of a longer industrial process are produced, and this phase could not have been developed without the contribution of the neighbouring areas where the sectors of the "basic" industry is prosperous.
Summing up, the Milanese production system exhibit vocations in the technological pole of the creative industry, in the fashion system, in the pharmaceutical sector and in the informationoptical sector. Although these specializations generate a limited space for innovation (confined to "peripheral" products), a substantial enlargement in that space emerges when extending the analysis of the comparative advantages to the next provinces where a cutting-edge manufacturing industry is tracked (especially, Lodi, Bergamo and Varese). A similar organic set of interconnected specializations allows for the creation of a vital industrial district with a remarkable strategic potential.
Some empirical evidence can be useful to confirm the idea that districts, which headquarters are often located in the metropolitan cities analysed, actually host the excellences of Italian exportoriented productions. In 2014, Italy export volume grew at 2.2% with respect to the previous year. If we look at some sub-sectors of the manufacturing (which average annual growth rate was 2.4%), we see that those industries, which we have already proved to be excellences according to the comparative advantage approach, outperformed both aggregate export growth as well as the manufacturing one. Particularly, the fashion system (production of clothes, textures, leather items and accessories) experienced a +5% export growth, the automotive +4.2%, the aircraft production +4.3%, the mechanical engineering sector +3.6% (source: Istat, Coeweb, Ateco taxonomy 2007). To put it simply, productions which characterise districts studied in the present paper exhibited in 2014 a better export performance than sectors which productions were not "excellences" in manufacturing areas.

Final remarks
The theoretical analysis put forward and the empirical exemplifications related to the metropolitan areas of the Law 56/2014 make some final reflections possible.
With an attempt to use a structured analytical device, it becomes clear how the set of knowhow, specialisations and local productive traditions, particularly in the cities, has a remarkable significance with regard to the potentiality of local productive systems.
Moreover, the endurance test offered by the variation of the borders observed demonstrates how to the variation of the analysed area correspond relevant deviations in terms of the potentiality of the productive vocations with greater differences. The instances of Bari and Florence are emblematic in this sense; and this in spite of the inability to precisely circumscribe the scope by only including, among the ex-provinces, the municipalities that were cohesive and metropolitan in the proper meaning of the term, because of the reasons of availability of statistical data pointed out.
A further interesting evidence is that of the coexistence trend in the nature, come to be held even by small and medium-sized cities, of development hub and centre of a self-contained system from the point of view of people's circadian rhythms; symbolic in this sense, within the heterogeneous and fragmented ex-provincial area of Turin, the case of the mechanical system of Ivrea and of the Canevese. Equally important is the indication of the necessity, especially in the institutional contexts of the new metropolitan authorities, to avail oneself of the flexibility permitted by law to set a new perimeter of the territory to include under the new governance structure, if the development hub it represents is to be enhanced: to make an example, how could the hub of advanced chemistry and pharmacology of Pomezia be maintained halfway inside the province of Rome (and thus under the government of the new authority) and halfway outside?
With these indications, the purpose is not that of sustaining to have proved what would serve in order to put back the country on the path of the valorisation of its own productive territorial assets; nevertheless, it is believed of having at least presented the theme in non exclusively rhetorical terms and in terms of principle, in order to help mature a different awareness of its implications.